MIASMA

T O T H E M EM ORY Pollution and

OF
MY M O TH ER
Purification in
early Greek Religion

ROBERT PARKER

C L A R E N D O N PRESS · O X FO R D
 Oxford University Press, Walton Street, Oxford 0 x 2 6d p 1996 PREFACE
Oxford New York
Athens Auckland Bangkok Bombay
Calcutta Cape Town Dar es Salaam Delhi
Florence Hong Kong Istanbul Karachi T h e text of this new printing rem ains unchanged, some tiny corrections
Kuala Lumpur Madras Μαάήά Melbourne aside. But the subject has, o f course, m oved on, an d it is only for tech­
Mexico City Nairobi Paris Singapore nical reasons o f book production th at the following updating is o f the
Taipei Tokyo Toronto
briefest. T h e m ost im portant new prim ary evidence1 is th at published by
and associated companies in
M. H. Jam eson, D. R. Jo rd an , R. D . K otansky, A Lex Sacra from Selinous,
Berlin Ibadan
Greek, Roman and Byzantine Monographs, 11 (1993). Side A o f the new text
Oxford is a trade mark of Oxford University Press (of the mid-fifth century BC?) th at they present prescribes sacrifice (after
a death, or killing?) to, am ong others, ‘the polluted T ritopatores, as to
Published in the United States the heroes’ ( t o l s Τριτοπατρ€ΰσί t o Î s μιαροΐς ϊιόσπερ τ ο ί ς h e p o e a i and
by Oxford University Press Inc., New Tork subsequently to ‘the pure (Tritopatores)’ (τοίς κ{α)θαροΐς): the same set
© Robert Parker 1983
o f ancestral spirits before and after purification, the editors suggest, not
Reissued in paperback 1996 two distinct sets. Side B contains rules, strikingly similar in some regards
to those found in B 28-55 o f the Cyrene cathartic law (Appendix 2
All rights reserved. No part o f this publication may be reproduced, below), for ‘purification from pursuing/polluting spirits (ελαστέροι)’
stored in a retrieval system, or transmitted, in any form or by any means, (which are thus firmly attested in an early prose text o f no m arginal
without the prior permission in writing o f Oxford University Press.
Within the UK, exceptions are allowed in respect o f any fair dealing for the
kind). Rites o f this type, and their N ear Eastern analogues or predeces­
purpose o f research or private study, or criticism or review, as permitted sors, are im portant exhibits in W . B urkert’s ground-breaking study o f The
under the Copyright, Designs and Patents Act, 1988, or in the case of Orientalizing Revolution: Near Eastern Influences on Early Greek Culture in the Early
reprographic reproduction in accordance with the terms o f the licences Archaic Age (H arvard, 1992). O ne m ay agree with him th at ‘the contin­
issued by the Copyright Licensing Agency. Enquiries concerning uum from the M esopotam ian culture to the M editerranean is there’
reproduction outside these terms and in other countries should be
w ithout accepting that the similarities he dem onstrates (purification by
sent to the Rights Department, Oxford University Press,
at the address above sucking-pig, by lustration from a branch, by asphalt, with an onion . . .)
necessarily point to strong influence at a determ inate period in the early
British Libray Catahguing in Publication Data archaic age. P. M . C. Forbes-Irving in Metamorphosis in Greek Myths
Data available (Oxford, 1990) greatly extends analysis o f the m ore extreme forms o f pol­
lution— cannibalism or family m urder or incest— by revealing the imag­
Library of Congress Cataloging in Publication Data
inative role th at they play in myth, as suprem e horrors which can only
Parker, Robert
Miasma: pollution and purification in early Greek religion. be escaped or effaced by transform ation out o f hum an form. Pollution’s
Bibliography. Includes index. indifference to motive has often been seen as a scandal: it is interesting
1. Purity, Ritual— Greece. 2. Greece— Religion to observe the renewed insistence o f m oral philosophers th at o u r own
I. Title notions of responsibility (some would say m oral responsibility) an d even
BL788.P37 1983 292 '.2 82-1783.5
legal liability in several ways respect th at which, crassly an d contingently,
ISB N ( h i 9-8 1 4 7 4 2 -2 actually occurs, not th at which is intended (see the essays entitled ‘M oral
L uck’ by B. Williams (in his Moral Luck, Cam bridge, 1981) an d T. Nagel
3 5 7 9 10 8 G 4 2 (in his Mortal Questions, Cam bridge, 1979)). As Williams writes o f Oedipus
Printed in Great Britain
[Shame and Necessity, Berkeley, 1993: 69) ‘T h e whole o f the Oedipus
*9 Ο Λ / n s i a y on acid-free paper by Tyrannus, that dreadful m achine, moves to the discovery o f just one thing,
/ 3 > 9 Biddles Ltd. th at he did it’, a fact which is as horrific for m odern as for ancient spec­
---- . N G^djfford & King’s Lynn tators because ‘we know that in the story o f on e’s life there is an author­
i ’’’ ■ r ' Ά V» f" v ity exercised by w hat one has done, and n ot merely by w hat one has
intentionally d one’. At the level o f high theory, Françoise H éritier and

1 Note too the new examples o f shrine ‘entry-rules’, SEG xxxvi 376 and 1221.
 VI Preface

Alain T estart have argued th at the role o f taboo is n ot to preserve dif­

ferentiation (so M . Douglas) b u t to create it, to prevent the association
o f like with like (see most recently A. T estart, Des mythes et des croyances:
CONTENTS
Esquisse d’une théorie générale, Paris, 1991). T o take one o r two illustrations
from many: the symbolic world o f the Samo, a people o f the U pper A B B R E V IA T IO N S A N D E D IT IO N S ix
V olta studied by H éritier, is said to be shaped by an opposition between
‘h o t’ and ‘cold’, and typical prohibitions am ong them are on making love Intro d u ctio n 1
(a hot act) on the ground (a h ot place), or to a lactating w om an (milk,
like sperm , being hot); in m any cultures, w omen are excluded from the 1. Purification: a Science of Division 18
S hunt and from sacrifice because they, like the victims o f these practices,
are animals that bleed. T o an am ateu r eye, the theory appears to p ro ­ 2. B irth and D eath 32
vide an exegesis o f the form o f taboos (or some am ong them) rather than
an explanation at any deeper level. O f the subjects o f the individual 3 . T h e W orks of A phrodite 74
chapters o f this book, the ‘W orks o f A phrodite’ have received the most,
and the m ost sophisticated, attention: from an extensive literature, let us 4 . T h e Shedding o f Blood 104
pick out H . van S taden’s subde argum ent that the modes o f treatm ent
applied to w om en patients in the H ippocratic corpus imply th at their 5. Sacrilege 144
need for ‘purification’ is m ore insistent th an that o f m en (‘W om en and
dirt’, in Helios 19 (1992), 7- 30); an d the counter-argum ent o f L. D ean- 6. C urses, Fam ily Curses, and the Structure of Rights 191
Jones ( Women’s Bodies in Classical Greek Science, O xford, 1994) th at male
fear o f pollution by w om en only grew up in the post-classical period, in 7. Disease, Bew itchm ent, and Purifiers 207
consequence o f new physiological theories and o f growing uncertainties
in the relations between the sexes. 8. Divine V engeance and Disease 235
R .C .T .P.
9. Purifying the City 257
Oriel College, Oxford
September 1 9 9 5
10. Purity an d Salvation 281

11. Some scenes from Tragedy 308

1983 PREFACE Epilogue 322

This book has developed from an O xford dissertation. I hope th at those A P P E N D IC E S

I thanked for their help with the thesis, and also its examiners, will accept
now a collective expression o f gratitude. For subsequent advice on whole 1. T h e G reek for T aboo 328
chapters I am very grateful to D r. J . N. Brem m er, D r. G. E. R. Lloyd,
and D r. C. Sourvinou-Inwood, an d on individual points or sections to 2. T h e C yrene C athartic Law 332
Professor A. M. Davies, D r. N. S. R. H ornblow er, D r. D. M . Lewis, and 3 . Problem s concerning ‘Enter pure from ..
Dr. M. E. Tucker. Mrs. A. M . C ripps kindly typed m uch o f the m anu­ requirem ents in sacred laws 352
script, and I am particularly grateful to M r. R. W . B. B urton for his care­
ful scrutiny o f the proofs. For financial aid I thank the Craven 4. A nim als and Food 357
com mittee, the Provost and Fellows o f O riel College, an d the Faculty
5. T h e R itual Status of the Justified Killer at Athens 366
B oard o f L iterae H um aniores in O xford. M y greatest debts are to H ugh
Lloyd-Jones, for inspiration and encouragem ent, and, for reasons too 6. T h e R itual o f Purification from Homicide 370
various to m ention, to my wife Joanna.
R.C .T .P.
Oriel College, Oxford
November 1 9 8 2
viii Contents

7. Exile and Purification of the Killer in Greek M yth 375

8. G ods particularly concerned with Purity 393

IN D E X E S 395 ABBREVIATIONS AND EDITIONS

For ancient au th o rs the abbreviations in The Oxford Classical

Dictionary,2 ed. N. G . L. H am m ond and H. H. Scullard, 1970, ix-xxii,
have been followed where available (with a few trivial exceptions
listed below ). For other cases those in L iddell-Scott-Jones, A Greek
English Lexicon,9 O xford, 1940 (LSJ), have been used, and where this
too offers n o thing an easily recognizable form has been chosen. Frag­
m ents are cited according to the num eration o f the following collec­
tions: H esiod, R. M erkelbach and M. L. W est, O xford, 1967; lyric
poets, P M G or SLG or PLF (see below) ; iambic and elegiac poets, M. L.
W est, Iambi et Elegi Graeci ante Alexandrum cantati, O xford, 1971—2;
pre-S ocratics, H . Diels, Die Fragmente der Vorsokratiker,6 revised by W.
K ran z , Berlin, 1951—2; historians, F. Jacoby, Fragmente der griechischen
Historiker, Berlin, 1923- ; Pindar, B. Snell/H . M aehler, Leipzig,
1975; A eschylus, N auck TG F (see below); Sophocles, A. C. Pearson,
C am b rid g e, 1917 (sam e num bers in S. R adt, G öttingen, 1977);
E uripides, N auck, TGF, except where H. v. A rnim , Supplementum
Euripideum, Bonn, 1913, or C. A ustin, Nova Fragmenta Euripidea in
Papyris Reperta, Berlin, 1968, are available (in these cases the editor’s
nam e is ad d ed ); A ttic comic poets except M enander, T . Kock,
Comicorum Atticorum Fragmenta, Leipzig, 1880-8; D orian comedy,
K aibel, CGF (see below); comic fragm ents known from papyri, Page,
GLP (see below) a n d /o r C. A ustin, Comicorum Graecorum Fragmenta
in Papyris Reperta, Berlin, 1973 (in these cases the editor’s nam e
is ad d ed ); M enander, A. K oerte/A . Thierfelder, Leipzig, 1959;
H ellenistic poets, Coll. Al. (see below). Specific editors are named
w hen o ther fragm ents are cited. H ippocratic texts are cited by the
ch ap ters o f E. L ittré, Œuvres completes d ’Hippocrate, Paris, 1839—61,
w ith references to his volume and page num bers in brackets (L. or
L ittré). B ut (H ipp.) Morb. Sacr. is cited by page and line in the edition
o f W. H. S. Jo n es, H arvard, 1923 ( J ) , and by section in the edition of
H . G rensem ann (Ars M edica 2.1), Berlin, 1968 (G.).
A bbreviations o f the epigraphical collections that are more com­
m only cited are listed below. For other items the abbreviations of LSJ
xii—xliii have been used.
F or periodicals the abbreviations of The Oxford Classical Dictionary2
have been used, w ith some exceptions and additions th at are listed
below.
X Abbreviations and Editions xi
Abbreviations and Editions
1. Abbreviations o f ancient authors, periodicals, collections o f texts, series, and
RAC Reallexikon fü r Antike und Christentum, ed. T. K lauser, Stuttgart,
general reference works
1950-
A B SA Annual o f the British School at A thens ; R E A. Pauly, G. Wissowa, and others, Real-Emyclopädie der klassischen
A JA American Journal o f Archaeology Altertumswissenschaft
A JP American Journal o j Philology R E A Revue des études anciennes
A nt. A ntiphon R E G Revue des études grecques
Ant. u. Chr. F. J . Dölger, editor and sole contributor, Antike und R E L Revue des études latines
Christentum R G VV Religionsgeschichtliche Versuche und Vorarbeiten
Ath. M itt. Mitteilungen des deutschen archaeologischen Instituts, Athenische R H R Revue de l ’histoire des religions
Abteilung „ R M L W . H . Roscher, Ausführliches Lexikon der griechischen und
Bacch. Bacchylides römischen Mythologie, Leipzig, 1884—1937
B E F A R Bibliothèque des écolesfrançaises d ’Athènes et de Rome Schw yzer E. Schwyzer, Dialectorum Graecarum exempla epigraphica
Buck C. D. Buck, The Greek Dialects, Chicago, 1955 potiora, Leipzig, 1923
Coll. A l. J . U . Powell (ed.), Collectanea Alexandrina, Oxford, 1925 SE G Supplementum Epigraphicum Graecum
CP Classical Philology « S G D I H. Gollitz and others, Sammlung der griechischen Dialektinschriften,
D ar.-S ag . C. D arem berg and E. Saglio, Dictionnaire des antiquités G öttingen, 1884—1915
grecques et romaines, Paris, 1877-1919 S IG Sylloge Inscriptionum Graecarum, ed. VV. D ittenberger, ed. 2,
H S Ç P Harvard Studies in Classical Philology Leipzig, 1898-1901, ed. 3, 1915-24
H T R Harvard Theological Review SL G D. L. Page (ed.), Supplementum Lyricis Graecis, Oxford, 1974
IG Inscriptiones Graecae. F or details see LSJ, xlii. IG I2 is being S olm sen/F raenkel F. Solmsen, Inscriptiones Graecae ad inlustrandas
replaced by IG I3, ed. D. M . Lewis, Berlin, 1981 - Dialectos selectae, ed. 4, revised by E. Fraenkel, Leipzig, 1930
K aib el, CGF G . K aibel, Comicorum Graecorum Fragmenta, Berlin, S S R Studi Storico-Religiosi
1899 SV F H. v. Arnim (ed.), Stoicorum Veterum Fragmenta, Leipzig, 1903-24
L S A F. Sokolowski (ed.), Lois sacrées de l ’Asie Mineure, Paris, 1955 T hesleff H . TheslefT (ed.), The Pythagorean Texts o f the Hellenistic
LSC G F. Sokolowski (ed.), Lois sacrées des Cités grecques, Paris, 1969 Period, A bo, 1965
L SS F. Sokolowski (ed.), Lois sacrées des Cités grecques, supplément, W ünsch R. W ünsch, TabellaeDefixionum = /G I I I . 3appendix, 1897
P aris, 1962 Z P E Zeitschriftfü r Papyrologie und Epigraphik
M /L R. M eiggs an d D. M . Lewis, A Selection o f Greek Historical
Inscriptions, O xford, 1969
M ichel C. M ichel, Recueil d ’inscriptions grecques, Brussels, 1900 I 2. Other works cited in abbreviatedform
Φ
N auck, T G F A. N auck, Tragicorum Graecorum Fragmenta,1 Leipzig,
1889 p A b t A. A bt, Die Apologie des Apuleius von Madaura und die antike
H Zauberei, R G W 4.2, Giessen, 1908
O F O . K ern , Orphicorum Fragmenta, Berlin, 1922
A dkins A. W . H . Adkins, Merit and Responsibility. A Study in Greek
Page, G LP D. L. Page (ed.), Greek Literary Papyri 1 (Poetry),
H a rv a rd , 1942 Values, O xford, 1960
#A lexiou M . Alexiou, The Ritual Lament in Greek Tradition, C am ­
PLF E. Lobel an d D. L. Page (eds.), Poetarum Lesbiorum Fragmenta,
O xford, 1955 bridge, 1974
A m an d ry P. A m andry, La Mantique apollinienne à Delphes (BEFAR
P M G D. L. Page (ed.), Poetae Melici Graea, Oxford, 1962
170), Paris, 1950
xii Abbreviations and Editions
A ndronikos M . A ndronikos, Totenkult (Archaeologia H om erica III
VV), G öttingen, 1968
A rb esm an n P. R. A rbesm ann, Das Fasten bei den Griechen und Römern,
R G W 2 1.1, Giessen, 1929
B arth F. B arth , Ritual and Knowledge among the Baklaman o j New
Guinea, O slo an d New H aven, 1975
B enveniste E. Benveniste, Le Vocabulaire des institutions indo-
européennes, 2 vols., Paris, 1969
B lack-M ichaud J . Black-M ichaud, Cohesive Force. Feud in the
Mediterranean and the Middle East, Oxford, 1975
Blum R. an d E. Blum, The Dangerous Hour. The Lore o f Crisis and
Mystery in Rural Greece, L ondon, 1970
B o n n er/S m ith R. J . B onner and G. E. Smith, The Administration of
justice from Homer to Aristotle, 2 vols., Chicago, 1930, 1938.
B orgeaud P. B orgeaud, Recherches sur le dieu Pan (Bibliotheca
H elvetica R o m ana 17), R om e/G eneva, 1979
B oyancé P. Boyancé, Le Culte des Muses chez les philosophes grecs, Paris,
1937
Boyce M . Boyce, A Persian Stronghold o f Zoroastrianism, Oxford, 1977
B rem m er J . N. B rem m er, The Early Greek Concept o f the Soul,
P rinceton, 1983.
B ru n e a u P. B runéau, Recherches sur les cultes de Délos à l ’époque helléni­
stique et à l ’époque impériale (B E F A R 217), Paris, 1970
B u rk ert, GR W. Burkert, Griechische Religion der archaischen und klassi­
schen Epoche, S tu ttg art, 1977
B u rk ert, H N ------ , Homo Necans. Interpretationen altgriechischer
Opferriten und Mythen, R G W 32, Berlin, 1972
B urkert, L S ------ , Lore and Science in Ancient Pythagoreanism, Cambridge
M ass., 1972, a revised English edition, translated by E. L.
M in a r, o f Weisheit and Wissenschaft: Studien zu Pythagoras, Philolaos
und Platon, N ürnberg, 1962
B u rk ert, S H ------ , Structure and History in Greek Mythology and Ritual,
Berkeley, 1979
B u solt/S w oboda G . Busolt, Griechische Staatskundei , 2 vols, (the
second revised by H. Sw oboda), M unich, 1920, 1926
B uxton J . B uxton, Religion and Healing in Mandari, O xford, 1973
C a lh o u n G. M . C alhoun, The Growth o f Criminal Law in Ancient
Greece, Berkeley, 1927
C am p b ell J . K . C am pbell, Honour, Family and Patronage. A Study o f
Institutions and Moral Values in a Greek Mountain Community, Oxford,
1964
Abbreviations and Editions xiii
C linton K . C linton, The Sacred Officials o f the Eleusinian Mysteries
(T ransactions of the Am erican Philosophical Society, new series,
64), P hiladelphia, 1974
Cook A. B. Cook, Zeus. A Study in Ancient Religion, 3 vols., C am ­
bridge, 1914—40
C ro issan t J . C roissant, Aristote et les mystères, Liège, 1932
D avies J . K . Davies, Athenian Propertied Families, O xford, 1971
D efradas J . D efradas, Les Themes de la propagande delphique (Études et
com m entaires 21), Paris, 1954 (ed. 2,-1972)
D etienne, Eugénies M . D etienne, ‘Violentes “ eugénies” ’, in M.
D etienne an d J . P. V ernant, La Cuisine du sacrifice en pays grec,
Paris, 1979
D etienne, Jardins M . D etienne, Les Jardins d ’Adonis, Paris, 1972
D e u b n er L. D eubner, Attische Feste, Berlin, 1932
Diels H . Diels, Sibyllinische Blätter, Berlin, 1890
D odds E. R. D odds, The Greeks and the Irrational, California, 1951
D odds, Progress E. R. D odds, The Ancient Concept o f Progress, and other
Essays, O xford, 1973
D ouglas M . Douglas, Purity and Danger. An Analysis o f Concepts o f /.
Pollution and Taboo, London, 1966 (cited from the Pelican edition,
1970)
D over K . J . D over, Greek Popular Morality in the Time o f Plato and /
Aristotle, O xford, 1974
D u m o n t L. D um ont, Homo Hierarchichus, Eng. trans., London, 1970
D urkheim E. D urkheim , The Elementary Forms o f the Religious Life,
trans. J . W. Swain, London, 1915
E delstein E. J . and L. Edelstein, Asclepius, a collection and interpretation
o f the testimonies, 2 vols., Baltimore, 1945
E delstein, A M Ancient Medicine, Selected papers o f L. Edelstein, ed.
O . an d C . Tem kin, Baltim ore, 1967
E itrem , Beiträge S. Eitrem , Beiträge zur griechischen Religionsgeschichte.
2, 3, K ristian ia, 1917, 1920
E itrem , Opferritus S. Eitrem , Opferritus und Voropfer der Griechen und
Römer, K ristiania, 1915
E v an s-P ritch ard E. E. E vans-Pritchard, Nuer Religion, Oxford, 1956
F arnell L. R. Farneil, The Cults o f the Greek States, 5 vols., Oxford,
1896-1909
F ehling D. Fehling, Ethologische Überlegungen auf dem Gebiet der
Altertumskunde (Z etem ata 61), M unich, 1974
xiv Abbreviations and Editions
F eh rle E. Fehrle, Die kultische Keuschheit im Altertum, R G W 6,
G iessen, 1910
F ontenrose J . Fontenrose, The Delphic Oracle, Berkeley, 1978
F o u cart P. F oucart, Les Mystères d ’Eleusis, Paris, 1914
F u g ier H . F ugier, Recherches suri ’expression du sacré dans la langue latine,
Paris, 1963
G a g a rin , Drakon M . G agarin, Drakon and Early Athenian Homicide
Law, Yale, 1981
G e b h a rd V. G ebhard, Die Pharmakoi in Ionien und die Sybakchoi in
Athen, diss. M unich, 1926
G e rn e t L. G ern et, Recherches sur le développement de la penséejuridique et
morale en Grèce, Paris, 1917
G ern et, Anthropologie L. G ernet, Anthropologie de la Grèce antique,
P aris, 1968
G ern et, Antiphon L. G ernet, Budé edition o f A ntiphon, Paris, 1923
G inouvès, R. G inouvès, Balaneutiké, recherches sur le bain dans
l ’antiquité grecque (B E F A R 200), Paris, 1962
G lotz G. G lotz, La Solidarité de la famille dans le droit criminel en Grèce,
P aris, 1904
G n o li/V e rn a n t G. Gnoli and J . P. V ernant (eds.), La Mort, les morts
dans les sociétés anciennes, C am bridge/P aris, 1982
G o ltz D . G oltz, Studien zur altorientalischen und griechischen Heilkunde,
Therapie, Arzneibereitung, Rezeptstruktur (Sudhoffs Archiv Beiheft
16), W iesbaden, 1974
G ra f F . G raf, Eleusis und die orphische Dichtung Athens in vorhellenistiscker
Zeit, R G W 33, Berlin, 1974
G riffin J . Griffin, Homer on Life and Death, Oxford, 1980
G ru p p e O . G ru p p e, Griechische Mythologie und Religionsgeschichte,
2 vols., M unich, 1897—1906
G u th rie , OGR W. K . C. G uthrie, Orpheus and Greek Religion,
L ondon, 1935
G u th rie , H G P W. K. C. G uthrie, A History o f Greek Philosophy, vols.
1 -3 , C am bridge, 1962-9
H a rriso n A. R. W. H arrison, The Law o f Athens, 2 vols., Oxford,
1968, 1971
H aslu ck M . H asluck, The Unwritten Law in Albania (ed. J . H.
H u tto n ), C am bridge, 1954
H eld en sag e C. R obert, Die griechische Heldensage, Berlin, 1920—6
( = p a rt 2 o f P reller/R obert, below)
Abbreviations and Editions XV
H erter, Dämonen H . H erter, ‘Böse D äm onen im frühgriechischen
V olksglauben’, Rheinisches Jahrbuch f . Volkskunde 1 (1950), 112-
43, rep rin ted in his Kinne Schriften, M unich, 1975, 4 3 -7 5 (from
w hich I cite)
H ertz R. H ertz, Death and the Right Hand, trans. R. and C . Needham ,
L o n d o n ,1960
Ja c o b y F. Ja c o b y , Atthis, the Local Chronicles of Ancient Athens, Oxford,
1949
J o rd a n B. J o rd a n , Servants o f the Gods (H ypom nem ata 55),
G öttingen, 1979
K u d lien F. K udlien, Der Beginn des medizinischen Denkens bei den
Griechen, Z ürich /S tu ttg art, 1967
K u rtz /B o a rd m a n D. C. K u rtz and J . B oardm an, Greek Burial
Customs, London, 1971
L a n a ta G . L a n ata , Mediana magica e religionepopolare in Greciafino ali’
età di Ippocrate, Rome, 1967
L atte, H R K . L atte, Heiliges Recht, T übingen, 1920
L atte, Kl. Sehr. K. L atte, Kleine Schriften zu Religion, Recht, Literatur und
Sprache der Griechen und Römer, ed. O . Gigon, W. Buchwald,
W . K unkel, M unich, 1968
L atte, Mord K . L atte, art. Mord in R E, reprinted in Kl. Sehr.
L atte, R R K . L atte, Römische Religionsgeschichte, M unich, 1960
L évi-Strauss C. Lévi-Strauss, La Pensée sauvage, Paris, 1962, cited
from the translation The Savage Mind, London, 1966
L évy-B ruhl L. Lévy-Bruhl, Primitives and the Supernatural, trans.
L. A. C lare, London, 1936
L ien h ard t G. L ienhardt, Divinity and Experience. The Religion o f the
Dinka, O xford, 1961
L inders T . L inders, Studies in the Treasury Records o f Artemis Brauronia
Found in Athens, Stockholm, 1972
L inforth I. M . Linforth, The Arts o f Orpheus, Berkeley, 1941
L ipsius J . H . Lipsius, Das attische Recht und Rechtsverfahren, 3 vols.,
Leipzig, 1905—15
L loyd G . E. R. Lloyd, Magic, Reason and Experience, Cam bridge, 1979
L loyd-Jones H . J . Lloyd-Jones, The Justice o f Zeus, Berkeley, 1971
Lobeck C . A. Lobeck, Aglaophamus, sive de theologiae mysticae
Graecorum libri tres, Regimontii Prussorum , 1829
M acdow ell, Law D. M . MacDowell, The Law in Classical Athens,
London, 1978
xvi Abbreviations and Editions
M acdow ell, Homicide D. M . M acDowell, Athenian Homicide Law in the
Age o f the Orators, M anchester, 1963
M euli, Ges. Sehr. K. M euli, Gesammelte Schriften, ed. T . Geizer, Basle/
S tu ttg a rt, 1975, 2 vols.
M ikalson J . D. M ikalson, The Sacred and Civil Calendar o f the Athenian
Year, Princeton, 1975
M o u lin ier L. M oulinier, Le Pur et l ’impur dans la pensée des Grecs
d'Homère à Aristote (Etudes et com m entaires 12), Paris, 1952
M u rr J . M u rr, Die Pflanzenwelt in der griechischen Mythologie, Inns­
bruck, 1890
M ylonas G. E. M ylonas, Eleusis and the Eleusinian Mysteries, Prince­
ton, 1961
N ilsson, GF M . P. Nilsson, Griechische Feste von religiöser Bedeutung mit
Ausschluss der attischen, Leipzig, 1906
N ilsson, GGR M . P. Nilsson, Geschichte der griechischen Religion, vol. i,
ed. 3, M u n ich , 1969, vol. ii, ed. 2, 1951, (the reference is to vol. i
unless otherw ise stated)
N ilsson, Op. Sei. Martini P. Nilsson Opuscula Selecta, 3 vols., Lund,
1951-60
N ock A. D. Nock, Essays on Religion and the Ancient World, cd.
Z. S tew art, 2 vols., O xford, 1972
O rfism o Orfismo in Magna Grecia (Atti del quattordicesim o convegno
di studi sulla M agna G recia), Naples, 1975
P ark e/W o rm ell H. W. Parke and D. E. W ormell, The Delphic
Oracle2, vols., O xford, 1956. P/W 50 = response n. 50 in vol. ii
P o llo ck /M aitlan d Sir F. Pollock and F. W. M aitland, The History of
English Law before the Time o f Edward I, ed. 2, 2 vols., Cam bridge,
1923
P re lle r/R o b e rt L. Preller, Griechische Mythologie, ed. 4 by C. Robert,
vol. i, Theogonie und Götter, Berlin, 1887—94
P ritc h e tt W . K . P ritchett, The Greek State at War, 3 vols., Berkeley,
1974-9
P /W cf. P arke/W orm ell above
R ead M . H . R ead, Culture, Health and Disease, London, 1966
R everdin O . R everdin, La Religion de la cité platonicienne, Paris, 1945
R ich ard so n N .J . R ichardson, The Homeric Hymn to Demeter, Oxford,
1974
R o b ert, Oidipus C. R obert, Oidipus, 2 vols., Berlin 1915
Abbreviations and Editions xvii
R obertson Sm ith W. R obertson Sm ith, Lectures on tht Religion of
the Semites, new edition 1894 (cited from the edition London,
1901)
R ohde E. R ohde, Psyche, ed. 2 H eidelberg, 1897, cited from the
tra n sla tio n by W. B. Hillis, London, 1925
R u d h a rd t J . R u d h ard t, Notions fondamentales de la pensée religieuse et
actes constitutifs du culte dans la Grèce classique, G eneva, 1958
Schw enn Fr. Schw enn, Die Menschenopfer bei den Griechen und Römern,
R G W 15.3, Giessen, 1915
Sim on B. Sim on, M ind and Madness in Ancient Greece, Cornell, 1978 ;
Snodgrass A. M . Snodgrass, The Dark Age o f Greece, Edinburgh,
1971
Srinivas M . N. Srinivas, Religion and Society among the Coorgs o f South
India, O xford, 1952
Steiner F. Steiner, Taboo, London, 1956 (cited from the Pelican
edition, 1967)
Stengel P. Stengel, Die griechischen Kultusaltertümei3, M unich, 1920
Stiglitz R. Stiglitz, Die grossen Göttinnen Arkadiens, V ienna, 1967
S troud R. S. Stroud, Drakon’s Law on Homicide (Univ. Cal. publ. in
class, stud. 3), Berkeley, 1968
T h o m as K. T hom as, Religion and the Decline o f Magic, London, 1971
(cited from the Penguin U niversity Book edition, 1973)
Töpffer J . T öpfler, Attische Genealogie, Berlin, 1889
T u rn e r V. W . T u rn er, The Ritual Process, London, 1969 (cited from
the Pelican edition, 1974)
v an G ennep A. van G ennep, The Rites o f Passage, trans. M. B.
V izedom and G. L. Calfee, London, 1960
V ern a n t, Pensée J . P. V ernant, Mythe et pensée chez les Grecs, 2 vols.,
Paris, 1965
V e rn a n t, Société J . P. V ernant, Mythe et société en grèce ancienne, Paris,
1974
V e rn an t, Tragédie J . P. V ernant and P. V idal-N aquet, Mythe et
tragédie en grèce ancienne, Paris, 1973
V ickers B. W. Vickers, Towards Greek Tragedy, London, 1973
W ach sm u th D. W achsm uth, Πόμπιμος b δαίμων, Untersuchungen zu
den antiken Sakralhandlungen bei Seereisen, diss. Berlin, 1967
W ä ch te r T . W ächter, Reinheitsvorschriften im griechischen Kult, R G W
9.1, G iessen, 1910
xviii Abbreviations and Editions
W h iteh ead D . W hitehead, The Ideology o f the Athenian Metic
(C am b rid g e Philological Society Supplem entary Volume iv),
C am b rid g e, 1977
W ilam ow itz, Glaube U . von Wilamowitz-MoellendorfT, Der Glaube
der Hellenen, Berlin, 1931-2 (cited from ed. 3, D arm stadt, 1959,
w ith altered pagination)
W illiger E. W illiger, Hagios, R G V V 19.1, Giessen, 1922
Z iehen L. Ziehen, Leges Graecorum sacrae e titulis collectae, 2.1, Leipzig,
1906
Z u n tz G. Z u n tz, Persephone, O xford, 1971
INTRODUCTION
A nyone who has sam pled a few of the most commonly read
G reek texts will have encountered pollution. In tragedy, the
plague a t the opening of the Oedipus Tyrannus is caused by it, it
precipitates C reon’s repentance in the Antigone, while Orestes in
the Oresteia, although he is driven to the m atricide by the fear of
one pollution, is seized by another after performing it. In
history, it plays, perhaps, a larger part than any other religious
m otif in the austere Thucydides. A Greek state in the fifth
century, we learn from him, m ight attribute a natural disaster
to a pollution it had incurred, and he shows us the Athenians
expelling the Delians from their island to ensure the purity of
this religious centre.1 In the fourth century, Aeschines could
envisage D em osthenes as the ‘dem on who pollutes all Greece’,
an d brings it to m isfortune.2 A glance at evidence of a different
kind, inscriptions regulating cult, shows how the concern for
p u rity affected the individual in his everyday religious practice.
T h e th re a t of pollution is, it seems, the dom inating concern of
the Superstitious M an o f T heophrastus.3
M any questions are w orth asking about a phenom enon of
this kind. T h e reader o f tragedy will wish to know w hether he is
confronted, in pollution, with a literary m echanism or a living
preoccupation. States intervened in the internal affairs of others
to ‘drive out the pollution’, or m ade war on account of it;4
pollution was usually a pretext, but the historian may be in­
terested in the unchallengeable validity assigned to such a
justification for aggression. T h e student of Greek values will
consider how the fear of pollution functions as an inhibiting
factor in a society whose dom inating values are of a different
kind. Som e have seen here a historical development; thus the
postulated grow th of pollution fears is central to the famous
1 Thuc. 1 .1 2 8 .1 , 5.1. C f. 1 .1 2 6 -1 3 5 , 4 .9 7 .2 - 9 9 ,3 .1 0 4 .1 - 2 ,5 .3 2 .1 for p ollu tion in this
a u th o r .
2 3 .1 5 7 f., a n d often .
3 T h e o p h r . Char. 16
4 H d t. 5 .7 2 .1 . W ars: see pp. 165 fT. below .
2 Miasma
hypothesis th a t describes the spiritual history of early Greece as
a transition ‘from sham e culture to guilt culture’.5 Still in the
sphere of values, a question arises about the relation of pollution
to m orality; the irrationality of the former, perhaps, makes it
h ard for a rational system of the latter to develop. The religious
historian m ay wonder how pollution relates to ‘sin’, prime
source of religious danger in a different tradition; this question
becomes o f central im portance in the case of those alternative
religions of the Greek world whose goal was salvation and
principal route towards it ‘purification’. T he subject is not
irrelevant even to the historian of science, since the Hippocratic
doctor, in seeing ‘im purity’ as a cause and symptom of disease,
is an heir to the prophet or oracle. T he origin of disease raises
the m ore general question of how the early Greeks, individually
and collectively, responded to the afflictions that befell them.
These problem s, and more, present themselves to the H el­
lenist from the Greek m aterial alone. F urther questions are
raised by anthropology, which shows that pollution belief is
closer to being a hum an universal than an idiosyncrasy of the
G reeks. W h at is it there for? It is not a product of the ill-focused
terro r th at perm anently invests the savage m ind, because that
terro r is an invention of nineteenth-century anthropology.6
Does it perhaps shore up those areas of the social structure and
value system th at lack any other sanction? T h a t would help to
explain the central problem of the divergence in these beliefs
betw een one society and another. For pollutions that derive
from involuntary acts, however, an explanation in terms of
som e o th er form of order which men seek to impose on their
experience will obviously be required. There are rules, too, that
govern the m inutiae of everyday life. ‘Don’t cut your hair or
your nails a t a festival’, urged Pythagoras. C an such trivial
injunctions be related to a broader system of Pythagorean
order, an d how could such a system be explained?
Pollution, then, is a pervasive phenomenon which raises
diverse questions. T he first problem is that of establishing a
w orking definition of the thing itself. T he title of this book seems
to announce a precise area of inquiry - a single Greek word
denoting, one m ight hope, an easily isolable them e whose clar-
5 D o d d s , C h . 2.
* S e e D o u g la s , 11 f.
3
Introduction
ity would be in happy contrast to the am biguity7 which sur­
rounds the concept of pollution in anthropological literature.
T h e hope proves delusive; the mia- word group is applied to a
diverse range of things, and if one isolates within it a category
th at seems to have real unity, the same criteria that have been
applied in order to constitute it dem and that phenom ena de­
scribed by different words should also be included. An English
exam ple will illustrate this simple point: ‘innocent’ thoughts
associate better with a ‘pure’ m ind than does ‘pure’ alcohol.
N ot m erely words are involved, of course, but forms of be­
haviour —avoidance, expulsion, ablution, and the like. As a
sim ple appeal to vocabulary will not serve, some further
a tte m p t to define w hat is here understood by pollution becomes
indispensable. T he approach may seem disconcertingly
scholastic, but it is clear in practice (witness the debate about
pollution in H om er) that discussion in this area is likely to be
conducted a t cross purposes w ithout an explicit definition of
term s.
T he basic sense of the mia- words is that of defilement, the
im pairm ent of a thing’s form or integrity. Things that in
English we term ‘dirty’ are a common source of such defile­
m ent, but there are defilements deriving from things that are
not dirty in themselves, or not deriving from m atter at all.
Miaind can be used for the pollution of a reputation through
unw orthy deeds, or of truth through dishonesty;8justice, law,
an d piety are in danger of defilem ent.9 This book treats, among
all the possible defilements to which mia- words could be ap­
plied, a sub-category that is to a considerable extent m arked out
by linguistic usage. T he verb miainö is more freely applied, but
w here the noun miasma or the adjective miaros (except in the
sense o f ‘revolting’) occur, they alm ost always10 refer to a condi-
7 S ee e .g . B u xton , 19 0 n. 2, Λ. S. M eigs, ‘A Papuan P erspective on P ollu tion ’, Man,
n .s. 13 (1 9 7 8 ), 3 0 4 - 1 8 .
8 S o lo n , fr. 32 .3 , Pind. Pyth. 4.100, Eur. Hel. 1000, cf. Pind. Nem. 3. 16 (a citizen ’s
d is g r a c e ‘d e files’ his agora), αίσχύνω sim ilarly used, Horn, II. 23.571, 6.209;
ά π ορρ νπα ίνω , S op h . fr. 3 14.159 R adt; κηλίόα προσβαλεΐν, Eur. Stheneboea, prologue
37 Γ ν. A rn im . 0 A e sc h . A g. 1669, Sept. 344, Eur. Supp. 378.
10 In H ip p . Flat. 5, 6 (6. 96, 98 Littré) miasmata in the air cause d isease. T h e reading
miasma in S E G x x v 4 4 7 .6 (A rcadia, 3rd c. BC:) is uncertain, as the editor D r. G . J . te
R ie le k in d ly inform s m e after re-inspection o f the stone (cf. Bull. Épig. 1969 n. 267); if
correct, it has a secu la r sen se, ‘offen ce’. M iaros o f ritual status, L S C G 56, LSS 115 A 10,
18; o f r itu a l/le g a l sta tu s, S E G xxvi 1306.25 f.,? D ie Inschriften von Ilion 25.86 (M ich el 524
C 1). It is n ot u sed o f e .g . dirty clothes.
4 M iasma
tion th a t h as som e, a n d usually all, o f the following characterist­
ics: it m akes the person affected ritu ally im pure, a n d thus unfit
to e n te r a tem ple: it is contagious: it is dangerous, an d this
d a n g e r is no t o f fam iliar secular origin. T w o typical sources o f
su c h a co n d itio n are co n tact w ith a corpse, o r a m urderer; a
p o llu te d re p u ta tio n , on the o th er han d , does not qualify on any
o f th e th re e counts. A specialization ra th e r like th a t o f miasma
ca n b e seen in its opposite, katharmos, w hich tends to be re­
s tric te d to a lim ited category ofcleansings. W hile kathaiw would
be th e n o rm a l v erb to use for w ashing a w ound, it w ould be odd
(th o u g h p o ssib le ) 11 to speak o f the process as a katharmos·, the
sacrifice, o n th e o th e r h an d , th a t rem edies the desecration of a
sa c re d grove is so d esc rib ed .12 T h is is not, therefore, a book
a b o u t G reek ideas o f d irt a n d defilem ent in general —a good
co m p re h en siv e tre a tm e n t o f th a t them e already exists 13 —bu t
a b o u t c e rta in dan g ero u s conditions to w hich the m etaphor of
d efile m e n t is often applied.
T h is m ean s th a t a large am o u n t o f evidence on w hat is merely
‘d isg u stin g ’ is excluded. D isgusting things in English sta rt w ith
th e p h ysically re p u g n a n t b u t include w hat is m orally o u trag e­
ous; in d e e d it is no t clear th a t a thing can be strongly disap ­
p ro v e d o f w ith o u t becom ing ‘disg u stin g ’. T h is is even m ore true
in G reek, in w hich miaros a n d its near synonym bdeluros are
a m o n g th e com m on est a n d strongest term s o f abuse. T hey can
b e a p p lie d to d irty h ab its —belching a t som eone, for in stan ce14—
b u t only in a m inority o f th eir uses do they concern w h at is felt
to be re p u g n a n t physically. O ften it is im possible to give them a
p re c ise r sense th a n ‘v illain’ or, m ore playfully, ‘rogue’. T h e
essence o f disgustin gness, how ever, seems to be deficiency in
11 PI. Soph. 226d —scarcely, in context, evidence for colloquial use.
12 S oph. OC 466.
13 M o u lin ier, passim. H is book contains an enorm ous am ount o f helpful lexicographi­
cal m ate ria l. I t is less strong on the subjects 1 shall attem p t to tackle, cf. V ernant,
Société, 121—40, H . Je a n m a ire , R H R 145 (1954), 99—104. It will be clear th at my usage
h a s n o th in g in com m on w ith th a t o f J . M . Redfield, Nature and Culture in the Iliad,
C h ica g o , 1975, 161 f. an d C h . 5, passim. Redfield here offers a brilliant analysis o f the
H o m e ric sense o f ord e r, b u t his use o f the language o f purification and pollution to
d esc rib e it h as little justificatio n in eith er G reek o r English usage.
14 e.g. A r. Vesp. 914, 1151.
Introduction 5
sham e; ‘d isg u stin g a n d sh am ele ss/b ra zen /b o ld ’ are co n stan t
c o n ju n c tio n s . 15 T h is b ad boldness can be directed ag ain st one­
self o r a g a in st o th ers; thus T im arc h u s, who p ro stitu ted him ­
self ,16 a n d M eid ias, w ho w an to n ly attack ed D em osthenes, are
b o th alike rev ealin g the miaria o f th eir n a tu re s .17 T ra ito rs and
law -b reak ers are miaroi, because it is sham elessness th a t causes
them to d isreg a rd n o rm al co n stra in ts . 18 T h e miaros is an
a n im a l , 19 lack in g th e self-control th a t is the first requisite o f life
in society. T h e criteria ju s t o u tlin ed , however, exclude beastli­
ness o f this k ind from the category o f pollution. N orm al people
try to avoid su ch miaroi, b u t to p re v en t them playing som e foul
trick, n o t from f e a ro f co n tam in atio n . I f one did try to include all
‘d isg u stin g ’ beh av io u r, pollution w ould becom e a category of
ala rm in g a n d p erh ap s vacuous com prehensiveness, since it
does n o t seem th a t in G reek term s disgustingness clings m erely
to a re stric ted set o f deviations; any outrageous act m akes its
p e rp e tra to r, view ed in a certain light, miaros.
11 is n a tu ra l to associate w ith miasma a few o th er w ords whose I
p rim ary reference is to d irt b u t w hich are also used in connec- ;
tion w ith co n tag io u s religious d a n g e r .20 M uch m ore im p o rtan t '
is th e n o u n agos, w ith its adjective enagês, ‘in agos’. Agos, too, it is
n a tu ra l to asso ciate w ith miasma, a n d no one w ho has discusssed
the su b je c t o f G reek pollution seem s ever to have th o u g h t o f
do in g o therw ise; the condition o f agos has all the three
c h a ra c te ristic s o f miasma m entioned above, a n d can actually be
referred to as miasma 21 (although the relation is not fully re­
cip ro cal). I t sh o u ld be em phasized, how ever, th a t the p artial
15 e.g. A r. Eq. 304, Pax 182—4, Ran. 465 f., X en. Hell. 7.3.6, Dem . 8.68, 19.17, (cf.
μιαρώ ς, ‘sh am elessly ’, 21.69); a t least 15 fu rth er instances occur. T h eo p h r. Char. 11.1.
defines β ό ελνρία , if the M SS are rig h t (D iels changed παιδιά to άναίδεια), as 'b latan t
an d rep re h en sib le je s tin g ’; this is the th in g in its m ildest form.
16 See A eschin. 1, passim. T h e link w ith πορνεία e.g. 54,88, 192, D em . 19.287. Sexual
p e rv e rsio r, .Arschin. 1.70; sexual violence, Dem . 19.309. H aving an erection, Ar. Lys.
989; m ak in g ad v an ces, Ar. Plut. 1069. βόελνττομαι as pro p er response to the sexually
im p u re, A r. Eq. 1288.
17 C o n n ectio n w ith άσέλγεια an d νβρις: e.g. Dem . 21.98,123,143; 47.81; Isae. 5.11.
18 T reac h ery : e.g. A r. Ach. 182, Eq. 239; D inarch. 3.18. Lawlessness: Andoc. 1.122,
Isae. 8.42, D em . 25.27, 35.26.
19 D em . 25.58, 43.83, 45.70 (άγριος), 58.49, D inarch. 1.50.
20 P rin cip ally μνσος, κηλίς, λϋμα, χραίνω.
21 Cf. A esch. Supp. 375 w ith 366, 619. F o r the respective consequences offlgoi and
miasma cf. A eschin. 3.111 an d Soph. Ο Γ 2 5 -3 0 .
6 M iasma
o v erla p b etw een the two concepts is perhaps d u e to a converg­
ence ra th e r th a n to a real sim ilarity in origin. O n this view they
w o u ld be tw o d istin c t form s o f contagious religious danger,
c re a te d by different acts a n d to som e extent conceived in diffe­
re n t w ays, w h ich w ere assim ilated because for the outsider their
p ra c tic a l consequences w ere the sam e. In sem antic origin, cer­
tain ly , th e re seem s to be no resem blance betw een the two term s.
T h e etym ology o f agos is controversial, b u t it now seems m ost
p ro b a b le th a t a n c ie n t scholars w ere right to connect it with
th e hag*- ro o t (hagnos, hagios) w hose sense is ‘to be revered,
s a c re d ’.22 E ven if the etym ological connection is fallacious, it is
p lau sib le th a t G reeks o f the classical period im agined it to exist,
a n d c e rta in th a t they did no t reg ard ‘p o llu tio n ’ as the basic
sense o f agos. T h e decisive text is the passage in the Oedipus
Tyrannus w here th e chorus refer to C reon, w ho has invoked upon
h im se lf a te rrib le curse should he prove guilty, as enagês,23 T here
is n o th in g p o llu ted ab o u t C reon, b u t he is ‘sacred ’ in the sense
th a t he h as su rre n d e re d him self conditionally to the gods who
w ill p u n ish him if his o a th proves false. (In L atin too an o ath is a
form o f self-consecration.24) I f it does, he will becom e enagês not
c o n d itio n ally b u t absolutely, a n d for outsiders, who will avoid
him for fear o f sh arin g his p u n ish m en t, his ‘sacredness’ will
a m o u n t to pollution. In different language we find the sam e
co n c ep tio n o f perilous consecration in an expression like ‘I have
freed m y self from Zeus o f xenoi to m ean ‘I have satisfied my
o b lig a tio n s as a xenos’,25 o r in the E um enides’ claim in
A eschylus th a t O restes, w ho has offended against them , is
‘c o n se c ra te d ’ to th e m .26 I t m ay even be th a t in Elis in the sixth
22 Sec P. C h a n tra in e an d O . M asson in Sprachgeschichte und Wortbedeutung, Festschrift A.
Debrunner, B ern, 1954, 8 5 -1 0 7 , for a valuable full discussion; cf. V ernant, Société.
1 3 4 -4 0 . B u rk ert, G R 405 denies the connection but adm its sem antic interference
b etw een th e tw o stem s. F or the rival d erivation from Sanskrit ‘a g a s \ sacrilege, sec e.g.
W illig er, 19 ff.; its p ro p o n en ts are req u ired , inter alia, to separate εναγής from εναγής,
a n d έναγίζω from its synonym s άγίζω an d καθαγίζω, and can m ake nothing οί αγος =
piaculum in S oph. A nt. 775 an d fr. 689 (puzzling adm ittedly on any view). For a difficulty
cf. p. 7 n. 31, below.
23 656, cf. 647,653. καθιερόω in a sim ilar context, Dem . 49.66.
24 Caput votis obligare, H or. Carm. ‘2 .8.6; exsecrari, cf. Fugier, 235.
25 PI. Ep. 7, 329b, cf. E ur. Flee. 345 πέψενγας τΰν έμάν ΊκέσιονΔ ία.
26 Eum. 304 καθιερωμένος; O restes in 451 protests th at he is now deconsecrated.
Introduction 7
c e n tu ry a d ecree o f outlaw ry could be expressed in the form ‘L et
him go aw ay to Z e u s .’27
In sev eral o th e r passages agos ap p ears as th e sanction in a
curse. T h e ‘P lataea o a th ’ specifies ‘L et there be agos for those
w ho h av e sw orn should they transgress th eir o a th .’28 W hen
H e ro d o tu s says th a t anyone w ho o b stru cts a p a rtic u la r right of
the S p a rta n kings is ‘held in the agos’, he is p ro b ab ly referring to
a p u b lic cu rse regularly p ro n o u n ced ag ain st offenders o f this
kind; ‘h eld in th e agos’ is closely p arallel to the com m on expres­
sion ‘held in a /th e curse ’.29 T h e offender, it seem s, is subjected
to a p erilo u s consecration. T h e arch aic R om an in stitu tio n o f the
leges sacratae has often been co m p ared , by w hich crim inals were
d e c la re d ‘sa c re d ’ to the god they h ad offended .30 T h ey were
co n se cra te d only in th e sense th a t they w ere m ade over for
p u n ish m e n t; from th e p o in t o f view of h u m an society they
becam e o u tcasts, to be killed w ith im p u n ity . Agos and enagês are
often co n stru c te d w ith a g o d ’s n am e in the genitive ,31 which
seem s to c o rresp o n d to the d ativ e o f the leges sacratae indicating
th e god to w hom th e offender is m ad e over. T his genitive is not
found w ith w ords like miasma w hose basic sense is defilem ent.
W h e re agos is n o t th e san ctio n to a curse b u t occurs as a result of
sacrilege, th e co n secratio n p resu m ab ly occurs spo n tan eo u sly .32
T h e p arallel is ev id en t betw een the offender who, becom ing
enagês, is a b s tra c te d from h u m a n society an d consigned to the
27 S O D I 1153 ( = M ichel 194, Schw yzer 415, Buck 63), as in terp reted by L atte, H R
6 2 - 4 . O n th e co n secratio n o f the dekatos in th e C yrene cath artic inscription see
A p p en d ix 2.
28 P. S iew ert, Der E id von Plalaiai, M un ich , 1972, p. 7 lines 50 f. F or enagês in a curse
see A eschin. 3.110, 12 1; agos? = curse, Soph. Anl. 256.
29 H d t. (5.56, cf. S o lm sen/F raenkel4 52.9, 54 A 20 f., M /L 30 B 34, PI. Leg. 881d,
Polyb. 12.6b.9.
30 B ib lio g rap h y in O g ilvie’s notes on Livy 3 .5 5 .5 -7 ; ad d Fugier, 236 IT. See too RAC
s.v. Anathema, on a co m p a ra b le Ju d a e o -C h ristia n institution.
31 (A esch.) Sept. 1017, T h u c. 1.126.2, 128.1—2, Aeschin. 3.110. It m u st be adm itted
th a t the c o n stru ctio n o f the noun agos often presents a problem on the C h a n train e/
M asso n th eo ry . W hile one expects the offender to be ‘in the agos’ (H d t. 6.56, and
enagês), in fact h e h im self is often virtually eq uivalent to the agos (Soph. O T 1426, T huc.
locc. cit., H d t. 5.72.1, A rist. Ath. Pol. 20.2), or an agos ‘happens to’ him (Aesch. Supp.
376, H d t. 6.91.1, P la ta e a o ath ). A esch. Eum. 167 has an άγος αιμάτων. C h a n train e/
M asson w ould p resu m ab ly have to explain these usages as an assim ilation in construc­
tion to miasma.
32 C f. D ü rk h eim , 320, ‘E very profanation im plies a consecration, b u t one which is
d read fu l, bo th for th e subject consecrated a n d for those who ap p ro ach h im .’
8 M iasma
gods for p u n ish m e n t, an d the anim al w hich, in the form of
sacrifice know n as enagizein, is b u rn t whole for a god w ith no
sh a re left for th e h u m a n w orshippers.
D iverse th o u g h they are in origin, miasma an d agos do, as we
h a v e n o ted , o verlap in usage. Every agos is probably also a
miasma, a n d agos is often co n stru cted as though it m eant pollu­
tion ra th e r th a n som ething like ‘avenging divine pow er ’.33 It
seem s, how ever, th a t, a t least in the early period, not all
miasmata a re age. W hile certain unavoidable physical conditions
a re miasmata, agos is a p ro d u c t o f avoidable even if involuntary
tran sg ressio n . A corpse, for instance, diffuses miasma, butagiw is
only c re a te d if a survivor denies it the divinely sanctioned right
o f b u ria l .34 T o c reate agos, the offence m ust pro b ab ly be directed
a g a in st th e gods o r th eir rules, as sim ple m u rd er seems not to do
so, w hile m u rd e r a t an a lta r certainly does .35 It som etim es
seem s as if w h a t causes agos is sim ply contact betw een miasma
a n d th e sacred . T h e historical instances recorded in H erodotus
a n d T h u c y d id e s all take the form o f killing in violation of
s a n c tu a ry . V io latio n of san c tu a ry w ithout bloodshed, however,
is sp oken o f as agos in A eschylus ,36 an d there is no connection
b etw een th e agos invoked in a n o ath or curse a n d polluting
o bjects. T o defile a sacred place by introducing miasma is one
w ay, b u t one w ay only, o f in cu rrin g th at perilous consecration
w h ich seem s ce n tral to the idea o f agos.
I t w as suggested earlier th a t miasma an d agos are p erhaps in
orig in tw o theoretically d istin c t forms of com m unicable religi­
ous d an g e r. T h e m ost im p o rta n t difference concerns the rela­
tion o f th e tw o to the gods. T o miasma gods seem irrelevant; it is a
d a n g e ro u s d irtin ess th a t individuals ru b off on one an o th er like
a p h y sical tain t. Agos by co n tra st has its source in a sacrilegious
33 See p. 7 n. 31.
34 S oph. Ant. 256 (w here, how ever, the m eaning ‘curse’ also seem s possible, cf. the
sc h o l.’s reference a d loc. to Bouzygean curses), (A esch.) Sept. 1017.
35 H d t. 6.91.1, T h u c . 1.126.2, 128.1—2.1 f a m u rd erer enters a sacred place, the place
in cu rs agos, A esch. Eum. 167. F or agos used o f plain m urder, however, see Aesch. Cho.
635 (a com pelling conjecture), Soph. O T 1426: in both cases the extrem e horror o f the
d eed p e rh a p s evokes th e m ore charged w ord. From the 4th c. agos an d enagës become
v irtu a l sy n o n y m s o f miasma an d miaros: T h eo p h r. ap. Porph. Abst. 2.29, p. 159.12 N.,
A le x a n d e r ap. D iod. 18.8.4, Polem on ap. M acrob. Sat. 5.19.26, Ap. Rhod. 3.203, 4.478,
A n th .P a l. 14.74.2.
36 A esch. Supp. 375.
Introduction 9
act, a n d th e enagës, as th e a tta c h e d genitive suggests, is in the
g rip o f a n av en g in g pow er; th e reason for avoiding him is not
Ica r o f co n ta m in a tio n b u t to escap e being engulfed in th e divine
p u n ish m e n t th a t aw aits him . W e en co u n ter h ere a crucial
a m b ig u ity in w h a t is u n d ersto o d by the term pollution. W hile
som e Scholars th in k o f it as th e im personal tain t, analogous to
d irt o r a n infectious disease, o th ers reg ard shared d a n g e r rath er
I lian th e m e ta p h o r o f co n tam in atio n as the essential. T h is could
be re p h ra s e d to say th a t one g ro u p confines pollution to miasma
i ii th e stric t sense, w hile th e o th er also includes agos. It w ould be
possible in th ese term s to offer a com prom ise solution to the
n oto rio u s p ro b lem o f pollution in H om er; while miasma cannot
he sh o w n to be p re sen t in him , agos (not the w ord, b u t the
exp erien ce) h e u n d en iab ly recognizes .37
Agos seem s, in fact, to p rovide m iddle g ro u n d betw een two
sou rces o f religious d a n g e r th a t are som etim es supposed quite
d istin c t, a n d , in som e cultures, m ay actu ally be so —on th e one
Ilan d , im p erso n al pollution, a n d on the other, the an g er o f a
perso n al deity . O n e reason for d ra w in g this d istin ctio n m ay be
the feeling th a t belief in divine an g er is prim itive b u t com ­
p re h en sib le, w hile belief in po llu tio n is wholly irratio n al. But,
even th o u g h p o llu tio n m ay o p erate ‘w ith the sam e ruthless
indifference to m otive as a typhoid g erm ’,38 divine anger is not
alw ays m e re discrim in atin g . T h e re is no p oint in avoiding
pollu ted sh ip m ates before p u ttin g to sea, only to fall in w ith
o th ers th e gods are an g ry w ith; you will finish a t th e bottom ju st
I lie sa m e .39 T h is is w hy the enagës, who is consecrated to an
av en g in g god, becom es in p ractical term s polluted. Divine
an g er, agos, a n d miasma can becom e inextricably intertw ined. In
A esch y lu s’ Supplices, we find th e consequence o f disregarded
su p p lic a tio n ex p ressed ,40 som etim es in close ju x tap o sitio n , as
miasma, agos, th e ‘w ra th o f Z eus o f S u p p lia n ts’, the hostility of
I he ‘all-d e stru ctiv e go d ’ from w hom even the d ead are not free,
37 L loyd-Jones, 74 f.
38 Dod,ds, 36.
39 S hipw reck is caused by, in general term s, injustice (Horn. Od. 3.133), im piety
(A esch. Sept. 6 0 2 - 4 ), p ollution (A nt. 5. 82, E ur. El. ! 350). Specific causes arc sacrilege
(A jax a n d C a ssa n d ra , O d y sseu s’ com panions and cattle o f sun), perjury (Kur. El.
I 355), a n d b lo o d -g u ilt (A nt. 5.82). For m uch fu rth er evidence see W ach sm u th , 265—
71.
40 See 366, 375 f., 385, 4 1 4 -1 6 , 478, an d for ju x tap o sitio n 6 1 6 -2 0 .
10 M iasma
a n d th e p resen ce o f an avenging Z eus perched on the roof-top,
p e rh a p s p o llu tin g 4’ a n d certain ly dam aging the house. T h e
sam e convergence can be seen on the ritu al level. In theory
sacrifice a n d purification m ay seem to be d istin ct operations,
th e o n e in te n d e d to ap p ease a deity a n d the o th er to efface an
im p e rso n a l p ollution. In p ractice, w h at is spoken o f as a p u ri­
fication often takes the form o f a sacrifice ,42 w hile the effects of
d iv in e an g er, a t least w hen it m anifests itself as a disease, can
so m etim es be w ashed aw ay .43
As a resu lt, it becom es ex trao rd in arily h ard to draw lines of
d e m a rc a tio n betw een pollution a n d the consequences o f divine
a n g e r. R eligious d a n g e r is alm ost alw ays potentially com m unal
in G reece; a p u n ish m e n t th a t is confined to the guilty parties
deserv es special co m m en t .44 I t m ay be m ore n atu ra l to envisage
th e m u rd e re r, for instance, as end an g erin g his associates, and
th e p e rju re r his d escen d an ts, b u t the o th er form o f co n tam in a­
tio n is possible in b o th cases .45 I f the consequences of different
offences h a d once been m ore distinct, they h ad been assim ilated
by th e d a te o f m ost o f o u r sources to an extent th a t m akes them
now alm o st inextricable. T h e re is no question o f form al ‘p u ri­
fic a tio n ’ from th e consequences o f perju ry ,46 b u t nor is there
from tem p le-ro b b in g , a n d the tem ple-robber is certainly
enagës,47 It is very h ard , therefore, to sep arate from pollution any
situ a tio n w h ere b re ach o f a religious rule has created danger. As
th e c h a p te r on sacrilege will show , quite m inor violations o f
sa n c tity a re ‘p o llu tio n s’ bo th in the sense o f causing com m unic­
a b le d a n g e r, a n d in th a t o f re q u irin g ‘purification’. T o unite all
th ese situ a tio n s in an undifferentiated category o f pollution
41 So th e M S in 650, b u t see P age's critical note.
42 H d t. 6.91.1, S oph. O C 466—92, E ur. H F 9 2 2 fT., w ith M o u lin ie r,88, R u d h ard t, 270,
P aus. 1.34.5 εστι ό ί κα θά ρσιον. . . θύειν, below, p. 209 on Epim enides. Such cathartic
sacrifice could b e d en o ted by a distinctive use o f έκθνομαι, H dt. loc. cit., Eur. fr. 912.12
(w ith ob ject o f the god ap p eased ), T h eo p h r. Char. 16.6 (a certain conjecture). Cf.
J . C a s a b o n a , Recherches sur le vocabulaire des sacrifices en grec, A ix-en-Provence, 1966. 96 f ;
th e re w ere ίεροποιοι έπι τα έκθύματα in A thens, Arist. Ath. Pol. 54.6, and for έκθνσίαι in
D elos see B ru n eau , 286—8. An εκθνμα m ay well have differed from the characteristic
form ol O ly m p ia n sacrifice, b ut the point rem ains th at this is a deistic rite.
43 S ec C h . 7.
44 D iod. 15.49.6, P lut. Timol. 3 0 .7 -9 .
45 e.g. E ur. El. 1355, Hipp. 1379.
4‘ B ut lor th e inform al possibility see O v . Fast. 5.681 f , cited .4«/.«. Chr. 6(1950), 73.
47 e.g. D iod. 16.60.1.
Introduction 11
would be u n accep tab le; th e lan g u ag e o f defilem ent is m ore
n a tu ra l in so m e cases th a n in oth ers, an d will n o t alw ays have
Ilie sam e im p licatio n s. I t seem s b etter, however, to o p erate w ith
.i co n cep t th a t is flexible a n d , a t its edges, ill-defined, th a n to
im pose a d e m a rc a tio n by force. C are will have to be taken in
d e te rm in in g w h a t, in a p a rtic u la r case, the label ‘pollution’
•ictu ally m ean s.
The p a rtia l convergence b etw een miasma a n d a w ord whose
etym ological connections seem to be w ith sacredness does not
M ipport th e often re p eated p arad o x th a t ‘the sacred is a t once
“ s a c re d ” a n d “ d efiled ” ’, because ‘in the savage m ind th e ideas
o f holiness a n d pollution are not yet d ifferentiated ’.48 These
<laim s are based on the Polynesian taboo, w hich does in fact
u n ite sacred th in g s a n d ritu al im p u rity w ithin th e single cate-
g( >ry o f p ro h ib itio n ;49 no w ord is available to in d icate the special
sta tu s o f th e one, it is said, w hich is n o t also ap p licab le to the
o th er. I t h as com e to be recognized, how ever, th a t taboo is
.i specialized p h en o m en o n q u ite u n su itab le for the indis­
c rim in ate in te rn a tio n a l ap p licatio n th a t it has often received .50
( lertainly, in G reek as in o th e r religions, there is a sim ilarity
betw een sacred a n d im p u re objects in th a t b o th are subject to
re stric tio n s .51 In different contexts th e one adjective hosios, in its
sense o f ‘safely av ailab le for profane u se’, can in d icate freedom
from e ith e r co n secratio n o r p o llu tio n .52 T h ere is, how ever, no
difficulty in G reek in distin g u ish in g betw een th e source of
re stric tio n in th e tw o cases; a sacred law spells o u t w ith welcome
clarity th e th ree estates o f th e w orld as ‘sacred, profane, and
p o llu te d ’.53 A G reek w ould be puzzled by the suggestion th at
t here is a n y th in g im p u re ab o u t the sacred, or vice versa. As we
hav e seen, im p u re things are miasmata, not age. T h e sacred is
in d eed co n tag io u s, in the sense th a t the offender falls in to the
pow er o f th e offended god; b u t, alth o u g h he becom es d a n g e r­
ous, a n d th u s ‘p o llu te d ’, for th e o u tsider, it is n o t pollution
48 M . E liad e a n d J . G. Frazer, cited by D ouglas, 18,20.
49 S teip er, 33—6.
50 S tein er, passim, esp. 35.
51 C f. T . O . B eidelm an, \V. Robertson Smith and the Sociological Study o f Religion, C hicago,
1974, 62 f.: ‘he saw th a t th e concepts o f holiness an d pollution both depended upon
restrictiv e rules o f avoidance, an d th a t form ally these prohibitions w ere alike.’
52 See A p p en d ix 1.
53 L S S 115 A 9 f.
12 M iasma
in th e god b u t avenging pow er th a t m akes him so. In Latin,
even th e lim ited connection betw een ‘sacred ’ a n d ‘accursed’
c o n ta in e d in th e use o f sacer in the leges sacratae cam e to be
p u z z lin g ;54 sim ilarly in G reek, if the etym ological link of agos
w ith hag- is correct, differentiation occurred early, through the
loss o f th e a sp ira te , betw een beneficial an d destructive forms of
c o n se c ra tio n .55
T h e tw o w ords th a t have been given such prom inence so far
a re n o t especially com m on. W hole literary genres can be found
from w h ich one o r the o th er is virtually absent. T h e verb miainö
is m o re often found in relatio n to pollution th an the noun
miasma, a n d th e com m on w ay o f saying ‘p o llu te d ’ is sim ply ‘not
c le a n ’ (katharos). O ften the language used in relevant contexts is
th a t o f hosiä, w h a t is religiously safe, ra th e r th a n specifically th at
o f p u rity . I t is as a focusing device th a t the w ords agos and
miasma h av e h ere been given alm ost em blem atic significance.
B u t alo n g sid e them can be set an o th er distinctively religious
w o rd w h ich by c o n tra st is very com m on. Hagnos m eans ‘p u re’,
b u t h as no etym ological connection w ith physical cleanliness .56
T h u s , th o u g h p ollution is elusive, p urity stands forth palpably;
a n d th e clear reality o f its opposite helps to justify the kind of
defin itio n o f p o llu tio n th a t we have adopted.
T h e deficiencies in the evidence for o u r know ledge o f Greek
p o llu tio n b elief will repeatedly be ap p a ren t. T h e gravest conse­
q u e n c e o f th e sta te o f the evidence is th a t it has been im possible
to c e n tre this stu d y on a p a rtic u la r tim e and place. A historical
a n d g eo g rap h ical synthesis becom es inevitable because no one
s ta te offers a corpus o f co n tem p o rary d ocum ents, hom ogeneous
o r no t, sufficiently dense to form a basis for a synchronic local
stu d y . E v en fifth-century A thens, for w hich literary evidence
is c o m p arativ ely a b u n d a n t, offers little forensic oratory, few
a c c o u n ts o f re le v an t behaviour, a n d alm ost noth in g by way of
ex p licit codes o f rules. T h e perils o f such a synthesis are obvi­
ous. P lace m atters; the G reeks w ere not a hom ogeneous cultural
m ass, a n d d id n o t see them selves so. Several historical instances
show th a t th e S p artan s, for instance, were ready to respect
54 See M a c ro b . Sal. 3.7.5; cf. YV. VVarde Fow ler, Roman Essays and Interpretations,
O x fo rd , 1920, 16 f.
55 B u t on th e an c ie n t lexicographers’ belief in am biguous w ords see A ppendix 1.
56 See p p. 147 if.
Introduction 13
religious claim s (such as th a t o f the festival calen d ar) even to
tlieir ow n obvious d etrim en t. Across the b o u n d ary in Argos,
how ever, lived m en fertile in legalistic devices for evading religi­
ous o b lig atio n s, a n d q u ick to exploit to th eir ow n ends their
n e ig h b o u rs’ p ie ty .57 T im e m atters even more; the passage from
the w orld th a t seem s to be im plied by H om er, w here the effects
o f literacy are scarcely perceived, a n d the com m unity has few
claim s as a g a in st the in d iv id u al household, to the society whose
a d m in istra tiv e com plexities are laid o u t in A risto tle’s Constitu­
tion o f the Athenians, did not leave religious values unaffected.
It is ob v io u s, for instance, th a t the im plications o f a belief like
I h at in collective p u n ish m en t change drastically w hen the col­
lectivity th re a te n e d is no longer a village com m unity, where
ev ery b o d y could know everyone else, b u t the city o f A thens,
co n v en tio n ally reckoned by the G reeks as co n tain in g 30,000
citizens. T h e a c tu a l range o f regional an d tem poral variation in
pollu tio n beliefs will certain ly not have been ca p tu red in this
book, b u t th e effort has been m ad e to presen t as differentiated a
I»icture as th e evidence perm its.
A n o th e r d elicate synthesis is th a t betw een different classes of
ev id en c e .58 T h e n o u n miasma, u b iq u ito u s in the tragedians, does
not o ccu r a t all in H ero d o tu s, T h u cy d id es, or X enophon. T his
m ig h t be tak en to prove th a t the w o rd ’s stylistic level is too high
lor p rose, th a t th e concerns o f tragedy are unreal, or sim ply th at
trag e d y a n d histo ry trea t different areas of experience. T he
sta tu s o f im ag in ativ e lite ratu re as evidence is, in fact, a p artic u ­
lar p ro b lem . M o d ern social histo rian s view such evidence with
suspicion; co u rt records, not extrapolations from Shakespeare,
form th e b ack b o n e o f a classic m odern study o f English po p u lar
relig io n .59 L itera ry texts can o n ly be safely exploited, it m ight be
arg u ed , to illu stra te the h u m an im plications o f beliefs and
a ttitu d e s th e existence an d significance o f w hich can be inde­
p en d e n tly estab lish ed . C lassical scholars, whose knowledge of
su b jects like po llu tio n derives largely from th eir reading of
trag ed y , h av e ten d ed to be less cautious, p artly because a lte rn a ­
tive sou rces o f in form ation on these subjects are h ard to find.
57 S ee p p . 154 f.
58 Cf. D over, 8—33; excellent rem arks on the relevance o f literary purpose in Lloyd-
Jones, 76 f.
59 K. T h o m a s, Religion and the Decline ojM agic, L ondon, 1971.
14 Miasma
T h e re a re m ost delicate problem s here, an d some differentia­
tion is re q u ire d . W orks o f a rt m ay convey incidental factual
in fo rm atio n th a t need no t be m istrusted. W e learn from
S h ak esp e are th a t the cold m aids o f E ngland give the nam e o f
‘d e a d m e n ’s fingers’ to certain flowers, and tragedy contains a
good d ea l o f reliable inform ation ab o u t m atters o f cult. W here,
on th e o th e r h a n d , a religious m o tif has an obvious function in
th e m ech an ics o f the plot, like the ghost in Hamlet or the plague
in th e Oedipus Tyrannus, no m ore can be inferred th an th at the
m o tif w as co m p reh en sib le to the audience; n eith er audience nor
p la y w rig h t n eed be com m itted to belief in the phenom enon.
R eligious them es th a t are em b ed d ed in the outlook o f a work are
h a rd e r to assess. ‘T h in k in g m ore than m ortal th o u g h ts’, ‘envy
o f the g o d s’, ‘the fam ily cu rse’: these, we believe, are prim e
religious d an g e rs in the ‘arch aic w orld view’. Yet it is chiefly
th ro u g h lite ra tu re th a t we h e a r o f them ; we scarcely know w hat
th e ir co rrelates m ay have been in the everyday experience o f the
fifth -c en tu ry A th en ian w ho w as so often invited to reflect on
th em in th e trag ic theatre.
T h e s p e c ta to r o f tragedy was also the spectator of comedy,
a n d it is in stru ctiv e to com pare the world-views of the two
genres. D ivine ju stic e m ay in tragedy often be an obscure ideal
in hopes o f w hich m en grope in pain; the gods o f O ld C om edy
a re d e c e n t sorts, w ho do th eir best to keep erring hum ans on the
rig h t tra c k .60 T h ey p u n ish w rongdoers ,61 b u t theirs is no savage
ju stic e ; unlike the D ionysus o f E uripides, they can forgive a
te m p o ra ry a b e rra tio n , a n d a h um ble plea for p ard o n will not
find th em a d a m a n t .62 T h is is a w orld against w hich no gran ­
diose re se n tm e n t is possible, a n d those w ho display it are ex­
plicitly m ark e d by th eir language as in tru d ers from the tragic
stage: Ό savage god, o destiny . . Λ 63 C om edy’s nearest equiva­
len t is undig n ified grum bling: ‘Som e god’s got the house into a
p ro p e r m ess .’64 Erinyes are know n from literatu re, not life,65
60 A r. S u b . 5 8 7 - 9 . T h is is the sta n d a rd civic view, cf. Solon, fr. 4 (an influential text),
fr. 1 1 .1 -2 , D em . 1.10,2.1, 18.153, 195, 19.254-6, Aeschin. 3.57, 130; w ith ‘luck'instead
o f g ods, D eni. 1.1, 4.12.
Ar. Nub. 3 9 5 -7 , 1458-62, Thesm. 6 6 8 -8 5 , Ran. 148-50.
62 A r. Nub. 1 478-80, /V v 6 6 8 f., cf. Vesp. 1001 f.; contrast E ur. Bacch. 13+4-9.
63 A r. S u b . 1264, cf. Pax 1250, Thesm. 1047, Ran. 310. 64 Ar. Vesp. 1474 f.
65 A r. Plut. 423, cf. Aeschin. 1.190, T im aeu s 566 FGrH Ir. 55; underw orld punish­
m en ts too are know n from art, Dem . 25.52.
Introduction 15
a n d th e big w ords w ith w hich trag ed y speaks o f crim e a n d
p u n ish m e n t a c q u ire here m ore hom ely m eanings .66 In a fourth-
( c n tu ry com edy, th e ‘curse o f th e P elo p id ae’, typical tragic
ι hem e, will be a conception to lau g h at, a n d th e only alastores,
polluting d em o n s, who are recognized will be m en -
philosophers, for in stan ce, w ho too little ap p reciate the value of
pleasu re .67 W h ile in high lite ra tu re th e seer is alw ays rig h t ,68 in
com edy h e is alw ays w ro n g .69 N ot ju s t com edy saw a cleft
betw een trag e d y a n d th e fam iliar w orld. W hen, in the fourth
cen tu ry , an o ra to r m o u n ts u n accu sto m ed religious o r em otional
heights, his o p p o n e n t will d raw him back by an accusation of
Iragddia, o r s h a m .70 It is in co n tin u atio n o f this usage th a t the
h isto rian w ho detects u n real p a tte rn s o f divine vengeance in the
events h e reco rd s is accused o f com posing ‘trag ic’ h isto ry .71
’1’hese facts do not, o f course, expose the tragic w orld-view as
m ere m elo d ra m a . I f som e o f its religious preoccupations
•.crmed u n re al in th e fourth century, they m ay not have done in
ι lie fifth; a n d th e relation o f com edy to trag ed y is n o t th a t o f real
life to ‘lies o f p o ets’, b u t o f o ne p o lar extrem e to a n o th e r .72 W h at
ι loes em erg e is th e crucial influence o f a literary w ork’s genre in
d e te rm in in g th e religious em phasis it contains. C ertain ly there
also scope for large diversity w ith in a genre (Iliada n d Odyssey,
I lo m er a n d th e epic cycle, th e th ree trag ed ian s), b u t works o f
I lie sam e g en re, because they have a com m on su b ject-m atter
an d , in A risto telian term s, a com m on aim , are liable to focus on
sim ilar a reas o f experience a n d belief to the exclusion o f others.
II one co n sults w orks o f different genre for inform ation ab o u t
ideas o f fate, o r th e co n tin u in g influence of the d ead on h u m an
life, o ne receives answ ers th a t, if n o t contradictory, are a t least
notab ly different in em phasis. T h u s it is p articu larly dangerous
“ άλιτρία, villainy, A r. Ach. 907; άτηρός, im possibly troublesom e, Vesp. 1299;
/ ρ ι ννων άπορρώξ, a real old curm udgeon, Lys. 811.
47 X e n a rc h u s, fr. 1.3 ap. A th. 63 f; cf. B aton, fr. 2.5; M oulinier, 266.
** P o u ly d am as in th e Iliad, T eiresias in Soph. OT, Anl., E ur. Bacch.
4# Ar. Pax 1047 fi'., A t. 521, 959 fF., C ratin u s, fr. 57,62, Callias, fr. 14, Eupolis, fr. 211,
112, E p ic h arm u s, fr. 9, ‘A ristoxenus’ ap. H ephaest. 8.3.27 C o n sb ru ch (K aibel, CGF,
|. 87); cf. L loyd, 1 7 n .4 1 .
70 LSJ, s.vv. τραγψόία, τραγωδέcp.
71 O n P lu ta rc h ’s d ero g ato ry use o f Iragikcs see P. de Lacy, A JP 73 (1952), 159-71,
ill id, su m m arizin g the d e b a te on ‘tragic histo ry ’, F. W. W albank, Polybius, Berkeley,
1972, 3 4 - 9 , w ith references.
72 A rist. Poel. 1448b 24 IT.
 16 M iasma
to b ase hy p o th eses o f cu ltu ra l change on works o f different
c e n tu ries th a t belong to different genres. T h e belief th at pollu­
tio n fears settled over G reece like a cloud in the post-H om eric
p e rio d is largely based on a com parison betw een tw o separate
g en res, eig h th - or seventh-century epic a n d fifth-century
tra g e d y ; b u t th e p rom inence o f m u rd er pollution in tragedy is a
c o n se q u en ce o f its preferred subject-m atter, a n d the phenom e­
n o n th a t req u ires exp lan atio n is not an upsurge of pollution
fears b u t th e em ergence o f a genre th a t so extensively explored
th e co n seq u en ces o f violence w ithin the family. If we take a
fifth -c en tu ry g enre w hose aim s are closer to the epic, the choral
lyric o f P in d a r, we continue to find pollution fears as incon­
sp icu o u s as in H o m er, w hile it is possible to read right through
th e w orks o f X e n o p h o n a n d scarcely becom e aw are th a t such a
1 th in g exists. T h e evidence o f one genre needs to be controlled by
c o m p a riso n w ith th a t o f others, literary an d non-literary. W hen
th is is do n e, som e conceptions best know n from high literature -
g o d -se n t delusion, for in stan ce —do tu rn o u t to occur, in slightly
a lte re d guises, a t every level .73
T h e sp asm o d ic a p p e a ra n c e o f pollution in literatu re raises a
final a n d m ore general p o in t o f m ethod. I t is no t ju s t in literary
tex ts th a t p ollution fears surprise now by th eir presence, now
th e ir ab sence. In the late seventh century, the great A thenian
fam ily o f th e A lcm aeonids in cu rred a celebrated pollution .74
T h e y w ere tried a n d exiled, a n d the very bones o f the dead
m e m b e rs expelled. M ore th a n h a lf a century later, Peisistratus,
la te r th e ty ra n t, m arrie d an A lcm aeonid girl for political
re aso n s, b u t w as unw illing to beget children from the polluted
stock; w h en th e g irl’s father, M egacles, learn t o f the insult, a
crisis in th e ir alliance resulted. A t the end o f the sixth century,
th e S p a rta n s arriv ed in A thens to expel the A lcm aeonid
C leisth e n es a n d his followers; th eir m otives w ere political, bu t
th e p re te x t w as to ‘drive o u t the agos’. T h e tain t still clung to the
slig h tly A lcm aeonid Pericles, a n d was exploited against him by
S p a rta in 432. Y et the A lcm aeonids h ad been able to re-
73 F o r com ic ate see p. 14 n. 62; for history (H dt. aside), X en. Hell. 6.4.3, A i t . Anab.
2.7.3; for o ra to ry , (Lys.) 6.22,27,32, A eschin. 3.117,133, Dem . 9.54; ate as a
m ech an ism o f d ivine p u n ish m en t, A ndoc. 1.113, Lycurg. Leoc. 91—3, Dem. 2 4 .1'11.
74 See H d t. 1.61, 5 .6 2 - 3 .1 ,5 .7 0 -2 ,6 .1 2 6 - 3 0 , T huc. Γ. 126-7, A rist. Ath. Pol. 20.2; cf.
D avies 3 6 8 -8 5 .
Introduction 17
instate them selves in A thens n o t long after their trial; th e sam e
M egacles w hose d a u g h te r P eisistratu s sp u rn ed was him self
i hosen as so n -in-law by C leisthenes o f Sicyon, in preference to
Ilie finest m en o f Greece; at th e end o f th e cen tu ry , the great
lem ple a t D elp h i, hom e o f th e p u re Apollo, w as b u ilt w ith funds
provided by th e tain ted family. In this case, it seem s th at
pollu tio n h a d soon ceased to be an actu al source o f religious
anxiety, a n d becom e in stead a n in h erited disgrace, one factor
■iniong o th ers in th e general re p u ta tio n of the family, which
enem ies w ould d en o u n ce a n d friends ignore. T h e explanation
here is p rim arily political, b u t o th er factors too could cause
i on cern a b o u t p o llu tio n to a p p e a r spasm odic or unpredictable.
I he in d iv id u a l w as m ost sensitive to the ad d ed th rea t w hen he
was m ost en d a n g e re d in o th er ways; thus we find fear o f pollu-
lion, like m an y o th e r religious co n cerns ,75 a t its m ost intense in
con n ectio n w ith seafaring. T h e consequences m ight prove th at
.i p a rtic u la r act, m erely do u b tfu l in itself, was in fact polluting,
o r th a t a n agos w hich m en hoped had been ‘sacrificed o u t’ was
slill active. T h e G reeks expressed som ething like this through
I he m e ta p h o r o f a pollution th a t ‘sleeps’ and then ‘w akes u p ’.76
I )u rin g S o phocles’ Oedipus al Colonus, the chorus are h alf
I>ersuaded th a t th e ir initial revulsion w as w rong an d O ed ip u s is
not a th re a te n in g person; b u t, w hen a terrifying th u n d erc lap is
su d d en ly h e a rd , th ey at once suppose th at the consequences of
a sso ciatin g w ith th e polluted have cau g h t up w ith th em .77 T h e
A th en ian s seem to have observed th eir fortunes d u rin g the
P elo p o n n esian w ar in o rd e r to evaluate their policy o f religious
I ink erin g w ith Delos; if b attle w ent ill, the right form ula h ad not
yet been fo u n d .78 T h e im plications o f all this for m ethod are
easily seen. A n ac co u n t o f pollution beliefs will be sterile and
u n re a l u n less it considers the com plicated process by which
b elief is tra n s la te d into behaviour.
75 W ach sm u th , passim.
76 A esch. Eum. 280, H d t. 7.137.1.
71 1482-4.
78 T h u c. 5.32.1.

1 P urificatio n is one w ay in w hich th e m etaphysical can be
PURIFICATION:
A SCIENCE OF DIVISION
In th e longest e x ta n t analysis o f the activity called kalharmos,
P la to in th e Sophist presents it as a n aspect o f the ‘science of
d iv isio n ’. O f th e kind o f division th a t retains w h at is better but
expels th e w orse, I do know th e nam e . . . every division of th at
k in d is univ ersally know n as a p urification .’1 T h e purifications
en v isag ed here by P lato are purely physical - w ashing, carding
o f wool, a n d th e like — b u t such physical acts of division are
re ad ily a n d unconsciously exploited to create divisions o f a
d ilfere n t kind. W e see this from D ickens’s law yer M r Jaggers:
1 em brace this opportunity of remarking that he washed his clients
oil', as if he were a surgeon or a dentist. He had a closet in his room,
fitted up for the purpose, which smelt of scented soap, like a per­
fum er’s shop . . . When I and my friends repaired to him at six o’clock
next day, he seemed to have been engaged on a case of a darker
complexion than usual, for we found him with his head butted into
this closet, not only washing his hands, but laving his face and
gargling his th ro a t. . .
A nd a g a in , afte r a d in n e r p arty , ‘I found him in his dressing
room su rro u n d e d by his stock o f boots, already h ard at it,
w a sh in g his h a n d s o f u s .’2 M r Jag g e rs, therefore, a m an whose
im p erio u s control o f his en vironm ent D ickens em phasizes,
s e p a ra te d th e different areas o f his experience by elaborate rites
o f lu stra tio n . J a g g e rs is an extrem e case, b u t this is a fam iliar
form o f b eh a v io u r. Few o f the o rd in ary in d iv id u al’s daily p u ri­
fications, clu stered as they are aro u n d points o f transition (be­
fore a n d afte r bed, on re tu rn from work, a n d so on), are w ithout
som e sym bolic content. W h a t is rem arkable ab o u t Jaggers is
th e ease w ith w hich he could rem ove u n p leasa n t associations by
1 PI. Soph. 226d.
2 Greal Expectations, C h . 26.
Purification: a Science o f Division 19
ilic ap p lic a tio n o f soap. H is clerk W em m ick required a
d ra w b rid g e for th e sam e purpose.
m ad e p a lp a b le . A lth o u g h it can p erh ap s o p erate as a divider in
•i q u ite n e u tra l sense, it m ore n atu ra lly separates higher from
low er a n d b e tte r from worse. Its m ost obvious use o f this kind in
( Jreece is to m ark off sacred areas from profane. ‘W e ourselves
lix b o u n d a rie s to the san ctu aries a n d precincts o f the gods, so
Ih a t n o b o d y m ay cross th em unless he be pure; a n d w hen we
e n ter we sp rin k le ourselves, n o t as defiling ourselves thereby,
b u t to w ash aw ay any pollution we m ay alread y have con­
tra c te d .’3 T h e re is a b u n d a n t evidence from literatu re, vase
p ain tin g s, a n d excavation for these stoups o f lu stral w a te r sited
at th e e n tra n c e to san ctu aries, for th e purification o f those who
e n tered . In inventories, they a p p e a r as p a rt o f a tem p le’s nor­
m al fu rn ish in g ; H ero, in his Pneumatica, tells o f a m echanical
device th a t gave forth lustral w ater a t the d ro p o f a coin .4 I t is
very rev ealin g for G reek conceptions o f the sacred th a t in
A th en s th e agora, civic a n d political centre o f the city, was
m ark e d off by sim ilar lu stral stoups. W h eth er the norm al
A th e n ia n w ould actu ally have purified him self before entering
is n o t know n, b u t certainly this w as th e b arrier beyond w hich
those d ep riv e d o f civil rig h ts m ight not pass, on th rea t of
p ro se c u tio n .5 A kind o f ring o f p u rity excluded th e disgraced
from co m m u n al life.
Fixed lu stra l bowls are first attested aro u n d the end o f the
sev en th ce n tu ry , b u t the custom because o f w hich they w ere set
u p is a lre a d y em b ed d ed in H om er. W ith o u t purification there is
no access to th e sacred . ‘R espect forbids me to p o u r a libation to
/ e u s w ith u n w ash ed h a n d s ’, says H ector, an d we find H om eric
c h a ra c te rs n o t m erely w ashing th eir h an d s b u t b ath in g and
3 H ip p o c. Morb. Sacr. 148.55 ff. J ., 1.46G .
4 Cf. S IG 3 index s.v. περιρραντήριον; H ero, Spir. 21. Full treatm en t by Ginouvès,
2 2 9 -3 1 0 (m y d e b t to this learned an d com prehensive work is very large). For the
earliest perirrhanteria see J . D ucat, B C H 88 (1964), 577 -6 0 6 . O n th eir function cf.
L u cian , Sacr. 13, Pollux 1.8.
5 See G . E. M. d e Ste. C roix, The Origins o f the Peloponnesian War, L ondon, 1972,
A pp en d ix 43; an d R. M a rtin , Recherches sur l ’agora grecque, Paris, 1951, 164—201, on the
agora's religious significance.
 20 M iasma
ch a n g in g th e ir clothes as a p re p a ra tio n for p ra y er or sacrifice .6
Before a n y sacrifice, the p a rtic ip a n ts w ere u nited by a rite of
sym bolic w ashing. L u stral w a te r w as carried ro u n d in a special
bow l, a n d those taking p a rt w ashed their h an d s in it, o r a t least
sp rin k led them selves or w ere sp rin k led .7 O nce the ritu al space
a n d th e p a rtic ip a n ts h ad been m arked off in this way, the
sacrifice could proceed. M o st sanctuaries h ad a spring near at
h a n d , a hallow ed source o f m aterial for ‘purifications a n d lus­
tra l w a te r ’.8 R eally elab o rate ritu als o f p re p ara to ry w ashing are
n o t a tte s te d , b u t the m ore closely involved psychologically the
m o rta l w as in the cerem ony to be perform ed, the g reater and
m o re form al the prelim inary requirem ents becam e. T h u s be­
fore in c u b a tio n , initiation, m ysteries, an d prophecy the bath
w as a re g u la te d a n d cerem onial ev e n t .9 A bout civic sacredness
w e a re less well inform ed, b u t Peisetairos is p robably following
a t least a n occasional A th en ian practice w hen he calls for w ater,
to clean se his h a n d s, a n d a crow n, before addressing the assem ­
bly o f th e Birds.10
G reeks o b served these custom s even though pu rity was not
a n obviously im p o rta n t a ttrib u te o f their gods. T h e gods ruled
th e un iv erse because they w ere pow erful a n d im m ortal, not
b ec au se th ey w ere p ure. I t w as in p ractical ra th e r than theologi­
cal term s th a t divine p u rity becam e an im p o rta n t conception.
S acred n ess is elusive, irreducibly m etaphysical; purity, though
also m etap h y sica l, can a t least be expressed sym bolically in
c o n c rete term s. C leanliness is, in fact, not a special p reparation
for w o rsh ip b u t a re q u irem en t for form al, respectful behaviour
o f a n y kind; th ere is no generic difference betw een the lu stra­
tio n s th a t p reced e a p ra y er a n d those th a t precede a m eal,"
6 H orn. II. 6.266—8, cf. e.g. 9.171 f., 16.228—30 (libation cup cleansed w ith sulphur),
Od. 4 .7 5 0 - 2 (b a th an d clean clothes), M oulinier, 26. L ater evidence e.g. Hes. Op. 7241'.,
E u r. E l. 7 9 1 -4 , M oulinier, 7 1 -4 .
7 H om . II. 1.449, an d passim. In classical tim es the sprinkling w as often done with a
to rch o r olive b ran ch dip p ed into the lu stral w ater, E ur. H F 928 f., Ath. 409b. For
d e ta ils see E itrem , Beiträge iii, 1-19, also G inouvès, 3 11-18.
" See p. 2 2 / n. 108 below.
9 In c u b a tio n : see p. 213 n. 31. In itiatio n , m ysteries: se eC h . 10 below, also Ginouvès,
3 8 0 - 6 (w a te r in stallatio n s in sanctuaries o f D em eter and o th er goddesses; cf. LSCG
65.37, 1 0 7 -1 2 ). P rophecy: e.g. schol. E ur. Phoen. 224, F ontenrose, 224, Ginouvès,
3 2 7 -4 4 . 10 Ar. A r. 463 f.
11 e -g· H orn. Od. 1.146, Ar. Av. 464. N o generic.difference: cf. J. G ould, JHS 93
(1973), 79 n. 34.
Purification: a Science o f Division 21
w hich is itself, for th e G reek, a cerem onial occasion. In both
cases th e p erson affected sheds a little o f his everyday self.
T h e se p u rificatio n s can be looked a t from the o th er side: not
only do th ey p re p a re the in d iv id u al for a special event, b u t they
also serve to lift th e event itself o u t o f the fam iliar p lane and to
im b u e it w ith san ctity . T h is w as often achieved by creating a
clear sp atial fram e for th e im p o rta n t occasion. Before every
m eeting o f th e council a n d assem bly a t A thens, a young pig was
killed a n d its corpse carried ro u n d th e circum ference o f the
ineetin g -p lace by special officials know n as perisliarchoi.12
T h o u g h q u ite different in form from any w ashing process, this
was still a ‘p u rifica tio n ’. Its function in creating a division was
so clearly felt th a t A risto p h an es, no d o u b t echoing p o p u lar
usage, co u ld sp eak o f taking a seat on th e Pnyx as ‘com ing inside
the p u rific a tio n ’.13 T em p les too w ere som etim es cleansed be-
iore festivals, a n d som e lexicographical sources tell of the
th e a tre , p u b lic buildings, ‘th e city ’, a n d m eeting-places in gen­
eral b eing tre a te d in th e sam e w ay .14 H ow ever th a t m ay be, the
n am e o f th e officials concerned, the ‘ro u n d the h e a rth leaders’,
show s th a t th e rite derives originally from household p ractice .15
These are clearly sym bolic acts, not m isguided provisions for
p u b lic hygiene; it w ould, one im agines, not have m ade sense to
purify th e assem b ly p ro v idently in advance, w hen no one was
th ere to see. O n a t least one occasion the custom perm itted the
A th e n ia n s a vivid sym bolic action. D u rin g a session o f the
assem b ly in 370, new s arriv ed from Argos o f the civil conflict
in w h ich 1,500 m en w ere killed. A fresh purification o f the
assem b ly w as o rd ered at o n ce .16 M u ch later, we hear o f the
M a n tin e a n s co n d u ctin g an elab o rate purification o f th eir land
afte r a tro o p o f m u rd ero u s C y n aeth an s had passed through.
S la u g h te re d an im als w ere carried ro u n d the city an d entire
12 M ost o f the sources are p rin ted by Ja c o b y in his com m entary on Istros 334 FGrH
lr. 16. T h e γαλή o f Ar. Eccl. 128 is pro b ab ly a com ic m istake. T h e rem ains w ent to the
crossro ad s, D em . 54.39.
13 Ar. Ach. 44.
14 P h o t., S u d a s.v. περιοτίαρχος, cf. H arp . s.v. καθάροιον, schol. Aeschin. 1 .2 3 .1user.
Cret. 4.146 (L S S 114) has a purification o f sh ip y ard s, according to the in terp retatio n of
G u ard u cci, a d toe.
15 Cf. E itrem , Opferritus, 177, R E 8.1280 f., 19.859. Hesych. s.v. περίστιον m akes this
dom estic p u rificatio n post-lunerary.
16 P lu t. Praec.Reip. Ger. 814b, cf. Diod. 15.57.3-58.
22 M iasma
te rrito ry .17 In these cases th e traditional rites o f division
w ere being re-exploited to express horror an d rejection, by
s e p a ra tin g th e citizens from the a b h o rre n t events.
T h o se w ith in o n e o f these purifying encirclem ents were
m a rk e d as h a v in g som ething in com m on. T h e purification of
th e boule a n d ekklësia defined the p artic ip a n ts as the A thenian
p eo p le in council o r assem bly, w hile m em bership o f a religious
co m m u n ity w as com m only expressed in term s o f ‘sharing lus­
tra l w a te r ’.18 T h is unifying function is well seen in the practice
o f p u rify in g a n arm y before the cam paigning season .19 Each
sp rin g , w h en th e M aced o n ian arm y reassem bled, it was
m a rc h e d b etw een the two halves o f a sacrificed dog, which
cre a te d w h a t h as been called a n ‘absorptive zone’ for all its
im p u ritie s. (T his, in cidentally, is the only form o f G reek puri­
fication for w h ich a really close n ear eastern analogue has been
d e m o n s tra te d .20) A fter the purification had, as it were, recon­
s titu te d th e m en as an arm y, they divided into tw o halves and
pro c eed e d to b eh av e as an arm y in sim ulated fight. P lutarch
re co rd s an id en tical rite for B oeotia, an d though he says nothing
o f its co n tex t it w as no d o u b t sim ilar. I f such an n u a l purifica­
tio n s w ere perform ed elsew here in G reece, they have left no
tra c e in th e sources, b u t an in cid en t in the Anabasis shows the
sam e ritu a l being exploited to w eld a n arm y back into a unity.
S om e o f th e G reeks h ad m ad e un au th o rized raids on villages,
a n d th e re h a d b een G reek casualties; w hen am bassadors came
to offer re s titu tio n o f the corpses, they w ere slain by the dead
m e n ’s co m p an io n s. Indiscipline seem ed to be increasing, and
so X e n o p h o n sum m oned an assem bly an d m ade a stirring
17 Polyb. 4 .2 1 .8 -9 .
18 e.g. A esch. Ag. 1037, Eum. 656, Soph. O T 240, Kur. Or. 1602, Dem . 20. 158, and
esp. A r. Lys. 1129 Γ. The point is m ade by G inouvès, 313.
15 See m ost recently P ritchett, iii, 196-202, w ith references. M acedonia: Livy 40.6.1 -
5; Polyb. 23.10.17 = S u d a s.v. έναγίζων; H esych. s.v. Ξανθικά. Boeotia: Plut. Quaest.
Rom. 111.290d. F or a possible purification o f a fleet before em barkation see Inscr. Cret.
4.146 (L S S 114), w ith G u a rd u c c i’s com m entary.
20 T o a v e rt a n evil om en, the H ittite arm y w as m arched through the halves o f a
sla u g h te re d p riso n er o f w ar: see O . M asson, R H R 137 (1950), 5 -2 5 , H. M. Küm m el,
Ersatzritualeflir den helhilischen König, W iesbaden, 1967, 150-68. Survival or recollection
o f th e p ra c tic e in A n ato lia, H dt. 7.39.3. G reco-R om an parallels for this cathartic ‘zone
d ’a b s o rp tio n ’ (M a sso n ’s term ) in M asson, loc cit., E itrem , Beiträge ii, 8 -1 6 . For a
H ittite ritual battle, com m em orating a historical victory, see H. Ehelolf, Sitz. Preuss. Ak.
Berl. 1 9 2 5 ,2 6 9 -7 2 .
Purification: a Science ojDivision 23
speech . ‘H o w a re we to p ray to the gods w ith a clear consci­
e n c e ’, h e asks, ‘if w e behave so wickedly? W h at city will receive
us, w h a t h o n o u r will w e enjoy a t hom e?’ H is eloquence
triu m p h e d ; th e troops insisted on th e restoration o f discipline
a n d p u n is h m e n t o f offenders, a n d X en o p h o n w ith th e seers’
su p p o rt p ro p o sed th a t th e w hole arm y be p u rified .21 R ep en t­
ance, ch a n g e o f h ea rt, rejection o f an arch y , reassertion o f the
a rm y ’s co rp o ra te id en tity as a disciplined unity: such was the
m essage o f this purification.
R esto red to itself after an extern al incursion, a com m unity
m ig h t express its sense o f recovered integrity by purifying the
places a n d elem en ts tain ted by th e in v ad er’s presence. A fter the
P ersian w ith d ra w a l in 479, th e G reek leaders a t P la taea con­
su lted D e lp h i ‘a b o u t a sacrifice’, b u t w ere told ‘not to sacrifice
before th ey h a d ex tinguished th e fire in the country, since it h ad
been p o llu te d by th e b arb arian s, a n d fetched p u re fire from the
com m o n h e a rth a t D e lp h i ’.22 As th e bringing o f new fire was an
a n n u a l cerem o n y in several p laces ,23 this was, like the A th en ian
re-p u rifica tio n o f th e assem bly, a n ad a p ta tio n o f a reg u lar ritual
for a specific expressive purpose. I t w as the m ost p o ten t renew al
a G reek co m m u n ity could undergo, since, lodged in th e indi­
v id u al h e a rth s o f houses a n d th e collective h ea rth o f the city, fire
was th e sym bolic m iddle p o in t aro u n d w hich the life o f the
g ro u p revolved. A lthough this is n o t recorded, there was no
d o u b t also m u ch p u rification o f surviving tem ples from the
P ersian p resence. T h e M essenians, it is said, once expelled all
E p ic u re a n p h ilosophers, a n d th en purified the shrines a n d the
e n tire s ta te .24
P u rificatio n , therefore, m arks off sacred places from profane,
creates special occasions, a n d unites individuals into groups. A
fu rth e r a re a o f experience w hich it helps to organize a n d articu ­
late is th e p ercep tio n o f time. Few people, in th eir inform al
th o u g h ts a b o u t tim e, consider a y ear as a succession o f 365
21 Anab. 5.7.13—35. A sim ilar purification o f the M acedonian arm y to end a period of
dissen sio n after th e d e a th o f A lexander, C u rt. Ruf. 10.9.11.
22 P lut. Arist. 20.4.
23 N ilsson, G F 173, cf. W . B urkert, C Q 20 (1970), 1-16. O n fire sym bolism X en. Lac.
13.3 is revealing.
24 A elian, fr. 39, p. 2 0 1 .1 3 -2 4 H ercher. Rom e was purified after the expulsion o f the
T a rq u in s , D ion. H al. Ant. Rom. 5.1.3, an d th e G auls, Livy 5.50.2, Plut. Cam. 30.3; cf. too
l ac. Hist. 4.53, Sil. Pun. 12.752.
24 M iasma
hom ologous days, or even tw elve cycles o f the m oon. I t is m ade
u p o f seasons, o f fesdvals, o f holidays, changing activities and
th e in te rv a ls betw een th e m .25 P urification, w hich removes dirt
from th e p a s t a n d so m akes re ad y for the future, is ideally suited
as a ritu a l to m ark transition. R om an exam ples are particularly
clear. T h e tubilustrium o f M arch a n d the armilustrium o f O ctober
in d ic a te d s ta rt a n d finish o f the cam paigning season .26 T h e
p e n te n n ia l lu stra tio n o f the people in arm s included both
than k -o lferin g s for the previous five years a n d prayers for the
follow ing ,27 a n d the w ord lustrum itself actually cam e, by way of
this cerem ony, to den o te a period o f time. As well as m arking
ch a n g e in a n e u tra l sense, purification is also, o f course, well
su ite d to satisfy th e urge periodically felt by m ost people to
m ak e a new s ta rt, a n d feel a ta in te d environm ent grow fresh
ag a in .
I t will, as often, be m ost convenient to consider A thens,
w h ere th e evidence is m ost a b u n d a n t. T h e chief public purifica­
tion lay in th e d isp atch o f scapegoats ,28 a practice which,
th o u g h it seem s in m ood a n d sym bolism to belong to a re­
stric te d ru stic com m unity, p ro b ab ly still took place in the
m e tro p o lita n A th en s o f the late fifth cen tu ry ,29 a n d cannot be
sh o w n to h av e been ab a n d o n e d even in, or after, the tim e of
A ristotle. T h e se w retched in dividuals w ere the an im ate form of
th e ‘offscourings’ (katharmata) w hich, in m any G reek purifica­
tio n s, w ere expelled from th e area o f h u m an h ab itatio n , carry­
in g im p u rity w ith them . S capegoats are said to have been sent
o u t in resp o n se to specific crises, such as d ro u g h t or plague, and
o b v io u sly th e cerem ony o u g h t prim arily to be discussed in the
c o n te x t o f co m m u n al reaction to danger. Since, however, such
rite s also h a d a fixed place in the festival cycle (not ju s t in
A th e n s, b u t a lm o st w herever in the G reek w orld they are known
to h av e b een perform ed), they m u st also count am ong the
25 Cf. E. E. E v a n s-P ritch ard , The Nuer, O xford, 1940, 9 4-108.
** L a tte , R R 117, 120.
27 S uet. Aug. 97.1, T . M om m sen, Römisches Staatsrecht3, Leipzig, 1887, 2.1.412 f. O n
Lustrum condere see R. M . O gilvie,y/W 51 (1961), 3 1 - 9
28 Cf. p. 258 below.
29 A r. Ran. 733.
Purification: a Science o j Division 25
p u rificatio n s th a t artic u la te th e m ovem ent o f the y e a r .30 Indeed
no reliab le in stan ces o f th e d isp atch o f scapegoats outside the
re g u la r seaso n al fram ew ork are anyw here recorded.
A t A th en s, scapegoats w ere sent o u t on T harg elio n 6, as a
p a rt o f th e T h a rg e lia , a festival o f Apollo; on T hargelion 7,
offerings o f th e still rip en in g corn, fu rth er first fruits, a n d the
G reek e q u iv a le n t o f the M aypole (eiresione) w ere carried in
solem n procession, an d new -corn cakes were b ak e d .31 T h e sam e
co n n ectio n w ith th e T h arg elia is found a t E ph esu s ,32 an d may
well be an c ie n t. T h e scapegoat ritu al has therefore som etim es
been seen as a m agical p ro tectio n for the new y e a r’s ripening
p ro d u c e a t a perilous tim e .33 T h e re is, however, no need to see
the re la tio n betw een the tw o days o f the festival in m agical
term s; on T h arg elio n 6, bad things are driven out, while on
T h a rg e lio n 7, good things are carried in, in a p a tte rn whose
a p p e a l on a n expressive level is self-evident. T h e scapegoats
w ere not, to o u r know ledge, led am o n g the crops, an d they were
said to pu rify th e city 34 a n d not th e fields. Scapegoat-like cere­
m onies w ere perform ed in o th er states a t different tim es ,35 and
A pollo, h o n o ra n d o f th e rites a t A thens, was a god m ore con­
ce rn e d w ith p u rificatio n th a n farm ing. C o n cen tratio n on the
h a rv e st o b scu res th e m ore general sense in w hich th e T h argelia
w as for A th en s a festival o f purification an d renew al. These
d ay s w ere p e rh a p s th e two m ost auspicious in the entire A the­
n ia n c a le n d ar. O n the sixth o f T h argelion, Socrates (and
A rtem is) w ere b o rn , P lato on the seventh. M ost o f the victories
o f th e P ersian w ars (A rtem isium , M ycale, Plataea) cam e in time
to be set on T h arg elio n 6, as well as b oth the b irth an d d eath of
A le x a n d e r .36 N ew fire, pow erful sym bol o f renew al, arrived
30 O n th e reg u lar/o ccasio n al co n trast see D eubner, 184—8. Even lor M assilia, where
th e so u rce sp eak s ol an occasional ritual only, w h at is described sounds regular (Scrv.
Aen. 3.57 = P etro n iu s, fr. 1). T h e D elphic rite o f the Septerion is in terp reted as ail
eig h t-y early ex pulsion o f a scapegoat by e.g. J . Fontenrose, Python, Berkeley, 1959,
4 5 3 -6 1 .
31 D eu b n er, 1 8 8 -9 2 . D eu b n er denies use o f the eiresione at the T h arg elia, bul cf.
N ilsson, O C R 125, V e rn a n t, Tragédie, 119 f., an d on the eiresiônë Burkert, S H 134.
32 H ip p o n ax , fr. 104.49.
33 N ilsson, O F 1 1 3 -5 , followed by D eubner, 192 f. 34 H ip p o n a x , fr. 5.
35 P o rp h . Abst. 2.54 (R hodes), D eubner, 187 f. (T erracin a).
36 P lu t. Quaest. Com. 717b, D .L. 2.44, Ael. VH 2.25. But for the b irth o f Apollo on
T h a rg e lio n 7 (N ilsson, O F 209) I can find no au th o rity (contrast Plut. Quaest. Oraec.
9.292e).
26 Miasma
from D e lp h i a t som e tim e d u rin g the m o n th T hargelion, very
p ro b a b ly d u rin g th e ac tu a l festival o f the T h a rg e lia .37
C le a n sin g is also the them e o f the two rem aining festivals of
T h a rg e lio n , the K a lly n teria a n d the P ly n teria ,38 w hich are
re g u la rly associated in the sources an d w ere p robably in fact
closely re la ted . M ost o f o u r know ledge of the Plynteria is due to
th e coincidence th a t w hen A lcibiades re tu rn ed from exile in
408, he unluckily chose the d ay o f this festival for his landfall in
A th e n s .39 O n this day, we learn, m em bers of the Praxiergidai
re m o v ed th e a d o rn m e n ts o f A th en a Polias’ an cien t image,
veiled it, a n d perform ed secret rites. Since the goddess’s im age
w as covered, this w as am ong the m ost inauspicious days o f the
e n tire year; th e tem ples w ere closed ,40 an d no A thenian would
hav e th o u g h t o f choosing it to begin an im p o rtan t undertaking.
I t seem ed th a t w ith h er covered h ead the goddess herself was
sp u rn in g A lcib iad es’ presence.
V ery little m o re is know n o f the festival th a n w hat can be
le a rn t from this anecdote. L iterary sources differ betw een
T h a rg e lio n 25 a n d 29 for its d a te ;41 a recently discovered dem e
c a le n d a r seem s to place it in the following m onth, but, as other
e p ig ra p h ic texts su p p o rt a celebration in T hargelion, its precise
lo catio n is a t p re se n t o b scu re .42 T h e nam e P lynteria indicates
th a t th e g o d d ess’s robes w ere w ashed before being replaced.
T h e tw o no b le girls w ho perform ed this function could be called
e ith e r loulrides o r pljntrides,43 a n d as loud is used of w ashing a
p erso n , w h ereas plunô applies to clothes, we can p erhaps infer
37 SIC? 711 w ith n. 8.
38 See, in ad d itio n to the festival handbooks, L. Z iehen in R E ‘2 1.1.1060-5, D. M.
L ew is, A B S A 49 (1954), 17-21, W . B urkert, ‘Buzyge und P alladion’, Zeitschrift f
Religions—und Geistesgeschichte 22 (1970), 356—68, an d , for possible P lynteria in Tegea, L.
K o e n e n , Z P E \ (1969), 7 -1 8 .
39 X en . Hell. 1.4.12, P lu t .A le. 3 4 .1 -2 .
40 Pollux 8.141: im possible to tell w h eth er all the tem ples (D eubner) or only some
(Z ieh en ) w ere affected. I G I 3 7.20—22, as restored an d interpreted by Lewis, loc. cit.,
h a s a tem p le locked th ro u g h o u t the m onth o f T hargelion.
41 Cf. D. M . Lew is, loc. cit., M ikalson, 1 6 0 -4 , B urkert, GR 347 n. 5.
42 L ines 52—4 o f th e T horikos calendar, SE G xxvi 136; for the text an d com m entary
see G . D u n st, Z P E 25 (1977), 243—64, J . L ab arb e , Thorikos, les Testimonia, G ent, 1977,
η. 50. B u t for P ly n teria d u rin g T hargelion cf. IG I3 7 (LSCG 15), 2 0 -2 , IG I3 246 (LSCG
2), G 26. As the sta rt is fragm entary, the evidence o f the T horikos calen d ar is not utterly
u n a ssa ila b le (one could su b stitu te K ally n teria, for instance); b ut the alternatives are
u n convincing.
43 P h o t, a n d H esych. s.v. loutndes.
Purification: a Science o f Division 27
th a t th e s ta tu e itself w as b ath ed . W h ere an d how this w as done,
if in d eed it w as done, is unknow n. (It has recently been shown
th a t th e procession in w hich th e ephebes escorted ‘P allas’ to the
sea alm o st ce rtain ly form ed p a rt o f a different festival.44) A bout
th e K a lly n te ria alm ost n o th in g is reco rd ed ,45 b u t its nam e
suggests ‘sw eeping clean ’, a n d it is tem p tin g to suppose th a t at
th e K a lly n te ria th e tem ple p recin ct was cleaned, as the im age
itself w as a t th e P lynteria.
T h e clean in g o f statu es was to som e extent a p ractical neces­
sity. A t E leusis, a special fu n ctio n ary is early attested , a n d at
O ly m p ia, th e sam e task is said to have been bestow ed as a
privilege o n th e d escen d an ts o f P h eid ias .46 I t m ight be
p erfo rm ed before sacrifice, o r accom pany th e p u rification o f a
sh rin e w ith an im al blood .47 Even in these cases, how ever, it was
a w ay o f cre a tin g a sense o f occasion in p re p a ra tio n for a rite;
a n d w h en , as a t A thens, it gave its n am e to an im p o rta n t public
festival, it h ad clearly acq u ired a sym bolic religious significance
q u ite d istin c t from the p ractical req u irem en ts o f cleanliness. A
m o n th -n a m e P ly nterion is attested for C hios, Paros, los, and
T h a so s, w h ich suggests th a t th e P lynteria m ay have been an
a n c ie n t Io n ia n rite .48
T h e b a th in g o f statu es, p artic u la rly the statu es o f goddesses,
in sp rin g s, rivers, or the sea was n o t rare in G reek c u lt .49 In
a d d itio n to re g u la r an n u al rituals, th e statu e m ig h t be taken out
for w ash in g , o r w ash ed on th e spot, if a tem ple w as polluted by
d e a th o r b lo o d sh ed .50 T h ere is no reason to look for a single
e x p la n a tio n for all such cases, because an im ag e-b ath m ay
im ita te a n y o f th e various m otives th a t an actu al goddess m ight
h av e for b ath in g . T h e b a th th a t H e ra ’s im age receives in
P la ta e a is p re -n u p tia l ,51 b u t it seems from C allim ach u s’ Hymn
44 By B u rk ert, op. cit.; d o u b ts alread y in C. J . H erington, Athena Parthenos and Athena
Polias, M a n c h e ste r, 1955, 30 n. 2. P ly n teria procession: Phot. s.v. ήγητηρία.
45 F o r its d a te see M ikalson, 164, an d for activities Lewis, loc. cit.
46 IG I 3 I A 14 (L S S 1), Paus. 5.14.5, C lin to n , 95, R E 19.1559 f.
47 LSC G 58.12 f., 39.26.
48 See· Λ. E. S am uel, Greek and Roman Chronology, M unich, 1972, index 2 (m onth
n a m e s). 49 Cf. G in o u v ès, 2 8 3 -9 8 .
50 E u r. I T 1199, L SC G 154 B 24 f., ? USA 79.14 f., 532 FG rH D (2). M erely an
ex ten sio n o f th e co m m o n p ractice o f purifying the precinct after a pollution, p. 145 n. 6.
51 P aus. 9 .3 .5 -9 , Plut. ap. Euseb. Praep. Evang. 3.1. 1—3 = fr. 157 S andbach; the
n u p tia l in te rp re ta tio n o f th e rite is secondary (cf. Nilsson, GF 5 0 -5 6 , R E 2 0 .2 .231 9 -2 5 ,
B u rk ert, S H 132—4), b u t th e b ath clearly belongs to it.
28 Miasma
th a t ‘P a lla s’ b a th ’ in A rgos is sim ply taken for b ath in g ’s sake ,52
a n d in th e case o f the P lynteria, th ere is no hint o f any specific
m o tiv a tio n for w h atev er cleansing the goddess received .53 She
w as, it seem s, p artic ip a tin g in the general renew al ch aracter­
istic o f th e season, and, by h er particip atio n , involving her
citizens too. I f th e d ay o f the cleansing itself was, for all the
A th e n ia n s, a n inauspicious tim e, the succeeding days will, by a
fa m iliar process o f c o n tra st ,54 have been a period of especial
lib era tio n , w ith life b eginning anew in purity. T h e ancient aition
re p re se n ts th e P ly n teria as the occasion w hen the w om en of
A theris first w ashed th eir clothes after their year-long grief for
th e d a u g h te r o f C ecrops A g lau ru s ;55 it was a tim e, therefore, of
revival a n d renew al o f hope.
T h e six th m o n th after T h arg elio n was M aim akterion.
T h a rg e lio n m e a n t the rising o f the Pleiades, the beginning of
su m m e r a n d o f th e harvest; M aim akterion the setting of the
P leiades, th e beginning o f w in ter a n d of the ploughing. T hese
w ere th e tw o tu rn in g -p o in ts o f the y e a r .56 Like T hargelion,
M a im a k te rio n w as characterized by cerem onies o f purification.
I t m ay h av e been d u rin g this m o n th th a t the ephebes solem nly
esco rted th e P alladion dow n to the sea a t P haleron for
c lea n sin g .57 I t w as certainly then th a t the P om paia were
c e le b ra te d .58 T h e evidence for this cerem ony is scanty, b u t we
a re told th a t, ‘am o n g the purifications’, the fleece o f a ram
52 Hymn 5. T h e m ythical baths th at serve as precedent are ju s t baths (vv. 5 -1 2 ,
7 0 - 4 ).
53 T h e once p o p u la r p re-n u p tial theory (e.g. Fehrle, 176 f.), well criticized by
G in o u v ès, 292 f., is still upheld by L. K oenen, Z P E 4 (1969), 14-18. Such theories
ignore o r d isto rt the ICallynteria.
54 Cf. D e u b n e r’s R o m an parallel, p. 22.
55 P h o t. s.v. Kallynteria, H esych. s.v. Plynteria. Glose correlation between aition and
r ite is so u g h t by A. M om m sen, Feste der Stadt Athen, Leipzig, 1898, 497-502.
56 Cf. H es. Op. 383 f., 614—17; H ippoc. Viel. 3.68 (6.594 L.); T heo p h r. Sign. 1.6 (fr. 6.
1.6, p. 117 W im m er), διχοτομεί τάν ένιαντύν Πλειάς τε δνομένη καί άνατέλλονσα; B C H Η.ί
(1961), 39 (E u ctem o n ); Pliny, Η Ν 18.280 namque vergiliae privatim attinent ad fructus, ut
quarum exortu aestas incipiat, occasu hiems, semenstri spatio intra se messes vindemiasque et omnium
maturitates conplexis. F o r ploughing in M aim akterion (denied by M ikalson, 86) cf.
D e u b n e r, 250, a n d th e link w ith th e setting o f the Pleiades (Hes. loc. cit.): a t 38° in 500
BC: th e P leiades rose M ay 20 an d set N ovem ber 3, according to E. J . Bickerm an,
Chronology o f the Ancient World1, L ondon, 1980, 112.
57 See B u rk ert, cited p. 26 n. 38 above.
58 E u sta th . a d H orn. Od. 22. 481, p. 1935. 5 f i (cited D eubner, 158 n. 2): cf. Hesych.
s.v. μα ιμάκτης:μειλίχιος, καθάροιος.
Purification: a Sdence o j Division 29
sacrificed to Z eus M eilichios (a dion) was carried o u t o f th e city.
T h e sense o f this rite is clear from the verb derived from it,
apodiopompein, to purify by expulsion; the fleece was an inani­
m a te scap eg o at, a n object th a t ab so rb ed evil an d was then
expelled. So one o fT h a rg e lio n ’s m ost im p o rtan t rites has a close
parallel in M aim ak terio n ; if th e b ath in g o f the P alladion were
m ore secu rely d a te d to this tim e, one could alm ost speak o f a
m irro r im age.
L ike th e T h arg elia, th e P ly n teria a n d the P om paia have both
been in te rp re te d as m echanism s o f ag ricu ltu ral m agic .59 T here
is a rev ealin g ly large elem ent o f th e a priori in such in te rp re ta ­
tions, since n e ith e r ritu al is ad d ressed to a farm ing god or
co n tain s th e least ag ric u ltu ral elem ent in its aitiology or
sy m bolism . T h o u g h it is tru e, as we have noted, th a t these
festivals o f renew al ap p ro x im ately coincide w ith im p o rtan t
m o m en ts in farm in g life, w h a t this proves is p erh ap s not th at
p u rifica tio n serves, ag ricu ltu ral ends, b u t th a t in a farm ing
c o m m u n ity th e em otional year, as it m ight be called, is shaped
a ro u n d th e ag ric u ltu ral y e a r .60 T h e obvious lan d m ark s th at
give sh a p e to th e dull succession o f days are events such as
h arv est, p lo u g h in g , or the sp ro u tin g o f the young corn. B ut the
info rm al c a le n d a r p u t to g eth er in this way acquires em otional
fu n ctio n s a n d can readily be festooned w ith sym bolic m eanings.
O n e has only to th in k o f th e associations of transience, b u t also
tra n sfo rm a tio n , th a t atta c h to the purely a rb itra ry new year in
o u r ow n society to see w h at h arv est tim e could have m ean t to a
G reek in ex tra -ag ric u ltu ral term s.
O n a sm aller scale, we find th e housefold purifications of
p riv a te cu lt artic u la tin g the experience o f tim e in the sam e way.
E ven w ith in th e m o n th , religion could distinguish between
su p erficially sim ilar d ay s .61 T h e concept o f dies fa sti a n d nefasti
m ay be asso ciated w ith R om e m ore th an w ith G reece, b u t the
hem ero lo g y o f H esiod o r his c o n tin u ato r was influential enough
to pro v o k e th e criticism o f H eraclitus; O rp h ic poets were
fascin ated by it, b u t the scholarly Philochorus, too, devoted a
m o n o g ra p h to th e them e, a n d P lu tarch in his com m entary on
59 D eu b n e r, 21, 158.
60 Cf. D u rk h eim , 349 f., ‘T h e seasons have only provided the o u ter fram ew ork lor the
o rg an izatio n [o f festivals] an d not the p rinciple upon w hich it rests.’
61 Cf. M ikalson, 13-2 4 .
30 Miasma Purification: a Science o f Division 31

H esio d c h a rac te ristic ally discovered scientific justifications for th e cro ssro ad s for H ecate on th e sixteenth, exactly h a lf a m onth
th e a n c ie n t beliefs .62 F rom the m eagre rem ains of this literature, a fte r th e n ew m o o n .68 T h ese tw o purifications o f p riv ate cult
w e le a rn , for instance, th a t the A thenians favoured the eight­ d iv id e th e m o n th as the p u b lic cerem onies o f T h arg elio n and
e e n th a n d n in e te e n th days o f the m onth ‘for rites o f purification M a im a k te rio n d ivide th e y ear. T im e is artic u la te d on both
a n d av e rsio n ’.63 T h e tu rn in g -p o in ts o f the m o n th in particu lar levels by th e sam e rh y th m s o f cleansing and renew al.
w ere m ark e d by re g u la r rites o f expulsion. A t the new m oon, T h e a c c o u n t o f this science o f division has been provisional,
th o se h o u seh o ld ers w ho could afford it sent o u t ‘m eals for a n d in som e respects one-sided. I t seem ed useful, however, to
H e c a te ’ to th e cro ssro ad s .64 T h ese m eals for H ecate are con­ begin w ith a w ay o f looking th a t relates purification to th e desire
sta n tly asso ciated w ith the purificatory offscourings throw n out for o rd e r, a n d th a t trea ts it as a form o f behaviour ra th e r th an as
in th e sam e place, an d , alth o u g h th eir exact relation to such a p ro d u c t o f a n explicitly fo rm u lated set o f ideas. T h e purifica­
relics 65 is n o t q u ite clear, it is certain th at like them they w ere a tions o f th is c h a p te r have m ostly n o t been in ten d ed to rem edy
w ay o f c a rry in g evil aw ay from the place o f hab itatio n , or at p o llu tio n s - lapses below th e level o f p u rity req u ired for every­
least o f p in n in g th e d angerous goddess a t the crossroads by d a y life - b u t to im p a rt a to uch o f sanctity, a state o f p u rity
p ro p h y la c tic offerings. A pollo’s D elian tem ple too w as purified ab o v e th e average. P urifications such as these create or restore
by p ig ’s blood once a m o n th .66 T h a t m odel o f restrained piety, v alu e ra th e r th a n av ertin g d an g er. T h e distinction, however, is
C le a rc h u s o f M eth y d rio n , cleansed a n d crow ned his statues n o t ab so lu te, as ‘sacred: not-sacred: p o llu te d ’ are points upon a
every new m o o n .67 B ut a t A thens, offerings w ere also sent ou t to c o n tin u o u s line. (.This trin ity m ay sound like an an aly st’s
a b s tra c tio n , b u t it ap p ears in ju s t this form in a G reek tex t.69)
62 H es. Op. 7 6 5 -8 2 8 ; H eraclitu s B 106 ap. Plut. Cam. 19.3; OF pp. 2 7 4 -9 ; Philochorus T o specific p o llutions, a n d to d an g er, we m ust now tu rn .
328 F G rH fr. 8 5 - 8 , 189 f.; Plut. fr. 100-112, 142 S andbach.
63 328 F G rH fr. 190 - b ut p erh ap s this referred originally to one m onth only,
M ik a lso n , 21.
64 A r. Plut. 5 9 4 - 7 w ith schol., A pollodorus 244 FGrH fr. 109.
65 D em . 54.39 distin g u ish es H eca te’s m eals from the rem ains o f pigs used to purify
assem b lies, L u cian , Dial. Mort. 1.1, Calapl. 7 from cathartic eggs. N o t all kalharmata
th erefo re becom e Hekataia. It is possible, how ever, th at Hekataia are the kalharmata left
by a specific form o f purification, th at o f the house (cf. T h eo p h r. Char. 16.7). O n the
o th e r h a n d , P lut. Quaest. Rom. 68.280c, ‘dogs are carried o ut to H ecate with the other
katharsia, p e rh a p s in d icates m erely th at the purpose of H ecate’s m eals was broadly
c a th a rtic , n ot th a t they them selves w ere exploited in a specific ritual of purification
before being taken out: the offering itself would have been the purification. In the
c o n fu sin g a n c ie n t controversy on όξνθνμια (see H arpoc. an d El. Mag. s.v. όξνθνμια·,
o th e r lexica ad d n o th in g ), Hekataia seem som etim es to be identified w ith kalharmata, but
not w ith the specific kalharmata from house purifications (which are, according to Didymus,
όξνθ ύμια ). P lut. Quaest. Conv. 708 I seem s to indicate that H ecate’s m eals were cooked.
A tteste d c o n stitu e n ts w ere magida (Soph. fr. 734, cf. p. 231 n. 141), puppies (Ar. fr. 204,
P lu t. Quaest. Rom. 68.280c, H esych. s.v. ‘Ε κάτης άγαλμα), and perhaps certain fish
(A n tip h a n e s ap. A th. 358 1 ). T hey w ere som etim es eaten, from poverty (Ar. Phil.
5 9 4 - 7 , ‘ I h e o p h r. Char. 16.4, L ucian, loc. cit.), or bravado (Dem . 54.39).
66 M o u lin ier, 106; B runeau, 93, cf. ibid., 270—4, 2 8 6 -7 for purification of the
1 hcsm o p h o reio n . P urification o f the sacred area, usually by p ig’s blood, before an
im p o rta n t festival o r on a reg u lar calen d ar basis was no d o u b t a general practice: see I G
I I 2 1672.126—7 ( = S IG 2 587, tem ple a t Eleusis an d priestess's house), IS C G 39.23 f.
(A p h ro d ite P an d em u s a t A thens, using a dove, because A phrodite abhorred pigs;
sim ilarly L SA 36.36, S a ra p is/Isis), L S C G 6 5 .5 0 ,6 6 ,6 7 f. (m y sterieso f A ndania), Aelius 68 P h ilo ch o ru s 328 FG rH fr. 86, cf. Jaco b y , ad loc., an d on the form o f offering
A ristid es 48.31 (A sclepieion o f P ergam um ). B o rg eau d , 230 f.
67 P o rp h . Abst. 2.16.

2 d e a th , A rtem is in th e Hippolytus ab an d o n s h er dying favourite:
BIRTH AND DEATH
E clipses, e a rth q u ak es, a n d m onstrous births are com m only
seen by p rim itiv e peoples as fearful portents. T h e fact is well
k no w n , a n d easy enough to u n d ersta n d ; these are phenom ena
th a t d is ru p t n a tu re ’s n orm al, observable course. It is a t first
sig h t m o re p arad o x ical th a t the m ost intim ately n a tu ra l o f all
ex p erien ces - begetting, b irth a n d d e a th - should also be seen
by p eo p le living close to n a tu re as p o ten t sourses o f im purity
a n d d an g e r. R eligious teachers a n d philosophers have rejected
th e w orld o f com ing-to-be a n d passing-aw ay w ith scorn. O ne
m ig h t su p p o se th e ir stance the late p ro d u ct o f speculation, a
k in d o f e stra n g e m e n t from n atu re , b u t the restrictions th at
h ed g e a ro u n d th e n a tu ra l processes th ro u g h o u t the world sug­
g est th a t it h as d e e p e r roots. A lm ost any book o f exotic travels,
a n y e th n o g ra p h ic study will tell o f the perils o f sexuality and
fertility, a n d th e m onstrous im p u rity o f the corpse. T his vast
diffusion does n o t m ean th a t the phenom enon is readily u n d er­
sto o d ; o n th e co n tra ry , it is one o f those universals or near
u n iv ersa ls th a t a re often taken to be com prehensible merely
b ec au se th e y a re com m on, a n d th a t prove u n d er investigation
too d e e p -se a te d , diverse, o r com plex for an y sim ple or single
e x p la n a tio n . T h e sam e im p u ritie s take on in different cultures
v ery d ifferent significance. In H induism , the uncleanness to
w h ic h th e b o d y is liable provides a theoretical basis for the caste
s tru c tu re , since th e low er castes are rendered im pure by w ash­
in g th e la u n d ry , c u ttin g the hair, a n d tending the corpses of the
h ig h er. In Z o ro astrian ism , by contrast, the im plications of
im p u rity a re n o t social b u t cosm ic; it is a w eapon o f the evil
p rin c ip le A h rim a n in his unceasing struggle w ith the creator,
O h r m a z d .1 B elief in the body’s im purity m ay be a phenom e­
n o n , like a n im a l sacrifice, a b o u t w hich little can usefully be said
in g e n e ra l term s.
1 See D u m o n t, 8 4 -9 3 ; Boyce, 94.
I
Birth and Death 33
T h e tw o n a tu ra l pollutions m ost often referred to in Greek
sources a re th o se o f b irth a n d d e a th .2 T o avoid th e pollution of
‘F arew ell. S acred law forbids m e to look upon the d ead , o r stain
m y eye w ith th e ex halation o f d e a th .’ H ippolytus necessarily
un d erg o es th e p o llu tio n o f d e a th a n d A rtem is necessarily shuns
it; th a t is th e in escap ab le difference betw een m ortal an d im m or­
tal. H u m a n sy m p ath y is all w ith H ip p o ly tu s w hen he com plains
‘Y ou find it easy to leave o u r long frien d sh ip ’; b u t it is a truism
o f G reek theology th a t E u rip id es has exploited to achieve the
p a th e tic effect .3 In th e Antigone, too, th e pollution o f d eath is
d ra m a tic a lly used. T h e d isp u te ab o u t Polyneices’ b urial has
b een c o n d u c te d in term s o f rights, deserts, and duties. W ith the
e n try o f T eire sias a n d his re p o rt we receive decisive p ro o f th at
A n tig o n e is in th e right. B irds o f prey have carried scraps o f the
u n b u rie d co rp se to the very altars, a n d all com m erce betw een
m a n a n d god is im possible. W hen an u n re p en tan t C reon insists
th a t m o rtals by th e ir acts ca n n o t pollute the gods, we can only
u n d e rs ta n d this rejection o f plain fact as lu n atic defiance.
T h ro u g h p o llu tio n , the universe has given an unam biguous
v erd ic t on th e m o ral q u estio n .4
As th ese exam ples have show n, the n atu ra l pollutions are
especially re p u g n a n t to th e gods. B irth or d ea th w ithin a tem ple
is sacrileg e ;5 th e sacred island o f Delos m ust be free from all
ta in t o f th e processes o f m o rtality .6 Even a h u m an who has
2 B irth a n d d e a th to g eth er: E ur. Cret. fr. 7 9 .1 7 -1 8 A ustin, D .L . 8.33, C hrysippus, ap.
F lut, de Stoic. Rep. 1044Γ—1045a, Schol. T heoc. 2.11/12 b, M en. Asp. 216 if., Porph. Abst.
4.16, p. 255.7 N auck.
3 E u r. Hipp. 1437 ii., cf. G riffin, 189; for the sam e m otif see E ur. Ale. 22 f., Ael. fr. 11,
M en. Asp. 97 f., H eliod. Aeth. 1.2.7. A pollo an d d eath inassociable, Aesch. Sept. 859 (cf.
Ag. 1075). 4 S oph. Ant. 9 9 9 -1 0 4 7 .
5 /6 ' 112 1035.10 πάτριον ϊσ τιν έν μηόενΐ τών τεμενών μ ή τ’ έντίκτειν μ ή τ’ έναποθνήσκειν,
P aus. 2 .27.1, 6 (E p id a u ru s), LSA 83, A nt. Lib. Met. 19.3. Leaving tem ple to die or give
b irth : T h u c . 1.134.3, X en . Hell. 5.3.19, Plut. Dem. 29.6, Ar. Lys. 742 f., SIG 3 1168.1.
D e a th in th e tem p les a sym ptom o f the extrem e dem oralization caused by the plague,
T h u c . 2.52.3. It is clear from these texts th a t these pu rity requirem ents applied to all
sa cre d p recin cts, n o t ju s t those o f specific gods.
6 T h u c . 3 .1 0 4 .1 -2 , C allim . Del. 276 f., S trab o 9.5.5 (486). As the latter 2 texts say
n o th in g o f the b irth tab o o , it has been suggested th at the D elians relaxed it once free
from A th en ian d o m in atio n : see B runeau, 48—52. T h e tem ple accounts list paym ents for
the disp o sal o f corpses w ashed u p on Delos; in the first such case the p u rch ase o f a ‘pig
for p u rificatio n ’ is m en tio n ed , b u t not subsequently (B runeau, loc. cit.). Plague
(fictional) on D elos as consequence o f a burial there, (Aeschines) Epistle 1.2. But the
p urificatio n sp a red to m b s o f ‘heroes’ (B ru n eau , 49).
34 Miasma
com e into c o n ta c t w ith b irth or d eath is excluded for a period
from w o rsh ip p in g the gods. As E u rip id es’ Iphigeneia com­
plains: ‘I criticize A rtem is’ clever logic. I f a m ortal is involved in
b lo o d sh ed , o r touches a new m other or a corpse, she shuts him
o u t from h er a lta r as polluted; b u t she herself takes pleasure
in h u m a n sacrifice.’ Auge reproaches A thena directly; if the
goddess is h a p p y to receive the dedication of blood-stained
spoils, how can she be an g ry th a t A uge has given birth in her
sh rin e? A n d yet she was; A uge’s crim e b ro u g h t barrenness on
th e e n tire la n d . M a n y people in the fifth century an d afterw ards
no d o u b t felt th e sam e unease as these E uripidean heroines
a b o u t such am o ra l rules o f im purity; the Stoic C hrysippus
b ra n d e d th em ‘irra tio n a l’; b u t the rules survived. I t was useless
to ap p ly b lu sterin g m oral dialectic to them because, though
p o llu tio n b elief m ig h t som etim es, as in the Antigone, reinforce a
p rin c ip le o f m o rality , these rules w ere essentially as am oral as
th e n a tu ra l processes them selves .7
T h e d esc rip tio n o f the rules th a t follows is necessarily eclec­
tic. T h e re is no G reek com m unity to w hich all the details that
will be m en tio n ed can be show n to have applied. For m any
p a rts o f th e G reek w orld there is no actu al evidence th a t such
re g u la tio n s existed a t all. W e never h e a r o f G reek com m unities
th a t ‘did n o t use p u rifications’, b u t it is clear from the local
re g u la tio n s w hich define how long im purity lasts a n d whom it
affects th a t a t this level, at least, there was no Panhellenic norm.
F o r d e a th ,8 w ith w hich we begin, a law of the fifth century lrom
Iu lis on K eos provides a fram ew ork ,9 while fu rth er details can
be a d d e d from A thens. T h e Iulis law is partially m odelled, as
several v erb al echoes show , on the funerary legislation of Solon,
7 E ur. I T 380—4; A u g e’s crim e, A r. Ran. 1080 + schol.; its consequences, Apollod.
3.9.1; h e r c o m p la in t, E u r. fr. '266 ap. C lem . Al. Strom. 7.23.5, p. 17 St. (the following
p o in t in C lem en t, ‘o th e r an im als do it’, is no d o u b t still E uripidean, cf. H dt. 2.64); on
E u rip id e s’ A uge see Z P E 4 (1969), 7 -1 8 .
8 O n d c a th -p o llu tio n see W ächter, 4 3 -6 3 , M oulinier, 7 6 -8 1 , N ilsson, GGR 9 5 -8 ,
G in o u v ès, 2 3 9 -6 4 . T h e basic account o f the funerary rites is R ohde, 162-74; cl.
K u rtz /B o a rd m a n , C h . 7, Alexiou, 4—23, a n d m ore generally the papers collected in
M e u li, Ges. Sehr., 1.301—435. See now the valuable discussion by C . Sourvinou-lnw ood.
‘A T ra u m a in Flux: D eath in the 8th century an d A fter', in R. Hiigg an d N. M .irinatos
(e d s.), The Greek Renaissance o f the Eighth Century B.C.: Tradition and Innovation.
S tockholm , 1983(?).
9 IG X I I 5.593 = S iG 3 1218 = LSC G 97. T h e Solonian law. Dem. »3.62; I speak of
‘S o lo n ’ for convenience w ith o u t w ishing to com m it m yself on the law 's origin.
Birth and Death 35
b u t differs from Solon in tre a tin g questions o f p u rity explicitly.
S o lo n ’s silence is ch a rac te ristic o f o u r evidence; alth o u g h funer­
a ry p o llu tio n was fam iliar a t A th en s, literary texts often fail to
speak o f it w hen tre a tin g o f d e a th or m ourning. I t is as though
being p o llu ted w ere, like w earin g d a rk clothes, ju s t one aspect of
th e s ta te o f m o u rn in g , a n d req u ired no special m ention.
A t th e m o m en t o f d ea th , th e house o f d ea th becam e polluted.
A special w a te r vessel was set outside, for the purification of
those co m in g o u t; this an d o th e r conventional tokens used to
in d ic a te a h ouse o f d ea th will have w arned those unw illing to
in c u r p o llu tio n n o t to en ter a t all. W e do not know w h eth er a
sp rin k lin g from this vessel was in itself full purification for those
w ho a tte n d e d th e w ake b u t not the funeral, o r w h ether
they w ere su b ject to fu rth e r restrictions. T h e w ater was fetched
from a n eig h b o u rin g house, as th e house o f d e a th ’s own supply
w as p o llu ted . A t Argos, we h e a r o f new fire b ro u g h t from next
door, a t th e conclusion of m o u rning, for the sam e re aso n .10 T h e
w om en o f th e household p re p are d th e corpse for th e cerem onial
lay in g -o u t a n d viewing; it w as w ashed, an o in ted , crow ned,
d re sse d in clean robes, generally w hite o r red, a n d laid upon a
bier strew n w ith b ran ch es a n d leaves .11 T h u s the d ead m an was
sym bolically m ad e p ure, in desp ite o f the co n tam in atio n all
a ro u n d him ; o f all those p resen t a t the wake, he alone w ore the
crow n, em b lem o f p u rity .12 C ertain obscure practices w hich the
I ulis law allu d es to p erh ap s o ccurred at this stage. ‘Do not p u t a
c u p u n d e r th e b ie r’ (p resu m ab ly d u rin g the laying-out), it
o rd e rs, ‘o r p o u r o u t th e w ater, or take the sweepings to the
to m b .’ In te rp re ta tio n here is guessw ork, b u t m odern p arallels 13
10 W a te r vessel: Ar. Eccl. 1033, E ur. Ale. 98 fF., an d lexicographers (R ohde, 188 n. 38).
W a te r b ro u g h t from nex t door: Pollux 8.65, Hesych. s.v. δστρακον. C ypress boughs
o u tsid e house o f death : Serv. Aen. 3.681, R ohde, 189 n. 39. New fire a t Argos: Plut.
Quaest. Gräec. 24. 297a. H ouses m ay even have been sealed off w hen d eath was
im m in en t, M en. Asp. +66 f.
" R ohde, 1 6 2 -6 w ith notes 36, 37, 40, 61; a d d C a llim . fr. 194.40-3. O n b ath in g the
co rp se see A ndro n ik o s, 2 - 4 , G inouvès, 239 f.; this was the p rep arato ry act th at carried
m ost sy m b o lic w eight. L ib atio n s to the corpse after burial could be spoken οΓα$χέρνιφ
o r λο υτρό ν, as th o u g h the process o f purification continued (Soph. El. 84, 434 etc.; cf.
P. S tengel, Hermes 57 (1922), 539 ff., G inouvès, 244).
12 D ead crow ned: R ohde, 189 n. 40; m o u rn ers not crowned: Arist. fr. 101 Rose3 ap.
A th . 675a.
13 B. S ch m id t, A R tV 24 (1926), 317 f., A R W 2 5 (1927), 82; M. G u ard u cci, SM SR 2
(1926), 8 9 -9 8 . R itu al sw eeping o f d e a th house a t Rome: L atte, R R 101; in Byzantium :
A lexiou, 25.
36 Miasma
m ak e it p lau sib le th a t the p oint o f these rituals was to banish
d e a th p o llu tio n from the house, the first two by catching it in a
vessel o f w a te r w hich was then poured away, a n d the third by
sw eep in g it o u t w ith the h o u sehold’s physical d irt an d deposit­
in g it w here it belonged, a t the tom b. If this is correct, it is
re m a rk a b le to find the C eans legislating against practices that
a re , it seem s, socially objectionable only in being superstitious,
a n d su p e rstitio u s only in the sense th a t they take too far th at
b elief in d e a th -p o llu tio n on w hich several o f the law ’s positive
re q u ire m e n ts are founded.
E arly o n th e th ird day occu rred the ‘carrying o u t’. Solon
re q u ire d th a t this should be perform ed before daw n; w hen the
e m p e ro r J u lia n passed a sim ilar m easure alm ost a thousand
y ea rs later, his aim was to p ro tec t passers-by from pollution,
b u t S olon w as p ro b a b ly m ore interested in discouraging the
o ste n ta tio n o f th e trad itio n al aristo cratic funeral by depriving it
o f a n a u d ie n c e .14 T h e place to w hich the body was carried lay
o u tsid e th e city, aw ay from the tem ples, and, o f course, no priest
a tte n d e d it. T h e disposal o f the body was the turning-point
w ith in th e seq u en ce of events th a t followed the death. Purifica­
tio n could now begin, a n d the activities o f everyday life be
g ra d u a lly re su m ed . A fter the funeral, it was trad itio n al for the
m o u rn e rs to w ash o r b a th e .15 T h e re followed, in the funerary
b a n q u e t, a n im p o rta n t reassertion o f the values of life an d of the
w ill to live; th e m o u rn ers resum ed the crow n, a n d sat down
to g e th e r to sh a re th e pleasures o f the ta b le .16 V ery probably,
14J u lia n , Epistle 136 B idez/C um ont. Solon: Dem . 43.62, cf. PI. Leg. 960a, Cic. Leg.
2.66, Z iehen, 264.
15 I re p e a t th is s ta n d a rd view on the au th o rity o f schol. R V Ar. Nub. 838, b ut w ithout
a b s o lu te confidence. T h e b a th th a t purifies the m ourners in LSCG 97 A 30 (the Iulis
law ) d id not necessarily follow the funeral im m ediately (cf. e.g. L SC G 124.4). T h u s the
fu n e ra ry b a th lacks firm early attesta tio n . T h e άπόνιμμα in the obscure exegetic
fra g m e n t o f C leid em u s ap. A th. 410a (323 F G rH fr. 14) is taken by K. M euli (in
Phylloboliaf ü r P. von derM ühll, Basle, 1945, 205 η. 1, = Ges. Sehr. 2.928 n. 2) as d irty w ater
in w h ich the m o u rn ers have w ashed, secondarily ‘offered’ to the dead as token o f a duty
p erfo rm ed ; th is is plau sib le, b ut such a n a ct m ight as well have occurred a t e.g. the
n in th -d a y rite as a t th e funeral itself. T h e lexicographical sources on the mysterious
έγχντρίστρ ια ι give them no o th er function at funerals th an pouring libations, although
in o th e r contexts they a re said to ‘purify the polluted’ (schol. (Plat.) M in. 315c, and
o th e r sources q u o te d a d . loc. by W . C . G reene, Scholia Platonica, Pennsylvania, 1938).
16 O n th e fu n era ry m eal see Fr. Pfister, R E s.v. Perideipnon. C row ning: Cic. Leg. 2.63.
R e a sse rtio n o f life a t the funeral: R. H u n tin g to n and P. M etcalf, Celebrations o f Death,
The Anthropology o f Mortuary Ritual, C am bridge, 1979, 3 4 -4 2 , 9 3 -1 1 8 , cf. of course
Birth and Death 37
how ever, th ey w ere not yet p erm itted to re tu rn com pletely to
n o rm al life. Ip h ig e n e ia ’s re m ark th a t A rtem is ‘keeps aw ay from
h er a lta r ’ a n y o n e w ho has ‘to u ch ed a corpse’ loses its point if the
p o llu tio n could be im m ediately effaced by ritu al w ashing. In
post-classical sac red laws, co n tact w ith d ea th norm ally causes
exclusion from th e sh rin e for a fixed period of days, a n d this is
su rely w h a t is im plied in E u rip id es .17 A lthough Iphigeneia is
sp ea k in g o nly o f A rtem is, it is m ost im plausible th at rules of this
kin d sh o u ld h av e been confined to one cult: all o u r o th e r evi­
d en c e suggests th a t all the O ly m p ian gods w ere equally con­
c e rn e d to keep th e n a tu ra l pollutions at a d istan ce .18 A t first
sig h t th e Iu lis law is m ore liberal. T h e legible p a rt o f the first
sid e ends: ‘those w ho are p o lluted . . . after w ashing .. . shall be
p u re ’, b u t th ere m ay well have followed a tem poral specifica-
itin era ry gam es. A m o u rn in g fast was p erh ap s seldom observed in classical tim es (one
case, A rr. Anab. 7.14.8), but m o urning a t S p a rta ended w ith a sacrifice to D em eter
(P lu t. Lyc. 27.4).
17 E u r. / Γ 380—4. S acred laws: LSCG 55.6 (A ttica, M en T y ran n o s, 2nd c. AD) 10
d ay s; L S C G 124.2—4 (E resus, unknow n cult,? 2nd c. BC) 20 days for a relative, 3 lor
a c q u a in ta n c e ; L S C G 139.13 (Lindos, unknow n cult,? 2nd c. AD) 40 days for a relative;
LSS 91. 1 3 -1 4 (L in d o s, A thene,? 3rd c. AD) 41 days for a relative, 7 lor w ashing a
c o rp se, 3 for e n te rin g the d e a th house; LSS 119. 3—4 (Ptolem ais, unknow n cult, ? 1st c.
BC) 7 d ay s; L SA 12. 7 - 9 (P erg am u m , A thene N ikephoros, after 133 BC.) one day fo ra
relativ e, im m ed iate access, after w ashing, from a ‘burial and carrying o u t’, i.e. a
n o n -re la tiv e ’s funeral; L SA 1 8 .7 -9 (M aeo n ia, M eter, 147-6 BC) 4 days f o ra relative,
tw o for an a c q u ain tan ce ; ? LSA 29.2; LSA 51.5 (M iletus, A rtem is, ? ls tc . BC) two days;
L SA 8 4 .6 - 9 (S m y rn a, D ionysus, 2nd c. AD) 10 days for a relative, 3 for acquaintance;
conceiv ab ly B C H 102 (1978), p. 326, line 15. Scraps of earlier evidence: LSCG 56.13
(C leon ae, early 6 th c. BC), b u t this p ro b ab ly refers to m u rd er, not n atu ral death ; LSS
31.10 ff. (T eg ea, ? 4 th c. BC), ap p aren tly prescribing a short period o f im purity after a
b u rial; ? L S S 106 (n eith er the first ed ito r no r Sokolowski offers a date); E ur. Ale. 1143-6,
A lcestis co n secrated to d eath for 3 days after retu rn from H ades. At C yrene pollution
laste d 3 d ay s after co n tact w ith a b irth , an d it w ould be strange if d eath-pollution was
effaced m ore q uickly (L S S 115 A 17—19). C oan priests were excluded from the house of
d e a th for 5 d ay s from the carry in g o ut; it is plausible th at m ourners should have been
ex clu d ed from th e sacred for the sam e period (LSCG 156 A 11, cf. 154 A 24—6, 3 9-41 ).
F o r d e a th in th e ho u se prev en tin g atten d an ce at a festival see A th. 46e—f, D.L. 9.43. If
su c h rules d id exist at A th en s, the silence o f th e epigraphical record suggests th at they
rem ain ed , in sign ifican t co n trast to the p ractice o f o th er states, at the level o f unw ritten
law s.
18 Cf. n otes to p. 33. D ocum ents like the C yrene law regulate w h at conditions pollute,
a n d for how' long; th ere is no question o f being p u re enough to visit one shrine b u t not
an o th er.
38 Miasma
tio n (‘on th e th ird d a y ’) or o th er qualification (‘b u t not enter
th e tem p les ’).19
T h e place th a t h ad h arb o u red the corpse required purifica­
tio n no less th a n the m ourners who had touched it. A t Iulis, the
h o u se o f d e a th w as sprinkled w ith sea w ater on the m orning
afte r the ca rry in g out; it was now pure, and contact w ith the
gods could be resum ed th ro u g h offerings at the h e a rth .20 (T he
b ereav e d m u st have ceased by now to be contagious, or the
h o u se w ould, o f course, have been im m ediately contam inated
an e w .) I f d e a th occurred in a public place, the entity requiring
clean sin g w as no longer the house b u t the w hole dem e, and
re sp o n sib ility for this task was specified with som e elab o ra­
tio n .21 E ven after the ‘carrying o u t’, pollution could still be
in c u rre d . T h e re is ap p e n d ed to the Iulis law a fu rth er regula­
tio n a p p a re n tly o f ra th e r later date. Funerals were succeeded by
fu rth e r rites perform ed a t the tom b, at gradually increasing
in terv a ls o f tim e, a n d it is the pow er o f these rites to pollute that
is h ere defined. ‘T h e council a n d people decided. O n the third
d a y (an d ) the an n u a l festival those w ho perform the rites shall
b e p u re, b u t no t en ter a tem p le .’22 T h ere follows an obscure
re g u la tio n a b o u t th e p urity o f the house in the sam e circum ­
stan ces. T h ese questions are decided, we note, by plebiscite and
n o t by c o n su ltatio n o f an oracle. It is not surprising th a t even
th e a n n u a l com m em oration o f the dead causes a m ild pollution.
Im p u rity lingers in the physical relics; the tim id a n d the pure
sh rin k from step p in g on a tom b, a n d a C oan law calls for
p u rifica tio n if a h u m a n bone or uncovered grave is discovered
19 T h e tra n sc rip t in IG X I I 5.593 in fact offers, after κα[0αρ]ονς έναι, εωι (from
A ttic -Io n ic εως, daw n?).
20 L S C G 9 7 A 1 4 -1 7 : for the reading in 16secG . K laffenbach, Philol. 97 (1948),372 f.
H o m e p urification a t A thens: A nt. Chor. 37, ? Dem. 47.70; for the house as recipient of
d e a th -p o llu tio n cf. E u r. Hel. 1430. A purificatory encircling o f the h earth m ay have
b een perform ed a t th is stage: see H esych. s.v. περίστιον
21 D em . 43.57 f., cf. Inscr. Cret. 4.76, LSC G 154 B 17—32.
22 L SC G 97 B 1—11. F o r the restoration καί, not έπί, in B 5 see Z iehen, 267 f.,
E. F re iste d t, Altchristliche Totengedächtnistage (Liturgiegeschichtliche Q uellen und
F o rsch u n g en 24), M ü n ste r, 1928, 112-14. it seems alm ost inevitable that here ‘the
th ird d a y ’ is co u n ted from th e funeral, not the death; otherwise it w ould coincide with
th e ekphora, a n d a special rule ab o u t the p urity of the celebrants would be unnecessary
(F re is te d t’s an sw er, op. cit., 96 n. 1, is inadequate: the only alternative is to suppose
th a t the rule on sid e B replaces th at on side A.)
Birth and Death 39
in a p u b lic p lace .23 T h is im p u rity p ro bably dim inished with
tim e; a co m m u n ity could, in a n em ergency, re-use its ow n old
g ra v e m o n u m en ts for b uilding - b u t a besieging force th at
ta m p e re d w ith th e tom bs o f th e besieged was ju stly punished
w h en disease su p erv en ed .24 As th e cult o f heroes w as celebrated
a t th e ir to m b s, th e p a rtic ip a n ts w ere som etim es req u ired to
purify them selves afterw ards, a n d people like priests w ho lived
in co n d itio n s o f special p u rity m ig h t be excluded from them
a lto g e th e r.25 E lab o ra te ap o tro p a ic precautions w ere taken in
A th e n s a t th e A n th esteria, w hen the dead re tu rn ed to e a rth .26
Ip h ig en e ia defined th e source o f pollution as ‘touching a
c o rp s e ’. T h a t is to p u t th e th in g in its m ost concrete form. B irth-
a n d m u rd er-p o llu tio n w ere co n tracted by en tering th e sam e
a re a o f social sp ace (typically th e sam e house) as the con­
ta m in a te d p erso n , a n d it w ould be su rp risin g if d eath-pollution
w ork ed differently. ‘T o u ch in g th e corpse’ m ight, o f course, have
form ed a p a r t o f th ç m o u rn in g ritu al, b u t this w ould be m erely a
tra n s la tio n o f social co n tact in to physical, an d w ould not prove
th e real p rim acy o f th e physical.27 T h ere m ay have been
a n o th e r d e te rm in a n t, m ore im p o rta n t th an ‘tou ch in g ’ or even
‘e n te rin g th e sam e ro o f’. In m an y societies, d eath -p o llu tio n is
sp re a d by re la tio n sh ip as well as co n tact:28 the d ead m a n ’s kin
a re c o n ta m in a te d from th e m om ent o f d eath , even if they are a
h u n d re d m iles aw'ay w hen it occurs. T h u s, in early Rom e, the
23 T o m b s: T h e o p h r. Char. 16.9, E u r. Cret. fr. 79.18 A ustin. C oan law: LSCG 154 B
1 7 -3 2 (if th e μ o f δά\μω ι in 17 is secure; ί£ρ]ώί would m uch im prove the sense).
24 D iod. 11.40.1, Lyc. Leoc. 44, P. M . F raser, Rhodian Funerary Monuments, O xford,
1977, 7; d isease, D io d 13.86.1—3. N o h arm , how ever, in overthrow ing a ty ra n t’s tom b,
P lu t. Tim. 22.2
25 P aus. 5.13.3; L SC G 154 A 22, 37; 156 A 8 -1 0 ; L SS 115 A 2 1 -5 ; Nock, ii, 577 f.
26 D eu b n er, 112.
27 T h e w om en w h o p rep ared the corpse o f course touched it. T w o H om eric m ourning
g estu res, to u ch in g the d ead m a n ’s chest and cradling his head, involved physical
c o n ta c t; the latter at least survived as a w o m an ’s gesture in classical tim es, b u t in
H o m e r th ey a re perfo rm ed only by th e d ead m an ’s closest associates, an d the typical
m a le gestu re a t th e classical prothesis seem s to have been a greeting from a d istance w ith
o u tstre tc h e d arm (11. 18.317; 23.18. 136 f.: 24.712, 724; G. N eum ann, Gesten und
Gebärden in der Griechischen Kunst, Berlin, 1965, 86; ibid., 89, for cradling). I know of no
G reek evidence for the farewell kiss (R om an texts in C . Sittl, Die Gebärden der Griechen
und Römer, L eipzig, 1890, 72). Cf. however, Plut. Pel. 33.8, T hessalians eager to touch
P e lo p id a s’ corpse, an d X en . Cyr. 7.3.8.
28 e.g. In d ia: D u m o n t, 88, H . O ren stein , Ethnology 4 (1965), 3; Borneo: H ertz, 39.
H u n d re d s o f miles: Lévy-B ruhl, 254 (T honga).
40 Miasma
im p o rta n t u n it o f pollution w as the 'familia funesta'. It is very
likely th a t this w as also the case in early Greece. T h e G yrene
c a th a rtic law states th a t birth, in co n trast to death, pollutes ‘the
h o u se [or, household] itself’. N othing is said ab o u t how death-
p o llu tio n o p e ra te d , b u t the necessary contrast w ould be p ro ­
v id ed if it affected the b ro a d er kin group ra th e r th an the
re stric te d fam ily resid en t u n d e r the sam e roof- In later sacred
law s, relatives are certainly polluted by a d eath for longer than
o u tsid ers, a n d it is h a rd to see w hy this distinction should be an
in n o v atio n . T h e re is, unfortunately, no clear evidence either to
p ro v e o r disprove th a t relatives could be autom atically polluted
w ith o u t com ing in to contact w ith the corpse .29
In th e Iu lis law th e pollution g roup is actually defined, but
th e sto n e becom es illegible a t a crucial point. I t begins ‘No
w o m a n shall go to the house w here a m an dies w hen he is
c a rrie d out, except those w ho are polluted [or, pollute
them selves]. T h e re shall be polluted [or, pollute them selves]
m o th e r, wife, sisters a n d d au g h ters, a n d in ad d itio n to these not
m o re th a n five w o m en .’ A t this point the reading becomes
u n c e rta in ,30 b u t a p ro b ab le reference to ‘children o f cousins’
suggests th a t the Iulis pollution g roup resem bled the A ttic
k in sh ip g ro u p o f anchisteia, w hich extended to cousins’ children
a n d u n d e r S olon’s law d eterm in ed the right o f p articipation in
m o u rn in g .31 T h e ‘no t m ore th a n five w om en’ m ight well be
re la tio n s by m a rria g e .32 T h e regulation is intriguing in its sug­
g estio n th a t pollution, ra th e r th a n being an incidental bu t
in e v ita b le b y -p ro d u ct, is a tem p o rary statu s to which not
everyone associated w ith the funeral is adm itted. I t would be
possible, a lth o u g h in this case illegal, for a w om an to enter the
29 Familia funesta: L atte , R R 49. T herefore the d ictato r Sulla divorced his dying wife,
P lu t. Suit. 35.2. C yrene: L S S 115 B 2 4 -7 . L ater sacred laws: p. 37 n. 17 above. N o clear
evidence: th e an ecd o tes in (P lut.) Cons, ad Apoll. 118c—119d (cf. parallels in th e T e u b n e r
a d loc.) are a m b ig u o u s an d unreliable: cf. th e conflicting reports in the Rom an
tra d itio n over th e ritu a l sta tu s o f H o ratiu s Pulvillus after his son’s d e a th (R E 8.2402).
30 H o ffm an n ’s re sto ratio n o f 28 f., com m only accepted, is im possible; see Bechtel on
S G D I 5398. I f th e g e n d er o f άλλον όίμεόένα in 29 is taken literally, no m en are polluted
ex cep t an y th ere m ay be am o n g the ‘ch ild ren ’ o f the preceding clause. M ore probably
th e m ascu lin e is generalizing, an d only the fem ale pollution group is here regulated.
31 M . B ro a d b e n t, Studies in Greek Genealogy, Leiden, 1968, 119-50.
32 N ote, how ever, th a t R. F. W illetts’s theory o f a special funerary role lor affines,
a d o p te d by A lexiou, 10 fT., is refuted by H . M eyer-L aurin, Gnomon 41 (1969), 162 f.,
H . J . W olff, Zeitschrift der Savigny-Stiftung f ü r Rechtsgeschichte, Römische Abteilung, 85
(1968), 4 2 2 -6 .
Birth and Death 41
house o f d e a th w ith o u t becom ing ‘one o f those p o llu ted .’ As this
in n e r g ro u p o f th e p o llu ted is n o t d eterm in ed sim ply by rela­
tio n sh ip , th ere was p resu m ab ly som e specific act or d u ty by
w hich it w as defined. O n e possibility is th a t they h an d led the
corpse, b u t this scarcely suits H e ro d o tu s’ rep o rt o f w h at is
a p p a re n tly th e sam e p h en o m en o n in S parta: ‘At the d e a th o f a
king tw o free in d iv id u als from each household, a m an and a
w o m an , m u st be p o llu ted [or, p o llute them selves]; th ere are
severe p en alties if they do n o t .’33 M o re p ro b ab ly the reference is
to self-defilem ent o f som e kind; fem ale m ourners in classical
tim es m ig h t still re n d th eir clothes, pluck o u t th eir hair, and
te a r th e ir cheek s .34 W h atev er the precise in terp retatio n , it is
c le a r th a t in b o th cases ‘being p o llu te d ’ is m ore like going into
m o u rn in g th a n catch in g a disease.
A m a n ’s statu s is seldom so clearly revealed as a t his passing.
T h e d e a th o f a stran g e r is m et w ith indifference in societies that
a re th ro w n in to turm o il by the d e a th o f a chief .35 It is h a rd to
believe th a t in G reece a fem ale slave w ould have been honoured
by th e sam e elab o ratio n o f fu n erary ritu al as the m aster of the
house, a n d it is tem p tin g to w o n d er w hether h er pow er to
p o llu te th e h o usehold m ay h ave been less too. T h e m any Greek
co m m u n itie s th a t to lerated the b urial o f children b u t not adults
w ith in th e settlem en t area p resu m ab ly felt th a t no g reat conta­
gion could pro ceed from such insignificant bones. F u rth e r than
this we c a n n o t go; th e w ritten sources tell us no m ore ab o u t this
th a n a n y o th e r asp ect o f th e d ea th o f the p o o r .36 T h ere is
evidence, how ever, th a t pollution m ight vary in intensity ac­
co rd in g to th e m a n n e r o f the d eath . No special ta in t attac h ed to
th e bed on w hich a m an died n atu rally , but, in a case of suicide
33 H d t. 6.58.1. P lut. Apoth. Lac. 238d, L ycurgus περιεΐλε τους μιασμονς, seem s to have
th e sam e sense; cf. Papyri from Tebtunis part / / , ed. E. M . H usselm an an d others,
M ic h ig a n a n d L o n d o n , 1944 (Michigan papyri 5), n. 243.11,244.17, m em bers o f guilds in
E g y p t in 1st cen tu ry AD fined if they refuse μιαίνεσθαι for d eath o f guild m em bers.
34 N eu m a n n , o p . cit., 86—9, cf. A esch. Cho. 2 2 -3 1 ; w earing o f soiled m ourning clothes
b a n n e d , L SA 16.6. F or self-defilem ent w ith d u st or m ud (com m on in H om er) cf. Eur.
Suppl. 826 f., Anth. Pal. 7.10. 3 - 4 (evidence from later antiq u ity in E. Reiner, Die rituelle
Totenklage der Griechen, T ü b in g e n , 1938, 43 n. 5); n o t washing, Eur. Or. 42.
35 H ertz, 76; m o re evidence in L. R. Binford, An Archaeological Perspective, New York
a n d L o n d o n , 1972, 220, 228 f., an d on children 234. O n child and slave burial in Greece
see B rem m er, a n d on dilFerential b urial G. B uchner an d L. C erchiai in G n o li/V ern an t,
2 7 5 -9 8 .
36 B u t no te D em . 47.70.
42 Miasma
by h an g in g , th e rope a n d the b ra n ch w ere destroyed or throw n
o u tsid e th e b o u n d arie s o f the city. T h e extra pollution here
obviously derives from th a t sam e m oral revulsion against
su icid e th a t cau sed p unitive m easures to be take against the
c o rp se .37 T h e body o f a soldier w ho died in b attle, by contrast,
scarcely p o llu ted the com rades w ho b u rn ed or buried him .
N o th in g is ever said on the subject in the m any relevant p a s­
sages in G reek h istorians, a n d X enophon describes him self
sacrificing to th e gods on the d ay after burying corpses too
p u tre s c e n t to be lifted from w here they lay .38 A t the m ost, the
su rv iv o rs m ay have purified them selves by w ashing. T h e dead
m a n ’s relatives, o f course, m ight well have ‘polluted them ­
selves’ on receiving th e news.
T h e ab sen ce o f pollution is h ere a m a tte r o f convenience, but
it m ig h t be m a d e in to a point o f ideology. In the classical period,
e x tra m u ra l b u ria l w as in m ost cities a strictly observed norm ,
b u t m an y com m unities set the tom bs o f their ‘founders’ and
‘sa v io u rs’ in th e agora itself .39 T h e connection betw een agora
a n d to m b p ro b a b ly persisted from the period w hen political
assem b lies o cc u rre d a t the grave m o n u m en t o f a heroic a n ­
cestor, o u tsid e th e settlem en t area, b u t by the historical period,
w h e n th e agora w as w ithin the city, special values m ust have
b een invoked in defence o f this b reach o f the rules. P lu tarch ’s
a c c o u n t o f th e d e a th o f A ratus, lib era to r of Sicyon, is revealing.
T h e S icyonians, th o u g h eager to bury him w ith the highest
h o n o u rs, w ere disconcerted by ‘a n old law forbidding burial
w ith in the w a lls’ to w hich ‘g re at superstition a tta c h e d ’. T hey
th ere fo re co n su lted D elphi, b u t on receiving a satisfactory
a n sw e r ‘sw itch ed from g rief to festivities, and clad in garlands
a n d w h ite clothes b ro u g h t his body u p to the city to the ac­
c o m p a n im e n t o f p ae an s an d choruses, an d choosing a con­
37 L S C G 154 B 33—6, P lut. Them. 22.2, H arp o cr. s.v. όξνθνμια. O nly a Diogenes
w o u ld e a t th e fruit o f a fig tree w here a m an had h anged himself, D .L . 6.61. Punitive
m easu res: A esch. 3.244, R ohde, 187 n. 33.
38 X en. Anab. 6.4. 9—13.
39 R. M a rtin , Recherches sur l'agora grecque, Paris, 1951, 194—201; cf. O . Broneer,
Hesperia 11 (1942), 128—61 ; G . V. L alande, H e sp e ria ^ (1980), 97 -1 0 5 . H ero tom bs left
d u rin g th e pu rificatio n o f Delos: p. 33 n. 6 above. N on-poiluting hero tom bs in Cyrene:
L S S 115 A 21—4. Political assem blies a t grave m onum ents: M artin, op. cit., 4 7 -5 6 . See
too N. S. R. H o rn b lo w er, Mausolus, O xford, 1982, 255 f.; C . B erard in G noli/V ernant,
2 7 5 -9 8 .
Birth and Death 43
sp icu o u s sp o t b u ried him like a fo under a n d sav io u r o f th e city’.
VVe m ay co m p are P lato ’s ac co u n t in th e Laws o f th e m agnificent
s ta te fu n erals w ith w hich his scru tin eers are to b e honoured.
D irges are fo rb id d en ; all p artic ip a n ts are to be dressed in w hite,
a n d choruses o f boys a n d girls will sing hym ns o f praise to the
d e a d m an . A t th e re a r o f th e fu n eral procession, if D elphi
agrees, th e p riests a n d priestesses are to walk, even th o u g h they
a re exclu d ed from all o th er burials; this one will not pollute
th e m .40 P la to is here ad o p tin g a n d p erh ap s ex p an d in g the
ideology o rig in ally developed for those who ‘show ed them selves
good m e n ’ by d e a th in th eir c o u n try ’s service. ‘T h e ir tom b is an
a lta r; in p lace o f lam en t they h ave rem em brance, g rief becomes
p ra is e ’, said Sim onides, a n d th e th em e was taken up in m any a
fu n e ral o ra tio n ; displays o f m o u rn in g in such cases m ight even
be co n tro lled by law. ‘H ere Saon sleeps a sacred sleep. Do not
say th a t good m en d ie .’41 By the qu ality of th eir lives, such
o u ts ta n d in g in d iv id u als have overcom e d eath itself. T h e ir p as­
sing, triu m p h ra th e r th a n defeat, cannot be m o u rn ed and
(P la to d ra w s th e consequence) can n o t be th o u g h t to pollute.
F o r th e m th e tired p latitu d e s o f ep itap h s - let no one lam en t my
p assin g - a re ac te d o u t in ritu al, a n d so the com m unity as a
w hole p a rtic ip a te s in th eir v icto ry .42 B ut w here an unw orthy
in d iv id u a l insists on being trea ted as non-polluting in d eath ,
p la g u e en su e s .43
T h e in d iv id u a l’s right to receive b urial was, o f course, sup­
p o rte d by pow erful social a n d su p e rn a tu ra l sanctions. T h e
‘co m m o n law o f th e G reeks’ agreed w ith the ‘unw ritten , un-
40 P lut. Arat. 53.2—4; PI. Leg. 9 4 7 b -d . Cf. (Plut. Tim. 39.3 on T im oleon’s funeral: όψις
μ έν ην εορτή πρέπο υσα, πάντω ν ^στεφανωμένων καί καθαράς έσθήτας φορούντων, idem , Pel.
33.5, p riests a t receip t o f P elopidas’ body.
41 S im onides, P M G 531.3—4, cf. Eur. I A 1437—48, 1466. Legal control: Z-SS64. 1 -4
(T h a so s, 5 th —4 th c. BC, cf. J . Pouilloux, Recherches sur Vhistoire et les cultes de Thasos 1,
P aris, 1954, 371—6). Saon: C all. Epigr. 9 Pf., 41 G /P : cf. G ow /P ag e’s note.
42 L ie n h a rd t, C h . 8 describes the rite w hereby aged D inka priests, carriers o f the ‘life’
of th e ir people, v o lu n tarily su b m itted to b u rial alive as a m eans o f publicly repudiating
n o rm a l physical d e ath . M o u rn in g w as forbidden. ‘F o r the rest o f the m aster’s p e o p le . . .
th e h u m a n sym bolic actio n involved in th e ‘artificial’ burial m ust be seen to transform
th e ex p erien ce o f a le a d e r’s d eath into a concentrated public experience o f v itality’
(317). L ie n h a rd t’s acco u n t is suggestive as a parallel for the passing ofG reek heroes like
O e d ip u s (m o u rn in g forbidden Soph. OC 1751-3) an d A m phiaraus. T h e ideology of
‘n o t m o u rn in g b rav e m en ’ perh ap s developed in S parta: cf. Plut. Ages. 29.7, m others of
L e u c tra d e a d th ro n g in g sh rin es in thanksgiving.
43 (A eschin.) Epist. 1.2.
44 Miasma
sh a k e a b le law s o f the gods’ in insisting th a t even the body o f an
en e m y sh o u ld be given u p after b attle for burial. T h ere was a
g en e rally recognized obligation for anyone w ho encountered an
u n te n d e d corpse - a drow ned m an on the seashore, for instance
- to perform a t least a token a c t o f burial; at A thens those who
n eg lec ted this m in im u m h u m an obligation were threatened by
o n e o f th e ‘B ouzygean curses ’.44 An un b u ried corpse was an
o u tra g e , a n d o n e possible consequence was pollution. Sopho­
cles, in th e Antigone, as we have seen, offered a rem arkable vision
o f th e form th a t this pollution took :45 scraps of the corpse,
d ro p p e d by birds o f prey on the altars, doused the sacrificial
fires, a n d do o m ed th e city to godlessness. No do u b t Greeks did
n o t n o rm ally co n sid er the n a tu re o f miasma so precisely, but it is
in te re stin g th a t w hen his d ra m a tic purpose forced Sophocles to
b e explicit he sh o u ld have given it so concrete a form. It is clear
th a t this, in context, is less a ‘n a tu ra l pollution’ th a n a cosmic
sa n c tio n o p e ra tin g a g ain st the violation o f a fundam ental social
p rin c ip le , th e in d iv id u a l’s right to burial. A drow ned m a n ’s
co rp se lying u n n o ticed b en eath the banks o f the Ism enus m ight
in th eo ry p o llu te the city no less severely; but it is scarcely an
a c c id e n t th a t th e one case o f pollution by a corpse w hich we find
vividly d esc rib ed derives from h u m a n crim e.
S o p h o cles’ p ic tu re o f the consequences of denying burial is
d ra m a tic a n d alarm in g . T h ese consequences, however, often
fail to a p p e a r in the contexts w here they m ight have been m ost
ex p e cted . In th e extended d ebates a b o u t burial in the Ajax and
E u rip id e s ’ Supplices, for instance, the law of the gods is brought
freely u n d e r a p p e a l b u t the th re a t o f pollution is never m en­
tio n ed ; w h en th e T h e b a n s in 424 refused to re tu rn the A thenian
d e a d , they w ere accused, according to T hucydides, o f illegality
a n d im p iety , b u t n o t specifically o f ‘polluting the gods’, an d in
d e c la rin g them selves only conditionally willing to re tu rn these
co rp ses th e T h e b a n s w ere obviously defying p o llu tio n .46 It is
44 L aw o f the gods: e.g. Soph. A j. 11301'., 1335, 1343, Ant. 77, 745, 749, E ur. Suppl. 19,
311, 526, 563, T h u c . 4.98.8. R equirem ent to bury: Soph. Ant. 255 + schol. (Bouzyges),
A el. V.H. 5.14, idem , fr. 242, N isb e t/H u b b a rd on H or. Carm. 1.28.23.
45 9 9 9 -1 0 1 5 .
46 T h u c . 4. 97.2—99; sim ilar T h e b a n conditions, X en. Hell. 3.5.24. For denial of
anairesis, ac tu a l o r th re a te n e d , cf. Diod. 17.25.6, Dem . 7.38. In G reek /b arb arian
re la tio n s the convention m ight not apply, H d t. 9.83, P lut. Tim. 25.5; cf. F. Jacoby, J H S
64 (1944), 42 f. T h e g en eral’s obligation to protect his own dead: Diod. 13.61.6 with
75.4, X e n . Hell. 1.7 (A rginusae generals), D iod. 17.68.4.
45
Birth and Death
n o t th a t, w hile ‘p o llu tio n ’ is ab sen t, ‘divine an g e r’ ag ain st the
v io lato rs o f th e divine law im pends. Such d eb ates can be con­
d u c te d in th e n a m e o f custom , ju stic e, a n d h u m an ity , w ithout
a n y serious ap p e al to the th re a t o f su p ern atu ral retribution.
T h e d iv in e o rigin o f th e law o f b u rial is a g u aran tee o f its justice
ra th e r th a n o f swift divine in terv en tio n in su p p o rt o f it. W hile it
m ay to som e ex ten t be tru e th a t th rea ts o f heav en ’s vengeance
a re sim p ly sq u eezed o u t by a su p erab u n d an c e o f h u m an indig­
n a tio n , it looks as if in th e late fifth century fear o f su p ern atu ral
p u n is h m e n t w as a ra th e r w eak an d rem ote d eterren t against
m a ltre a tm e n t o f th e corpse.
O n e reaso n for this m ay h av e been th a t the obligation to
g ra n t b u ria l w as never ab solute. It was not m erely in the
u p su rg e o f p o p u la r fury or sim ilar ungovernable passions that
ex cep tio n s o ccu rred ; it was the ‘com m on law o f the G reek s’, not
ex clu d in g th e A th en ian s, th a t tem ple-robbers a n d th eir like
sh o u ld be ‘th row n, o u t u n b u rie d ’, an d m any of the Antigone’s
a u d ie n c e will h av e voted for th e decree th a t declared ‘it shall
n o t be p e rm itte d to b u ry A n tip h o n in A thens or in the territory
th e A th e n ia n s co n tro l.’ N ot even P lato in the Laws w as p re­
p a re d to g ra n t all his citizens an inalienable right to burial in
th e ir n ativ e la n d .47 It is often suggested th a t the G reeks, in
c o n tra s t to m o d ern E uro p ean s, w ere abnorm ally sensitive to
th e fate o f th e ir corpses. T h e p resu m ed m odern indifference is
ex a g g era te d ; th e th o u g h t o f a relativ e’s body devoured by
scav en g ers is as hateful as it ever was, and the reason why this is
n o longer a h a u n tin g fear is p erh ap s m erely th a t it is m ost
47 P o p u lar fury: N ie. D am . 90 FG rH fr. 60, Diod. 16.16.4. (cf. Plut. Dion 3 5 .5 -7 ).
D y n a s t’s h a tre d : D iod. 17.118.2 (the possibility, Q uin t. C u rt. 7.2.32, 8.2.12).
O lig a rc h ic b ru ta lity : T h eo p o m p . 115 FGrH fr. 96 (H yperbolus throw n in sea), Lys.
12.21 ,?96. T y ra n t throw s offenders in sea as m eans o f terror, T heopom p. 115 FGrH fr.
227 (cl. R E S u p p l. 7.1605 f.). E nem y’s revenge: Plut. Nie. 28.5, cf. Dem. 29.5. 'C om m on
law a b o u t tem p le ro b b e rs’: Diod. 16.25.2, cf. 16.35.6 (throw ing in sea), A eschin. 2.142
(p u sh in g o v er cliff, cf. D em . 19.327, P aus. 10.2.4). T raito rs an d tem ple-robbers at
A th en s: X en . Hell. 1.7.22, cf. T huc. 1.138.6; subverters o f the second A thenian league,
S IG 3 147.62. T h e p u n ish m en t exercised a t A thens: T huc. 1.138.6 ( Themistocles),
(P lu t.) .V Orat. 8 3 3 - 4 (A n tip h o n ), Lyc. Leocr. 112-15 (P hrynichus), Plut. Phoc. 37.3—5,
L ys. 19.7, cf. H y p . 1.20,4.18 K enyon. T h o se long dead dug up and expelled: Nie. Dam .
loc. cit., T h u c . 1.126.12, Lyc. Leocr. 1 1 2 -1 5 , Plut. Dion 53.2, Alex. 77.2. Plato: Leg.
854e—855a, 873b, 909c. M ore in B rem m er. C ynics (e.g. S V F 1.253 )and E picurus (D. L.
10.118) d en ied the im p o rtan ce o f burial, b u t this was o f course not wholly new, cf.
H eraclit. B 96, E u r. fr. 176. O n the futility o f punishing corpses see Aesch. fr. 266,
N loschion, frr. 3, 7 Snell (cf. C Q 31 (1981), 417) and already? Horn. II. 24.54.
46 Miasma
unlikely to occur. T h e corpse is now sacrosanct, beyond the
re a c h o f re w ard a n d p u n ish m en t; even the bodies o f suicides are
o ften a d m itte d to consecrated g round, an d it is long since the
co rp se o f a d e b to r w as last subjected to d istra in t .48 T h e real
difference is th a t in the an cien t w orld the corpse enjoyed no
s u c h ex em p tio n . T re a tm e n t o f corpses rem ained one o f the
m e a n s by w hich m en could h u rt, hum iliate, o r honour one
a n o th e r, express co n tem p t or re sp ect .49 T h is is why the them e
co u ld be o f c e n tral im p o rtan ce in g re at works o f literature. It
w as th e p o te n tia l for hum iliatio n th a t was particu larly strongly
felt. In trag ed y , th e victim ’s relatives lam ent, not his exclusion
from th e u n d erw o rld , b u t his dishonour; ‘not to be buried in
A ttic a ’ is only the fam iliar p u n ish m en t o f ‘deprivation o f hon­
o u r ’ in its m ost ex trem e form. P lato is eager for his citizens to
realize th a t th e corpse is a valueless, unperceiving th in g ,50 bu t
c o n tin u e s d esp ite h im self to tre a t it as a vehicle o f honour in the
h u m ilia tio n s he inflicts on it.
I t looks as if, in the case o f the tra ito r a n d tem ple-robber, the
law o f th e gods en su rin g the right o f burial ceased to apply. O ne
m ig h t even co n clu d e th at, w ith th eir honour, they lost the
p o w e r to pollute. T h is w ould lead to the paradox th at, while no
fu n e ra l o r to m b is p u re except th a t o f the o u tstan d in g servant of
th e co m m u n ity , the only corpse th a t will not cause pollution if
left u n b u rie d is th a t o f the public enem y .51 T h ere are m any
G reek s w hom one could im agine m ain tain in g th a t this was
in d e e d th e case —A th e n s’ ferocious L ycurgus w ould be one —
b u t th e conclusion w ould go beyond the evidence. T h e A the­
n ia n s ‘th rew o u t’ th eir crim inals ‘u n b u ried beyond the bounds
o f A ttic a ’; this allow ed the relatives to arran g e for burial, and
48 C l. M rs H e n ry W ood, East Lynne, L ondon, 1861, P art 1, C h. 10, for an instance of
d is tra in t. F o r A th en s see Diod. 10.30. J a n B rem m er rem inds me th a t m odern corpses
a re used for m edical research - b u t they have to be d o n ated for the purpose. Requisi­
tio n in g w ould be u n th in k ab le.
49 F or H o m e r an d the an cien t n ear east see Griffin, 4 4 - 7; on H om er, J . P . V ern an t in
G n o li/V e rn a n t, 4 5 -7 6 .
50 Leg. 959c.
51 L ogically th ere is a p o ten tial tension betw en corruptio optimi pessima an d corruptio
optimum non attingit. Z o ro astrian s an d H in d u s th u s arriv e a t opposite conclusions as to
w h e th e r the d e a th o f a person o f high o r low sta tu s pollutes m ore (Zoroastrians: The
Zend Avesta, p a r t 1, The Vendidad, trans. J . D arm esteter, O xford, 1880, F arg ard 5, §6,
27—38; H in d u s: S. J . T a m b ia h in J . G oody (ed.), The Character o f Kinship, C am bridge,
1 9 7 3 ,2 0 9 -1 2 ).
Birth and Death 47
even if th ey d id not, the pollution w ould a t least fall outside
A ttic territo ry . O th e r states’ m odes o f m altreatin g th e corpse -
ih ro w in g it over a cliff, or into th e sea —w ere pro b ab ly intended
to exclu d e all possibility o f b u rial, b u t they did serve to dispose
o f th e rem ain s aw ay from h u m an h a b ita tio n .52 T h e in ten tio n of
th ese m eth o d s will scarcely h av e been to evade pollution —they
w ere ra th e r a sym bolic rejection o f th e m alefactor - b u t they
p ro b a b ly h ad th e effect th a t th e q u estion of pollution did not
need to be raised. Prolonged p u blic exposure o f the corpse, as
p re scrib ed by C reo n in Antigone, w as n o t the p ractice o f any
G reek state, a n d w h en m en tio n ed is treated as shocking .53 T h e
law o f th e gods gave w ay before society’s ab om ination o f certain
o f its enem ies; po llu tio n m ight have do n e so too, b u t the evi­
d e n c e does n o t tak e us so far .54
T h e Antigone a t first sight p resents an exception. Polyneices is
a tra ito r; C reo n exposes Polyneices; the gods are angry, and
p o llu tio n resu lts. T h e law o f the gods has refused to give way. As
w e n o ted , C re o n ’s trea tm e n t o f Polyneices’ corpse differs from
th e A th e n ia n p ractice, b u t this o b servation scarcely provides a
so lu tio n : few will be convinced th a t C reo n ’s tragic erro r lay in
leav in g th e body on the plain, a n d th a t all w ould have been well
h a d h e carried him off to be eaten by the birds beyond the
b o u n d arie s. T h o u g h the p a rtic u la r m ode of hum iliation chosen
by C reo n is an ag g rav atin g factor, it is the ju stic e o f giving
P olyneices an y th in g o th er th an a p ro p e r burial th a t is in dis­
p u te .55 It need not follow, how ever, th a t the rights vindicated
for th e h ero Polyneices belonged also to a sham eless pilferer of
52 S ee p. 45 n. 47. B u rial occurs outside A ttica (although no A thenian m ay p artici­
p a te ) Plut. Phoc. 37. 3 - 4 , cf. PI. Leg 909c. A t A thens, ‘throw ing into th e p it’ was
p rim a rily a m ode o f execution (see e.g. G ernet, Anthropologie, 3 0 8 -1 1 ; L atte in R E
S u p p l. 7. 1608 f. is hyper-sceptical). T h ere is no evidence th at recovery o f the body by
relativ es w as legally forbidden, although it m ay no t have been practically feasible. T h e
M S S read in g in X en . H e li 1.7.20 indicates a d istinct use of the pit for post mortem
ex p o su re, b u t sh o u ld p ro b ab ly be em ended (cf. J . Diggle, CÄ31 (1981), 107 f.). Even if
co rp ses w ere left in th e pit, they were at least out o f the way. C orpses were throw n o u t at
Ά λ μ ν ρ ίό ε ς (H esych. s.v., cf. A r. fr. 132); we d o not know whose. Cf. too Plut. Them.
2 2 .2 .
53 P lu t. Nie. 28.5 ( = Ί 'im aeus 5(i6 F G rH fr. 101), Phylarchus 81 FGrH fr. 45 ap. A th.
52 ld .
54 C h a rito n 1.5.5 offers a novel tw ist, unfortunately not attested elsewhere: μή θάψητέ
μ ε , μ ή μιάνητε τήν γήν, αλλά τό άσεβϊς καταποντώοατε σώμα.
55 R ightly (ag ain st B ow ra an d E hrenberg) G. C erri in G n o li/V ern an t, 121-31.
48 Miasma
sa c re d plate. Sophocles m ay o f course have believed, in advance
o f his age, th a t p unitive action against the corpse is in all
circ u m sta n ces a n outrage, b u t the m oral prem iss o f the play
d o es no t seem to be th a t ‘even traitors are hum an, and deserve a
m in im u m o f re sp ect’. N othing encourages us to view Polyneices
in this lig h t.56 Polyneices is a hero who has led an arm y to
defeat, a n d C re o n ’s offence is the fam iliar one o f denying an
en em y th e rig h t to burial: this is w hy the corpse rem ains on the
b attlefield w here it fell. It is the treatm en t o f enem ies and not of
villains th a t is in question.
W e tu rn now from d ea th to b irth .57 A ccording to C ensorinus,
p ro b a b ly echoing V arro, ‘in G reece they treat fortieth days as
im p o rta n t. F o r th e p re g n an t w om an does not go out to a shrine
before th e fo rtieth d ay . . .’ (from the m om ent th at she becomes
a w a re th a t she is p re g n an t?).58 A ritu al exclusion o f forty days
so u n d s m ore S em itic th an G reek, but in Greek m edical texts the
fo rty -d ay perio d is o f p a rtic u la r im portance precisely in relation
to p re g n a n c y a n d b irth ; d u rin g the first forty days after concep­
tion, for in stan ce , m en stru atio n continues, an d m iscarriage is a
c o n s ta n t d an g e r, w hile by the end o f this period the em bryo is
fo rm ed a n d the m ale child begins to m ove.59 T h e dangerous
tra n s itio n a l p erio d therefore lasts forty days, an d d u rin g this
p e rio d , if C en so rin u s is right, the m other is excluded from
c o m m u n a l life.60 Seclusion a t the onset of pregnancy is a widely
56 See esp. v. 731. N o t even C reon is ever allowed to use the word ‘traito r' of
Polyneices. F o r a different view see C erri, op. cit., and S. Fuscagni in M . Sordi (ed.).
Religione e politica nel mondo antico, M ilan, 1981, 6 4 -7 2 . In M oschion’s Pheraioi, by
c o n tra st, the rig h t o f b u rial w as perhaps vindicated even for a tyrant: see C Q 31 ( 1981 ),
417.
57 O n b irth -p o llu tio n see M oulinier, 6 6 -7 0 , G inouvès, 2 3 5 -8 , W ächter, 2 5 -3 6 .
58 De die natali 11.7.
59 W . H . R oscher, ‘D ie T esserak o n tad en und T esserakontadenlehre der G riechen
u n d a n d e re r V ö lk er’, Ber. Sächs. Ges. Wiss. 61.2 (1909), 2 8 -3 4 , 40, 8 5 -1 0 1 : sec esp.
C e n so rin u s, loc. cit., A rist. Hist. An. 7.3. 5 8 3 a2 7 -5 8 3 b l5 . O n w ide diffusion of
gynaecological forties cf. G. E ichinger Ferro-Luzzi, Anthropos 69 (1974), 148-52.
V iew s on th e tim in g o f these m atters w ere how ever very' various in G reece, cf. E. N ardi,
Prciurato Aborto nel mondo greco romano, M ilan, 1971, 9 3 -1 1 5 , 123-32.
6,° T h e 2 p eriods w ould not o f course coincide, as m ost of the 40 days after conception
w ould p ass before th e m o th er perceived her pregnancy. P erhaps in o rder to bring them
in to coincidence, R oscher, op. cit., 28, 30, d a te d the m other's seclusion from “the
w e d d in g a n d co n cep tio n ’. B ut we would surely have heard of a 40-day seclusion
im m ed iately follow ing th e w edding: an d w hat o f subsequent pregnancies?
Birth and Death 49
a tte ste d rite o f p assa g e ,61 an d , a t th e first pregnancy, the Greek
w o m an m ig h t have been exposed to th e an g er of A rtem is .62
T h e re is, how ever, no trace o f C en so rin u s’ rule in any other
source, a n d he, o r even V arro , w rote at a tim e w hen Greek
p u rity rules w ere not necessarily restricted to ancient norm s.
L a te r in p reg n an cy , the w om an was not excluded from tem ­
ples b u t positively expected to visit th e m .63 W e do hear, how­
ever, o f a few sacred rites an d places forbidden to her, and there
m ay well h ave b een m an y m ore. P reg n a n t w om en, a n d suckling
m o th ers, w ere excluded from th e m ysteries of D espoina at
L ycosu ra, w hile ‘no anim al o r w om an in need o f E ileithyia’
w ould v en tu re u p o n the hill in A rcad ia w here R heia gave b irth
to Z e u s .64 T h e se are clearly instances of th at sam e logic of
o p p o sitio n w hich som etim es req u ired chastity for the service of
A p h ro d ite; it is from rites o f fertility, a n d a m ythical birthplace,
th a t those a b o u t to give b irth are d eb arred . Except in these
special contexts, the p reg n an t w om an w as not herself polluting,
b u t it is in tere stin g th a t she was p articu larly exposed to pollu­
tion by others. W 'hen E u rip id es’ Iphigeneia is ab o u t to lead the
p o llu te d O restes th ro u g h the streets, she issues a special w arn ­
ing to th ree categories of person —to priests, p reg n an t women,
a n d those a b o u t to m a rry .65 T h e p reg n an t w om an is, as we
w ould say, d elicate, b u t it is to ritu al dangers th a t this delicacy
re la tes, a n d , as the parallel w ith those ab o u t to m arry shows, it
is from th e ch an g e she is ab o u t to undergo, and not h er present
physiological condition, th at her delicacy derives. T hese two
classes o f p eo p le are exposed to pollution because they are on
th e brink.
O n th e consequences of b irth, the earliest explicit evidence
com es from th e C y ren e c a th a rtic law .66T h e text is fragm entary,
b u t it p ro b a b ly specified th at th e m o th er only polluted those
w h o en tered th e sam e roof u n d er w hich she lay; it certainly
61 v an G en n ep , C h . 4.
62 Cf. schol. T h eo cr. 2.66b, L. D eubner, J D A I 40 ( 1925), 211 f. The evidence o f LSS
115 B 1 -2 3 is relev an t to this whole section, b u t so obscure th at it m ust be left to
A p p en d ix 2.
63 A rist. Pol. 1335b 12-16.
64 L S C G 68.12—13, C allim . Jov. 11 —13. C allim . Dem. 130—2, at least as presented by
th e poet, is different: a concession an d not a rule.
65 I T 1 226-9.
66 L SS 115 A 16—20, B 2 6 -7 , cf. A ppendix 2.
50 Miasma
ru le d th a t a n y o n e w ho did so was im pure for three days, but did
n o t p ass th a t im p u rity on fu rth er. C o n tact w ith a new m other is
a n o rm a l source o f pollution in sacred laws, an d three days is
a b o u t th e av erag e period o f exclusion ,67 but the C yrene law is
iso la te d a n d v alu ab le in the specification th at pollution is incur­
re d by ‘e n terin g th e sam e ro o f’ a n d not ‘touching’. A nother
p assa g e o f th e sam e law m akes b irth-pollution pollute ‘the oikia
its e lf’. Oikia could be in terp re ted eith er physically, which
w ould m ake it synonym ous w ith ‘the roof’ o f the previous
re g u la tio n s, o r socially, the household; the second in ter­
p re ta tio n in tro d u ce s a theoretical inconsistency w ith the other
ru le, as a m e m b e r o f the household could if he w ished stay away
from th e physical house d u rin g the period o f im purity, b u t in
p ra c tic e b o th rules w ould no d o u b t leave the sam e individuals
p o llu te d . (L et us no t suspect the G reek h u sb an d o f exploiting
his w ife’s p o llu tio n as an excuse to shun the house during a
try in g tim e.) B irth seem s to have polluted a m ore restricted
circle th a n d e a th ; th e fu rth e r kin w ere certainly not affected
au to m a tic a lly , a n d there is no evidence th at they w ere expected
to in c u r p o llu tio n by visiting the house d u rin g the time of
im p u rity . T h is reflects the fact th a t d ea th s (and m arriages) pull
to g e th e r the social g ro u p m ore insistently th an b irth s .68
The first b a th o f m o th er a n d child was an im p o rtan t occa­
sion. T h e story o f a g o d ’s b irth was scarcely com plete w ithout
m e n tio n o f it, a n d even for m ortals the w ater m ight be fetched
67 T h o se w ho com e in to co n tact w ith the m o th er recover their pu rity on the 2nd day
(L S A 12.7), on th e 3rd d ay (LSA 5 1 .6 - 1 0 ) , after 3 days (ZÄS9I.15; ? LSC G 124.7; cf.
L S C G 154 A 24, 39; 156 A 13), on the 7th day (LSS 54.5, an oriental cult); in LSS 119.6
th e p eriod is lost. B C H 102 (1978), p. 325, line 6 (Isis) seems to attest the 9th day, but
th e reference m ig h t be to th e m o th er herself; for this a n d the sim ilar problem concern-
ing L SC G 171.16 (10 days) see A ppendix 3. F or the m other polluted longer th an others
see L SC G 1 2 4 .5 -8 , L S S 91.15. In LSA 5 1 .6 -1 0 a dog th at gives b irth pollutes like a
w o m an , no d o u b t because dogs share a roof w ith m en (sim ilarly cats in m odern Greece.
B lu m 4 7 f.; in L SS 91.11 m iscarriages o f w om an, dog, an d donkey all pollute alike). In
L S S 115 B 2 4 - 7 m iscarriag e pollutes like d e a th if the foetus is ‘distinguishable', i.e. has
reco g n izab le lim bs, a n d , if not, like birth. (Views varied as to when this articulation
sh o u ld occur, cl. p. 48 n. 59. A ristotle advocated early abortion before the ad v en t of ζωή
a n d αισθησις, Pol. 1335b 24—6). In LSCG 154 A 24, ? 39 m iscarriage pollutes like birth;
th e re sto ratio n w hich m akes it pollute like de a th in th e closely com parable LSCG 156 A
13 is therefore q u estio n a b le . For the pollution o f m iscarriage in later sacred laws
see A p p en d ix 3.
68 A rist. Elh. Nie. 1165a 18-2 1 . H usband: cf. S. Beckett, Compatir. 16.
Birth and Death 51
from a special s p rin g .69 As is often the case, how ever, the act
o f physical w ash in g was not sufficient to re-establish purity.
T h e d etails o f th e fu rth e r rites th a t followed the b irth are an
u n h a p p y tan g le o f conflicting an d deficient lexicographical evi­
den ce, b u t it is p ro b ab ly right to d istinguish two m ain cere­
m onies, th e amphidromia on the fifth d ay a n d the nam e-giving on
th e te n th .70 A t th e amphidromia, th e child was carried at a run
a ro u n d th e h e a rth . D etails are u n certain , but the m ain point of
th e cerem ony w as p ro b ab ly to b rin g th e new m em b er into
c o n ta c t w ith th e h o u seh o ld ’s sacred centre, at w hich brides and
new ly -b o u g h t slaves w ere also p resented. Ideas o f purifying the
ch ild by fire m ay also have been p resent, although th e sources
do not say so .71 O n th e sam e day, p ro b ab ly as a p a rt o f th e sam e
cerem o n y , th e w om en who assisted at the birth are said to have
b een p urified, b u t n o th in g reliable is recorded ab o u t the
m e th o d .72 T h e m o th e r had p resu m ab ly ceased by this tim e to
be an active source o f pollution, or all those who atten d ed the
fifth-day cerem onies w ould have gone aw ay infected; certainly
on C os even priests w ere allowed to ven tu re in four days after
th e b ir th .73 N o th in g is recorded o f any special ritu al for the
p u rifica tio n o f th e house. T h e m o th er p erh ap s en tered the state
o f ‘im p u re, b u t not p o llu tin g o th ers’, a n d only retu rn ed fully to
p u rity a n d n o rm al life w ith the nam e-giving and sacrifice on the
te n th day. E ven after this the cycle o f read ju stm en t pro b ab ly
h a d som e w ay to ru n . J u s t as d ea th was followed not ju s t by
69 G ods: Hymn. Horn. Ap. 120 -1 , C allim . Jov. 14-16, Paus. 8.28.2, 8 .4 1 .2 -3 ,
G inouvès, 235, 238. S pecial spring: e.g. C allim . fr. 65. T h is bath occurred im m ediately;
D o n a tu s on T e r. Andr. 483.3 ( = M en. fr. 36) attests a postponed b ath post puerperium,
b u t a p p a re n tly for R om e ra th e r th an G reece.
70 See m ost recently L. D eubner, Rh. Mus. 95 ( 1952), 374—7.
71 P resen tatio n : F ustel de C oulanges, La Cité antique1*, Paris, 1919, 54 (w ith the
lu stra l ex p lan atio n ); N ilsson, G F 115 f. P urification by fire: R ohde, 318 n. 72; J . G.
F ra z e r, A p p en d ix 1 to Loeb A pollodorus. An ordeal: O . G ruppe, Bursian Jahresb. 137
( 1908), 342 f.. B .p h il. Woch. 26 ( 1906), 1137-9, stressing Pl. Tht. 160e. Decision b y genos
w h e th e r to re a r child: G lotz, 41. Cf. fu rth er L. D eu b n er in J . H astings (ed.),
Encyclopaedia o f Religion and Ethics, E dinburgh, 1909, s.v. Birth ( Greek and Roman); Eitrem ,
Opferritus, 173—7; V e rn a n t, Pensée, i. 158—64; G . S. K irk in Entretiens Hardt 27, 5 6 -6 1 .
72 P urificatio n o f w om en: schol. PI. Tht. 160e, Suda s.v. amphidromia, A postolius 2.56.
Schol. Pl. Tht., loc. cit. m akes the w om en do the running, and be purified thereby, but
in A p o st., S u d a, an d H arp . s.v. amphidromia, an d by im plication Pl. Tht. 160e, the
ru n n e rs are m ale.
73 L S C G 154 A 24, 39, ? 156 A 12-13. N ote how ever, th e Superstitious M a n ’s refusal
to go έπΙ λεχώ, for fear o f p o llution, T h eo p h r. Char. 16.9.
 Miasma
th ird - a n d n in th - b u t also by th irtie th -d ay rites, so we hear of a
jo y fu l ‘fo rtieth -d ay festival’ after b irth . Pollution in both cases
coincides w ith the intense early period o f the gradual process of
a ssim ila tin g ch a n g e.74
A s several instances have already shown, sacred persons
w ere re q u ire d to hold them selves a t the sam e distance from the
n a tu ra l pollu tio n s as the gods them selves. Any contact with
d e a th m ig h t im p a ir the w holeness necessary for divine service.
I n M essene, it is said, if a p riest o r priestess lost a child, he or she
w as forced to re n o u n ce the office; amphithaleis, children both of
w hose p a re n ts w ere still alive, h ad an im p o rtan t ritu al role.75
T w o C o a n in scrip tio n s th a t prescribe rules of p urity for certain
local priests a n d priestesses are largely concerned w ith keeping
th e m free from the ta in t o f b irth a n d d e a th .76 It is m ost plaus­
ible, a lth o u g h no t strictly d em onstrable, th at all C oan priests of
im p o rta n t cults w ere subject to sim ilar restrictions. A priest or
priestess m ay n o t e n te r a house o f d ea th until five days after the
c a rry in g o u t, m o u n t a hero shrine, or eat of the offerings m ade in
h ero ic o r c h th o n ic cult; if he discovers the corpse of a suicide, he
c a n n eith e r ignore it nor touch it, an d m ust sum m on a passer­
b y .77 A house w h ere a birth o r m iscarriage has occurred he m ust
av o id for th re e days. D etails o f dress a n d diet are influenced too.
T h e priestesses o f D em eter seem to be forbidden to w ear clo­
th in g m ad e from d ead anim als, o r to eat m eat slaughtered in a
p a r tic u la r w ay (p e rh ap s stra n g le d ).78 O n e of the sam e C oan
74 10 d ay s for th e m o th e r seem to em erge from E ur. El. 654 w ith 1124-1133.
P ro b a b ly 10 d a y s for m other, 3 for others in LSCG 124.7-8 (Eresus, ? 2nd c. BC. cf.
p. 355), 21 d ay s for m o th e r in the late L SS 91.15. 40th - day festival, C ensorinus ΰ ..V.
11.7 (no ea rlie r atte s ta tio n ). Sacrifice by a λεχώ, LSCG 77 D 13, by wom en ‘walking out
th e lochia', LSA 52 B 10.
75 P a u s .4.12.6, N ilsson, GGR 118w ith bibliography.
76 L SC G 154 A 2 1 - 4 5 , 156 A 7 -1 6 : interpreted by R. Herzog, A R W 10 (1907),
4 0 0 - 15, idem ., ‘H eilige G esetze von K os’, Bert. Abh. 1928.6, 17, 2 0 -5 . T hey concern
p riestesses o f D em eter an d the priest o f Z eus Polieus, but the inscriptions to be set up in
v a rio u s o th e r sa cre d places ( LSCG 154 A 16-18) presum ably contained rules for the
p rie sts o r priestesses o f the relevant cults. F or such rules cf. Phrynichus comicus, Ir. 70a
ό' άνάγκα ’αθ' ίερεϋσιν καθαρενειν, φράσομεν.
77 C o rp se o f a suicide: LSC G 154 B 33—6, well interpreted by Herzog with reference to
S erv. Aen. 6. 176: cum pontificibus nefas esset cadaver videre, magis tamen nefas fuerat si visum
insepultum relinquerent.
78 C lo th in g m ad e from d e a d anim als: θνηαείόιη, cf. passages in L SJ s.v., Pl. Leg. 956a
έλέφ α ς ôt άπολελοιπότος ψνχήν σώματος ούκ εύα γίς ανάθημα, LSCG 124.14, 17 θνασίόια
a n d skiiis b a n n e d , L SC G 65.23 leather san d als perm itted only if m ade from sacrificial
a n im a ls (d e a th by sacrifice no longer counts as d eath, X en. Λ nab. 4.5.35); tanner
Birth and Death 53
in scrip tio n s gives an exam ple o f th e purification th at m ight be
re q u ire d if th e ta in t o f d ea th did to uch sacred ground: the g o d ’s
s ta tu e is, it seem s, carried o u t o f th e p recinct to be w ashed, a
sacrifice o f p ro p itiatio n is offered, a n d the whole shrine is
p u rifie d .79 C h ristia n ity takes control o f the n atu ra l processes,
a n d p a rtic u la rly o f d eath ; th e funeral rites are conducted by a
p riest, a n d it is beside the ch u rch th a t th e body is laid to rest.
G re ek religion rejects it w holly .80
Before co nsidering the in terp re tatio n o f these pollutions, wc
m u st ask w h a t they entailed in practice. T h e only attested
co n seq u en ce o f b eing polluted is th a t o f exclusion from the
tem ples. It m u st have involved m ore th a n this, or the in ter­
m e d ia te co n d itio n know n from Iulis o f ‘p ure, but excluded from
th e tem p les’ w ould not differ from it, b u t the additional elem ent
c o u ld h ave been no m ore th an the p ro p e rty o f passing pollution
on to oth ers, a n d thus ren d erin g th em in tu rn unfit for access to
sh rin es. O f the m o re rigorous restrictions attested in m any
societies n o th in g em erges in o u r sources. W e cannot know, for
in stan ce , w h e th e r a m an subject to funerary pollution would
e n g ag e in ag ricu ltu re, o r a w om an w ho h ad assisted a t a lying-in
c o u ld go h om e an d cook for h er household. ‘H esio d ’ dis­
co u rag ed in terc o u rse after a funeral, a n d some m ay have
h ee d ed him , as th ere is later evidence th a t contact w ith d eath
w as felt to en d a n g er th e rep ro d u ctiv e processes; special restric­
tions m ig h t be placed on a tten d a n ce at funerals by w om en of
c h ild b e a rin g age. T h ey seem to have shared, in atten u a te d
form , th e ‘d elicacy ’ o f th e p re g n an t w om an. A plausible guess
m ig h t be th a t those w ho took pollution seriously would stay at
im p u re , A rtem id . 1.51, p. 59.4 Pack; for Rom e cf. O v. Fast. 1.629 f., T L L s.v . morticinus.
B are feet are often required for ritual, b u t not necessarily for this reason
(J. H eck e n b ach , De nuditate sacra sacrisque vinculis, G iessen, 1911, ( R G W IX . 3). 2 3 -3 1 ).
R e sto ratio n o f th e forbidden food ]ktwv in LSCG 154 Λ 27 is problem atic: πνικτών in the
N ew T estam ent sense o f ‘stran g led ’ is im plausible, as it has a different sense in
G reek cuisin e (L S J s.v .); for o th er suggestions see Herzog, o p cit., 23.
79 L S C G 154 B 24—32: cf. E ur. /T 1 0 4 0 f., 1176—7, 1199—1201 ; in the L indian tem ple
record, 532 FG rH D (2), the tem ple roof is rem oved for 3 days to adm it purifying rain;
for D elos see p. 33 n. 6 above.
80 N o priests at funerals, PI. Leg. 947d, ? E rin n a v. 19 in Page, GLP 488 (cf. Ci. M .
B ow ra in Greek Poetry and Life, Essays presented to Gilbert Murray, O xford, 1936, 334).
P o rp h . Abst. 2.50. The lau rel, sacred p lan t par excellence, not used at funerals, C allim fr
194. 3 9 -4 3 .
54 Miasma
h om e, ‘engage in no serious u n d erta k in g ’, an d avoid exposing
them selves to persons in delicate ritual conditions, w hile the
m o re casual w ould sim ply keep aw ay from the tem ples .81 It is
also no t clear how far the chain of pollution m ight extend away
from th e original co n tam in atin g object. T he only evidence on
th e p o in t is th e section o f the C yrene law w hich, in the case of
b irth , d eclares th e chain broken after the first link: ‘the person
in the house shall be polluted himself, b u t shall not pollute
a n y o n e else, w h erev er he goes ’.82 W e never hear of pollution
c o n tra c te d a t several rem oves, b u t it is im possible to prove that
th e C y ren e re g u latio n is typical or ancient.
I t is in ev itab le th a t, given the ch aracter of o u r sources, we
know even less ab o u t the em otional th an the practical im plica­
tions o f these pollutions. It does not seem to be the case that the
objects a society declares im pure necessarily evoke a response of
p a rtic u la r fear or revulsion in its m em bers, or th at disgusting
th in g s are a u to m atically im p u re .83 A pollution attaching to a
tru ly d isg u stin g object m ay be extended by logical elaboration
to so m eth in g q u ite innocent; the m ark of the corpse is set upon
le a th e r shoes. T h e re is, as we have seen, som ething impure
a b o u t th e tom b; b u t in vase paintings m ourners seem to
a p p ro a c h it in a m ood ra th e r o f sentim ental pilgrim age than
terrified p ro p itia tio n .84 It is unlikely th at the w om an who had
j u s t en su red th e survival of a house by bearing an heir felt much
d e g ra d e d by h er pollution. Some G reek texts do indeed speak of
corpses as repulsive, a n d tending them as dirty w ork ,85 but we
d o no t know th a t th e w om en who helped at a birth had to steel
them selves for th e task.
T h e first p relim in ary to an interpretation should be to ask
w h a t th e G reeks them selves felt the pollutions of birth and
d e a th to be. T h e y clearly saw the two pollutions as sim ilar, since
81 P ure b u t excluded from tem ples: LSCG 97 B 7 -1 1 . D eath and reproduction: p. 70
n. 123 below. P lausible guess: conflated from Arist. Ath. Pot. 56.4, Plut. Ages. 29.6
(s ta y in g a t hom e), X en . Hell. 1.4.12, Eur. /7 Ί 2 2 7 —9.
82 L S S 115 A 1 7 -2 0 .
85 Boyce, C h . 5, passim, m uch the best account 1 know of w hat pollution feels like.
N ote too P. M. K ab erry , Aboriginal Woman, Sacred and Profane, London. 1939. 238—ID
(m e n stru a tio n d an g ero u s ra th e r than disgusting).
84 See D. C . K u rtz , Athenian White Lekythoi, O xford, 1975, Plates 18-22 .nul iii.ui>
o th e rs, C . S ittl, Die Gebärden der Griechen und Römer. Leipzig. 189(1. 71 (kisses lilimn .11
grav e).
85 E ur. Supp. 767, Pl. Resp. 439e, Arist. Poet. 1448b 12.
Birth and Death 55
th ey co n stan tly spoke o f them to g eth er, and this conjunction
c o n d e m n s a n y ex p lan atio n based m erely on em otional re­
sponse; even H e ro d o tu s’ T h ra c ia n s, w ho w ept at b irth s and
ce le b rate d d eath s, recognized th a t in term s o f ap p ro p riate em o­
tio n al response th e tw o events have n o th in g in com m on. A t one
p erio d , im p ressed by the first revelations o f com parative an ­
throp o lo g y , sch o lars did not d o u b t th a t pollution was synonym ­
o us w ith th e presence o f an evil d em o n ;86 u n fortunately, the
sw arm in g nam eless dem ons to w hich they m ade co n stan t ap ­
p eal scarcely a p p e a r in the G reek sources before N eoplatonism .
E ven if th e an im ist in te rp re ta tio n h ad been b etter founded in
th e evidence, it w ould have rem ain ed to ask w hy certain o ccu rr­
ences, a n d n o t oth ers, should have let loose a sw arm o f these
d em o n s in to th e w orld. T h e early texts trace these pollutions to
m o re co n crete origins: again a n d again, they refer to ‘the corpse’
a n d ‘th e w o m an w ho is lying in ’, a n d in Sophocles, as we have
seen, d e a th -p o llu tio n is n o th in g o th er th a n scraps o f the corpse.
T h e corpse ro tted a n d the w om an bled; once the corpse was in
th e g ro u n d , th e first purifications could be perform ed, while the
G re ek w o m an p ro b ab ly rem ain ed im p u re for ten days after
b irth , a b o u t th e tim e for w hich p o st-p artu rie n t bleeding w ould
n o rm ally co n tin u e. (Like m en stru atio n , lochial bleeding was, in
scientific lan g u ag e, a ‘p u rificatio n ’).87 T h u s the m etaphysical
miasma ra d ia te d o u t from a physical centre. In the sam e way,
th e im p u rity o f sexuality was caused by sem en, while the pollu­
tion o f th e m u rd e re r was expressed in the im aginary stain of
blood on his h an d s. T hese facts w ere noted by T ylor, w ho was
in te re ste d in p rim itiv e lustratio n s because o f th eir survival, in
form s su ch as sp rin k lin g w ith holy w ater, into the form al religi­
ous p ra ctice o f his ow n day. H e saw this as a process w hereby
86 e.g. A rb esm an n , Fehrlc, W ächter, E itrem ( Opferritus, 97), an d , influentially,
R ohde. Not ju s t p rim itives explain the inexplicable in term s o f dem ons, in despite of
evidence. C o n tra s t H e rte r, Dämonen.
87 L S J s.v. κάθαρσις l i a . T h e scientific texts m ake this purification last m uch longer,
42 d a y s fo ra girl, 30 fo ra boy, H ippoc. Nat. Puer. 18 — M ul. 72 (7.500,8.152 L .); 30 fo ra
girl, 40 for a boy, A rist. Hist. An. 7.3 583a 3 0 -3 2 ; 40 days, Censorinus D .N. 11.7. T hese
figures obviously relate to the total period o f lochial discharge (in fact, norm ally 3 -4
w eeks); m odern doctors distin g u ish w ithin this an initial period o f ab u n d an t bleeding
(th e su b seq u en t d isch arg e scarcely contains blood), for w hich, m edical colleagues tell
m e, 10 d ay s w ould be a reasonable outside estim ate. I am suggesting th at th e ritual
im p u rity is based on this initial period, but ad m it th at the scientific texts d o not
recognize th e distin ctio n .
56 Miasma
o rig in ally p ra c tic a l m easures o f hygiene were ritualized, ren ­
d e re d sym bolic, a n d so fossilized an d preserved.
I t is th e p la in e s t p ro o f o f the o rig in al p rac tica lity ol p roceedings now
p a s s e d in to fo rm alism , to p o in t o u t how far th e cerem onial lu stratio n s
still k eep th e ir co n n ex io n w ith tim es o f life w hen real p urification is
n e c e ssa ry , h o w fa r they still co n sist in form al clean sin g o f the new ­
b o rn ch ild a n d th e m o th e r, o f the m a n sla y e r w ho has shed blood, or
th e m o u rn e r w h o h a s touched a co rp se .88
It is, how ever, precisely the ‘passage into form alism ’ th at re­
q u ire s ex p lan a tio n . W hile m u n d an e d irt yields to w ashing, it is
only after a fixed period o f tim e th a t pollution can be w ashed
aw ay. D irt does n o t discrim inate, b u t pollution is liable to afflict
a d e a d m a n ’s relatives m ore th an outsiders, an d the status o f the
d ec ease d m ay influence the intensity o f the pollution. T ylor
h im se lf q u o te d exam ples o f peoples w ho practised elaborate
cerem o n ial lu stratio n s, alth o u g h conspicuously indifferent, in
th e ir daily lives, to w h a t he regarded as the sim plest principles
o f cleanliness a n d hygiene. ‘T h e D a rd an ian s of Illy ria’, Greek
e th n o g ra p h y n o ted w ith interest, ‘only take three b ath s in their
lives: w hen th e y ’re b o rn , w hen they m arry, and w hen they die.’
C lean lin ess is often a n im p o rta n t p a rt o f purity, but a dirty robe
m ay be ritu ally far p u re r th an a clean o n e .89 In draw ing a tte n ­
tion to the d irtin ess o f the im pure, T ylor was p erhaps indicating
n o t, as he su p p o sed , the real basis o f such im purity, bu t the
co n c rete vehicle th ro u g h w hich m ore ab stra ct realities were
conveyed. M a ry D ouglas has suggested th a t a society m ay use a
su p p o se d physical im purity as an unconscious sym bol upon
w h ich it focuses fears or concerns o f a m uch b ro ad er social
c h a ra c te r .90 I t is obvious, for instance, th a t in G reece the blood
on th e m u rd e re r’s h an d s clings th ere because of the act he has
co m m itte d : crim e, no t dirt, is at issue. T h a t is perhaps an
e x tre m e case, a n d w e need not suppose th at physical pollution
alw ay s bears so heavy a m etaphorical burden; b u t it w ould be
cu rio u s if th e p o llu tio n o f corpse an d m other did indeed derive
88 E·. B. T y lo r, Primitive Culture4, L ondon, 1903, vol. ii, 429.
89 D a rd a n ia n s: N ie. D am . 90 F G rH fr. 107. Robes: Srinivas, 105, cf. ibid., 82 f for die
H in d u w o m a n ’s n u m ero u s b a th s after b irth , only certain of w hich im prove her purity.
G . B a ch elard , L ’Eau et les rêves, Paris, 1942, 192, observes ‘Le C afre ne se lave le corps
q u e lo rsq u 'il a l’âm e sale’.
90 D ouglas, C h . 7.
Birth and Death 57
solely from beliefs ab o u t d an g ero u s forms an d conditions of
m a tte r, a n d ow ed n o th in g to feelings ab o u t th e g reat hu m an
ev en ts a t th e ce n tre o f w hich they lie.
T y lo r stressed th a t pollution focuses aro u n d real dirt; m edi­
cal m ateria lism stresses th a t it focuses aro u n d real danger.
M e d ic a l m ateria lism is W illiam J a m e s ’s term for the a tte m p t to
sh o w th a t, w ith in ap p a re n tly a rb itra ry an d superstitious religi­
o us laws, so und hygienic principles are enshrined. M aim onides,
w ho in th e tw elfth cen tu ry in terp re ted the ab o m in atio n s of
L eviticu s in term s o f p ractical dietetics, still has m any followers
to d ay . T h is m eth o d of exegesis, offering as it does a ready
co m p ro m ise betw een religious a n d scientific tru th , is p artic u ­
larly w elcom e to ed u cated , ra tio n alist ad h eren ts o f faiths th at
re q u ire th e o b serv an ce of such rules. T h u s J . J . M odi, a Parsee
h im self a n d a u th o r o f the s ta n d a rd work on Parsee ritual,
ex p lain s th e Baresknum, th e elab o rate A vestan purification cere­
m ony for m o u rn ers, as a tech n iq u e for isolating the corpse itself
a n d all w ho com e in to co n tact w ith it, and so preventing the
s p re a d o f infection. T h e Drug Nasu, th e fly dem on o f d ea th th at
settles on th e corpse, is sim ply an im ate contagion; the
m o u rn e r’s seclusion, a p ro to -q u a ran tin e. O n e ch a p te r actually
b ea rs th e title O l d Ira n ia n P urification an d M odern Plague
O p e ra tio n s ’. A special n u m b er o f the Health Education Journal
w as q u ite recently devoted to articles by, am ong others, H indus,
M u slim s, a n d B u ddhists, illu stratin g the hygienic principles
em b o d ied in th eir ritu al rules. In the G reek world, it comes as
no su rp rise to find P lu tarch app ly in g m edical rationalism to
ex p lain p u zzlin g religious tra d itio n s .91
C e rta in ly , rules o f p u rity m ay resem ble rules o f hygiene, and
m ig h t even on occasion have beneficial hygienic effects. It is
so m etim es claim ed th a t m edieval Jew s escaped the w orst rav­
ag es o f th e p lag u e th ro u g h observance of the code o f Leviticus.
F u m ig a tio n by su lp h u r, m uch practised by the Greeks, does
disinfect; b u t one m ay d o u b t w h eth er they ap plied it with
sufficient rig o u r to receive m uch em pirical confirm ation o f the
91 O n m edical m aterialism see the exposition an d critique o f Douglas, 4 1 -4 , which I
follow closely. J . J . M odi, The Religious Ceremonies and Customs o f the Parsees2, Bombay,
1937, esp. 9 8 -1 0 1 , 149 ff.; The Health Education Journal 1 7 .1 ,M arch 1959; Plut, de Is. el
Os. 3 8 3 a-c, Quaest. Com. 670f-671a. For hygienic rationalizations of beliefs concerning
fem in in e po llu tio n s cf. G. E ichïnger Ferro-Luzzi, Anthropos 69 (1974), 154.
58 Miasma
efficacy o f th eir religious rule, a n d su lp h u r was as valuable in
th e tre a tm e n t o f bew itched livestock as in the purification of a
h ouse. W e are p ro b a b ly dealing h ere w ith a case o f coincidence
b etw e en a su b sta n c e ’s sym bolic a n d scientific appropriateness
for a p a rtic u la r ta sk .92 Scientific rationalism would be hard
p re sse d to ex p lain w hy the sprinkling o f the house o f d eath at
Iu lis h ad to be done by a free m an, o r w hy the house’s fire and
w a te r b u t no t its tables a n d blankets suffered especial pollution.
C le a rly the sp rin k lin g was a cerem onial act, not to be entrusted
to th e slaves w ho w ould perform the m ore m u n d an e tasks of
h o u se cleaning; as for fire an d w ater, it is because they are
p a rtic u la rly c a p ab le o f p u rity th a t they are particularly liable to
p o llu tio n . E m p irical observation o f the facts of contagion could
scarcely h av e led to th e belief th a t pollution affected a dead
m a n ’s relatives m ore severely th a n outsiders, a n d it w ould be
easy to am ass fu rth e r exam ples to show how few sound m edical
p rin cip les G reek religious rules in fact contain. T h e point is not
m erely th a t these laws w ere not in practice m edically effective,
b u t ra th e r th a t they w ere not conceived in this light at all. A
b re a c h o f th em m ig h t in theory lead to disease, b u t th a t is
e q u a lly tru e o f the obligation to sacrifice or any o th er religious
ru le. Som e G reeks w ere aw are, a t least in tim e o f plague, that
d isease could be tra n sm itte d from person to person ,93 bu t there
is n o evidence th a t they norm ally saw corpses as a source of
infection. O n e historical text th a t does, unusually, speak o f
u n b u rie d bodies as a d an g er to h ea lth explains this d an g er in
te rm s o f co rru p tio n o f the air ra th e r th a n direct contam ination
o f th e survivors by the corpse .94 A nd, though gods are ageless
a rid diseaseless, they are p articu larly exposed to pollution.
I t is tem p tin g to seek an analogy betw een these natural
p o llu tio n s a n d th e veil o f silence d raw n over m any aspects o f
m a n ’s physical n a tu re in m odern w estern society. Excretion,
d ecreasin g ly b u t still in large m easure sexuality, increasingly, it
is claim ed , d e a th , are d irty things th a t are to be hidden aw ay
a n d n ev er spoken of. W e have got into the h ab it o f referring to
th e se as ‘ta b o o ’ subjects. T h e re is, how ever, an im portant
d ifference betw een these m odern taboos an d those recorded by
92 Cf. L évi-S trauss, 12.
93 See p. 219 below.
94 D iod. 17.64.3, cf. Q u in t. C u rt. 5 .1 .1 1
Birth and Death 59
an th ro p o lo g y : in th e one case th e tab o o is a veil, while in the
o th e r it seem s ra th e r to act as a m ark er. T h e u ltim ate objective
m ay p e rh a p s be sim ilar in the tw o cases, b u t the in term ed iate
tac tic s are q u ite different. T h e re w ere certainly things th at
G re ek society g enuinely sought to hide from view, b u t in assign­
in g th e w om en w ho h a d assisted at a b irth, for instance, to a
sp ecial categ o ry o f th e polluted they w ere draw ing atten tio n to
th e ev en t ra th e r th a n obscu rin g it. W e are said today to practise
‘d e a th a v o id a n c e ’; we take no last farewells, shield children
from all know ledge o f the disru p tiv e event, an d refuse to ack­
now led g e explicitly th a t d ea th is near; w hen it is im m inent, we
a b a n d o n th e dy in g m an to the care o f professional nursing
staff .95 T h is is m ore like the b eh av io u r o f G reek gods th an G reek
m en ; m o rtals are n o t entitled to hold them selves aloof from
p o llu tio n . F u n ctio n alism w ent so far as to explain prim itive
ta b o o as a m ech an ism for en su rin g the ap p ro p riate social in­
v o lv em en t in occasions like b irth a n d death: the h u sb an d is
forced by a tab o o to rem em b er th a t his wife is lying in .96 T he
fu n c tio n a list h as n o t proved his co ntention th at the ritu al su r­
ro u n d in g these crises creates concern ra th e r th an expressing it,
b u t h e is no d o u b t rig h t to assum e th a t this, like all ritu al, is a
m e a n s o f d isp lay a n d not disguise.
T h e m ost im p o rta n t co n trib u tio n to an u n d erstan d in g of
th ese p h e n o m e n a rem ains th a t o f v an G ennep, w ho in a
c e le b ra te d w ork d em o n strated the very general h u m an
te n d e n c y to ritu alize im p o rtan t transitions o f every type - in
sp ace, in tim e, in social statu s - and illustrated the typical
s tru c tu re o f su ch ritu als, w hereby the person undergoing the
tra n s itio n is w ith d raw n from his previous surroundings,
m a in ta in e d for a tim e in an in term ed iate state, an d finally
re in te g ra te d in to society u n d er new conditions or at a new level:
rite s o f se p a ra tio n lead to a period o f transition concluded by
rites o f in c o rp o ra tio n .97 T his extension over tim e is, it seems, the
95 P h. A ries, The Hour o f our Death, L ondon, 1981 ( = L ’Homme devant la mort, Paris,
1977), 5 5 9 -6 0 1 , 6 1 1 -1 4 .
96 A . R. Radcliffe B row n, ‘T a b o o ’, in his Structure and Function in Primitive Society,
L o n d o n , 1952, 1 3 3 -5 2 . C riticism s in Steiner, C h. 10; cf. how ever the sym pathetic
co m m en ts o f B a rth , 166 f.
97 v a n G en n ep , passim; th e trip a rtite schem a, 11. T h e transitional stage is studied,
b u t in relatio n to in itiatio n only, by V. W. T u rn e r, The Forest o f Symbols, C ornell, 1967,
93—111 a n d idem ., The Ritual Process, L ondon, 1969, C h. 3.
60 Miasma
w ay in w hich societies em phasize the changes th a t are m ost
im p o rta n t to them . F or the individual, the ritu al stages provide,
w h e re necessary, a program m e for em otional ad ju stm en t to the
crises o f his experience: in bereavem ent, for instance, he m ust
in d u lg e his g rief for a fixed period, an d then set it aside .98 T he
m a n y ritu als th a t accom pany b irth an d death in G reece fit
n ea tly en o u g h into van G e n n ep ’s schem e. For the central
c h a ra c te r, o f course, n a tu re has done the work, and rites of
s e p a ra tio n are no t needed; b u t after the physical event a cul­
tu ra lly p re scrib ed in term ed iate period m ust elapse before the
b a b y is a d m itte d , a t nam e-giving, to the society o f the living,
a n d before th e last rites consign the corpse to the ground, and
th e soul, still flitting h ith erto am o n g the living, to the w orld of
th e d ea d . As th e tran sitio n s undergone by those who die or are
b o rn tran sfo rm th e w orld o f th eir associates, they too become
su b je c t to rites o f passage. A t b irth , it is prim arily the m other
w h o is w ith d ra w n from norm al society an d requires rein­
c o rp o ra tio n , b u t th e o th e r relatives, by atten d in g nam e-giving
rite s a n d the like, acknow ledge an d assim ilate the change that
h a s com e over th e family. A fter a d ea th , all the relatives and
asso ciates e n te r a n ab n o rm a l state, know n as m ourning, in
w h ich fam iliar p u rsu its, interests, dress, a n d d ep o rtm en t are in
v a ry in g degrees forbidden. O bsessed, actually or convention­
ally, by m em ories o f th e d ead, they are d u rin g this transitional
p erio d h alf-dead them selves. (T o accep t d ea th an d reject the
p e rio d o f lim bo can be, as the hero o f C a m u s’s L ’Etranger found
o u t, a serious social crim e.) D uring the period o f m ourning, a
tw o-w ay tra n sitio n occurs: the d ead m an moves from the land of
th e living to th a t o f the spirits, w hile the survivors re tu rn from
d e a th to life. T h e last rites finally incorporate or reincorporate
d e a d a n d living respectively in th eir p ro p e r com m unities. As
A chilles says to A gam em non: ‘A t daw n, urge your m en to fetch
w ood, a n d provide all th a t the corpse should have w hen it goes
to th e d ark n ess below; so th a t all the quicker u n tirin g fire may
b u r n him aw ay from o u r sight, a n d the people m ay tu rn to their
98 C f. G . G o rer, Death, Grief and Mourning in Contemporary Britain, L ondon, 1965, 72-8.
a n d esp. the com m ent q u o ted on p. 75, ‘the week o f g rief gives you tim e to get over all
th e w orry a n d w h a t n o t . . . Even though it seem s outlandish a t the tim e, it really is a
h e lp . . . y o u ’re aw ay for a week an d get over all yo u r grief. You get it all concentrated in
o n e w eek’ (ap ro p o s o rthodox Je w ish m ourning rites).
Birth and Death 61
ta sk s .’99 In early G reece, as in o th er societies, this ritu al of
s e p a ra tio n w as so effective th a t the recipient was now ‘d e a d ’
even i f i t h a d been perform ed for him in error, in his absence.
P lu ta rc h tells us th at: ‘A nyone for w hom carrying o u t and
b u ria l h ad been perform ed, as th o u g h he were dead, was con­
sid e re d im p u re by th e Greeks, a n d they w o u ld n ’t let such a one
asso c ia te w ith them selves, or e n te r a tem p le.’ It was only after
D e lp h i h a d devised a ritu al o f rein teg ratio n , in the form o f an
ela b o ra te p an to m im e o f re b irth , th a t such u n fortunates could
be re a d m itte d to society at all .100
I t is obviously in th e context o f the ritu alizatio n o f transitions
th a t th e p o llu tio n s o f b irth a n d d e a th belong, an d specifically in
th e in te rm e d ia te stage. In a b rillian t book, M ary D ouglas has
su g g ested th a t p o llu tio n is in g eneral a p roperty o f th e betw ixt
a n d betw een; th a t w hich falls betw een or violates the categories
in to w hich a given society divides external reality is accounted
by th a t society im p u re .101 In Leviticus, for instance, edible land
99 H orn. //. 23. 4 9 - 3 3 . Cl', the N'uer address to the dead m an: ‘F riend, this beast is
yours. N ow tu rn yo u rself to the ghosts. T u rn yourself'aw ay from us’ (E vans-P ritchard,
146); th eir m o rtu a ry cerem ony is called the ‘cu ttin g ofT of the dead. F or a now classic
stu d y o f m o rtu a ry rites see H ertz.
100 Quaest. Rom. 2641 —265a: a striking m odern case, F,vans-Pritchard, 152 f. T o be
falsely rep o rted d ead was a κακός δρνις (E ur. Hel. 1051) in the 5th century, b u t not
in to le ra b le in a good cause (Soph. El. 58—64); we do not know w hether, if the funerary
rites h ad not been p erfo rm ed , P lu tarc h ’s ritu al w ould still have been necessary.
101 D ouglas, passim; for her earlier and la te r thou g h ts on the topic cf. h er collected
p a p e rs, Implicit Meanings, Essays in Anthropology, L ondon, 1975, an d the reader, Rules and
Meanings, ed. M . D ouglas, L ondon, 1973. E. R. L each develops sim ilar ideas in 'A n im al
C a te g o ries an d V erb al A b u se’, in E. H. L enneberg (ed.), New Directions in the Study o f
Language, M a ssach u setts, 1964, 23—63 (rep rin ted in P. M a ran d a (ed.), Mythology,
Selected Readings, L o n d o n , 1972, 39—67): he is effectively criticized b y j . H alveson, Man
n.s. 11 (1976), 505—16, cf. n.s. 12 (1977), 527 f. S. J . T am b ia h offers a theoretical
m od ificatio n in J . G oody (ed.). The Character o f Kinship, C'antbridge, 1973, 191 f.,
(p o llu tio n located in th e o v erlap betw een two categories, not the gap betw een them )
a n d a p ractical ap p licatio n in 'A nim als are Good to T h ink and Good to P rohibit',
Ethnology, vol. viii, n. 4 (O cto b er 1969), 424—59, m ostly reproduced in Rules and
Meanings, 127—66. T w o elem ents in the original theory should now, it seems, be
je ttiso n e d : (1) the an alogy betw een how perceptual skills are learnt by the child, and
how cu ltu ral categ o ries are eith er learnt by the child or created by societies: cf.
P. H e rsh m a n , Alan n.s. 9 (1974), 292—4, C. R. H allpike, The Foundations ojPrimitive
Thought, O x fo rd , 1979, 69—71 ; (2) the notion th at bodily em issions are im pure because
they create a m b ig u ity ab o u t bodily boundaries: lor a m ore plausible view cf. A. S.
M eigs, ‘A P a p u a n P erspective on Pollution’, Man n.s. 13 ( 1978), 3 0 4 -1 8 . The defini­
tion ol d irt as ‘m a tte r out o f place' requires, a t least, qualification, cf. Meigs, loc. cit.,
H allpike, op. cit., 160 n. 4. H allpike plausibly suggests th at the p rim ary form o f 'd ir t’ is
faeces; an d on tre a tm e n t o f faeces ethology m ight well offer guidance.
62 Miasma
a n im a ls are defined as ‘w hatsoever p a rte th the h o o f . . . and
c h e w eth th e c u d ’. T h e pig is therefore an abom ination, because
‘th o u g h he d ivide the hoof, yet he chew eth not the cu d ’. Lacking
o n e o f th e necessary characteristics of the acceptable dom estic
a n im a l, he is a m o n ster. As a general theory of pollution, this is
n o t w holly convincing; no t all pollutions can be seen as pro­
d u c ts o f category violations, a n d it is not clear th at prim itive
societies are necessarily m ore disconcerted by classificatory
a n o m a lie s th a n w e a re by, say, the am biguous status of the
to m a to .102 B ut in th e case o f the rites o f passage, the theory has
a n obvious plausibility. Persons in the transitional condition
a re by definition betw een statuses, a n d it is not h ard to see the
co rp se a n d th e new bab y as situ ated at the interstices betw een
tw o w orlds. T h e corpse, in p artic u la r, is anom alous both soci­
ally (no longer in h u m a n society, not yet am ong the dead) and
p h y sically (all the o u tw ard m arks o f a living person, bu t life­
less). T h ese tran sitio n al beings do not, however, fall betw een
categ o ries b ecause the existing categories cannot hold them .
T h e re is no in trin sic classificatory problem ab o u t the new -born
b ab y ; he is alive en o u g h w hen he enters the world, an d it is only
b ec au se o f th e ch a rac te ristic stru c tu rin g o f transitions th a t he
m u s t be su sp en d e d for a period in lim bo —m ore a p lan t than a
h u m a n being, says P lu ta rc h 103 — before being adm itted by
b a p tism , n am in g , o r sim ilar cerem onies to the com pany of the
living. H ere, it is not the case th a t the logic of classifications has
g e n e ra te d a m isfit w ho therefore evokes a reaction of alarm ; on
th e c o n tra ry , a disconcerting being has been declared a misfit
by special m a n ip u la tio n o f the classificatory processes. T he
b ein g is d isco n ce rtin g not on logical, cognitive, or classificatory
b u t o n the sim ple em otional grounds th a t it is h ard to adjust to
decisive change. I t is interesting th a t, in d eb a te ab o u t the burial
o f corpses, we d o find arg u m en ts th a t appeal to the breach of
ca te g o ry b o u n d arie s. By refusing burial, C reon is ‘keeping here
w h a t belongs to the gods below ’; in such a case, ‘T h e gods above
a re b eing p o llu ted , a n d the gods below are not getting w h at is
th e irs .’ W e h av e seen, however, th a t it is outrage at unm erited
102 Cf. D ouglas herself, Implicit Meanings, 288; J . G oody, The Domestication oj the Savage
M ind, C a m b rid g e, 1977, 45; G . S. K irk in Entretiens Hardt 27, 4 4 -7 .
103 Quaest. Rom. 102. 288c.
Birth and Death 63
d is h o n o u r th a t causes this p a rtic u la r d em arcatio n to becom e so
im p o rta n t. I t looks as if d eclarin g an object or a person a misfit
m ay be 3 ra tio n a liza tio n for unease, distaste, o r anger th at is felt
on o th e r g ro u n d s .104
W e re tu rn to v an G ennep. T h e pollutions o f b irth a n d d eath
re la te to th e d iso rien tatio n ac tu ally o r conventionally produced
by the g re a t crises in h u m an existence. N ot all crucial tran si­
tio n s po llu te, how ever. It is n o t en ough to say th at m arriag e is
too joyful an occasion to be polluting, because b irth is joyful too.
T h e real difference seem s to be th at, while m arriag e is a con­
tro lled event, b irth a n d d eath in tru d e on h u m an life a t their ,
ow n p leasu re. T h ey are an irresistible ‘irru p tio n o f the bio­
logical in to social life ’.105 A lth o u g h they are n a tu ra l events, they
a re also v iolations o f order; th e d ead o r dying m an a n d the
p a r tu r ie n t w om an h ave lost control o f th eir own bodies, an d the
social g ro u p m u st sta n d back pow erless while crucial changes
a re w orked u p o n it. T h e acco m p an y in g rites o f passage can be
seen as reassertio n s o f control; th e bab y , th ru st rudely into the
w orld by n a tu re , still requires social acceptance, an d the shade
w ill n o t be a b le to reach the w orld o f the dead unless the due
rites are p erfo rm ed . T h is co m m an d eerin g o f the n atu ra l p ro ­
cesses by society th ro u g h ritu al is so effective th a t w hen ritual
a n d ph y sical facts conflict, physical statu s yields to ritu al; living
m en for w hom fu n erary cerem onies have been perform ed have
b ee n d eclared ‘d e a d ’, an d , as we saw, d ead the u n fortunates
m u s t rem ain . M arriag e, by co n trast, is not an in trusion th at
re q u ire s sealin g off, b u t is itself a harness set upon the rebellious
body. V ico defined it as a ‘ch aste carn al union consum m ated
u n d e r fear o f som e d iv in ity ’; it purifies the physical. In viewing
th e po llu tio n s o f b irth an d d ea th in this way, we are, though
d o u b tin g D o u g las’s specific location o f im p u rity in the betw ixt
104 Cf. now M . B eard, J R S 70 (1980), 20, w ith references. C reon: Soph. Ant. 1070 f.,
cf. Lys. 2.7. By th e 4 th c., category am biguity was a subject for ch ild ren ’s riddles, PI.
Resp. 4 7 9 b -c . See too LSJ s.v. έπαμφοτερίζειν.
105 D u m o n t, 99, cf. 8 8 - 9 , ‘It can be seen th at im purity corresponds to the organic
asp e c t o f m an. Religion generally speaks in the n am e o f universal order; b u tin this case,
th o u g h u n aw are in this form o fw h a t it is doing, by proscribing im purity it in fact sets up
a n op p o sitio n b etw een religious an d social m an on th e one hand, an d n atu re on the
o th e r .’ P. H e rsh m a n , Man. n.s. 9 (1974), 290 claim s ‘Pollution is essentially th a t which
c a n n o t be co n tro lle d .’
 64 Miasma
a n d betw een, ac cep tin g h er b ro a d er insight th a t fear o f pollu­
tio n is a p ro d u c t o f th e urge for o rd er a n d control.
N a tu ra l p o llu tio n s are, it ap p e ars, com plex. T hrough
sy m b o lism o f d irtin ess th a t derives from the events them selves
(th e ste n c h o f the corpse, the m ess o f the birthroom ),106 there is
conveyed the d isru p tio n w ro u g h t in the social g ro u p ’s steady
ex isten ce by physical events th a t are o f crucial im portance to it
b u t bey o n d its control. T h is relation to the experience of the
social g ro u p is p a rtic u la rly clear in the case o f funerary pollu­
tio n . A ristotle observes th a t m o u rn in g is a form of ‘hom oeo­
p a th y ’ o f th e m o u rn ers w ith the d ep a rted . D eath-pollution, too,
is a kin d o f tem p o ra ry p artic ip a tio n in the condition o f the dead
m a n , w ho is th ro u g h th e decay o f the corpse ‘foul’ (miaros). In
H o m e r certain ly , a n d possibly in historical tim es too, the
m o u rn e rs ‘befouled’ them selves w ith d u st in sym pathy; pollu­
tio n is a tran sp o sitio n o f this sy m p ath etic befoulm ent to the
m e ta p h y sic a l p lan e. ‘Being p o llu te d ’ is a kind o f m etaphysical
s u it o f m ou rn in g . T h is, and not an erroneous theory o f co n ta­
gion, is w hy the d ea d m a n ’s relatives are m ore polluted than
o u tsid e rs, a n d w hy w hen m ourning is forbidden pollution m ay
be ex clu d ed to o . 107
In d e e d , it ca n be very h ard to distinguish betw een the
socially p re scrib ed consequences o f ‘p ollution’ an d o f ‘g rief’.
T h o s e w ho h av e a tte n d e d a funeral are excluded from sacred
p lace s because they are polluted, b u t sacrifice is a joyous occa­
sion for G reeks a n d th u s they w ould have been excluded an y ­
w ay by th e conventions o f m ourning. (Sim ilarly, it w ould, we
so m etim es h ear, be sacrilegious to m ention D ionysus in connec­
tio n w ith d e a th ; 108 it w ould also, o f course, be bad form to think
o f th e festive god in such a context.) A fath er who goes ab o u t his
n o rm a l business d esp ite the d ea th o f a child is blam ed for his
u nfeeling violation o f m o u rn in g ra th e r th a n for ‘polluting the
te m p le s ’.109 I t is as though the gods by th eir concern for purity
106 O n su ch focusing sym bols cf. H ertz, 82—3, V . VV. T u rn e r, The Forest o j Symbols,
C o rn e ll, 1967,98.
■107 A rist. fr. 101 R ose3 ap. A th. 675a. C orpse miaros, H orn. II. 24.420.
108 H d t. 2.86.2, D em . 60.30, Pl. Menex. 238b. Cf. p. 70 n. 123 on the incom patibility o f
d e a th a n d sexuality.
105 A eschin. 3.77; a sim ila r attack , Isocr. 19.40, a n d cf. Lys. 1.14, (P lut.) Cons, ad
Apoll. 118c—119d.
Birth and Death 65
m erely enforce th e b eh av io u r th a t is ap p ro p riate in purely
h u m a n te rm s .110
T h e se o b serv atio n s do not claim to be exhaustive. T h e belief
th a t ‘corpses a re d irty ’ is clearly a possible vehicle for num erous
em otio n s. In th e case o f b irth , it is surely likely th at the Greeks
will h av e asso ciated th e im p u rity o f m o th er a n d child d u rin g
th e first few d ay s o f life w ith th eir very real physical peril du rin g
th a t perio d . P o llution would th u s have helped to define a n d so
lim it a perio d o f d a n g e r an d anxiety; the cerem ony en d in g it
w ould be a ritu al expression o f the hope th at the child, having
su rm o u n te d th e initial dangers, now belonged to this w orld and
w ould live o n .111 T h e polluted m o u rn er too surely felt in contact
w ith a n a b o m in a b le pow er. F u n era ry pollution is not explained
by m a n ’s fear a n d h atre d o f d eath , or b irth -p o llu tio n would be
inexplicab le; it m u st none the less surely have becom e a focus
for th ese feelings. In such an a rea O c k h am ’s razo r is too b lu n t
a n in stru m e n t.
In G reek b elief these pollutions h ad a fu rth er special ap p lica­
tion, th eir role in sep a ratin g gods a n d m en. (‘Special’ here does
n o t m ean e ith e r u n iq u e or recent; th e sam e ap p lication occurs,
in v ary in g degrees, in R om an religion, H induism , and
Z o ro a s tria n is m .112) W hile in m ost tribal societies it is the p ro­
te c tio n o f fellow h u m an s a g ain st these n atu ra l pollutions th a t is
th e m ain concern, in G reece real d an g er seems only to occur if
th e gods a re exposed to them . T h u s it is on the altars, not am ong
110 T h e relatio n o f ‘p o llu tio n ’ to ‘m o u rn in g ’ is in general u ncertain. Is the period o f
a b n o rm a lity follow ing a S p a rta n king’s d e a th (H d t. 6.58.3, X en. Hell. 3.3.1, A rist. fr.
6 11.10) ‘m o u rn in g ’ (H d t.) or ‘pollu tio n ’ (X en.; for a funerary period as not hosios cf.
E u r. Antiope 80, Page C L P p. 68)? M o u rn in g periods could be protracted: e.g. Lys. 1.14
(som e restrictio n s till 30th d ay ), LSA 16 (3 m onths!). P resum ably ‘pollution’ lasted less
long; b u t the m a tte r is obscure.
111 ‘T h e ritu a l d an g ers w hich are believed to threaten the process o f conception and
b ir th are, to som e ex ten t, a reflection o f p rag m atic anxieties ab o u t real d angers . . .
w h ere p reg n an cy an d ch ild b irth not infrequently lead to the d eath of both m other and
c h ild ,’ B uxton, 214. F re q u en t death s in 1st week, Arist. Hist. An. 588a 8 -1 0 . In terco n ­
n e c tio n o f p hysical a n d ritu a l dan g ers in m o d ern Greece, Blum , 12 f , 19 f., 111 (64) (in
th ese acco u n ts m o th er an d child are clearly m ore im perilled by their ow n pollution
th a n are o u tsid ers).
112 H in d u ism : C .J . Fuller, X lann.s. 14 (1979), 473; L. D um ont, Une sous caste de VInde
du sud, P a ris/T h e H ag u e, 1957, 345, cf. 210; G. Eichinger Ferro-L uzzi, Anthropos 69
(1974), 1 3 1 -3 . W e find here a sliding scale: im purity of any kind d eb ars from the
tem p les, w hile severe p ollution also im pedes dom estic and social activities, cl. herro-
L uzzi, loc. cit., S rinivas, 106. Z oroastrians: e.g. Boyce, 100.
66 Miasma
th e houses, th a t Sophocles’ birds o f prey d ro p the scraps of
P olyneices’ corpse, a n d , as we have seen, it is h ard to identify
a n y c e rta in consequence o f co n tact w ith natu ral pollutions
a p a r t from exclusion from the tem ples. T h is exclusion from the
sa c re d is no d o u b t in origin, as we have seen, sim ply a n exclu­
sion from social life in its festive forms; there is no celebration,
n o feeling o f co m m u n ity , w ithout sacrifice. It certainly comes to
seem , how ever, as if the real b a rrie r th a t pollution sets u p is not
b etw een m an a n d m an b u t betw een m an and gods. By banning
b irth , d e a th , a n d also sexuality from sacred places, the Greeks
em p h asiz e th e g u lf th a t separates the n a tu re of god an d m an.
O n o n e level, o f course, the gods have m uch in com m on with
m a n in these respects: they u n d erw en t birth, an d engage in
sex u al activity. B u t w hereas for m en birth and sex are p a rt of a
cycle th a t ends in the grave, the gods enjoy the benefits o f the
flesh b u t no t its ills. (P hilosophers w ere to seek to free the gods
from th e ta in t o f the physical altogether.) Excluded from a
te m p le because o f the b irth o f a son, a G reek is rem inded,
p e rh a p s , th a t his son has been born to replace himself, and die
in his tu rn , w hile the gods persist in splendid im m ortality .113
T h is ac co u n t m ay be taken roughly to represent the situation in
A th e n s in th e fifth century. In the H om eric world, it has often
b een a rg u ed , a ttitu d e s w ere very different. D espite the count­
less d e a th s d esc rib ed in H om er, there is no hint o f miasma
affecting the living. T h e heroes m ay re tu rn to their norm al
p u rs u its after a funeral w ithout ap p a ren tly even w ashing. In
p a rtic u la r, th e ab so lu te revulsion o f the gods from scenes o f
d e a th seem s to b e m issing. T h ey m ingle in b attle w ith the dead
a n d dying, a n d do no t disdain to touch a corpse. T h e river god
S c a m a n d e r, th o u g h com plaining th a t Achilles is obstructing
his stre a m s w ith d ead bodies, does not speak o f this as a
d e se c ra tio n . T h e only pollution know n to H om er, on this view,
113 G olden age m y th s often stress th at originally there was neither sexuality nor
d e a th : C. von F ü re r-H aim en d o rf, Man n.s. 9 (1974), 5 4 0 -2 , 548. T h e cattle of the sun
a r e n o t born a n d d o n ot d ie, Horn. Od. 12. 129-31. Em pedocles saw sexuality as
c h a ra c te ristic o f th e flaw ed world o f ‘strife’ (see p. 301 below).
Birth and. Death 67
is sim p le d irt; for his w orld th e m etap h y sical contagion of d ea th
w o u ld be a co n c ep tu al im possibility .114
T h e p o stu la te d gro w th in pollution fears conflicts, for w h at
it is w o rth , w ith P lu ta rc h ’s p ictu re o f the great arch aic legisla­
tors g ra p p lin g w ith powerful su p erstitious fears a tta c h e d to
d e a th .115 It is m ore im p o rta n t th a t th e argum ents them selves
will scarcely b e a r investigation. P ollution belief in som e form
u n d o u b te d ly existed in the classical period, and yet it is easy to
find p assages w here classical au th o rs seem oblivious to it.
D e a th s in H o m er o ccur in b attle, b u t th ere is no evidence th a t
soldiers w ere ever p o lluted by th e d ea th s o f their com rades. If
A pollo, p u re st o f gods, h andles S arp e d o n ’s corpse in the Iliad
w hile his sister shuns th a t o f H ip p o ly tu s in E uripides, th a t
difference derives from a p e rm a n e n t am biguity in the relation of
god to m an ra th e r th a n a tran sfo rm atio n in belief. T h e gods do
ca re for th e ir h u m a n favourites, b u t fate a n d the m ortality o f the
fav o u rites im pose u p o n th a t care lim its w hich m en m ay resent.
In th e p a rtic u la r contexts, H o m er is em phasizing the care,
E u rip id e s th e lim its; b o th poets are p o rtray in g an aspect o f the
d iv in e n a tu re , n o t tran sc rib in g ritu al rules. O n a b ro ad er level,
th e g u lf b etw een m ortal and im m o rtal th a t E uripides expresses
th ro u g h A rte m is’ flight before pollution is the u nshakeable first
p re m iss o f H o m e r’s religion .116 Even in a fifth-century poet, the
sam e A pollo su b jects h im self to w orse pollution th a n in H om er;
th e th ird Pythian show s the p u re o ne snatch in g a new -born baby
from a corpse. T h e poet who w rote this lived, none the less, in a
city th a t k ep t tem ples a n d tom bs well separated .
T h e re is in fact a connection betw een d ea th a n d d irt in
H o m e r, a lth o u g h it m ay not be ju stified to speak o f pollution in
th e classical sense. Physical cleanliness is an im p o rtan t expres­
sion in th e poem s o f w holeness a n d propriety. T h e heroes w ash
before m eals, a n d w ould be ash am ed to pray to the gods when
d irty ; before p o u rin g an im p o rta n t libation, Achilles first
114 Cf. Stengel, 156 f.; idem , Hermes 41 (1906), 241 (= Opferbraiiche der Griechen,
L eipzig, 1910, 28 f.); M . M . Gillies, C'Q 19 (1925), 7 1 -4 ; M oulinier, 2 5 -3 3 . Funerals:
G illies cites II. 23.257 f., 24.801 f., Od. 12.10 ff. Corpses: II. 16.666-83, 24.612.
S c a m a n d e r: II. 21. 2 1 8 -2 1 . Z enodotus ath etized II. 16.666-83, as im posing in ap p ro ­
p r ia te w ork on th e ‘griefless o n e’.
115 Sol. 12.8, Lyc. 27.1.
116 II. 5 .4 4 0 -2 , Griffin, C h . 6, passim. In II. 22. 213 Apollo in fact ab an d o n s the
d o o m ed H ecto r, b u t this is in a sense cause as well as consequence o f the com ing death.
68 Miasma
purifies th e c u p w ith su lp h u r, th en washes it in w ater, and
finally w ashes his own hands; it is a rite o f cerem onial sprinkling
t h a t unites the p a rtic ip a n ts at every sacrifice in a sacred circle.
C le a n clothes a re essential to a display of respect o r a sense of
w e llb ein g .117 R eactions to disaster, by contrast, com m only
focus u p o n a physical defilem ent, incidental though it m ay
seem to th e real im p o rt o f w h at h as occurred. ‘H is head, beauti­
ful before, lay in the d u s t’: so H o m er sum s up the hum iliation of
th e d ea d H ecto r. T h e fall of P atroclus is prefigured in the fall of
th e helm et: ‘th e crest was befouled w ith blood an d dust: yet
before it h a d n o t been p erm itted for the horse-hair helm et to be
befouled w ith d u st, b u t it p ro tected the head a n d fair brow of a
god-like m an , A chilles .’118 W e are dealing, certainly, w ith a
g re a t p o e t’s pow er to express the a b stra c t through the concrete,
b u t, unless H o m er invented the w hole system o f ritu al w ashing
t h a t h e describes, cleanliness m u st also have been an uncon­
scious sym bol o f good o rd er in the society th at he knew .119 This
sy m b o lic significance o f physical integrity is one reason why it
m a tte re d th a t th e corpse should not be m utilated. Achilles’
im m e d ia te re actio n to the news o f P atro clu s’ d ea th is therefore
m o st in terestin g : ‘T ak in g grim y d u st in both his h ands he
p o u re d it over his head, a n d befouled his fair lace.’ Achilles
p o llu te s h im self in his grief; later, urged to w ash off the battle
m ire th a t clings to him from his conflict w ith H ector, he d e ­
c lares w ith a n o ath : ‘N o w ater m ay com e n ea r my head, before I
h a v e set P atro clu s on the flam es, h eap ed him a m onum ent, and
c u t m y h a ir for h im .’ H e speaks o f this refusal to w ash as a
religious o b lig atio n (themis); th a t m ay be the language o f pas­
sion, b u t self-pollution a n d not w ashing w ere probably, if not
fixed rules o f m o u rn in g , a t least trad itio n al m odes for the
ex p ressio n o f grief .120
' 17 Cf. p. 20. C lea n clothes: cf. O . P. T a p lin , Greece and Rome 27 (1980) 9-11
" ’‘ II. 22.402 f., 16.795-9.
H o m eric sym bolism is finely described, w ith reference to the concrete symbolism
o f e arly social, political, an d religious life, by Griffin, C h. 1. O n such sym bolism
L. G e rn e t, ‘D ro it et p ré d ro it en G rèce an cien n e’, L ’Année Sociologique, 3e serie (1948—9),
P aris, 1951, 21 —119 ( = G ern et, Anthropologie, 175 fT.) w as m ost im portant.
120 II. 18. 2 3 - 5 , 2 3 .4 4 -6 . F o r self-pollution cf. II. 22.4 1 4 ,2 4 .1 6 3 -5 ,6 4 0 , Od. 24.3161.;
for n ot w ashing, Hym Hom. Cer. 50. A ndronikos observes, p. 2, th at not w ashing cannot
h a v e been a rule, as A chilles is urged to w ash, II. 23.39-41; b ut it m ay have been a
co m m o n practice.
69
Birth and Death
T h is co n d itio n o f Achilles m ay seem to differ from the p ollu­
tion o f th e m o u rn e r in classical tim es. T h e d irt on Achilles is
self-inflicted a n d physical, not au to m a tic a n d m etaphysical;
c o n se q u en tly his condition is n o t contagious, he will becom e
clea n ag ain as soon as he w ashes after th e funeral, a n d th ere is
no sugg estio n o f d an g er. U n d e r ex am ination, these differences
lose th eir co m fo rtab le clarity. T h e d istinction betw een physical
a n d m etap h y sical pollution becom es u n certain if one considers
th a t th e one m ay be a sym bol for the o ther. A t Iulis, as we have
seen, th e m o u rn ers pro b ab ly p o lluted them selves physically,
a n d afte r th e fu n eral it will have been the m arks o f this self­
d efile m e n t th a t they w ashed oil'; b u t th e physical pollution was
also m etap h y sical, or the law w ould not have needed to specify
th a t, by w ashing, the m o u rn ers recovered th eir purity.
M e ta p h y sic a l p o llu tio n certainly was present a t Iulis, because
th e house o f d e a th req u ired purification from it. C onditions in
fifth -c en tu ry Geos p ro v e nothing a b o u t Achilles; b u t it should
be n o ted th a t, as long as he rem ain ed d irty, he was by H om eric
e tiq u e tte exclu d ed from social life an d divine cult no less than
th e classical m o u rn er. T h u s he w as subject to the only practical
co n seq u en ce o f pollution th a t is actu ally attested at any date.
A n obvious difference is th at, in th e H om eric ideology, norm al
life resum es im m ed iately after th e fu n eral ,121 while pollution
m a y cling to th e classical m o u rn er for a fu rth er sp an o f days; but
H o m e r ten d s so to prolong the lam en tatio n at the laying-out
th a t th e p erio d o f ab n o rm al funerary tim e is actually longer
th a n in classical p ra c tic e . 122
T h e a rg u m e n t ends, inevitably, in uncertainty; the evidence
is n o t o f th e rig h t kind. It shows a t least th at the sym bolism o f
p o llu tio n w as a lre ad y linked to d e a th in H om er, even ifit lacked
th e m etap h y sical extension it w as later to receive. N othing,
how ever, conclusively proves the extension to be a later d e­
v e lo p m en t. A sh a rp co n trast betw een H om er and the fifth
c e n tu ry will alm o st certainly be founded on over-em phasis of
th e im p o rtan ce o f d eath -p o llu tio n for the latter, an d will give it
a n u n d eserv ed pro m in en ce am ong the sym bolic expressions of
121 II. 2 3 .5 2 -3 .
122 //. 24. 784—7, Od. 24. 63—5; P atroclus how ever rem ains laid o u t lor one day only,
A n d ronik o s, 9.
70 Miasma
m o u rn in g o f w hich it form ed a p art. I f we do not find explicitly
a tte s te d in H o m er the idea o f d e a th as an inauspicious event
w hose d an g e ro u s influence persists for those who come into
c o n ta c t w ith it, it is certainly p resen t in H esiod or his con­
tin u a to r: ‘D o n o t beget a child on your retu rn from a n ill-
o m en ed b u rial, b u t from a feast of the gods.’ No contact is to be
p e rm itte d betw een p rocreation a n d d ea th , an d ‘b u rials’ are
o p p o sed to ‘feasts o f th e gods ’.123 In a later au th o r we would
ack n o w led g e th a t as pollution belief. All th a t is lacking is the
sy m b o lic connection betw een ‘ill-om ened’ and ‘d irty ’, and that
co n n e ctio n , given the place o f physical pollution in the
sym bolism o f m o u rn in g , lay close a t hand.
T h e re is a t all events no d o u b t th at, if a corpse is denied
fu n e ra ry rites in H om er, the consequences m ay be m ore than
m ere ly physical. H ecto r w arns Achilles th at, if m utilated, he
m a y ‘becom e a w ra th o f the gods’ ag ain st him; E lpenor issues a
sim ila r w a rn in g to O dysseus, an d in the last book o f the Iliad
A c h illes’ co n d u c t does stir the gods to indignation an d interven­
tion. T h e lan g u ag e used is th a t o f divine an g er an d not pollu­
tio n , b u t th e significance o f this distinction is easily over­
em p h asized : in b o th cases a h u m a n rule is receiving
s u p e rn a tu ra l su p p o rt. W e find here, on an im p o rtan t issue, a
c le a r co n tin u ity o f value betw een H om er an d the fifth
c e n tu ry .124
A rchaeological evidence on the disposition o f graves m ight
a lso provide g u id an ce on early attitu d e s to pollution. As we
h av e n oted, e x tra m u ra l burial was the norm in alm ost all
classical G reek cities. It w ould be shocking to m ingle the dw el­
lings o f the d ea d w ith those o f the living, still m ore with those o f
th e gods. C h ristia n b u rial ad sanctos was a sh arp break with
123 H es. Op. 7 3 5 -6 . Cf. B uxton, 149: ‘T h e action o f procréation belongs to life and
m u st n ot be in tro d u c ed in to situations associated w ith death. T o mix the two is
d e a th -d e a lin g .’ T h e sam e incom patibility m eans th a t death-pollution can be elfaced
elsew h ere by ritu a l co p u latio n , R. G. W illis, Man n.s. 7 (1972), 376. Solon banned
w o m en u n d e r 60 from atte n d a n c e at funerals, except for close relatives, Dem. 43.62.
This will h ave served his general aim o f reducing the scale of funerals, b ut in excluding
fertile w om en in p a rtic u la r he m ay have been influenced by religious motives; even to
th e m agnificent funerals o f scrutineers, Plato only ad m its virgins and women past
c h ild b e a rin g (Leg. 947d). F or R om e cf. C ensorinus Ü .N . 2.2 (no blood sacrifice on
b irth d a y s), C IL I 2 p. 231 (F asti P raenestini), C arm en tis, goddess o f birth, and morticina
o p p o se d .
124 II. 22. 358, Od. 11.73, II. 24. 33-7 6 .
Birth and Death 71
p a g a n p ra c tic e . 125 B ut this sensitivity to the place o f b u rial was
n e ith e r u n iv ersal in th e G reek w orld n or im m em orially ancient.
G re ek tra d itio n knew th at, in th e old days, burials m ight be
m a d e actu a lly in the houses o f the living, a n d it also knew of
h isto rical G reek cities th at disposed o f their d ead w ith in the
h a b ita tio n a l area. B oth these trad itio n s have been confirm ed
arch ae o lo g ic ally ,126 an d excavation seem s also to have show n
th a t th e ru le on e x tra h a b ita tio n a l burial was less strictly ap ­
p lie d in early cen tu ries th an in th e classical period. It is tem p t­
in g to co rrelate this increasing desire to sep arate th e d ead from
th e living w ith increasing fears o f pollution. S om ething of the
k in d is su ggested by P lu tarch in the ex p lan atio n he offers of
L y c u rg u s’ fu n erary legislation:
A n o th e r a re a th a t he o rg an ized a d m ira b ly w as th a t o f b u rial . . . he
d e s tro y e d su p e rstitio u s fears ab so lu tely b y allow ing th e b u rial of
co rp se s in th e city, a n d th e sitin g o f to m b s n ear to tem ples; th u s he
m a d e th e y o u n g m en th o ro u g h ly fam iliar w ith sights o f this kind, so
th a t th ey felt n o d is tu rb a n c e or a la rm a t th e th o u g h t o f d e a th , as
th o u g h it p o llu te d a n y b o d y Who to u ch ed a corpse o r w alked betw een
to m b s .127
T h is m oral stiffening th rough in trah a b itatio n al burial, so well
exem plified in w arlike S p arta, does not seem to have extended
to th e o th er in tern a l-b u ry in g city, inbelle Tarentum.
T h e re are tw o difficulties o f p rin cip le in such an argum ent. A
society m ay fear pollution from d eath and the fresh corpse
w ith o u t ex ten d in g th a t fear to th e site o f the g rave .128 Secondly,
even if th e g rave is felt to pollute, it need not be placed right
125 \V. V ollgraff, ‘In h u m atio n en terre sacrée d an s l’an tiq u ité g recque’, Mémoires
présentes par divers savants a l'Académie des Inscriptions, xiv. 2 (1951), 3 1 5 -9 8 , sought to
e sta b lish hellen istic an alo g ies for burial on sacred ground, m istakenly: see L. R obert,
Opera Minora Selecta 4, A m sterd am , 1974, 124 f. Shocking to call an a lta r a tom b, Ar.
Thesm. 888. (P lu t. Arist. 20. 6, on burial o f E uchidas in shrine o f A rtem is Eukleia, is a
p u zzle). O n th e orig in s o f depositio ad sanctos see Ph. A ries, The Hour oj our Death, London,
1981, 3 0 -4 0 ; P. B row n, The Cult o f the Saints, C hicago and L ondon, 1981, Ch. 1.
126 (PI.) M in. 315d, — M iddle H elladic house burials. Intern al b urial at T aren tu m ,
Polyb. 8.28.6, cf. W alb an k , ad . loc. and K u rtz/B o ard m an , 308 f.; a t S p arta, Plut. Lyc.
27. ί , Inst. Lac. 238d, P aus. 3 .1 4 .1 -3 , cf. Ά ρ χ . Αελτ. 19 ( 1964), A 123 ίΓ., 2 8 3 -5 , ABSAY2.
(1905—6), 281, 13 (1906—7), 155 1Γ. Paus. 1. 43.3 shows internal burial to have been
ex cep tio n al a t M eg ara, n ot, as is som etim es supposed, norm al.
127 Lyc. 27.1.
128 N u er a re indifferent to graves (E v an s-P ritch ard , 145), M a n d ari bury' w ithin the
h o m este ad (B uxton, 114); b o th have death -p o llu tio n beliefs.
72 Miasma
o u tsid e the h a b ita tional a re a .129 Every settlem ent has its nooks
a n d in terstices w hich nobody feels to be p art of their own
living-space, a n d this was pro b ab ly particularly true in the
strag g lin g villages a n d tow nlets o f early G reece. T h e disposal of
th e corpse is a ‘ca rry in g o u t’, b u t the necessary psychological
s e p a ra tio n can b e achieved w ith o u t the corpse in fact being
tak en very far. T h u s, even if, in a given city, a com plete change
from in tra m u ra l to ex tra m u ral burial could be dem onstrated
ov er a certain period, th a t change w ould not necessarily be
evid en ce for new beliefs ab o u t d e a th o r the corpse. (W e have
le a rn t th a t the theology o f in h u m atio n an d crem ation need not
be very different.) In fact, on the archaeological evidence at
p re se n t available, such a com plete change cannot be found. If
w e ignore the two in tern al-b u ry in g cities, S p arta and
T a re n tu m , it seem s to be true th a t from the M ycenaean period
o n w a rd s ex tra h a b ita tio n a l burial was everyw here the norm ,
th o u g h by no m eans a n inflexible one. In the M ycenaean period
in te rn a l b u rial w as ra re ;130 in the S ubm ycenaean period and
D a rk A ge, it is attested at A thens, L elkandi, and perhaps
Io lk o s ,131 b u t in each case the evidence m ainly concerns
c h ild re n , a n d th ere is no h in t th a t for adults external burial was
n o t th e rule. A t A thens, for instance, a com plem entary p attern
o f in tra m u ra l child b u rial a n d ex tra m u ral ad u lt crem ation has
b een su g g ested .132 In tra h a b ita tio n a l burial has recently been
claim ed for C o rin th , a n d the A rgolid; but in the case o f the
A rgolid it has been pointed ou t th a t it is extrem ely difficult,
b o th chronologically a n d topographically, to be sure th at when
a p a rtic u la r g rave w as d u g it w as.felt to fall w ithin the living-
sp a c e o f a p a rtic u la r village .133 F or later centuries, a cem etery in
A th en s, on the low er slopes o f the acropolis beside the agora,
129 In M a d a g a sc a r, tribes th a t buried w ithin an d w ithout the settlem ent were equally
a fra id to ap p ro a c h a tom b (A. van G ennep, Tabou el totémisme à Madagascar, Paris, 1904,
66 f.).
130 V . R. d ’A D esborough, The Greek Dark Ages, L ondon, 1972, 276.
131 S n o d g rass, 144 f., 361; D esborough, op. cit., 276 f., cf. 369.
132 S n o d g rass, 144 f.
133 C o rin th : C . K. W illiam s II an d J . E. Fisher, Hesperia 42 (1973), 4. Argolid:
R. H äg g , Die Gräber der Argolis in submykenischer, protogeometrischer und geometrischer Zeit,
i, U p p s a la , 1974, 87—91, a good discussion w hich em phasizes th a t the perception of a
s h a rp d istin ctio n betw een in tern al a n d external burial is a product of urbanization;
even H a g g ’s cau tio u s claim s a re d o u b ted by P. C ourbin, Rev. Arch. 1977, 328.
Birth and Death 73
h a s been th o u g h t to provide im p o rta n t evidence: it contains
a d u lt b u rials from th e late geom etric period to the end of the
six th ce n tu ry , b u t is then a b a n d o n ed . (A few sixth-century
b u ria ls h ave also been found in o th er p a rts of the city .)134 B ut
th is proves a h a rd e n in g o f a ttitu d e s a t the end o f the sixth
c e n tu ry o nly if th e cem etery, w hile in use, was w ithin the old
city-w alls; a n d this is u n c e rta in .135
I n a h istory o f attitu d e s to these pollutions, therefore, th e first
solid ev en t is the act o f P eisistratus, who, according to
H e ro d o tu s, ‘ro o ted his ty ran n y firm ly . . . taking hostages from
th e A th e n ia n s . . . a n d purifying Delos in accordance w ith the
o ra c le s ’.136 (H e rem oved all graves from the area w ithin sight of
th e tem ple.) It w ould be easier to u n d ersta n d his m otives if we
k n ew w h a t ‘th e o racles’ were th a t o rd ain ed it, an d if we could be
su re in w h a t sense th e purification helped to ‘root’ P eisistratu s’
ty ra n n y . A p o in t o f obvious significance is th at, w h atever the
o rig in o f th e oracle?, the effective im pulse to purify Delos cam e
n o t from the D elians them selves b u t from P eisistratus. By this
d isp la y o f co n cern for the sanctity o f the great Io n ian religious
c e n tre , A th e n s’ ru ler stren g th en ed his city’s claim tb general
sp iritu a l p a tro n a g e o f the Io nians, and to a position o f influence
a m o n g th e islan d s a t th e centre o f w hich Delos lay .137
1,4 R. S. Y oung, Hesperia 20 (1951), 6 7 -1 3 4 , esp. 131-3. The ban on intram u ral
b u ria l at A th en s is know n from Gic. F am. 4.12.3.
135 F. E. W in ter, Hesperia, Supplement 19, 1982, 199-204.
136 H d t. 1.64.2.
137 Cf. H . W . P ark e, ‘Polycrates and D elos’, GQ_40, (1946), 105-8. For a different but
n o t irreco n cilab le view see B urkert, GR 310: p a rt o f a process of theological elaboration
ol th e im p licatio n s o f the trad itio n al m o rtal/im m o rtal opposition.

3 sac red w ith o u t im posing a n y re strain ts on sexual activity. W e
TH E WORKS OF APHRODITE
‘A p a rt from E g y p tian s an d G reeks’, says H erodotus, ‘alm ost
th e w hole o f the rest o f m ankind copulate in sacred places and
go in to shrines w ith o u t w ashing after sleeping w ith a w o m an .’ 1
In G reek ideology, therefore, sexual activity is in som e sense
in c o m p a tib le w ith th e sacred. Such activity is, o f course, indis­
p u ta b ly n a tu ra l; for m an an d w om an intercourse is themis, that
w h ich is n a tu ra l a n d rig h t .2 It th u s jo in s b irth and d ea th to form
a trio o f in esca p ab le h u m an processes from w hich the gods
re q u ire in su latio n . As H ero d o tu s indicates, this takes two
form s, physical sep a ratio n (no intercourse in sacred precincts)
a n d lu stra tio n (w ashing after intercourse before entering a
sh rin e ). B oth are well a tte ste d independently. C au tio n ary tales
d esc rib e th e d ra m a tic re trib u tio n th a t strikes those who copu­
late in sh rin e s ,3 w hile a long series o f sacred laws regulates
access to tem ples ‘from a w o m an ’ o r the like. T he earliest of
th ese p erm its im m e d ia te entry after intercourse d u rin g the
n ig h t (passage o f tim e here replaces w ashing as a m ode of
s e p a ra tio n ), b u t requires w ashing after intercourse by d a y .4
1 H d t. 2.64. T h e sta n d a rd collection o f m aterial is Fehrle; there is a sane survey by
H . J e a n m a ire in Mystique et continence, T rav aiix Scientifiques du viic C ongrès Inter­
n a tio n a l d ’A von, Les É tu d es C arm clitaines, 1952, 51—60.
2 H om . //. 9.276.
3 e.g. H d t. 9.116—20 (A ryactes, in P rotesilaus’ shrine); cf. B urkert, H N 72, Fehrle,
242, a n d , for the rule, X en. Ages. 5.7, A lciphron, Epistles, 4.1, Ach. Fat. 5.21.4. But note
p. 76 η. 8 below . T h o u g h th e crim e is com m only located in the tem ple o f a virgin
g o d d ess, the case o f A ry actes show s this not to be essential. R itual origins for such
sto rie s are often suspected (e.g. F. G raf, SSR 2 (1978), 75); such rituals would
th em selv es, how ever, be based on abnorm ality. In Ziehen, n.61 — Buck, n. 64.
S chw yzer, 412 (O ly m p ia , 6 th c.) fornication in a sh rin e ap p aren tly requires penal
sacrifice an d p urification only.
4 L S S 115 A 11 If. (m isin terp reted by Sokolowski). B ut ap p aren tly even alte r w ashing
th e re w as a certain sh rin e from w hich the w orshipper w as d eb arred , line 13. Im m ediate
e n try a fte r w ash in g is allow ed by LSCG 124.9 ( ‘from a w om an’), 55.4.(‘a w om an’),
L S A 18. 9 - 1 3 ( ‘a.w o m an ’), ?51. 10-13, B C H 102 (1978), p. 326 .14 ( ‘from aphrodisia).
Im m e d ia te e n try , a fter w ashing, from licit intercourse, LSC G 139. 14-17 (‘from lawful
in te rc o u rs e ’), I S S 91.17 (‘in terco u rse’). LSA 12. 1 -3 (‘o n e’s own wife or husband )
The Works o f Aphrodite 75
S u ch a rule p ro tec ts th e ideological b a rrie r betw een sex a n d the
see this in th e fam ous scene in Lysistrala w here M y rrh in e is
m ak in g excuses to h er lustful h u sb a n d . ‘[If I yield to you] I
w o n ’t be p u re en o u g h to go back up to th e acropolis.’ N o trouble
a b o u t t h a t’, answ ers K inesias, ‘you can wash in th e C lepsydra
fo u n ta in .’ T h is passage also show s th a t, though m ost o f our
eviden ce co n cerns the p u rification o f m an ‘from a w om an’,
th e re is no difference in the p u rification th at w om an requires
‘from a m a n ’.5
S u ch rules obviously have n o th in g directly to do w ith m oral­
ity. L a te r sacred law s do try to assim ilate them to m oral san c­
tio n s by d istin g u ish in g in p o in t of p u rity betw een licit and illicit
in te rc o u rse (w ith a p ro stitu te o r som ebody else’s spouse), and
e x c lu d in g th e w o rsh ip p e r from th e sh rin e for a period o f days
a fte r illicit con tacts; b u t the early texts speak m erely o f p u rity
‘from a w o m a n ’.6 T h ey are n o t p ro d u cts of asceticism , as they
re q u ire no ab stin en c e; even th e later laws th a t contain sanc­
tio n s ag a in st sexual irregularities norm ally allow the effects of
licit in terc o u rse to be sim ply w ashed aw ay. N or is it easy to see
th e m as expressions o f a stro n g internalized feeling th at the
sex u al act is d eg rad in g or disgusting. H ippolytus, w ho does see
p e rh a p s om its even w ashing. L onger periods o f p u rity are required by LSA 29 .4 —6 (cult
o f Aleter Gallesia, “o n e ’s ow n wife’, 2 d ay s), L S S 54.4 (a Syrian god, ‘w o m an ’, 2 days),
119. 7—9 (u n k n o w n cu lt in E gypt, ‘m en from wom en, and vice versa’, 2 day s), LSCG
171.17 (a p riv ate C o a n fo undation, ‘w o m an ’, 3 days), an d the new regulations (? 2nd c.
AD) for th o se u n d erg o in g incu b atio n in the Asklepieion o f Pergam um (Altertümer von
Pergamon, viii. 3, ed. C. H ab ich t, Berlin, 1969, p. 168, 11-14, ‘aphrodisia\ two days.
R u les for those m erely en terin g the sh rin e were perh ap s less strict, as H . W örrle, the
e d ito r o f th e new law , notes, p. 181). Sexual pu rity is required w ithout specification of a
p eriod by L S C G 95.5 (‘w o m an ’), L SS 59. 15—16 (‘w om an’), 108.1 ( ‘aphrodisia’).
A d d itio n a l im p u rity d eriv es from intercourse w ith a courtesan in LSA 29.7 (an extra
d a y ), LSS 91.18 (30 ex tra days), with som eone else’s spouse LSA 12. 5—6 (an extra
d a y ); for th e stress o n licitness cf. LSC G 139.14. In LSA 18. 13—15 the p ro stitu te m ust
r e m a in p u re for 2 d ay s before entering. P erm an en t exclusion after ‘law less’ contacts in
L S S 91.19, an d the exceptional LSA 20. 25—50. E x tra period o f p u rity required ‘from
d e flo ra tio n ’ L SC G 139.18, LSS 91. 12. A p art from L SS 115 (and, on Sokolowski’s
d a tin g , L S A 29) none o f these texts is earlier th an the 2nd century BC; several are very
late. T h e earliest evidence is ? H om . Od. 8. 364 f.; im purity o f sex, Porph. Abst. 2.50,
4.20.
5 A r. Lys. 912 ί. A few o f the texts in the preceding note also m ake intercourse pollute
b o th p a rtn e rs equally.
6 L a te r sacred law s: n. 4 above. ‘From a w o m an ’: L SS 115 A 11, LSCG 151 A 42.
76 Miasma
sex in this w ay, is not p resen ted by E uripides as a typical Greek.
A n a u r a o f sh am e does indeed su rro u n d sexuality, bu t its source
seem s to be e m b a rra ssm e n t ab o u t bodily functions ra th e r than
g u ilt. It is w ith in the general stru c tu re o f respectful behaviour
a n d d ec o ru m th a t these rules find th eir place. Sex is a private
affair; th o se w ho a re w illing to ‘couple openly’ are characterized
b y this as extrem e b a rb a ria n s .7 K eeping private things private
is a m ark o f social d istan ce or respect; to perform a private act
d e lib e ra te ly in th e presence o f a n o th er indicates either intim acy
o r c o n te m p t. O ld C om edy, a rum b u stio u s a n d sham eless genre,
sp eak s openly Of sexual a n d bodily functions th a t politer forms
o f d isco u rse are a t pain s to conceal. T h e insulation o f sex from
th e sac red is m erely a specialized case o f the general principle
th a t sex u al activity, like o th er bodily functions, requires dis­
g u ise in form al contexts. T h e sym bolic veil th at, by w ashing,
th e w o rsh ip p e r sets u p betw een his sexual activity an d the gods
is a n expression o f respect, ra th e r like p u ttin g on clean clothes
before a p p ro a c h in g a shrine. I f lovers som etim es yielded to the
te m p tin g seclusion of ru stic precincts, they m ay have reassured
th em selves w ith th e th o u g h t th a t the easy-going country deities
w o u ld not sta n d u p o n form alities .8
‘H e sio d ’ reveals sim ilar ideas in their original context. ‘Do
n o t expose y o u r sham eful p arts, w hen you are bespattered with
seed, before th e h e a rth ’, he w arns. W e find here clearly indi­
c a te d th e physical fact th at, in G reece as elsew here, decisively
s h a p e d th e sym bolism th ro u g h w hich attitu d es to sexuality are
ex p ressed . Sex is dirty; it involves a bodily em ission. T h e d irti­
ness is the sam e w h a te v er the m oral statu s o f the act (and may
in d e e d afflict the m ale against his w ill ).9 It is obvious, however,
th a t th e b elief o r assertion th at ‘sex is d irty ’ is seldom u n in ­
7 X e n . An. 5.4. 33—4, cited by D over, 206; cf. G. H enderson, The Maculate Muse, Yale
1975, 3 - 5 . T h e chronological developm ent noted bv D over, 207 does not concern us
h ere.
8 Sed faciles Nymphae risere, V irg. Eel. 3.9. F estivals, an d grottoes, of Pan particularly
in v ite such transgression: Ar. Lys. 911; E ur. Ion. 936—9; R. H crbig, Pan, der griechische
Bocksgott, F ra n k fu rt, 1949, 48 on his Plate xxxv n. 4, ‘Liebesopfer eines ländlichen
P a a re s im H eilig tu m un d in Beisein P ans’; Ael. Ep. 15, cf. A lciphron, Epistle 4.13.16.
B o rg eau d com m ents, 229, ‘L a transgression, d an s ce cas, est rituelle.'
9 H es. Op. 733 f. F or έκμιαίνομαι = ejaculate, voluntarily or involuntarily, see Ar.
Ran. 753 w ith schol. F o r possible pollution by wet d ream s in the C vrene law see p. 342
below .
The Works o j Aphrodite 11
fluenced by th e em otional a n d social significance o f the act to
w h ich it relates: the d irt becom es sym bolic. H esiod’s w arn in g is
a g a in st ‘exposing o n e ’s sham eful p arts when stain ed w ith seed’,
o r m ak in g a n o p en display o f a fact th a t should be hidden; a n d it
is the h e a rth th a t he seeks to p rotect. T his is p artly d ue to
re sp ect for fire, a p u re elem ent w hich is liable, it seems, to
m etap h y sica l co n tam in atio n by this p artic u la r form of d irt; for
th e sam e reaso n , a c h a rac te r in H ip p o n ax ap p aren tly ‘hides the
fire’ before m ak in g love . 10 B ut th ere is also p erh ap s a sym bolic
o p p o sitio n b etw een the h earth , public centre o f th e household,
p lace o f light, a n d th e sexual act, privately perform ed in d ark ­
ness in th e in n e r recesses. A fu rth er co n trast arises from the fact
th a t th e h e a rth is a goddess, a virgin, who sits all d ay at hom e
like an u n m a rrie d d au g h ter. K eeping sexuality aw ay from the
h e a rth is th u s also a w ay o f m ain tain in g , on the sym bolic level,
th e d istin c tio n betw een the m arried a n d u n m arried estate s .11
In resp ect o f sexual p u rity , as o f m any o th er areas o f concern
a b o u t p u rity , ‘H e sio d ’ offers alm ost the only evidence as to how
it affected d aily life. W e are left to w onder, an d to doubt,
w h e th e r resp ect o f this kind for the hearthfire persisted long.
F o r H e ro d o tu s, it is a p eculiarity o f B abylonian spouses th at
th ey purify them selves after intercourse before touching any
h o u seh o ld u ten sil .12 T h e re m ay, how ever, have been everyday
co n tex ts in w hich, in rustic com m unities, sexual p u rity was
re q u ire d . T h e p la n tin g an d harvesting of the olive, bee-keeping,
a n d th e p re p a ra tio n o f food au· tasks to be perform ed, according
to R o m an a g ric u ltu ra l w riters, by children, or the ab stin en t, or
o n ly afte r p u rific a tio n .13 U n fo rtu n ately , we cannot be sure of
th e p ro v en an ce o f these rules. T h e hellenistic ag ricultural
trea tise s th a t to som e ex ten t lie beh in d the R om an w riters had
u n d e rg o n e n o n -G reek influences, a n d cannot be assum ed to
reflect p rim ev al lo re .14 Bees’ a n tip a th y to sexuality, however, is
10 H ip p o n . fr. 104.20 \V., in terp reted by M . L. W est, Studies in Greek Elegy and Iambus,
B erlin, 1974, 143. But for lam p s left alight see Ar. Eccl. 7—9 w ith U ssher’s note.
11 C f. V e rn a n t, Pensée, i. 129-48.
12 H d t. 1. 198.
13 O live: P allad iu s 1.6.14 ( Graeci iubent), cf. Geoponica 9 .2 .5 -6 ; bees: e.g. Colum ella
9.14.3; food: C o lu m ella 12.4.3.
14 C olum ella 12.4.2 asserts C a rth ag in ian influence. O n the lost G reek litera tu re see
E. O d e r, C h . 25 o f F . S usem ihl, Geschichte der griechischen Literatur in der Alexandrinerzeit,
vol. i, L eipzig, 1891.
78 Aliasma
a w ell-attested G reek belief'; there is som e reason to think th at
u n m a rrie d girls m ay have been en tru sted w ith the preparation
o f food, a n d P lu ta rc h p erh ap s records (the text is in doubt) that
p u rity w as re q u ire d in o rd e r to begin the h arv est .15 W e should
th ere fo re co n sid er the R om an rules as a t least possible witnesses
to G re ek p ractice. T h ey ap p e a r to work partly on a level of
sy m p ath y (pure trees a n d anim als dem an d purity o f the farm er),
p a rtly th ro u g h m etaphysical extension o f a requirem ent of
p h y sical p u rity (food m u st be clean), an d partly on the prem iss
th a t im p o rta n t a n d d elicate operations (the harvest) should be
a p p ro a c h e d w ith the sam e respect as is paid to the gods. It is
h a rd to go fu rth e r, w hen the context o f these rules is so insecure.
A n d this is, u n fo rtu n ately , the sum o f o u r know ledge of
d o m estic re q u irem en ts o f sexual purity.
T h e se p a ra tio n betw een religion a n d sexuality w hich we
d iscu ssed earlier is, o f course, very restricted. T his is tru e even
on th e th eo retical level, to say nothing o f the frequent practical
ex p lo itatio n o f festival licence for purposes o f sexual adven­
tu r e .16 T h e gods observed the decencies ,17 bu t m any o f them
w ere sexually active; on e a rth , alth o u g h at some festivals a
sex u al allusion w ould have been as untim ely as in ch u rch today,
to th e effect o f very m an y others it w as central. Sacred activities
in G reece are as a class distinguished from the profane by their
g re a te r dignity, b u t they differ greatly in them selves in sol­
em n ity a n d p ro p rie ty . A festival like the P an ath en aea may
reflect the dignified decorum o f public an d social order, bu t the
m essy m ysteries o f fertility also have their place in religious life.
E ven w ith in the class o f festivals th a t exploit sexuality ex­
plicitly, th ere is g re at divergence; a solem n m ystery like the
sa c re d m arria g e o f D ionysus at the A thenian A nthesteria
clearly differs as m uch in m ood from the deliberately o u t­
ra g eo u s obscenity o f the H aloa as do both from the straightfor­
w a rd in dulgence o f A phrodisia an d the like. It is possible that,
for the m ore frankly hedonistic rites, p urity will not have been
re q u ir e d ;18 b u t the co n tra st betw een the intrinsic dignity of
15 Bees: p. 83 n. 37 below . U n m arried girls: p. 80 n. 25 below. H arvest: Piut. Qiiaest.
Com. 655d, w ith H u b e rt’s note in th e T e u b n e r.
16 e.g. Lvs. 1.20, M e n . Epit. + 5 1 -4 , Phasm. 95 IT., Sam. 3 8 -4 9 .
17 H o rn .’//. 1 4 .330-6, P ind. Pyth. 9.40 !'.
18 C f. Carmen Priapeum 14, an d p. 76 n. 8 above.
The Works o j Aphrodite 79
religious activ ity , a n d the possibly d isrep u tab le ch a rac te r of
a c tu a l rites, p re su m a b ly m ean t th a t G reeks som etim es purified
them selves in respectful p re p a ra tio n for acts they w ould have
b een a sh a m e d to perform in everyday life.
W e h av e so far considered only the req u irem en t to w ash after
in te rc o u rse before en terin g a shrine. Religious rules o f sexual
p u rity w ent beyond this, b u t from this point onw ards, for lack of
d etailed evidence, it becom es h a rd to speak w ith real precision.
T h o u g h P au san ias offers a good deal of inform ation ab o u t
virgin priestesses a n d the like, a congeries o f facts ab o u t the
p ra c tic e o f different p arts o f G reece in the second cen tu ry A D is
a n u n reliab le basis on w hich to reco n stru ct early views on cultic
ch a stity . It is foolhardy to assum e th a t the term s on which
p riesth o o d s in a p a rtic u la r cult w ere held could never change . 19
T h e safest p ro c ed u re is obviously to con cen trate on the com ­
p ara tiv e ly w ell-docum ented case of A thens. Even here, how ­
ever, large g ap s in o u r know ledge m ake a certain vagueness
in ev itab le.
A t A th en s, as elsew here in th e G reek world, m any ritual
fu n ctio n s fell to those who because of th eir age were necessarily
p u re - th e ‘in ta c t’ boys a n d girls o f C atu llu s’ h y m n .20 In post-
classical m edical a n d ag ricu ltu ral w riters, the virgin an d the
‘u n c o rru p te d boy’ a re credited w ith m agical pow ers th a t are
obvio u sly conceived as deriving from p u rity .21 It is not clear,
how ever, th a t it was because o f th eir p u rity th a t such
m in is tra n ts w ere chosen for the classical rites, still less th at
p u rity was co n sid ered m agically effective. O ften it w ould be a
rev ersal o f th e tru th to say th a t the children em body the p u rity
th a t the cerem ony dem ands; on the contrary, the rite’s sole
fu n ctio n is as a stage in the induction o f the children to ad u lt
life. 1n a fam ous passage o f the Ljsistrata, the chorus boast o f the
fo u r ritu al roles th a t they perform ed w hen little girls, as ‘bearers
o f secret th in g s’, ‘co rn -g rin d ers for A th en e the lead er’, ‘bears at
19 F o r D elos see B ru n eau , 63, 504—6.
20 Poem 34, su n g by puellae et pueri integri. E vidence in Fehrle, 112-25.
21 F ehrle, 5 4 -8 ; for th e m uch favoured ‘u rin e o f a n in tact boy’ cf. texts in T.L .L . s.v.
impubes, col. 706 bo tto m , M . W ellm an n ’s edition o f D ioscorides, vol. ii, p. 381, index s.v.
ο ύρο νπ α ιό ύςά φ θ ό ρο υ. Cf. in general H . H erter, ‘D as unschuldige K in d ’Jahrb. f . Antike
u. Christentum 4 (1961), 28—36.
30 Miasma
B ra u ro n ’, a n d ‘b ask et-b earers ’.22 O f these, it is certain that
b e in g ‘a b e a r a t B ra u ro n ’ is a case o f the kind ju s t m entioned;
th e bears w ere no t chaste representatives o f the A thenian
p eople, b u t little girls seeking ritu al protection a n d p reparation
before the on set o f w om anhood. A lthough this is m ore contro­
versial, being a ‘b e a re r of secret th in g s’ was p robably also
o rig in ally one stage in a young g irl’s in itiatio n .23 T h ere is no
re aso n to see the post o f ‘b ask e t-b earer’, attested in m any cults,
as a n in itiato ry survival, b u t the choice o f a m aiden for it seems
to reflect the division o f lab o u r in household cult, w here subor­
d in a te roles a re assigned to the c h ild re n .24 O nly in the case of
A th e n e ’s ‘co rn -g rin d e r’ is it plausible to see chastity as integral
to th e role, since a b u n d a n t com parative evidence is available
for sexual p u rity being req u ired in the p rep aratio n o f food .25 O f
co u rse, p u rity m ay have com e to seem requisite for all these
p o sts, since it is the distinctive characteristic o f the unm arried
girl. T h u s the co u rte san H ab ro to n o n in M en a n d er can jokingly
claim , afte r th ree days abstinence, th a t she is now pure enough
‘to ca rry th e b ask et o f the goddess ’.26 B ut it w ould not have
seem ed a p p ro p ria te to use som e o th er category o f ritually
p u re person (an old w om an, for instance) to serve as basket-
b e a re r. R itu a l functions are divided o u t betw een the different
ag e a n d sex g ro u p s o f society, a n d basket-bearing has fallen to
th e u n m a rrie d girl; this is h er co n trib u tio n to religious life. It is
a q u estio n o f sta tu s differentiation ra th e r th an purity. T he
freq u en c y o f m aid en choirs th ro u g h o u t the G reek w orld has
b e e n p u t in a new light by the d em o n stratio n th a t the chorus
w as th e in stitu tio n th ro u g h w hich young girls were educated for
w o m a n h o o d .27 W e h av e only to read the song com posed for one
s u c h ch o ru s, A le m a n ’s first partheneion, to see th a t the fresh
22 A r. Lys. 641 —7, cf. A. B relich, Paides e Parthenoi, Rom e, 1969, w ith C . Sourvinou-
I n w o o d ,y / « 9 1 (1971), 172 -7 .
25 B urkert, H N 171, w ith references; b ut note th e reservation o f P. Y idal-N aquet in
Faire de l ’histoire, iii, e d .J . Le G off & P. N ora, P aris, 1974, 154. A m agical interpretation
o f th e ir virginity in D eu b n er, 12.
24 Cf. Ar. Ach. 253 f. F o r th eir diffusion see D eubner, index s.v. Kanephoren.
25 P lu t. Quaest. Rom. 85, w ith H . J . Rose, ad loc. an d in Mnemos. n.s. 56 ( 1928). 79 f.
F o r the d a n g e r o f sexual co n tam in atio n o f food cf. p. 99 below on Posidippus, fr. 1.
26 M en. Epit. 440.
2’ C . C a la m e , Les Choeurs de jeunes filles en Grèce archaïque, Rome, 1977, vol. i, passim,
esp. C h . 4.
The Works ojAphrodite 81
ch a rm o f y o u n g creatu res h elped to m ake them m inistrants
especially p leasin g to th e g o d s .28
In th e m y th s to ld by A ttic poets, young people are sacrificed
(o th e gods as well as serving them . O n e text states explicitly
th a t only th e u n m a rrie d can be used for this p u rp o se .29 T h e
e x p la n a tio n m ay be th a t th e p rim ary m yths o f this type, which
estab lish ed th e story p a tte rn , w ere reflections o f the mock-
d e a th s o f in itiates, persons by definition not m a rrie d .30 It is
h a rd to see w hy, in term s o f sim ple pathos, the sacrifice of a
y o u n g b rid e or groom w ould not have been equally effective.
( )n th e o th e r h a n d , since u n m arrie d children are pro p erty of the
fa th e r in a w ay th a t m arried are not, the existing p attern is
necessary to create A g am em n o n ’s (like A b ra h a m ’s) dilem m a.
A t all ev en ts, h ere too statu s seem s a m ore plausible ex p lan a­
tion th a n m ere p u rity .
Som e festivals, though celebrated by the sexually active,
w ere ch arac te rized .b y an em p h atic anti-sexual ethos. T h e m ost
im p o rta n t ex am p le at A thens is th e T h esm o p h o ria .31 M en were
exclu d ed ; c a u tio n a ry tales d escribed the repulse, in one case
even th e c a stra tio n , o f m ale in tru d e rs .32 T h ere is evidence,
p e rh a p s n o t w holly reliable, th a t in som e G reek states all the
p a rtic ip a n ts w ere req u ired to a b stain from sex for a p rep arato ry
p erio d before th e festival, an d it is certainly tru e th a t in A thens
th re e day s ab stin en c e was d em an d ed from th e w om en who
w ere m o st involved in the ritu al. T h e branches or m ats on
w hich th e ce le b ran ts sat cam e from a p lan t believed to have an
28 O th e r A th en ian ritu a l roles reserved lor parthenoi: w ashing A th en a’s im age, at the
P lv n teria (D eu b n er, 18 n. 8); celebrating a pannychis, a t the P an ath en aea (ibid., 25);
m a rc h in g in su p p lica tio n to the D elphinion (ibid., 201 n. 8). Roles reserved lor paides
(th e u b iq u ito u s ep h eb ic processions aside): choirs at T h arg elia and Dionysia (ibid.,
I 9 8 n .2 . 140 n. 1); the O sch o p h o ria (ibid., 143 f.); carrying the eiresiônê (ibid., 1 9 9 n .9 );
being 'b o v from th e h e a rth ' (ibid., 75). O f these, p u rity is m ost likely to be relevant to
w ash in g A th e n a ’s im age, a n d being boy from the hearth.
M E ur. Phoen. 944 f. O n such sacrifices cf. Schw enn, 121-39.
3U The obvious case is Iphigeneia.
31 B urkert, GR 3 6 5 -7 0 ; useful collection o f testim onia in K. D ahl, Thesmophoria, En
graesk Kvindefest, O p u sc u la G raeco latin a 6, C openhagen, 1976; on the anti-sexual ethos
see especially D e tie n n e, Jardins, 151-5; on fem inine self-assertion, idem , Eugénies,
passim, an d on th e d issolution of social o rd er W. B urkert, CQ n.s. 20 (1970). 1-16.
32 H d t. 6. 134.2, A elian, fr. 44, W . B urkert, o p cit., 12. Exclusion o f m en from
tem p les/festiv als o f D em eter a n d K ore is com m on throughout G reece, W ächter, 130 f.,
L S C G 63.10, LSA 6 1 .8 -9 .
82 Miasma
a n ta p h ro d isia c effect.33 T hese regulations seem to have several
levels o f significance. R epelling the m ale is a kind o f sym bolic
p re c o n d itio n for th a t assertion o f independence which, by their
n a tu re , ‘w o m en ’s festivals’ in a m ale-dom inated society repre­
sen t. In som e cases, such as the L em nian fire festival, this
te m p o ra ry rejection o f norm al p attern s o f existence seem s to be
th e rite ’s m ain p o in t.34 A t the T hesm ophoria, however, the
w o m en do no t m erely secede for secession’s sake, bu t because
th e y have specific w ork to do. In the ritu al context, the ideologi­
cal division o f lab o u r betw een the sexes becomes absolute; as
w a r belongs to m en, so fertility belongs to wom en, a n d their
ritu a l lab o u rs w ould be ruined by any contact w ith the male.
S u p erim p o sed on this sim ple antithesis, at least in the case of
th e T h esm o p h o ria , is the idea th at, in o rd er to ensure the
h e a lth y co n tin u an c e o f society, w om an m ust su b o rd in ate her­
se lf to its strictest norm s. T h e T h esm o p h o ria is a festival of
p io u s a n d godly m atro n s, from w hich all disorderliness is ex­
c lu d e d ;35 w om an is here tam ed, strip p ed of the ap p a ra tu s of
” A ll p articip an ts: th e P ythagorean lady philosopher T heano, asked after how many
d a y s w ith o u t in terco u rse a w om an w as ’pure enough to go down to the T hesm ophor-
eion , an sw ered th a t after intercourse with her spouse she was pure at once, and after
in terco u rse w ith anyone else, never (Clem . Λ1. Slrom. 4.19.302. 1 -3 St, T heo, Progymn.
5, p. 9 8 .3 - 7 S pengel, T h e o d o re t, Graec. A jj. Cur. 12.73; the sam e mol w ithout reference
to a specific sh rin e in D .L . 8.43 an d Stob. 4.23.53 H .) By im plication, therefore, ‘going
d o w n to the I hesm o p h o reio n ’ did norm ally require prelim inary abstinence. But the
a n e c d o te , unlikely anyw ay to be pre-hellenistic, refers to no specific com m unity or
festival (on T h e a n o ’ cf. v. F ritz in R E s.v. Theano, col. 1380). O v id ’s Ceres festival with
9 d a y s of ab stin e n ce (Met. 10. 4 3 4 -5 ), often quoted in this context, in fact reflects the
R o m an Sacrum Anniversarium Cereris (H. Le B onniec, Le Culte de Céres à Rome, Paris. 1958,
4 0 8 - 1 0 ); this cerem ony is G reek in origin (Le Bonniec, 386 f.), b ut probably derives
from o n e o f the extended D em eter festivals o f M ag n a G raecia (ibid., 4 2 0 -3 ) ra th e r than
d ire c tly from th e 1 h esm ophoria. Ar. fr. 317, cf. 329, indicates abstinence at the
T h esm ophoria, possibly p rep arato ry . 3 days abstinence: schol. L ucian 276.5 Rabe
(D e u b n e r, 40 n. 5). A n tap h ro d isiac plants: F ehrlc, 139—54.
34 Cf. B u rk ert, loc. cit. B urkert suggests the sam e (or the Skira (H N 164), a t which
w om en chew ed garlic to keep th eir m en aw ay (Philochorus 328 FG rH fr. 89), b ut they
m ay h av e had positive work to do. The exclusion o f m en from D ionysiae rites (W ächter.
132) is, of course, rebellious. It is alm ost alw ays from rites o fD e in e te r an d Dionvsus
th a t m en are sh u t o u t (W äch ter, 1 30-3). F or the concept o f ‘w om en’s festival' cf. LSCG
36. 8 - 1 2 , Ar. Tkesm. 834 f. O n the religious role o f wom en cf. Kur. Melanippe Desmotis. fr.
6. 12—22 v. A rnim .
35 ^ r - Thesm. 330, cf. O tû tn n e , Jardins, 152, Eugénies, 196 f., quoting C allim . fr. 63.
1~ a g ain st B urkert to prove exclusion ofparlhenoi. E xclusion o f non-slave concubines
is n ot certain , how ever: cl. M en. Epit. 749 f. (m ore im portant than L ucian, Dial. Meret.
2 .1), on w hich D eu b n er, 54 is special pleading. A thenian exclusiveness w as probably a
The Works o j Aphrodite 83
sex u al a ttra c tio n ,36 forced to sit, fasting, on the h ard ground.
T h e c e le b ra n ts o f th e T h esm o p h o ria are term ed ‘bees’, the pure
ty p e o f id eal w o m an h o o d .37 (T h e respectable bees are, of
cou rse, likely to h ave relished th e conspicuous distinction m ade
a t th e T h e sm o p h o ria betw een them selves a n d row dy dog-
w om en o f d u b io u s stock.) A final layer of significance derives,
p e rh a p s , from co n trast. S exual abstin en ce is re q u ired before
a n d d u rin g th e T h esm o p h o ria precisely because, w ith o u t sexu­
ality , th e re can be no fertility. T h e ritu al focuses atten tio n on
th e id ea o f p ro d u c tiv e sexual union by a paradoxical tem porary
in sisten ce u p o n its opposite. E verything m arks th e period of
ab stin e n c e as ab n o rm al; virgins, w ho are p erm an en tly pure,
h av e no p a rt in th e rites .38
A sim ilar a rg u m e n t allows us to see an o th er A thenian
w o m en ’s festival, th e H alo a, as affirm ing the sam e m oral norm s
as th e T h esm o p h o ria , alth o u g h by opposite m eans. T h e tone
h ere w as licentious; priestesses w hispered to the m arried women,
u rg in g th em to a d u lte ry .39 T h e festival th u s challenged the
ru les, b u t n o t w ith in ten t to overthrow them ; once the festival
w as over, th e rules reasserted th eir claim s w ith renew ed
insisten ce.
F o r th e logical counterpoise to such w om en’s rites, we m ust
look o u tsid e A th en s. A t several places in G reece, w om en were
special dev elo p m en t: co n trast for E retria L. D oria, Cahiers du centre Jean Berard 5,
N ap les, 1979,621'.
36 S acred law s from P eloponnesian cults o f D em eter T hesm ophoros or sim ilar god­
desses b a n em b ro id ered robes, p u rp le robes, m ake-up, gold o rnam ents: LSS 32, 33, ?
28, L S C G 68, 65. 1 6 -2 3 . Such g arb denotes the prostitute: P hylarchus 81 FG rH fr. 45,
D iod . 12.21.1, C lem . AI. Paed. 2.10. p. ‘2 20. 6 - 9 St. Schol. Soph. O C 680 records: φάαι
τά ς θεάς άνθίνοιςμή κεχρήσθαι άλλά καί ταΐς θεσμοφοριαξούααις τήν των άνθινων στεφάνων
άπειρήσθαι χρήσιν ; it is tem p tin g to su ppose th a t the ban extended to ‘flowered robes’,
ty p ical d ress o f the p ro stitu te (Sud. & Phot. s.v. έταιρ&ν άνθινων). T h e long list of
fem ale g a rm e n ts in A risto p h an es’ second Thesmophoriazousai (fr. 320, cf. 321) perhaps
relates to su ch rules.
37 A p o llo d o ru s 244 F G rH fr. 89, cf. L. Bodson, Ιερ ά Ζώια, Brussels, 1978, 25 ff., for
bees a n d D em eter. Bees a n d sexual virtue: M. D etienne, ‘O rp h ée au m iel’, in Faire de
! 'histoire. e d .J . L e G o ffa n d P. N ora, iii, P aris. 1 9 7 4,56-75, H. F. N orth , Illinois Classical
Studies 2 (1 9 7 7 ), 3 5 -4 8 . Cf. the oath o f m arital fidelity in LSCG 65.8. Such ideals are not
confin ed to D em eter cu lts, however: cf. the skolion PM G 90 i.
38 Cf. Λ. D . N ock, ‘E u n u ch s in A ncient R eligion’, A R W 2 3 (1925), 2 5 -3 3 , reprinted
in N ock, i, 7—15.
39 Schol. L u cian 280. 16-17 R abe (D eubner, 61 n. 5).
84 Miasma
ex clu d ed from the cu lt o f H eracles, while, in Phocis, the hero
b o re the title ‘w o m an -h ater ’.40 W e have no evidence about
sex u al restrictio n s im posed on o rd in ary p articip an ts in these
cu lts, b u t a t Phocis the priest was bound, exceptionally, to a
y e a r’s celibacy. A t first sight there is a contradiction between
th e lusts o f the m ythological H eracles an d the m isogynism of his
cu lt. I t can p e rh a p s be resolved by seeing H eracles as the hero
w ho perform s m ale activities, including seduction an d procrea­
tio n , su p rem ely well, b u t requires protection from certain kinds
o f fem inine influence to preserve his excellence. Even
m ythologically, H eracles is u n d e r th rea t from wom en; a god­
dess dogs him , a q u een enslaves him , his wife ‘m an -slau g h terer’
ev e n tu a lly destroys him . T h ro u g h H eracles, we discern a m ore
g en e ral sense o f m asculine force endangered by the arts of
w om en; the id ea is com prehensible on the level o f the seductress
w ho u n m an s m en, o r the wife w ho poisons them , b u t also in the
term s o f Book 6 o f the Iliad.*1 Iso lated from such dangers, the
w o rsh ip p e rs o f H eracles the w o m an -h ater p rep are them selves
for th e w ork o f m en. A ctual sexual abstinence in p re p ara tio n for
h u n tin g a n d w arfare is not dem o n strab le in the historical
p erio d ; w here ab stin en ce is attested , as for athletes, it can
p e rh a p s be ex p lain ed p rag m atically .42 It is, how ever, probably
sig n ifican t th a t the sep aratio n o f the sexes was particularly
em p h asiz ed in connection w ith certain characteristically
m ascu lin e activities: w om en m ight not set foot in the council
h o u se o f a t least one G reek state, or in the stadium at
40 L S S 63, LSA 42 A, Ael. N A 17.46, Plut. De Pyth. Or. 20, 4031' (Phocis). Cf. L. R.
F a rn e il, Greek Hero Cults and Ideas o f Immortality, O xford, 1921, 162 f.; C h. Picard, B C H
47 ( 1923), 2 4 6 -9 ; B. B ergquist, Herakles on Thasos, U ppsala, 1973, 85.
41 S eductress: H om . Od. 10.340 f. Poisons: E ur. fr. 464, M en. fr. 718.9. A ntiphon 1,
D em . 25.79. Iliad 6: J . K akridis, Homer Revisited, L und, 1971, C h. 3, Griffin. 6,
VV. S eh ad ew a lt, Von Homers Welt und Werk4, S tu ttg a rt, 1965, 207—33.
42 W ar: pace G . M u rra y , The Rise o f the Greek Epic4, O xford, 1934, 133. Hes. Scut. 14-22
is a r a th e r different case. H u n tin g : B urkert, S H 118, cf. H N 72 n. 12, sees a reflection in
e.g. H ip p o ly tu s; b u t th e w ell-attested link o f h u n tin g and virginity in G reek m ythology
seem s ra th e r to reflect th e values and activities o f an age set, cf. M . D étienne, Dionysos
mis à mort, P aris, 1977, C h. 2. Athletics: A eschylus Theori/Isthmiastae, 29—31, with
L lo y d -Jo n e s’s co m m en t, L oeb A eschylus vol. ii, p. 544; PI. Leg. 839e—840a: B urkert,
H N 117 n. 43.
The 14 orks o f Aphrodite 85
O ly m p ia .43 I t w as p ro b ab ly for sim ilar reasons th a t women
w ere som etim es excluded from th e cults of Poseidon, Zeus, and
Ares, all em p h atically m asculine gods. A b rief en try in a sacri­
ficial c a le n d a r from M ykonos is revealing: ‘T o Poseidon
Phvkios, a w h ite lam b w ith testicles. W om en not ad m itte d ’.44
W hile these cults th a t em p h asized sexual division w ere quite
com m o n , it w as only seldom , to ju d g e from the surviving evi­
d en ce, th a t th e lay m an was re q u ired to keep him self pure in
p re p a ra tio n for a festival. A p a rt from the T h esm o p h o ria, only
tw o in stan ces are attested early, an d even these, like the
T hesm o p h o ria on the m ore sceptical in terp retatio n , concern
not th e w hole body o f p artic ip a n ts, b u t restricted groups who
w ere to play an im p o rta n t p a rt in th e ritual. At th e A th en ian
A n th e ste ria , th e w om en w ho p re p are d the archon basileus' wife
lor h er sacred m arriag e to D ionysus swore th at they w ere ‘pure
from u n clea n things in general, an d especially from intercourse
w ith a m a n ’.45 T h e re seems to be a p aro d y of o ath s o f this kind in
th e Lysistrata, a n d they m ay well have been m uch m ore frequent
th a n we k n o w .46 In the attested case, the sym bolic point was
su rely to keep the m ystic union w ith the god free from all tain t of
m erely h u m a n sexuality. For the sam e reason, in m yth, gods
alw ay s chose virgin b rid es .47 D u rin g the festival o f Zeus Polieus
43 C o u n cil house: A th. 150 a (N au cratis); S. G. M iller, The Prytaneion, C alifornia,
1978, 11, sta te s th a t the rule applies m ore widely, but w ithout citing evidence. If
m a g istra te s celeb rate A ph ro d isia at the end o f their term (X en. Hell. 5.4.4), th at is
p a rtly because A p h ro d ite is p atro n ess o f m agistrates, but also an expression o f re­
sp o n sib ility laid a sid e (P lu t. Comp. Cim. et Luculi. 1.3, cf. F. C ro issan t an d F. Salviat,
B C H 90 (1966), 4 6 0 - 71 ). S tadium at O lym pia: W ächter, 126 (m aidens were adm itted
b u t n o t m a rrie d w om en, Paus. 6.20.9).
44 L S C G 96.9. W om en excluded from cults o f Zeus: LSCG 109, L SS 88b, 89 (the
ex cep tio n al exclusion o f wom en from the cult o f A thene A potropaios in L SS 8 8 a, b
seem s to deriv e from h er close relationship to Z eus). From cult o f Ares: Paus. 2.22.6 f.,
w ith N ilsson, G F 408. From the sh rin e of the Anakes at Elateia: LSCG 82. The
c h a ra c te r o f th e violently m isogynist hero E unostos (Plut. Quaest. Graec. 40) is uncer­
tain . U n c e rta in cultï'2-SCG’ 124. 18-20. A few further exclusions, W ä c h te r, 126—9; cf.
H a llid a y on P lut. Quaest. Graec. 40.
45 (D em .) 59.78; on the m arriage, B urkert, H N 25 5 -6 3 . O n abstinence by laym en cf.
F ehrle, 1 2 6 -1 5 4 , a n d p. 82 n. 33 above.
46 A r. Lys. 181 —237, no te esp. the hieratic w ord άταύρωτος in 217. A nd in M en. Epit.
440 αγνή γάμων γάρ, φασίν, ημέραν τρίτην\ήόη κάθημαι, note esp. φασίν.
47 Cf. P. M a as, Kleine Schriften, M unich, 1973, 66. T h e m others o f Plato and Alexan­
d e r w ere, in p o p u la r story, avoided by th eir hu sb an d s after the divine visitation th at
sow ed the fam ous sons, D .L .3.2, Plut. Alex. 2.6, Fehrle, 3.
86 Miasma
on C os, the citizen a p p o in ted to sacrifice the bull was required
‘to keep p u re for a night from w om an a n d [? m a n ]’.48 In this
case, it is h a rd to find an explanation in the c h a rac te r of the
ritu a l, beyond the fact th a t this was a sacrifice o f high dignity
a n d im p o rtan ce. If m ore evidence were available, we m ight find
th a t p re lim in ary abstinence could be im posed on any laym an
w ho w as to p a rtic ip a te significantly in a ritual o f especial sol­
em n ity , w h atev er the source o f th at solem nity m ight be. It is
very plausible, for instance, th at there were rules o f this kind for
th e E leu sin ian initiate; no trace, however, rem ains.
A b stin en ce was pro b ab ly som etim es observed in response to
o racles or o th e r divine signs. W hen he consulted D elphi about
his childlessness, A igeus was told not to indulge sexually for a
fixed period, a n d it is not im plausible that the m ythological
resp o n se reflects ac tu a l o ra cu lar p ra ctice .49 By the logic of
c o n tra st, th ere is an obvious suitability in refraining from sexual
c o n ta c t as a p re p a ra tio n for procreation. Religious fear m ight
lead to ab stin en ce, if we accept the im plications o f the story that
K in g Agis sh u n n ed his wife for ten m onths after an earth q u ak e
in th e n ig h t .50 B ut in such cases it is not purification from the
ta in t o f sexuality th a t is desired.
I f ab stin en c e is som etim es required o f laym en perform ing
priestly functions, it m ight afortiori be expected o f priests. T h e
id ea o f religious ab stin en c e was certainly a fam iliar one. W hen
E u rip id e s’ E lec tra reveals th a t h er husb an d has never a p ­
p ro a c h e d her, h e r b ro th e r asks at once ‘Is he u n d er som e sacred
re q u ire m e n t o f p u rity ?’51 It should be stressed a t once, however,
th a t ‘sacred re q u irem en ts o f p u rity ’ th a t im posed long periods
o f ab stin e n c e w ere exceptional. In the classical period, m ost
p riests a n d priestesses th ro u g h o u t the G reek w orld w ere either
m a rrie d people co n d u ctin g norm al family lives, w ho m ay at the
48 L SC G 151 Λ 42—4 (only the ‘sla u g h te rer’ is so bound, not all particip an ts, pace
F eh rle, 155 n. 1). A ccording to Nilsson, G f 21, chastity is required because this is a
g u ilty sacrifice, like the A thenian B uphonia. Is the link w ith H estia (lines 19. 2ii)
relev an t?
49 E u r. Med. 6 6 5 -8 1 , Plut. Thes. 3.5.
50 P lut. Ale. 23.9, Ages. 3.9; the story is based on a m isu n d erstan d in g o f Xen. Hell.
3 .3.2, b u t m ight none th e less reflect a real possibility.
51 Kur. El. 256, cf. Tro. 501; cf. Fehrle, 75-111 (in terp retatio n s very dubious). O n
G re e k p riests see S tengel 3 1 -4 8 , Z iehen in R E s.v. Hiereis, Burkert. GR I5 7 -6 3 : a
m o n o g ra p h is req u ired .
The Works o j Aphrodite 87
m ost h av e been b o u n d to tem p o ra ry periods o f chastity, or
m a rrie d people p ast the age o f freq u en t sexual activity. It is
becau se they are not the ru le th a t we h ear specifically from
P au san ias o f ‘virgin priestesses’ a n d th e like. A ny m ore rigorous
re q u ire m e n ts w ould be su rp risin g in a society w here sacred
functio n s w ere often a tta c h e d to political office, a n d ten u re of a
p riesth o o d seldom form ed the cen tre o f the h o ld er’s existence.
A b o u t th e m arital statu s req u ired for the m any priesthoods
in A th e n s ,52 explicit evidence is alm ost entirely lacking. O f the
careers o f in d iv id u a l in cu m b en ts we som etim es know som e­
thin g , b u t in th e case, for instance, o f a statu e o f a priestess
d e d ic a te d by h er son, it is im possible to be su re w h eth er she
a lre a d y h eld th e office d u rin g h er child-bearing period. Plato
a n d A risto tle agree th a t in th e ideal state priesthoods should be
assig n ed to th e elderly. P lato ’s specification is p articu larly in ­
terestin g : ‘T h e m an who is going to be sufficiently p u re for
d iv in e service, as sacred laws req u ire, should be over 60 years
o ld .’53 A p riesth o o d , therefore, m ight im pose req u irem en ts of
a b stin e n c e w hich a younger m an w ould find h a rd to observe
a n d , as P lato does not seek to ju stify o r explain such req u ire­
m en ts, they m u st have been fam iliar in A thenian practice. His
re m a rk does not in d icate in itself w h eth er abstin en ce w ould be
ex p ected th ro u g h o u t ten u re o f the office (a year, by P lato ’s rule)
o r m erely for a few days in p re p ara tio n for p artic u la r cere­
m onies. A y e a r’s ab stin en ce for a m ale is once a ttested , b u t that,
as w e h av e seen, is outside A ttica, a n d in the fiercely anti-sexual
c u lt o f H eracles ‘w o m an -h ater ’.54 Since m any priesthoods were
h eld ‘for life’, it is m ore a ttra ctiv e to suppose th a t they entailed,
a t m ost, sh o rt p eriods o f p u rity . Eleusis is the only cult for w hich
re stric tio n s o f this kind are m entioned in the sources, a n d there,
th o u g h th e view th a t th e h iero p h a n t was b o und to p erm an en t
c h a stity from th e m om ent o f taking up office can n o t be formally
re fu ted , it is m o re plausible to suppose, since he could retain his
wife w hile in office, th at he w as sim ply req u ired to rem ain
52 The a n tiq u a te d book of J . M a rth a , Les Sacerdoces Athéniens, Paris, 1882, has not been
rep laced , ex cep t for Eleusis, on w hich see C linton. For priestesses cf. H. M cClees, A
study o f women in Attic Inscriptions, diss. C olum bia, 1920, 5 -1 6 , 45; Jo rd a n , 2 8 -3 6 .
53 PI. Leg. 759d, A rist. Pol. 1329a 2 7 -3 4 . P lato specifies the sam e age for priestesses.
54 P lut. De Pyth. Or. 20, 403f.
88 Miasma
c h a ste for a p erio d before the m ysteries .55 D etails, unfortu­
n ately , are no t av ailab le eith er for this or for any o th er cult in
A ttic a o r G reece as a w hole .56
As th e typical im age o f the priest in A ttic literature is of an old
m a n , it is likely th a t P lato a n d A ristotle, in their preference for
th e old, a re reflecting a tendency o f A thenian p ra ctice .57 But we
h e a r explicitly o f aged m in istran ts only in the case o f A th en a’s
sa c re d lam p, te n d e d by w om en w ho h a d ‘finished w ith sex ’,58
a n d th ere is no d o u b t th a t A th en ian s could be app o in ted to
p rie sth o o d s w hile still in th eir sexual prim e. Lysim ache,
p riestess o f A th e n a Polias for sixty-four years, obviously took up
office w hile still young; she alm ost certainly h ad offspring, and
it is scarcely p lausible to assum e th a t the sixty-four years of
office b egan only after h er child-bearing days w ere over .59 T he
first priestess o f A th en a Nike p erh ap s held office for alm ost half
55 cf. F o u cart, 171 —3; TopfFer, 54; B urkert, H N 313; C linton, 44. T h e relevant facts
are: (1) v ario u s texts, o f w hich A rrian, diss. Epict. 3.21.16 is the earliest, refer to the
h ie ro p h a n t’s ab stin e n ce, often referring it to the use o f an tap h ro d isiac drugs; (2) Paus.
2.14.1, in a list o f différences betw een the h iero p h an ts o f Phlius an d Eleusis, m entions
th a t th e form er m ay ‘take a wife, if he w ishes’; it is h ard to see why Paus. includes this
p o in t unless it co n stitu tes o n e o f the differences. N ote, however, th at P aus. speaks of
‘ta k in g ’ a n d n o t ‘h a v in g ’ a wife; (3) num erous children o f hierophants are attested; they
m a y , o f course, h ave been begotten before th eir father was hierophant; (4) I G II2 3512
show s th a t a h ie ro p h a n t could have a wife while in office. (2) an d (4) are readily
reconciled on th e view th a t a hiero p h an t could retain a wife while in office but not
a c q u ire one. I t is conceivable, though scarcely credible, th at the hierophant could have
a v o id e d all sexual c o n ta c t w ith the wife he retained; b u t A rrian, loc. cit. uses the term
ά γνενω , n o rm ally ap p lied to tem porary abstinence. F or the idea οϊπροειρημένον ήμερων
άριθμόν άγνενειν see D em . 22.78, quoted p. 97 below.
56 N o n -A ttic evidence for ad hoc priestly hagneia is alm ost non-existent. N othing in
L S C G 154 A, 156 A 1 -1 6 ; LSC G 156 B 2 9 -3 5 m ay have treated the subject, b u t is
b ey o n d reco n stru ctio n ; L SC G 83.40 is vague (as is LSA 79.6).
*7 S oph. O T 18, a n d in general H om . //. 1. 26, 6. 29 8 -3 0 0 (A ntenor, the priestess's
h u s b a n d , is a όημογέρων, 3. 149), Hes. fr. 321. Cf. C linton, 44.
58 P lut. Num. 9.Î 1; on th is lam p cf. R. Pfeiffer, Ausgewählte Schriften, M unich, I960,
4 —7. B u rk ert, w ith o u t arg u m en t, seems to identify these aged atte n d a n ts o f the lam p
w ith th e a c tu a l priestess o f A thena Polias (in whose tem ple the lam p w as), H N 168 n.
59, G R 337; b u t the case o f e.g. L ysim ache, discussed in the text, refutes this. O ne m ight
r a th e r see th e la m p ’s a tte n d a n t in the kind ol aged διάκονος to the Polias priestess
m e n tio n e d P aus. 1.27.4 (cf. IG I I 2 3464). F or priestly office held by those who have
‘d o n e w ith sex’ cf. F eh rle, 95 n. 1, Paus. 2.10.4, 7.25.13, G RBS 14 (1973), 6 5 -7 3 .
59 64 years: Pliny, N H 34.76. O ffspring: IG I I 2 3453, w ith D. M . Lewis, ABSA 50
( 1955), 4—6, w ho ib id ., 7—12 collects the evidence for priestesses o f A thena Polias (cf.
D avies, 1 7 0 -3 ). T h e d a u g h te r o f Polyeuktos (Lewis, n. 4) had a husband while
p riestess ( I G I I 2 776. 2 2 -3 0 ); the h u sb a n d a t som e d a te h ad a son, presum ably by her
(IG I I 2 5610, D avies, 72). C hrysis (Lewis, n. 10) had ‘d escendants’, IG I I 2 1136.15. In
fav o u r o f seeing th e P olias priestess as post-sexual, there is only (Plut.) Λ\ orat. 843b on
The Works o f Aphrodite 89
a ce n tu ry , a n d th e reg u latio n th a t established her post con­
tain ed no specification ab o u t m arital s ta tu s .60 It has even been
a rg u e d th a t A risto p h an e s’ L y sistrata a n d M yrrhine, the latter
a t least w ed d ed to a d em an d in g h u sb an d , are none o th er than
th e p riestesses o f these two cults of A th e n a .61
T h e tw o m o st im p o rta n t m ale officials of the E leusinian cult,
th e h ie ro p h a n t a n d the d ad u c h , could both be m arried,
a lth o u g h th e h iero p h a n t was p ro b a b ly not p erm itted to take a
new wife w hile in office. Priests in o th er cults, it m ay be
assu m ed , will n o t have been m ore restricted th a n the hiero­
p h a n t .62 T h e ch ief female officials a t Eleusis, the priestess of
D e m e te r a n d K o re an d th e h iero p h an tid s, h ad norm ally been
m arrie d , a n d , th o u g h it is not dem o n strab le th a t their
m arriag es co n tin u ed while they held office, there is no positive
evid en ce to th e co n trary ; th e fact th a t they m ight live in special
‘h o u ses o f th e p riestess(es)’ does not seem necessarily to exclude
th e p resen ce o f a h u s b a n d .63 T h e priestess o f Nem esis at
R h a m n u s could be a m other, b u t here too it is possible th a t she
h a d ‘finished w ith sex’ before assu m in g office .64 F o r the cult of
D e m e te r T h esm o p h o ro s, how ever, a n honorific decree fortu­
n ately pays trib u te to a lady in d isp u tab ly eq u ip p ed with a
h u s b a n d w hile serving as priestess .65
P h ilip p e (Lew is, n. 11) w ho ‘afterw ards becam e priestess o f A thene, but before th at
D iokles m a rrie d h e r a n d begot . . In view o f th e o th er evidence cited, th at m ust
rep re se n t a n isolated case, or a t m ost a later developm ent. O th e r m arried wom en
involv ed in A th en e cult: S E G xxiv 116, IG I I 2 2342.31.
60 F irst priestess SE G xii 80; regulations IG I ' 35 = M /L 44. For the chronological
p ro b lem s see M /L . O r w as this post annual?
61 D. M . Lewis, A USA 50( 1955), 1—7; note how ever K .J . Dover, Aristophanic Comedy,
L o n d o n , 1972, 152 n. 3. J o r d a n believes (35) th at A thena Partheiios m ust have had a
d istin c t priestess, a virgin.
62 H ie ro p h a n t: p. 88 n. 55 above. D aduch: C linton, 67. T h ere rem ains, as a possible
E le u sin ian celib ate, th e m ysterious ίερενς παναγής (C linton, 95 f.), b u t we can only
guess w h at restrictio n s governed him , an d his very existence before the late 1st century
BC: relies o n a resto ratio n in IG I 3 6 C 48. F or o th er A thenian priests who had wives
w hile in office cf. e.g. IG I I 2 3629, 4076 (exegete), 4851.
63 Priestesses o f D em eter an d K ore w ith children: nn. 3, 6, 10, an d 16 in C linton's
catalo g u e (68—76). H iero p h an tid es w ith children: nn. 3 ,4 ,5 , 7, 10, 11 (8 6 -9 ). Houses:
C lin to n , 20, 71 (th e celib ate P ythia had a special house at Delphi, Parke/W orm ell 44 n.
84). Cf. n. 65 below.
64 I G I I 2 3462. F o r a priestess o f Helios d edicated by her son cf. ibid., 3578.
65 Hesperia 11 (1942), p. 265 n. 51, whose evidence rem ains im p o rtan t even if th e cult
is o f a dem e an d not, as its ed ito r thinks (2 7 0 -2 ), o f the state. IlC lin to n , 71, were right
th a t this p riestess o f th e T hesm ophoroi w as none o th er th an the priestess o f D em eter
90 Miasma
In c o n tra st to th e considerable evidence for those who were
o r h a d been m arrie d , it is very h a rd to identify virgin priests or
priestesses a t A thens. T h e priestess o f A rtem is a t B rauron was
p e rh a p s one, b u t m ight, if so, have renounced her office at
m a rria g e .66 V irgin priestesses o f A rtem is reflect the n atu re of
th e goddess a n d , m ore p articu larly , o f her young adherents;
ra th e r th a n reflecting a general ideal o f virginity, they em body
th e values o f a p a rtic u la r age se t .67 T h e priestess o f D em eter
T h e sm o p h o ro s has now been released, as we saw, by inscrip-
tio n al evidence from the p erp etu al virginity to w hich a scholion
on L u cia n co n d em n ed her; a n d scholars have been too hasty in
esta b lish in g a com m unity o f celibate priestesses at Eleusis .68 A
a n d K ore (b u t w hy th en the different title?), the problem about the m arita l status o f the
la tte r w ould be resolved. A t all events, the new text refutes the wild fantasies (surpris­
ingly acce p te d by B urkert, G R 368) o f schol. L ucian Dial. Meret. 7.4, p. 279. 21 Rabe,
a n d Timon 17, p. 11‘2 .5 R. L ucian him self in Tim. 17 (the other passage is vaguer) refers
to a m an w ho does n ot ap p ro ach his young bride, b u t neglects her, καθάπερ ιέρειαν τή
Θ εσμοφόρφ τρέφων ôià παντός τοϋ βίον. I am not sure th at this im plies m ore than
te m p o ra ry ab stin e n ce ( ‘for life’ can be referred to the boorish m an, not the priestess),
b u t even if it does, it will not m ake the priestess o f T hesm ophoros a virgin. Anyway.
L u c ia n surely allu d es n ot to A ttica but to the kind o f rite th at lay behind the Sacrum
Anniversarium Cereris (p. 82 n. 33), in w hich sim ilar regulations for the priestess are found
(u n less indeed we a re d e a lin g w ith a late developm ent th at alfected the cults both ol
‘C e re s’ a n d ‘T h e sm o p h o ro s’). F u rth e r evidence for priestesses m arried while in oilice:
IG 112 1316 (cf. SE G xvii 36), 3607, 3725, Posidippus, fr. 26.21 ap. A th. 377b.
66 There is no d ire c t evidence (for the allusions to h e r d . J o rd a n , 34; the identification
o f a m a rrie d in cu m b en t in CJ 74 (1979), 361 is unfounded.) Indirect evidence, not
conclusive, com es from E u r. I T 1462— 3 (the virgin Iphigeneia to be first priestess, cl.
130—1). H er responsibilities (L inders, 52 f.) seem too g reat for a young girl. IG 112 2874
gives a lifelong priestess o f A rtem is, probably in the cult a t O inoe (cf. S. Solders, Die
ausserstädtischen Kulte und die Einigung Attikas, L und, 1931, 30).
67 O n virgin priestesses o f A rtem is cf. Fehrle, 9 8 - 1 0 2 - whose arg u m en t, however,
th a t th e presence o f virgin m inistrants proves the goddess to have been originally a
fertility m o th e r is q u ite m isguided. T h is relation o f contrast betw een god and servant
do cs oc c u r (P aus. 2.10.4 is th e paradigm case, cf. Burkert, GR 162), b ut perhaps not lor
c o n tr a s t’s sake; F eh rle’s ow n m aterial shows virgin priestesses to be m uch com m oner in
th e c u lt o f virgin goddesses, m arried w om en observing abstinence, or wom en finished
w ith sex’, in th e cu lt o f D em eter an d the like. U . Pestalozza, Religione mediterranea,
M ila n , 1951, 2 3 5 -5 9 ( = S M S R 9 (1933), 173-202) has wild speculation on ‘S aeerdotie
S acerd o tesse im p u b eri nei culti di A thena e di A rtem ide’.
68 Pollux 1.35 re a d s in B ethe’s text: ιεροφάνται, όςιόοϋχοι, κήρνκες, σπονόοφόροι,
ίέρειαι, παναγείς. T h e co m m a before παναγείς should probably go, as there is nothing
distin ctiv ely E leu sin ian a b o u t ϊέρειαι, but 2 M SS o f Poll, give not ίέρειαι b u t ιερείς, and
th is sh o u ld surely be acce p te d , as the ΐερεϋςπαναγής is inscriptionally attested (Clinton,
96), a n d πα ναγής for Pollux ( 1.14) is a m ale title. E ntries like Hesych. παναγείς ΑΘήνηοι
ϊέρεια ι, ibid., πα ναγία ιέρεια ήτις ον μίσγεται άνόρί (cf. Foucart, 214 η. 5), seem to be
The Works o j Aphrodite 91
fine im ag in ativ e p o rtra it o f a virginal tem ple serv an t we do
have, how ever, in E u rip id es’ Io n .69 T h is fresh, innocent, d e­
vo ted y o u th w arn s us ag ain st in terp re tin g the ideal o f purity, in
the case o f th e young, too narrow ly. In Ion, an in tact body
houses a m in d u n co n tam in ated byjoyless an d cynical thoughts.
Io n ’s ch a ste tem p le service, how ever, is no m ore th an a stage in
his life.
A fter this su rv ey o f the evidence, som e ten tativ e conclusions
a b o u t cu ltic ch astity can now be d raw n . A possible line of
in te rp re ta tio n w ould be s tru c tu ra l .70 T h e G reeks, like o th er
societies, d iv id ed th e affairs o f the w orld into sacred a n d p ro­
fane, a n d this b asic dichotom y was n atu ra lly extended into the
sex u al sp h ere. P rofane life is, necessarily, sexual; to ap p ro ach
th e sacred m en m u st therefore becom e asexual. A m inim um
division from the profane is achieved by the w o rsh ip p er who
w ash es before access to the tem ples, a m axim um by the lifelong
virg in priestess. Such an acco u n t is p ro b ab ly an accu ra te d e­
sc rip tio n o f som e p a rt o f a G reek ’s sentim ents: the closer a
m o rtal com es to sacred objects, the m ore ac u te becom es his
n eed for sexual p u rity .71 I f an ex p lan atio n is needed as to why
sex u ality is d ra w n into the co n tra st betw een sacred an d profane
a t all, it m u st lie in th a t em b arrassm en t ab o u t bodily functions
discu ssed earlier, w hich n atu rally aligned sexuality w ith the
less h o n o u ra b le pole in the antithesis. W e m ust em phasize,
how ever, th a t th e opposition betw een gods a n d m ortals is not
th e only one in play; there is also th a t betw een statuses (u n m ar­
ried a n d m arrie d ) a n d betw een sexes. T h e significance of
ch a stity o r ab stin en c e varies accordingly: the h iero p h an t before
th e m y steries seeks to free h im self from the tain t o f the physical:
th e P h o cian H eracles priest shu n s w om en to p ro tect his god
b a se d on c o rru p t texts o r m isu n d erstan d in g and are perhaps influenced by C h ristian ­
ity. R eferences to 'p riestesses' in an E leusinian context (F oucart, 215) can be applied to
th e h iero p h an tid s an d the priestess o f D em eter an d K ore (so too the probably Eleusi­
n ia n ίρ εία ς . . . σαόφρονα Κνπριν ίχούσας o i IG 112 3606.15, which anyw ay does n o t imply
a b so lu te celib acy ). F or pro b ab ly chaste D em eter priestesses in H erm ione, b u t only
a tte s te d late, cf. T h . R ein ach , B C H 32 ( 1908), 505.
69 H is v irginity, 150. O th e r virginal tem ple servants in E ur. I T 130.
70 E. R. L each, Culture and Nature or La Femme Sauvage, Stevenson L ecture for 1968,11.
71 W hence, in th e p ractice o f later an tiq u ity , m ost m agical acts d em and prelim inary
ab stin e n ce: F ehrle, 50 n. 6.
92 Miasma
from c o n ta m in a tio n by the fem inine: the virgin priestess of
A rte m is provides a m odel o f virtue ap p ro p riate to young girls
n o t yet in flower. T h e language o f ‘p u rity ’ is used in all three
cases, b u t it can in d icate resistance to several d istin ct forms o f
co n ta m in a tio n .
It is also im p o rta n t to note th a t these ritual rules give rise to
no positive ideal o f chastity. A bstinence, though equipping the
w o rsh ip p e r to a p p ro a c h the tem ples an d sacred objects of the
gods, does no t re n d e r him godlike himself. Even H ippolytus by
his p u rity w ins the favour of b u t one goddess. It is less in order
to b e a c e rtain kind o f person th a t chastity is required th a n in
o rd e r to e n te r certain places, to u ch certain objects, view certain
sig h ts.72 It m ay h av e been m ore com m on to ap p o in t hum ble
sa c rista n s th a n a c tu a l priests from am ong those who, by
re aso n o f th e ir age, w ere necessarily pure, because their
m u n d a n e d u ties b ro u g h t them into m ore frequent contact w ith
th e te m p le.73 C o n n ected w ith this is the failure to assign positive
v alu e to the self-denial th a t continence dem ands. C ontrol over
sex u al d esire w as indeed, in G reek values, an im p o rtan t p a rt of
b o th m ale a n d fem ale excellence, b u t this was ra th e r because
in d u lg e n c e m ig h t divert both sexes from their essential virtues
th a n b ecause self-m ortification was esteem ed in itself. T h e
a n sw e r o f th e C ynics to the problem of lust was subm ission
w ith o u t em otional com m itm ent, not resistance.74 In ritual
p ra ctice, it is h a rd to find even a lim ited a tte m p t to present
c o n tin en c e positively as an aspect o f self-m astery. Young
c h ild re n a n d th e aged are chosen for the posts th a t would
im p o se real re stra in ts on the sexually m ature, w hile sacred fires
a re g u a rd e d , n o t by p erp etu al virgins, b u t by w om en w ho ‘have
d o n e w ith sex’.75 P u rity a n d innocence m ay be associated with
th e ‘in ta c t boys an d girls’ o f cult, b u t they are treated as the
v irtu e s o f a p a rtic u la r stage o f life ra th e r th an as a general ideal,
o r a necessary consequence o f celibacy. It is for m arriage th at
A rte m is’ ‘b e a rs’ are p re p arin g them selves.
T o these generalizations, the ‘virgin priestess for life’, where
72 D em . 22.78 (entering, touching); (D em .) 59.73, 85 (sacrificing, seeing, entering,
d o in g ), 78 (touching).
73 e.g. P lut. Num. 9.11, E ur. Ion 150, Paus. 2.10.4.
74 D over, 208, 212 f.
75 E u r. Ion 150, P lut. De Pyth. Or. 20, 403Γ, Pl. Leg. 759(1, Plut. Num. 9. I I.
The Works o f Aphrodite 93
she ex isted , w'as an exception. T h e re is a significant distinction,
seldom d ra w n sufficiently sh arp ly , betw een the virgin priestess
w ho laid d o w n h er office a t m arriag e, and the ‘virgin priestess
for life’; for th e one, h er office w as a m ere p relim in ary to the
n a tu ra l goal o f w o m en ’s life, w hile in the o th er case it was a
su b s titu te for it. It m ay th u s be no coincidence th at firm evi­
d en c e for th e ‘virgin priestess for life’ is very h ard to fin d .76 T he
p ro p h e te ss a t D elphi was in theory a m aiden (the god could not
possess a b o d y given over to th e pleasure o f a m ortal) an d was
ce rtain ly b o u n d to strict ch astity d u rin g her ten u re o f office, but
in p ra ctice th e p o st was norm ally filled by an old w om an, who
will, since sp in sterh o o d was n o t a recognized estate, certainly
o n ce h av e been m a rrie d .77 As w e have seen, no such priesthoods
a re securely a tte ste d a t A thens. T h e m ost interesting evidence
com es o nce ag ain from E uripides. His T heonoe is an d will
re m a in a virgin; she is w ra p p ed in a m ysterious sanctity, pecul­
iarly rig h teo u s, a n d has ab n o rm al u n d erstan d in g of the inner
counsels o f th e gods. T h e re is no d o u b t th a t positive religious
cap acities a re h ere being trea ted as d ep en d en t on virginity.78 It
m ig h t be w rong, how ever, to relate T h eo n o e’s pow ers merely to
p u rity from a physical tain t. W ith d raw al from the sexual
s tru c tu re o f society brings w ith it w ithdraw al from the social
s tru c tu re , a n d it seem s th a t T h eo n o e can subm it herself to the
gods so co m pletely because like subm ission is not required of
h er by a h u sb a n d . T h ere is a kind o f analogy betw een such a
w o m an a n d sacrificial anim als, o r sacred land, ‘let go’ by m or­
tals for th e use o f the gods.79 T h is conception m ay often have
76 T h e only case I can find is Paus. 9.27.6, H eracles’ priestess a t T hespiae. In the
m ore reliable tra d itio n th e ‘L ocrian m aid en s’ served for a year only, cf. F. G raf, SSR 2
( 1978), 61 —79 (w ith persuasive in itiatory in terp retatio n ).
77 P ark e/W o rm ell, 1.35; cf. 36 for a Pythia w ith progeny. A pollo’s prophetess at
A rgos, γυνή άνόρός εννής είργομένη (Paus. 2.24.1), was p erhaps also post-sexual. T h e
co n cep tio n o f th e pro p h etess as the g o d ’s bride (Fehrle, 7 IF., 79; K. L atte, H T R 33
(1940), 9 - 1 8 , R E 18. 84Ü; B urkert, H N 143) is hin ted at m ythologically, esp. in the
ligure ol C a ssa n d ra (A esch. Ag. 1202—12), but w as certainly not enacted ritually in
( »rcece; the sacred m arriag e in P a ta ra is for H dt. ( 1.182) a foreign custom , tinged with
c h a rla ta n ism . It is in a less precise sense th a t the prophetess is reserved for the god (on
th is reserv atio n cf. Kur. Tro. 2 5 1 -8 ). O ld w om en a t D odona too, S trab o 7.7.12, cf.
L. Hod so n , Ιε ρ ά Ζώ ια, Brussels, 1978, 101 1Γ.
78 E ut . Hel. 1 0 -1 5 ,8 6 5 -7 2 ,8 7 6 -9 1 ,8 9 4 ,9 3 9 , 1006-8. R. K an n ich t {EuripidesHelena,
I leid o lb erg , 1969, i, 75) points out th at T h eo n o e’s vocation ap p ears as a su b stitute for
m a rria g e in 12 f.
79 Set* Kur. Tro. 41 2.
94 M iasma
b een im p o rta n t w here virgin priestesses existed; an d the inde­
p e n d e n c e th a t virginity offered w as certainly significant in
fo rm in g the im age o f virgin goddesses.80 But it is h a rd to know
how far fam iliar G reek attitu d e s are expressed through the
exotic T h eonoe.
T h e reg u la tio n s w e have considered so far have been con­
c e rn e d , alm o st exclusively, w ith sexuality, not sexual m orality.
O n ly in the hellenistic period, as we have noted, does in ter­
co u rse w ith a n o th e r m a n ’s wife becom e m ore polluting than
in te rc o u rse w ith o n e’s own. A C o an law cited earlier shows, if
co rrectly resto red , th a t on the ritu al level no distinction is
d ra w n betw een heterosexual a n d hom osexual co n tac t.81 N ot all
th e form s o f union, how ever, w hich create im purity in the later
law s w ere su b ject to m oral stigm a in the classical period. In
sleep in g w ith a co u rtesan , o r a boy, there w as no necessary
d isg race. A t A th en s, it is really only th e adulterous m ale who
w as su b ject to legal penalties a n d m oral disapproval, b u t free,
so far as the evidence goes, from ritu al disabilities. A nd it is
w o rth sta tin g w ith som e em phasis th at the two classes o f sexual
o lfen d e r w hom society m ost savagely condem ned w ere p erm a­
n e n tly ex cluded from the shrines. T h e re was no question of the
c o n v icted ad u lteress, o r the m ale p ro stitu te, acq u irin g the right
to w o rsh ip w ith respectable citizens through a sim ple period of
a b s tin e n c e .82
T h e se exclusions are, o f course, aspects of atimia, deprivation
o f citizen rights, a n d sim ilar restrictions w ere placed on other
classes o f d iscred ited persons, such as state debtors and
d e s e rte rs .83 D eserters are excluded from the shrines not because
they a re p o llu te d a n d dangerous, b u t because they have sacri­
ficed th e ir rig h t to a place in the com m unity o f citizens. It is
80 See W ilam ow itz on E ur. / / / ' 834, B urkert, GR 284 f.
81 L SC G 151 A 42 (ad d ressed to a m an), άγνεύεσθαι γνναικάς και ά [ ]ς w here either
άνόρός o r αρσενος is h ard to avoid (άμίόος P aton-H icks, cf. A th. 150a, too dem and-
in g ly ). T h is text a p a rt, th e possibility o f hom osexual contact seems not even to be
en v isag ed in early ritu a l rules.
82 A d u ltero u s w om an: (D em .) 59. 85—7; A eschin. 1.183; prostituted m ale: Aeschin.
1. 1 9 ,2 1 , 160, 164, 188; D em . 22.30, 73, 77; 24.126.
83 H a rriso n , ii, 16 8 -7 6 ; Ci. E. M . de Ste. C roix, The Origins o f the Peloponnesian War,
L o n d o n , 1972, 397 f.
The Works o f Aphrodite 95
n a tu ra l to in te rp re t th e case o f th e sexual offenders in th e sam e
w ay; a lth o u g h th eir deeds a re d escribed in th e language o f
p o llu tio n ,84 it is because they are disgraced, n o t because they
a re d an g e ro u s, th a t they are b an n e d from religious life. T h e
p o in t is a n im p o rta n t one, as it distinguishes th e position in
G reece from th a t o f th e m an y societies w here sexual irregu­
larities a re in d eed seen as sources o f religious dan g er, causing
d isease o r crop failure.85 T h e G reeks very p ro b ab ly saw incest
as a p o llu tio n o f this kind, alth o u g h this is surprisingly h ard to
d e m o n s tra te ; b u t in expressing th e wish th a t sodom y and
a d u lte ry co uld becom e u n th in k ab le acts like incest, P lato ack­
now ledges th eir a c tu a l statu s to be different.86 I t is h ard to show
th a t th e ad u lteress or m ale p ro stitu te is en d an g ered or d an g er­
ous on a n y s u p e rn a tu ra l level. If the people o f C ym e avoided as
‘im p u re ’ th e stones on w hich they exposed adulteresses to
h u m iliatio n , th e ascrip tio n o f im p u rity was itself p a rt of the
pro cess o f h u m iliatio n . O n e can im agine th a t a w ronged A the­
n ia n h u s b a n d m ig h t have purified the m arriage bed, or m ight
h av e feared the consequences if he disobeyed the legal req u ire­
m e n t to p u t his ‘p o llu te d ’ wife aw ay, b u t such unease would
be a co n seq u en ce o f society’s m oral con d em n atio n of the
ad u lte re ss ra th e r th a n its cause. T h e w orst au to m atic p u n ish ­
m e n t for sexual crim es is no m ore th a n attack by bees; hostile to
sex u ality in an y form , they especially a b h o r ad u lterers, and
stin g th em savagely, disgusted by th eir sm ell.87
A q u estio n arises ab o u t the ap p lication of the term ‘pollu­
tio n ’. I f we m ean by it beh av io u r th at is felt to su b v ert th e m oral
fo u n d a tio n o f society, so th a t the guilty persons m ust be ex-
84 A d u ltery ‘p o llu tes’ the bed: E ur. Hipp. 1266, Or. 575, Hel. 48 (of. Hec. 366), Anth.
Pal. 3.5.2; it p o llutes those threaten ed b y it: E ur. Hipp. 6 0 1 -6 , 6 5 3 -4 , 946, ? Soph.
Inachus, fr. 269 a 24 R a d t. Αίσχύνω sim ilarly used: Hes. fr. 176.7, E ur. Hipp. 408, 420,
944, 1165, 1172, M en. Sam. 507, and regularly to denote rape. A dulteress excluded from
sh rin es ‘to p rev en t p o llutions’, (D em .) 59.86. M ale pro stitu te ‘im p u re in body’:
A eschin. 1.19, 188; 2.88; o n the ‘unclean m o u th ’ he acquires see below.
85 D ouglas, C h . 8.
86 PI. Leg. 8 3 8 a -9 a .
87 C ym e: P lut. Quaest. Graec. 2, 2 9 1 e-f; for arch aic institutions preserv ed a t C ym e see
A rist. Pol. 1269a 1 - 3 , L atte, H R 32, an d for public hum iliations o f ad u lterers (and
o th ers) H a llid a y on P lu t. loc. cit.; L atte, Hermes 66 (1931), 155—8 = Kl. Sehr. 290—3;
L loyd-Jones in Dionysiaca, Studies presented to Sir Denys Page, C am bridge, 1978, 58 f.; p.
195 below . P u rify in g th e bed: cf. Prop. 4.8.83—6. Legal requirem ent: (D em .) 59.87.
Bees: P lu t. Quaest. Nat. 36.
96 M iasma
pelled if th a t society’s essential values are to be p reserved,88
m a le p ro stitu tio n a n d fem ale a d u lte ry are clearly pollutions;
th e y u n d e rm in e respectively the essential qualities o f the m an -
his m ascu lin ity 89 - a n d o f the w om an - h er stew ardship o f the
p u rity o f th e stock.90 W ith in pollution so defined, however, we
w ill also h av e to include desertion, d e b t to the state, and m ost
o f th e ca p ita l crim es. I f we req u ire o f pollution th a t it be con­
tag io u s, a n d d an g e ro u s on a su p e rn a tu ra l level, all these
offences, in clu d in g the sexual ones, will be excluded. W hat
m a tte rs , o f course, is not d isp u te ab o u t the presum ed essence
b e h in d a w ord, b u t a clear distinction betw een separate
p h e n o m e n a . F or the sake o f such clarity it m ight be helpful to
p u t th e sexual offences in a category of ‘m etaphorical m oral
p o llu tio n s’: pollutions because they are so described,
m e ta p h o ric a l b ecau se they are no t contagious or dangerous in
th e sam e sense as, for instance, m u rd er, and m oral to em phasize
th a t th e im p u rity o f the ad u lteress has quite d istin ct origins
from th a t o f th e corpse. W e are dealing w ith breaches o f social
ru le s —ju s t like desertion in b a ttle — w hich are spoken o f as
p o llu tio n s b ecau se they derive from ‘d irty ’ acts.
T h e response to these offences is interesting. T h a t there
sh o u ld be u n c h a ste w om en a n d boys in the w orld is no m atter
for concern; they provide, indeed, a useful outlet for the not
u n re a s o n a b le desires o f honest m en. I t is only am ong the pos­
sessors o f ‘h o n o u r’ (full citizen rights) th a t they are o ut o f place.
O ffen d ers a re not exiled o r p u t to d eath b u t deprived of
‘h o n o u r’ a n d forced to find a place am id the flotsam o f foreign­
ness a n d vice th a t laps a ro u n d the citizen body.91 T h e atimoi are
in a n alm o st literal sense the ‘o u t-casts’ o f A th en ian society.
In re sp ect o f the relation betw een ritu al p u rity a n d m orality,
th e conclusion m u st be th at, even i f ‘p u rity ’ is in itself am oral,
stro n g factors o f a different kind kept the m orally discredite^
from the a lta r. I t is q u ite m isleading to view G reek religion, at
88 A eschin. 1.183, adu lteress excluded tva μ ή τάς άναμαρτήτονς τών γυναικών
άναμιγννμένη όιαφθείρη.
89 Political enem ies revealed by th eir sexual practices as ‘w om en’: A eschin. 2. 129,
179, H yp. fr. 215, (D em .) E p. 4.11.
W o m an as ste w ard in general: T . E. V. Pearce, Eranos 72 (1974), 16—33.
91 Cf. W h ite h e a d , 67 n. 109. B ut citizens m u st go o u t to enjoy the floating world: to
in tro d u c e meretrices to the m arita l hom e is ap p allin g , A ndoc. 4.14, T er. Ad. 747.
The Works o f Aphrodite 97
least in its p u b lic aspects, as stan d in g beyond or b en eath the
m o ral d e m a n d s o f society. T h is religion was rigorously statu s­
con scious, a n d statu s, as we h av e seen, could be affected by
m o ral co n d u c t. Like its m ag istrates, those w ho p ray ed on the
c ity ’s b e h a lf h a d to be tru ly rep resen tativ e o f it; they were
re q u ire d to be o f especially p u re stock,92 an d they could not
hav e lived licentiously, since those who did w ere unfit to be
citizen s a t all, let alone to re p resen t the citizen body. Aeschines
tells th e A th en ian s not to be su rp rise d at the failures th a t beset
th e m , w h en th ey have a m an like T im arc h u s to d raft prayers on
th e ir behalf. T h u s forms o f b eh av io u r th at are, in them selves,
sh am efu l b u t n o t dan g ero u s, becom e actu al sources o f religious
d a n g e r w h e n p e rp e tra te d by those holding an office th a t d e­
m a n d s a n h o n o u ra b le in cu m b en t. D em osthenes’ den u n ciatio n
o f A n d ro tio n concludes: ‘you have A n d rotion as re p airer of
sa c re d vessels. A ndrotion! w h a t could be a worse offence
a g a in s t th e gods? T h e m an w'hose jo b it is to en ter tem ples,
to u ch lu stra l w a te r a n d sacred baskets, an d take charge of the
c u lt w e p ay to th e gods, o u g h t to have kept p u re not ju s t for a
fixed n u m b e r o f days, b u t th ro u g h o u t his life, from the things
A n d ro tio n h as d o n e .’ D em ochares, having p ro stitu ted ‘even the
u p p e r p a rts o f his b o d y ’, w as ‘unfit to blow the sacred fire’. T h e
sp eech ag a in st N eaera is exten d ed testim ony to the shock felt a t
te n u re o f a n im p o rta n t p riesthood by a w om an o f sham eful
life.93 O n e m ay a d d th a t it is very doubtful w h eth er a convicted
a d u lte re r, ritu ally p u re th o u g h he m ay have been, w ould have
b ee n co n sid ered a su itab le ca n d id a te for a priesth o o d .94
As w as n o ted earlier, it is difficult to prove th a t incest is a
‘p o llu tio n ’. H ere, too, problem s arise ab o u t the definition of the
term . In c e st is now here spoken o f as a miasma, a n d it does not
seem th a t it w as even form ally illegal at A thens, m uch less th at
th e offender w as publicly expelled to purify the state.95 In one
p assa g e in E u rip id es, how ever, O ed ip u s is said, im m ediately
a fte r a reference to his m arriag e, to be ‘polluting the city ’, and
92 (D em .) 59.92, PI. Leg. 759c, Ar. Pol. 1329a 29 I'., L SA 73 .4 -8 .
93 A eschin. 1.188, D em . 22.78, A rchedikos, fr. 4 = Polyb. 12.13.7 (sam e point against
D em o sth en es 76 FG rH fr. 8, a n d cf. A eschin 2.23, 88), (Dem .) 59.72-117.
94 cf. A ndoc. 1. 1 2 4 -9 , on Callias.
95 See H a rriso n , i, 22 n. 3, M . B ro ad b en t, Studies in Greek Genealogy, Leiden, 1968, 155.
A dkins, 110 n. 17 excludes incest from the pollutions on these grounds.
98 M iasma
th e idea o f religious d a n g e r is p resen t in the com m on claim th at
su c h a m a tc h is anosios, offensive to the gods.96 T h e incestuous
co u ld be socially isolated w ith o u t exile, by exclusion from sacri­
ficial co m m u n ities a n d m arriag e exchanges.97 It was believed in
la te r a n tiq u ity th a t C im on in cu rred a ctu al ostracism because of
his relatio n s w ith E lpinice, a n d the recently discovered ostra-
con th a t urges him to ‘get o u t a n d take his sister w ith him ’
suggests th a t this m ay indeed have been a factor. T h e Aeolus of
E u rip id e s, s te rn e r th a n H o m e r’s, p u t his incestuous d au g h ter
to d e a th .98 O n a n im aginative level, a n analogy is clearly felt
betw e en incest (a n d o th er gross sexual offences) a n d the worst
p o llu tio n s. In th e m yth o f O ed ip u s, it is associated w ith p a rri­
cide. A connection betw een sex a n d eating, a n d thus betw een
fo rb id d e n sexual co n tac t a n d forbidden food, is said to be found
th ro u g h o u t th e w orld, a n d becom es alm ost explicit in the m yths
o f T h y estes, T ereu s, a n d C lym enus. T hyestes seduced his
b ro th e r’s wife, T ereu s his wife’s sister, C lym enus his daughter;
all w ere su b seq u e n tly forced to eat th eir ch ild ren ’s flesh. Plato,
too, associates cannibalism a n d incest, while A eschylus’
D a n a id s ask, in reference to a forced an d perh ap s incestuous
m a rria g e , ‘H o w could a bird th a t eats an o th er bird be pure?’99
In c e st, p a rtic u la rly th a t betw een generations, is, therefore, one
o f th e su p re m e h o rro rs o f the im agination th at define by con­
tra s t th e norm s o f o rd ered existence. It lies in a sense beyond
p o llu tio n , b ecau se it is beyond purification.
In m ost societies, sexual beh av io u r is regulated, in addition
to th e re stra in ts o f decency a n d m orality, by a canon o f the
‘n a tu r a l’. T h is canon m ay declare illicit any form o f sexual
re la tio n b etw een certain classes o f people, such as m em bers of
th e sam e sex, o r it m ay forbid certain acts even w hen perform ed
by those betw een w hom sexual c o n tac t is in itself perm issible. It
96 K ur. Phoen. 1050; S oph. Ö C 946, Ar. Ran. 850, PI. Leg. 838b; pollution language in
A d . N A 6.39. cf. too (P lu t.) Par. Min. 19a, 310b.
97 G lotz in D ar.-S agl. s.v. Incestum, 450.
98 C im o n : first in (A ndoc.) 4.33, Plut. Cim. 4 .5 - 7 . T h e ostracon: see p. 270 below.
O th e r im p u ta tio n s o f incest in orators: Lys. 14.28, Lys. fr. 30 G ernet, a n d cf. Andoc.
1 .1 2 4 -9 , (A ndoc.) 4.22, ? Isae. 5.39. Aeolus: see N auck, TGF, 365 f.
99 PI. Res p. 5 7 1 c-d ; A esch. Supp. 226, cf. μιαίνειν γένος in 225. O n the relevance o f
in cest to Supp. see A. F . G arvie, Aeschylus'Supplices: Play and Trilogy, C am bridge, 1969,
2 1 6 —20; J . K . M a c K in n o n , CQ 28 (1978), 74—82. N ote th a t απτομαι is used both of
sexual a n d d ie ta ry crim e.
The Works o f Aphrodite 99
is h a rd to know q u ite w h a t belongs in such a category in Greece.
T h e ‘u n n a tu ra l vice’ m ost fam iliar from o ur ow n cu ltu re w as, it
seem s, n o t seen as such, w h eth er perform ed hom osexually or
hetero sex u ally , before P la to .100 O ra l sexual acts do n e by a m an
conform to th e definition in th a t they are considered revolting
even w hen n o t m orally sham eful. ‘A nyone w ho d o esn ’t abom i­
n a te su ch a m a n ’, says a n A risto p h an ic chorus ab out
A rip h ra d e s, w hom it revealingly describes as th e ‘in v en to r’ of
su ch p ractices, ‘will never d rin k from th e sam e cup as u s.’ As we
h av e seen, a n o th e r com ic poet d eclared th at D em ochares h ad in
th e sam e w ay m ad e h im s e lf‘unfit to blow the sacred flam e’. A
cook m ig h t, it seem s, claim th a t a rival indulged in these
p leasu res, a n d so w ould ta in t the food, while th e kiss o f such a
m a n w as to be av o id ed .101 T h ese texts agree in expressing
rev u lsio n ag a in st th e p ractice th ro u g h a strikingly physical view
o f th e ‘p o llu tio n ’ th a t it causes. T h e state is n o t en d angered by
th e fact th a t p eo p le do such things, b u t those who do them
beco m e very d irty , a n d th eir m isused m ouths co n tam in ate all
th ey to u ch o r b re a th e on. T h e source of the revulsion seems to
be th e offence ag a in st the b o d y ’s hierarchy. T h e m ost h o n o u r­
a b le p a rt o f th e body, a n d the p urest, is th e head, an d of the
h e a d th e p u re st p a rt should be the m outh, w hich receives food,
u tte rs p ra y er, a n d im p lan ts ch aste kisses;102 it is th u s in p a rtic u ­
la r d a n g e r o f co n tam in atio n by co n tact w ith d irty an d sham eful
100 K . J . D over, Greek Homosexuality, London 1978, 60, 165—170; heterosexual anal
in te rc o u rse , ibid. 100—1, a n d note H d t. 1.61. 1 (‘irreg u lar’, n o t ‘u n n a tu ra l’). T h is is not
su rp risin g in a c u ltu re w here hom osexuality w as probably once a req u ired phase in a
y o u th ’s ed u catio n : J . N . B rem m er, ‘An E nigm atic Indo-E uropean Rite: P aederasty’,
Arethusa 13.2. (1980), 2 7 9 -9 8 ; cf. P. C artled g e, P C P S n.s. 27 (1981), 17 -3 6 .
101 A rip h rad e s: A r. Eq. 1280-9; he ‘p o llu tes’ his tongue, an d ‘soils’ his chin, licking
th e ‘d isg u stin g d e w ’. D em ochares (an d D em osthenes): p. 97 n. 93. A cook: Posidippus,
fr. 1.5—6 ap. A th . 6 6 2 a (som e textual u n certain ty ). A kiss: Ar. Eccl. 647. O ral sexual acts
by th e m ale a re n o t p o rtray ed on vases, G. H enderson, The Maculate Muse, Yale, 1975,
51, D over, op. cit., 99—102. H en d erso n ’s belief (5 1 -2 ) th at no disgust is felt a b o u t such
p ractices is b ased on a failure to d istinguish betw een oral sex perform ed by m en and
w om en ; the la tte r is not disgusting, a t least w hen perform ed by hetairai, but m ight
p e rh a p s becom e so if perform ed by citizen wives (H . D. Jocelyn, PCPS n.s. 26 (1980),
1 2 -6 6 , pro v es th e ab u siv e term laikazein to refer to fellation by the w om an; in later texts,
for w h ich see C o u rtn e y on Ju v e n al 6.51, a n d A rtem id. 4.59, p. 283. 8—16 Pack, oral sex
p o llu tes m an a n d w om an alike).
102 O n th e h e a d cf. L S J s.v. κεφαλή-, the fact th a t it can becom e/ζίαρά proves its norm al
p u rity . R eligious im p o rtan c e o f ‘p u re m o u th ’: Aesch. Eum. 287, Supp. 696. K isses: Kroll
on C a t. 79.3.
100 M iasma
o rg a n s. C o n sid e ratio n s o f social hierarchy, how ever, become
confused w ith this sim ple h ierarch y o f the body. W hen per­
fo rm ed by a w om an, such acts are not revolting, because
w o m a n is n a tu ra lly d eg rad ed in relation to m an; even w hen
d o n e by a m an , it is only w hen com bined w ith the absolute
se lf-d eg ra d atio n o f hom osexual p ro stitu tio n th a t they are suffi­
cien tly o u trag e o u s to becom e a focus for political ab u se .103
In c e st, finally, th o u g h it was p ro nounced ‘n a tu ra l’ by n a tu re ’s
aggressive su p p o rters, was surely in conventional b eliefju st the
o p p o site, as it violated an u n w ritten law .104
T h ro u g h o u t this discussion o f sexual m atters, one idea th a t
h a s p lay ed no p a r t is th a t o f the in h ere n t im purity o f wom en,
m a n ife ste d th ro u g h m en stru atio n . T h is absence is surprising,
b o th b ecau se m en stru a tio n is viewed as a pollution by innum er­
a b le societies, a n d p artic u la rly because it com m only acts as a
sy m b o l on w hich m e n ’s attitu d e s o f suspicion a n d hostility
to w a rd s w om en can focus.10S F or its statu s as an unconscious
sy m b o l o f this kind, evidence from New G uinea in p artic u la r is
q u o te d , w here, it is said, fear o f m en stru al pollution is m uch
s tro n g e r a m o n g trib es w ho ‘m arry [the dau g h ters of] the people
w e fig h t’ th a n a m o n g those w ho m arry from friendly tribes;106
th a t m ay be a n extrem e case, b u t a connection o f som e kind
betw e en m e n stru a tio n an d w o m an ’s status as a n inferior,
th re a te n in g , o r m istru ste d being is w id esp read .107 T h ere un-
103 W o m an : see n. 101. P olitical abuse: no o rato r, to my knowledge, accuses oppo­
n e n ts o f d oing such th in g s to women.
104 X e n . Mem. 4.4.19—23, PI .Leg. 838a—b;: n a tu re ’s supporters: SV F 1.256, 3. 743—6,
cf. E u r. fr. 19.
105 D ouglas, 173 fF., cf. her Implicit Meanings, Essays in Anthropology, L ondon, 1975,
C h . 4.
106 M . J . M eggitt, ‘M a le -F e m a le R elationships in the H ighlands o f A ustralian New
G u in e a ’, American Anthropologist, 1964, vol. 66, special publication on New Guinea, the
Central Highlands, ed. J . B. W atson, 2 0 4 -2 4 ; cf. e.g. M . S trath ern , Women in Between,
L o n d o n a n d N ew Y ork, 1972, C h. 7, an d , for fu rth er references on sexual pollution in
th is a re a , A. S. M eigs, M an n.s. 13 (1978), 304—18. M . R. Allen, however, Male Cultsand
Secret Initiations in Melanesia, M elbourne, 1967, 54, draw s attention to societies that
m a rry friends b u t h av e stro n g sexual pollution beliefs, and E. F aith o rn (in Towards an
Anthropology o j Women, ed R. R. R eiter, New York an d L ondon, 1975, 127-40), points
o u t th a t m e n stru a l blood is only one o f a n u m b er o f dangerous bodily w astes, produced
by b o th sexes.
107 See, to cite only m o d ern M e d ite rran ean parallels, J . Pitt-R ivers, People o f the
Sierra2, C hicago, 1971, 197; J . C utileiro, A Portuguese Rural Society, O xford, 1971,99, 276;
B lu m , 46 (12), cf. 99( 16); J . du Boulay, Portrait o f a Modem Greek Village, O xford, 1974,
The Works o f Aphrodite 101
d o u b te d ly existed in G reece co n sid erab le unease an d suspicion
a b o u t w o m en th a t could have found expression in this way. T h e
first w o m an cam e to m an as ‘a beautiful evil in place o f good’,
b rin g in g w ith h e r disease a n d old ag e .108 As we have seen,
w o m an th re a te n e d m a n ’s virility, his valour, a n d his life. A p art
from th e d ire c t d am ag e she could do him , there w as also, m ore
th re a te n in g still, th e pow er she possessed to b rin g d ish o n o u r on
his n am e. L u stfu l a n d u n co n tro llab le, she was th e weak link in
th e fam ily ch ain o f h o n o u r.109 A nd w hen she ab an d o n ed the
m o d est, subm issive role th ro u g h w hich society wisely sought to
re stra in h e r volatile n atu re , the whole stru c tu re of ordered
existen ce w as th ro w n into je o p a rd y . P ru d en t states h ad insti­
tu te d a special m ag istracy o f ‘w o m en-contrailers’. 110
It does n o t m a tte r th at this a ten d entious selection from the
m an y w ays in w h ich G reeks could view wom en. Such attitu d es
existed, a n d could readily h ave been expressed in term s of
im p u rity . W e do occasionally find w om en spoken o f as ‘d irty ’ or
‘re v o ltin g ’; a n d th e fact th a t th e ideal w om an was com pared to
th e p u re bee p erh ap s in d icates w h a t m ight be th o u g h t of the
re s t.111 W h a t ca n n o t be d em o n strate d is a connection betw een
this id ea a n d m en stru atio n , or any strong fear of m enstrual
blood as a p o llu tin g force. P u rity from m en stru al co n tam in a­
tion only a p p e a rs as a condition for entering a tem ple in late
102—3;J . O k ely in S. A rd en er (ed.), Perceiving Women, L ondon, 1975,55—86. A dm ittedly
here too, as in the R o m an ag ricu ltu ral w riters disc ussed in the text, though m enstrual
blood sym bolizes th e d an g ers in h eren t in female sexuality, w hat is directly im perilled
by it is th e life o f farm an d field.
108 H es. Op. 5 7 -1 0 5 , Theog. 570—602, cf. D over, 99—102 (w om an sham eless, deceitful,
‘p ro m p t to devise evil’, vindictively jealo u s, ungrateful, sp read er o f m alicious gossip),
an d p a rtic u la rly J . G o u ld , ‘Law, C ustom an d M yth: Aspects of the Social Position o f
W o m en in C lassical A th e n s’,J H S 100 (1980), 3 8 -5 9 .
109 E u r. fr. 662; ‘p o llu tio n ’ inflicted on m arriage bed, p. 95 n. 84 above; ‘doglike
m in d ', H es. Op. 67 w ith W est’s note, P. F riedrich, The Meaning o f Aphrodite, C hicago,
1978,135.
1,0J e o p a rd y : e.g. A esch. Cho. 5 8 5 -6 3 8 . M agistracy: A r. Pol. 1322b 3 7 - 1323a 6 ,
B u so lt/S w o b o d a, i, 494 n. 1.
111 Ale. fr. 346.4, w ith p articu lar reference to lust (D. L . Page, Sappho and Alcaeus,
O xfo rd , 1955, 305; W . R osier, Dichterund Gruppe, M unich, 1980, 258 n. 344, unconvinc­
ingly suggests a specific reference to lu b rication); Ar. Lys. 253, 340; M en. fr. 718.6; but
in all these th e sense ‘revolting’ - a sense in w hich miaros is often applied to individual
m en - is m ore p ro m in e n t th an th at o f d irty . F or d irty types o f w om en see Sem onides 7.
F em ale sexual secretions dirty, Ar. Eq. 1285 (whence, in part, revulsion a t cunnilingus).
102 Miasma
sa c re d law s o f non-G reek c u lts.112 M en stru atio n is, in fact,
so m e th in g a b o u t w hich, outside m edical texts, we hardly hear.
I t seem s alm o st c e rtain th a t th ere will have been rituals relating
to it, b u t n o th in g o f the kind is re c o rd e d .113 Possibly it was a fact
so secret a n d sh a m in g th a t it could not be alluded to a t all, even
to th e ex ten t o f re q u irin g p u rity from it in a sacred law. It is,
c e rta in ly , alm o st th e only bodily function w hich O ld C om edy
n e v e r m e n tio n s .114 C h an ce allusions in scientific texts, however,
su g g est th a t the w idespread taboo against intercourse during
m e n stru a tio n w as no t o b serv ed .115 T h is is particu larly su r­
p risin g in th a t, alth o u g h the process o f m enstruation was
co m m o n ly spoken o f as a ‘p u rificatio n ’, one m ight, by analogy
w ith o th e r p urifications, expect the w aste m a tte r discharged
th e re b y to be p a rtic u la rly im pure.
W h a t seem s to be the only early evidence for m agical p ro p er­
ties o f m e n stru a l blood com es, curiously, from A ristotle, who
a sse rts th a t th e m e n stru a tin g w om an dim s the m irro r in front of
w h ich she stan d s. T h is detail re ap p ea rs in R om an agricultural
w riters in co m p an y w ith fu rth er powers; m enstrual blood sours
w ine, blights trees a n d crops, b lu n ts knives, kills bees, rusts
U 2Z ,5554. 7 - 8 . ? 91.16, 119.13, LSC G 55.5, B C H 102 (1978), 325 line 9. Cf. Porph.
Abst. 2.50, an d p ro b ab ly H eliodorus Aeth. 10.4.5.
1.3 A co n n ectio n betw een m en stru atio n an d the T hesm ophoria is suggested by
K . K eren y i, Zeus and Hera, L ondon, 1975/6, 157, cf. B urkert, GR 369 f., an d D etienne,
Eugénies, 213. A. M o m m sen, Philol. 58 (1899), 3 4 3 -7 , argues th at, in the tem ple
in v e n to rie s o f A rtem is B rau ro n ia, ράκος som etim es indicates a valuable garm ent. H e
su g g e sts th a t it acq u ire d th is m eaning, via th at o f offering, from a custom o f young girls
d e d ic a tin g th eir first m en stru al rags to A rtem is [ράκος in this sense Geopon. 1. 14.1,
10.67.3, P lut. Quaest. Conv. 700e,j cf. G oltz, 229 f.) But there is no evidence for such a
c u s to m , an d A eolian βράκος m eans ‘ro b e’ in S appho, fr. 57, T heoc. 28.11. L inders
a rg u e s (58 I.) th a t ράκος does m ean ‘rag’ in the records. I suspect there was originally a
co n n ectio n betw een the three ‘polluted d a y s’ a t the end o f the m onth (p. 158) and
m e n s tru a tio n . M e n stru atio n naturally fell, according to A ristotle, a t the m onth’s end
(H ist. A n. 582a 35—6 , Gen. An. 738a 16—22, 767a 1—13); on the m oon an d m enstruation
cf. Cl. P reau x , La Lune dans la pensée grecque, B russels, 1973, 88 f.
1.4 K . J . D over, Greek Homosexuality, L ondon, 1978, 173, m entioning the possi­
bility th a t th e ‘th in g s a m an m ay not nam e’ d edicated by a retired hetaira in Philetas
1.5 G o w /P a g e (A nlh. Pal. 6.210) are m en stru al towels. If so, th a t is relevant to
A. M o m m sen ’s theory m entioned above; b ut one ra th e r expects the reference to be to
so m e th in g salacious.
1.5 A rist. Gen. An. 727b 12—23, H ippoc. Nat. Xlul. 8 (7 .3 2 4 L ittré). T h e excuse in Ach.
T a t. 4.7.7. (άνόρί σννελθεΐν ον θέμις d u rin g period) com es therefore from some o ther
tra d itio n . O f course, w illingness to have intercourse durin g m enstruation does not
n ecessarily m ean th a t m en stru atio n is positively evaluated (cf. Blum, 46 (12)), al­
th o u g h it m ay (B uxton, 212).
The Works o f Aphrodite 103
m etals a n d m ad d en s dogs. (It can be p u t to beneficial use too,
b u t even h ere its utility, like th a t o f a poison, lies in its d estru c­
tiveness. L o custs fall d ead to th e g ro u n d a t sight o f a girl at her
first m e n s tru a tio n .116) It is u n fo rtu n ately im possible to tell
w h e th e r this w hole com plex o f beliefs already existed in A ris­
to tle ’s tim e .117 W h a te v er th eir d ate, it is not to m ale h ealth and
p o ten cy th a t th e m en stru atin g w om an poses a th rea t.
O n e m ig h t o f course argue, n o t im plausibly, th a t it is indeed
m ale fear o f w om en th a t endow s m en stru al blood w ith such
fearful pow ers, a n d only by a F reu d ian process o f displacem ent
is its d e stru c tiv e force d iv erted to o p erate ag ain st p lan ts and
field. T h e only text, how ever, th a t m akes w om an, by h er m ere
p h y sical n a tu re , a source o f d a n g e r to m an is ‘H esiod’, who
w a rn s ‘L et a m an not clean his skin in w ater a w om an has
w a sh e d in. F o r a h a rd p en alty follows on th a t too for a tim e.’118
T h e re is no reason to see in th a t a reference to m enstru atio n . It
is n o n e th e less in terestin g , as co n tain in g the idea o f co n tam in a­
tion; b u t it finds no echo in later texts. In the classical period, to
ju d g e from th e surviving evidence, the th rea t w hich w om an
poses to o rd e re d society proceeds not from the d ark recesses of
h e r body b u t o f h er m in d .119
116 A rist. de somniis 459b 2 3 -4 6 0 a 23; Pliny, H N 7.64, 28. 7 8 -8 0 , C olum ella 11.3.50,
Geoponica 12.20.5, 25.2; locusts: Col um . 10.357 IT., 11.3.64, Ael. N A 6.36, Pliny, H N
17.266, 28.78, Geop. 12.8.5 f.; m en stru al blood, o r indecent exposure by a woman,
a v e rts hail an d w hirlw in d s from vineyard: Plut. Quaest. Conv. 700e, Pliny, H N 28.77,
Geop. 1.14.1; m ed icin al powers: Pliny, H N 28.82—6. Cf. H. W agenvoort, Roman
Dynamism, O x lb rd , 1947, 173-5.
117 F o r the locust ch arm , C olum . a t 10.358 cites ‘D ard an u s’, a t 11.3.64 ‘D em ocritus'
o n a n tip a th ie s (for th e link o f the two cf. Pliny, H N 30.9); Pliny, H N 28.78 cites
M etro d o ru s o f Scepsis, w ho claim s th e discovery to have been m ad e in C a p p ad o cia.
O n m edicinal pow ers Pliny, H N 28. 82—6 qu o tes various unrevealing authorities.
118 H es. Op. 753—5 . T o the sam e kind o f context belongs the idea o f wom an ‘burning
u p ' a n d ‘w ith erin g ’ m an , Op. 704—5, cf. D etienne, Jardins, 224 f.
119 B ut n o te S im on, 242, 260—6, on the H ip p o cratic doctor’s ‘need to be ig norant' of
th e in side o f th e fem ale body.

4 to allow this p o llu ted m an to e n te r divine precincts an d pollute
T H E SHEDDING OF BLOOD
T h a t th e blood o f his victim clings to the h an d o f a m urderer,
a n d , u n til cleansed, d em an d s his seclusion from society, is a
b e lie f a tte ste d in a bew ildering variety of literary, oratorical,
h isto rical, m y th o g rap h ical, a n d pictorial sources - although
o th e r sources preserve a stolid a n d no less perplexing silence on
th e sam e subject. T w o texts illu stra te the m a tte r in som e detail,
B ook 9 o f P la to ’s Laws, an d the Tetralogies ascribed to A ntiphon.
T h e Tetralogies are an obvious startin g -p o in t for a discussion,
a lth o u g h no t an ideal one. T h e ir sophistical au th o r, w hether
A n tip h o n o r a n o th e r, is p erh ap s not him self com m itted to the
d o c trin e s h e m an ip u lates, an d m ay not know w here to draw the
line in his im itatio n o f belief; a n d the relation o f these h ypotheti­
cal cases to a c tu a l legal process is a long-standing difficulty. O n
th e o th e r h a n d , it is now generally agreed th at the audience to
w h ich they a re add ressed is A th e n ia n ,1 w hich m eans th a t they
c a n be co n fro n ted w ith the one body o f hom icide law th a t is well
k n o w n to us; a n d the very form o f the tetralogy, designed to
show how th e sam e topic o f a rg u m e n t can be exploited an d
re -ex p lo ited by bo th p arties, m eans th a t the full po ten tial o f the
a rg u m e n t from pollution is here displayed as in no o th er text.
I t m ay be useful to give a su m m ary o f such arg u m en ts in the
o rd e r in w h ich they ap p ear. T h e ‘you’ o f the speakers refers to
th e ju ro rs . T h e first tetralogy concerns a case o f p rem editated
h o m icid e; the d efen d an t denies his involvem ent.
First speechfo r the prosecution: It w ould be against o u r interest to
p ro se c u te an in n o cen t m an a n d let the guilty escape. T h e whole
city is p o llu te d by the guilty m an until he is prosecuted, and if
w e connive a t this by charging the innocent, the guilt for this
p o llu tio n o f the city becom es ours, a n d the p u n ish m en t for the
1 S e e G e r n e t, Antiphon, 8 - 1 3 ; K . J . D over, C Q 44 (1950), 58; M . G agarin, O RBS 19
(1978), 2 9 1 -3 0 6 .
The Shedding o f Blood 105
m istak e you w ould m ake falls u p o n us. T h u s the w hole pollu­
tion falls u p o n us if we act u n ju stly (3). I t is ag ain st y o u r in terest
th e ir sa n c tity , o r pass on his co n tam in atio n to the innocent
by e a tin g a t th e sam e tab le w ith them . T his is th e kind o f thing
th a t causes crops to fail, a n d affairs in general to go w rong. T h e
vote you are a b o u t to cast concerns y o u r own interest: m ake this
m a n b e a r his sins on his own h ead , a n d purify the city (1 0 -1 1 ).
First speech fo r the defence: As I am innocent, I will not pollute
th e sh rines. I t is m y o p p o n en ts who, by prosecuting the inno­
ce n t a n d lettin g go th e guilty, cause crop failure (11).
Second speech fo r the prosecution·. As his guilt is m anifest, in
seeking a c q u itta l h e is m erely asking you to transfer his own
p o llu tio n u p o n yourselves (9). If you acq u it him unjustly, the
d e a d m a n w ill n o t be a visitan t a g ain st us, b u t you will have him
u p o n y o u r m inds. So avenge th e victim , p u nish th e killer, an d
clean se th e city. T h u s will you be free o f the pollution you would
o th erw ise in c u r on the guilty m a n ’s b eh alf (1 0 -1 1 ).
Second speech fo r the defence: R em em b er the victim ’s rig h t to
veng ean ce. I f you condem n m e, the real culprit will never be
foun d (11).
T h e seco n d tetralogy concerns a boy killed by an o th er a t a
javelin p ractice. B oth p arties agree th at the d ea th was acci­
d e n ta l.
Prosecution 1: M y son’s d eath , if unavenged, will be a source of
religious an x iety to us. Exclude the killer from the places the law
re q u ires, a n d do n o t allow the w hole city to be polluted by him
<2>Prosecution
· 2: Even if the killing was a sim ple accident, the
killer sh o u ld p ay the penalty; it m ay, however, be a ta in t sent
a g a in st him by th e gods for som e act o f im piety (8). As the whole
p o llu tio n is liable to be transferred to you, take great care. D on’t
involve yourselves in the killer’s pollution (1 1 -1 2 ).
Defence 2: T h e victim killed him self, in effect; thus he cannot
be said to be unav en g ed (8). T h e d ead boy, pun ish ed by his own
fault, can leave no form o f v isitant against anybody. B ut if an
in n o c e n t boy is destroyed, this will be a source o f religious
an x iety to those w ho condem n him (9).
T h e th ird tetralo g y relates to a d eath in a braw l. T h e fact of
th e killing is ag reed , b u t the degree o f provocation disputed.
106 Miasma
Prosecution 1: T h e victim o f m u rd e r leaves behind him the
a n g e r o f th e avenging spirits, w hich acts as an agent of god’s
ven g ean ce o n b e h a lf o f one robbed o f the divine gift o f life.
T h o se w ho ju d g e unjustly bring this anger, a pollution th a t does
n o t belong to them , into th eir ow n houses (3). If we, the dead
m a n ’s n a tu ra l avengers, prosecute the innocent, we will have
th e v isita n ts o f the victim actin g as avenging spirits against us;
a n d we will ourselves be guilty o f m u rd e r (4). W e have therefore
p ro se c u te d th e guilty party; do you punish him a n d cleanse the
city (5).
Defence 1\ C o n sid e r your ow n interest. Should you acq u it p ie
un ju stly , b ecau se the prosecution has failed to convince you,
th e d e a d m a n ’s visitan t will tu rn ag ain st the prosecution and
n o t a g a in st you. B ut if you condem n m e unjustly, it is against
you a n d no t the p ro secu to r th a t I shall tu rn (?) the anger o f the
av e n g in g sp irits (8). A cquit m e: thus shall we all avoid defile­
m e n t as b est we can (9).
Prosecution 2: W e ad ju re you, on b eh alf o f the victim , to
a p p e a se the w ra th o f the avenging spirits by this m a n ’s death,
a n d so cleanse the w hole city (7).
Defence 2\ D o n ’t kill an innocent m an. If you do, the dead
m a n ’s av en g in g sp irit will still be a visitant against the guilty
(perhaps —the text is corrupt), a n d the innocent m an by his d eath
will d o u b le th e pollution o f the avenging spirits against his
killers (10).
T h e first q u estio n raised by these texts ab o u t the pollution of
b lo o d sh ed is the a p p a ren tly sim ple one o f w h a t it is. T his
q u estio n has, o f course, a n obvious answ er, w hich is fundam en­
tal to the w ay in w hich the m u rd erer is norm ally described,
im ag in ed , a n d p o rtrayed: his pollution is the blood o f his victim
clinging to his h a n d s .2 In these speeches, however, this obvious
a n sw e r is en tirely disregarded. M ost openly, perhaps, in the
th ird tetralogy, b u t by im plication th ro u g h o u t the work, pollu­
tion a p p e a rs not as a m ess o f blood, b u t as the anger o f the
v ictim , o r o f avenging spirits actin g on his behalf, against the
m a n w ho h as ro b b e d him o f the life th a t is his right. ‘A ppease
th e w ra th o f the avenging spirits by this m a n ’s d ea th , a n d so
clean se th e w hole city ’, ‘the pollution of the avenging sp irits’,
2 e.g. A esch. Eum. 41 f., a n d the com m on expression ον καθαρός τάςχείρας.
The Shedding o f Blood 107
‘b rin g u p o n them selves the an g er o f the avenging spirits, a
p o llu tio n th a t does not belong to th e m ’ - such expressions
illu stra te u n am b ig u o u sly , in th e case o f m u rd er, th a t ‘d em o n ic’
in te rp re ta tio n o f pollution w hich we noted to be u n attested for
b irth a n d d e a th .3 T h e literal im age o f m u rd er-p o llu tio n as a
sta in on th e h an d s, w here it does ap p e ar, is m anifestly a symbol
o f so m eth in g beyond itself, since the stain is invisible; the
Tetralogies u n ab a sh ed ly su b stitu te th e thing sym bolized for the
sym bol.
T h e id ea th a t it is his v ictim ’s an g er th at m akes th e m u rd erer
d an g e ro u s o r en d an g ered is not confined to the Tetralogies. Plato
refers to a n ‘an c ie n t m y th ’ w hich explains th e killer’s exile in
ju st these term s, w hile X e n o p h o n ’s C yrus can even appeal to
th e m u rd e re d m a n ’s pow er to send o u t ‘avenging d em ons’ as
a n ack now ledged fact w hich will su p p o rt the m ore doubtful
g en e ral p ro p o sitio n o f the so u l’s survival.4 Legend told how the
re g en t P au san ias w as h au n ted by the ghost o f a B yzantine girl
lie h ad su m m o n ed ‘for h er sh am e’ a n d accidentally killed.5 T he
id en tificatio n b etw een pollution a n d the victim ’s anger is o b ­
vious in an expression like ‘the pollution com ing from th e dead
m a n ’.6 T h e E rinyes, above all, are an im ate agents of pollution
w ho em b o d y th e an g e r ofo n e slain by a kinsm an. A lthough they
are not form ally identical w ith pollution (rath er they ‘arrive
w here a m a n hides bloody h a n d s ’),7 there is no difference
betw een its effects a n d theirs, a n d the operations o f the two are
norm ally co-extensive; even w here, in the exceptional poetical
con cep tio n o f A eschylus’ Eumenides, they continue th eir assaults
a lte r th e m u rd e re r’s h ands are clean, the evils w ith w hich they
(h re a te n A th en s for h arb o u rin g th e m u rd erer are fam iliar
clfects o f p o llu tio n .8 This co-extensiveness of pollution a n d the
v ictim ’s a n g e r is im plicit in th e form al rites o f purification, in
w hich ‘w ash in g off the blood’ is followed by ap p easem ent; the
3 Tetr. 3 y 7, d 10 (cf. K . J . M a id m e n t’s note, ad loc. in the L oeb), a 3; cf. R ohde, 215
... 176:
4 PI. Leg. 8 6 5 d -e , cf. 8 7 2 e-8 7 3 a; X en. Cyr. 8.7.18.
5 P lut. Cim. 6.4—7, De sera 555c, Paus. 3.1 7.8—9. O n h aunted houses in antiq u ity see
I )odds, Progress, 157 n. 2.
6 S oph. 0 7 "3 1 3 - unless the genitive is objective, as it seems to be in E ur. fr. 82. Cf.
S oph. Ο Π 012.
7 A esch. Eum. 3 1 6 -2 0 .
“ 7 7 8 -9 2 .
108 M iasma
sa m e is tru e o f the savage self-protective devices o f m urderers,
w h o ‘w ipe off’ o r ‘spit o u t’ the v ictim ’s blood, and seek to
in c a p a c ita te him for revenge by m u tilatio n .9 In exem pting
from all legal sanctions, therefore, the killer w ho h ad been
p a rd o n e d by his dying victim , th e A thenians w ere not bidding
d efia n ce to p o llu tio n ,10 b u t acknow ledging its source.
A g ain st th e identification o f pollution an d angry spirits, it has
b een o b jected th a t such spirits are virtually confined to tragedy
a n d the Tetralogies. In forensic oratory, history, a n d comedy,
alastores a n d alitêrioi are no t su p e rn a tu ra l beings b u t polluted,
sacrilegious, d an g e ro u s h u m a n s .11 T h e contrast, how ever, be­
tw een A eschylus a n d the Tetralogies on the one side, an d the
o ra to rs a n d com edians on the oth er, p erhaps reflects the differ­
en ce betw een different ages as m u ch as th a t betw een literature
a n d life. T h e atm o sp h ere o f A eschylus a n d the Tetralogies is too
th ick w ith sp irits for everyday h ab itatio n ; but they perhaps, by
th e ir im ag in ativ e exaggeration, set before us the fundam ental
s tru c tu re o f p o p u la r belief. T h e evidence of language is re­
v e a lin g .12 T h e sam e w ord (prostropaios) can be used o f the pollu­
ted killer him self, o f the victim ’s polluting blood, and o f the
v ictim h im self in his anger, o r his avenging sp irits;13palamnaios
is a p p lie d to th e killer, the dem ons th a t attack him , a n d the
(dem onic) p o llu tio n th a t ra d ia tes from h im ;14 w ords likemiastôr,
9 Cf. A p. R hod. Argon. 4. 6 9 9 -7 1 7 , 4 7 7 -9 ; R ohde, 180 f., 5 8 2 -6 .
*° D em . 37.59, E u r. Hipp. 1447—51 (explicitly said to purify T heseus), PI. Leg. 869a
(w h ere a req u ire m e n t o f purification rem ains). Sim ilarly, Iolaus would be polluted by
th e sacrifice o f his d a u g h te r but n ot by her v o luntary self-oblation, E ur. Heracl. 558 f.
11 M o u lin ier, 2 5 9 -7 0 . B ut there are exceptions: for su p e rn atu ral alitêrioi cf. Andoc.
1.130, PI. Ep. 7. 336b; s u p e rn a tu ra lpalammioi, X en. Cyr. 8.7.18. M oulinier’s position is
critic iz ed by V e rn a n t, Société, 127, 132 f.
12 O n m ost o f these w ords see W . H . P. H atch , H SC P 19 (1908), 157-86.
13 K iller: A esch. Eum. 176,237,445; E ur. H F \ 259, ? cf. Heracl. 1015. T h e w ord is also
used o f su p p lia n ts, A esch. A g. 1587, S oph. A j. 1173, Phil. 930, OC 1309, esp. killers,
A esch. Eum. 41, 234. T h e re has clearly been sem antic interference betw een the ‘tu rn ­
in g ’ o f the s u p p lia n t to a n a lta r, an d the ‘tu rn in g ’ o f pollution against those it infects.
V ic tim ’s blood: E u r. H F 1161, Ion 1260. V ictim , or his spirits, as visitant: ? Aesch. Cho.
287, A nt. Tetr. 1 y 10, 2 <5 9 (a ‘p re-a n im isd c’ n eu ter), 3 a 4, /38, δ 10; polluting visitant,
w ith o u t specific reference to m urder, A eschin. 2. 158, Eupolis, fr. 120. προατρέπομαι of
th e victim tu rn in g h is a n g e r against the killer: A nt. Tetr. 3 β 8 (by conjecture), PI. Leg.
866b.
14 K iller: A esch. Eum. 448, Soph. El. 587, Tr. 1207, Phryn. C om . fr. 58. Demons:
X e n . Cyr. 8.7.18. D em onic pollution: E ur. / 7 Ί 2 1 8 (doubted by J . Diggle, Studies on the
Text oj Euripides, O x fo rd , 1981, 88 f.; b u t for dem ons sh o o tin g o u t from a polluted person
see e.g. E ur. Med. 1333).
The Shedding o f Blood 109
alastör, a n d alitërios w ork in very sim ilar w ay s.15 T h ese are
re m a rk ab le sets o f m eanings. T h e killer is prostropaios, b u t so is
the victim ; th e killer, a palamnaios himself, is also attack ed by,
a n d em an a tes, s u p e rn a tu ra l palamnaioi. T h e unifying factor is
(he p o llu tin g act, w hich sets up a chain of ab n o rm al relations
betw een h u m a n s —victim , killer, associates o f killer —th e con­
nectin g links in w hich are s u p e rn a tu ra l powers. I t is h ard to
believe th a t sem an tic configurations o f this kind correspond to
no felt reality.
T h e re are, certainly, m any passages w here the pollution of
m u rd e r is referred to a n d yet there is no suggestion th at the
av en g in g sp irits o f th e victim are a t work. O ften, no d o u b t, the
claim th a t a p a rtic u la r in d ividual ‘has im p u re h a n d s’ is a
re p ro a c h o r a sim ple descrip tio n o f w hat, in custom ary term s,
his ritu al s ta tu s now is, ra th e r th a n a r expression o f a real belief
(h at s u p e rn a tu ra l d angers th rea ten . Even w here the idea of
d a n g e r is ce rtain ly present, how ever, it is not necessarily de­
rived from th e d ead m a n ’s anger. N othing is said ab o u t the
victim in th e fam ous passage o f the forensic A n tip h o n th at
ex p lain s how th e ships th at a m u rd e re r sails in run into danger,
a n d th e sacrifices he a tten d s go a w ry .16 Even though the
E rinyes tak e acco u n t, rem arkably, o f the fact th at O ed ip u s slew
his fa th e r u nw ittingly, a n d do not in the ex tan t texts pursue
him , O e d ip u s rem ain s one of the great polluted figures o f Greek
lite ra tu re ; in th e Hercules Furens, it does not seem to be the
w ra th o f his tin y ch ild ren th a t m akes the m ighty H eracles so
po llu ted . In th e la tte r two cases, b oth o f them instances of
kin-killing, it seem s th at pollution derives not from the w rong to
th e victim , b u t from the violation o f the o rd er o f the family;
th ere is expressed th ro u g h it universal shock, not the p articu lar
a n g e r o f th e victim an d his kin. O f th e A ntiphon passage, and
o th e rs like it, we can p erh ap s say th at alth o u g h m urder-
po llu tio n derives its dynam ic charge from the victim ’s w rath, it
ca n to som e ex ten t retain th a t dynam ism even w hen separated
15 O n alitërios see H atch , op. cit., 157-62; on alastör, Fraenkel on Aesch. Ag. 1501, and
on Z eu s A lastoros, Cl. Rolley, B C H 89 (1965), 4 5 4 -6 . K illers are miastores, A esch. Cho.
944, S oph. El. 275, 07*353, Eur. El. 683, Andr. 615; they are threaten ed by miastores,
A esch. Eum. 176 f., Soph. EL 603, Eur. Med. 1371. Erinys can work sim ilarly, cf. Soph.
EL 112 w ith 1080.
16 A n t. 5. 8 2 -4 .
110 M iasma
from it. I f it seem s a rb itra ry to rep resen t the ‘avenging’ pollu­
tio n o f the Tetralogies as p rim ary, a n d o th er forms as derived, the
ju stific a tio n m u st be th a t it is as an avenger th a t m urder-
p o llu tio n a p p e a rs in the texts w here its threats are m ost vividly
p re sen ted .
T h e significance o f this identification w ith the victim ’s anger
is th a t it affects the w ay in w hich pollution is diffused. A ccord­
in g to th e logic o f the m e ta p h o r o f ‘defilem ent’, it ought to
o p e ra te , in D o d d s’ phrase, ‘w ith the sam e ruthless indifference
to m otive as a typhoid g erm ’, 17 a n d there are certainly texts in
w h ich it is re g ard e d as liable to do ju s t this. In E uripides, for
in sta n c e , the b a rb a ria n king covers his head w hen O restes
p asses ‘so as no t to get a polluting spirit upon m e’. 18 (The
p ro te c tiv e device is as m echanically conceived as the threat.) In
th e Tetralogies, by co n trast, pollution ap p ears as a stern and
d is c rim in a tin g u p h o ld er o f the m oral order. A lthough diffusion
b y physical c o n ta c t is not excluded, the pollution th a t is em ­
p h a siz e d a tta c h e s to those w ho, by om ission or com m ission,
o b s tru c t the v ictim ’s right to revenge. It threatens in the first
in sta n c e the d e a d m a n ’s kin, should they fail to find an d prose­
c u te the tru e killer, a n d secondly the ju ry , should they fail to
c o n d e m n him . In this case the prosecution w ould have done
th e ir d u ty , a n d be safe, b u t the ju ro rs, and through them the
city w hich they rep resen t, deeply endangered. F or Plato too,
p o llu tio n ‘com es ro u n d to’ kinsm en o f the victim who fail to
b rin g a pro secu tio n . B ecause they m ake pollution operate in
th is d iscrim in a tin g way, both au th o rs can identify it w ith the
‘v e n g e a n c e ’ o r ‘en m ity ’ o f the g o d s.19 Before courts existed, it
w as th e th re a t o f the sam e fierce bu t purposeful pollution that
im p o sed on O re stes his terrible revenge.20 Even w hen pollution
is sp re a d by sim ple contact, it rem ains purposive, though in a
different sense; the purpose is to im pose social isolation upon
th e killer, a n d those w ho suffer th ro u g h involuntary association
17 D o d d s, 36.
18 E ur. I T 1218.
19 A n t. Tetr. 1 a 3, y 9,11; Tetr. 2 γ 11-12; Tetr. 3 a 3,4, β 8; PI. Leg. 866b, 871b.
Id en tificatio n : A nt. Tetr. 3 a 3 , PI. Leg. 871 b (in E ur. fr. 82 the gods ‘avenge’ pollutions).
20 A esch. Cho. 2 6 9 -9 6 (cf. E ur. Or. 5 8 0 -4 ); ‘attacks o f the E rinyes’ are spoken of, but
th e sy m p to m s are precisely those o f pollution. It w as the sam e w ith Alcm aeon: E ur. fr.
69 w ith the testim onia.
The Shedding o f Blood 111
w ith him are un lu ck y victim s o fa rule chiefly aim ed a t o th ers.21
In tu rn in g to consider how m u rd er-p o llu tio n is in cu rred , we
e n te r a q u ick san d . T h e p ro secu to r in th e second tetralogy states
in ten tio n to b e irrelev an t to pollution; Sophocles’ O e d ip u s can
d eclare, ‘p u re before the law, unknow ing did I com e to th is.’22
All m u rd erers are excluded from sacred places; yet the chorus
in th e Agamemnon in one place im ply th at it w as only the
com plicity o f his wife in the m u rd er of A gam em non th a t caused
‘pollu tio n o f th e lan d a n d the c o u n try ’s gods’.23 C reon in Sopho­
cles a t first supposes th at, by leaving a few scraps o f food in the
lo m b to w h ich he consigns A ntigone, he is ‘p u re in respect of
lliis g irl’; la te r in the play, h u m b led , he acknowledges him self
h er killer.24 It is possible to a tta c h pollution either to the physi­
cal ag en t, o r to the person u ltim ately responsible for the act. A
d ialo g u e in E u rip id es presents this tension in extrem e term s:
Menelaus: D o you m ean to say you deserve to live?
Orestes·. Yes, a n d be a king . . .
Menelaus: Yes, y o u ’d be ju s t th e m an to h an d le holy w ater.
Orestes: W h a t p revents me?
Aienelaus: A n d perform sacrifice before battle.
Orestes·. H a v e you the right to do so?
Menelaus: O f course. M y h an d s are clean.
Orestes: B u t not y o u r h e a rt.25
S u cn co n ten tio n s w ere no d o u b t often heard. W hen E u ripides’
A chilles, o u tra g e d to learn o f the proposal to sacrifice
Ip h ig en eia, says th a t he him self is polluted by the abuse o f his
n a m e in th e plo t to lure her to Aulis, he is obviously expressing
m o ral revulsion in ritu al term s.26 Since the stain on the m u r­
d e r e r’s h a n d is in fact invisible, it is ju s t as possible to dispute
w h e th e r a p a rtic u la r person is touched by pollution as by guilt,
a n d , since its social consequences are serious, ju s t as necessary.
21 In PI. Euthphr. 4c E u th y p h ro , not obviously by w ay o f p aradox, in fact confines
p o llu tio n to conscious association.
22 Tetr. 2 a 2; S oph. OC 548.
23 A esch. Ag. 16441'.
24 S oph. Ant. 7751*., 889, 1339-46.
25 E u r. Or. 1600—4. For the ascription o f pollution to a person only distan tly responsi­
ble lo r a d eath cf. E u r. Andr. 614 f. Persons m orally responsible for a d eath (even one
th a t d o esn ’t occur) spoken o f as ‘killers’: Soph. Aj. 1126, O T 534, Eur. Hel. 280, Med.
1364, A ndoc. 1. 58.
26 E u r. ΙΛ 9 3 8 -4 7 .
112 M iasma
A m id all this am biguity, it w ould be reassuring to tu rn to the
p recisio n o f a code o f rules. A t A thens, both sets o f the exegetes
co u ld be co n su lted on religious questions arising from violent
d e a th ; o f th eir trad itio n s, unfortunately, virtually nothing is
know n. A law o f the early sixth cen tu ry from C leonai probably
tre a te d m u rd er-p o llu tio n , b u t no certain or even probable in­
fo rm a tio n can be ex tracted from it. T h e G yrene cath artic law
c o n ta in s reg u latio n s for the p urification of the autophonos; it is
fru stra tin g th a t we can n o t be su re w hether this m eans ‘kin-
k ille r’, ‘killer w ith o n e’s ow n h a n d ’, or m erely ‘killer’. A hel-
len istic sacred law from L ato in C rete seems to declare p u re the
p e rp e tra to r o f certain forms o f involuntary hom icide (pushing a
p erso n in a fire, or p o u rin g boiling w ater over h im ).27 From such
d e su lto ry scrap s o f inform ation there is little to be learnt. F or an
e x ten d e d code we m u st tu rn to P lato ’s in Book 9 o f the Laws. Its
m o st strik in g feature is the ac u te sensitivity to circum stances
w ith w hich h e credits pollution. H e lists a series o f conditions
u n d e r w hich ‘the killer w ould rightly be p u re ’: killing of a night
thief, o r o f a footpad in self-defence: killing o f a person sexually
v io latin g a relative o f the killer: killing in defence of a relation.28
(E lsew here he declares pure the m an who kills in self-defence or
d u rin g civil strife, even, rem arkably, if the victim is a brother;
re trib u tio n ag a in st a hom icidal slave is also non-polluting.29)
P la to has lim ited the A th en ian category of justified killing to
ac ts w hich positively serve social o r family solidarity. T hose
a c c id e n ta l killings, in athletics, m ilitary training, or w ar, w hich
a t A th en s fell in to the sam e justified category, have been tran s­
ferred to the low est level o f his class o f ‘involuntary acts of
violence!; they ca rry no penalty, b u t require purification.30
T h u s P lato n ic pollution can distinguish betw een deliberate,
ju s tifia b le hom icide, wholly p ure, a n d non-culpable, acciden­
tal h om icide, w hich by robbing the state of a useful life31 causes
a m ild p o llu tio n . P lato continues to legislate for the m anifold
form s o f hom icide w ith sim ilar casuistry. T he killer o f a slave, by
27 E xegetes: Ja c o b y , 4 1 -5 1 . Cleonai: LSC G 56, w ith bibliography. Cyrene: L SS 115 B
50, cf. p. 351 below . L ato: L SS 112.
28 874 b - c .
29 8 6 9 c - d ; 868 b - c .
30 865 a - b .
31 Cf. 831a.
The Shedding of Blood 113
a c cid en t or in an g er, m ust u n d erg o ‘m o re and g re ater’ purifica­
tions th a n th e p e rp e tra to r o f accid en tal non-culpable hom icide,
b u t a p a rt from co m p en satin g th e ow ner is subject to no further
sa n c tio n .32 A nyone w ho ‘in v o lu n tarily ’ kills a free m an, except
in th e p a rtic u la r m itig atin g circum stances already m entioned,
m u st suffer exile, w hich is for P lato a form of purification, for
periods v ary in g according to th e c h a rac te r of the d eed .33 T he
w hole g ra d u a te d scale o f p u rification an d p u n ish m en t reaches
ils c u lm in a tio n in th e d elib erate p arricid e, for w hom d eath itself
is too little; the m agistrates carry his naked corpse to a
crossro ad s o u tsid e th e city, take each a stone a n d cast it at his
liead, to ‘cleanse the whole city’, an d then h url the body un-
Ijuried o v er th e b o u n d aries o f th e lan d .34
It w ould be m isleading to say th a t P lato has m oralized
pollu tio n , a lth o u g h he is m oving in th a t direction. T races o f a
m ateria l, o b jective conception rem ain .35 Purely accidental
killings m ay re q u ire purification, a n d even exile. T h e m an who
m u rd ers th ro u g h a hired assassin is, Plato insists, ‘polluted in
soul’ a n d m u st be punished exactly like the physical killer; but
P lato allow s him th a t b urial in his native land th a t the delib­
e ra te m u rd e re r is norm ally d en ied .36 Pollution distinguishes in
term s o f social o rd e r as well as m oral intention, and reacts
differently to th e killing of slave by free m an and free m an by
slave. It is clear, however, th a t though Plato surely regards
pollution as a real thing a n d no legislator’s fiction, he is not
m oved by an in d iscrim in ate h o rro r o f shed blood.
M a n y o f th e P lato n ic differentiations und o u b ted ly derived
from A ttic p ractice. For A thenians, as, ap p aren tly , for all
G reeks a t all tim es, blood shed in b attle could sim ply be w ashed
off.37 T h e p e rp e tra to r o f ‘ju stified hom icide’, or at least certain
32 865 c, 868 a.
33 See the tab le in W . K noch, Die Strafbestimmungen in Platons Nomoi, W iesbaden, I960,
162 f.
34 873 b.
35 Cf. R everdin, 177 if.
36 872a.
37 T h e only text suggesting th at G reek soldiers purified them selves form ally after
b a ttle is A esch. Sept. 679—82. It can n o t be allowed to weigh ag ain st the total silence o f
th e h isto rian s, w hose im plication is echoed by Eur. Ion 1334 καθαρός άπας τοιπολεμίονς
υς αν κτάνη, PI. Leg. 869 d καθάπερ πολέμιον άποκτείνας εστω καθαρός; cf. A ndoc. 1. 97,
a n d th e d eclaratio n o f w ar on the S p artan helots by the ephors. A regular purification
a fter h u n tin g is claim ed only by A rr. Cyn. 33.
114 M iasma
categories of it, w as form ally considered pure; private scruples
m ig h t h av e caused him to seek purification, b u t no one could
p ro se cu te him for en terin g the tem ples w ithout it.38 C ertain
killings in lite ra tu re th a t fall into no precise legal category are
p ro b a b ly th o u g h t o f in roughly these term s. T h u s in E uripides,
th o u g h H eracles does speak o f ‘cleaning his h a n d s’ from the
killing o f L ycus, it is clear th a t this absolution will be autom atic
a n d final; th e sim plest ritu al can efface the blood o f a villain.39
T h e ‘in v o lu n ta ry ’ killer, by contrast, incurred exile, an d could
n o t re tu rn before he h ad ‘sacrificed a n d been purified’.40 P lato’s
p a rtia l extension o f pollution to the a u th o r o f a m u rd er as well
as its a g e n t is also A ttic. By law, ‘the deviser was subject to the
sam e pen alties as the m an w ho did it w ith his h a n d ’ and, since
‘th e d ev iser’ could th u s be prosecuted for m urder, he will have
b een ex cluded from the shrines for the period before the trial.
VVe h e a r o f an inform er, w hose m urders w ere perform ed by
legal process, bein g sh u n n ed ‘like a polluting dem o n ’, ‘as a
m u rd e r e r’, a n d eventually being b ro u g h t to trial for entering
th e sac red places alth o u g h ‘m anifestly’ a killer.41
T h e real pro b lem th at the subject presents has begun to
em erg e in this discussion. In assessing the intensity of pollution,
w e h av e been ap p ealin g not m erely to ritual criteria b u t also to
legal p en alties. T h is has been in accord w ith the practice of
A ttic a u th o rs , w ho com m only tre a t exile itself as a form o f
p u rific a tio n .42 T o consider m erely the n u m b er an d intensity of
ritu a l lu stratio n s th a t a p a rtic u la r act required w ould be quite
to m iss th e ir conception o f w h a t the im plications o f pollution
a re . R itu a l a n d legal statu s are assim ilated to the extent th at in
co n tex ts o f hom icide ‘p u re ’ a n d ‘not subject to legal sanctions’
a re often sy nonym ous.43 A few acts, such as the killing of a
38 See A ppendix
39 Kur. H F 940. T h e ritu a l envisaged is unclear; 923 speaks o f ‘rites to purify the
h o u se ', 940 a n d 1145 of ‘cleansing the h a n d s’, but w hat H eracles seems to be preparing
is a no rm al sacrifice (R u d h a rd t, 270, M oulinier, 88). O dysseus purifies his house, but
n o t him self, a fter th e m u rd e r o f the suitors (Od. 22. 4 8 1 -9 4 ).
40 D em . 23.72.
41 D eviser: A ndoc. 1. 94, cf. Aeschin. 1.172 w ith 2.148. Inform er: Lys. 13.79,81,85-7
(cf. M acD o w ell.Homicide, 131-3).
42 e.g. A esch. Ag. 1419 f., Cho. 1038, Kur. Hipp. 35, A nt. Tetr. passim, PI. Leg. 8 6 5 d —e;
see too N ie. D am . 90 F G rH fr. 45.
43 See A pp en d ix 5.
The Shedding o f Blood 115
d e p e n d e n t, m ay have req u ired purification, an d even som e
seclusion o f th e killer, w ith o u t being subject to legal penalty,
h u t an A th e n ia n w ould p ro b ab ly have said th a t they were
‘polluting, b u t n o t sufficiently p o llu tin g to req u ire exile’, ra th e r
th a n ack n o w led g in g them as a real exception to the principle.
(A n o th e r special case, w hich will be m entioned later, is th at of
kin-killing.) T h is correlation betw een legal an d ritu al req u ire­
m en ts, how ever, im poses the q u estion of the causal relation
betw een th e two. D o they coincide because the th rea t o f pollu-
tion is one facto r am o n g others, o r even the do m in atin g factor,
th a t th e leg islato r took into account? O r has pollution, a religi­
ous id ea a n d not, in itself, a pow erful d eterm in a n t of action,
w rap p ed itself ro u n d the law like ivy ro u n d the oak an d proudly
claim ed th e sh a d e th e la tte r casts to be its own?44
T h e difficulty w ith the first a p p ro a c h is th at it is h ard to give
m ean in g to th e id ea of a fear o f pollution th at is som ehow quite
d istin c t from all th e o th er m otives th a t d eterm in e responses to
hom icide. I f it is fear o f pollution alone th a t causes the killer to
be exiled, th e u n a ttra c tiv e conclusion seems to follow that,
w ith o u t it, his v ictim ’s kin w ould let him live on unm olested. It
is also im possible on this hypothesis to explain why, in G reek
society, som e form s o f killing are entirely pure, while even in
societies th a t req u ire purification after w ar an d the h u n t these
socially ap p ro v ed form s o f killing are m uch less polluting th an is
m u rd er. Som e h ave supposed th at, though G reek responses to
hom icide derive in origin from such fam iliar motives as the desire
lor revenge, p o llu tio n in tru d ed a t a p artic u la r historical
m o m en t to p u sh th e in stitu tio n in a new direction. T h u s it has
often been arg u ed th a t the first codification of m u rd er laws by
D raco in th e sev en th cen tu ry w as a response to th e grow th of
h ith e rto unkn o w n fears o f p o llu tio n .45 If this w ere tru e, the
novel fears w ould them selves aw ait an explanation. B ut the
p o stu la te d tran sfo rm atio n , su d d en a n d otherw ise inexplicable,
44 Λ view close to this is well p u t by M acD ow ell, Homicide, 1—5, 141-50; but n o tcT . J .
S au n d ers, J H S 85 (1965), 225. G ag arin , Drakon, 164-7 sees doctrines of pollution as
po st-D raco n ian .
45 e.g. by E. M eyer, Geschichte des Altertums, I I I 2, S tu ttg art, 1937, 528—34;
W ilam o w itz, Das Opfer am Grabe3, Berlin, 1907, 8 f.; B onner/S m ith, i.53. V igorous and
elVective criticism in C a lh o u n , 2 5 -4 1 ; a su b tle discussion by L. G ernet, Annales 10
(1955), 530—3. G lotz, 225—37 has a m ore ingenious v arian t, well criticized by
I \ F au co n n ct, L'Année sociologique 10 (1 9 0 5 -6 ), 4 7 5 -8 .
116 M iasma
in th e tre a tm e n t o f killers sim ply does not occur. D raco passed a
h o m icid e law , b u t in a society in w hich au th o rity w as gradually
b ein g ce n tralized , m u rd e r w as unlikely to be left uncontrolled
by law , since public cognizance o f hom icide seems to be a
d istin c tiv e m ark o f a centralized political system .46 H e o r a
successor p ro b a b ly forbade the acceptance o f blood-m oney (the
d etails a re q u ite o bscure)47 b u t such a restriction on the power
o f in d iv id u a l citizens to b a rte r w ith life an d d e a th could also
h av e been p re d ic ted as p a rt o f the process o f centralization of
a u th o rity . E ven if these reform s w ere justified as a defence
ag a in st p ollution (as is the b an on accepting com pensation in
th e O ld T e s ta m e n t),48 a self-m oving fear o f pollution was
clearly no t th e ir tru e inspiration. B ut it has often been noted
th a t th ere is no m ention o f pollution in the surviving portions of
D ra c o ’s law , a n d o th e r a u th o rs o f the period w ho have m uch to
say a b o u t ju stic e a n d the w elfare o f the com m unity (Hesiod,
A rc h ilo ch u s, A lcaeus, Solon) do no t seem to be h au n ted by the
sp ec tre o f th e u n p u n ish ed m u rd e re r lurking in its m idst.
T h e a lte rn a tiv e conception, w hich m akes pollution a kind of
sh ad o w y sp iritu a l Doppelgänger o f the law, is therefore more
attra c tiv e . N o t ju s t D raco ’s b u t all surviving hom icide laws
ig n o re it alm o st entirely.49 Som etim es the appeal to pollution in
classical a u th o rs ap p e ars as alm ost a rationalization o f an in­
stitu tio n w hose h istorical origins we can actually see to be
different. T h e ‘in v o lu n ta ry ’ killer u n d e r A th en ian law was re­
q u ire d to go in to exile until p ard o n ed by the v ictim ’s kin, for a
p erio d th a t could in theory be indefinite alth o u g h it seems
no rm ally to h av e been fixed a t a year. T h is w ithdraw al can be
ex p lain e d as a response to pollution. Plato quotes a n ‘old m yth’:
th e free m an , freshly d ead a n d angry a t his p re m a tu re and
46 J . B eattie, Other Cultures, L ondon, 1964, 156. O n the introduction o f legal process
for hom icide see E. R uschenbusch, Φόνος, Historia 9 (1960), 129—54; L atte, Λlord;
G a g a rin , Drakon; above all H . J . Wolff, ‘T h e O rigin o f Ju d icial Legislation am ong the
G re e k s’, Traditio 4 (1946), 3 1 -8 7 . T h e 6 th -cen tu ry Sicilian hom icide law S E G iv 64 is
u n fo rtu n a te ly too m u tilated to be revealing.
47 See for different views R ohde, 211 n. 154; G lotz, 314 f., 439 f.; B onner/S m ith, ii.
196—8; L atte , Mord, 284 = Kl. Sehr. 387 (a different nuance Kl. Sehr. 274); M. G agarin,
G R B S 20 (1979), 303. Survival or revival of blood-m oney outside A ttica: lnscr. Prien. 84;
M ich el 524 C 20 f. (Ilion).
48 N u m b ers 3 5 :3 1 -3 .
49 T h e only A ttic exception (M acD ow ell, Homicide, 148) is Dem . 23.72.
The Shedding o j Blood 117
violent en d , ca n n o t b ea r to see his killer at large in the places he
h im self freq u en ted , an d m ay seek revenge.50 In m ythology and
I lom er, how ever, we find the involuntary an d even the
justified’ killer (of later classification) subject to p erm an en t
exile, a n d th e force th a t drives him o u t is not ju s t pollution but
p u rsu it by th e v ictim ’s kin.51 T h e classical in stitu tio n is a
m itig ated su rvival o f the pre-legal procedure. T h e re is, of
course, a n elem en t in such a case th a t in m odern term s appears
irra tio n a l, a n d th a t some m ight wish to explain th ro u g h fear of
pollu tio n . T h e v ictim ’s kinsm en, faced by the m onstrous fact of
Iiis extin ctio n , decline to take acco u n t o f m otive. ‘H e did it
in v o lu n ta rily ’, says the p ro secu to r in the second tetralogy, ‘b u t
I he affliction he b ro u g h t u p o n m e is n o t less th a n it w ould have
been h ad he do n e it w ith in te n t.’ (H e adds th at even ap p a re n t
accid en ts m ay be in stru m en ts o f divine vengeance; the sugges­
tion is m ost revealing, b u t n o t essential to his p o in t.52) An
in n o cen t m a n has suffered violent d eath , a n d his d ea th m ust be
la k e n o u t on its ‘cau se’, regardless o f w h eth er an in tention to kill
was p re se n t o r even possible.531n A thens, anim als or inanim ate
o bjects th a t h ad caused d eath w ere tried a t the P ry tan eu m and,
if found guilty, expelled beyond the boundaries o f A ttica.54 T he
a c c id e n ta l killer h ad therefore to try to d em onstrate, not th at he
w as m o rally in n o cen t, b u t th a t he w as not causally responsible
for w h a t o ccu rred at all.55 B ut the basis of the in stitution seems
no t to be fear of pollution b u t the urge to exact retrib u tio n , and
be seen to ex act it, for an injury th at has been received.
S im ilar p ractices are found in societies which lack the m etaphor
o f blood p o llu tio n , a n d the reason why the irrational or in an i­
m ate killer is ‘cast outside the la n d ’ is by assim ilation to the fate
50 Leg. 8 6 5 d -e .
51 II. ‘2 3 .8 5 -8 , Od. 2 2 .2 7 -3 2 . Kxile lor accid en tal killing in the m yths e.g. οΓ'Aetolus,
C e p h a lu s, O 'xylus, P erseus (A ppendix 7), an d the story of A drastus (H dt. 1. 35), tor
ju stifiab le hom icide in the m yth of’H y ettus (Hes. fr. 257). Cf. G lotz, 491'. Life-exile tor
accid en tal ho m icid e in classical S p arta, X en. An. 4.8.25.
52 Tetr. 2 a 2, γ 7 -8 .
53 See G ern et, 3 0 5 -8 8 on the 'objective crim e’, also Dover, 152 f., 159; on ‘the cause
o l'd e a th ’ G ern et, 3 6 8 -7 1 , A dkins, 103-7.
54 M acD ow ell, Homicide, 8 5 - 9 , cf. PI. Leg. 873e-874a. Sim ilar practices elsewhere
in G reece, P aus. 5.27.10, 6.11.6.
55 A dkins, 103—7. Such evasions are, however, prim itive, not sophisticated, cf. e.g.
th e p u n ish m en t o f the axe at the B ouphonia (Paus. 1.24.4), and tor the form of
a rg u m e n t alread y A esch. Cho. 923.
118 Miasma
o f th e ra tio n a l killer. (By a sim ilar assim ilation, hom icidal pigs
in th e M id d le Ages were h u n g .56) T h e idea th a t the hom icidal
ax e is po llu ted , o r th a t the victim w ould be angry if his acciden­
tal killer w ere not expelled, is a secondary elaboration upon the
p rim a ry d esire for retrib u tio n .
A n o th e r clear in stan ce o f the w ay in w hich the concept of
p o llu tio n fits ro u n d the legal o r pre-legal institution is the status
o f th e killer in exile. His taint, ineffaceable at hom e, disappears,
o r a t least becom es open to purification, as soon as he reaches
foreign soil. T h e victim is only angry, it is said, a t seeing the
k iller ro a m in g a t large in the places he him self once frequented
(h e sh ares, therefore, the pro b ab le feelings o f his surviving
k in ).57 A few crim es were, it is som etim es claim ed, so h o rren d ­
ous th a t no city w ould provide refuge for their p erp etrato rs58
(th e re is a h in t in Sophocles th a t at A thens the A reopagus m ay
h av e in terv e n ed in such cases), b u t in general the principle of
‘b u t th a t w'as in a n o th e r c o u n try ’ m u st have applied. ‘T hose
w ho a re in exile for killing, once they m ove to a n o th er city, are
n o t tre a te d as enem ies by those who receive th em ’, says
L y c u rg u s.59 B ut w hile the p e rm a n e n t exile was p u re in respect
o f his new su rro u n d in g s, b u t still polluted should he seek to
re tu rn hom e, for the invo lu n tary killer tem porary exile was
itse lf a kind o f cleansing, d u rin g w hich his pollution ‘was ru b ­
bed off’ or ‘fell asleep ’ (since ‘tim e purifies all things’),60 ready
to be finally rem oved by purification w hen he cam e back to his
n a tiv e soil.
A fu rth e r in sta n c e w here pollution offers an ex tra explana­
tion o f a n in stitu tio n th a t can also be explained in o th er term s is
56 E. P. E vans, The Criminal Prosecution and Capital Punishment o f Animals, London,
1906; on d e o d a n d (penal su rren d er o f hom icidal objects) see K en n y ’s Outlines o f
Criminal L aw , 19th edn. by J . VV. C . T u rn e r, C am bridge, Î966, 7 f. In old English law
‘Legis enim est qui inscienter peccat, scienter emendet’ vel. sim. (Leges Henrici Primi, cd. L. j .
D o w n er, O xford, 1972, 88.6a, 9 0 .11a, 70.12b: followed how ever in each case by a
reco m m en d atio n to m ercy, see D ow ner’s references on 70.12a), ‘the thought o f m an
sh all n o t be trie d , for the devil him self know eth not the thought o f m a n ’ (a late medieval
law y er cited P o llo ck /M aitlan d , 2.474).
57 PI. Leg. 865e; for th e k in’s sentim ent cf. the quotation in C am pbell, 198: ki
su d d e n ly saw him there, d rinking an d p u ttin g on airs. I rem em bered th at his brother
V asili d re w blood from m y brother. I could not sta n d it.’
58 S oph. O C 9 4 4 - 5 0 (A reopagus), E ur. E l. 1194-1200, H F 1286-90, Hipp. 1066 Γ,
M ed. 847—50, Lys. 12.35, (Lys.) 6.16,30, Lvc. Leoc. 133.
59 Leoc. 133, cf. D em . 23.39.
60 A esch. Eum . 238, 280, 286.
The Shedding o f Blood 119
th a t o f th e m in o r A th en ian co u rt ‘a t P h re a tto ’, w hich h eard the
case o f a n y o n e accused o f d elib erate m u rd er w hile already in
exile for hom icide. It m et on th e shore, a n d th e accused m an
a p p ro a c h e d it from the sea; he was n o t allow ed to set foot on
land , b u t p lead ed his case from the b o at.61 O bviously these
reg u latio n s can be in terp re ted as a device to pro tect the land
from p o llu tio n , a n d this is no d o u b t how m an y A thenians
u n d ersto o d th em . E qually, how ever, they p ro tected th e exile
h im self w ho, if he set foot in the forbidden territo ry , becam e an
o u tla w to be killed w ith im p u n ity . T h e site o f th e court em p h a­
sizes w ith form al arch aic sym bolism th at the exile is not b reak­
ing bo u n d s.
P o llu tio n ’s lack o f real coercive force of its ow n is clear from
the case o f th e victim w ith o u t a p atro n , m entioned by A nti­
p h o n .62 H e is p ro b ab ly th in k in g o f the slave killed by his own
m aster, a lth o u g h it is not im possible th at the child killed by his
fa th e r is also envisaged. ‘Even w hen a m an kills som ebody he
co n tro ls him self, so th at there is no one to avenge him , he still, in
re sp ect for cu sto m a n d the gods, purifies him self an d keeps
aw ay from th e places laid dow n by law, thinking th a t this will be
b est for h im .’ U n less A n tip h o n is being disingenuous, we m ust
assu m e th a t th e killer here avoids the forbidden places not for
fear o f a p ro secu tio n by seizure, b u t from p rivate scruples.
H a v in g sh ed blood, he fears th e consequences (disease, m ad ­
ness?), a n d hopes to escape them by v o luntary subm ission
(A n tip h o n does not say for how long) to the restrictions th at
w ere n o rm ally im posed on the killer perforce. T his is indeed
m ost striking, a n d w arns ag ain st too external a view o f w hat
p o llu tio n m eans. T h e dan g er, however, seems to be confined to
th e killer him self, since his pollution does not req u ire him to be
d riv en into exile o r subjected to an y legal restrain t. T h e son who
ac tu a lly p ro secu tes his fath er for causing an u n p ro tected de­
p e n d e n t’s d e a th , claim ing to fear pollution, is b ran d ed thereby
as a fa n atic.63
61 M acD ow ell, Homicide, 8 2 - 4 , cf. Paus. 2.29.10, PI. Leg. 8 6 6 c -d , 867e (scrutiny at
th e frontier).
62 (i.4, cf. 5.87. Plato requires no m ore th an purification in such cases, Leg. 865d,
868a.
63 PI. Euthphr. 3e—4d. T h e killing o f one dep en d en t by ano th er, the sam e passage
show s, req u ired purificatio n , an d punitive m easures against the hom icide, but not of
co u rse legal process.
120 M iasma
P ollution by itself m akes n o th in g happen. B ut to speak o f it as
a ra tio n a liz a tio n is unjustified, because there is no reason to see
it as chronologically secondary, w hile to trea t it as m erely the
religious shadow o f a legal in stitu tio n w ould sep arate the two
a re a s in a w ay th a t seem s u n tru e to archaic G reek attitudes.
I t m ay be helpful at this p oint to seek guidance from ethno­
g ra p h y .64 T h e belief th a t killing pollutes is very w idespread
a n d , in its d etailed ap plication, very diverse. T h is diversity is
its e lf im p o rta n t. Som e pollutions th reaten the killer only, others
th e killer a n d his v ictim ’s kin, others again the w hole society.
S om e can be cleansed im m ediately o r after a short period, while
o th e rs d e m a n d the expulsion o f the offender from the group.
T h e s e divergences are found bo th betw een societies an d in the
tre a tm e n t by th e sam e society o f different forms o f killing. O ften
socially ap p ro v ed killing causes a m ild pollution an d sham eful
killing a very severe one. A n o th er a n d even m ore im p o rtan t
p o in t th a t em erges from an y careful eth n o g rap h ic account is
th a t p o llu tio n is a kind o f in stitu tio n , the m etaphysical justifi­
ca tio n for a set o f conventional responses to the disruption of
n o rm a l life th ro u g h violent d ea th . K illing causes pollution ju s t
as, a m o n g us, d e a th causes m ourning; because o f pollution
v ario u s av o idances are practised, ju s t as because o f m ourning
black suits a re do n n ed . C lassical scholars, by contrast, have
te n d e d to in te rp re t the p h enom enon in term s o f em otions,
cre d itin g th e G reeks w ith th a t h o rro r o f spilt blood th a t they
im ag in e they them selves m ight in the circum stances have felt.
B u t th o u g h su ch h o rro r is no d o u b t the source o f the im agery of
p o llu tio n , it c a n n o t explain the w ay in w hich ‘the polluted’
c o n tin u e to perform a set o f stan d ard iz ed acts, a n d continue to
be av o id ed by o utsiders, how ever little horror m ay be felt in the
p a rtic u la r case. A clear exam ple o f pollution as a kind o f insti­
tu tio n com es from the N u e r.65 W hen a killing occurred, the
64 See e.g. J . G . F ra zer on P ausanias ‘2 .7.7, idem ., The Golden Bough3, iii, London,
1 9 1 1 ,1 6 5 -9 0 ; P. B o h a n n an (ed.), African Homicide and Suicide, P rinceton, 1960;
references in I. S c h a p e ra , ‘T h e Sin o fC a in ’,Journal o f the Royal Anthropological Institute, 85
(1955), 33—43; references in following notes. F or differential pollution note e.g. G. M.
W ilson, in P. B o h a n n an , op. cit., 182 (m an slau g h ter pollutes the killer, m u rd er the
w hole co m m u n ity ); D ouglas, 106 f. (only the killing o f fellow -tribesm en causes hunting
failu re).
65 See E v a n s-P ritc h a rd , 2 9 3 -7 , idem , The Nuer, O xford, 1940, 152—5; sim ilarly
a m o n g M a n d a ri, B uxton, 227 f.
The Shedding o f Blood 121
m u rd e re r h im self req u ired im m ed iate purification. H e received
it from a chief, in w hose house he th en lived until the feud was
settled . T h e re rem ain ed a k ind o f relation o f pollution between
th e kin o f th e victim a n d o f th e killer. I f a m em ber o f eith er cam p
a te o r d ra n k from a vessel belonging to the o th er, he would
su rely die. T h e pollution w ould also com e into effect if a third
p a rty in a d v e rte n tly used the vessels of both sides. T h is state
c o n tin u e d u n til cattle w ere p aid in com pensation, to buy the
d e a d m a n a lev irate bride, a n d the feud was w iped out by
sacrifice. T h e se san ctions obviously gave sym bolic expression
to th e social g u lf created betw een th e two sets o f kin by the act of
killing. W h e n th e o rd e r dislocated by the m u rd er was restored,
the p o llu tio n en d ed . T h ey also o p erated as a discreet pressure
to w ard s settlem en t, since th e need to g u ard ag ain st a third
p a rty settin g th e pollution off im posed tiresom e restrictions on
all con cern ed .
P ollu tio n , therefore, is not so m u ch a ratio n alizatio n as a
vehicle th ro u g h w hich social d isru p tio n is expressed. N aturally
it is closely asso ciated w ith the d ead m a n ’s anger, b u t even this
is ju s t a n o th e r w ay o f expressing the sam e sense o f disruption.
S ince th e d iso rd er is the pollution, any action th a t restores the
no rm al eq u ilib riu m of things becom es a purification. A word
th a t is often found in this context is the verb hosiö or aphosid,
w hich conveys th e idea o f restoring religious norm ality and
th ere b y p u ttin g o neself in the clear. T h e ways o f doing this are
vario u s. T h e v erb is used in relation to exiling a killer, to
b rin g in g him to co urt, an d to hurlin g stones at a p arricid e’s
n ak ed corpse. A householder w ho cannot prosecute the killers
o f a n elderly fem ale d ep en d en t, because she is u n related to him,
can only purify his house, but, since this is all he can do, even
this p u ts him in th e clear.66 T h e crucial point is th a t w hatever
can be d o n e sh ould be done. As we have seen, pollution nudges
th e v ic tim ’s relatives into bringing a prosecution to avenge their
kin sm an , a n d th en tu rn s its atten tio n s to the ju ro rs and,
th ro u g h th em , th e entire com m unity.
T h e o rd e r w hose restoration pollution dem ands is not sim ply
a m a tte r o f peace a n d quiet. T h e solution for w hich it works is
no t th e easiest, b u t the one w hich reflects the society’s sense o f
66 E u r. Or. 515, PI. Euthphr. 4c, Leg. 873b, 874a, Dem . 23.73,47.70. Cf. A ppendix 1.
 122 M iasma
w h a t is p ro p e r. T h u s, though the N u er pollution encourages an
h o n o u ra b le settlem en t betw een the two sets of kin, it renders
sh am efu l co n nivance w ith o u t p ay m en t o f cattle im possible.
S tro n g religious sanctions o f the sam e kind are also found in
G reece. A n tip h o n explains th a t m u rd er trials are held in the
o p e n a ir n ot m erely to protect the ju ro rs from pollution, b u t also
‘so th a t th e p ro secu to r should not share a roof w ith his kins­
m a n ’s slay er’,67 a n d num erous texts speak o f voluntary associa­
tion w ith a k in sm a n ’s killer as the w orst o f crim es, com pulsory
asso c ia tio n as the bitterest o f degradations. In E uripides’
Andromache, M en e la u s can profess to be shocked th at Peleus is
p re p a re d to e n te r the sam e roof as A ndrom ache, ex-wife o f the
b ro th e r o f th e killer o f P eleus’ son th a t she is.68 T h is sense o f a
sp ecial c o rru p te d relation created betw een families by the act of
killing is reflected sem antically in the w ord authentês, w hich in
ea rly usage is co n stru cted w ith a dative of disadvantage: X is
authentês to Y if he has killed one o f Y ’s kin.69 M u rd e r w ithin the
fam ily creates the sam e kind o f relationship o f pollution be­
tw een the killer a n d the surviving relatives. F or expiable forms
o f kin-killing (fratricide in anger, killing o f child by p aren t, but
n o t the reverse, in anger) P lato in the Laws im poses restrictions
o n th e killer even after his re tu rn from exile. H e m ay not resum e
a n y form o f association w ith his family, because he has d e­
stro y ed one o f th eir relatives, a n d if he does, both he an d they
a re liable to p rosecution for im piety. Plato is certainly reflecting
A th e n ia n se n tim e n t here, alth o u g h we know nothing o f the
form al legal po sitio n .70 ‘T h e law ’ forbids H eracles to bury the
67 5. 11, cf. A rist. Ath. Pol. 57.4. O . W einreich, H erm es 56 ( 1921 ), 326—31 ( = Kl. Sehr.
i, A m ste rd a m , 1969, 5 5 2 - 7 ) refers this custom to the purifying pow er o f sun and rain
(cl. 532 FG rH , p. 513 p a ra . 2); A n tip h o n ’s explanation in term s o f the symbolic
m e a n in g o f th e sh a re d ro o f (cf. II. 9.640, L SS 115 A 16-20, and m any texts ab o u t to be
cited on th e ‘authentês re la tio n ’) is far superior.
68 Kur. Andr. 654—9. O n the ‘authentês relatio n ’ between families an d social groups cf.
. S oph. El. 2 6 2 -7 6 ,3 5 8 ,5 8 7 ,1 1 9 0 , Ο Γ 821 f., E ur. Andr. 170-4, T huc. 3.58 (!), Isae. 9.20,
(A ndoc.) 4.22, A rist. Ath. Pol. 18.6, Dem . 18.257: all this abhorrence lies behind Horn.
II. 24.505 I. T h e relatio n d isregarded, Hes. Scut. 11.
69 1,. G ern et, Droit et société dans la Grèce ancienne, Paris, 1955, 2 9 -3 8 ( = R E G 22 (1909),
13—32); see esp. H d t. L I 17.3, Soph. El. 272, E ur. Andr. 172, ΙΑ 1190, Tro. 660, Rhes.
873, A n t. 5.11. T h e special usage w ith dativ e o f disadvantage (expressed or implied)
th a t G ern et estab lish es rem ains a fact even ifC h a n tra in e ( ’Αφιέρωμα Τριανταφυλλίόη,
T h essalo n ik i, 1961, 8 9 -9 3 ) is right to m ake the basic m eaning ‘responsible’. F. Zucker,
‘A u th e n te s un d A b leitu n g en ’, Sitz. Leipz. 107.1962.4, does not yet know C hantraine.
70 PI. Leg. 8 6 8 c -8 6 9 a . B ut cf. n. 79 below.
The Shedding o f Blood 123
ch ild ren he h as slain, a n d it w ould be sinful (not themis) for
A g am em n o n , h av in g sacrificed one o f his children, to em brace
the o th ers;71 in a historical case, continued association by the
victim ’s b ro th e r w ith a p resu m ed father-killer leads to a prose­
cu tio n for im p ie ty .72 Society’s p ractical pow er to insist upon
w h a t it believes to be rig h t is here very weak. K in-killing is
u tterly a b o m in a b le ,73 b u t since revenge (and subsequently p ro ­
sec u tio n 74) belongs exclusively to the victim ’s relatives, w ho are
also relatives o f th e killer, the pressu re tow ards connivance is in
p ra ctice very stro n g .75 T h e fath er w ith two sons, one o f whom
kills th e o th er, is left helpless in his old age if h e expels the
offender as he should. P ollution does its best to reassert the
claim s o f th e victim ag ain st those of convenience (or even, as in
th e Oresteia, ag a in st those o f b ro a d er social ord er). Fam ily
m em b ers w ho d isreg ard it invite divine p unishm ent. Too m uch
tru s t is p laced in th e gods’ clem ency, we hear, by the ‘father
w ho sh ares his house w ith sons w ho have shed k indred blood’.76
I ts p ra c tic a l success, o f course, in forcing kin-killers into exile is
h a rd to assess. E xpulsion o f hom icidal relatives by the rest of the
clan , a n d v o lu n tary w ith d raw al ‘in obedience to the law ’, are
b o th found in m ythology,77 b u t ab o u t the fate o f actu al kin-
killers in A th en s th ere seems to be no scrap of evidence.
71 E u r. H F 1361, IA 1191 f.; cf. Aesch. Cho. 909, H d t. 3.50.3.
72 D em . 22.2. T h a t ‘A th èn es ap p liq u a sy stém atiquem ent ce prin cip e’ (Reverdin,
188, cf. G lotz, 436—8) the evidence o f one m alicious prosecution does not dem onstrate.
M acD ow ell, Homicide, 9 f. goes too far in inferring th at failure to prosecute in any
hom icid e case could lead to an im piety charge; the case in question in Dem . 22.2 is one
o f p arricid e, a n d the ollence is ‘association’ w ith the killer, not failure to prosecute;
w here the killing h ad occurred betw een fam ilies, ‘association’ would not norm ally
arise. T h e an alo g y w ith P lato (Leg. 866b, 87 lb , etc.) is m isleading here, as P lato is very
free w ith im piety charges.
73 O n th e h o rro r o f kin-killing, w hich is in fact ‘self-killing’ (cf. p. 351 below on aulo-
co m p o u n d s), see e.g. A esch. Sept. 681 f., E u r. H F 1074—6, Med. 1268—70, fr. 82, PI. Leg.
872c - 873b. Its tain t m ay persist ab ro ad , cf. m ost o f the passages cited p. 118 n. 58.
P arricid e fo rbidden b u rial in native land?: Soph. OC 407. For discussion see G lotz, 44
f., 2 3 2 - 6 , 3 2 1 - 3 ,4 3 4 - 8 .
74 See m ost recently M. G ag arin , O R B S 20 (1979), 302-13.
75 See I. S ch ap era, op. cit.; B lack-M ichaud, 2 2 8 -3 4 .
76 E u r. fr. 645.4.
77 E xpulsion: H orn. II. 2.665 f., 16.573 f., H dt. 1.35.3, Apollod. 3.12.6; G lotz, 44 f.
V o lu n ta ry w ith d raw al by ty ran ts m historical rom ance: Nie. D am . 90 FGrH fr. 45, 61,
P arth , Amat. N a n . 14.5. O th e r v oluntary responses to pollution, Apollod. 2.4.12, 2.7.6.
N ote too th e m oral p ressu re supposedly exerted on the tyrant Periander, w ho had killed
liis wife, by his son, H d t. 3.50—53.
124 M iasma
B ecause p o llu tio n expresses a sense o f disorder, little or none
o f it resu lts from killings th a t are felt to be quite appropriate.
T h is is w hy its d em an d s an d those of the law norm ally coincide
so closely. W here public sen tim en t swings in favour o f a
p a rtic u la r form o f killing (of the ad u lterer, or the tyrant) there
p o llu tio n gives w ay w ith o u t a fight. Even in civil strife, pollution
w as h eld e ith e r n o t to be in cu rred a t all, or to be willing to yield
to a sim ple collective purification.78 In the occasional cases
w h e re th ere is a clash o f interest betw een pollution an d m oral
feeling o r th e law , pollution is still stan d in g out in favour of a
p rin c ip le o f o rd e r w hose validity in norm al circum stances is
u n iv ersa lly accep ted . P arricide is the m ost ap p allin g o f acts. A
sla n d e ro u s im p u ta tio n o f it is one o f the ‘unsayable things’ and
liab le to legal action, a n d the possibility th at it m ight occur
in a d v e rte n tly is a n objection o f self-evident validity to the sex­
u al co m m u n ism o f the Republic.19 (T h e p artic u la r h o rro r is that
su ch cases w ould n o t even be recognized, an d thus the ‘custom ­
a ry so lu tio n s’ could not be applied; this d an g er im pressed
A risto tle him self.) In co n tra st to the norm al p a tte rn , such a
v io latio n o f the o rd e r o f the fam ily can, it seems, be polluting
th o u g h legally p u re .80 It would be h a rd to prosecute O edipus,
w ho killed his fa th e r unknow ingly in self-defence, an d yet he is a
d a n g e ro u s m a n to encounter. B ut it is, of course, the crucial
im p o rta n c e o f the fa th e r’s inviolability th a t causes the pollution
to spill over even on to invo lu n tary cases; the h o rro r is even
in cre ased by the fact th a t the violation o f fu n d am en tal o rd e r has
o c c u rre d a t ra n d o m . In the sam e way, it is because the m other’s
rig h t no t to be killed by h er son is in general unquestioned that
p o llu tio n a tta c h e s to O restes a n d A lcm aeon, even though in the
p a rtic u la r case th e ir act m ay be justified.
It is obvious th a t m urd er-p o llu tio n differs in im p o rtan t re­
sp ec ts from those caused by b irth an d death. All these pollu­
tio n s a re p ro d u c ed by breaches o f order, b u t the source of
d is tu rb a n c e is q u ite distinct in the different cases. M urder-
p o llu tio n is caused by an u n n a tu ra l act, an d for this reason is
v irtu a lly identified, as we saw, w ith the anger o f the m an
78 PI. Leg. 869c, X en . An. 5.7.35, Paus. 2.20.2. In m ost cases o f stasis we h ear nothing
a t all of ritu a l consequences.
79 Lys. 10, passim; A r. Eccl. 638—40; A rist. Pol. 1262a 31 f. ‘Solutions’: exile followed
by dissolution o f th e fam ily, as in Plato, G lotz suggests, p. 234.
80 S oph. OC 548.
The Shedding o f Blood 125
u n n a tu ra lly killed. T h is an g e r th en directs itself in ways th at in
th eo ry enforce the expulsion o f th e killer from the com m unity,
liirth - a n d d ea th -p o llu tio n , by contrast, m erely cause those
m ost affected to lie low for a while.
T h e a p p ro p ria te context for beliefs o f this kind ab o u t
m u rd er-p o llu tio n is surely a society th a t lacks m ore form al legal
in stitu tio n s. T h ey express a n d focus concerns th a t can n o t be
d isch arg ed th ro u g h fixed channels o f procedure: if O restes had
been tak en in ch arge by a policem an, there w ould have been no
need for the Erinyes. (In this m odified sense, there is tru th in the
often ex pressed view th at m u rd er-p o llu tio n is too ‘p rim itive’ a
belief to be a n in novation of the seventh century.) As a result,
m an y aspects o f th e in stitu tio n ’s original workings m ust rem ain
o bscu re. B ut one d etail th a t we can point to w ith some plausi­
bility, b ecau se it survived w ith various ram ifications into the
age o f th e o ra to rs, is the p ro clam atio n against th e killer.81 As
D ra c o ’s law takes fam iliarity w ith it for gran ted , it is evidently
very an c ie n t. In th e historical period, the proclam ation by the
v ic tim ’s k insm en was su p p lem en ted by one by the archon
basileus, w h ich form ally excluded the killer, in the period before
trial, from ‘lu stral w ater, libations, m ixing bowls, shrines,
a g o ra ’.82 I t is scarcely rash to infer th a t the original relatives’
p ro c la m a tio n w as to the sam e effect. O ed ip u s’ proclam ation
a g a in st L a iu s’ killer in Sophocles probably gives a fair im pres­
sion: ‘I forbid an y o n e in th e land . . . to receive or address the
m an , or a d m it him to prayers to the gods or sacrifices, o r give
him lu stral w ater; b u t let all th ru st him from the house.’83 It
seem s th a t p ro clam atio n s o f this kind were often respected in
th e early period, because th ere is no real trace in legend of
th e kind o f blood feud fam iliar from m any non-centralized
societies.84 In ste a d o f rem aining w ith his kinsm en to fight it out,
o r seeking refuge w ith a pow erful lord in his own land, the killer
p ersu ad es th e v ictim ’s relatives to accept blood-m oney, o r flees
81 / 6 ' I3 104 (M /L 86) 20, M acD ow ell, Homicide, 23—6; on the origins see L atte. Mord,
283 f. = Kl. Sehr. 386; G ern et, Anthropologie, 2 2 7 -9 .
82 D em . 20.158.
83 S oph. Ö T 2 3 6 -4 1 ; cf. H d t. 3 .5 0 -5 3 .
84 T h is p o in t, crucially im p o rtan t (and constantly neglected) in relation to the
su p p o sed H o m eric indifference to pollution, is em phasized by Nilsson, G F 99 n.l,
W ilam o w itz, Kleine Schriften, 5.1, Berlin, 1937, 120. Clontrast the protection th at for
o th e r offences could be sought w ithin the sam e com m unity, H om . Od. 16.424-30.
126 Miasma
to a n o th e r c o u n try w here he is purified an d starts life anew . T h e
a d v a n ta g e s o f su ch a convention, w hich saved the G reeks from
th e ravages o f feud, are obvious, b u t its im aginative vehicle
m u st h av e been the ‘p o llu tio n ’ o f the killer, w hich d eb a rre d his
co u n try m e n , how ever sym pathetically disposed, from shelter­
in g th e sh e d d e r o f a ‘fellow trib e sm a n ’s’ blood.85
I f th e p ro p e r place for a belief in m urder-pollution is in a
society w ith o u t courts, we w ould expect it to w ither aw ay or
c h a n g e in m ean in g once courts are established. T h is would not
necessarily h a p p e n im m ediately, because tim e would be
n e e d e d for the courts to en tren ch them selves a n d win recogni­
tion as a satisfactory form o f procedure. Pollution tem porarily
a c q u ire s a new function, as a th re a t directed by the origi­
nal avengers a g a in st the su rro g ate avengers, the ju ro rs, and
th ro u g h th em a g a in st the city th a t they represented. W e see this
m o st clearly in the Eumenides, w here the Erinyes, defeated in the
first o f all m u rd e r trials, at once propose to tu rn their m alice
a g a in s t th e w hole city by w hich O restes has been acquitted.
(N o rm a l p rosecutors, like those o f the Tetralogies, can merely
w a rn o f th e d an g e rs o f pollution, b u t because of their dual
n a tu re as pro secu to rs a n d an im a te pollution the Erinyes can
also inflict it.) T h e n u m erous a n d distinctive oath s sworn at
h o m icid e trials seem to have been in tended to transfer responsi­
b ility for a false decision from the ju ro rs to the perjured
p a rtic ip a n ts .86 A fter A eschylus a n d A ntiphon, however, the
d a n g e rs o f p ollution seem to recede. Even w ithin ‘A n tip h o n ’,
th e re is a no ticeab le co n tra st betw een the Tetralogies, w here the
a rg u m e n t from p ollution recurs w ith obsessive regularity, and
its m erely in te rm itte n t presence in the forensic speeches. T he
sp ee ch Against the Stepmother (ad m itted ly a w eak case probably
u n d e rta k e n only in obedience to the d ead m a n ’s order) contains
no reference to the tem ples she pollutes, no a tte m p t to trace the
w o rk in g o f th e divine curse in h er life after the m urder, no
th re a ts o f divine vengeance against the prosecutor’s conniving
h alf-b ro th ers; a lth o u g h the crim e itself is repeatedly spoken of
85 F or the concept o f killing a n emphylos (adm ittedly designating, in some cases,
kin sm an ra th e r th a n tribesm an) see Horn. Od. 15. 273, Hes. fr. 190.2, Pind. Pyth. 2.32,
PI. Resp. 565e, Leg. 871a, E p horus 70 FG rH fr. 100, P. Oxy. 1241, col. 3, 28 if., T heophr.
ap. P o rp h . Abst. 2.27, Paus. 2.20.2.
86 M acD ow ell, Homicide, 90—100. O n the m otivation see Aeschin. 2.87 f.
The Shedding o f Blood 127
as a n im p iety , the only suggestion th a t its consequences m ay be
su p e rn a tu ra l is th e final sentence, ‘I think the gods below too
ca re for those w ho have been w ro n g ed .’ T h e defen d an t in the
H e ro d es case does indeed ad v an ce the celebrated arg u m en t
from safe co n tac t as a p ro o f o f his innocence (8 1 -4 ), but,
th o u g h he rem in d s the ju ro rs th a t they have often in th e past
com e to re g ret cap ital sentences (6 9 - 71,91), and speaks o f such
a false v e rd ic t as ‘n o t ju s t a m istake b u t also an im p iety ’ (88,91,
92), he does n o t claim th a t they have suffered as a consequence,
or will necessarily do so if they go w rong in the presen t case. O n
th e c o n tra ry , w hen a m an is u n justly executed, ‘along w ith his
bod y his ho p e o f revenge dies too.’ H is friends will n o t care to
av en g e him ; even if they do, ‘w h a t good will th a t do him once
h e ’s d e a d ? ’ (95). T h e d efen d an t in th e speech On the Choreutês
p o in ts o u t to the ju ro rs the solem nity o f th eir charge (3—6), b u t
does n o t ad v a n ce fu rth er tow ards a th rea t th an the rem ark th at
a co rrec t v erd ic t is desirab le ‘p rin cip ally for the sake o f the gods
a n d piety, b u t also for your ow n sakes’ (3). A rgum ents fam iliar
from th e Tetralogies recur, b u t in the m ost m u ted tones. An
u n ju s t a c q u itta l is ‘less religiously offensive’ (kosiôteron) th an an
u n ju st co n d em n atio n ; in the Tetralogies, it is a question o f which
w ay th e av en g in g spirits will tu rn .87 T h e ju ry can n o t ‘transfer
th e re sp o n sib ility ’ for an u n ju st decision upon anybody else;
h e re too, in th e Tetralogies, it w ould be pollution o r spirits th at
th e ju ro rs could not ev ade.88
In o th e r o rato rs, su p e rn a tu ra l th reats have receded even
from th e position they occupy in the forensic A ntiphon.
G o rg ia s’ Palamedes contains only the faintest h in t th a t an unjust
c o n d e m n a tio n m ig h t be a source o f d an g er to the assem bled
G reeks; th e consequences on w hich the o ra to r really insists are
th o se o f rem orse a n d everlasting disgrace.89 T h e idea th at the
gods are w a tc h in g the ju ro rs as they vote is not extinguished in
fo u rth -c e n tu ry o rato ry — it occurs p articu larly in cases o f im ­
piety — b u t it has settled dow n as no m ore th an one arg u m en t
87 5.91; Tetr. 3/3 8. X en. Hell. 1.7.19 m akes an unjust capital condem nation ‘a great
offence ag ain st th e gods’ w ith o u t explicit m ention o f pollution.
88 6.6, cf. 5.89; Tetr. 3/38.
89 G o rg ias B 1 1 .3 4 -6 . T h e ‘im pious d eed ’ they will have on th eir consciences perhaps
h in ts a t d an g er.
128 M iasma
a m o n g m a n y .90 E ven in religious cases, the injustice of the deed
is often em p h asized m ore th a n the im piety.91 T h e first speech of
L ysias, a defence in a case o f ju stified killing, is quite free from
th e la n g u ag e o f pollution, a n d it ap p ears only fleetingly even in
th e p ro se cu tio n o f E rato sth en es.92 A com parison is difficult
b ec au se o f the a c cid en t93 th a t has preserved for us three m urder
sp eech es o f A n tip h o n b u t none o f later date except for Lysias I,
w h ich is a defence. B ut it is reasonable to suppose that, in a
fo u rth -c e n tu ry p rosecution, m u rd e r w ould have been presented
as a th re a t to society on a secular far m ore th an on a religious
level. T h is sec u larizatio n p ro b ab ly has com plex causes, bu t it is
te m p tin g to suggest as one o f them th a t m urder-pollution had
o u tliv e d its utility. T h e prom inence o f pollution in the Laws is
c h a ra c te ris tic o f th a t w ork’s profound religious conservatism .
T h e a p p ro a c h ad o p ted here p u ts no em phasis on fear and
h o rro r. T h e p o llu te d m u rd erer is by definition dangerous, but
th is does not m ean th a t fear w as the origin o f the belief, nor even
th a t, p ro v id ed the p ro p e r procedures were followed, the danger
p re se n te d by the killer wras an y m ore a source o f anxiety than
th e high-voltage cables th a t ru n through our cities. T h e idiom
w as, o f course, well suited to express any fear or h o rro r that
m ig h t a c tu a lly be felt, as in the case o f O edipus, b u t th at does
n o t tell us a n y th in g ab o u t the origin o f the belief.
O n th e o th e r h an d , since the d octrine o f pollution does
p o stu la te in ten se dangers, it w ould alw ays be possible for an
in d iv id u a l o r a com m unity to w orry w h eth er the custom ary
p ro c e d u re s w ere a d e q u a te in o rd e r to cope w ith them . T h e
p o llu te d m u rd e re r lurking u n d etected could becom e a source of
im a g in a tiv e terro r. T h e re is obviously a question here about
in tim a te feelings th a t we are scarcely equipped to answ er, but
su c h evidence as is available suggests th a t intense anxiety was
n o t th e norm . I f we consider the different sets o f people in­
volved, it is the killer him self w hose peril is m ost frequently
90 (L ys.) 6 . 13 μή βούλεσθε εις υμάς τήν αιτίαν ταύτην περιτρέψαι, Dem . 19. 220 (cf. 239)
μ ή ■■■ύμείς τή ν ά ρ ά ν κ α ϊ τήνέπιορκίανοίκαό'Ιείσενέγκησθε, Dem . 29.4, ? 43.84, 59.109 (cf.
126), L ycurg. Leocr. 146 (these last two texts both claim th at public responsibility
before the gods for a n in d iv id u al’s m isdeeds only begins once they are brought to trial),
A eschin. 2.87 f.
91 e.g. L ysias 30.
92 12.99.
93 K .J . D over, Lysias and the Corpus Lysiacum, Berkeley and Los Angeles, 1968, 6.
The Shedding o f Blood 129
m en tio n ed . T h e m u rd e re r goes m ad , an d not in th e elaborate
m ythological histories o f O restes a n d A lcm aeon alone; in the
Hippolytus, th e n u rse reacts to P h a e d ra ’s d eran g em en t by asking
‘Are y o u r h a n d s clean o f blood, child?’ and A m p h itry o n in the
Heracles su p p o ses for a m om ent, rem arkably, th at the hero has
been d riv en m a d even by his ju stified revenge ag ain st Lycus.
The sam e belief is still attested in fourth -cen tu ry texts.94 T h e
killer o f a p a re n t, according to a ‘doctrine o f priests o f o ld ’
reco rd ed by P lato, is surely destin ed him self to perish a t the
h an d s o f a child, in this in carn atio n or a n o th e r.95 F or the th rea t
to th e v ictim ’s kin, should they fail to seek revenge, the m ost
e lo q u en t testim o n y is A pollo’s oracle to O restes in the Choephori,
w hich m en tio n s cancerous diseases, leprosy, an d m adness.96
W e d o n o t h ear, how ever, o f any defaulting avenger, m ythologi­
cal o r h istorical, w ho was actu ally believed to have been
afflicted in this w ay. As to the d an g ers undergone by those who
asso ciate w ith th e killer, A n tip h o n states th at they are dem on­
stra te d by n u m ero u s instances, w hile O restes too, in Aeschylus,
can p o in t to his ‘h arm less asso ciatio n ’ w ith m any households as
p ro o f o f his p u rity .97 B ut it is in terestin g th at the only specific
risk w hich A n tip h o n refers to is th at o f shipwreck. T h ere is no
su gg estio n th a t disease or m adness is contracted by contact
w ith a m u rd e re r and, th o u g h th e possibility is envisaged in
tra g e d y ,98 no exem plary m ythological tale is based upon it.
W h en in m y th a p u rification proves ineffective, this is revealed
th ro u g h th e k iller’s renew ed m adness and not the affliction of
his associates. X en o p h o n offers a purely secular version o f the
belief: ‘So far have m en gone in th eir precautions ag ain st m u r­
d e r th a t m any h ave m ade a law th a t not even the m an who
asso ciates w ith th e m u rd erer should be p u re .’99 T h ere is,
finally, th e d a n g e r to the com m unity at large. T h is ap p ears to
be excellently attested ; the pro secu to r in the first tetralogy
94 M u rd e r m adness: sec A ppendix 7 on the m yths o f A lcm aeon, H eracles and
Ip h itu s, Ixion, a n d O restes; Aesch. Cho. 1055 f , E ur. Hipp. 316, H F 966 f , Or. 339, Xen.
Cyr. 8.7.18, PI. Leg. 865d—e, P lut. Cim. 6.4. But m urder-m adness is a com m on belie lin
societies th a t d o not talk o f m urder-pollution.
95 PI. Leg. 872c.
96 2 7 8 -9 6 .
97 A n t. 5. 82, A esch. Eum. 285. In Ael. VH 8.5 blood-guilt causes co n trary winds.
98 E u r. Or. 793.
99 Hiero 4.4.
130 M iasma
w a rn s the ju ro rs th a t unavenged m u rd ers lead to crop-failure,
w hile th e Oedipus Tyrannus opens w ith all n atu re o ut o fjo in t as a
c o n se q u en ce o f th e d e a th o f L aius. Specific instances, however,
p ro v e su rp risin g ly h a rd to discover. In m yth, w hen plague
follows u p o n the m u rd e r o f a n individual, the victim is norm ally
som eone especially d e a r to the gods (priest, prophet, or poet),
th e g u ilty p a rty no t a p riv ate citizen b u t the whole com m unity,
a n d th e p u rp o rt o f the story aitiological. H istorically, we do not
find afflicted states in stitu tin g h u n ts for the m urderers in their
m id st; the com m onest religious explanation for public disaster
is sacrilege, a n d the only kind o f killing th a t seem s to be
id en tified as a cause is the collective m assacre, w ith num erous
v ictim s a n d co m m u n al responsibility, w hich was already a
so u rc e o f sca n d al before the affliction o cc u rre d .100 Even in
S ophocles, p lag u e w ould not p erh ap s have bitten so deep had
n o t b o th victim a n d killer been kings, and one the father o f the
o th e r. I t seem s th a t the a u th o r o f the Tetralogies has taken the
d o c trin e o f p o llu tio n to a theoretical extrem e som e way beyond
th e level o f u n ease th a t in p ractice it created.
W e tu rn in conclusion to the factor th a t has long bedevilled
d iscu ssio n o f this issue, the ‘silence o f H o m er’. It has delib­
e ra te ly b een p o stp o n ed to the end, so th at readers w ith little
ta s te for sp ec u la tio n ab o u t the unknow able can pass straight on
to th e follow ing ch ap ter.
H o m e r’s silence was first noted in an tiq u ity . ‘W e d o n ’t find
th e killer bein g purified in H om er, b u t either going into exile or
b ein g killed in tu rn (or, paying com p en satio n )’, says one
sch o liast, a n d an o th er, detectin g a t one point a n allusion to
p u rific a tio n , co m m ents ‘p erh ap s an anachronism , like “ the
tru m p e t s o u n d e d .” ,|01 It should be em phasized th a t these two
texts, a lth o u g h often taken to in d icate th a t H om eric m a n ’s
a ttitu d e to hom icide w as relaxed, ad m it the opposite interpre­
ta tio n j u s t as readily. T h e H om eric killer cannot m erely be
p u rified , b u t m u st flee instead. It is interesting to contrast the
first securely a tte ste d purification, th a t of Achilles in the
100 See pp. 273 il', below.
101 Schol. T . //. 11.690 (the am biguous w ord is άντιτίνοντα); schol. T . II. 24.480.
The Shedding o f Blood 131
Aethiopis (a p oem p erh ap s o f th e m id-seventh c e n tu ry ).102 H av ­
ing, u n d e r p ro v ocation, slain T h ersites, A chilles sailed to
Lesbos (a te m p o ra ry sym bolic exile), sacrificed, was purified,
a n d rejo in ed th e G reek arm y. W e see here, som e say, the first
im p rin t o f th e novel d o ctrin e o f pollution: Achilles w ould not
h ith e rto h av e been incom m oded for p u ttin g dow n such a low
fellow. T h e a lte rn a tiv e to p u rificatio n for Achilles, however,
m ig h t well h av e been p e rm a n e n t exile, h o u n d ed o u t by
T hersites’ cousin, D iom ede. It has sim ilarly been suggested
th a t th e A th e n ia n s first established th eir exëgêtai pythochrêstoi to
m ake possible th e re tu rn o f the tain ted A lcm aeonids. ‘You were
all too lax ’, exclaim s O vid o f th e ancients, ‘in thinking th at the
g rim crim es o f bloodshed can be w ash ed aw ay in river w a te r.’103
P u rificatio n is n o t m entioned in H om er; th e custom ary re­
spon ses to hom icide th a t a p p e a r in th e poem s, how ever, are
q u ite reco n cilab le w ith th e in stitu tio n we have p o stu lated , and
m ay even be tak en to p resuppose it.104 T h ere is no possibility for
th e H o m eric killer, any m ore th a n for the ‘p o llu ted ’ killer o f
classical tim es, o f finding refuge w ithin his own country. H e
m ay, it is tru e, be ab le to p ersu ad e th e victim ’s kin to accept
c o m p e n sa tio n (this was p ro b ab ly p articu larly com m on in cases
o f a c c id e n ta l k illin g )105 but, as the A frican evidence shows, the
p a y m e n t o f blood-m oney is not irreconcilable w ith a doctrine o f
p ollu tio n . N o ab so lu te m oral revulsion is felt ag ain st deliberate
killing, w h ich m ay even be a subject for b o astin g ,106 b u t this
102 O C T H o m e r v, p. 105. 28 1Γ. The arg u m en ts advanced for d atin g A rctinus (W.
S c h m id /O . S täh lin , Geschichte der Griechischen Literatur, i.i, M unich, 1929, 211 f., G. L.
H uxley , Greek Epic Poetry, L ondon, 1969, 144) are fragile. O th e r purifications ascribed
by la te sources to early poets (adduced by L loyd-Jones, 73) m ust be treated with
c a u tio n (C alh o u n , 26—9). W hereas in Hes. Scut. 13 A m phitryon m erely ‘supplicates’
the T h e b a n s, in A pollod. 2.4.6. a n d hyp. D, E to Hes. Scut. (pp. 2 6 9 -7 1 , Rzach, ed.
maior) he is purified there. W e can n o t therefore be confident th at Procris w as really
‘p u rifie d ’ a t T h eb es in the Epigoni (O C T H o m er v, p. 115, fr. ii). O n the supposedly
H esio d ic sto ry in Schol. D. II. 2.336 see p. 382 below. But the purification ascribed by
P roclus to th e Aethiöpis is unlikely to be a late accretion, as sub seq u en t accounts o f
T hersites’ d e a th ignore it (R E s.v. Thersites, 2 4 6 1 -3 ). Stengel, 157 claim s th at the
p u rificatio n is o f early a n d untypical form because sacrifice precedes purification; but
cf. D em . 23.72, ? L S S 115 B 58.
103 Exegetes: Ja c o b y , 40 f., 272 n. 225. O v. Fast. 2.45 f.
104 F o r th e evidence see B o n n er/S m ith , i. 15-22.
105 Cf. H aslu ck , 239 f. B ut the definition o f the ‘accid en tal’ can depend as m uch on
the m utual disposition o f the two kin groups as the facts of the case, Black-M ichaud, 19 f.
106 e.g. Od. 13. 258 11'.; cf. H asluck, 228, a n d for public indifference C am pbell, 201.
132 Miasma
could scarcely be looked for in a society w ithout centralized
a u th o rity , w here the th re a t o f violence is the ind iv id u al’s only
final p ro te c tio n a g a in st en c ro ach m en t by his neighbour. In the
classical period, by co n trast, killing is m uch closer to being the
sam e ab so lu tely horrific ac t th a t it is today. It is ‘im pious’, a
‘p u b lic offence’, ‘am on g the w orst o f crim es’.107 ‘M an -slay er’ is,
like ‘te m p le -ro b b e r’, a term o f everyday abuse, an d orators
co n co ct m u rd e ro u s plots in w hich they claim th eir opponents to
h a v e been in v o lv ed .108 T h is change in a ttitu d e is obviously due
to th e d ev e lo p m e n t o f altern ativ e institutions through which
th e in d iv id u a l can vindicate his rights. A rm s are no longer
w o rn , a n d the only m otives for killing th a t can rem ain are
sh am efu l. B u t it is again clear from the eth nographic evidence
th a t p o llu tio n m ay a n te d a te m oral revulsion against killing.
T h e d iso rd e r th a t it expresses is not m oral b u t social, a d istu rb ­
a n c e o f the eq u ilib riu m betw een two family groups. It m ay
becom e a vehicle for m oral revulsion once this is felt, b u t this is a
k in d o f re a p p lic a tio n .109 Even in H om er, how ever, as in most
n o n -c e n tra liz e d societies th a t condone honest killing,110 there
exists a special category o f sham eful killings th a t are fiercely
c o n d e m n e d . (M o st o f the m u rd ers o f m odern society w ould fall
in to this category.) K illing ‘by ste a lth ’, later condem ned in all
c irc u m sta n ces, is still adm issible in defence o fh o n o u r, bu t it is
c le a r from th e case o f A egisthus th a t killing for m aterial and
sex u al g ain invites divine punishm ent; we see the gods
107 S ee e.g. A nt. Tetr. 3 a 2, A nt. 5.10, D em . 21.45. A m ong the W est L ocrians, by the
ea rly 5th cen tu ry , the killer an d his genea w ere exiled in perpetuity, and his house
d estro y ed ( M /L 13. 12—14); a t A thens the property o f the d eliberate killer was
confiscated, in a p p a re n t co n trast to the H om eric practice (Od. 13. 258 f.) - a penalty-
reserved for serious crim es against the com m unity.
108 M an-killer: M en. Dysc. 481 w ith S a n d b a c h ’s note, and note PI. Euthphr. Id
Im p u ta tio n s o f m u rd er: Lys. 1 0 .1 ,2 6 .8 -1 3 , Isae. 8.41, 9.17, Isoc. 18.52, D em 21. 104,
2 2 .2 ,5 9 .9 , A eschin. 1.172,2.148.
It is te m p tin g to suppose th at in G reece a pollution originally confined to killer
a n d victim ’s kin becam e extended for this reason to the whole com m unity (so
D u rk h eim in his review o f G lotz, L ’Année sociologique 8 (1 9 0 3 -4 ), 469), particularly
th ro u g h the in stitu tio n o f courts; b ut the killer’s exile suggests th at he was always
g en erally p olluted. F o r th e sam e reason it is unsatisfactory to suppose th at pollution
o rig in a lly a tta c h e d only to sham eful killers (miaiphonoi).
110 See e.g. H asluck, 2 4 4 -5 ; P. P. Howell, A Manual o f Nuer Law, O xford, 1954,40,42,
55; P o llo c k /M a itla n d , i.52, ii.458n. 1 ,4 8 6 on non-em endable offences; B lack-M ichaud,
117 f.
The Shedding o f Blood 133
them selves d iscussing it.111 K in-killing, strongly condem ned by
public o p in io n , will surely n o t have escaped the notice of the
E rinyes, a n d th e th rea t o f divine an g e r is a powerful d eterren t
from g u est-k illin g .112 It is very revealing th a t legalistic
s tra ta g e m s, ra th e r like th a t o f C reo n in the Antigone, are em ­
ployed to av oid th e literal ta in t o f these kinds of m urder. T he
offensive p erso n is m arooned, or d isp atch ed ab ro ad to be killed
by a stra n g e r, or sent o u t to face im possible d angers in the
h u n t.113 T h e s u p e rn a tu ra l d an g ers th a t are a p p a re n t in these
cases are not m entioned in connection w ith o rd in ary killings,
becau se th e ch ief responsibility for achieving revenge lies not
w ith th e gods b u t w ith th e v ictim ’s kinsm en. F ear o f disgrace is
(he c h ief p ressu re th a t drives th e H om eric kin to seek revenge,
a n d in som e m o d ern M e d iterran ea n feuding societies it is the
only o n e .114 (T h e re is therefore no reason to see the H om eric
p ictu re as a n idealized aristo cratic rendering of an institution
w hose real basis is th e p e a sa n t’s fear o f g h o sts.115 T h ere is less
eviden ce for fear o f the d ead in H esiod than in H om er.) But in
H o m er the d ead can intervene to nudge the living and rem ind
th e m o f th e ir du ties; m a ltre a tm e n t o f a corpse provokes divine
revenge, a n d the E rinyes ensure th at each m em ber of a family
S tealth : Od. 13. 258 ff.,//. 7. 142-6, co n trast Soph. Tr. 2 7 4 -9 , Pind. Pyth. 2.32. It
w as p articu larly after όολοκτασίαι th a t the killer sought to p rotect him self by ‘spitting
o u t’ the blood, Etyrn. M a g n p. 118. 31—6 citing Aesch. fr. 354 an d Ap. Rhod. 4.479.
A egisthus: Od. 1. 3 5 -4 7 ; tor possibly violent public response to a sham eful killing cf. Od.
16. 3 7 6 -8 2 .
1,2 E special in h ib itio n s ag ain st kin-killing: //. 9 .4 6 1 , Od. 10.441. T h e Erinyes uphold
rig h ts, w h eth er o f p a re n ts, eld er b rothers, o r beggars ( Lloyd-Jones« 75), living o r dead
( Öd. 1 1. 280). G uest-killing: Od. 14.406, 21. 28, II. 24. 5 8 3 -6 (cf. Kur. Hec. 2 5 - 7,714 (f.,
789 fr.)
113 S oph. Ant. 775 f., Od. 3. 2 6 7 -7 2 , II. 6. 155 (Γ., w ith T zetzes on Lycophron 17. An
e r ra n t sen t aw ay to be killed: A pollod. 1. 8. 4 ( = H esiod, fr. 12, Periboea); given away
for killing: A pollod. 3. 9. 1 (Auge); exposed to m ortal dangers: Apollod. 3. 13. 3 (Peleus).
F or m aro o n in g cf. the Philoctetes m yth, E u r. Hec. 1284-6, Paroem. Gr. Coislin., p. 123
G aisfo rd s.v. Ά νά γνρ ο ς (testim onium to A ristophanes Ά νάγνρος).
1.4 Od. 24. 433—6. Cf. e.g. H asluck, 219—260, C am pbell, 193—203, Black-M ichaud,
passim.
1.5 So e.g. Stengel, 156: ‘u nd wo ritterliche A delsgeschlechter herrschen, trotzig ihrer
K raft v e rtrau e n d , wie die hom erischen basileis, findet d er ängstliche G lau b e der
nied rig en B ürger schw er E ingang’. For em phasis on the ‘p easan t’ basis o f Hom eric
society cf. H . S tra sb u rg er, Gymnasium 60 (1953), 97—114; P. A. L. G reenhalgh, Historia
21 ( 1972), 532 f.; P. W alcot, Greek Peasants Ancient and Modern, M anchester, 1970, 16—19.
134 Miasma
p ay s to the o th ers th eir d u e .116 It seems inevitable th at the
v ic tim ’s kin w ould have been exposed to su p ern atu ral danger as
w ell as p u b lic scorn if they failed to seek revenge.
T h u s, o f th e b u n d le o f p h en o m en a th at constitute, or are
ex p lain e d by, pollution in the classical texts we find in H om er
th e k iller’s exile, divine an g er provoked by p a rtic u la r forms of
killing, a n d th e p o ten tial at least for ghostly sanctions against
in activ e kin. T h e actu al m e ta p h o r o f pollution is absent, but
th e re exists an e p ith e t miaiphonos (it is applied to Ares) which
m ean s, p re su m a b ly , ‘one w ho kills in a polluting w ay’ an d in
la te r texts is ap p lied to the m ost culpable m u rd e re rs.117 T he
c e le b ra te d silence, therefore, reduces itself alm ost entirely to
th e m a tte r o f th e actu al rite o f purification. O f the exiled killer,
H o m e r says m erely th a t he ‘m akes supplication to ’ a powerful
p rin ce , w ith o u t m ention o f p u rifica tio n .118 Even in later texts,
how ever, th e re q u est for p urification ap p ears m erely as a sub­
division o r special aspect o f supplication. Zeus is god o f the one
bec au se he is god o f the other, an d in ritual rules from both
A th e n s a n d C y ren e the killer seeking purification is a ‘sup­
p lia n t’. 119 H is m ost pressing re q u irem en t is for a hom e in which
to s ta rt life anew . By consenting to purify him , the foreign lord
ac c e p ts the o bligation to provide one, and this is the source of
th e p urified m a n ’s strong indebtedness tow ards h im .120 But
th is, the really im p o rta n t service, is already provided by the
H o m e ric lo rd w ho ‘receives’ a hom icidal ‘su p p lia n t’. I f the
a c tu a l rites o f p urification w ere in troduced in post-H om eric
tim es —a n im p o rta tio n from L ydia, p e rh a p s121 —the im portance
of this in n o v atio n was slight. It is h ard to accept, how ever, th at
su ch rites w ere a com plete novelty, though m odification and
ex ten sio n in th e ir app licatio n there m ay well have been. T hey
a re no t p ro d u c ts o f the sam e kinds o f anxiety as the com pulsive
116 11. 23.65—107, 22.358; lor the E rinyes see L loyd-Joncs, 75. For the dead m a n ’s
claim s see E. B ruck, Totenteil und Seelgerät im griechischen Recht, M unich, 1926, 27-34,
R o h d e, 38, G lotz, 5 9 - 7 6 , above all II. 24. 5 9 2 -5 .
' 17 LSJ s.v. μιαιφόνος
1,8 e .g .//. 16.574.
' 19 356 F G rH fr. 1; L S S 115 B 5011'.
120 t o r this see H d t. 1.44, E ur. Stheneboea, prologue 22—5, p. 44 v. A rnim , Apollod.
3.13.3, a n d e contrario Ix io n ’s crim e ag ain st Zeus.
121 G . G ro te, History o f Greece2, L ondon, 1883, i.25, citing H dt. 1.35 - but sceptics
m ig h t reg ard H d t.’s rem ark as a story-teller’s im provisation.
The Shedding o f Blood 135
w ash in g s o f th e p a tie n ts o f F reu d , b u t cerem onial expressions,
exp lo itin g co n crete sym bolism , o f social realities. Bloodshed
has cau sed th e killer’s exclusion from society an d , to perm it his
read m issio n , th a t blood m u st be w ashed away. T h ese rites are
perfo rm ed in th e classical period not by vag ab o n d priests but
by h ig h -b o rn rep resen tativ es o f th e com m unity; a N estor would
d o u b tless h av e been h ap p y to preside over such a cerem ony.122
P u rificatio n o f th e su p p lian t in his new hom e ab ro ad is deeply
e m b e d d e d in m ythology an d , in the autonom ous prince who
ac ts as p u rifier, presupposes a figure who was becom ing extinct
in th e a rc h a ic p e rio d .123 I f H o m er h ad been lost, indeed, and
only th e m ythological evidence survived, no one w ould have
d o u b te d for a m o m en t th a t these rites w ere prim eval. It is
te m p tin g to revive the unfashionable view th at, in ‘su p p licat­
in g ’ a foreign lord, the H o m eric killer im plicitly requests p u ri­
fication; th e ac tu a l ritual is o m itted by the poet, not by the
society th e poet d esc rib es.124 T h e a u th o r o f the H esiodic Shield,
w h o ce rtain ly lived in a period w hen purifications were
perfo rm ed , w as co n ten t to describe a killer’s arrival in a new
c o u n try in term s o f su p p lic a tio n .125 T h e passages th a t have
b een q u o te d to prove th a t H o m er cannot have been fam iliar
w ith th ese p ractices are q u ite inconclusive.126
122 See A p p en d ix 6. D odds designates the ritu als 'elab o rate a n d m essy’ (36); were
H o m eric ritu a ls a s a ru le an y th in g else?
123 See A p p en d ix 7. T h e m yth o f Ixion is based upon the arch aic institution ol
b rid e -p ric e (D io d . 4 .4 9 .3 ; on b rid e-p riced '. A. M. S n o d g ra ss,JH S 9 4 (1 9 7 4 ), 114-25).
124 K . O . M u ller, Aeschylos Eumeniden mit erläuternden Abhandlungen, G öttingen, 1833,
137. F or the su b seq u en t d eb ate cf. references in G lotz, 228 n. 3, B onner/S m ith, 16 n. 1,
C a lh o u n , 16 ii.2. N ilsson, G G R 91 f., L loyd-Jones, 83 revive M u ller’s position, without
c itin g him . M ü lle r’s positive arg u m en ts for the presence o f m u rd er purification were
fallacious. H e o b serv ed (134 n. 10) th a t schol. T . II. 24.480 ap p aren tly read άνάρύς ές
ά γνίτov; it notes τάν όέ καθαίροντα καί άγνίτην ελεγον. But thé trad itio n m akes excellent
sense, as su p p lica tio n w as m ade άνόρός ές άφνειοϋ (see A ppendix 7), an d m urder
p u rificatio n w as not perform ed by a specialized ‘purifier’; άγνίτης, w hich first appears
in L yco p h ro n 135, is a form ation o f a kind very com m on in technical and poetical
H ellen istic G reek, ra th e r rare in H om er: cf. G. R edard, Les Noms grecs en -ΤΗΣ, -ΤΙΣ,
P aris, 1949, 1 1 0 -1 5 , 260 n. 2. (B ut M ü ller’s reading is accepted by E. Fraenkel,
Geschichte der griechischen Nomina agentis a u f -τήρ, -τωρ, -της. ii, S trasb u rg , 1912, 128 n. 2,
an d defended by W illiger, 4 9 -5 2 .) T h e thambos o f the spectators in II. ‘2 4.482 need imply
n o m o re th a n su rp rise an d curiosity, cf. Od. 7. 144—5. N or does Od. 23.118 1Γ. support
M ü ller, cf. L ipsius, 9 n. 25.
125 Hes. Scut. 13.
126 Od. 15.223 11'., ‘2 2 .4 8 0 -9 4 . As to the first, the sea-shore was no place for a formal
p urificatio n (T h eo cly m en u s does not even perform a formal supplication, co ntrast Od.
-7.133 11'.); as to the second, O dysseus considers the killing o f the suitors justifiable
hom icid e w ith no co m pensation payable (cf. A ppendix 5).
136 Miasma
W e h av e so far discovered no really surprising discontinuity
betw e en H o m e r a n d the fifth century. I t is, however, som etimes
claim ed th a t th o se heroes w hose m onstrous pollution fills the
A ttic stag e a re view ed by H o m er an d o th er early poets w ith a
c e rta in c o m p la isa n c e .127 In the Odyssey, O restes is an exem plary
figure, u n tro u b le d by Erinyes. Still in the Odyssey, O edipus lives
o n as king in T h eb e s after the discovery of his crim es, while in
th e Iliad h e is h o n o u re d w ith funeral gam es like any other hero;
h e p e rh a p s even, in one o f the old T h e b a n epics, m akes a new
m a rria g e . A n o th e r early poem m ay have let A lcm aeon, with his
m o th e r s blood upon him , m arch out against T h eb es at the
h e a d o f the Epigoni. K illing a p a re n t is, it seems, ju s t one o f the
o rd in a ry ups a n d dow ns o f a h e ro ’s career.
S u ch a conclusion becom es paradoxical, if one considers the
c h a ra c te r ol th e m yths them selves; is not their point to im agine
th e u n im a g in a b le ? 128 W e w ould have to suppose th a t the trage­
d ia n s rediscovered in these stories th a t original significance
w h ich the early poets h ad forgotten. It does not seem, however,
th a t H o m er w as unconscious o f the horror o f the events he
a llu d es to, even th o u g h it does no t suit his im m ediate purpose to
em p h asiz e it. H e presen ts O restes as a glorious an d prosperous
figure, as do all the fifth-century poets, in the long term , except
E u rip id es; to do so, however, he finds it necessary to focus
a tte n tio n on th e d e a th o f A egisthus, a n d suppress all allusion to
th e m a tric id e .129 (Som e have even supposed th a t it did not yet
form a p a rt o f th e legend.) H is O ed ip u s, though king, is suffer­
in g all th e p ain s th a t the Erinyes o f a m other can create;130 if
H o m e r knew o f a n y fu rth er m arriage, this detail too he su p p res­
sed. As for lost poem s, we ca n n o t assess their m oral colour, or
h ow they tre a te d these delicate incidents in their heroes’
careers. T h e m ere existence o f a m yth th at allow ed O edipus to
m a rry a g a in c a n n o t properly be used as an arg u m en t, because
127 G lotz, 233 1'., D odds, 36. F or the evidence on these legends see A ppendix 7.
O n O restes d ilem m a see H asluck, 217. The correct response would have joeen lor
M e n e la u s to kill C ly taem n estra ; w as it to p revent this tam e solution th at h<? was sent
w a n d e rin g so long in E gypt?
129 Cf. A. Lesky, R E s.v. Orestes, 968 f.; M . D elcourt, Oreste et Alcm'eon, Paris, 1959, 21,
89; J . G riflin.y/ZV 97 (1977), +4 n. 32; co n trast the clear statem ent ol'H es. fr. 23 a 30.
130 Od. 11. 2 7 5 -8 0 : all here is m ysterious an d dire. For the sufferings cf. Hes. fr. 193.4
π ο λ νκ η ό ίο ς Ο ίδιπόόαο (p ro b ab ly in the context o f his funeral gam es), Ibvcus SL G S
222.5.
The Shedding o f Blood 137
m y th ical persons a ttra c t to them selves stories o f diverse origins
a n d ten d en cies, a n d it is left to th e poets to e x tra ct from them
such co h eren ce as they c a n .131 E ven O e d ip u s’ life as a w an d er­
ing o u tc a st, su blim ely im ag in ed by Sophocles, seem s to have as
its o rig in n o th in g m ore significant th a n A th en ian pretensions
to possess his grave.
B ecause these stories have n o th in g to do w ith w h a t is typical
o r legally ex e m p la ry ,132 th e im aginative response to them of
poets becom es elusive evidence. F or th e fifth-century tragedian,
O re s te s ’ situ atio n has been fu rth e r rem oved from everyday
exp erien ce by th e fact th a t a n O restes o f the day w ould perhaps
h av e so u g h t redress th ro u g h th e c o u rts.133 T o ask w h at tre a t­
m e n t a n ac tu a l O e d ip u s w ould have received is a rath er
fa n ta stic q u estio n , b u t it is clear even from the Oedipus Coloneus
th a t th e issue w ould have been co n troversial;134 th e poet, how­
ever, h as o th e r in terests th a n the precise ritu al statu s of in ­
v o lu n ta ry incestuous parricid es. T o descend to this level, it
is in stru c tiv e to co m p are th e ch o ru s’s horrified response in
A esch y lu s’ Septem to the im p en d in g fratricide w ith P lato’s regu­
latio n s on th e su b ject in the Laws. ‘T h e d e a th like this of two
b ro th ers, one slain by the o th e r - this is a pollution w hich can
nev er grow o ld ’, say the chorus. F o r Plato, fratricide is ‘p u re ’ in
civil strife, a n d req u ires three years exile, ad m itted ly associated
w ith d isso lu tio n o f the family, w hen it occurs th rough anger;
only th e m u rd e r o f a b ro th er in cold blood d em an d s the severest
p e n a ltie s .135
131 C f. M . D elco u rt, Oedipe, ou ία légende du conquérant, Liège, 1944, ix: ‘Il n ’y a pas
d ’O e d ip e prim itif. C e q u i est prim itif, ce so n t les thèm es qui, en s’articu lan t les uns aux
a u tre s so n t d ev en u s d ’ab o rd les gestes d ’O edipe, puis sa vie et enfin son caractère.’
H . Je a n m a ire , Rev. Phil. 213 ( 1948), 163, speaks o f ‘une biographie rom anesque do n t il
é ta it réservé à d e g ran d es artistes de d égager l’élém ent tragique’.
132 In the case o f O restes, L. G ern et, the legal historian, insists on this. Annales 10
(1955), 531.
133 Kur. Or. 5 0 0 -4 . F or the possibility o f prosecuting a kinsm an (denied by Glotz,
437) cf. Ar. fr. 585, PI. Euthphr. 4a, Poll. 8.117.
134 S oph. O C 4 2 7 -4 4 , 76 5 -7 1 (cf. 07*1438 f., ask the god). T h e reception o f O edipus
a t A th en s is sim ilarly conten tio u s, co n trast 225 If. (chorus), 5 5 1 -6 8 (T heseus), 9 4 4 -
50 (C reo n ).
135 A esch. Sept. 681 f., PI. Leg. 868c, 8 6 9 c-d , 8 7 3 a-b . C ases o f wholly accidental
kin-killing P lato u n fo rtu n ately does not consider. S entim ental trad itio n w as eventually
to d eclare th e d a u g h te rs o f Pelias p ure from the accidental killing o f th eir father, see
A p p en d ix 7 s.v. Peliades.
138 Miasma
T h e re is no need therefore to p o stu late a su dden transform a­
tio n in th e eig h th o r seventh centuries. B ut it is w orth consider­
ing som e o f th e ex p lan atio n s th a t have been offered by those
w ho believe in this tran sfo rm atio n , because o f their relevance to
o u r m ain them e. T h e m ost p o p u la r has been the nascent influ­
ence o f D e lp h i, a n d a chronological observation seems to lend
s u p p o rt. D elp h i rose to prom inence in the post-H om eric
p e rio d , exactly w hen, it is claim ed, the need for purification was
first m ak in g itself felt. T h e first attested purification from m u r­
d e r, th a t o f A chilles in the Aethiopis, was preceded by sacrifice to
‘A pollo, A rtem is a n d L eto’.136 T h e ch a rac te r o f D elphi’s influ­
en ce has been defined in various ways. Som e see the essence of
th e D elp h ic d o c trin e as the ab so lu te d eb t of vengeance to the
d e a d m a n ’s soul, a n d detect its expression in D ra co ’s ban on
c o m p e n s a tio n .137 O th e rs stress ra th e r the need for purification
a n d expiation. D id no t Apollo h im self serve A dm etus for a year,
a n d su b m it to com plicated rites after the m u rd er o f the dragon
P ytho? T h e re has even been talk o f ‘the new religion o f expia­
tio n ’ (Sühnereligion), founded by ‘the D elphic c h u rc h ’ in the
e ig h th c e n tu ry .138 Som e ap p eal, w ith o u t specific reference to
D elp h i, to the c h a ra c te r o f Apollo as the god par excellence of
p u rity a n d cleansing, whose p rophets, the archaic M en o f God,
c a rrie d to all corners of G reece their m ission o f healing,
a p p e a se m e n t, a n d p u rifica tio n .139
T h e G reeks, o f course, spoke o f Apollo, the ‘ancestral exe-
g e te ’, w ith im m ense respect, an d w ould not have scorned the
id ea th a t he h a d exercised a civilizing influence upon their lives.
N o n e the less, w hen the word ‘c h u rc h ’ appears in the context of
G re ek religion, it is h ard not to discern behind Apollo an d his
D e lp h ic serv an ts the im age o f p ro p h ets an d priests of a very
d ifferent kind. A pollo, it seem s, in tro d u ced into G reek religion
th a t sp iritu a l a n d m oral elem ent in w hich it h ad been hitherto
so la m e n ta b ly deficient.
136 O C T H o m er v, p. 105.28 Ii.
157 W ilam ow itz, Das Opfer am Grabe3, Berlin, 1907, 14 ff.; Glaube, ii. 36. Sim ilarly
R o h d e, 174 ΙΓ.
158 L. D eu b n er, Neue Jahrb. 43 (1919), 403; m ore cautiously Nilsson, GGR 632—7,
6 4 7 - 5 2 , B urkert, G R 232, an d cf. G lotz, 237.
139 I.. G ern et, Annates 10 (1955), 541. In his com m entary on Laws I X (Paris, 1917,
122) he h a d credited ‘la religion apollinienne' w ith the doctrine o f grad u ated pollution
a c c o rd in g to responsibility. F u rth e r doxography in D eliadas, 12 f.
The Shedding o f Blood 139
A categ o rical den ial o f all D elp h ic influence is o u t of the
q u estio n , w h en so little is know n, b u t it is su rprising how
m eagre th e solid evidence in favour o f it tu rn s o u t to be. It is
unsafe, for in stan ce, to d raw an arg u m e n t from the n a tu re o f the
god him self. A pollo, it is true, is in th e fifth cen tu ry the ‘purifier
o f m e n ’s hou ses’, he who ‘w ashes aw ay ’ evil; from his ep ith et
Thoibos verbs m ean in g ‘purify’ are fo rm ed .140 T h is cleansing
fun ctio n is obviously a n aspect o f A pollo’s healing function, and
is th erefo re likely to be very a n c ie n t.141 W ith m u rd er purifica­
tion, how ever, A pollo has, on the level o f cult, no connection; his
priests d o not perform it, at D elphi or, very probably, anyw here
e lse .142 T h e evidence is extensive th a t the god at whose altars
m u rd e re rs so u g h t purification w as Zeus; he acq u ired this func­
tion, w h ich fell to him n atu rally as god of su p p lian ts, w hen he
perfo rm ed for Ixion th e first o f all such rites, an d he never
su rre n d e re d it to his so n .143 Apollo, by contrast, was a god of
o racles w ho becam e an a u th o rity on m u rd er purification be­
cau se p o llu tio n w as an issue on w hich, like o th er o racu lar
g o d s ,144 he w as repeatedly consulted. W hen he cleansed
O re stes in A eschylus, he was perform ing a task th at would
norm ally h av e fallen to a h u m an purifier. H e felt him self re­
sp o n sib le becau se his ow n oracle h ad enjoined the m atricide,
140 A esch. Eum. 62 f., PI. Cra. 405b, L S J s.v. άφοίβαντος, φοιβαίνω, φοιβάω, φοιβος.
141 See B urkert, G R 232, Rh. Mus. 118 (1975), 19:
142 R. R. D y e r ,J H S 8 9 (1969), 3 8 -5 6 , pointed this out, correcting a general m iscon­
c ep tio n . The conclusion is not w eakened by the fact that, pace D yer, O restes in
A eschylus obviously w as cleansed by A pollo a t D elphi. T h ere is no single cult o f Apollo
to w hich c a th a rtic rites w ere definitely attach ed . T h e ancient tem ple o f A pollo T hearios
a t T ro izen (P au s. 2. 31. 6—9) claim ed to hav e been th e site o f O restes’ purification, but
in o rd er to ex p lain a b an q u etin g custom (cf. the Choes aition a t A thens, E u r. IT 9 4 7 if.),
not a c a th a rtic ritu al. H is cure was elsew here linked with Z eus (Paus. 3. 22. 1) and
A rtem is (P h erecy d . 3 FG rH fr. 135a). H eracles’ purification at A m yclae from the blood
o f Ip h itu s (A pollod. 2. 6. 2) is not necessarily connected w ith the cult o f Apollo there.
A pollo K ath arsio s, sch o lars note w ith surprise, does not exist (R E 10.2519).
143 A esch. Eum. 717 f. (Ixion). For Zeus K ath arsio s see H dt. 1.44.2, A p. Rhod. 4.708
f., ps.-A rist. Mund. 401a 23 f., Pollux 8.142, Cook, ii.ii. 109 7 n .2 , 1100 n.l. F or cults see ?
L S S 65.4, LSA 56.11, Plut. Thes. 12.1 (aition for a cath artic cult o f Zeus M eilichios at the
old b o u n d a ry o f A ttica), Paus. 5.14. 8, schol. E u r. Tro. 90 = 8 4 / r6V //fr. 38. O th e r titles
o f Z eus are o f co u rse also relevant, M eilichios (Paus. 2.20.2), Phyxios (Paus. 3.17.9),
P alam n aio s, H ikesios, A lastoros. See F arnell, i.64—9, J . W . H ew itt, H SCP 19 (1908),
(»1—120, N ilsson, GG R 411—17, M. H . Ja m eso n , B C H 89 (1965), 159—65, C. Rolley,
B C H , ibid., 4 5 4 - 6 . N ilsson’s view th a t Apollo replaces Zeus as god o f expiation (GGR
11 7) se em s to m isconceive th e sep arate relation o f the two to the process.
144 Cf. S E G \ix V 2 7 .
140 Miasma
b u t h e w as n o t th e god to w hom the rites w ere addressed. His
role in th e Aethiopis is isolated, a n d p uzzling.145 T h u s it seems to
be a reversal o f history to suppose th at, once the oracle o f the
p u re A pollo a tta in e d P anhellenic im portance, it inevitably
sp re a d th e d o c trin e th a t m u rd e r dem an d s purification through­
o u t G reece. In this area, it w as the functioning o f the oracle
itse lf th a t m ad e the god into a ‘purifier o fo th er people’s houses’.
A rg u m e n t from the g o d ’s original n atu re is m istaken. It
m ig h t none th e less be the case th a t the D elphic priesthood,
c o n sta n tly confronted by inquiries on ju s t these m atters, was
re sp o n sib le for th e creation (or at least system atization) and
diffusion o f a new doctrine. T h e controversial question of
D e lp h ic teac h in g arises h e re .146 W as D elphi a true fountain-
h e a d o f new w isdom , or a sounding-board th a t amplified
p e rh a p s b u t d id n o t create its clients’ typical religious concep­
tio n s a n d p reoccupations? T h e d octrine of purification is an
excellent test case. P lato in the Laws subm its certain aspects oi
his legislation on this subject to the D elphic god an d his in­
te rp r e te rs .147 W h a t Apollo is req u ired to expound, however, is
th e ritu a l, w hile th e m ore im p o rta n t issues of exile, punishm ent,
a n d p a rd o n P lato h im self determ ines. M oreover, P lato’s arti­
ficial sta te is re lia n t on D elphi because it lacks an cestral trad i­
tions o f its ow n. It is very doubtful to w h at extent Delphi
in flu en ced even the ritu al of historical G reek states; a glance at
th e sacred law s show s th a t they followed divergent local trad i­
tions, no t directives from the centre o f the earth. Even in the
case o f th e g re a t c a th a rtic law o f C yrene, w hich is alm ost unique
a m o n g sac red law s in presen tin g itself as an o ra cu lar response
o f A p o llo ,148 it is generally agreed th a t the actual regulations,
b o th in d ialec t a n d content, w ere form ulated in C yrene itself.
E ith e r th e asc rip tio n to A pollo w as sim ply fictitious, or the laws
h ad been sent ready drafted for the god’s formal approval, which
he, w ith his deep-seated respect for local tradition,149 had no cause
145 Is T h e rs ite s’ scapegoat n a tu re (below , p. 260) relevant? A pollo was god o f the
T h a rg e lia .
146 R eferences to earlier discussion in D efradas, 12 f. O n D efradas’s w ork see
H . Berve, Gnomon 28 (1956), 174-81, L. G ernet, Annales 10 (1955), 5 2 6 -4 2 , H.
J e a n m a ire , R H R 149 (1956), 2 3 1 -5 , P. A m an d ry , Rev. Phil. 303 (1956), 268-82.
147 C f. G . R. M orrow , Plato's Cretan City, P rinceton, 1960, 4 2 3 -7 ; Ja co b y , 13—15.
148 See A p p en d ix 1.
149 X en . Mem. 1 .3 . 1, 4. 3. 16, Isoc. Paneg. 31.
The Shedding o f Blood 141
to refuse. A t A thens, the situation was probably very similar.
Though th e exêgëtai pythochrêstoi have som etim es been seen as
local re p resen ta tiv e s o f th e D elphic god, A pollo’s role was
confined to selectin g these in terp re ters from a list o f candidates;
o nce in office, they will have exp o u n d ed essentially A thenian
lore w ith o u t reference to D e lp h i.150 T h e m ost im p o rtan t p u ri­
fications a t A th en s, those o f ‘su p p lia n ts’, are perform ed by the
K u p atrid exegetes, a college o f obvious an tiq u ity w ho had
n o th in g to d o w ith D e lp h i.151
O ccasio n ally , every state w as forced to look beyond its own
recognized ritu al procedures. W hen plague raged or crops
failed, th ere w as no o th er recourse b u t D elphi. In the Oedipus
Tyrannus, A pollo m akes a long-forgotten crim e the cause of the
city ’s m isfortunes, b u t no single d ep en d ab le historical parallel
can be q u o te d .152 T h o u g h it concerns a different oracle, the
q u estio n p u t by th e D od o n aean s to th eir Zeus is revealing: ‘Is
it b ecau se o f som e m o rtal’s pollution th at we are suffering this
sto rm ?’153 T h e suggestion cam e from the citizens them selves.
A p a rt from a m oralizing story o f obviously post-classical
o rig in ,154 th ere rem ains only the evidence o f m yth. It has re­
p ea te d ly been arg u ed th at ours is a D elphic O re ste ia ,155 which
em b o d ies th e teaching th at killing is som etim es a duty, but
■ilw ays req u ires purification. T h e connection lies n ea r at hand
(it w as m ad e in a n tiq u ity 156) w ith the co u rt o f Apollo
D elphinios a t A thens, w hich tried cases of justified hom icide,
150 Ja c o b y , 30—3. E ven if th eir patria, unlike those of the E u p atrid exegetes, were
sa n ctio n ed by D elphi (Jacoby, 33, 38), the C yrene inscriptions show s how such a
san ctio n is p ro b ab ly to be understood. Even in P lato we can infer a sim ilar procedure.
D espite th e role o f ‘pro p h ecies’ an d ‘the g o d ', the detailed draft legislation in sacred
m a tte rs is th e w ork o f exegetes, priests, nomotketai {Leg. 828a-b, 871 c-d ).
151 See Ja c o b y , loc. cit.; su p p lian ts, 356 FG rH fr. 1.
152 See C h . 9.
153 S E G xix 427. O n th e w ay th at m ost oracles tell their clients w hat they expect or
w ant to h e a r see T h o m a s, 257, w ith references.
154 A clian, 1 7 /3 .4 4 , P arke/W orm ell, nn. 575—6: the young m an who killed his friend
w hile seeking to defend him is pure, w hile he who abandoned him is polluted. For the
m oral to n e o f this cf. e.g. th e ‘oracles’ Anth. Pal. 14.71, 74 (P /W 591—2), below, p. 324.
P /W 339 has a ‘P y th ian purification o f P h o eb u s’, in P/\V 74 A pollo banishes polluted
in q u ire rs from his tem ple.
155 e.g. D efrad as, 160-204. D elphic influence on the m yth is still asserted by
11. H o m m el, Antike und Abendland 20 (1974), 15.
156 ])cm 23.74. But his trial, o f course, was on the Areopagus.
142 Miasma The Shedding o f Blood 143

a n d also w ith the various expiations undergone by the god him ­
self.157 As a vehicle for establishing th a t a category of justified
h o m icid e exists, how ever, a case o f m atricide is unnecessarily
p ro b le m a tic ; ju stifie d hom icide seems not in fact to have re­
q u ire d p u rifica tio n at A thens, a n d the choice o f the D elphinium
as a co u rt need have noth in g to do w ith D elphic doctrine
(A pollo D elphinios is a n older god th a n Apollo of D elphi).158
A g a in st an y a tte m p t to exploit the evidence of m yth, there is an
o b vious objection o f principle: it is not necessary o r even p lau­
sible to su p p o se th a t w henever the D elphic god appears in a
m y th , he ow es his place there to his priests. It is h a rd to see, for
in stan ce , w h a t m otive a devotee would have for ascribing to
A pollo u ltim a te responsibility for O ie ste s’ m atricide. W e seem
r a th e r to be d ea lin g w ith the invention o f a story-teller whose
c h ie f in te re st w as the psychology o f the m ortal. H ow could
O re ste s b rin g h im self to slay his ow n m other? T h e answ er was
obvious: only a t the instance o f a g o d .159
A still m ore hypothetical source of influence, possibly con­
n e c te d w ith D elphi, is C re te .160 F rom C rete cam e Epim enides
to A th en s; to C re te w ent Apollo him self for cleansing from the
blood o f th e d rag o n , to a tow n q u ite obscure in historical
tim e s .161 T h e re are o th er hints too th a t C rete was a land of
a n c ie n t renow n in the arts o f p u rificatio n .162 P erhaps it was
from th ere th a t these rites w ere reintroduced into Greece; an
a v a ila b le ch a n n el w ould have been the C retan priests who,
acco rd in g to trad itio n , served Apollo in the early tim es at
D e lp h i.163 So speculative a reco n stru ctio n can n eith er be re­
futed n o r confirm ed. T h e new need in G reek society th a t en ­
co u rag ed the im p o rtatio n o f the rites w ould anyw ay rem ain to
be identified.
A m o re in tere stin g possibility concerns O rp h ism . O rpheus,
we learn , ta u g h t m en ‘rites a n d to ab stain from m u rd e r’, and
‘m u tu a l s la u g h te r’ w as p ro b ab ly presen ted in O rp h ic poetry as
a ch a ra c te ristic o f m a n ’s b a rb a ric p a s t.164 It seem s alm ost cer­
tain th a t som e connection exists betw een the cen tral im p o rt­
ance o f ‘not killing’ (anim als or m en) in O rp h ism a n d the new
h o rro r o f killing th a t w as developing, as we saw, in a society th at
w as sh ed d in g its arm s. B ut the eccentric religious m ovem ent,
t h o u g h it m ay have focused a n d intensified these attitu d es, can
scarcely h ave created them from n o th in g and then foisted them
on society a t large.
T h is histo rical excursus ends negatively. N othing has
em erg ed to explain the p o st-H o m eric transform ation. But, very
pro b ab ly , th ere was n o th in g to explain.

157 See A pp en d ix 7 s.v. Apollo.

158 O n th e god see now F. G raf, ‘A pollon D elphinios’, M H 36 (1979), 2 -2 2 . His
te m p le is su ita b le as a court because he is a god intim ately associated w ith civic life,
G ra f, 7—13. I can find no evidence for the assertion (W achsm uth, R E 4. 2513, H erter,
R E S u p p l. 13.1092) th a t defendants w hose plea o f justified, hom icide w as adm itted at
th e D elp h in io n w ere th en purified there. O n the ritual statu s o f justified hom icide see
A p p e n d ix 5.
159 Cf. M . D e lc o u rt, L ’Oracle de Delphes, P aris, 1955, 179; Oreste et Alcméon, Paris, 1959,
103—12. F ontenrose, 109 ad m its possible D elphic influence - b ut exerted to publicize
th e o racle, n ot in stru c t the Hellenes.
160 See e.g. L. D eu b n er, NeueJahrb. 43 (1919), 3 9 4 -5 .
161 See P aus. 2.7.7, 2.30.3; 10.6.6—7, 10.7.2, 10.16.5; hypothesis C to Pind. Pyth. (p.4
D ra c h m a n n ); \V. Aly, Der Kretische Apollonkult, T ub in g en , 1908, 49—52.
162 C re ta n purifiers, A elian, VH 12.50; C re ta n asceticism , F ur. Cretans, fr. 79 Austin. 163 Hymn. Hom. Ap. 388—544, cf. P. B ourboulis, Apollo Delphinios, T hessaloniki, 1949,
In a n O rp h ic trad itio n , purificatory m aterials com e from C rete (O F 156). K illersoften :(5-H, λ ΐ. G u ard u cci, S M S R 1 9 -2 0 (1 9 4 3 -6 ), 8 5 -1 1 4 , G. L. Huxlev, GRBS 16 (1975),
ilee th e re (A pollod. 3.15.8, Porph. Abst. 2.29, Certamen, 2 3 7 -8 in O C T H om er v, 119—24.
p. 234), b ut p erh a p s m erely as a safe refuge. 164 Ar. Ran. 1032, O F 292, G raf, 34 11'. Cf. PI. Leg. 8 7 0 d -e, 872d-873a.

1

5 I ion o f the sh rin e. T h e re is n o th in g intrinsically im p u re ab o u t a
SACRILEGE
A c h a p te r on sacrilege in a book ab o u t pollution perhaps re­
q u ire s ju stific atio n . In co n tra st to m u rd er, there is, it m ight be
arg u e d , n o th in g d irty ab o u t tem ple-robbing; the tem ple-robber
seeking cleansing in a foreign hom e is not am ong the stan d ard
p erso n n el o f m ythology, an d sacred laws, or exègetic traditions,
re la tin g to th e p urification o f the sacrilegious are not attested.
S acrilege has som etim es, therefore, been excluded, im plicitly or
explicitly, from th e possible categories of pollution even in
serious a n d significant discussions.1
‘P u rificatio n s’ after acts o f sacrilege do occur, how ever, and
in good n u m b er. T h ey escape notice because they relate to
m in o r a n d , as it m ight seem , technical offences. I f purification
from sp e c ta c u la r form s o f sacrilege is not attested, th a t is rath er
b ecau se su ch offences are inexpiable th a n because no contagi­
ous d a n g e r a tta c h e s to them . In m ythology, while the m urderer
flees a n d is purified, the m an w ho fights the gods suffers im­
m e d ia te a n d d ra s tic p u n ish m en t. B ut on the day to day level of
cu ltic p ractice, since sanctity is defined in term s o f purity,2
m in o r in fractio n s are treated as pollutions th a t m ust be m et by
p u rifica tio n . P urification will obviously be req u ired if som e­
th in g in trin sically polluting is allow ed to come into contact w ith
th e sa c re d ,3 b u t the principle is b ro ad er than this.4 In several
P elo p o n n esian cults o f D em eter, for instance, the participants
a re fo rb id d e n the use o f elegant an d alluring clothing,5 and in
1 Cf. G e rn e t’s rem ark , Recherches, 36 O n p a ra ît (Glotz au moins) ne vouloir
co n sid érer, en fonction de l’idée de souillure, q u e l’hom icide: m ais l'hom icide est un
d élit récent; des anciens délits —essentiellem ent sacrilèges —on ne d it rien .’ Sacrilege is
explicitly excluded by A dkins, 110 n. 117.
2 A bove, p p. 19Γ.
3 See p. ‘27 n. 60, p. 33 n. 6, Z iehen, n. 61 = Buck, n. 64, Aesch. Eum. 167, 715 f., Ant.
Tetr. l a 10, β 11.
4 Cf. H . J . S tukey, T A P A 67 (1936), 295.
5 L SS, p. 71 f o r a list. Purification: L SS 33. D edication: LSCG6Ü, ? LSS 32. Ideology of
th ese festivals: p. 83 n. 36 above.
Sacrilegi 145
one case it is specified th a t a tran sg ressio n will req u ire purifica-
p u rp le gow n (indeed the offending object is som etim es required
lo be d e d ic a te d to th e goddess); b u t it is polluting in this context
becau se it offends ag ain st the ethos o f a festival th a t requires
w om en tem p o ra rily to renounce the p a ra p h e rn a lia o f sexual
a ttra c tio n . A p o llu tio n like this is wholly m etaphysical, unlike
(hose th a t h av e been considered in previous ch ap ters, w hich
at least h ad th eir origin in tan g ib le im purities. Several other
sac red laws d em an d a purification o f th e shrine in th e event of
tran sg ressio n , a n d if such d o cu m en ts h ad m ore com m only
specified a p en alty , th e list could no d o u b t have been ex ten d ed .6
A fu rth e r difference betw een these cases and those discussed
in earlier c h a p te rs is th at th e object o f purification is the shrine
a n d n o t th e g uilty h u m an . T h is is because, th rough such acts,
the p u re gods suffer defilem ent. T h is conception cam e to be
criticized as cred itin g m en w ith an u n acceptable pow er over
im m o rta ls,7 b u t it u n d o u b ted ly existed in p o p u lar speech,8 and
m u st be co u n ted as a n anom aly in trad itio n al belief. In the case
o f su p p lia n ts, defilem ent m ight be said to fall on the em blem of
th eir san ctified statu s, the su p p lian t crow n.9 T h o u g h m ortals
c a n p o llu te th e gods, how ever, th e gods do not seem to suffer by
it; th e idea, found in som e m ythologies,10 of divine pow er w an ­
ing b e n e a th clogging pollution is not attested in Greece. O n the
c o n tra ry , it is u p o n the offending m ortal th at th e pollution
6 L SC G 136, allow ing a pack-anim al to en ter the shrine of A lectrona at Ialysus, or
e n te rin g w earin g leath er shoes or products o f the pig; LSCG 152, throw ing cakes into
the sprin g s in a sh rin e o f A sclepius and the N ym phs in Cos; LSCG 15+ B 1-16, various
offences; L S S 115 A 2 6 -3 1 , m aking an illicit sacrifice at C yrene; L SS 28, 31, both
o b sc u re. In L SC G 76, 149, ‘p ro p itiatio n ’ is required. Any alteration, however
necessary , o f tem p le goods o r fittings req u ired an άρεστήριον (propitiatory cake): see
L S J s.v., Stengel, 134, Sokolowski on LSCG 32.58.
7 S oph. Ant. 1044, E ur. H F 1232: perh ap s u n d er sophistic influence, VV. Schm id,
Philol. 62 ( 1903), 9. T h e cautious form ulation in Lys. 2.7 m ight be a response to such
criticism s: ιερώ νμιαινομένω ν τούς άνω θεούς άαεβειαθαι. In m odern p o p u lar H induism
o p in io n s seem to differ by region as to w h eth er the gods can be polluted: s c e C .J . Kuller,
M an n.s. 14 ( 1979), 469, w ith references.
8 A esch. Ag. 1645,? E u r. I IF 757, E ur. Heracl. 264, Ion 1118, fr. 368, PI. Leg. 917b,
A lcip h ro n Ep. 4.1. M oulinier, 256 f. in terp rets expressions like τό άγος τής θεού (p. 7 η.
31 ) as th e pollu tio n suffered by the goddess; th at is scarcely the expression’s origin, but
it m ay som etim es h av e been so understood.
’ E u r. Heracl. 7 1.
10 B urkert, S H 89
146 Miasma
re b o u n d s; he falls into the pow er o f the god whose p urity he has
v io lated , becom ing enagês. W hen, in Oedipus at Colonus, O edipus
u n w ittin g ly desecrates the grove o f the E um enides, the chorus
in sist th a t he perform a ‘purification o f these goddesses’, who
h av e been p o llu te d by him . B ut a subsequent rem ark by the
ch o ru s show s th a t it is m ortals, not deities, who are endangered:
‘If .you d o th a t, I will be h ap p y to associate w ith you; otherw ise,
stra n g e r, I w ould be frightened ab o u t you.’“ Practically, there­
fore, the difference betw een sacrilege an d o ther forms of pollu­
tion d isap p ea rs.
I t w as no t m erely in the n arrow ritual sphere th a t a violation
o f religious rules w as seen as a pollution. Pollution occurs if the
d ivinely san c tio n e d rights of su p p lian ts are violated in any way,
not m erely if they are slain a t the a lta r ,12 an d even an offence of
th o u g h t can be spoken o f in the sam e way: he who denies the
efficacy o f divine vengeance ‘lawlessly defiles the gods’. 13 It may
be m ore com m on in such cases to use the language o f divine
a n g e r o r revenge ra th e r th an th a t o f pollution, but, as we have
seen, th e w o rd -g ro u p round agos form s a bridge betw een w hat
a re anyw ay no t tw o sharply circum scribed concepts.14 T he
re ality o f infectious religious d an g er is the sam e in ei ther case,
w h a te v e r lan g u ag e it is described in.
T h e se ideas o f polluted tem ples, su p p lian t crowns, an d gods
a re m erely a specialization o f a very general tendency to envis­
ag e d ev a lu a tio n , the failure to pay honour w here honour is
d u e , in term s o f defilem ent. T h e consulate w ould be polluted,
R o m an a risto c ra ts felt, should a new m an a tta in it; in Greek,
h o n o u r, tru st, ju stic e , an d piety are all liable to tain t, an d this is
also th e co n tex t in w hich the polluted m arriage bed belongs,
a lth o u g h h ere the th reaten ed ideal has found a concrete symbol
for itself.15 I t m ig h t be h ard to find a language in w hich degra-*
d a tio n a n d defilem ent are not connected. In the idea o f polluted
gods, therefore, a form o f co nceptualization is a t work w hich is
no less n a tu ra l th a n th a t w hich sees pollution issuing from the
sta in on th e m u rd e re r’s hands. T h e different source o f the sense
" S oph. OC 466, 4-90-2.
12 A esch. Supp. 375, Kur. Heracl. 71, 264.
13 E ur. H F 757 - b ut cf. G. \V. B ond ad loc.
14 pp. 8 IT.
15 Sail. Cat. 23.6; above p. 3 nn. 8, 9 and p. 95 n. 84.
Sacrilege 147
o f d efilem en t in th e two cases, how ever, leads to th a t difference
in its diffusion th a t w as m en tio n ed earlier: ju s t as an insult
pollutes th e h o n o u r o f the p erson insulted, b u t leaves its
p e rp e tra to r p u re, so too sacrilege initially defiles the gods an d
th eir sacred places a n d only reb o u n d s upon the guilty h u m an
by w ay o f p u n ish m en t. W here, o f course, the consequences of
d e v a lu a tio n ca n n o t be tu rn ed back in this way u p o n the source
o f d isresp ect, th e value itself disap p ears.
It is possible th a t the adjective hagnos, the sta n d a rd term used
to express th e p u rity of the w o rsh ip p er, originally denoted no
m ore th a n th a t respect w hich access to the sacred required.
Hagnos is u n u su al in th e G reek religious vocabulary in th at it
ca n be a p p lied to b o th gods a n d m en. I t used to be assum ed th at
th e m ean in g in b o th cases was th e sam e, ‘p u re ’ an d m ore
specifically ‘c h a ste ’; b u t th ere was always difficulty in seeing
w h a t th e ch a stity o f Zeus, Apollo, a n d D em eter consisted in,
a n d , even w hen used of m ortals, hagnos is a vague term , w hich
re q u ires q u alificatio n from its context to describe p u rity from a
specific ta in t su ch as sexuality. It is etym ologicajly related to
hazomai, ‘I feel o r disp lay rev eren ce/resp ect’, an d w hen applied
to gods, th e ir precincts, or th eir festivals, seems to m ean not
‘p u r e ’ b u t ‘d em an d in g resp ect’.16 S trong su p p o rt for this in­
te rp re ta tio n com es from the parallel case o f semnos, a n adjective
w h ich u n q u estio n ab ly m eans ‘rev eren d ’ and is used in ju s t the
sam e contexts as hagnos. A rtem is, G races, N ym phs, M uses,
N ereids, P oseidon, D em eter, K ore, A thene, Apollo, C htho-
n ian s, Z eus, personified ab stractio n s, divine im ages, seats of
gods, rivers, fires, a n d aither all receive both epithets, an d there
a re no sig n ifican t areas in w hich one is ap plied b u t not the
o th e r .17 W h ile som e deities, alm ost certainly on the basis o f cult,
a re given th em as by rig h t,18 they are applied to others where
th e id ea o f aw esom eness is p articu larly ap p ro p riate. Ajax as-
16 W illiger, 37—72. In prose, hagnos in its sense o f venerable w as su p p lan ted by Aagios,
a p p lie d to tem ples, rites, m ysteries, a n d subsequently deities, b u t never to pure
m o rtals: see W illiger, 72-^84, M oulinier, 281 f., Benveniste, ii, 2 0 2 -7 .
17 Cf. W . F e rra ri, ‘D u e N ote su ka g n o sS tu d . Ital. di Fil. Class. 17 (1940), 3 3 -5 3 .
F e rra ri in this v alu ab le stu d y suggested th e parallelism but did not work it out to its
lim it; I tried to d o so in pp. 329—35 o f my O xford doctoral thesis (1977, sam e title as this
book), b u t h av e o m itted th e d etailed evidence here, as anyone who cares to d p so can
recov er it from lexica.
18 S ee IC X I V 204, 431, Stiglitz, 6 4 -5 , R M L 1.1814 f.
148 Miasma
sa u lte d C a ss a n d ra in the tem ple o f ‘hagna Pallas, she who of all
th e blessed gods is m ost terrible to sacrilegious m o rtals’. W hèn
th e gods divided u p the ea rth , H elios was absent, a n d ‘they left
h im , a rev eren d god, w ithout a sh are o f la n d ’: a god, th at is, who
little deserv ed to be treated in such a way. In ‘they show respect
for th e su p p lian ts o f reverend Z eus’ the etym ological connec­
tion is allu d ed to explicitly.19 Hagnos an d semnos or its cognates
a re b o th used, often together, in the sam e highly charged way to
d e n o te th e inviolable sanctity o f m ysteries,20 supplication,21
s a n c tu a ry ,22 o a th s,23 o r any overriding claim .24 W hile the ren­
d e rin g ‘ta b o o ’ th a t is som etim es proposed for hagnos is imprecise
- one can disp lay respect for a person or place w ith o u t actual
av o id an c e - it brings ou t the inhibitions upon norm al freedom
o f actio n th a t hagneia im poses. ‘N ow is the tim e o f the god’s
festival a m o n g the people, a festival th a t dem ands respect: who
w o u ld d ra w a bow today?’25
T h is in te rp re ta tio n of divine hagneia leaves its relation to
h u m a n hagneia problem atic. T h e parallel w ith semnos fails us
h ere, b ecause the tw o words, so closely com parable as epithets
for gods a n d th e ir possessions, diverge com pletely w hen used of
m o rtals. Semnos m oves outside the specifically religious sphere
b u t keeps the sense o f ‘req u irin g resp ect’; hagnos rem ains princi­
p ally religious b u t now m eans ‘u n co n tam in ated , fit to approach
th e g o d s’, w ith no very obvious u n d erto n e o f ‘reverend’. It does
n o t a p p e a r in this sense in H om er, b u t this is perh ap s coinci­
d en c e, as we find in H esiod the instruction to perform sacrifice
‘re v ere n tly (hagnös) an d purely ’, which* m ight already be the
19 A lcaeus, S L G 262. 16-1 9 , Pind. 01. 7. 59 f., Aesch. Suppl. 652 f. T h e verb is
obviously felt as closely related also in Aesch. Eum. 885, άλλ'εΐμέν άγνόν έστί j οι Πειθοΰς
σέβας.
20 A r. Ran. 386 f., cf. Hymn. Hom. Dem. 476—9.
21 ίκ ίτα ι ό'ίεροί τεκ α ί αγνοί, oracle o f D odona ap. Paus. 7. 25. 1, the only case o f hagnos
being ap p lied to a m o rtal in the sam e sense as to a god (for semnos used sim ilarly cf.
A esch. Eum. 441).
22 A esch. Suppl. 223 f. iv άγνφ . . . ΐζεοθε, i.e. a t the altar: cf. Eur. Andr. 253, 427, H F
715, Suppl. 33, 359.
" άλλ ' άγνόν δρκον adv κάρα κατώμοσα, E ur. Hel. 835, cf. Soph. Phil. 1289, and lor the
άγνόν σέβας of th e gods Soph. O T 830, E ur. Cycl. 580, cf. Aesch. Eum. 885.
24 σέβας όέμηρώ ν άγνόν (so C a n te r for άγιον) ονκ έπηόέσω, Aesch. fr. 135.
25 H om . Od. 21. 258 f. W illiger, 38 renders hagnos ‘religiöse Scheu erw eckend’;
V e rn a n t, 136 says ‘hagnos et hagios m arq u en t la distance, la barrière à ne pas franchir, le
m y stère à respecter . . . ce qui rend le divin, en tan t que tel, in touchable’ (the final
p h ra se p e rh a p s goes too fa r).
Sacrilegi 149
fixed p h ra se it later b ecam e.26 U sed w ithout specification,
hagnos in d icates fitness to w o rsh ip ,27 freedom from religious
co n tam in atio n o f every kind; priests at C yrene are d istin ­
guish ed from laym en as hagnoi, a n d a E u rip id ean chorus-leader,
after statin g in g en eral term s th a t his life is hagnos, goes on to list
the specific p o llu tio n s th a t he avoids.28 W here it m eans ‘ch aste’,
t his lim itatio n is ind icated by context or by an ad d ed genitive; it
ca n also acco rd in g to context express purity from blood-guilt,
b irth , a n d d e a th .29 Such h u m an hagneia is essentially negative,
freedom from this o r th a t pollution; it is the necessary m inim um
if a god is to heed th e w o rsh ip p er’s prayers, and its absence will
n o t go u n av en g ed , b u t it does not bridge the gap betw een god
a n d m an . C o n secratio n is expressed th rough w ords from the
root hosi-, n o t hagn-.30 T o gods hagnos is always applied affec­
tively, to express th e sp eak er’s a ttitu d e to them ra th e r th an to
convey in fo rm atio n , b u t of m ortals it m erely states an objective
fact a b o u t th eir ritu a l statu s. It is difficult, therefore, to recon­
cile d ivine a n d h u m a n hagneia by saying th at the m ortal, by his
hero ic ab stin en c e, com es to share in the divine aw esom eness.31
T h e tw o usages do, certainly, som etim es converge. T h e
w o rsh ip p e r expressed his ‘resp ect’ for the ‘san ctity ’ o f a sacred
place chiefly by p ro tectin g it from pollution; thus its hagneia was
for him defined a n d felt in term s o f purity. T h e E rinyes in
A eschylus w a rn Apollo th a t if he continues to p atro n ize blood-
g u ilty perso n s, his oracle will no longer be hagnos.31 An oracle
26 Op. 336 f., cf. W est’s note an d Ion. fr. 27.5.
27 See e.g. A esch. Suppl. 364,696, X en. Mem. 3.8.10, S O D /5 112, LSCG 130, indices to
LSC G , LSS, L SA , F ehrle, 48.
28 L S S 115 A 21, 24; E u r. Cretans, IV. 7 9 .9 -2 0 Austin.
29 See e.g. M en. Epit. 440, E ur. Hipp. 316, LSA 12.1-9, indices to LSCG, LSS, LSA,
s.v. hagnos, hagneuo.
50 E u r. Cretans, fr. 79.15 A ustin, Ar. Ran. 327, 336, Pl. Resp. 363c, M. H. V an der
V'alk, Mnemos. 103 ( 1942), 125 f., and on the negative ch aracter o f hagneia W illiger 53 f.
B ut for a special use o f άγνίζαι in tragedy see A ppendix 1.
31 F r. Pfister, Phil. Wochenschr. 1923, 359 f., R E Suppl. 6.153; but see W illiger, 53, and
on th e lack o f positive esteem for chastity above, p. 92.
32 Eum. 7151’., cf. e.g. A nt. Telr. 1 a 10, ß 11, X en. Ages. 11.2, T h eo p h r. ap. Porph. Abst.
2.19, p. 4 9 .8 - 10 N auck. T his is W illiger’s explanation, 5 5 -6 0 , lor the ‘p u re’ sense of
hagnos. O th e r coincidences/interferences betw een the two forms o f hagneia: hagni a p ­
plied to a virgin goddess certainly cam e to be understood as ‘ch aste’, (A rist.) Probl.
894b 34 f., a n d p erh ap s in e.g. Aesch. Suppl. 1030-2, Ar. Ran. 875; hagnos o f a sacrifice
m ean s som etim es ‘so lem n ’ (Soph. Tr. 287, cf. 756, X en. Symp. 8.9), som etim es ‘p u re’ in
th e sense o f bloodless (T h u c. 1. 126.6, PI. Leg. 782c, T h eo p h r. ap. Porph. Absl. 2.31,
p. 162.1 V ) .
150 Miasma
w as n o rm ally hagnon in the sense o f ‘d em anding respect’; here,
how ever, w here it is co n trasted w ith pollution, the word obvi­
ously also co n tain s the idea o f undefiled, and the divine and
h u m a n senses becom e inseparable. B ut it is not clear th a t such
cases o f in terferen ce or coincidence betw een the two senses of
th e w ord a re sufficient to explain the original bifurcation, an d it
m a y be w o rth considering the possibility th a t hagnos began as a
Ja n u s -fa c e d adjective like aidoios, indicating both sides o f a
re la tio n involving respect: ‘d em an d in g respect’ o f gods, ‘dis­
p la y in g re sp e c t’ o f m en. T h e o th er two archaic verbal adjectives
in -nos (semnos a n d deinos)33 are norm ally confined to a passive
sense, b u t G orgias a n d Isocrates found it possible to apply
semnos actively,34 ‘respectful’. I f this q u ite speculative
h y p o th esis w ere correct, it w ould show th at ‘p u rity ’ is merely
th e m ost d istin ctiv e aspect o f th a t ‘respect’, the lack o f which
defiles the g o d s’. It is a t all events clear th at the hagneia which
fits a m o rtal to a p p ro a c h the gods is not in conceptual origin a
m a tte r o f physical cleanliness. Hagnos never m eans ‘clean ’ in a
se c u la r context, an d even in reference to ritual p urity is not
n o rm ally ap p lied to in an im ate objects such as the clothes of
w o rsh ip p e rs. T h at w ould, perhaps, be as bizarre as to speak of
‘re sp ectfu l’ clothes. W here it is used o f an object —a precinct,
lu stra l w ater, even an axe35 —it establishes for th a t object, as
so m eth in g sacred, a claim to reverence.
W h e re th e b a rrie r o f respect th a t hedges round the sacred is
v io lated , p o llu tio n occurs. T h e characteristic form o f this re­
sp ec t is, o f course, inhibition. S acred things are com m only
su rro u n d e d by interdiction; a sim ple exam ple comes from
T h u c y d id e s, w ho m entions a spring in Boeotia w hich was ‘not
to be to u c h e d ’ for any except cultic purposes.36 T h e ‘untouch­
a b le ’ sp rin g suggests o th er fam iliar phenom ena o f Greek
religion — things not to be spoken, or m oved, places not to be
e n te re d . D u rk h e im supposed th a t the sacred is typically defined
33 A. D e b ru n n e r, Griechische Wortbildungslehre, H eidelberg, 1917, 159.
34 G orgias B 6, p. 286.12 D /K , Isocr. Bus. 25 (w hich also contains a unique active use
ol hagios, W illiger, 83 f.). Sophocles’ ph rase εΰαεπτος άγνεία (O T 864) is a nice
illu stratio n of hagneia as reverence, an d this m eaning would suit the first attestation of'
th e w ord in reference to m ortals, Hes. Op. 336 f , perform sacrifice άγνώς και καθαρώς.
35 See F errari, o p cit., M oulinier, 40, an d Pind. fr. 34.
36 4.97.3.
Sacrilege 151
by ju s t su ch a com plete a b stra ctio n from o rdinary h u m an use,
as th e o p p o sitio n betw een th e categories o f sacred a n d profane
is o n e th a t ad m its of no co m prom ise.37 B ut th o u g h the gods
m u st obviously m ake d em an d s o f m en in o rd er to insist on the
reality o f th e ir presence a t all, th ere is no logical reason w hy
these d em an d s should take the form o f prohibitions ra th e r than
co m m an d s, n o r w hy sacredness should be d eterm in ed nega­
tively (this sp rin g is not used for profane purposes) ra th e r th an
positively (this sp rin g is used for ritu al). In p ractice it m ay be
h a rd to discover, or even conceive of, a religion in w hich there is
no co n n ectio n betw een sacredness an d in terd ictio n ,38 b u t it is
ce rtain ly th e case th a t th e em phasis given to this connec­
tion varies b o th betw een religions an d w ithin them . Both
P a n a th e n a e a a n d E leusinian M ysteries are sacred occasions of
th e highest im p o rtan ce in the sam e state; this does not m ean
th a t b o th a re su b ject to th e sam e intensity of interdiction. T he
n o rm al G reek w ord for sacred, hieros, does not co n tain in itself
th e no tio n o f ‘fo rb id d e n ’, b u t m erely m arks o u t things th a t are
in som e w ay associated w ith the gods.39 T h is association often
b u t n o t alw ays takes the form of ow nership; sacred diseases,
sac red w ars, a n d sacred days are nob o d y ’s property, while
D elos is th e sacred island o f Apollo even though m ost of the
te rrito ry belongs to individual D elians. W hen sacred things are
c o n tra s te d to hosia,40 things over w hich the gods have no claim
a n d th a t m ay be used freely w ith o u t offence to them , ‘sacred ’
h as com e to en tail ‘restricted ’, b u t th e ch aracter o f the restric­
tio n will v a ry from case to case, an d hieros is often used w ithout
th e o p p o sitio n to hosios being eith er stated or im plied. Hagnos/
hagios differs from hieros in em phasizing the m ajesty of the divine
- a cooking-pot used in a tem ple, though certainly hieros, 41 is not
n ecessarily hagnos — but, as we have noted, the respect th a t is
re q u ire d o f th e w o rsh ip p er need not be synonym ous with
avo id an ce.
37 D urk h eim , 299—325 an d passim; criticized e.g. by S. Lukes, Émile Durkheim, His
L ife and Work, L o n d o n , 1973, 2 4 -8 .
38 S tein er, 129 f. *
39 B u rk ert, G R 402 f. N ote D em . 21.16. O n the expression άνθρωπος ιερός see
B u rk e rt, 403 n. 5.
40 L atte , H R 55 n. 16, 75 n. 40, 114; B usolt/Sw oboda, 514.
41 See e.g. A r. Nub. 254.
152 Miasma
W h e rea s D ü rk h e im p erh ap s saw the relation betw een pro­
fane a n d sac red as unique, it has been pointed ou t th a t it merely
reflects, in intensified form, the p a tte rn s o f respectful behaviour
th a t a re found in everyday life.42 E ach m an is a tem ple, su r­
ro u n d e d by a temenos on w hich no outsider m ay in tru d e w ithout
d u e cause; an incursion is felt as a pollution, an d it is this sense
o f p erso n al defilem ent th a t is p erh ap s the unconscious model
for th e p o llu tio n o f gods, shrines, a n d values in general. T he
m o re resp ected a person is, the less conceivable does it become
to ta m p e r w ith his clothes, e n ter his room unasked, or even,
th o u g h p ro p e rty rights are no t here in question, m ake free w ith
his n a m e o r occupy his special seat. H u m an aw esom eness,
how ever, like divine, dem an d s positive as well as negative
trib u te ; ce le b rato ry d inners h ere take the place o f sacrificial
feasts. A n d even th e m ost revered individual is not condem ned
to a b so lu te un to u ch ab ility ; a w orking relationship can be estab­
lished w hich will allow his w orshippers some access to him and
som e ex p lo itatio n o f his resources, even though the original
re sp ect will no t p erh ap s survive u nchanged too great an
in tim acy .
T h e a rg u m e n t th a t respect for gods differs only in degree
from resp ect for m en can be rep h rased to say th a t there are
m a n y sac red objects outside tem ple walls. In som e societies,
th is seem s to be tru e even on a linguistic level, sacredness being
d e te rm in e d no t by relation to su p ern atu ral pow ers bu t by a
p a rtic u la r a ttitu d e o f reverence associated w ith it. I f I appeal to
you b y w h a t you hold m ost sacred, I p robably have a value
ra th e r th a n a religious relic in m ind. A lthough hieros seems not
to be used in this w ay (the a p p ro p ria te w ord w ould be hagnos), a
G reek could certain ly m ake a sim ilar appeal in the nam e of a
v alu e o r v alued object in ad d itio n to the gods or instead of
th e m .43 I t w as even possible to seek san ctu ary a t the tom b of
o n e ’s o p p re sso r’s father ra th e r th a n a t a sacred place.44 T he
h e rb a list m ig h t extend his display o f respect for the m ysterious
a n d te m p e ra m e n ta l p lan ts he culled to the point of bringing
42 E. GofFman, ‘T h e N atu re of Deference an d D em eanor’, American Anthropologist 58
( J u n e 1956), 473—502, also in his Interaction Ritual, H arm ondsw orth, 1972, 47—96.
43 e.g. H orn. II. 22.338.
44 E u r. Hel. 980—7, T im ae u s 566 FG rH fr. 50 ap. A th. 520b, cf. F ontenrose, 309. For
su p p lic a tin g th e o p p resso r’s wife s e e j. N. B rem m er, Mnemos. 33 (1980), 366 f.
Sacrilegi 153
th e m offerings,45 w hile m en w ho insp ired exceptional aw e by
th e ir d eed s rev ealed them selves as gods thereby a n d earned
cult. A m o re m u n d a n e in stan ce of the p ractical diffusion o f
sac red n e ss w ould be the g arla n d a n d th e values associated with
it. A lth o u g h it is left to th e o b serv er to say th a t the g arlan d is a
m ark o f co n secratio n , anyone w ho w ore one enjoyed, in p rin­
ciple, th e benefits and suffered th e liabilities o f sacredness
(resp ectiv ely inviolability, a n d th e obligation o f p u rity ).46 T h e
g a rla n d m ark e d w ith a certain san ctity m any areas o f G reek life
o u tsid e th e strictly religious sphere; th e p articip an ts at d in n er
p a rtie s p u t one on, b u t certain p u b lic offices too w ere ‘garlan d -
w e a rin g ’.47 T h is last d etail shows, as do the lustral stoups
a ro u n d th e agora a n d the p relim in ary purification o f the assem ­
bly,48 th a t th e co m m u n ity itself w as in a sense a sacred entity.
T h e sacred , therefore, ap p e ars as the intensely venerable
ra th e r th a n th e ab solutely o th er. T h o u g h sacredness is com­
m on ly su rro u n d e d by in terd ictio n , the relation betw een m en
a n d th e th in g s th ey hold in aw e is m ore com plex th an one of
sim p le avo id an ce. T h e gods’ claim s over things they actually
ow n a re n a tu ra lly large, a n d to cede ow nership to the gods is
obv io u sly a n o tab le ab n eg atio n by m en; on the o th er hand,
d iv in e rig h ts in th ese cases m ay be reduced to p roperty rights,
so th a t th e m o rtal is scarcely m ore restricted in his use o f sacred
p ro p e rty th a n o f any o th er p ro p erty not his own. Like other
p ro p rie to rs, gods let o u t th eir land an d lend th eir m oney at
in te re st. W h e re sacred things are not ow ned by the gods, the
p ro h ib itio n s, if an y , th a t derive from sacredness are very varied.
A th e n ia n ‘sacred trirem es’, for instance, were not exem pt from
use in w a r.49 T o ap p reciate th e flexibility of sacredness in its
45 T h e o p h r. Hist. PI. 9.8.7.
46 A r. Plut. 21, A eschin. 1.19. Rem oval o f the crow n in contexts o fd eath or execution,
A rist. fr. 101 Rose1, ap. A th . 675a, Ath. Pol. 57.4, Lycurg. Leocr. 122, (Plut.) Cons, ad
Apoll. 119a; pollu tio n o f su p p lian t crow n, E ur. Heracl. 71. Note too PL Phd. 58a—c. T h e
essen tial po in t a b o u t crow ns was m ade by W ilam ow itz on E ur. H F (377; subsequent
w ritin g s (listed by B u rk ert, GR 101 n. 5 an d YVachsmuth, 312) have obscured the issue
by treatin g a sym bol as a m agical device (G anszyniec in R E s.v. Kranz is a partial
ex cep tio n ).
47 Ar. A v. 4 6 3 -5 , A eschin. 1.19. A certain sacredness o f public office ap p ears also in
th e restrictio n o f arch o n sh ip s, like priesthoods (p. 175 n. 177), to the physically intact,
L ys. 24.13.
48 A bove, pp. 19 an d 21. O n the close relation between piety and patriotism see
D ov er, 250 f , idem , Talanta 7 (1975), 26.
49 VVachsmuth, 285.
154 Miasma
p ra c tic a l im p licatio n s (as well as to confront one o f his own
sacra), a n E n g lish m a n m ight su b stitu te the notion o f ‘royal’; the
re stric tio n s placed upon the co m m o n er’s use of crow n property
a re n o t o f o ne kind (royal parks an d crown jew els), while an
occasion can be ‘ro y al’ w ithout being ow ned by the crown or
su b je c t to restriction. In w h a t follows, different categories of
sac red things will be considered to see w hat are the im plications
of th e ir sacred n ess for h u m an activity, an d w h at the conse­
q u en c es o f th eir desecration.
T h e in stitu tio n o f ‘holy d a y s’ is one th a t few people, perhaps,
a sso c ia te w ith G reece; certainly the Jew ish sa b b a th was an
o b serv an ce th a t m ight ea rn the contem pt o f a G reek a u th o r.50
T h e re existed, none the less, a special term , hieromênia,sx to
d e n o te th e sac red tim e associatéd w ith a festival, an d it is
co m m o n ly in the context o f restrictions placed upon profane
a c tiv ity th a t w e find it used. T h e original sense of the w ord m ust
b e ‘sac red m o n th ’, although it is unclear w h eth er any Greek
s ta te re ta in e d so long an o bservance in the historical period.52 It
w as th e hieromênia o f the C arn e ia th a t prevented the S partans
from m a rc h in g o u t in full force to T herm opylae, perhaps also to
M a r a th o n ,53 a n d on several o th er occasions Peloponnesian
sta te s w ere o b stru cted in th eir cam paigning by the hieromêniai of
th e m ost im p o rta n t festivals.54 G reeks enjoyed festivals, bu t it
c a n n o t h av e been m ere love o f pleasure th a t caused serious
m a tte rs to be neglected in all these cases. T h ere is a religious
o b lig atio n here, a n d one th at, rem arkably, could not be satis­
fied by m a g istrates acting on b e h a lf o f the people, b u t fell upon
th e e n tire citizen body. W hen the obligation becam e intoler­
ab le, it w as co u n tered , like o th er binding religious rules, by
so p h istic evasion ra th e r th an sim ple neglect; to avoid cam ­
50 A g a th a rc h id e s o f C n id u s, 86 FGrH fr. 20.
51 G . R ougem ont, ‘L a H ierom enie des P ythia et les “ trêves sacrées” d ’Eleusis, de
D elp h es et d ’O ly m p ie ’, B C H 97 (1973), 75 -1 0 6 , a l p. 81.
52 R ougem ont, op. cit., 8 6 -9 . N ote also p. 26 n. 40 above.
53 H d t. 7.206.1, 6.106.3 (b u t against referring the latter to the C arneia see P ritchett,
i, 1 1 6 -2 1 ).
54 M a in texts: H d t. 9.7.1, 9.11.1, T huc. 4.5.1, 5 .5 4 .2 -4 , 5.75.2, 5.82.3, X en. Hell.
4.5.11 : cf. H . P opp, Die Einwirkung von Vorzeichen, Opfern und Festen a uf die Kriegführung der
Griechen, E rlan g en , 1957, 75—122, P ritchett, i, 121—6. O nly m ajor festivals seem to have
c o n stitu te d hieromêniai ol this kind, an d the dem an d s even o f these m ajor festivals seem
to have varied: the G ym nop aidia could be postponed (T huc. 5.82.3), the whole
S p a rta n arm y w as not required for the H y ak in th ia (X en. Hell. 4.5.11).
Sacrilege 155
p aig n in g d u rin g a hieromênia, the Argives m an ip u lated the
c a le n d a r so as to stick fast on .th e last available profane d ay .55 It
is h a rd to be clear w h eth er th e suspension o f activity, and
especially o f w arfare, w as essential to the ideology o f the
festival, o r a m ere negative b y -p ro d u ct of the positive req u ire­
m en t to ce le b rate the rites a t the due time. I f the original
hieromênia w as in d eed a m o n th , ab sten tio n from profane activity
m u st surely h ave been valued in itself, as the positive celebra­
tions could scarcely last so long. A D elphic response preserved
by ch a n ce in D em o sth en es shows th a t a period o f com m unal
in actio n m ig h t still be o f religious ad v an tag e in the fourth
c e n tu ry .56
In c e rtain circum stances, a hieromênia m ight prevent aggres­
sion a g a in st a sta te as well as by it. T h e term was ap plied to the
truces, betw een th e host state a n d the p articip atin g states, th at
p e rm itte d th e celeb ratio n o f th e A ttic Eleusinia an d the four
P an h ellen ic a th le tic festivals even in tim e of w ar.57 T hese truces
a re n o rm ally seen as com pacts for m utual benefit, a n d it is
c e rta in ly tru e th a t all G reek states strongly desired the o p ­
p o rtu n ity to p a rtic ip a te in the festivals, and th at w ithout the
tru ce s this w ould have been im practicable. B ut it is not im ­
p lau sib le th a t they grew o u t o f a feeling th a t respect is d u e to an
e n e m y ’s festivals no less th an to his shrines. C ertain ly the
sac red tru ces h ad a special ch aracter, in th at, unlike norm al
truces, d ec en t states w ere expected to accept them au to m ati­
cally if proffered in good faith.58 In 387, the Argives, th reaten ed
by a S p a rta n invasion, h ad recourse to th eir usual stratag em of
d e c la rin g a sacred truce (p ro b ab ly th at for the N em ean
g a m e s )59 ‘n o t w h en the tim e cam e due, b u t w hen the S p artan s
55 T h u c . 5.54.3 (cf. P lu t. Alex. 16.2). Sophism in connection w ith supplication, p. 184
n. 219, w ith oath s e.g. H d t. 4.154.4, and cf. L atte in R E s.v. Meineid 348 = Kl. Sckr. 369 f.
56 D em . 21.53.
57 G. R o u g em o n t, op cit.; F. J . F ern an d ez N ieto, Los Acuerdos Belicos en la Antigua
Grecia, S an tiag o , 1975, i, 147-84. T h e O lym pic inviolability was in th e 4 th c. claim ed,
falsely, to h av e once extended to the Eleans them selves: see W albank on Polybius
4 .7 3 .6 -7 4 .8 .
58 C f. A eschin. 2.133—4. As for X en. Hell. 4.5.1—2, the S p artan s no d o u b t denied the
A rg iv es’ rig h t to proclaim an Isth m ian truce.
59 See references in Popp, op. cit., 144 n. 229. T h e identification dep en d s on dating
th e in cid en t to 387, a N em ean year; it is recom m ended by the fact th at a N em ean truce
d id exist, w hile truces for local festivals are not attested until m uch later (Busolt/
S w o b o d a, 1263). P ritch ett, i, 123 plum ps, w ithout argum ent, for the C arneia; an
a n n u a l festival w ould, o f course, have served th e A rgives’ gam e better than a biennial.
156 Miasma
w ere a b o u t to in v a d e ’. W hen confronted by this stratagem
previously, th e S p artan s, characteristically, h ad w ithdraw n; on
this occasion, the invasion w ent ah ead , but only because
A gesipolis h a d providently checked w ith Zeus a t O lym pia that
it w as ‘safe to reject a truce unjustly offered’, an d subsequently
co n firm ed w ith A pollo a t D elphi th a t on this point he ‘agreed
w ith his fa th e r’.60 F or a S p a rta n arm y to disregard a N em ean
tru c e w ith o u t this explicit sanction w ould have been im possi­
ble, a n d b reach es o f the festival truces are in general very rare.61
A m u ch b ro a d e r ideal is suggested by the P lataean s’ com ­
p la in t ag a in st T h eb es in T hu cy d id es th at they w ere set upon
‘d u rin g a hieromênia’;62 as there is no question here o f a Panhel-
lenic festival, this seem s to im ply th a t all sacred occasions
sh o u ld be ex e m p t from attack. B ut, although the truces for the
g am es a n d Eleusis are p erh ap s specializations from some such
o rig in al ideal, th ere is no o th er secure evidence for its p er­
sisten ce in th e classical p erio d ,63 a n d a good deal o f evidence
th a t it w as n o t in practice observed; festivals w ere the ideal
m o m e n t for a su rp rise a tta c k .64 Such m ethods m ay have been
so m e w h a t im p ro p er, b u t they w ere not com parable to violation
o f a pu b licly h erald ed , an d accepted, festal truce; in the hel-
len istic p erio d , it w as by proclaim ing th eir own festal truces that
local festivals so u g h t to p ro tect them selves.65
In c o n tra s t to th e Peloponnese, there is no question a t A thens
o f sac red tim e interfering w ith m ilitary activity. P resum ably her
festivals could be perform ed satisfactorily even w ith the arm y
a w a y .66 W h a t we do find a t these times is a suspension of
im p o rta n t asp ects o f the life o f the com m unity. Festivals were
60 X en . Hell. 4 .7 .2 -3 .
61 A eschin. 2.12 (cf. D em . 19, 2nd hypothesis, para. 3); A thenian decree in Hespena 8
( 1939), 5 - 1 2 ; im p u te d b reach , T huc. 5.49 —5.50.4. Sanctions ag ain st violation, Thuc.
loc. cit., L SC G 78.47—9. F or observance, note T h u c. 8.9.1-8.10.1; b u t for possible
su sp en sio n o f th e O ly m p ia in 428 and 424 see P ritch ett, i, 120 n. 26. H ost states could
a p p a re n tly d ecline to offer the truce to enem ies, T h u c. 8.10.1; the E leusinian truce,
possibly for this reaso n , seem s to have been ineffective d uring the P eloponnesian war,
X e n . Hell. 1.4.20.
62 T h u c . 3 .5 6 .2,65.1.
63 Cf. how ever p. 155 n. 59.
64 H d t. 6.87, T h u c . 3.3.3, X en. Hell. 5.2.25—36, Aeneas T acticus 4.8 (cf. B urkert, SH
174 n.20), P lut. Pelop. 5.
65 B u so lt/S w o b o d a, 1263.
66 T h e battle o f N axos w as won in 376 d u rin g the E leusinia, Plut. Phoc. 6.7 (noted by
P ritc h e tt, i, 121 n. 28).
Sacrilege 157
no t necessarily g en eral days o f rest (stalls were open an d b u ild ­
ing w ork w en t o n )67 b u t courts, council, assem bly and other
a d m in istra tiv e bodies did n o t hold sessions d u rin g them .68
E x cep tio n s, w hile not actu ally illegal, were stigm atized as
highly irre g u la r. T h ere is no reason to go beyond the obvious
ex p la n a tio n suggested by a fo u rth -cen tu ry decree,69 th at people
w a n te d tim e off to celeb rate th e rites an d enjoy them selves; the
festival c a le n d a r d eterm in ed th e p a tte rn o f holidays, in an
asso ciatio n th a t seem s to us, p erh ap s, m ore self-evident th an it
is becau se it persists in o u r ow n cu lture. If it were tru e th a t the
assem b ly ten d ed to avoid the m onthly ‘sacred d ay s’ o f O ly m ­
p ia n gods even w hen they w ere not the occasion o f a public
festival, this m ight in d icate a m ore disinterested respect for the
sacred ; b u t th e fact is very u n c e rta in .70 A more specific ideology
o f th e festival is suggested by D em osthenes: ‘the city gave each
o n e o f us a g u a ra n te e ag ain st being subjected to any u n pleas­
a n tn e ss o r o u trag e at this tim e, by m aking it a hieromênia’, and
‘w h en you w ere all celebrating a hieromênia, and a law existed
th a t a t this tim e no-one should w rong anyone else either
p u b licly o r p riv ate ly .’71 B ans on the seizure of debtors and the
execu tio n o f crim inals h ap p en to be attested in connection with
p a rtic u la r festivals,72 an d it is very likely th at they extended to
all p u b licly recognized hieromêniai. T h e festival is a tim e of
peace, w h en even legally sanctioned violence is inadm issible.
S pecial legal pro ced u res p ro tect the peace against ‘w rong doing
67 M ikalson, 203.
68 A r. Nub. 620, (X en .) Ath. Pol. 3.2—8, Lys. 26.6, A th. 98b (courts); council and
assem b ly , M ikalson, 186—204, w ith D. M . Lewis, CR n.s. 27 (1977), 215 f. For the
exp ressio n άφ ίοίμοι ήμέραι see L SS 14.47 f.
M Ap. A th . 4.17 le, cf. P .J . Rhodes, The Athenian Boule, Oxford, 1972, 30; cf. Plut. Nie.
28. F o r th e sam e reason there m ay have been a tendency lor business to proceed during
w o m en ’s festiv als, from w hich m en w ere excluded anyway: see M ikalson, 189, and lor
T h e b e s X en . Hell. 5.2.29. At A thens, however, there were no meetings on a t least the
ce n tra l d a v o f the T h esm o p h o ria (Ar. Thesm. 80); conceivably it counted as ‘polluted’
(P ritc h e tt, iii, 212).
70 D. M . Lew is, CR n.s. 27 (1977), 215 f.
71 D em . 24.31, 29.
72 Deni. 2 1.11, 175 f., PI. Phd. 58a—c. X en. Hell. 4.4.2 makes the stay o f execution at
festivals P anhellenic. T h e law against ‘placing a su ppliant branch’, a m ode o f initiating
legal actio n , d u rin g the F^leusinia, belongs in this context (Andoc. 1.1 1 0 -6 ). C hains
an d sa créd n ess in com patible, E ur. I T 468 f. F or the taint of punitive legal procedure cf.
p. I 75 n. 177.
158 Miasma
co n c ern in g th e festival’.73 P u n ish m en t u n d er these procedures
seem s to h av e been fiercer th a n for a com parable offence com ­
m itte d in pro fan e tim e; an individual w ho w hipped an enem y at
a festival w as p u t to d e a th .74 i t was a p a rtic u la r affront to
A th e n ia n feeling th a t the execution o f Phocion should have
o c c u rre d on a sacred day, a n d th a t the city ‘should not even
h a v e been p u re o f public bloodshed d u rin g a festival’.75
T h e sam e conception is to be found in the hellenistic period in
several G reek cities, w here festivals w ere the occasion of
a ‘tru c e ’ involving holidays for children and slaves, perhaps
th e release o f prisoners from chains, an d a cessation of legal
a c tiv ity .76 T h e earliest o f this non-A thenian evidence is a law
from T h aso s, p erh ap s o f the late fourth century, specifically
fo rb id d in g ‘d en u n c ia tio n a n d seizure’, the m ost aggressive
m o d es o f d irec t legal action, d u rin g a list of nam ed festivals.77 A
p assa g e in H o m er th a t has alread y been quoted shows th at the
p e a c e o f the festival is an early value: ‘Now is the tim e of the
g o d ’s festival am o n g the people, a festival th at dem ands respect.
W h o w ould d ra w a bow today?’78
T h e re w as a t A thens a n o th er g roup of days ab stracted from
n o rm a l use, th a t o f ‘im p u re d ay s’, on w hich tem ples w ere closed
a n d ‘n o b o d y w ould begin an y serious u n d ertak in g ’.79 A few
festival days co u n ted as im p u re because of the inauspicious
rites perform ed on th e m ,80 b u t no one w ould have thought of
d e sc rib in g th e P a n ath en a ea, for instance, or D ionysia as im ­
p u re , a n d th ere w ere im pure days, connected originally,
p e rh a p s, w ith phases o f the m oon,81 th a t w ere distinct from
festivals. In m u ch the sam e w ay in S p arta, w arfare m ight, it
73 M acD ow ell, Law, 194—7 (whose in terp retatio n is m ore prag m atic than that
a d o p te d here).
74 D ein. 21.180.
75 P lut. Phoc. 37.2, cf. ibid., 2 8 .2 -3 for feeling about the arrival o f a M acedonian
g a rriso n d u rin g the m ysteries.
76 L. R o b ert, Etudes Analoliennes, Paris, 1937, 177-9; F. J . F ernandez N ieto, Los
Acuerdos Belicos en la Antigua Grecia, Santiago, 1975, i, 151 n. 1.
77 L S S 69, cf. F. S alviat, B C H 82 ( 195sf, 198.
78 Od. 21.2581'.
79 X en. Hell. 1.4.12. Cf. J . D. M ikalson, ημέρα άποφράς, A JP 96 (1975), 19-27; a
different view, in som e respects less convincing, in P ritchett, iii, 209-29.
80 P lvnteria: X en . Hell. 1.4.12, Plut. Ale. 34.1, cf. p. 26 above; Choes: Phot. s.v. μιαρά
ήμερα (b u t cf. M ikalson, loc. cit.); o th er possibilities in P ritchett, iii, 211 f., 215.
81 P ritch e tt, iii, 2 0 9 -1 4 .
Sacrilege 159
seem s, be im p ed ed not ju s t by a cerem ony b u t also by an
u n p ro p itio u s p h ase o f th e m o o n .82 T h u s n o rm al days were
c o n tra ste d to b o th ‘sac red ’ a n d ‘im p u re’ days. T h is is an in ­
stan ce o f th e k ind o f p h en o m en o n th at has often been explained
by th e su p p o sed p rim itive confusion o f th e sacred a n d the
unclean. 83 O n su ch a view, th ere w ould originally have been a
single categ o ry o f forbidden days, only g radually sep a rated out
into tw o classes o f holy a n d im p u re. T w o causes are not neces­
sarily id en tical, how ever, m erely because they lead to the same
effect (th e ab stra c tio n o f a day, or thing, from n o rm al use); and,
th o u g h b o th w ere subject to restriction, sacred an d im p u re days
w ere n o t o f th e sam e ch aracter. O n the contrary, th eir relation
w as one o f p o larity , an d activities especially ap p ro p ria te to one
(like sacrifice or m arriage) w ould be u n th in k ab le on the o th er.84
T h e p o larity is p artic u la rly m ark ed in the case of legal process;
w hereas on sacred days even the execution o f ju d g em en t
ceased , th e A reopagus, m ost solem n in stru m en t o f public
ju stic e , held its sessions on ‘im p u re d ay s’.85
T h e festival peace is a lim ited and tem porary assertion of the
v alu e o f co m m u n ity ag ain st the divisions th a t characterize
n o rm al, pro fan e life.86 T h e G reeks them selves saw religion as
th e cohesive force in every kind o f social grouping. B ut the
o p en n ess a n d u n g u ard ed n ess th a t m ade the festival a tim e for
th e exp erien ce o f fellowship also m ade it an open invitation to
th o se w ho w ished to su b v ert th a t fellowship. M u rd e r a t the
sacrifice is a freq u en t th em e,87 w hile A eneas T acticus, m ost
p ra g m a tic o f G reeks, points o u t th at festivals are the com ­
m o n est occasion for uprisings w ithin the state.88 It is surprising
to n o te th e ex ten t to w hich in such cases feelings ab o u t the
ju s tic e o f th e cause prevailed even in th ird parties over religious
scru p les. A n o b serv er w ho disapproved politically was likely
also to ex perience revulsion at the im piety,89 b u t m u rd er at a
82 Cf. P ritc h e tt, i, 1 1 6 -2 1 , on H d t. 6.106.3.
83 Cf. p. 11.
84 Lys. fr. 53 T h alh eim (5 G ertiet), cf. PI. Leg. 8 0 0 d -e.
85 Pollux 8.117, Etvm. Magn., Et. Gud. s.v. αποφράδες, P ritchett, iii, 210. For the
im p u rity o f p u n itiv e legal procedure cf. PI. Ep. 356d-357a, Arist. Ath. Pol. 57.4.
86 Cf. T urner, C hs. 3—5.
87 E ur. El. 774—858, Andr. 1085—1165, E phorus 70 FG rH fr. 216, Nie. D am . 90 FGrH
fr. 52, D iod. 14.12.3, P lut. Timol. 1 6 .5 -6 , cf. Dion 56.6, Polyaenus, Slral. 1.23.2.
88 22.1 7; for instances see ibid., 17.3, D iod. 13.104.5, X en. Hell. 4.4. 2—4.
89 X en. Hell. 4 .4 .2 -4 .
160 Miasma
festival is no t explicitly identified as agos in o u r sources. A fter a
v io latio n o f sa n c tu a ry p e rp e tra te d in the best o f causes, the
A th e n ia n s found it necessary to purify their city an d expel the
g u ilty fam ily; yet a t th eir drinking parties they regularly
ce le b ra te d H a rm o d iu s a n d A ristogeiton for killing the ty ran t ‘at
th e sacrifice o f A th e n e ’.90 I f political developm ents had been
different, o f course, m ore m ight have been heard o f how the
a c c u rse d p a ir ‘polluted the hieromênia'.
F ro m sacred tim e, we tu rn to sacred space. L and ‘taken
o u t’ for the gods is a w ell-recognized category, com m only
m e n tio n e d , for in stan ce, in connection w ith the ap portionm ent
o f la n d a t th e fo u n d atio n of a colony.91 A curious incident o f the
330s show s the A th en ian s concerned not to profiteer in land at
th e expense o f th e gods. T h e territory of O ro p u s h ad fallen to
A th e n s a n d w as to be divided am o n g the tribes, b u t the suspi­
cion arose th a t a p o rtio n o f it belonged by right to A m phiaraus.
L uckily th e god in question was pro p rieto r of an oracle and
co u ld be co n su lted directly a b o u t his claim s. A com m ission was
se n t to settle th e issue by in cu b atio n in the shrine.92
S acred lan d , like any oth er, was m arked ou t by boundary
sto n es, w hich n orm ally m ad e a declaration ab o u t ow nership.
‘[S acred area] o f C h iro n ’, an early exam ple announces.93 T he
god occu p ied a n a re a ‘cut off’ (temenos), b u t did no t differ in this
from an early king; it was only after H om er th a t the word
temenos, a p p a re n tly secular in origin, becam e specialized in its
fa m iliar religious sense.94 In the classical period, land belonging
to th e gods fell in p ractice into two categories.95 T h e first was
th a t w hich w as genuinely a b stra c te d from h u m an use an d left
u n cu ltiv a te d ; the second, also term ed sacred,96 w as let ou t for
90 A lcm aeonids: p. 16 above. H arm odius: P M G 895.
91 e.g. T h u c . 3.50.2, IG I 3 46 A 1 4-15 (M /L 49), PI. Leg. 738d, (A rist.) Rh. Al. 1425b
22. S acred lan d is n ot a legal category (H arriso n , i, 235), but the point does not concern
us here.
92 H y p erid es, Euxen. 14-1 7 , cf. L. R obert, Hellenica 11-12, Paris, 1960, 194 fT.
93 O n horoi see M . G u ard u cci, Epigrafia Greca iv, Rome, 1978, 4 6 -7 3 , w ith biblio­
g ra p h y , p. 73; C h iro n , ibid., p. 48.
94 K . L atte , R E s.v. Temenos, 435. F or temenos in L inear B, see M . G érard Rousseau,
Les Mentions religieuses dans les tablettes mycéniennes, Rom e, 1968, 208.
95 L atte , op. cit.; for leased sacred land see Stengel, 19-21, O . S chultess in R E s.v.
M isthosis, D. B e h ren d , Attische Pachturkunden (V estigia 12), M unich, 1970, 55 If., an d on
th e extensive D elian evidence B usolt/Sw oboda, 64 n. 4.
96 P. G u ira u d , La Propriétéfoncière en Grèce, P aris, 1893,368, c itin g / G X 1 V 645.98 and
passim-, cf. IG I 2 377 (M /L 62) 16, 21.
Sacrilegi 161
a g ric u ltu re like an y o th er lan d . T h e unw orked lan d w ould
n orm ally be th a t im m ed iately su rro u n d in g the sacred buildings
or a lta r, w hile th e b ro a d er p erip h ery , a n d fields belonging to a
sh rin e b u t n o t a tta c h e d to it, m ig h t be cultivated; it seems,
how ever, from leases th a t a g ric u ltu re will have lap p ed right
ro u n d th e bases o f certain m inor p riv ate shrines.97 T h e d istin c­
tion o f th e tw o kinds of land is show n clearly by an A th en ian
d ecree o f 418 to 417 regulating the san ctu ary o f C odrus, N eleus,
a n d Basile; th e hieron itself is to be fenced off, b u t the temenos
leased o u t to becom e an olive o rc h ard of at least 200 trees.98 It
looks as if th e o rc h a rd is being created o u t o f h ith erto un­
cu ltiv ated lan d , th e change o f use being sanctioned by a sim ple
decision o f th e people. W hen, in th e m id-fourth century, the
possibility w as raised o f b ringing p a rt o f the E leusinian ‘sacred
orgas' u n d e r cultiv atio n , o ra c u la r au th o rity was sought by a
cu rio u s a n d elab o rate procedure; D elphi disapproved, an d the
la n d re m a in ed u n tilled .99 T h a t p a rtic u la r trac t o f land h ad so
long been con tro v ersial th a t special procedures w ere necessary,
b u t in th e case o f a norm al sacred field it was p resum ably felt,
ra th e r su rp risin g ly ,100 th a t a h u m an decision was sufficient,
b ecau se leasin g w as a re-definition ra th e r th an a negation o f the
g o d ’s claim .
U n w o rk ed sacred lan d w as certain ly supposed to be p u re
97 e.g. S IG 3 963, L SC G 47, IG 112 2501. In a T h esp ian lease of c. 230,100 feet are to be
left free a ro u n d th e sh rin e o f M eilichios, B C H 60 ( 1936), 182 f.
98 IG V 84 (L SC G 14), cf. R. E. W ycherley, A B SA 55 (1960), 6 0 -6 .
99 L S C G 32.23 if. In 26 I w ould su p p lem en t not έν]τός (as editors) b u t έκ]τάς τών όρων,
a n d su p p o se th e A th en ian s to be asking leave to cultivate the γη αόριστος o f T huc.
1.139.2, th e a re a a ro u n d th e edges o f the ‘sacred orgas’ in its narrow sense. (O n this
view , the d ecree will be u sing ‘sacred orgas’ in 30 in a broad sense to include the
‘u n b o u n d e d la n d ’, w hile T h u cy d id es co n trasts them .) A year or so later, an A thenian
c a m p a ig n ag ain st M eg ara for en cro ach m en t on the orgas was concluded w ith the orgas
m a rk ed o u t anew , and the έσχα τια ίοσαιησανπρόςτήιόργά όι left uncultivated, άνελόντος
το ϋ θεοϋ λώιον και αμεινον είναι μή έργαζομένοις, 324 FGrH fr. 30, cf. 328 FGrH fr. 155.
T h e o racle referred to is likely to be the one solicited in LSCG 32 (P. F oucart, Memoiresde
I ’Academie des Inscriptions et Belles-Lettres 38 (1909), 179-82), w hich would confirm that
L S C G 32 co n cern s έσχατιαί, land outside th e boundaries. But even if th at connection is
false, it is im p lau sib le th a t, w ithin a year o r two o f a cam paign ag ain st M egara for
e n c ro a c h m e n t, the A th en ian s should be thinking o f bringing the orgas itself under
cu ltiv atio n . O n this M eg arian w ar see Ja c o b y on Philochorus 328 F G rH fr. 155, G.. L.
C aw k w ell, B C H S 2 ( 1969), 3 2 8 -3 2 .
100 C o n tra s t e.g. LSC G 7 2 .3 -8 , ibid., 129, for o racu lar consultations in com parable
s itu a tio n s, a n d cf. SIG 3 987. In S IG 3 965. 1 5 -1 7 the deine Peiraeus specifies rental term s
for such temeni as ‘it is possible and themiton to bring u n d er cultiv atio n ’.
162 Miasma
from the ta in ts o f b irth , d e a th , a n d sexuality discussed in pre­
vious c h a p te rs, a n d also no d o u b t from bodily functions. (The
p o llu tio n o f tem ples could o f course occur in the m ost literal
w a y .)101 In 424, th e T h e b a n s m ade it a serious charge against
th e A th e n ia n s th a t th eir troops h a d cam ped in the sacred
p re c in c t a t D elion, ‘co n trary to the G reek cu sto m ’, and were
d o in g th ere ‘all th e things th a t m en do on profane g ro u n d ’.102 A
te m p le sh o u ld be situ ated aw ay from the com m on p ath , so that
all w ho a p p ro a c h it can ensure th a t they are p roperly p u re .103 It
seem s th a t in A ttica requirem ents o f p urity also in theory
ex te n d e d to w orked sacred land. P rivate cult organizations th at
let o u t th e ir land regularly req u ired the ten an t to trea t the
p ro p e rty ‘as a sacred p lace’.104 Lease contracts for publicly
o w ned p recincts do not contain explicit rules ab o u t purity, but
th e y m ig h t h av e form ed p a rt o f the ‘law ab o u t temenê’105 upon
w h ich su ch c o n tra cts are based; certainly a decree o f the R om an
p e rio d d eclares th a t b irth a n d d e a th are traditionally forbidden
‘in all tem em \ an expression w hich in context ought to include
those th a t a re ren ted o u t.106 W orked sacred land thus probably
re ta in e d a m easu re of notional sacredness beyond the fact th at
th e re n t w as p a id to the god; there were, o f course, degrees of
san c tity , a n d such lan d was not sacred enough to be used, for
in stan ce , as a place o f san ctu ary . (It w ould be interesting to
know w h e th e r a pious in v ad er m ight have felt obliged to exem pt
101 A r. Vesp. 394, Ran. 366, an d for defilem ent as a political w eapon Hesych. s.v. iv
Π νθ ίφ χ έσ α ι (a reference I owe to Sim on H o rn b lo w er). L ater evidence in Fehling, 34; cf.
C o u rtn e y on J u v e n a l 1.131.
102 T h u c . 4.97.3. F o r occupation o f shrines and hero precincts through pressure of
sp a ce cl. I hue. 2.1 7 .1, an d LSA 55 w ith Sokolowski. O n the rules o f w ar in Greece see
J . de R om illy i n j . P. V ern an t (ed.), Problèmes de la guerre en Grèce ancienne, Paris, 1968,
2 0 7 -2 0 . A ctual fighting in the ternenos a t O lym pia, X en. Hell. 7.4.28-32.
103 X en. Mem. 3.8.10.
104 L SC G 47.5—7; IG I I 2 2501. 4, 15; H . W . Pleket, Epigraphica i, L eiden, 1964, n.43;
cf. B e hrend, op. cit., 96 If.
' 105 IG I3 84.25 (LSC G 14).
106 IG I i 2 1035. 10 f., cf. B ehrend, op. cit., 68. O n the date, w hich hovers betw een 1st
c. BC a n d 2nd c. AD, see S E G xxvi 121, w ith bibliography. T h e decree seeks to rem edy
th e illicit a p p ro p ria tio n o f temenê, clearly cultivated fields an d not tem ple precincts, by
p riv a te persons. T h e w ord temenos can be used o f sacred land o f any kind, K . L atte, R E
s.v. Temenos, 435. T h e trees in rented temenê are som etim es protected in the lease (e.g. IG
I I 2 2494. 15 t., IG X I V 645. 135 f.), but p ro b ab ly not for specifically religious reasons;
th e sa m e is tru e o f th e ban on constructing tom bs in IG X IV 645. 137 (although note the
in sisten ce in this context th a t this is ‘sa cre d ’ land).
Sacrilege 163
leased temenê from ravaging.) A req u irem en t o f p u rity m ight
even in special circu m stan ces ex ten d to land th a t w as ow ned by
p riv ate in d iv id u als, a n d was sacred to a god only in the sense of
being p a rtic u la rly d e a r to him . T h e sanctity o f Delos, religious
ce n tre o f th e A th en ian em pire, becam e so in ten se th a t the
A th en ian s found it necessary to expel the existing graves from
(lie sacred islan d , a n d pro tect it in fu tu re from all ta in t o f birth
a n d d e a th ;107 it w as useless for the individual D elian to com ­
p lain th a t h e h a d the rig h t to die on his own p ro p erty if he
chose. T h e sacred n ess o f o th er, o b scu rer ‘sacred islan d s’ was
p ro b a b ly o f this k in d .108 S acredness could thus exercise effective
claim s, d istin c t from those o f ow nership, even over som ething in
w hich p ro p e rty rig h ts did exist. A lexander, som e m ain tain ed ,
h ad in cu rred th e w ra th o f D ionysus by sacking th e go d ’s city of
T h e b e s, for all th a t he sp ared th e sacred places.109 T h e p atro n
god d ess o f A th en s w ould not to lerate the ab u se by individuals
o f olive stu m p s grow ing on th eir ow n la n d .110
U n tille d sacred land could be spoken o f as ‘let go’, w hich
brin g s o u t w h a t it has in com m on w ith days, anim als, and
p erso n s ‘let g o ’ in favour o f th e g o d s.111 I t w ould be interesting
to know w h e th e r a t any period real religious ren u n ciatio n had
been p ra ctise d , in th e sense th a t large areas o f good lan d were
left u n u sed . Som e gods in historical times h ad su b stan tial areas
leased o u t, b u t it can n o t be assu m ed th a t they h a d ow ned this
la n d from o f old, a n d th a t it h ad originally, like the Neleus
s a n c tu a ry , lain uncu ltiv ated . A lthough the gods do not seem
co m m o n ly to h ave b ought la n d ,112 they could acquire it by gift
o r confiscation, an d , in the case of colonies, m ay have received
in th e o rig in al allo tm en t fields in ten d ed from th e first to provide
th em w ith a rev en u e th ro u g h leasing. T h e two su b stan tial areas
o f u n w o rk ed la n d th a t are fam ous, because violation o f them
h elp ed to p re cip ita te two m ajor G reek w ars, m ay b oth be
special cases. T h e E leusinian ‘sacred orgas’ lay on the boundary
107 T h u c . 3.104.1—2. S exuality they tolerated.
108 F or th e categ o ry cf. X en . Cyn. 5.25 (dogs excluded).
109 E p h ip p u s 126 F G rH fr. 3, Plut. Alex. 13.4.
1,0 Cf. p. 165 n. 120.
111 L S C G 32.30, cf. L S J s.v. άνετος, άφετος.
112 O n this an d the following see P. G u irau d , La Propriétéfoncière en Grice, Paris, 1893,
3 6 2 -7 .
164 Miasma
b etw een A th en s a n d M e g a ra ,113 a n d its original consecration
m ig h t h av e o cc u rre d th ro u g h a m u tu al renunciation o f dis­
p u te d territo ry . O th erw ise the ex p lan atio n probably lies in the
c h a ra c te r o f the divinities concerned; the goddesses of agri­
c u ltu re d e m a n d th e ren u n ciatio n o f one trac t o f ‘swelling, teem ­
in g ,’ la n d (orgas) in exchange for the fructification o f the rest. As
for th e C irrh a e a n p lain below D elphi, its special statu s cam e to
be ex p lain e d th ro u g h a curse im posed upon the wicked
C irrh a e a n s after a sacred w a r;114 the explanation m ay be ficti­
tio u s, b u t th ere w ould not have been scope here for a p ro p a­
g a n d is t’s in v en tio n if unw orked land h ad been a fam iliar
p h e n o m e n o n . W e do no t h e a r o f any state renouncing a large
a re a th a t fell u n am biguously w ithin its own territory. Sm aller
tra c ts, how ever, th a t w ere ‘b e tte r unw orked’, like the A thenian
P elarg ik o n , m ay well have been com m on. Som e A thenians
a ttrib u te d som e p a r t o f the d isasters o f the Peloponnesian w ar
to th e o c c u p atio n o f the P elargikon th a t it m ade necessary.115
T h e sa n c tity o f the unsullied m eadow , w here ‘no shepherd
v e n tu re s to feed his flock, a n d iron h as never com e’, is unforgett­
a b ly evoked by E u rip id e s.116 T h e E uripides o f p o p u lar concep­
tio n m ig h t h av e been expected to dism iss reverence for such
p laces as ra n k sup erstitio n ; the poet himself, however, invests
his m eadow w ith pecu liar m oral value. In m ythology and
h isto rical m oralizing, the consequence o f grazing, burning, or
felling a sacred grove m ay be d e a th , m adness, or a curious and
h u m ilia tin g d ise a se .117 Such su p e rn a tu ra l dangers were not in
them selves sufficient d eterren t, as the protection o f unw orked
113 O n its id en tificatio n see U . K a h rste d t, Ath. M itt. 57 (1932), 9 f., E. M eyer in RE
s.v. Megara, 159. O n the etym ology o f orgas see L SJ and the etym ological dictionaries;
E . N o rd en , Aus altrömischen Priesterbüchem, L u n d , 1939, 2 2 -4 , followed by Nilsson, GGR
179 n. 7, w as m isled by lexicographical evidence in his in terp retatio n o f orgas as ‘wild
w ooded la n d s c a p e ’, cf. L atte , Philol. 97 (1948), 155 n. 1 = Kl. Sehr. 102 n. 13.
114 A eschin. 3 .1 0 7 -1 1 2 . F irst reference to this sacred land, LSCG 78.15 ff. (380-79
BC). F ictitious: N . R obertson, C g n .s . 28 (1978), 3 8 -7 3 , b ut cf. Historia 29 (1980), 242 f.
115 T h u c . 2.17. 1 -2 . F or concern over the Pelargikon cf. IG I 3 7 8 .5 4 -7 (LSCG 5),
P ollux 8.101, N ilsson, GGR 79 n.6. All this anxiety arose, it has been suggested (Ziehen,
22 n. 14), from a m isq u o ted a n d m isunderstood o racu lar fragm ent, Anth. Pal. 14.73.1,
TÖ Π ελασγικόν ",Α ρ γο ς άμεινον.
116 Hipp. 73—81. Cf. Hymn. Horn. Ven. 264—8.
117 H esiod, fr. 43 (a) 5—9, (b), (c) (E rysichthon), H dt. 6.75.3 bis (Cleom enes),
9.116.3, 120 (A rtay ctes), S u d a s.v . Ά ναγνράσιοςδαίμω ν.
165
Sacrilegi
groves a n d m eadow s is one o f th e com m onest them es o f sacred
la w s.118 T h e need for h u m an legislation does not, o f course,
d isc re d it th e go d ’s powers; divine vengeance operates through
the h u m a n s w ho ‘com e to th e go d ’s a id ’. 119 A t A thens, th e d eath
p en a lty o rig in ally th rea ten ed an y o n e who tam p ered with
A th e n a ’s sacred olives, b u t th a t is quite ex ceptional;120 in
g en eral, sacred law s treat p astu rin g a flock or cu ttin g w ood in a
g o d ’s grove as a n offence ag ain st property ra th e r th an a
th re a te n in g sacrilege. W hile theft o f sacred goods was a capital
offence, th e p u n ish m e n t here is a fine, an d varies according to
the d am ag e actu ally caused. A n e rra n t sh ep h erd m ay be let
off w ith a n obol for each an im al pastu red , b u t to fell one of
A sclep iu s’ stately cypresses in C os costs 1,000 d ra c h m a s.121
O n e law even d eclares th at, provided the p ro p er price is paid to
th e god, sacred wood m ay be used for ‘sacred, profane and
u n c le a n ’ p u rp o s e s.122
S acred la n d did, how ever, have a special statu s in th at
offences a g a in st it offered a p articu larly valuable h an d le for
po litical m an ip u latio n . A ccusations un d er this head could be
co n stru c te d in su ch a way as to th rea ten crippling penalties,
w h ich m a d e th em a n ideal device for the p u rsu it o f personal
v e n d e tta o r class strife behind a veil o f legal process.123 Sim ilar
m eth o d s p roved no less serviceable in in ter-state relations. A
G reek sta te w ould not ad m it to going to w ar w ithout ju s t
c a u s e ,124 a n d no n e could b e ju s te r th a n ‘com ing to the god’s a id ’
118 See Sokolow ski’s lists, LSS, p. 143, LSCG, pp. 72, 211.
119 F o r th e expression see Aesch. Sept. 14, Soph. O T 136, H d t. 8.144.2, Ar. Lus. 303,
X e n . Hell. 1.2.6, A eschin. 3.120, Diod. 16.25.1, ‘2 8.3, etc., LSCG 177.139. O bservers
h av e so m etim es m isu n d ersto o d this kind o f relation, Steiner, 142 f. In Isocr. 18.3
h u m an p u n ish m e n t is said to avoid the delays o f divine. Cf. Dio C ass. 51.8.3.
120 A rist. Ath. Pol. 60. 2, cf. Lysias 7.
121 L SC G 136.32, 150 A 5. O th e r attested penalties: cuttin g or carrying oil'w ood,
L S C G 37.15, 50 d rach m as (the fine seem s sm all, b u t the legislator seems to have had
m in o r offences like carry in g o lfb ro k en branches chiefly in m ind); 65.78 If., discretion­
ary ; 91.10, 100 d rach m as; LSS 81.10,? 100 d rach m as per tree; T h u c. 3.70.4, one stater
p e r vine-pole; Pollux 8.101 (b u t cf. IG I3 7 8 .5 4 -7 ), 3 drach m as; p astu rin g herds, LSCG
91.11, ? 67, ? 79.29 (f., S1G3 963.37 f., confiscation o f the herd; LSCG 116.12, hemiekton
p e r an im al; b oth offences. LSCG 84.14, ? 50 drachm as. I have not been able to find out
w h at p en alty co m p arab le dam ag e to profane property would carry; but for the olfence
cf. PI. Leg. 843c-d.
122 LSS 115 A 8 -1 0 .
123 I huc. 3.70.4—5.
124 Pl. Ale. 109c, cf. H . Bengtson, Historia 12 (1963), 100-4.
166 Miasma
in defence o f sacred land. T h e p a th betw een cynicism and
g u llib ility is here a n arrow one, an d h ard to follow. W hile it is
tru e th a t ‘a G reek state cham pions the cause o f “ its ow n” deities
. . . to avoid the an g er o f these deities’,125 it is also tru e th at
religious offences w ere often exploited, or condoned, for politi­
cal ends w ith b la ta n t o pportunism . Both the third and fourth
sac red w ars began w ith accusations o f cultivating the sacred
C irrh a e a n p lain , a n d in bo th cases the political designs o f the
accu sers a re p a lp ab le an d u n d is p u te d .126 W hen Aeschines
la u n c h e d the second o f these accusations, the illicitly cultivated
la n d w as, he m ain tain ed , clearly visible from the terrace where
th e D elp h ic A m phictyons held th eir sessions;127 yet, in this as in
th e e a rlier case, the en cro ach m en t h ad ap p aren tly been
to le ra te d until political hostility provoked action. Political
m o tiv a tio n is m u ch less certain in the A thenian interventions
a g a in st M e g a ra over the sacred orgas, b u t th at m ay partly be
b ec au se so little is know n a t all o f the detailed background to
th ese events. T h e fam ous M eg arian decree th at, m any con­
tem p o ra ries felt, p re cip ita te d the Peloponnesian w ar is h ard to
call in evidence on either side; it professed itself a response to
M e g a ria n ab u se o f the orgas,126 b u t the suspicion, rife in ancient
tim es, th a t P ericles’ m otives w ent deeper th an th a t has not yet
b een w holly alla y e d .129 In the m id-fourth century, further
tro u b le o ver the orgas led to actu al w ar against M egara. T here is
no reaso n in this case to d o u b t th a t the A th en ian s’ concern was
religious, as they gain ed no tem poral adv an tag e from the suc­
cessful ca m p a ig n th a t they conducted; b u t there m ay have been
p a rtic u la r reasons w hy A th en ian sensitivity to M egarian im pi­
ety becam e a c u te a t ju s t this tim e, a n d we should not neglect as
a facto r sim ple h u m a n resen tm en t th a t the territory which the
A th e n ia n s piously denied them selves should suffer encroach­
m e n t by the ‘cursed M eg arian s’. 130
125 K .J . D over, A J P 87 (1966), 207.
126 See e.g. P ark e/W o rm ell, i, 222, 236. H . W . Parke, Hermathena 53 (1939), 65-7 1 ,
a rg u e s for a sim ilar accusation in connection w ith the second sacred w ar.
127 A eschin. 3. 118 f. A supervisory p rocedure was supposed to have been established
in 380/79, LSC G 78. 15-21.
128 T h u c . 1.139.2, P lut. Per. 30.
129 D espite Geoffrey de Ste. C roix’s fine book, The Origins o f the Peloponnesian War,
L o n d o n , 1972.
130 F or the w ar see D em . 13.32, 3.20, cf. 23.212, 324 FGrH fr. 30, 328 FGrH fr. 155,
p. ICI n. 99 above. F o r the offence o f encroachm ent on profane land cf. SIC’3 56.25.
Sacrilege 167
E v en fu rth e r a b stra c te d from h u m an use th a n the ‘unsullied
m ead o w ’ is th e a re a to w hich en try is absolutely forbidden, the
abaton. In som e religions, it is said, th e m ore sacred a thing is,
th e less to u ch ab le it becom es, so th a t absolute holiness im plies
ab so lu te inaccessibility. T h a t conception can be only p artially
ap p lie d to th e G reek evidence. T h e classical tem ple, it is true,
co n tain s in th eo ry an in n er room th a t is not to be entered
(adyton); b u t archaeology h as show n th a t a com m on early form
o f tem ple, p e rh a p s indeed th e earliest form, was built round a
c e n tra l h e a rth a n d used for co m m u n al feasting.131 As for the
abaton, th e inaccessi ble grove o r area like th a t o f th e Eum enides
a t C o lo n u s, th ere is no reason to think th a t it was ever the
n a tu ra l form for th e precin ct o f the m ost powerful gods. A l­
th o u g h th e ra tio n a le o f a p a rtic u la r abaton is som etim es obscure,
th e clea rer cases suggest th a t divine force o f a p a rtic u la r kind,
n o t d iv in ity in g eneral, w as hedged ofTin this way. In m arking
o u t abata o f perso n s struck by lightning, an d of the Eum enides,
th e G reeks w ere p ro tectin g them selves against th e universe’s
d e stru c tiv e a n d avenging p o w ers.132 I t is not a coincidence that
pow ers o f ju s t th e sam e kind w ere pro p itiated th rough the
sacrificial ritu al know n as enagismos, w hich involved the total
131 See H . D re ru p , Arch. Anz. 1964, 199—206, idem , Griechische Baukunst in geometrischer
Zeit (A rchaeologia H om erica I I.O ) , G öttin g en , 1969, 123—8. S nodgrass, 408—13
a rg u e s for th e h e a rth tem ple as the earliest form ; co n trast B urkert, GR 150. T h e reality
o f th e h e a rth tem p le is, how ever, disp u ted by B. Bergquist, Herakles on Thasos, U ppsala,
1973,61.
132 Abata w h ere lightning has stru ck e.g. E ur. Bacch. 10, Pollux 9.41, IG I I 2 4964,5
(Z eu s K a ta ib a të s ), cf. W . B urkert, G lotta 39 (1961), 208—13 w ith references. Abaton of
th e T rito p ateres: IG I2 870 (cf. IG I I 2 2615, SE G xxi 650). O f the E um enides: Soph. OC
126. O f a heroized d ead w om an: IG X II. 3 suppl. 1626 (SIG3 1223). O f the Hyacinthids:
E u r. Erechtheus, fr. 65.87 A ustin. U nspecified: IG X I I. 3. 4 5 3 -5 (probably tom bs, cf.
S IG 3 1223 co m m en tary ); IG X I I. 3 suppl. 1381; X I I. 5. 255; LSCG 121; LSS 34, 128
(access to th e last 3 m ig h t have been occasionally perm itted); inscr. in F. C ourby, Le
Portique d'Antigone (Delos 5), Paris, 1912, 9 7 - 102 (a further A thenian instance is cited
ib id ., 101 n. 2); The Inscriptions o f Cos, ed. W. R. P aton and E. L. Hicks, O xford, 1891, n.
8, line 11 (cf. S. M . Sherwin-VVhite, Ancient Cos (H ypom nem ata 51), G öttingen, 1978,
135 η. 283). T h e p h en o m en o n is com m oner in A rcadia, Paus. 8.30.2, 38.6 (Zeus
L ykaio s), 8.31.5 (a grove, unspecified), 36.3 (cave o f Rhea, open to priestess). T h is is all
th e evidence for actu al abata I have found (the idea o f an u n tread ab le island is
co m p reh en sib le, A rr. Indica 37.4, but I know no G reek instances). Shrines open once a
y ear only a re found, D em . 59.76, Paus. 6.20.7, cf. PI. Criti. 116c, b u t the h abitual
inaccessib ility h ere is p erh ap s a by-product o f the positive idea o f perform ing ritual in a
sp ecial sp o t o n ce a year. E xtensive m aterial on the them e, uncritically arran g ed , in
J. \Y . H ew itt, ‘T h e M a jo r R estrictions o f Access to G reek T em p les’, TAPA 40 (1909),
*83-92.
168 Miasma
d e stru c tio n o f the sacrificial offering. G re at goddess though she
w as, it w ould not have m ade sense to shun A th en a thus, or
refuse all sh are in offerings m ade to her.
T h e san c tity o f tem ples a n d sacred im ages, a n d the divine
a n g e r strik in g those w ho violate them , are alm ost too fam iliar to
n eed illu stra tio n :133 Even w hen A lexander razed T hebes to the
g ro u n d to in tim id a te the G reeks, he spared its sacred places.134
11 w as m u tila tio n o f divine im ages th a t, at A thens in 415, led to
a s p e c ta c u la r series o f im piety trials. I t is w orth pausing over
th e events o f 415, because they are b e tte r docum ented th a n any
c o m p a ra b le in cid en t in G reek religious history, an d are often
th o u g h t to illu stra te the reserves o f superstitious fear th at were
re a d y to com e b u b b lin g th ro u g h once a crisis h ad cracked the
sm o o th surface o f late fifth-century rationalism . Even scholars
w ho d id n o t g ru d g e the G reeks th eir belief in gods have spoken
o f ‘en o rm o u s h ysterical fuss’ a n d ‘real religious h y steria’, and
th e p roceedings did not escape the censure o f T hucydides,
w ho n o ted th a t ‘good m en w ere convicted on the testim ony of
b a d ’.135 All o u r sources agree, how ever, in show ing th at the
h y steria, if h y steria it was, was not exclusively religious.
W h e n all th e herm s w ere m u tilate d by unknow n persons in a
single n ig h t, sh o rtly before the Sicilian expedition set sail, the
A th e n ia n s re a c te d w ith ‘an g er a n d fear’. R epeated em ergency
m eetings o f council a n d assem bly w ere held, a board o f in­
v estig ato rs estab lish ed , an d rew ards offered for inform ation,
from a n y q u a rte r, ab o u t the m u tilatio n and any o th er acts of
im p iety . W h en these m easures provoked the revelation (or
fa b ricatio n ) th a t A lcibiades a n d others h ad parodied the
m ysteries, a la rm only grew , a n d som ething like a w itch-hunt
d e v e lo p e d .136 G e n u in e anxiety ab o u t the religious conse­
133 See e.g. A esch. Ag. 3 3 8 -4 2 ,5 2 7 , Pers. 8 0 9 -1 5 ; H d t. 1.19. 1 -2 (accidental burning
o f tem p le causes disease), 8.33, 53.2, 109.3, 129. 2—3, 143.2; instantaneous self-inflicted
p u n ish m e n t for those w ho ta m p e r w ith im ages, H d t. 5.85. 1 -2 , Dem . 24.121. T h e good
g e n eral respects enem y hiera, X en. Ages. 11.1; o n e’s ow n side burns tem ples by accident
only, H d t. 5.102.1, D iod. 16.58.6 (contrast P aus. 10.35.3).
134 A rr. Anab. 1. 9.9.
135 D o dds, 191, M . P. N ilsson, Greek Folk Religion, New York, 1961, 122; T huc. 6.53.2.
O n these events I h av e found m ost useful D. M. M acD ow ell’s edition o f A ndocides On
the Mysteries (O xford, 1962), an d K . J . D over in A. W. G om m e, A. A ndrew es, K .J .
D o v er, A Histoncal Commentary on Thucydides, vol. iv5 O xford, 1970. 264—88.
136 P lut. Ale. 18.8, A ndoc. 1.40, 36, T h u c. 6.27.2 w ith À ndoc. 1.27-28, 40.
Sacrilege 169
q u en c es o f im p iety w as certain ly felt. I t was inform ation ab o u t
‘o th e r acts o f im p iety ’, not o th e r th rea ts to the state, th a t was
solicited by offers o f rew ards a n d im m unity, an d A ndocides
vividly p re sen ts th e lan g u ag e o f religious d an g er in w hich the
first su ch d e n u n c ia tio n was m ade. ‘Y ou, A thenians, are ab o u t
to sen d o u t this enorm ous expedition a n d expose yourselves to
d a n g e r. B u t I will show you th a t y o u r general A lcibiades has
alon g w ith o th ers been paro d y in g the m ysteries . . Λ 137 T his
b rin g s o u t tw o ch a rac te ristic religious preoccupations, the
especial need for sound relatio n s w ith the gods w hen dan g er
th re a te n s, a n d the p a rtic u la r d an g e r th a t atten d s on an im pious
co m m an d e r.
T h e o th e r side to th e affair is in d icated by T h u cy d id es. T h e
m u tila tio n w as believed, he says, to be not ju s t a ‘bad om en for
th e v o y ag e’, b u t also ‘p a rt o f a revolutionary conspiracy for the
overthrow o f the dem ocracy’.138 T his is not T hucydidean bias,139
as it is ju s t as clear from th e o th er accounts th a t the m ost
ex tre m e m anifestatio n s o f p an ic w ere p ro d u cts o f political fear.
T h e a tm o sp h e re o f a w itch -h u n t so vividly evoked by A ndoci­
d es w as created by dem agogues w ho insisted th at a large con­
sp ira c y w as a t work, a n d w hen th e A thenians left th eir hom es to
s p e n d a n ig h t u n d e r arm s in th e T h eseum , they w ere a n ticip at­
in g a n in v asio n from w ith o u t tim ed to coincide w ith an uprising
from w ith in .140 T h e m u tilatio n w as held to be a pledge in crim e
given to one a n o th e r by th e co n sp irato rs;141 h ad it been m erely a
d ru n k e n p ra n k , it w ould have rem ained an act o f im piety, but
th e re w ould have been no g re at cause for a la rm .142 Even if the
p ro fa n a tio n o f th e m ysteries was not originally seen as. p a rt of
th e co n sp iracy , it certainly cam e to be associated w ith it, and
th e ju ro rs responsible for the condem nation pro b ab ly thought
th a t th ey w ere castin g th eir votes against treacherous as well as
im p io u s m e n .143
137 T h u c . 6.27.2, A ndoc. 1.11.
138 6.27.3.
139 So D o d d s, 202 n. 78.
140 A ndoc. 1.36,45, T h u c. 6.61.2.
141 A ndoc. 1.67, cf. D over, op. cit., 286; for such pledges cf. Diod. 13.112. 4, Thuc.
3.82.6.
142 P lut. Ale. 18.8.
143 P iut. Ale. 21.3, T h u c . 6.61.1; on the tim ing see M acDowell, op. cit., 184 §4.
170 Miasma
‘All you people ever think ab o u t is conspiracies’, says a
c o n serv ativ e in A ristophanes, m ocking the irrationality of
d e m o c ra tic fe ars.144 B ut the m u tilatio n o f the herm s was a
sy stem atic u n d erta k in g , not a dru n k en exploit, w hich de­
m a n d e d investigation even if it did not constitute p ro o f of
c o n s p ira c y .145 T h e G reeks believed th a t respect for the gods and
re sp ect for th e law s o f m en w ere p ro d u cts of the sam e inhibitory
p ro c e ss ,146 a n d this being so it is presum ably tru e th a t the
p e rso n p re p a re d to defy society in the one respect m ight also
defy it in th e o th er. T h e real irratio n ality lay in the conduct of
th e in v estig atio n , as T h u cy d id es points out. It is striking to
ob serv e how quickly in this charged atm osphere one accusation
o f im p iety led to an o th er, a n d all w ere believed;147 b u t it is also
n ecessary to ask w h a t h ad charged the atm osphere. I n a differ­
e n t political context, a su rp risin g n u m b er of the im pious
w retch es w ere able to re tu rn to A thens a n d resum e public
life.148
S acred e q u ip m e n t belongs on sacred prem ises, an d is not to be
p u t to pro fan e u se .149 A ctually to steal sacred property, of
course, is an offence the enorm ity o f w hich is indicated in
several w ays. A t A thens the tem ple-robber, like the traitor, was
d e n ie d b u ria l in his native land, an d m any G reek states chose
m eth o d s o f execution ap p a ren tly intended to p revent any form
o f b u r ia l.150 W h en A lexander issued his fam ous ‘recall o f exiles’
d ec re e in 324, only tem ple-robbers an d m urderers were ex­
c lu d e d .151 T h e em otional charge attac h in g to the offence was
144 A r. Vesp. 488 f., cf. L. W oodbury, Phoenix 19 (1965), 180.
145 O n th is a n d the following sec D over, op. cit., 285 f.
146 B u rk ert, GR 372 f., cf. e.g. Lys. fr. 53 T halh eim , 5 G ernet, καταγελώντες τών θεών
καί τών νόμων τών ήμετέρων.
147 1n terestin g m odern parallels in G. G rote, A History o f Greece, new edn. in 10 vols.,
L o n d o n , 1888, vi, notes on pp. 1 1 ,3 7 ,4 7 ,4 9 .
148 (l.y s.) 6.13 f., A ndoc. 1. 35, 53, 55. F or p articu lar cases note A lcibiades and 2
figures w ho a p p a re n tly retu rn ed in his w ake, A deim antus and Axiochus, both politi­
cally active in 407 (M /L , p. 246). M ost o f them , though, presum ably had to w ait for the
recall o f exiles in 404. T h e affair could be joked ab o u t, Ar. Lys. 1094, Plut. Ale. 20. 6 -7 .
145 L SC G 116.22—5, L SS 24, 27, 117, L SA 74; disrespect to sacred property, Andoc.
4.29, D em . 21.16, 22.73.
150 A thens: X en. Hell. 1.7.22, cf. Diod. 16.25.2. T hrow ing over a cliff, Aeschin. 2.142,
P lu t. Praec. Reip. Ger. 825b; throw ing in the sea, Diod. 16.35.6.
151 D iod. 17.109.1.
Sacrilege 171
sufficient to m ake ‘tem p le-ro b b er’ a term o f everyday abuse;
o ra to rs exercised ingenuity in devising ways in w hich their
o p p o n e n ts h a d d ep riv ed the gods o f th eir d u e a n d so fell in to the
a b o m in a te d ca te g o ry .152 D isrespect for sacred m oney was a
m ark o f extrem e social decay, the behaviour o f a ty ra n t or
b a r b a r ia n .153
T h e te m p ta tio n to tem p le-ro b b ery was, o f course, enorm ous,
p a rtic u la rly in tim e o f war. N ow here else was so m uch m ovable
w e alth so read ily av a ila b le .154 A t the sta rt of the Peloponnesian
w ar, th e A th en ian s seem p ru d e n tly to have rem oved the tem p ­
ta tio n by tran sferrin g to the acropolis the treasu re from m ost of
the o u tly in g san ctu aries. T h e pious N icias, we are told, delib­
e ra te ly d elay ed in o rd e r to allow the Syracusans tim e to guard
th e ir rich O ly m p eio n ag ain st his eager tro o p s.155 B ut although
su ch p re cau tio n s w ere necessary, piety surely posed its own
re stra in ts. L o o tin g m ight o ccur on im pulse, b u t the sacred
p laces o f th e enem y w ere never an explicit target; w hen a
d e d ic a tio n sen t by the ty ra n t D ionysius fell into A thenian
h a n d s on th e w ay to D elphi, they kept it to pay th e m ercen a­
rie s ,156 b u t w ould scarcely have a p p ro p riated it ifit had reached
a tem ple. If b arb a ria n s sacked G reek shrines, th a t was why
b a r b a ria n invasions o f G reek territory were alw ays u nsuc­
cessfu l.157 T h e P anhellenic san ctu aries displayed th eir w ealth,
u n p ro te c te d a n d unm olested. I t w as not, it seems, until the
h ellen istic perio d th a t the trad itio n al inviolability o f such places
ca m e u n d e r serious th reat, an d even then at the h an d s o f pirates
livin g on th e fringes o f the G reek w o rld .158
152 A buse: L S J s.v. ιερόσυλος. O rato rs: D em . 22 .6 9 -7 1 , 24. 111 f., 120 f., Γ29 f., 137,
49.65, cf. Lys. 30.21, Isae. 5.44. F ram in g on a charge ο i' hierosulia occurred in legend
(A eso p ), an d n o d o u b t also in reality: A rist. Pol. 1304 a 3, Plut. Praec. Reip. Ger. 825b.
S everal sta te s ex ten d ed th e category o f hierosulia to include a variety o f related offences,
L a tte , H R 8 3 - 6 .
153 Social d ecay: Solon, fr. 4.12, Soph. Ö T 8 8 3 —96. T y ran t: X en. Hieron 4.11, Diod.
14.67.4. B a rb a ria n : n. 157 below.
154 Cf. D iod. 16.56.6 on D elphi.
155 IG I3 52 (M /L 58) A 18-2 2 , cf. M /L , p. 158; Plut. Nie. 16.7, cf. Demetr. 30.2.
G u a rd s em ployed ag ain st dom estic tem ple-robbing: LSCG 60 w ith Sokolowski.
156 D iod. 1 6 .57.2-3.
157 See the passages o f H d t. 8 cited in p. 168 n. 133 (cf. 1.105), Diod. 14.63. 1 -2 , 70.4,
7 6 .3 - 4 ,7 7 .4 .
158 E. Schlesinger, Die griechische Asylie, diss. G iessen, 1932,63—8. D ionysius, a tyrant,
sack ed a rich E tru scan tem ple, Diod. 15.14.3—4, A lexander’s governors w ere unreli­
ab le , A rria n Anah. 7.4.2.
172 Miasma
T h e p e rp e tra to rs o f the m ost striking exception b ro u g h t upon
them selves a to rre n t o f execration. U n d e r pressure from the
D e lp h ic A m p h icty o n y in 356, the Phocians am azed G reece by
o c c u p y in g the u n p ro tec ted D elphi; to finance the w ar th at
follow ed they w ere eventually com pelled, despite their initial
p ro te sta tio n s, to m elt dow n D elphic treasure into coins for
m e rc e n a rie s.159 O v e r the details o f th eir subsequent fates pious
h isto ria n s gloated w ith a n insistence th a t recalls early C hristian
lite ra tu re on th e d e a th s of the persecutors. Philom elus hurled
h im se lf from a cliff, O n o m arch u s was crucified, Philon died
u n d e r to rtu re, w hile P hayllus suffered the prolonged p unish­
m e n t o f a lingering disease. I f P halaecus lived on for a while, the
g ods w ere hold in g him in reserve for fu rth er hum iliation and
to rm e n t.160 As for th e troops, m an y w ere killed by the avenging
A m phictyones; som e were b u rn t to d eath when, m iraculously, a
te m p le w h ere they h ad sought refuge took fire.161 T h e m er­
cen aries w ho escaped cam e in the end to no good, although
som e w ere exploited by the divine providence to do Tim oleon
good service before perishing them selves.162 K ing A rchidam us
w as p u n ish e d for his involvem ent by loss of burial, the A the­
n ia n s a n d S p a rta n s by loss o f lib erty .163 T h e P hocian wife who
received E rip h y le ’s necklace o u t o f the D elphic spoils was killed
by h e r son, o r a t least plotted m u rd e r against her h u sb an d - we
to u c h h ere u p o n tragic h isto ry ’s freest fantasies.164 L ater
so u rces m ay have in troduced elaborations, b u t this religi­
o u s in te rp re ta tio n was already firm ly fixed in the earliest ac­
c o u n ts .165 In th e concluding treaty, the Phocians involved in the
159 M a in source, D iod. 16.2 3-39, 5 6 -6 4 : cf. N. G. L. H am m ond, J H S 57 (1937),
4 4 —78, P ark e/W o rm ell, i, 221—31.
160 D iod. 16.61.1—3, 56.4.
161 D iod. 16.35.6, 5 8 .4 -6 (in Paus. 10.35.3, however, the T h eb an s light the m atch).
162 D iod. 16.63.5, 78.4, P lut. Timol. 3 0 .7 -1 0 .
163 P aus. 3.10.5, in th e context o f T heo p o m p u s 115 FGrH fr. 312 (cf. 232), Diod.
16.64.1. ‘D io d o ru s, how ever, entirely fails to m ention the fate which overtook Thebes,
th e a r d e n t a n d persevering cham pion o f Apollo, eleven years later’, com m ents C . T . H.
R . E h rh a r d t, T h e T h ir d S acred W ar, unpublished B. L itt, thesis, O xford, 1961, 73.
164 D iod. 16.64.2, D am ophilus 70 FG rH fr. 96, T heopom pus 115 FG rH Îv. 248; m ore
in P ark e/W o rm ell, i, 231 f., nn. 3 0 -1 .
165 E la b o ratio n s: see P arke/W orm ell, i, 228. T h e earliest accounts: cf. preceding
note. U n fo rtu n a te ly D io d o ru s’ source is unknow n. H e can scarcely be D am ophilus,
b ecau se o f th e d iscrep an cy betw een Diod. 16.64.2. an d 70 FGrH fr. 96. H am m ond's
a s c rip tio n o f D iod. 16.64 to a different source from 16.61-3 (CQ 31 (1937), 83) is
unco n v in cin g ; 64 is the clim ax to which 61—3 lead.
Sacrilege 173
‘tem p le-sa ck in g ’ w ere declared ‘accu rsed ’, an d , som e fifteen
years later, A eschines could in te rp re t th e tu m u ltu o u s events of
th e in terv e n in g years th ro u g h th e o p eratio n o f divine favour or
en m ity to th e various states, in acco rd an ce w ith th eir role in the
sac red w a r.166
T h e P h o cian action, how ever, was less m onstrous an d u n ­
ex a m p le d th a n th eir eventual failure an d disgrace has m ade it
a p p e a r. By th e end o f the fifth cen tu ry , various exceptions to the
th eo re tic al u n to u ch ab ility o f sacred m oney h ad come to be
a c cep ted . A tem p le m ight lend o u t its spare resources in coined
m oney to in d iv id u als a t in terest, ju s t as it leased its temenë for
cu ltiv atio n ; b o th p ractices are found in the accounts o f Apollo’s
tem p le on D elo s.167 M ore analogous to the Phocian case was the
tra d itio n , first attested at A th en s in connection w ith the S am ian
c a m p a ig n o f 4 4 0 ,168 of financing w ars by public borrow ing from
tem p le funds. T h e gods could not refuse these loans, for w hich
th ey received no security an d on w hich the in terest m ight be
re d u ced to a n alm o st nom inal ra te ;169 such borrow ing w as not
con fin ed to coined m oney b u t exten d ed to dedications, which
w o u ld be m elted dow n, an d even, in Pericles’ fam ous phrase, to
th e very gold on th e im ages o f the gods. If the god-fearing felt
u n ea se a b o u t th ese m easures, th eir dissent is not recorded, and
th e fact th a t th e loans m ade by the gods d u rin g the Peloponne­
sian w a r w ere for th e m ost p a rt never repaid can n o t be show n to
h av e been a source of serious guilt to the A thenians. As an initial
ju stific a tio n , th ere was the in terp re tatio n o f such requisition as
m e re borrow ing, w ith in terest payable; T h u cy d id es’ Pericles
insists on this. It w as p ro bably m ore im p o rtan t psychologically,
a n d in th e eyes o f the G reek w orld, th at the treasures which
w ere now b eing used in defence o f A thenian interests h ad been
d ed ic a te d , for th e m ost p art, by A thenians, in hon o u r o f A thens’
d iv in e p atro n ess. It was th eir ow n m oney, in origin, th at they
w ere exploiting, a n d in a cause o f w hich the goddess herself
w ould approve.
166 D iod. 16.60.1, A eschin. 3 .1 3 2 -4 .
161 C f. IG I3 248, N em esis o f R ham nus; ibid., 402, Delos, loans, and leases (M /L
53,62).
168 IG I3 363 (M /L 55). For Pericles see T h u c. 2.1 3 .4 -5 .
169 M /L , p. 215; in the early years o f the w ar it seems to have been charged a t more or
less the going rate (H . T . W ade-G ery, CR 44 (1930), 163—5, A. B. W est. TAPA 61
(1930), 234 f )
174 Miasma
T h is w as th e crucial difference betw een the A thenian and the
P h o cian c a se .170 T h e P hocians u ndoubtedly professed to be
m erely ‘b o rro w in g ’ D elphic funds, b u t few of the offerings in
q u e stio n will h av e been m ade by P hocians, an d th eir claim to
a d m in is te r D elphi a t all was fiercely disputed. W h a t rendered
th e P hocian actio n so outrageous, therefore, was not the exploi­
ta tio n o f sacred p ro p e rty in itself, b u t the fact th at Phocians
w ere tu rn in g to th eir ad v an tag e w h at m em bers of o ther states
h a d piously re n o u n ced . T h e re is an analogy w ith feelings about
th e cu ltiv atio n o f sacred land, w here too it is h u m an indignation
th a t fires the sense o f sacrilege. A cts like th at o f the Phocians
h a d often been m ooted in the p ast, som etim es w ith ra th e r more
ju stific a tio n , b u t never carried into practice. D uring the Ionian
rev o lt, H e catae u s h ad proposed th a t C roesus’ dedications in
A p o llo ’s sh rin e at D idym a should be p u t to use; a t the sta rt of
th e P elo p o n n esian w ar, there w as talk on the S p artan side of
ex p lo itin g D elp h ic a n d O lym pic treasure; in the 360s, w hen the
A rc a d ia n league h ad occupied O lym pia, an d was in a position
ex actly an alogous to th a t o f the Phocians at D elphi, m oderate
o p in io n w ith in it h a d sw ung ag ain st expropriation o f the trea­
su re ‘lest we leave the gods a com plaint against o u r children’.171
By th e ir initial u n d erta k in g to leave the dedications in tact, the
P h o cian s them selves im plicitly condem ned their subsequent
ac tio n .
D esp ite th e ir sacrilege, how ever, the Phocians did not lack
sy m p a th y a n d prom ises of assistance. T he A thenians and
S p a rta n s , u n ited to Phocis by hostility to T hebes, m ay initially
h av e h o p ed th a t it w ould be possible to conduct the w ar w ithout
b ro a c h in g th e sacred funds; w hen th a t hope failed, they did not
w ith d ra w th e ir su p p o rt, a n d it was not because o f religious
scru p le s th a t it proved in the m ain ineffective. It would be
in te re stin g to know w h a t at this tim e the feelings o f god-fearing
p ro -P h o c ia n s w ere. Some h ard y spirits m ay have argued th at
th e gods forgive even crim es com m itted u n d er duress, an d tu rn
th e ir a n g e r ag a in st the au th o rs o f the constraint (in this case
T h e b e s ).172 M ore, no do u b t, will have seen in the affair a choice
170 Cf. G . L . C aw kw ell, Philip o f Aiacedon, L ondon, 1978, 6 4 -6 .
171 H d t. 5.36.3, T h u c . 1.121.3, 143.1 (a h in t o f sacrilege in the word κινεΐν?), Xen.
Hell. 7 .4 .3 3 -5 , cf. 6.4.30.
172 Cf. T h u c . 4 .9 8 .5 - 7 .
Sacrilege 175
o f evils. D esp ite m oralizing stories o f in stan tan eo u s p u n ish ­
m en t, m ost G reeks th o u g h t o f th e consequences o f sacrilege as
‘b a d h o p es’ for th e fu tu re .173 O n this occasion, th e expectation
o f d is ta n t evil m u st h ave seem ed preferable to the presen t th reat
o f d o m in a tio n by T hebes.
By th e term s o f th e final settlem en t the defeated Phocians
w ere re q u ired to repay th e full am o u n t they h ad ap p ro p ria ­
te d .174 A tre a ty on any o th er term s w ould have been h ard to
conceive of, as d eb ts to gods w ere a serious m atter. N ot all states
claim ed the right, as did A th e n s,175 to a n n u l sacred no less th an
p u b lic d eb ts by plebiscite. (T h e A th en ian s a p p e a r here as m ore
E ra stia n th a n m ost G reeks, a n d it is possible th a t th eir exploita­
tion o f sacred resources d u rin g th e P eloponnesian w ar, w hich
w as trea ted ab o v e as an acceptable m odel, was in fact viewed in
som e q u a rte rs as sacrilegious.) T h e Eleans in 420, eager to
p la c a te th e S p a rta n s in exchange for the retu rn o f Lepreon,
offered to rem it th eir h a lf of th e fine they had im posed on them
for b re a c h o f th e O lym pic truce, and pay them selves the half
th a t w as d u e to the g o d .176
A m in o r b eneficiary of sacredness is the priest, l h e priest is a
k in d o f w alking tem ple; he avoids those categories o f polluted
p erso n s w ho a re d eb a rre d access to sh rin es,177 b u t in retu rn
a c q u ire s a claim h im self to th e inviolability o f the sacred place.
H e is defiled if p rofane h an d s are laid upon his u n to uchable
ro b e s .178 W h en A lex an d er ca p tu red T hebes, he m ade over all
th e territo ry for d istrib u tio n am o n g his allies ‘except the sacred
p a r ts ’, a n d enslaved all the in h ab itan ts ‘except the priests and
173 F o r ‘g o o d ’ o r ‘b a d ’ hopes consequent o n co nduct see e.g. PI. Resp. 331a, Isoc.
8 .3 3 - 4 , D em . 19.240, X en . Ages. 1.27, M en. fr. 494.
174 D iod. 16.60.1—2. C a p tu re d P hocian arm s were destroyed, ibid., 3. For the sm elt­
ing a n d re-d ed icatio n by th e pious O p u n tia n L ocrians o f the coins struck by the
P h o cian s from th e tem ple treasure see P arke/W orm ell, i, 229.
175 Dem . 24.55; for the counter-evidence see L atte, H R 51 f.
176 T h u c . 5.49. 5. Cf. Bulletin Épigraphique 92 (1979), n. 185 for a decree specifying th at
rev en u e from H eracles’ sacred q u arries a t Eleusis m ay not be diverted for non-sacral
p u rposes.
177 F o r p riestly p u rity see pp. 52 (b irth an d d eath ), 86 fi. (sexuality) ; also PI. Ep.
3 5 6 d -3 5 7 a (no co n tact w ith im prisonm ent, execution), Leg. 759c (priest to be true-
b o rn , physically in tact, p u re both in h im self an d in descent from pollutions such as
bloo d -g u ilt): cf. A rist. Pol. 1329 a 29—30 (citizen), A naxandrides, fr. 39.10, LSCG 166.9
w ith Sokolow ski’s note (physically in tact), LSA 73. 6 - 8 (citizen stock for three
g e n eratio n s).
178 E u r. I T 798 f.
176 Miasma
p riestesses’. H e m ay have h a d his H om er in m ind; for H om er
d escrib es how the sage O dysseus spared a ca p tu red priest of
A pollo, a n d how the h ea d stro n g A gam em non b ro u g h t plague
u p o n the G reek host th ro u g h disrespect to another. A lcibiades
even released c a p tu re d priests w ith o u t a ra n so m .179 In certain
circ u m sta n ces, th e o rd in ary individual could benefit from the
sa c re d in violability by d onning the em blem o f sacredness, the
crow n, o r by perform ing tem porary service to a god. T h e blows
th a t M eid ias d e a lt D em osthenes w hen he was D ionysus’
khorêgos sm ote, the victim tells us, religion itself.180 F or the sam e
re aso n it m ig h t be dan g ero u s to o b stru ct the due perform ance
o f rites. W h en , in the fourth century, H elike an d B oura
w ere engulfed by a tid al wave, traditionalists countered loose
scientific talk o f the s u b te rra n e a n com pression o f air by dis­
covering a n offence o f this kind by the two villages against
P o se id o n .181 P u n ish m e n t even aw aited those w ho harm ed a
g o d ’s h u m b le r d ep en d en ts, his sacred herds. T h e com panions
o f O d y sseu s le a rn t this to th eir cost, while, in the late sixth
ce n tu ry , a citizen o f A pollonia is said to have been blinded by
his fellow citizens because he fell asleep on w atch a n d let wolves
d e v o u r th e sacred cattle o f the s u n .182 X enophon once found
h im self in possession o f a n old a n d sickly sacred horse. Fearing
it m ig h t die n a tu ra lly (and so be lost to the god?), he fattened it
u p for im m e d ia te sacrifice.183
R itu a l is a final area o f m anifold restriction. Every sacred
p re c in c t a n d every festival h ad its ow n distinctive rules; of
p re c in c t rules we have som e know ledge through surviving ex­
a m p les o f th e inscriptions set up a t the entrances, while for
179 A rria n Anab. 1.9.9, Plut. Alex. 11.12; H orn. Od. 9 .197-201, //. 1.9-100 (but note
II.5.76—83, Od. 2 2 .3 1 0 -2 9 ); P lut. Ale. 29.5 . O n C reu sa’s crim e see p. 185 n. 224.
180 D em . 21.126. O n th e crow n see p. 153 above.
181 D iod. 1 5 .4 8 -9 , cf. H . B. G ottschalk, Heraclides o f Pontus, O xford, 1980, 94 f.
D elian s c o n d em n ed to perm an en t exile for expelling A m phictyones from tem ple, IG II2
1 6 3 5.134-140. .
182 H orn. Od. 12.374—419, H d t. 9.93.1—3. O n sacred cattle o f the sun cf. Hym. Horn.
Ap. 4 1 1 -1 3 , B u rk ert, S H 94. T o eat o f the sacred fish in A reth u sa’s pool a t Syracuse
m e a n t in sta n t d e a th , Diod. 5.3.6, from T im aeus. O n sacred flocks see LSCG 79 with
S okolow ski’s co m m en tary , ibid., 67.15, Stengel, 93 f. U nfortunately the penalty for an
offence a g ain st them seem s n ot to be specified.
183 Anab. 4.5.35.
Sacrilege 177
festival ru les w e a re d ep e n d en t on ch an ce allusions in literary
sources. M a jo r categories o f rule co n cern :184
(1) rig h t o f access to the sh rin e or festival (typical excluded
categories are m en, w om en, slaves, foreigners)
(2) co n d itions o f p u rity req u ired for access
(3) dress
(4) m ode o f sacrifice
(5) fo rb id d en objects (e.g. knots, sw ords, m etal objects, pack
an im als, skins).
A full tre a tm e n t would req u ire an in terp retatio n o f m ost
o f G reek religion a n d m uch o f G reek society, a n d cannot be
a tte m p te d here. T h ro u g h the first class o f rules, m ajor d em ar­
catio n s o f society find intensified expression. W e have seen, for
in stan ce , how the division o f capacities and duties betw een the
sexes is a c c e n tu a te d in the ritu al sp h ere.185 R ules o f class 3
su p p o rt rules o f class 2 (cleanliness) o r 1 (avoidance o f the cloth­
ing o f the p ro s titu te ).186 Sacrificial rules also relate to d em arca­
tion, b u t o f th e divine ra th e r th a n h u m an sphere. Incense,
typical ac co m p an im en t o f O ly m p ian sacrifice, is n o t to be used
in th e cult o f th e un d erw o rld goddess H e cate.187 (Seldom , u n ­
fo rtu n a te ly , is th e m ean in g o f such a regulation as perspicuous
as th is.) R u les o f class 5 are d iv erse;188 they derive p artly from
co n c ern for p u rity , p artly from the sym bolic classification of
a n im a ls a n d an im al pro d u cts, p artly from m agical dangers, and
no d o u b t from o th e r motives besides. C ertain rituals, most
n o ta b ly th e E leu sin ian m ysteries, are fu rth er protected by rules
o f secrecy, a n d it is p ro b ab ly here th at we find at its m ost
m ark e d in G reek religion the connection betw een sacredness
a n d in terd ictio n . Superficially, secrecy divides profane know­
ledge from g u ard ed sacred knowledge; it is pro b ab ly m ore
im p o rta n t, how ever, th a t a division is thereby created betw een
th o se w ho h ave access to this knowledge and those denied it.
T h e secrecy o f th e T h esm o p h o ria em phasized the separation of
184 C i. H . J . Stukey, TAP A 67 (1936), 2 8 6 -9 5 , an d , on a crucial principle for the
in te rp re ta tio n o f these do cu m en ts (the laws only specify m istakes that are likely to be
m a d e), H. Seyrig, B C H 5 \ (1927), 197 f.
185 p p . 82 fl'.
186 See p. 83 n. 36.
187 L S S 133.
188 See L S C G 65, 68, 124, 136, 154b, 170; LSS 28, 32,3 3 , 59, 60, 91; LSA 6, 14.
178 Miasma
th e sexes, th a t o f the E leusinia the exclusive good fortune, in the
a fte r life, o f th e blessed élite.
B reach o f ritu a l regulations m ight, as we have seen, require
p u rific a tio n o f the shrine. O ften the sanction is not stated, and
w e a re left to w o n d e r w h eth er it w ould be an uneasy conscience,
a p u rifica tio n , or ac tu a l legal action. In cases ju d g e d serious,
p ro se c u tio n w as certainly a possibility. A t A thens in the fourth
ce n tu ry , a h ie ro p h a n t who h ad m ade an illicit sacrifice to please
a lad y w as convicted; his p u n ish m en t is unfortunately not
re c o rd e d .189 A sp ec ta cu lar violation is said to have been
p e rp e tra te d by the d a u g h te r o f N eaera, who, though disquali­
fied on tw o co u n ts (non-citizen birth, and p rostitution), suc­
c e ed ed in becom ing wife of the archon basileus an d perform ing the
so lem n rites re q u ired o f the basileus’ wife. T he A reopagus called
h e r h u sb a n d to acco u n t for m arrying such a creature, but,
a ss u re d th a t he h ad done so in ignorance, and w ould now put
h e r aside, took no fu rth e r a c tio n .190 T h e council seems to have
ta k e n a n in tere st in an o th er such case, although here too no
a c tu a l p u n itiv e m easures are re co rd e d .191
T h e re ca n be little d o u b t th a t A thenian courts w ould have
been p re p a re d to strike h a rd in defence of the Eleusinian
m ysteries. In 415, there w ere ag gravating political factors, but
p erso n s su sp ected o f profaning the m ysteries w ould surely have
b een co n d em n ed to d e a th a n y w ay .192 T h e reasons for this were
n o t exclusively religious. Eleusis h ad too im p o rtan t a place in
th e im age the A th en ian s had o f them selves as benefactors and
civilizers o f G reece for any attack on it to be tolerable. T h ere is
no h isto rical evidence to show w h a t m ight have h ap p en ed if
co m ed y ’s fan tasy h ad been fulfilled, an d a m ale u surper d e­
te c te d a t th e T h esm o p h o ria , but we do have in this connection
189 (D em .) 59.116 f. As it would have suited the o ra to r’s arg u m en t to em phasize a
severe sentence, we can p erh ap s infer th a t it w as not severe. T h e h iero p h an t’s real
offence, o f course, concerned sta tu s and rights as m uch as religion: he m ade a sacrifice
th a t belonged to a n o th e r official. For an accusation o f illicit sacrifice see Lycurgus, fr.
4.1, in th e L oeb Minor Attic Orators, ii, ed. J . O . B urtt, 1954, p. 142.
190 (D em .) 5 9 .7 2 -8 4 .
191 Isae. 6.49 f.
192 F or A eschylus’ p rofanation see N auck, TGF, p. 28. O n e o f the few secure facts
a b o u t D iagoras o f M elos is th a t his offence w as against the m ysteries: see e.g. schol. RV
A r. A v. 1073 —the decree m ay o f course have been passed in the wake o f the H erm es
affair. O n D iagoras see F. Jacoby, Berl. Abh. 1959. 3, L. W oodbury, Phoenix 19 (1965),
1 7 8-211.
Sacrilege 179
tw o good in stan ces o f th e p o p u lar su b -literary genre o f ‘im piety
in sta n tly p u n is h e d ’. 193 M iltiades, seeking to en ter th e tem ple of
D e m e te r T h esm o p h o ro s on P aros for nefarious purposes, was
seized by p a n ic a t th e entran ce; as he fled he w renched his hip,
a n d d ied from an infection th a t en tered the w ound. B attos of
G yren e in sisted on view ing th e g o d d ess’s forbidden mysteries;
th e p riestess trick ed him by show ing him harm less things, but
th e w om en ce leb ran ts, b lo od-stained from sacrifice, rushed on
him a n d u n m a n n e d him . In b o th cases the goddess prevented
th e im p iety b u t p u n ish ed th e in ten t; the existence o f the stories
proves the in ten sity o f the taboo.
In d iscussing som e cultures, it w ould be necessary at this
p o in t, h av in g listed th e restrictions th a t b ar the passage from
p ro fan e to sacred , to tu rn to th e fu rth er set th a t b a r the p ath
back. D esacralizatio n after co n tact w ith the gods is som etim es
ju s t as n ecessary as sacralizatio n in p rep aratio n for it. T h e
w o rsh ip p e r in th e O ld T esta m e n t w ashes after touching a
sac red book, o r g arm en t. W h a t th e pagans told Isaiah was not
‘s ta n d by thyself, com e not n ear to m e, for I am holier th an
th o u ’ b u t ‘for I w ould sanctify y o u ’. 194 In catholicism today the
co m m u n io n chalice m ust be w iped (‘purified’) after th e m ass
before a p ro fan e person like th e sacristan can h an d le it.195
S acred rules h ave th u s becom e ‘tw o-w ay’;196 the gods m ust be
p ro te c te d from h u m an profanity, b u t m en too m u st be p ro ­
te c te d from d iv in e sanctity. O b jects like clothes once used in a
religious co n tex t becom e u n u sab le for any o th er p u rp o se .197
193 H d t. 6 .1 3 4 -6 , Aelian, fr. 44 H ercher, cf. D etienne, Eugénies. For the genre cf. p. 168
n. 133, A elian, fr. 10,35—7 ,4 3 . It was obviously cultivated in priestly circles, (Lys.) 6.1,
A ndoc. 1.29, a n d th e Strafwunder in the E p id a u rian tem ple record. Such stories seem to
be univ ersal; in m o d ern In d ia, untouchables a b o u t to exercise th eir newly acquired and
hotly d isp u ted rig h t o f tem p le en try are driven off by bees, deb arred by m ilitan t cobras,
d ro w n ed o r eaten by crocodiles d u rin g p rep arato ry b ath in g (E. B. H a rp e r ed., Religion
in South Asia, W ash in g to n , 1964, 180). F o r 16th-century E ngland see J . C arey, John
Donne, Life and A rt, L o ndon, 1981, 21; for 19th-century O xford, G. Faber, Oxford
Apostles2, L o n d o n , 1936, 442. T h e b ro ad er genre o f ‘im piety p u n ish ed ’, perhaps after
lo n g delay (P lu t. Ser. Num. Vind.), acq u ired a certain intellectual respectability in the
co n tex t o f philo so p h ical d eb ates on divine ju stice, cf. H. B. G ottschalk, Heraclides o f
Pontus, O xford, 1980, 95. There is a g reat deal o f m aterial o f this kind, often with
E p ic u rean s as victim s, in th e fragm ents o f A elian’s On Providence an d On Divine Appear­
ances (pp. 1 9 0 -2 8 3 H erch er, cf. R E 1.486).
194 L eviticus 16:23-4; Isaiah 65:5; R obertson Sm ith, 4 5 0 -3 .
195 E. des Places, La Religion grecque, Paris, 1969, 376.
196 D ouglas, 18.
197 R o b ertso n S m ith , loc. cit.
180 Miasma
(T h is is yet a n o th e r source o f the d octrine o f the ‘prim itive
confusion o f th e sacred a n d the u n clea n ’.) In G reece, however,
it seem s th a t the sacred becom es contagious only in particular
circu m stan ces. A G reek n atu ra lly w ashed before a rite, bu t it
w ould be stra n g e to find him w ashing again after it, as a H ebrew
d i d .198 H e m ig h t have done so after p articip atin g in the cult of
h eroes, o r ‘gods o f aversion’, 199 b u t these were pow ers o f a
special kind, w hose precincts w ere som etim es ‘u n trea d ab le’,
a n d offerings to w hom w ere com m only b u rn t whole w ithout
h u m a n p a rtic ip a tio n . T h e ‘exit’ or desacralization process is
little developed in the ritu al o f O ly m p ian sacrifice.200 It was
c u sto m a ry for the E leusinian in itiate to dedicate the clothes in
w h ich he w as in itia te d ,201 b u t this is not necessarily because
th e y h a d becom e too sacred for norm al use; it m ay be an
ex a m p le o f th e com m on p ractice o f m arking a transition in
s ta tu s th ro u g h an a p p ro p riate dedication. N orm ally it w as only
th ro u g h tran sg ressio n th a t infection occurred; the offender
a g a in s t a religious rule becam e enagês, while the object intro­
d u c e d illicitly into a shrine h ad to be left behind.
T h e aspects o f sacredness th a t have been discussed so far
re p re se n t, one m ight say, the self-protection of the gods, the
ru les by w hich they m ark out th eir place in the world. T hough
su ch rules h av e social im plications, they cannot be seen as
specific p ro d u c ts o f the social process, a n d w ould not if rem oved
leave an im m e d ia te cleft in social life. T h e m antle o f sacredness
is also extended, how ever, over crucial areas o f the relations
b etw e en m en. O ccasionally, p erh ap s, it w ould be ap p ro p riate
to speak o f a conscious exploitation o f the sacred inviolability
for p ra g m a tic ends: fines for secular offences, for instance, were
h a r d to evade a n d im possible to recover if they w ere declared
p a y a b le to a g o d .202 M ore com m only the process seems quite
198 L SC G 151 B 23 looks like a case: ‘the priest sacrifices a n d is sprinkled with
s e a -w a te r,’ b u t it is n ot certain th a t the w ashing follows the sacrifice.
199 P o rp h . Abst. 2.44. O n th e im purity o f hero cult see p. 39 n. 25.
200 H . H u b e rt a n d M . M auss, Sacrifice: Its Nature and Functions, trans. YV. D. H alls,
L o n d o n , 1964, 4 5 - 9 ; indeed the utility o f the concept in a G reek context is question­
a b le , cf. G . S. K irk in Entretiens Hardt 27, 68—70.
201 A r. Plut. 845. ο νκά ποφ ορ ά sacrificial rules are p erh ap s m ore relevant.
202 L atte , H R 48—61. F or a law declared ‘sa cre d ’ see M /L 13.14. F or dedication of
p ro p e rty by in d iv id u als to pro tect it see L atte, H R 100, an d for dedication in connection
w ith m an u m issio n , ib id ., 101—11.
Sacrilege 181
u n stu d ied . A p ro m in e n t exam ple o f this unconsciously applied
sacred n ess is th e p ro tectio n afforded by the gods to su p ­
p lia n ts.203 As th e in stitu tio n seem s to some ex ten t to have
develo p ed in th e period covered by o u r sources, it will be
co n v en ien t to consider the H om eric and classical evidence
sep a rately . E ven in H om er, ‘su p p licatio n ’ occurs in diverse
contex ts, an d , w h a t is m ore im p o rta n t, w ith very differing
p ow er to co n stra in the person su p plicated. F rom the m any
possibilities, tw o form s can be singled out, w hich m ight be
term e d ‘h elp m e’ a n d ‘spare m e’ supplication. In the first, the
su p p lia n t en tered territory controlled by the person su p ­
p licated , p erform ed a ritu al act o f self-abasem ent, an d m ade a
re q u est. S uch su p p licatio n was com m only addressed to a
m em b er o f a d ifferen t com m unity, a n d the ‘help m e’ su p p lian t
(literally ‘c o m er’) is clearly assim ilated to the stran g e r.204 A
c h a ra c te ristic req u est, m ade by exiled hom icides, for instance,
w as for ad m issio n to the foreign com m unity, w hich is why a
social g ro u p like th e p h ra try w hich w ould be confronted by
ap p e a ls of this kind m ight h o n o u r Zeus o f S uppliants am ong its
p a tro n d eities.205 Such a su p p lian t had an absolute claim not to
be h a rm e d by th e person he had supplicated; this claim,
g u a ra n te e d by Z eus o f S u p p lian ts,206 w as an intensification of
th e s tra n g e r’s sim ilar claim , g u aran teed by Zeus o f Strangers.
H is rig h t to receive the aid he sought was p erhaps not absolute,
b u t if the re q u e st w as reasonable it w as certainly very strong. If
th e su p p licatio n w as successful, a social bond was created th at
e n ta ile d lastin g m u tu a l obligations.
‘S p are m e’ su p p licatio n is the ap p e al for m ercy in b attle in
ex c h an g e for ran so m . ‘S u p p licatio n ’ here is a term o f conveni­
ence, because, alth o u g h ‘help m e’ an d ‘spare m e’ supplication
exploit the sam e ritu al gestures, the second would p erhaps not
h av e been describ ed by H om er as hiketeia. Lycaon, entreatin g
m ercy, g ro u n d s his ap p eal on the claim th at he is, because o f a
203 S e e jo h n G o u ld 's fine article 'Hiketeia', J H S 93 (1973), 7 4-103. O n rejection of
su p p lic a tio n as pollu tio n see p. 146 above. A n ostracon perhaps accused A risteides of
a n oll'ence ag ain st su p p lian ts, M /L , p. 42. Cf. too p. 152 n. 44.
204 J . G o u ld , op. cit., 9 0 -4 .
205 Sec R. H erzog, ‘H eilige G esetze von K os’, Berl. Abh. 1928, 35, and lbr related
p h ra try gtid sC . Kollev, B C H 89 ( 1965), 45».’
206 H om . 11. 24.570.'
182 Miasma
p rev io u s en c o u n te r, ‘like a s u p p lia n t’ o f the ‘help m e ’ category
to A chilles.207 I f he needs to argue th a t he is ‘like a su p p lian t’,
c lasp in g A chilles’ knees can scarcely have m ade him into a real
s u p p lia n t. T h is is not a m ere linguistic quibble; the conse­
q u e n c e is th a t the ‘sp are m e’ su p p lian t has no Zeus o f S up­
p lia n ts to invoke in his defence, a n d thus nothing like an abso­
lu te title to m ercy .208 T h e victor often in practice takes pity,
b ec au se the b a ttle s u p p lia n t’s self-abasem ent has a strong em o­
tio n a l a p p e a l, bo th intrinsically209 a n d by association w ith true
su p p lic a tio n ; b u t for a hero to d isp atch a m alefactor who has
c lu tc h e d his knees does not p resen t a m oral p roblem .210
T h e role o f H o m e r’s Zeus o fS u p p lian ts is readily understood.
P la to ex p lain s th a t offences against strangers a n d suppliants
a re p a rtic u la rly offensive to the god because such people are
p a rtic u la rly helpless;211 all th a t needs to be ad d ed is th a t the
god is safeg u ard in g a necessary social institution, because, in a
w o rld w here in n o cen t people are constantly being driven into
exile, th e av en u e o f rein teg ratio n th a t supplication provides is
in d isp en sab le.
In classical supplication, tw o changes can be observed.
W h e n th ere is a n y kind o f th re a t o f constraint, the supplication
n ow norm ally takes place a t an a lta r212 and, although the
a p p e a l for p ro tectio n in a foreign country still occurs, more
co m m o n ly the altars serve as a place o f refuge eith er in norm al
207 //. 21.75. F o r th e claim to be a n actual su p p lian t.co n trast Od. 5.450,9.269, 16.67.
T h e only case I find in H o m er o f a hik- w ord applied to a ‘spare m e’ supplication is II.
22.123, hikesthai (less form alized th a n hiketeuöo r hiketes).T his is a m ovem ent tow ards the
classical convergence o f lissomai a n d hiketeuö (for a weak use o f the latter cf. Od. 11.530).
O n the tw o stem s cf. A. C orlu, Recherches sur les mots relatifs a Vidée de prière (E tudes et
c o m m e n ta ire s, 64) P aris, 1966, 298 f. Benveniste, ii, 253 f. argues th at one becom es a
hiketês by ‘reach in g th e knees’ (IL 18.457, Od. 5.449, 9.266 f.) of the person supplicated.
B ut th e em p h asis on ‘reach in g ’ the knees m ight be a secondary application of the idea of
a rriv in g co n tain ed in hiketes; B enveniste m istranslates Lycaon’s plea (‘je saisis tes
g en o u x ; j e suis ton su p p lia n t’) to m ake it fit. G ould, op. cit., 93 n. 100a suggests on
d ille re n t g ro u n d s th a t ‘help m e’ supplication is prim ary, ‘spare m e’ ‘m erely a crisis
ex ten sio n , a m etap h o rical a d a p ta tio n ’.
208 F o r th e g o d s’ p rim a ry concern w ith stra n g e r supplication see esp. Od. 5. 4 4 7 -9 ,
9 .2 6 9 -7 1 . Od. 14.278—84 is a p artial exception, b u t this is half-way betw een the two
form s o f su p p lica tio n .
209 B u rk ert, S H 4 5 -7 .
210 Od. 22. 3 1 0 -2 9 .
2.1 Leg. 729e-730a.
2.2 A lread y a possibility in H om er, Od. 22.334—6: o th er early instances, the CyIonian
c o n s p ira to rs, a n d , p ro b ab ly , A lcaeus in exile (fr. 129, 130).
Sacrilege 183
w a rfa re o r civil strife. T h ese two developm ents to g eth er are
re sp o n sib le for th e p ecu liar d ilem m as th at constan tly su r­
ro u n d e d th e classical in stitu tio n . T h e su p p lian t now in theory
enjoys ab so lu te inviolability, because he shares the sacredness of
th e a lta r to w hich he clings;213 a n d this inviolability is available
to th e m alefacto r in the m iddle of th e society w hich he has
w ron g ed . F o r th e S p artan s, once th e traito r P au san ias had
re a c h e d an a lta r, no satisfactory line of co n d u ct w as avail­
a b le .214 T o leave him to die o f starv atio n in the sacred area
m e a n t p o llu tio n , b u t, as they found, to carry him o u t ag ain st his
will w hen on th e p o in t o f d e a th was no better; the only policy
safe in religious term s - free p ard o n - was socially intolerable.
F ierce sp irits m ig h t claim th a t religious p rotection did not
e x ten d to villain s,215 b u t, in th e case o f P ausanias, Apollo of
D elp h i h a d d eclared th at it did. W hereas the H om eric in stitu ­
tion w as socially functional, classical su p p lication m ay ap p ear
a n ab u se; th e s u p p lia n t’s san ctity is no longer a necessary
religious shield held by society over an im p o rtan t social rela­
tion , b u t a concession d ragged from an unw illing society by
d e sp e ra te in d iv id u als exploiting the logical consequences o f the
s a n c tity o f sacred places.216
T h e real d ilem m a th a t this situ atio n created is illu strated by
th e p ro p a g a n d a w a r in w hich S p a rta an d A thens engaged
before th e o u tb re a k of the ac tu a l P eloponnesian w ar.217 All
th re e o f th e ou trag es th en cited took the form o f killing of
su p p lia n ts, a n d all th ree w ere still rem em bered alth o u g h they
h a d o cc u rre d far in the past, one of them alm ost 200 years
before. In none o f th e three cases, w ith a p artial exception,218
d id d e a th ac tu a lly o ccur on sacred ground; in two o f them the
213 Cf. vividly Com. Adesp. 239 A ustin ( = Page, G L P n . 48), fr. 1.12-13, 18-20.
214 I huc. 1.134. I he later king P ausanias w ho, th reatened by a cap ital charge, lived
o u t his life as a hiketês in th e temenos o f A th en a at T eg ea, was a t least o u t o f the country,
P lut. Lys. 30.1.
215 L ycurg. Leoer. 128 f., E ur. fr. 1 0 4 9 (= Oedipus, fr. 98 A ustin), cf. Eur. Heracl. 2591'
216 T a c . Ann. 3.60. (F o r this reason the A thenian acropolis was g uarded to prevent
ru n a w a y slaves achieving san ctu ary , IG I3 45.) T h e co ntrast is not seriously affected by
tw o rid ers th a t cau tio n d em ands: ( 1) ‘dysfunctional’ supplication, though not attested,
m ay well have occu rred a n d caused problem s in the early period; w hat would O dysseus
h ave d o n e if a su ito r h ad reached an altar? (2) it is n atu ral, given the n atu re of the
evidence, th a t only the p roblem atic cases from th e classical period should be recorded.
217 T h u c . 1.126.2-135.1.
218 T h e n v iç o f T h u c . 1.126.11.
184 Miasma
v ictim s w ere lu red o u t th ro u g h prom ises an d then killed, while
in th e th ird the su p p lian t, d eb ilitated by starvation, was carried
o u t to die. T h e p re ssu re to avoid d e a th a t the altars was obvi­
ously intense, b u t to have succeeded in doing so was no
g u a ra n te e ag a in st p o llu tio n .219 In two cases, the victim s were
tra ito rs or su b v erte rs o f the constitution and, in th e third,
p erso n s o f th e low est rank. In two o f the cases, again, the state
c o n c ern e d h ad a tte m p te d public purification, while, in the
th ird , the S p a rta n s believed th a t their m u rd er o f helot su p ­
p lia n ts in a sh rin e o f Poseidon h ad caused the great earth q u ak e
o f 462.220 A n o th e r exam ple o f the long-lasting effects o f such a
crim e com es from H e ro d o tu s.221 D u rin g civil strife in A egina in
th e 490s o r 480s, th e rich m u rd ered 700 of the people, one of
w hom w as clu tch in g a tem ple door. T h e A eginetans w ere ‘u n ­
a b le to sacrifice o u t this pollution (agos), b u t w ere expelled from
th e ir island first’ (in 431 ).
I t w as h a rd in norm al w arfare for m en seized by the jo y of
killing222 to acknow ledge an absolute title to m ercy, b u t it was
even h a rd e r in the ex tra o rd in arily em bittered atm osphere of
civil strife, w hen th e su p p lian t sp ared today m ight head a
co u n te r-c o u p tom orrow . V iolations in civil w ar becam e very
c o m m o n .223 T h is is pro b ab ly the context in w hich belongs a
2,9 C a su istic atte m p ts a t evasion w ere none th e less popular: see G ould, op. cit., 82 f.,
a n d for treach ery a d d schol. D em . 1.5 (cf. Plut. Alex. 42.1 ), for fire M enander, Perinthia
1 ff. w ith S a n d b a c h ’s note. A ctual killing on sacred ground is rare even am ong the
v io latio n s listed below . F aced by fire or force the su ppliant will ideally ensure
m ax im u m pollution by stay in g put (Eur. Andr. 2 5 8 -6 0 , Ion. 1255-60), b u t is scarcely to
be c o n d em n ed for n o t d oing so (cf. Plut. Dem. 29. 5: contra, A. P. B urnett, Catastrophe
Survived, O xfo rd , 1971, 160).
220 E xpulsion o f A lcm aeonids, 1.126.12; statu es for A thene Chalkioikos, 1.134.4;
e a rth q u a k e , 1.128.1. A ccording to later trad itio n , P ausanias required m ore elaborate
ap p e a s e m e n t, P lu tarc h , fr. 1 2 6S andbach. 221 6.91.
222 X en. Hieron 2.15 f.
223 V iolation in stasis o r com parable situtions: H d t. 5.46.2, T huc. 3.81.5, X en. Hell.
2.3.52—6 (cf. Lys. 12.96), 4.4.3, 6.5.9, 7.2.6,? A rist. Pol. 1303a 29—31, A rr. Anab.
1.17.12, P lut. Dem. 28.4, cf. 29.5 f., Praec. Reip. Ger. 825b, Q uint. G urt. 10.9.21. In
w arfa re: H d t. 6.79 f., A rr. Anab. 1.8.8 (Diod. 17.13.6), Paus 10.35.3 (contrast Diod.
16.58.4—6); by b a rb a ria n s: H d t. 8.53.2, Diod. 13.90.1. O bservance in warfare: T huc.
3.28.2, X en. Hell. 4.3.20 (cf. Ages. 2.13, 11.1), Diod. 11.92, 14.53.2-3, A rr. Anab. 2.24.5,
cf. D iod. 13.67.7. O n T h u c . 3.75, 81 see G ould, op. cit., 83. For later evidence see
P. D u crey , Le Traitement des prisonniers de guerre dans la Grèce antique, Paris, 1968, 295—300.
M u rd e r a t a lta rs in m ythology: e.g. Apollod. 1.9.8, idem , epit. 2.13, 3.32, 5.21. For
m u rd e r a t th e alta rs cau sin g d isaster in m ythologized history see F ontenrose, 76. It is a
very com m on th em e o f art: see K . Schefold, M H 12 (1955), 138 f., cf. G. Roux, Antike
Kunst 7 (1964),36f.
Sacrilege 185
fam ous in scrip tio n from M a n tin e a o f the fifth century, w hich
excludes from th e shrine o f A th en e Alea a g ro u p o f nam ed
in d iv id u a ls w ho h av e polluted it w ith bloodshed; the b an ex­
ten d s to th e ir d escen d an ts, a n d is to last for ever.224 It w as hard
too for th e victors to respect th e w om en w ho fled to the altars
w h en a city was sacked. T h e L o crian Ajax failed, an d the
co n seq u en ces for him self, his com panions, a n d his descendants
w ere d ire .225
E ven w h en su p p licatio n re tain ed its tru e H om eric form of a
re fu g ee’s p lea for reception, it m ight provoke anxiety an d guilt.
T h e s u p p lia n t h ad an ab so lu te claim not m erely to be done no
h a rm by th e in d iv id u al o r city a t w hose a lta r he sat, b u t also to
receive from th em positive p ro tectio n against his enem ies; the
in v o lu n ta ry host was not free to step back a n d leave the
p u rsu e rs to risk th e consequences o f violation, as it w as his own
a lta r th a t w ould be defiled.226 S uicide at the altar, a final
reso u rce w h ereb y the ‘spare m e’ su p p lian t could b ra n d an
in d elib le p o llu tio n on the enem ies he was otherw ise powerless
to h a rm ,227 w as also available as a th rea t for use by th e ‘help m e’
su p p lia n t. T h is w as the w eapon by w hich the D anaids in
A eschylus forced a n unw illing Pelasgus to accept th eir suppli­
c a tio n a t th e cost o f p ro b ab le w a r.228 Faced by a sim ilar choice,
a h isto rical sta te m ig h t well have responded differently; b u t the
C y m aean s, u n d er pressure from Persia to surrender a suppliant,
w e n t to co n sid erab le lengths to ensure his safety, an d the
C h ia n s, w ho b etra y ed him , treated as defiled th e land they
received as re w a rd .229
D esp ite re p e a te d violations, the sanctity o f the su p p lian t
224 S chw yzer 661, S o im sen/F raenkel4 5, Buck 17, cf. Latte, H R 4 5 -7 . C reusa, who
a tte m p te d to m u rd e r a tem ple servant on sacred ground, was threaten ed w ith stoning,
E u r. Ion. 1 112, 1237.
225 /lieu Persis, O C T H o m er v, p. 108.2-6, cf. already Od. 3 .1 3 4 -5 , 4.499-511;
A lcaeu s, S L G 262, w ith H. L loyd-Jones, G R B S 9 (1968), 137; on the L ocrian m aidens,
F. G raf, S S R 2 (1978), 6 1 -7 9 . For rap e in a tem ple duly punished by stoning see Paus.
8.5.12.
226 H d t. 3.47.3, E u r. Heracl. 255 f.
227 T h u c . 3.81.3.
228 Supp. 4 5 9 -7 9 , cf. H d t. 7.141.2, E ur. Hel. 985—7. O n m orally coercive suicide cf.
G ern et, Anthropologie, 297 f. Are the corpses hanging in tem ples th at occasionally
a p p e a r, w ith o u t ex p lan atio n , in the sources (532 FGrH D (2), p. 513,? LSCG 154 B
33—6) a c tu al cases of th is ? Cf. too O v. Met. 7.603.
229 H d t. 1.1 5 7 -6 0 . A novella ab o u t successful w ar in defence of suppliants, Diod.
12.9.
186 Miasma
re m a in e d a t the end o f the fourth century a significant con­
s tra in t u p o n ac tio n .230 Some sanctity likewise still clung to
a n o th e r asp e ct o f social relations b ro u g h t directly u n d er divine
p ro te c tio n , th e o a th .231 G ods w itnessed it, sacrificial ritual ac­
co m p a n ie d it, a n d it w as com m only tendered in a sacred place.
U p o n its san c tity d ep e n d ed innum erable' relations, both w ithin
sta te s a n d betw een them : th ro u g h his office as g u ard ia n of
o a th s, Z eus was au to m atically g u ard ia n also of social m orality.
‘T h e o a th is w h a t holds the dem ocracy together. O u r society is
m a d e u p o f th ree p a rts, the m agistrates, the ju ro rs, the private
citizens. E ach one o f these tenders this pledge.’232 T h e hum blest
citizen w as th u s co nstantly forced to choose betw een ‘respect­
in g ’ a n d ‘defiling’ the gods. It is no t surprising th a t perjury
sh o u ld have been the first offence for w hich post-m ortem
p u n ish m e n t is a tte s te d .233 In serious oaths, the sw earer invoked
d e s tru c tio n on h im self a n d his d escendants in case o f perjury,
a n d m ig h t have p re sen t in person the children by w hom he
sw o re .234 T h is th re a t to d escen d an ts is constantly m entioned in
c o n n e ctio n w ith broken o ath s.235 W hen observance o f an oath
w as in to lerab le, a casuistic pseudo-fulfilm ent m ight be attem p ­
te d ra th e r th a n sim ple violation.236 M erry rogues exploited the
in s titu tio n a t every period; a d irect line runs from Autolycus,
a d m ire d in H o m er for u n su rp assed skill in ‘thieving an d oath s’,
230 A sm all in stan c e, on a n everyday level, Plut. Alex. 42.1.
231 L atte , H R 5—47, 96—101, idem , in R E s.v. Meineid (Kl. Sehr., 367—79), Dover,
246—50, B u rk ert, G R 377—82 w ith bibliography, G lotz in D ar.-S ag s.v .Jusjurandum.
232 L ycurg. Leocr. 79. H ow can one have peace w ith the S partans, ask the A charnians
( A r. Ach. 308), οίσιν ο ντε βωμός οντε πίστις ονθ' ορκοςμένει ?
^33 H orn. II. 3.278 f., 19,259 f., cf. Ar. Ran. 150.
234 D estru ctio n o t'genos invoked in oaths: M /L 5 .4 7 -5 1 , IG I3 14.17 (M /L 40), ibid.,
37.55 (M /L 47), S IG 3 3 6 0 .5 2 -5 , Ar. Ran. 587 f., A ndoc. 1-.31, 98.1, 126, A n t.-^ .ll f.,
D em . 2 3 .6 8 ’ 24.151; w ith p resen tatio n o f children, Lys. 32. 13, Dem. 29.26, 54, 54.38.
D e stru c tio n o i'genos in curses: M /L 13.15; 30, passim; Schw yzer 632 A 23, 634 B 46 f.,
S o p h . 4/. 1178, A r. Thesm. 3 4 9 f.r D em . 19.70Γ., Aeschin. 3.11 l.C f. to o A r .Plut. 1102—9.
M u c h m ore in G lotz, 5 7 2 -4 , Pease’s note on C ic. Nat. D. 3.90, R. L attim ore, Themes in
Greek and Latin Epitaphs, U rb a n a , 1962, 112-14, an d cf. SIG 3, index s.v. έξώλης.
235 H orn. II. 3.298—301 ,4 .1 6 0 —2, Hes. Op. 282—5, T heog. 199—208, L ycurg. Leocr. 79.
O n th e p u n ish m e n t o f p erju rers see also H es. Op. 219—24, 8 0 3 -4 , Theog. 231 f.,
E m p ed o cles Β 115.4, A r. Nub. 395—7, X en. Anab. 2.5.7, A ntiphanes, fr. 233, epigram in
Polvb. 4.33.3. O n e person can urge an o th e r to perjury by taking the p u nishm ent on
himself*, A r. Lys. 914 f., P lut. Arist. 25.1. T h e re are o f course o aths and oaths: no one
ex p ects m uch from a lover (H es. fr. 124) or salesm an (H dt. 1.153.1, PI. Leg. 917b).
236 A bove, p. 155 n. 55.
Sacrilege 187
to L y san d e r, w ho ‘ch eated boys w ith knuckle-bones, m en w ith
o a th s ’, w hile th e o rd in ary w ord for p erju ry contains etym ologi-
cally no fu rth e r id ea th an th a t o f ‘o a th ’; b u t the curren cy was
n o t d e v a lu e d .237 T h e fate o f G lau cu s, whose fam ily was ob­
lite ra te d root an d b ran ch alth o u g h he had done no m ore than
c o n su lt the P y th ia ab o u t violating his pledge, m ight serve as a
w a rn in g .238 T h e very A th en ian clubs th a t existed to provide
m u tu a l aid in co u rt cases, w here necessary th ro u g h perjury,
w ere fo u n d ed u p o n o ath s.239 ‘I ’m willing to b rin g m y children
w h e rev er you w ish’, says a lady in Lysias, ‘a n d sw ear the oath.
B ut I ’m n o t so desp erate, or m ad for money, as to com m it
p e rju ry by m y ch ild ren before I d ie .’240 T h e skilled logographer
w ou ld n o t h ave h o p ed to im pose on the ju ry w ith an arg u m en t
th a t w as lu d icro u sly naïve. T h e a rb itra to r in a p riv ate case
m ig h t, if re q u ired to ten d er an o ath , reach a different verdict
th a n he w ould have done w ith o u t it;241 the shadiest of
c h a ra c te rs could be sh am ed o u t o f a deception by the challenge
to a n o a th w h ich w ould b ra n d him publicly as a p erju rer;242 an
e n e m y ’s p erju ry is claim ed to be beneficial, since it secures for
o n e ’s ow n side th e favour o f the gods.243 Perjury influenced the
237 A utolycus: H orn. Od. 19.395 i. H ere too deceitful oaths were no d o u b t involved
r a th e r th an p lain p erju ry ; for the art cf. II. 15.41 w ith schol. A T ad. loc. L ysander: Plut.
Lys. 8 .4 - 5 (the mot w as ascribed to oth ers too, L atte, R E s.v. Meineid, 350 = Kl. Sehr.,
372). W ord for p erjury: M . L.eum ann, Homerische Wörter, Basle, 1950, 79-9 2 .
238 H d t. 6.86 α 2 - ô .
23° οννωμοσίαι, cf. E. Leisi, Der Zeuge im Attischen Recht, diss. Z ürich, 1907, 118 f.,
G . M . C a lh o u n , Athenian Clubs in Politics and Litigation, A ustin, T exas, 1931, 7 7 -8 2 . O n
the ev id en t m en d acity o f m any witnesses cf. Leisi, op. cit., 114 ff., u nfortunately paying
m ore a tten tio n to o ra to rs’ assertions ab o u t o p p o n en ts’ witnesses th an the direct
eviden ce o f the texts. ‘D er grösste F ehler in d er attischen R echtsprechung’, concludes
L eisi, ‘ist d er M angel an E hrlichkeit im C h a rak ter des A theners.’ N ot all evidence,
ex cep t in m u rd e r trials, w as delivered on o ath , Leisi, 5 7 -6 6 , H arriso n , ii, 150-3.
240 Lys. 32.13, cl. D em . 29.26,47.73, 57.22, 57.53. For the o th er side see D em . 54.38,
a n d for b o th sides (A rist.) Rh. Al. 1432a 3 3 - 1432b 4.
241 D em . 29.58, 52.301. (the speaker claim s έμού r if ίχνενμεν ö q k o v ο ν ό έ ν αύτφ ίμελεν,
μ εθ ' όρκον όέ ίσως αν ούκ ήόίκησεν àtà το α ΐτο ϋ ΐόιον). F or the ju ro r’s concern for his oath
cl. A eschin. 3.233: il he votes unjustly b ορκος, δν όμωμοκώς δικάζει, ονμπαρακολονθών
α υ τό ν λυπεί. O a th as a co n strain t on the witness, A nt. 5.11 f.
242 D em . 33.14, cf. p ro b ab ly 59.60, and for the ‘purificatory o ath ’ Horn. II. 23.579 If.,
L a tte , H R 5—28. C o n tra s t A r. Nub. 1232—6. O bviously his loss o f social credibility in the
ev en t o f p erju ry is as im p o rta n t a co n strain t on the sw earer as the fear o f divine anger.
243 X en . Hell. 3.4.11 ( = Ages. 1.13; for A gesilaus’ own firm ness o f o ath , ibid., 3.2);
H orn. II. 4 .2 3 4 -9 .
188 Miasma
fo rtu n e s no t ju s t o f individuals b u t o f nations. F or X enophon,
S p a r ta ’s defeats at the h ands o f T h eb es in the fourth century
w ere p u n ish m e n t for h er seizure o f th a t city in defiance of an
o a th .244 T h e S p a rta n s them selves tho u g h t th at they fared ill in
th e A rc h id a m ia n w ar because they had entered u p o n it in
c o n tra v e n tio n o f the th irty years peace, sanctioned, o f course,
by o a th s .245
A clea r final exam ple o f religious protection for a n exposed
b u t in d isp en sab le social function is th a t o f the h erald .246 His
h e ra ld ’s staff in h a n d , he m oves, inviolable, from state to
s ta te ;247 in clu d ed in an em bassy in tim e o f w ar, he extends his
p ro te c tio n to the o th e r m em b ers.248 W hen the m u rd er o f a
h e ra ld occurs, the indignation o f the offended state is loud,249
b u t th e offenders too m ay in calm er mood recognize the need for
ex p iatio n , w hile pious outsiders will w atch for signs o f divine
a n g e r. T h e w ra th o f T alth y b iu s w hich beset the S p artan s for
th e m u rd e r o f D a riu s’ herald was tem porarily allayed by their
a tte m p t a t ap p e asem en t, b u t ‘woke u p ’ again som e sixty years
a fte r th e crim e, a n d found a paradoxical fulfilm ent w hich
p ro v e d to H e ro d o tu s the divine c h a rac te r of the affair. How the
A th e n ia n s w ere pu n ish ed for the sam e crim e H erodotus is not
ce rta in , b u t h e does not d o u b t th a t divine anger should have
fo u n d som e expression.250
T h is survey has covered a very large nu m b er o f the religious
d a n g e rs to w hich G reek life w as exposed. M ost o f the situations
w h e re H e ro d o tu s detects the o peration o f divine vengeance, or
244 Hell. 5.4.1. T h a t seizure itself, however, punished T hebes for violation of an oath,
Isoc. 14.28.
245 T h u c . 7.18.2.
246 Cf. L. M . W ery, Ant. Class. 35 (1966), 468—86, P. D ucrey, Le Traitement des
prisonniers de guerre dans la Grèce antique, Paris, 1968, 3 0 1 -4 . ■
247 T h u c . 1.53.1 w ith schol., Dem . 51.13, R. Boetzkes, Das Kerykeion, diss. M ünster,
1913. T h e h e ra ld ’s rig h ts m ight be suspended, however, in an ακήρυκτος πόλεμος·, see
J . L. M yres, C R 57 ( 1943), 66 f.
248 H orn. Od. 10.102, T h u c. 2.12.2, 4.118.6. In tim e of peace, am bassadors are
th eo retically inviolable anyw ay by the sta te to w hich they are sent, (D em .) 12.3-4 (for
a n infraction X en . Hell. 5.4.22). O n e explanation for the destruction o f Sybaris was
d iv in e a n g e r over a n offence against heralds, Phylarchus 81 FGrH fr. 45. It is not clear
w h e th e r violence ottered to an em bassy travelling from an enem y to a 3rd p arty breaks
th e rules, T h u c . 2.67, Hell. Oxy. 7.1.
249 P lut. Per. 30.3 (A nthem ocritus, cf. Paus. 1.36.3), (Dem .) 12.2-4.
250 H d t. 7.133—7. P au sa n ias offers an answ er, 3.12.7: T alth y b iu s’ w rath sm ote the
fam ily o f M iltiades.
Sacrilegt 189
w h ere sp eak ers in T h u cy d id es claim th at religious rules have
been vio lated , fall u n d e r one o f these headings. D espite certain
ch an g es o f em p h asis (as in th e case o f supplication), the forms of
religious re sp ect req u ired th ro u g h o u t this long period are in
th e ir b ro a d o u tlin es rem ark ab ly constant. A lexander p erhaps
le a rn t from O d y sseu s the im p o rtan ce o f sparing priests, a n d the
H o m e ric sac red festival w here the bow is not to be draw n
prefig u res th e legal truce o f hellenistic cities. T h e com m on view
th a t fear o f p o llu tio n is virtually unknow n to H o m er is obvi­
ously based on an im plicit exclusion o f the pollution of sacrilege,
w h ich is u b iq u ito u s in him.
All th e p o llu tio n beliefs here discussed m ight be seen for­
m ally as fu n ctio n in g to m ain tain a category distinction, th at
betw een gods a n d m en. T h e sam e category distinction is con­
sta n tly th re a te n e d , an d as co nstantly reaffirm ed, in the m any
m yth o lo g ical stories o f h ard y heroes scaling the heavens,
m o u n tin g th e beds o f goddesses, or challenging the gods to tests
o f s tre n g th o r skill.251 T hese stories are the m ythological corre­
late o f th e rules w ith w hich we have been concerned. It is
som etim es said explicitly th a t categories are confused when
these rules are violated. X erxes, for instance, th rough disre­
sp ect to tem ples ‘p u t sacred a n d p riv ate dwellings on the sam e
level’.252 W e need to ask, how ever, why the m ain ten an ce o f this
p a rtic u la r category distinction was o f such im portance. P opular
feeling a b o u t th e u ltim ate th rea t to it, denial of the gods’ very
existence, is revealing. T h e prosecutions o f intellectuals for
im p ie ty a t A th en s in the late fifth century are an obscure
a r e a ,253 b u t A risto p h an e s’ Clouds provides clear evidence that
a th e ism w as felt to lead directly and necessarily to the dissolu­
tion o f social m orality. T h e institution m ost obviously
th re a te n e d was the o ath , b u t the connection betw een religion
a n d a n o rd e red society w ent m uch further th an this. O rd er
d e p e n d s on aidôs, self-restraint expressed through respect for
recognized values. Aidôs is a q u ality th at you have o r you lack; it
is th e sam e aidôs, o r absence o f it, th a t expresses itself in speech,
d ress, d e p o rtm e n t, sexual behaviour, relations w ith m en, and
251 e.g. H orn. II. 2 .5 9 5 -6 0 0 , 5 .4 0 5 -9 , 2 4 .6 0 2 -9 ; Hcs. fr. 3 0 .1 -2 3 , 51 177 11-12
252 H d t. 8.109.3.
253 See m ost recen tly K. J . D over, Talanta 7 (1975), 24-54.
190 Miasma
a ttitu d e to gods. T h e inviolable m eadow of a god is a fit sym bol
for th e ch a stity o f a virtuous youth, as both are protected by
aidös,254 R espect for the gods entails an ultim ate restrain t in
co n d u c t, a w illingness to stick a t som ething. S paring an en­
e m y ’s sacred places is, like re tu rn in g his dead for burial, a
re co g n itio n o f the m inim um right to respect enjoyed by any
h u m a n 255 (or, a t least, by any G reek; b arb arian s cannot abso­
lu tely rely on such respect). T h e m an, by contrast, w ho flouts
religion despises, in A th en ian eyes, ‘bo th the gods and our
la w s’.256 I f he will engage in a conspiracy to annihilate sacred
im ages, he is unlikely to feel scruples ab o u t subverting the
d em o cracy . T h e su p erio r pow er o f the gods m ust be vindicated
even in m orally n e u tra l areas like th a t of prophecy to keep
society sound. O th erw ise, religion will decay, an d there will be
no fu rth e r m o tiv atio n for ‘reverent p urity in w ord an d d eed’.257
W h e n ju stic e does triu m p h in the w orld, this is confirm ation
th a t th e gods are there. T h e G reek w ho then eagerly exclaims
‘th e gods exist’ is an n o u n cin g m ore th a n a fact ab o u t cate-
g o n es.
254 Kill Hipp. 73—81. Cl. A esch. Ag. 371—2 on (he tram pling o f άθικτων χάρις.
255 T h erefo re a n alleged violation o f sacred places is m et by refusal o f burial, T huc.
4.97. 2 - 9 9 , w hile tem ple-ro b b ery norm ally incurs this punishm ent.
256 See p. 170 n. 146.
257 S oph. 0 7 " 8 6 3 -9 1 0 .
258 M en. Dysc. 639, cf. Aesch. Ag. 1578 w ith Fraenkel.
6
CURSES, FAMILY CURSES, AND
THE STRUCTURE OF RIGHTS
W h e n m o rtals violated the sacred in the directest o f the ways
th a t w ere discu ssed in th e previous ch ap ter, th e consequence
w as th a t ‘a n agos cam e upon th e m .’ Agos is here a spontaneous
a n d a u to m a tic p ro d u c t o f tran sg ressio n .1 As was noted in
th e in tro d u c tio n , how ever, it could also be invoked against
offenders in curses: ‘let an agos com e upon those who have
sw o rn th e o a th should they transgress it.’2 T h o u g h curses often
d e m a n d sim ply th a t the offender should ‘be destroyed him self
a n d his fam ily’, they som etim es specify fam iliar consequences
o f po llu tio n : crop-failure, sterility o f anim als, m onstrous births
o f h u m a n s .3 T h e A m p hictyonie o ath contains a provision of this
kin d , a n d con tin u es ‘A nd m ay they never sacrifice w ithout
offence to A pollo o r A rtem is or . . .’. I t is the im possibility of
sacrificing ‘w ith o u t offence’ th a t, according to A ntiphon, often
in d icates th e presence o f a pollution, a n d th a t revealed to the
S p a rta n s th a t ‘th e w ra th o f T alth y b iu s h ad struck upon th em ’
for th e m u rd e r o f X erx es’ h erald s.4
A lth o u g h th e agos o f sacrilege is in principle au to m atic, while
th a t o f a curse d ep e n d s upon p u blic proclam ation, the distinc­
tion is little m ore th a n a form al one. In cases o f sacrilege, the
d iv in e cu rse w as often su p p o rted by a h u m an one; in 415, after
th e p ro fan atio n o f the m ysteries, ‘th e priests a n d priestesses
sto o d facing w est a n d cursed [the offenders] an d shook their
p u rp le robes, acco rd in g to the ancient custom .’ (O nly one
1 H d t. 6.91.1.
2 ‘P lataea o a th ’: see p. 7. Lines 4 0 - 6 specify the n atu re o f the άγος. T h ere m ay be an
in v o catio n o f agos in the defixio, W ünsch, n. 90, b 6.
1 Soph. O T 2 6 9 -7 2 (w here the connection w ith pollution is explicit); A m phictyonie
o a th . ap. A eschin. 3.111; Eupolis, Demes 31 f., Page, GLP, p. 208 = n.92 A ustin fr.
1 .3 3 -4 . For later ep ig rap h ic evidence see L. R obert, Études épigraphiques el philologiques,
P aris, 1938, 313 w ith n. 3, citing SIG 3 360.55 f„ 5 26.40-7, 527.8 5 -9 0 , Schwyzer
1 9 8 .2 3 -5 .
4 A nt. 5.82, H d t. 7.134.2; cf. LSA 16.25—7, w ith Sokolowski’s note.
192 Miasma
g e n tle lady refused to take p a rt, saying th a t she w as ‘a priestess
o f p ra y ers a n d not o f curses’.)5 It w as the spoken verdict of a
h u m a n trib u n a l or, th ro u g h o ra c u la r consultation, o f a god6
th a t confirm ed the presence o f agos, an d the sacrilegious re­
ceived th e ir m ost lasting ta in t w hen they were ‘w ritten up on
th e p illa r as offenders against the gods’. A nd if the divine curse
a g a in s t sacrilege often had to aw ait h u m an confirm ation in
o rd e r to becom e fully effective, in the m any arch aic G reek
co m m u n itie s w here the m agistrates pronounced curses in a d ­
v an c e a g a in st c e rta in categories o f treacherous behaviour,7 the
offender w as in theory ‘held in the agos’ (the expression comes
from H e ro d o tu s, in this context)8 from the m om ent o f his crime
j u s t as securely as if he h ad robbed a tem ple. As a result o f this
convergence betw een curses th a t are au to m atic an d those that
a re p ro claim ed , it can be difficult in a p artic u la r case to decide
w h ich o f the tw o is in question. T h e passer-by w ho covers a
co rp se p erfu n cto rily w ith soil ‘to escape agos’ m ay either be
av o id in g th e ta in t caused by neglecting a fundam ental divine
law , o r m ore specifically the curse regularly pronounced against
su c h offenders a t A thens by a m em ber of the priestly family of
B o u zy g ai.9
B etw een agos in its tw o forms there is, in fact, a deep sim ilar­
ity. A n y o n e can u tte r a curse, b u t the pow er to curse effectively
is n o rm ally confined to certain categories - kings, parents,
p riests, m ag istrates, a n d the like —w ho represent w hatever in
society m ost d em a n d s rev eren ce.10 H ip p o ly tu s’ ‘I f only m or­
tals could curse the gods’ is a b itte r acknow ledgem ent th a t this
p o w e r is, in fact, d ep e n d en t upon the hierarchy o f a u th o rity .11
5 (L ys.) 6.51, P lut. Ale. 22.5. Purifications, by contrast, were perform ed facing east,
p. 225.
6 A rist. Ath. Pol. 1, D iod. 16.60. 1, Andoc. 1.51, above, p. 185 n. 224.
N o te too C h ry se s’ p ra y e r, H orn. II. 1.37-42.
7 E. Z ie b a rth , Hermes 30 (1895), 57—70; idem , in R E s.v. Fluch; G lotz, 569—76;
R. V allois, B C H 38 ( 1914), 250 -7 1 ; L atte, H R 61 - 8 8 . 8 6.56.
9 S oph. Ant. 256 w ith schol.; on Bouzygean curses see T öpfier, 139, W. Schulze,
Kleine Schriften2, G ö ttin g en , 1966, 191.
10 R. V allois, op. c i t.—an im p o rtan t article; cf. D ouglas, 127.
11 E u r. Hipp. 1415; for the distinctive construction o f araios w ith dative of dis­
a d v a n ta g e cf. E ur. Med. 608, PI. Leg. 931c (em pow ered to curse), Aesch. Ag. 237, Soph,
fr. 399, E ur. / 7 ’ 778 (w orking h arm through a curse). In Soph. Tr. 1201 f. (H eracles to
H y llu s); εΐ ôè μή, μενώ σ'ιέγώ /καϊ νέρθεν ών άραιος είσαεϊ βαρύς, the w ord άραίος seems
a c tu a lly to have becom e a noun, ‘curse-dem on’. O n the word cf. W . H. P. H atch , H SCP
19 (1908), 1 5 7 -8 6 .
Curses, Family Curses, and the Structure o f Rights 193
T h e re is th u s a clear sim ilarity betw een the agos th a t seizes the
sacrilegious a n d th e curse p ro n o u n ced against those who vio­
late w h a te v er is socially ‘sac red ’. T o som e extent social sanctity
even h as s u p e rn a tu ra l forces w orking autom atically in its
defence; th e E rinyes o f a w ronged father will p ro b ab ly seek
reven g e w ith o u t form al invocation in a curse. For an idea of
th e p o ten tial aw esom eness o f a curse invested w ith th e full
solem n ity o f p u b lic au th o rity , w e can tu rn to th e Oedipus
Tyrannus, w h ere O e d ip u s pro n o u n ces one against th e unknow n
killers o f L aiu s. I t is not th e least o f his torm ents, after the
rev elatio n , th a t he has im posed so terrible a sentence upon
h im self.12
P u b lic curses o f m agistrates w ere aim ed ag ain st behaviour
t h a t d irectly th re a te n e d p u blic w ell-being or order. T h e earliest
a n d m ost fam ous exam ple com es from T eo s;13 we learn from an
in sc rip tio n p e rh a p s o f the early fifth century th a t the m agis­
tra te s w ere re q u ired , th ree tim es a y ear at public festivals, to
invoke d e stru c tio n upon anyone using poisons (or m agic
spells?) a g a in st th e T eian s, interfering w ith the im port o f corn,
re sistin g th e a u th o rity of th e m agistrates(?), co n ducting or
co n d o n in g p iracy ag ain st th e T eian s, b etraying th eir territory,
o r ‘d evising any evil concerning the com m onw ealth o f the
T e ia n s in re sp ect eith er o f th e G reeks or b arb a ria n s’; m agis­
tra te s a b u sin g th eir au th o rity w ere probably also included in
th ese curses, w hich extended in each case to the family o f the
offender. B oth in its inclusions an d its om issions (theft, m urder,
arso n , ad u ltery , a n d the like) th e T eian inscription is typical of
th e in stitu tio n ; at S p a rta subversion o f regal privilege, at
A th e n s seeking o r su p p o rtin g ty ran n y , treatin g w ith the M ede,
b e tra y in g th e city, taking bribes ag ain st the city’s interest,
d eceiv in g th e council an d people, ad u lteratin g the coinage(?),
a n d ex p o rtin g vital foodstuffs w ere subject to curses, while the
citizen s o f th e T a u ric C hersonese bound them selves by oath
(w ith curse san ctio n s) not to com m it a very sim ilar range of
offences.14 In A th en s, a t least, these curses w ere not an assertion
12 2 3 6 -7 5 , 744 I'., 1 381-2. 13 M /L 30.
14 S p arta: H d t. 6.56. A thens: m ain text Ar. Thesm. 332—67, cf. P. J . Rhodes, The
Athenian Boule, O x fo rd , 1972, 37; curses ag ain st food exports (clearly not a p a rt o f the
re g u la r curses before assem bly an d council), Plut. Sol. 24.1. T a u ric Chersonese: S I(?
360. F or the ran g e o f offences countered by curses see esp. L atte, H R 6 8 -7 7 , with much
fu rth e r evidence.
194 Miasma
o f th e m ag istra te s’ a u th o rity b u t a n expression o f the m ood of
th e people, w ho ‘pray ed along’ w ith the heralds who pro­
n o u n ce d th e m .15 T h e sacred pow er whose potential an g er they
exp ressed w as indeed, in this case, society. (It is interesting th at
th e people o f A thens, no less th a n th eir gods, h ad ‘unspeakable’
m ysteries, aporrhêta, p ro tected by ju s t these curses.16)
P a rt o f the p o in t was perh ap s th a t m any o f these offences
w ere p a rtic u la rly h a rd to g u ard against on a hu m an level; but
d e te c tio n w as certain ly not entirely im possible, an d the ques­
tion arises o f w h a t tre a tm e n t from his fellows the m an consigned
to d iv in e p u n ish m e n t m ight receive. U pon the killer o f Laius,
O e d ip u s im poses a form o f excom m unication: ‘L et no one
receive him , o r speak to him , o r m ake him a p a rtn e r in p ray er or
sacrifice to th e gods, or give him lustral w ater, b u t let all thrust
h im from th e m .’ A story in H ero d o tu s has the ty ran t P eriander
u sin g exco m m u n icatio n o f this kind as a punishm ent, a n d there
a re h isto rical instances o f p ublic m alefactors being subjected to
w h a t a p p e a rs a t first sight to be a spontaneous social ostracism ,
b u t co u ld be a survival o f a m ore form al earlier in stitu tio n .17 It
seem s unlikely, how ever, th a t the seething public indignation
w o u ld alw ays h av e been satisfied to express itself in so negative
a form . T h e old A ttic law ag ain st ty ran n y m ade the offender
atimos in the a rc h a ic sense, an outlaw to be killed w ith im­
p u n ity ,18 a n d it is h a rd to see w h at objection there could be to
killing an y b o d y ag a in st w hom the curse ‘let him p erish ’ had
b een publicly p ro n o u n ced . W e are dealing, in fact, w ith ju s t the
k in d o f offence w hich was liable to provoke particu larly violent
form s o f p o p u la r revenge - destru ctio n o f the h o u se,19 stoning,20
15 A r. TKesm. 3 3 1, 352.
16 L ys. 31.31, A r. Thesm. 363, cf. S IG 3 360.26.
17 S oph. O T 2 3 8 -4 1 ; H d t. 3 .5 1 .2 -5 2 .6 ; H d t. 7.231, Lys. 13.79, X en. Hell. 1.7.35,
D em . 25.61 (cf. D in arch u s 2.9); cf. W . Schulze, loc. cit. (p. 192 η. 9), also PI. Leg. 881
d —e, X en . Lac. Pol. 9.4—6.
18 A rist. Ath. Pol. 16.10, discussed m ost recently by M. H . H ansen, Apagoge, Endeixis
and Ephegesis against Kakourgoi, Atimoi and Pheugontes, O d en se, 1976, 7 5-80.
19 M /L 1 3 .9 -1 4 (proposal to reassign land; m u rd er); H d t. 6.72.2, T huc. 5.63.2
( S p a rta n kings w ho failed as generals through suspected corruption); A r. Nub. 1484
(sacrileg io u s teachings); K ra te ro s 342 F G rH fr. 17, (Plut.) X Oral. 834 a (Phrynichus
a n d A n tip h o n , betray al); Nie. D am . 90 FG rH fr. 60 (K ypselids, tyranny); D iod. 12.78.5
(A rgive gen erals).
20 e.g. H d t. 5.38 (ty ra n t), 9.5.2 (treacherous proposals, cf. Ar. Ach. 2 0 4 -3 6 , Lycurg.
Leoc. 71), T h u c . 5.60.6 (general w ho fails to press hom e ad v an tag e), X en. Anab. 5.7.2
(g en eral w ho ‘deceives’ tro o p s), 6.6.7 (‘tra ito r’), Diod. 13.87.5, 91.3, PI. Ep. 7.354d
Curses, Family Curses, and the Structure o f Rights 195
ex p u lsio n o f th e corpse u n b u rie d .21 A lready in the Iliad, H ecto r
tells P aris th a t he deserves ‘to have a stone tunic p u t on h im ’ for
th e affliction h e has b ro u g h t on his h o m elan d .22 It w ould be
ra sh to assu m e th a t before the in stitu tio n of special form s o f
p ro c e d u re —a t A th en s, eisangelia d ealt w ith such cases23 —the
c rim in a l w as sim ply left to th e gods to punish. A recently-
p u b lish e d frag m en t o f the T eia n curses seems to show th at
th e re , a t least, th e cu rse could entail o u tlaw ry .233
It is clear th a t, th o u g h an offender o f this kind m ay form ally
b e ‘a c c u rse d ’ o r ‘in th e agos\ the im p o rta n t fact ab o u t him is not
th a t h e is a source o f religious dan g er. T h e th rea t he poses is on a
se c u la r level - he pollutes not the gods b u t the constitu tio n 24 —
a n d th ere is no q u estio n , as th ere can be in cases of m u rd er or
even in v o lu n ta ry sacrilege,25 o f his being avoided th rough u n ­
ease ab o u t s u p e rn a tu ra l consequences even by those sym ­
p a th e tic a lly disp o sed to him m orally. As a possibility, before
d e te c tio n , he is certain ly feared intensely, b u t once caught, the
feeling he provokes is in d ig n an t rage. T h e sam e distinction, as
w e h av e seen, also applies to a m ild er form of social rejection;
th o u g h th e m u rd e re r m ay be d eb a rre d from the agora and
sa c re d places to p ro tec t them from pollution, in the case o f the
m ale p ro stitu te o r m an w ho has ‘throw n aw ay his shield’,
exclusion is sim ply a m ark o f disgrace, an d the only pollution
his fellow citizens w ould suffer th ro u g h his presence is the social
o n e o f m ixing w ith a rogue.26 (O ffenders o f this kind were in
m a n y a rch aic com m unities subjected to h u m iliating p u n ish ­
m en ts ra th e r like th e stocks; these involved a ‘ta in t’, b u t it was
( ‘tre a c h e ro u s’ g en erals), Plut. Sol. 12.1 (asp iran ts to tyranny); see further the scholars
cited by F ehling, 63 n. 258, an d to r stoning o f leaders ibid., n. 262.
21 A bove, p. 45 n. 47. 22 Horn. II. 3.56 f.
23 H ypereid es, Euxen. 7 - 8 , 29; cf. m ost recently M acD ow ell, Law, 183-6; P. J .
R h o d e s,./ / / 5 9 9 (1979), 103-14.
2M Chiron 11 ( 1981 ), p. 7 face (b), 5—9; cf. SE G xxvi 1306. 2 3 -6 (partially vindicating
G lo tz, 465, ag ain st L atte , H R 69, n. 21 ).
24 X e n . Hell. 2.3.23, 26, 51. T h e verb used is λυμαίνομαι, which is referred by lexica
n o t to λύμα (pollution) b u t λύμη (outrageous injury); though this is generally correct (cf.
th e figura etymologica in E u r. Hel. 1099), it seem s likely th at in m any contexts Greeks will
also h a v e h eard λύμα in the w ord (note e.g. E ur. Bacch. 354, Hipp. 1068, o f adultery;
X en . Hell. 7.5.18, a stain ed rep u tatio n ; and above all A r. Eq. 1284, im p u re sexual
p leasu res, also th e sem an tic interference betw een λύμα and λύμη them selves, L S J s.w .).
F o r th e G re e k ’s sense o f being u n d er th reat in secular term s this is a crucial w ord-
gro u p .
25 e.g. S oph. OC 4 9 0 - 2 , E u r. /T 9 4 9 - 5 7 .
26 A bove, pp. 9 4 -6 .
196 Miasma
u p o n th e victim s th a t it fell.27) O n the o ther h an d , it is probably
tru e even in cases o f sacrilege th a t the prim ary public response
is o n e o f rage ra th e r th a n o f fear. T h e form s o f m ob ju stic e that
w ere m en tio n ed e a rlier —stoning a n d the like - w ere all applied
to sacrilege as well as treach ery ,28 b u t not, in the m ain, to other
categ o ries o f offence,29 a n d seem to testify to sim ilar feelings in
th e tw o cases. S to n in g the sacrilegious m ay have been a m eans
o f a v e rtin g divine w ra th ,30 b u t no one ever cast a stone in a
m erely p ru d e n tia l spirit.
T h e do m estic correlate to the cursing pow er o f the king or
m a g is tra te is, o f course, th a t o f the father. ‘A p a re n t can curse a
ch ild m o re effectively th a n anyone can do it to anybody else,
q u ite rig h tly ’, says Plato, a n d as instances o f curses which
ev ery o n e agrees to have been fulfilled he cites those o f O edipus
a g a in s t his sons, A m y n to r ag a in st Phoenix, a n d Theseus
a g a in st H ip p o ly tu s.31 Such curses are, in the epic, adm inistered
b y th e E rin y es,32 w ho are g u ard ia n s o f the stru ctu re o f family
a u th o rity (y o u n g er sons norm ally have no E rinyes);33 a m other
c a n ‘curse E rin y es’ ag a in st h er son, a n d it is as ‘curses’ th at they
d e sc rib e them selves w hen form ally asked their identity by
A th e n a in A eschylus.34 T hese are m ythological conceptions too
elev a ted for everyday speech,35 a t least in classical A thens; the
v ^ lu e to w hich they relate, how ever, is fundam ental, as is clear
in p a rtic u la r from A risto p h an es’ p o rtray al o f m oral an arch y in
th e Clouds.36 In its defence, P lato organizes sanctions w hich take
us rig h t b ack in to the sphere o f public curses a n d outlaw ry.
A n y o n e failing to p ro tec t a p a re n t from assault by a child is
27 A bove, p. 95 n. 87. ‘T a in t’: X en. Hell. 3.1.9.
28 Cf. p. 45 n. 47, p. 194 n. 19; stoning o f the sacrilegious, Fehling, 63 n. 260. Sam e law
lor tem p le-ro b b ers a n d traito rs, X en. Hell. 1.7.22.
29 H o u se-d estru c tio n for m u rd er am ong the L ocrians (M /L 13.13) is an exception.
T y ra n ts a n d defective leaders are often stoned; stoning for o ther categories o f offence is
som etim es envisaged, b u t ‘so g u t wie nie a n tik un d historisch’, Fehling, 63.
30 A lcaeus, S L G 262.
31 Leg. 9 3 1 b - c .
32 e.g. H om . 11. 9.454, 5 6 6 -7 2 .
33 11. 15.204.
34 H om . Od. 2.135 f. (note the fear it inspires); Aesch. Eum. 4 1 7 .11 is as curses relating
to rig h ts th a t they a re constantly constructed w ith a genitive o f the w ronged party: cf.
E. R o h d e, Kleine Schriften, T ü b in g e n and L eipzig, 1901, ii, 2 3 3 -5 , w ith the qualification
o f D o d d s, 21 n. 37.
35 o f ' άγορενεις/άρά,ς τε οτιτ/εράς καί Έριννας, Ap. Rhod .3 .7 1 0 —1.
36 1 321-450.
Curses, Family Curses, and the Structure o f Rights 197
‘h e ld in a cu rse o f Z eu s o f k in d re d ’, a n d the m an convicted of
su ch assa u lt is to be b an ish ed to th e co untryside a n d excluded
from th e sh rin es for ever; any free m an w ho as m uch as speaks to
him m ay n o t e n te r city, sh rin es o r m arket-place w ith o u t being
p u rifie d .37 (S eco n d ary though it ap p e ars, the contagious im­
p u rity o f the m o ral leper here receives from Plato ch aracteristic
em p h asis.) P lato goes to extrem es, b u t u n d er A ttic law convic­
tion for m a ltre a tm e n t o f p a re n ts entailed atimia, a kind of
m itig a te d o u tlaw ry .38 Even P lato ’s uncom fortable im age of
a g e d p a re n ts as ‘living sh rin es’ is reflected in the claim th a t they
sh o u ld receive h o n o u rs ‘eq u al to those o f the gods’.39 D isrespect
to th e m is sacrilege, a p o llu tio n ,40 an d d an g er atten d s upon it.
F e a r o f a p a re n ta l curse is, in the epic, a real con strain t upon
ac tio n , a n d th e occu rren ce o f one is a d ire event w hich m ay lead
to a d ra s tic re actio n .41 I f less is h ea rd of it, outside a m ythologi­
cal co n tex t, in th e fifth and fourth centuries, th at m ust in large
p a r t be becau se th e rights o f p a re n ts had received such effective
p ro tec tio n in sec u la r law.
T h e curses considered so far have su p p o rted the stru ctu re of
a u th o rity , a n d this is their m ost characteristic function. It is,
how ever, to rig h ts ra th e r th an raw pow er th a t they relate, an d if
th e y com m only consort w ith a u th o rity th a t is because the rights
o f co m m u n ities a n d p aren ts are in fact very extensive. Even the
stro n g ca n p e rh a p s not curse effectively unless w ronged, while
th e w eak a c q u ire th e pow er to do so in so far as th eir recognized
rig h ts a re infringed. T h e disguised O dysseus can suggest, ten ta­
tively it is tru e, th a t ‘beggars have E rinyes’;42 the m yth o f the
h o u se o f T a n ta lu s shows à ch ario teer an d a younger bro th er
im p o sin g effective curses, an d a d au g h ter w ith E rinyes;43 ‘even
dogs h ave E rin y es’, the p roverb says (they are, after all,
m em b ers o f th e h o u seh o ld ).44 E u rip id es’ M edea not merely
u tte rs curses ag a in st C reousa, b u t in a m ore serious sense ‘is’ a
37 Leg. 8 8 1 b -e .
38 A eschin. 1.28.
39 Leg. 869b, 931a, A eschin. 1.28.
40 C f. (A eschin.) Epistle 2.5.
41 H om . Od. 2.135 f.; 11. 9.454 IÏ., 566 11'.
42 H o m . Od. 17.475.
43 A esch. /4^. 1433, cf. ibid., 237, E ur. Med. 1389; on the rights protected by Erinyes
see E. W ü st, R E Sup p l. 8.1161'.
44 M a c a riu s 3.54.
198 Miasma
cu rse ag a in st J a s o n , w ho has w ronged h er m ore deeply.45 This
is, in th eo ry a t least, the difference betw een the curse and the
b in d in g spell; th e form er has its ow n intrinsic pow er, while the
la tte r, a n a c t o f aggression u n su p p o rted by right, needs
re in fo rcem en t th ro u g h m agical techniques, the im purity of the
g rav e, a n d invocation o f infernal pow ers. (In practice, no
d o u b t, those w ho believed them selves w ronged often had
re so u rce to defixiones as well as m ere curses.46) T his pow er o f the
w ro n g e d to curse effectively relates to the m ore general way in
w h ich the w orld som etim es operates to redress injustice. In
H e ro d o tu s, in p a rtic u la r, p u n ish m en t often com es upon indi­
v id u als for violent acts th a t are not affronts to the gods in any
d ire c t sense.47 B ut though the possibility exists, it is noticeable
th a t stories o f the ‘w ronged w idow ’s curse’ type are not at all
co m m o n in G reece. T h e S p artan defeat a t L eu ctra in 371 gave
rise to one fam ous instance; they lost, it was said, because o f a
cu rse im posed on them centuries before by one Skedasos of
L e u c tra , w hose d a u g h te rs died after being rap ed by passing
S p a rta n y o u th s a n d w ho then him self com m itted suicide over
th e ir to m b .48 I n this case, how ever, it w as obviously the exist­
e n ce o f a to m b o f ‘th e L eu ctrian m aid en s’ at the site o f the battle
th a t d e te rm in e d the form o f the story. O ne reason for the
sc a rc ity o f stories o f this kind m ay be th a t they tend to be
su b su m e d u n d e r th e ‘w ronged s u p p lia n t’ type; b u t the fact
p ro b a b ly also in d icates som ething ab o u t the general ethos of
G re ek cu ltu re.
It is n a tu ra l to consider, in connection w ith curses, the doctrine
o f in h e rite d fam ily guilt.49 Several interrelated ideas need to be
45 E u r. M ed. 607 f.
46 See W ü n sch , n n . 98, 102, 158 for the claim by the a u th o r o f a defixio to have been
w ronged. O b jects, by the sam e title, can try to curse those w ho steal o r violate them
(S ch w y zer 272; IG X IV 865). O f th e tom b-curse, however, I know no explicit early
in sta n c e (Schw yzer 272 need not be one, L. H . Jeffery, The Local Scripts o f Archaic Greece,
O x fo rd , 1961, 348).
47 C f .J . E. Pow ell, A Lexicon to Herodotus, C am bridge, 1938, s.v. τίσις.
48 F u llest version (P lu t.) Am . Narr. 773c-774d, already known to X en. Hell. 6.4.7. Cf.
F o n ten ro se, 147 f., B u rk ert, S H 74. Suicide here, as often, increases cursing power. For
s im ila r stories see p. 107 above (the regent P ausanias); Plut. Quaest. Graec. 12,293d-f
(C h a rila ).
49 Cf. G lotz, 560—83; J . T . K akridis, Ά ρ α ί, A thens, 1929, 141—68 (w ith the com m ent
o f R. V allois, R E A 34 (1932), 98 f.), D odds, C h . 2.
Curses, Family Curses, and the Structure o f Rights 199
m e n tio n e d , n o t all o f them involving E rinyes a n d curses, w hich
te n d to sh ad e in to one a n o th e r even though they are p erh ap s
th eo re tic ally sep arab le. T h e first a n d com m onest is the fam ous
d o c trin e o f Solon a n d m any later m oralists: sooner o r later Zeus
p u n ish e s all w ro ngdoers, a n d if they escape them selves, ‘their
in n o c e n t ch ild ren p ay for th eir deeds, or th eir d escendants
a fte rw a rd s .’50 P erju ry is th e offence m ost com m only punished
in this w ay, b u t a n y o th er m ig h t be; the m oderates in the
A rc a d ia n league, for in stance, d ecided not to touch the sacred
tre a s u re a t O ly m p ia ‘lest w e leave th e gods a co m plaint against
o u r c h ild re n ’.51 A slightly different tone is in tro d u ced w hen it is
specified th a t th e an cestral crim e is one o f bloodshed. T h e basic
co n c ep tio n re m a in s the sam e, b u t em phasis shifts from the
im a g e o f th e slow -grinding m ills o f god to th a t o f a pollution
w h ich has ta in te d th e stock.52 In a m uch stro n g er form, this idea
o f th e in te rn a l co rru p tio n o f th e fam ily is -central to the m yths of
th e houses o f L ab d a cu s a n d T a n ta lu s .53 In co n trast to the
p re c e d in g cases, it seem s essential here th at the crim es o f the
p a re n ts a re violations o f the o rd e r o f the family, an d lead to
sim ila r vio latio n s o n th e p a rt o f th eir children.54 B oth m yths in
th e ir m ost e x ten d e d form do in d eed begin w ith acts o f violence
a g a in s t o u tsid e rs,55 b u t b oth in th eir cen tral an d earliest-
a tte s te d core p o rtra y a fam ily th a t, th ro u g h the m ost m anifold
perv ersio n s, is g n aw in g o u t its ow n heart. T h e im plicit logic is
su g g ested by P in d a r’s su m m ary o f the m yth o f O edipus: the
E rin y s, seeing O e d ip u s slay his father, proceeded to ‘slay his
50 S olon 1 3 .2 5 -3 2 . In respect o f o ath s cf. pp. 186 f: above, and m ore generally Theog.
731—52, A esch. Eum. 934, E ur. fr. 980, Lys. fr. 53 T h alh eim (5 G ern et), (Lys.) 6.20,
Isocr. Bus. 25, D o d d s, 33 f., D over, 260; specific instances will follow.
51 X en . Hell. 7.4.34.
52 μιαιψ όνον η σύγγονον/παλαιώ ν προγεννητόρων, E ur. H ipp. 1379 f.; cf. Aesch. Supp.
265, παλαιώ ν αιμάτω ν μιάσματα. F or p ollution language in reference to past kin-killing
see e.g. A esch. Ag. 1460, Cho. 649 f.
53 M a in texts o n in h erited guilt or curse: A esch. Sept. 6 5 3 -5 , 6 9 9 -7 0 1 , 7 2 0 -9 1 , Ag.
1 0 9 0 -7 , 1 1 8 6 -9 7 , 1309, 1338-42, 1460, 1 4 6 8 -8 8 ,1 4 9 7 -1 5 1 2 , 1565-76, 1600-2; Soph.
E l. 5 0 4 -1 5 , Ant. 5 8 3 -6 0 3 , OC 3 6 7 -7 0 , 9 6 4 -5 , 1299, E ur. El. 6 9 9 -7 4 6 , 1306f., I T
1 8 6 -2 0 2 , 987 f., Or. 8 1 1 -1 8 , 9 8 5 -1 0 1 2 , 1 5 4 6 -8 , Phoen. 3 7 9 -8 2 , 8 6 7 -8 8 , 1556-9,
1 5 9 2 -4 , 1611.
54 Cf. PI. Leg. 8 7 2 e -7 3 a , cf. 729c.
55 Pelops an d M yrtilus: attested in Soph. El. 5 0 4 -1 5 , b u t excluded in Aesch. Ag.
1 192f. L atu s a n d C h ry sip p u s: n o t in A esch. Sept. 742 ff. T h e origin o f this m otifis quite
u n c e rta in : see L loyd-Jones, 119—21, an d ag ain st D eu b n er’s analysis of Peisander, 16
F G rH fr. 10, M . D elco u rt, Oedipe ou la légende du conquérant, Liège, 1944, xii ff.
200 Miasma
sons by m u tu a l sla u g h te r’.56 W ith this conception o f the family
crim e th a t leads au to m atically to fresh crim e is constantly
in te rtw in e d the id ea o f the actu al spoken curse w hich brings
d e sc e n d a n ts to h arm . Im p recatio n s ag a in st their ow n kin were
u tte re d by O e d ip u s, T hyestes, and, in one variant, Pelops,57
a n d , in th e ex ten d ed form s o f the legends, the T a n ta lid and
L a b d a c id woes w e n t back to curses by the outsiders, M yrtilus
a n d P elops.58 S uch a curse seem s m erely to express in words
w h a t p o llu tio n w ould have achieved anyw ay in its ow n in­
a rtic u la te w ay, a n d it can be difficult, though scarcely im­
p o rta n t, to d ecide w h eth er the alastores an d Erinyes referred to
in a p a rtic u la r passage are spontaneous products o f transgres­
sion, o r d u e to a spoken curse.
P o stp o n e d p u n ish m e n t o f the kind envisaged by Solon, seen
by som e as p a rtic u la rly ‘divine’, w as criticized by others as
m o rally re p u g n a n t.59 C ertain ly there w as nothing quite like it in
h u m a n ju stic e, by w hich sons m ight be punished w ith their
fa th e rs b u t n o t n orm ally instead o f them . T h e conception on
w h ich th e tragedies are based, how ever, seem s to be one of
g re a te r m oral subtlety. W hen the sm itten H eracles recalls th at
his fa th e r m a rrie d th e d a u g h te r o f a m an he h ad killed, and
c o m m en ts ‘w h en the foundations o f a house are ill laid, the
d e sc e n d a n ts are b o u n d to suffer’,60 his proposition has an
o b v io u s p lau sib ility in term s w hich are not m erely those of
p o llu tio n , o r divine an g er, w aking up late to sm ite the innocent
in th e second g en eratio n . A gam em non a n d A egisthus are not
in n o c e n t victim s, a n y m ore th an the Polyneices a n d Eteocles of
S ophocles; even A ntigone is a savage d a u g h te r o f a savage
s ire ,61 a n d it is in C ly ta em n estra th a t the curse o f the Pelopid
lin e finds em b o d im en t. ‘A godless act breeds m ore such after,
tru e to its ow n ty p e .’ It is th ro u g h h u m a n sin an d folly, ‘m a d ­
ness in re aso n in g a n d a n Erinys o f the m in d ’, th a t the house’s
56 01. 2 .3 8 -4 2 .
57 H ellan icu s 4 F G rH i'r. 157, cf. Heldensage, 217.
58 M y rtilu s: A pollod. Epit. 2.8. Pelops: B yzantine hypothesis to Aesch. Sept., in
Aeschyli Tragoediae Superstites, ed. VV. D indorf, O xford, 1851, vol. iii, 297.
59 H d t. 7.137.2, T h eo g . 7 3 1 -4 2 ; cf. D odds, 33 f.
60 E u r. H F 1258-62.
61 S oph. Ant. 471 f. F or the p a re n ts’ m oral deficiencies reappearing in the child cf.
E u r. Hipp. 337—43. F or G reek views on m oral inheritance (by no m eans uniform ), see
D over, 83—95.
Curses, Family Curses, and the Structure o f Rights 201
trag ic d estin y is w orked o u t, n o t in a series o f external afflictions
b ese ttin g th e in n o c e n t.62 Even w hen one o f the agents is in fact,
like O re ste s, in n o cen t, it is a com pulsion created by p ast crim es
th a t driv es him to his terrible act. W e see here th e special
c h a ra c te r o f th e fam ily crim e, for w hich rem edy m u st be sought
‘n o t from o u tsid e, b u t from them selves, th ro u g h savage bloody
co n flict’. In these circu m stan ces it is n o t surprising to find the
d o c trin e o f d u a l m o tiv atio n becom ing explicit. ‘T h a t you are
in n o c e n t o f th is m u rd e r, w ho will b ear witness? B ut the dem on
o f th e race m ig h t be a n accom plice.’63
I t is so m etim es suggested th a t th e id ea o f in h erited guilt, in
w h a te v e r form , is a p o st-H o m eric developm ent, a p ro d u c t of
D e lp h ic teac h in g o r o f a creeping sense of g u ilt.64 D ivine re­
v en g e a g a in st th e w hole family, how ever, is certainly attested in
H o m e r, j u s t w h ere o ne w ould expect it, in connection w ith
o a th s. Z eu s p u n ish es perju ry in th e end, if not at once, an d
o ffenders ‘p ay for it a t a high cost, w ith th eir ow n heads, their
w ives a n d c h ild re n ’.65 It is tru e th a t w h a t is envisaged here is a
d e la y e d re ck o n in g strik in g b o th th e crim inal a n d his family, not
th e co m p lete p o stp o n em en t o f p u n ish m en t to the guilty m a n ’s
ch ild re n ; b u t it is h ard to see how anyone w ho accepted
th e fo rm e r p o ssibility w ould b e offended by the latter. It is,
ce rta in ly , p lau sib le th a t the belief in delayed p u n ish m en t
h a rd e n e d so m ew h a t in th e arch aic age, the period th a t saw the
d e v e lo p m e n t o f th e O rp h ic d o ctrin e o f inherited guilt. W here a
H o m e ric G reek, faced by u n acco u n tab le m isfortune, concluded
‘I m u s t be h a te d by Z eu s’, o r ‘I m u st have com m itted some
offence ag a in st th e g o d s’,66 one o f the fifth century m ight ra th e r
th in k o f som e u n d efin ed an cestral fault: ‘Such was the will o f the
gods; p e rh a p s they w ere angry w ith m y family from of o ld .’67 It
is n o t clear th a t such a change o f em phasis is o f any great
im p o rta n c e . U n co m fo rtab le though the doctrine o f in h erited
g u ilt a p p e a rs to us, an xiety is not necessarily its origin. It
62 A esch. Ag. 7 5 8 -6 0 , S oph. Ant. 603.
65 A esch. Cho. 4 7 2 - 4 , Ag. 1505-8.
64 e.g. K ak rid is, op. cit., 141, D odds, 36.
65II. 4 .1 6 0 —2, cl. 3.300 I., Hes. Op. 282—5 (the latter very close to the Solonian
d o c trin e ). F or affliction o f w hole families see II. 6. 2 0 0 -5 , Od. 2 0 .6 6 -7 8 ; the H om eric
Z eu s ca n h a te a w hole fam ily, Od. 11. 436.
66 H om . II. 21.83, Od. 4.377 f.; cf. still H d t. 6.12.3, M en. Asp. 215.
67 S oph. OC 964 f., cf. E u r. Hipp. 8 3 1 -3 , 1379-81.
202 Miasma
p ro te c ts th e belief in divine ju stic e ag a in st crude em pirical
re fu ta tio n ; for the sam e reason, perjury, typical cause o f in­
h e rite d p u n is h m e n t in later texts, is alread y liable to post­
m o rte m p u n ish m e n t in H o m er.68 T h o u g h in some contexts it
a p p e a re d u n ju st, in o th ers it could vindicate the gods: C roesus,
d e p riv e d o f his kingdom d espite rich offerings to D elphi, was
m erely being asked to h a n d back, after a generous period of
u su fru c t, w h a t his a n c e sto r h ad w rongfully acq u ired .69 Poets
a n d h isto rian s m ig h t devise an cestral offences as a kind of
e x p la n a to ry h y pothesis to im pose p a tte rn on d isp ara te events;
th u s H e le n a n d C ly ta em n estra bo th b etray ed their h usbands
b e c a u se th e ir fa th er, T y n d a re u s, h ad om itted a sacrifice to
A p h ro d ite .70 T h e d o c trin e w as p erh ap s not even an im p o rtan t
so u rc e o f anxiety. In n o c e n t suffering w as a fact of experience
w h ich m ig h t be exp lain ed in term s o f in h erited guilt, b u t this
n eed n o t m ean th a t, w hen not afflicted, the innocent lived in
fear. W h en th e rich A th en ian is p ersu ad ed by an O rp h e u s ­
in itia to r’ to p ro tec t him self from the consequences o f ancestral
sin by sacrifice,71 this is perh ap s sim ply a transposition of
sacrifices he m ig h t anyw ay perform ‘for good luck’.
T h e in h e rite d g u ilt o f tow ns an d com m unities was perh ap s a
m o re serious p reo ccu p atio n . O ften, o f course, it was the actual
o c c u rre n c e o f d isa ste r th a t provoked the pious to look for an
a n c e s tra l crim e to ex p lain it; m ost obviously th a t is true, as we
saw , o f the S p a rta n defeat a t L euctra. W e do not know how
seriously, before th e d isaster o f 431, the A eginetans h ad the
sacrileg io u s deeds o f the 490s o r 480s on their m inds.72 T he
o b lig a tio n accep ted by the L ocrians to pay ‘m a id en -trib u te’ for
a th o u s a n d years in expiation o f A jax’s crim e long seem ed
s p e c ta c u la r evidence for a com m unal sense o f guilt; it has
re c e n tly b een a rg u ed convincingly th a t the institution
o rig in a te d in tem ple service o f a fam iliar kind, a n d only
a c q u ire d its special ch a ra c te r by a process of secondary
re in te rp re ta tio n , a n d p erh ap s m ere m isu n d erstan d in g by out-
“ II. 3.278 f.
69 H d t. 1.91.
™ S tesichorus, fr. 223 Page.
71 PI. Resp. 364c.
72 p. 184 above.
Curses, Family Curses, and the Structure o f Rights 203
sid e rs .73 O n th e o th e r h an d , an cestral g u ilt clearly influenced
th e a c tu a l b eh a v io u r o f the A th en ian s w hen they expelled the
D e lian s from th e ir islan d in th e belief th at they h ad been
‘c o n se cra te d a lth o u g h they w ere im p u re because o f some
a n c ie n t offence’.74 I f T h u cy d id es accep ted th a t A th en ian m oti­
v a tio n in this case w as religious, it is not for us to disagree; b u t it
is in te re stin g to note th a t an cestral like o th er guilt is easier to see
fro m o u tsid e th a n from w ithin.
A fu rth e r d ev elo p m en t in th e arch aic period, it has been
su g g ested , is th e ten d en cy to link to g eth er originally d istinct
m y th s to form th e ch a rac te ristic trag ic vision o f a family o r race
afflicted th ro u g h th ree or four g en eratio n s.75 In the Iliad’s
a c c o u n t o f how A g am em n o n received his sceptre, there is no
h in t o f ta in te d stock; th e Cypria first m ad e him a T a n ta lid .76 T h e
ex ten sio n , how ever, o f the O ed ip u s saga into the th ird g en era­
tio n th ro u g h th e ex p ed itio n o f th e Epigoni is already m entioned
in th e Iliad,77 w hile th e crim es o f th e T a n ta lid house involved
m o n stro u s a n d m arvellous elem ents th a t H om er m ight well
h a v e p referred to keep o u t o f sight. Even if such a developm ent
co u ld be d e m o n stra te d , it w ould be h ard to know w h a t conclu­
sio n sh o u ld be d ra w n a b o u t the tem p er o f the age from the fact
t h a t poets d e te c te d this p a tte rn in the fortunes o f tw o m ythical
houses. N oble fam ilies co ntinued to b o ast their descent from the
T a n ta lid o r L a b d a c id line.78
F ew o f th e ideas discussed so far w ould be likely to have m uch
in flu en ce on b eh av io u r, except to the extent th a t individuals
m ig h t be en co u rag ed , or discouraged, in th eir crim es by the
p ro s p e c t o f th e reckoning being p o stponed to their descendants.
T h e y d o not, th a t is to say, isolate a recognizable category o f
73 F . G raf, SS R 2 (1978), 61 —79; differently in details, but not im plication,
F o n te n ro se , 131 —7. S im ilarly in H d t. 7.197 an an cestral sin is invoked to explain a
sin g u la r religious req u irem en t.
74 T h u c . 5.1.
75 F . W eh rli, ‘T y p ologische R ichtungen d e r griechischen S agendichtung', in his
Theoria und Humanitas, Z ü rich , 1972, 7 1 -8 7 ; he refers to K. Schefold’s arg u m en t that
scen es o f vio len t crim e becam e p o p u lar in a rt in the early 6th century, M H 12 (1955)
138 f.
76 II. 2.101—8, Cypria, fr. 11.4 Allen. Even in the 5th century, the sp lendour o f the
P elo p id s can p rev ail o v er th eir sufferings, Pind. 01. 1.89.
77 4 .4 0 4 -1 0 .
78 A lcaeu s, fr. 70.6, P ind. 01. 2.35—47, Islhm. 3.17; on the descent o f the E u p atrid s at
A th e n s from O restes see TöpfTer, 176 f.
204 Miasma
p o llu te d persons, s p ru n g from crim inal ancestors. At A thens,
o n e fam ily did notoriously find itself in this position, b u t it
is a little su rp risin g th a t no specific case besides th a t of the
A lcm aeo n id s can be q u o ted . A considerable n u m b er o f children
w ere, how ever, d eprived o f ‘h o n o u r’ (i.e. in rough term s citizen
rig h ts) b ecause o f th e ir fa th e rs’ offences. In addition to various
specific crim es for w hich this penalty is said to have been
im p o se d ,79 one text p erh ap s im plies in general th a t the children
o f m en p u t to d e a th by th e state becam e atimoi.80 Such heredit­
a ry p u n ish m e n ts could still be im posed in the fourth century, as
w as finally d e m o n stra te d by the discovery of E u crates’ law
a g a in s t ty ra n n y .81 B u t it is clear th a t the ch ild ren ’s loss o f rights
is a c o n tin u a tio n in m itig ated form o f the earlier practice, also
w ell a tte ste d , by w hich they sh ared th eir fath er’s atimia in the
sen se o f o u tlaw ry a n d w ere liable to be killed w ith him if
c a u g h t.82 T h e m ain in ten tio n o f the in stitu tio n is the pru d en tial
a n d p u n itiv e one o f destroying the public offender ‘root and
b r a n c h ’,83 a n d a n y c a th a rtic m otivation is quite secondary. It is
in con n ectio n w ith subversive offences th a t the inherited
p u n is h m e n t is specifically attested (aspiration to tyranny,
b e tra y in g th e city, ac cep tin g bribes for the harm o f the people).
O n ly by inference from the ra th e r doubtful rule m entioned
e a rlie r can it be co n cluded th a t the children o f m en executed for
m u rd e r o r tem p le-ro b b in g becam e atimoi·, this g ran ted , it re­
m a in s possible th a t th e m u rd e re r’s children retained their
rig h ts if he chose to re tire into exile before the verdict.84 In the
Oedipus at Colonus, Ism en e reports th a t Eteocles an d Polyneices
in itially re n o u n ced th e ir claim to the th ro n e o f T hebes because
o f th e ‘co rru p tio n o f th e ir race from o f o ld ’; bu t though their
s u b s e q u e n t ch ange o f h e a rt was im pious, the specific point th at
th ey w ere d isqualified for kingship by pollution does no t receive
e m p h a sis.85
79 M . H . H an sen , Apagoge, Endeixis and Ephegesis against Kakourgoi, Atimoi, and
Pheugontes, O d en se, 1976, 71 an d 73.
80 D em . 25.30.
81 5 £ G x ii 87, cf. H an sen , op. cit., 72.
82 See H a n se n , op. cit., 7 5 -8 0 .
83 C f. F raenkel on A esch. Ag. 535 f.
84 CF. A nt. Tetr. 1/3 9.
85 O C 367 if. In 90 F G rH fr. 45 a L ydian king goes into exile for 3 years to expiate a
m u rd e r co m m itted by his father.
Curses, Family Curses, and the Structure o f Rights 205
W h a te v e r th e ir legal statu s, th ere w ere, certainly, social
m e a n s by w h ich th e ch ild ren o f a p o lluted father could be m ade
to feel u n clean . It w as open to an y sacrificial com m unity to
m ak e its ow n decision as to w ho was accep tab le as a m em ber.
A b o v e all, th e m arria g e prospects o f the children an d p artic u ­
larly th e d a u g h te rs w ere jeo p ard ized : ‘w ho will m arry m e /h e r/
y o u ?’ is, in these contexts, a co n stan t re frain .86 B ut this too is a
difficulty n o t confined to the polluted b u t shared by the socially
d isc re d ite d in gen eral. E u rip id es’ H elen m entions th a t because
o f h e r d isg race no one is w illing to m arry H erm ione, and the
s a m e p ro b lem confronts the d a u g h te r o f a state d e b to r.87 It is
in te re stin g th a t O e d ip u s, in his p o rtra it o f the w retched life th at
a w a its his p o llu ted d au g h ters, seem s to d raw colours from
A n d ro m a c h e ’s p ictu re in H o m er o f the h ard sh ip s th a t A stya­
n a x will h ave to u n d erg o as a m ere o rp h a n .88 In th e second
g e n e ra tio n , p o llu tio n m ay indeed be som ething to be held
a g a in s t a fam ily, b u t as a ‘re p ro a c h ’89 not sh arply different in
kin d from a n y o th e r d am ag e its re p u tatio n m ight incur. A
c e rta in resid u al u n ease is a p p a re n t in P lato ’s specification th at
c a n d id a te s for th e priesth o o d should be investigated to ensure
‘th a t th ey com e from th e p u rest possible families; the c a n d id ate
h im se lf m u st be u n ta in te d by bloodshed and all such crim es
a g a in s t th e gods, a n d so m ust bo th his p a re n ts.’90 Plato is in
g e n e ra l opposed to in h erited guilt even for the w orst crim es. O f
th e ch ild re n a n d fam ily o f the m an executed for im piety he says:
‘I f th ey grow u p different from th eir father, they should be given
d u e c red it for th eir noble achievem ent in transform ing evil into
g o o d .’91 It is th erefo re significant th a t the one h ered itary dis­
q u alifica tio n for priesth o o d th a t Plato specifies should be the
ta in t o f b lo odshed. It is also significant, however, th at this
d isq u alificatio n should be confined to the narrow ly religious
sp h ere.
86 S oph. O T 1192—1502, Kur. A ndr. 974—6, El. 1198—1200; on the A lcm aeonids sec p.
16 above.
87 E u r. Hel. 933, (D em .) 59.8.
88 S oph. O T 1486 if., cf. Horn. //. 22.490 if.
89 S oph. O T 1494. In Eur. Supp. 2 2 0 -8 , A d rastu s, by contracting a m arriage alliance
w ith Polyneices, has 'm u d d ie d ' his ‘b rig h t’ house by contact with one th at is 'u n ju st',
‘sick ’, a n d ‘u n fo rtu n a te ’ (‘sick’ in relation to pollution also Kur. I T 693) - revealingly
v ag u e term s (cf. p. 2 1 9 o n contagious bad luck).
90 Leg. 759c.
91 855a, cf. 9 0 9 c - d , b u t note 8 5 6 c -d .
206 Miasma
T h is a rg u m e n t invites us to consider ancestral pollution o f an
u n fa m ilia r kind. In A ristophanes, th e accusation, T say th at
you a re from th e fam ily o f those w ho sinned against the god­
d e s s’, is c o u n tered by ‘I say th a t your g ra n d fa th e r w as one o f the
b o d y g u a rd s o f H ip p ia s.’92 T h e ju x tap o sitio n in A ristophanes
su g g ests th a t the tw o form s o f tain t w ere not felt to be radically
d ifferen t in kind. T h e sam e p illar on the acropolis bore in­
sc rib e d , for p e rp e tu a l contum ely, the nam es both o f th e sac­
rileg io u s a n d o f tra ito rs.93 O f the tw o tain ts, one was perhaps
ea sie r to efface th a n the other. W hen the S p artan s b ro u g h t up
a g a in s t Pericles the A lcm aeonid crim e, his p o p u lar support,
a c c o rd in g to P lu ta rc h , only increased,94 b u t no P eisistratid
co u ld set foot in A th en s in the fifth century. I f it was through
h o stility to ty ra n ts th a t th e A lcm aeonids in cu rred pollution, it
w as su re ly th e ir carefully n u rtu re d re p u ta tio n for the sam e
q u a lity th a t h elp ed to cleanse it.9S O n e has only to read the
sp eech es o f Lysias to discover how chronic, contagious, and
h e re d ita ry , in consequence o f the oligarchic revolutions, the
ta in t o f ‘h a tre d o f the people’ h ad becom e. A nd though it m ay
be h a rd , in th e strictly religious sphere, to discover instances o f
in h e rite d innocence to set alongside in h erited guilt, in civic life
th e y exist in a b u n d a n c e . D istinctions bestow ed by the A thenian
p eo p le on foreign benefactors regularly extended to th eir sons.96
T h e ap p e a l to a n c estral credit is one o f o rato ry ’s stan d ard
th em es; as a consequence o f an act o f sacrilegious m u rd er
p e rfo rm ed by th e ir ancestors, for purely personal m otives, the
d e s c e n d a n ts o f H a rm o d iu s a n d A ristogeiton enjoyed tax relief
a n d free d in n e rs in p e rp e tu ity .97
92 A r. Eq. 445-H .
93 I.v c u rg . Leoc. 117.
94 Per. 33. 1 -2 .
95 H d i. 6.121.1.
94 e.g. IG I3 2 3 ,2 7 ,9 2 , 95, 181.
97 e.g. Isae. 5.47. E x ploitation by A ndocides o f his ancestors’ hostility to tyranny,
1 .1 06,2.26.
7
DISEASE, BEWITCHMENT AND
PURIFIERS
A slave in M e n a n d e r is critical o f his m aster’s hypochondria:
W hat do I suggest you do? If there had really been anything wrong
with you, then you’d have had to look for a real cure. But there isn’t.
Find an imaginary cure for your imaginary disease and persuade
yourself that it’s doing you some good. Get the women to wipe you
round in a circle and fumigate you. Sprinkle yourself with water
draw n from three springs, with salt and lentils added.1
T h is p assage illu strates b o th th e sem i-m edical use of
religious tech n iq u es o f purification, a n d th e co n tem p t in w hich
su c h m eth o d s w ere held by en lightened A th en ian s o f the fourth
c e n tu ry . T h e sam e co n tem p t em erges from a fragm ent of
D ip h ilu s w h ich describes th e m ost fam ous purification of
m ytholo g y , th a t o f th e d au g h ters o f Proetus by M elam pus.
‘C le a n sin g th e d a u g h te rs o f P roetus an d th eir father P roetus the
son o f A bas, a n d th e old w om an to m ake five in all, w ith one
to rc h a n d o n e squill for all those people, an d su lp h u r a n d pitch
a n d m u ch -reso u n d in g sea, d raw n from the deep a n d gentle-
flow ing o c e a n .’2 D ip h ilu s’ m an n er, in m etre (hexam eter),
la n g u a g e (H o m eric expressions), a n d th o u g h t is th at o f
b u rlesq u e ; he ridicules th e notion th a t one torch an d one squill
co u ld serve to cleanse five people, an d seems to have transferred
to th e leg en d ary M elam p u s th e healing m ethods o f the lowest
c o n te m p o ra ry ch a rla ta n s. T h e g re at seer em erges as a p ed lar o f
su p e rstitio u s m u m m ery . I t is a hostile observer again, the
H ip p o c ra tic a u th o r o f On the sacred disease, who gives the m ost
d e ta ile d p ictu re o f su ch p ractitio n ers a t work. T hese ‘magi,
pu rifiers, begging-priests, frau d s’ w ho ‘purify (epileptics) w ith
blo o d as th o u g h they w ere p o llu te d ’ are, he claim s, m erely
1 Phasma 5 0 -6 .
2 D ip h ilu s, fr. 126.
208 Miasma
‘u sin g th e divine as a cloak a n d a shield for their ow n helpless­
ness in n o t h a v in g an y useful rem edy to ap p ly ’.3
On the sacred disease shows th a t epilepsy was a typical object of
p u rifica tio n . So too w as m adness, as we see from the casual
re m a rk th a t A risto p h an e s’ Bdelycleon h ad h ad his lunatic
fa th e r ‘w a sh e d a n d p u rified ’.* B ut a n anecdote telling how K ing
C leo m en es, in th e g rip o f a long disease, tu rn ed to the purifiers
for aid suggests th a t physical illness could be tackled in the
sam e w ay.5 T h e re is reason to think th a t certain skin diseases, in
p a rtic u la r, w ere p o p u larly seen as pollutions th a t could be
w à sh e d aw ay; a n d a cure for im potence is found in H ipponax
w h ich w ould alm o st certainly have been spoken o f as a purifica­
tio n .6 In m y th , even a plague afflicting an entire com m unity
co u ld b e effaced by a p ro p h e t sprinkling lustral w ater from a
b ra n c h o f lau rel to th e acco m p an im en t o f m agic w ords.7
I t is unlikely th a t the purifier h a d alw ays been the degraded
fig u re w ho a p p e a rs in m ost o f o u r texts. Some developm ent is
a lre a d y a p p a re n t in th e co n trast betw een the easy contem pt of
th e fo u rth -c e n tu ry com edians, a n d the need felt by the H ip p o ­
c ra tic a u th o r to a tta c k his m agical rivals a t length. A fragm ent
o f S ophocles refers w ith a p p a re n t respect to a ‘purifier of the
a rm y , skilled in th e rites o f w iping off (disease)’,8 an d it has
o ften b een p o in ted o u t th a t the ch a rlatan s o f On the sacred disease
re sem b le E m pedocles, w ho claim ed like them the pow er of
co n tro llin g th e w eath er, p ractised as a healer, a n d w rote a poem
w ith th e title Katharmoi.9 (A lthough katharmos here refers to
esc a p e from th e cycle o f in carn atio n , it is plausible th at
E m p ed o cles also tre a te d diseases by purification.) W e see the
p u rifie r’s o rig in al prestige above all in the fam e o f M elam pus,
m o st illu strio u s m em b er o f a fam ous m an tic fam ily.10 Every­
3 M orb.Sacr. 140.11.3-8, 1 4 8 .3 6 -8 J., 1.10.40G . 4 Vesp. 118.
5 P lu t. Apophth. Lac. 223e( 11). Purification from disease at Rom e: T ib . 1.5.11-12,
O v . Ars. Am. 2. 329 f.
6 Skin disease: see below . H ipponax, fr. 92, in terp reted by L atte, Hermes 64 (1929),
385 f. = Kl. Sehr. 464 f.
7 C allim . fr. 194. 2 8 - 3 1 , w ith C lem . Al. Strom. 5.8.48, p. 359 St., cited by Pfeiffer,
a d loc.
8 F r. 34, cf. G raf, 106. 9 See e.g. L loyd, 37 f.
10 Cf. H esiod, fr. 203. See on M elam pus Heldensage, 5 8 -6 0 , 196-202, 246-53;
K . H a n e ll, Megarische Studien, L und, 1934, 101-5; G. R adke in R E s.v. Proittdes; Nilsson,
G G R 613 η. 2; I. Löffler, Die Melampodie, M eisenheim , 1963; A. H enrichs, Z P E 15
(1974), 2 9 7 -3 0 1 . D exicreon in Plut. Quaesi. Graec. 5 4,303c is m odelled on M elam pus.
Disease, Bewitchment, and Purifiers 209
b o d y knew how he cu red the d au g h ters o f P roetus, or the
w o m en o f A rgos, from their frenzied w an d erin g th ro u g h the
P elopo n n ese. T h e re are, too, th e g re a t w an d erin g healers o f the
a rc h a ic a g e .11 S everal are associated w ith Apollo, as was
M e la m p u s h im self.12 A t A pollo’s behest, the seer Bacis
c lea n sed th e S p a rta n w om en o f a n o u tb reak o f m adness.
T h a le ta s th e C re ta n by his m u sic stayed a plague th at was
ra g in g in S p a rta . A b aris ta u g h t b o th A th en ian s an d S p artan s
sacrifices to a v e rt th e onset o f p la g u e .13 T h ese a re shadow y or
leg e n d a ry figures, b u t we ap p ro ach firm historical reality w ith
E p im en id es, a n d a tta in it in E m pedocles.
T h e re is, how ever, a difficulty ab o u t assim ilating the p u ri­
fiers o f th e fifth a n d fourth centuries to th e arch aic healers. In
th e earliest reference, it is not as a p u rifier b u t as a ‘healer-seer’
(iatromantis) th a t M elam p u s a p p e a rs .14 T h e purifier treats
sy m p to m s by a m agical tech n iq u e, w hereas the healer-seer
diag n o ses th e d isease’s cause. H e can th en prescribe the a p p ro ­
p ria te cure, w h ich need not take th e form of a purification. Both
for th e diagnosis a n d the p rescrip tio n his skill as a seer is
re q u ire d . In C alch a s in Iliad 1 we have a sim ple exam ple o f a
h ea le r-se e r a t w ork. H e identifies A pollo’s anger over C hryseis
as ca u se o f th e p lag u e, a n d tells th e G reeks w h at restitu tio n to
m ak e, a n d w h a t sacrifices to perform , in o rd er to appease the
g o d .15 E ven E p im enides, w ho gave his nam e to the first com ­
p re h en siv e stu d y o f an cien t p u rificatio n s,16 cleansed the
A th e n ia n s, acco rd in g to trad itio n , not by lu stratio n b u t by an
in g en io u s sacrificial ritu al designed to p ro p itiate offended
11 See o n th em R ohde, 2 9 9 -3 0 3 , L. G ern et Le génie grec dans la religion2, Paris, 1970,
1 1 8 -2 1 , N ilsson, G G R 6 1 5 -2 0 , D odds, C h. 5, Burkert, L S 1 4 7 -6 5 ,1. P. C ulian u , .SSÄ 4
(1980), 2 8 7 -3 0 3 .
12 H orn. Od. 15.245, H es. fr. 261, A pollod. 1.9.11. For his D ionysiae connections see
H d t. 2.49, P aus. 1.43.5, B u rk ert, H N 189 f.
13 S chol. Ar. Pax 1071 = 115 FG rH fr. 77; (Plut.) Mus. 4 2 .1 146b = P ratinas, P M G
713 (iii); A pollonius, M ir. 4, A r. Eq. 729, Sud. s.v. Abaris.
14 H es. fr. 37.14; cf. Bacch. 1L 9 5 -1 1 0 (w ithout M elam pus), schol. M V Horn. Od.
15.225 (b u t see Ä £ 2 3 . 1.120).
15 H orn. 11. 1. 9 3 -1 0 0 . F or o th er instances see e.g. 3 FGrH fr. 33 (with Heldensage.
58 f.); A pollod. 3.3 (P olyidus an d G laucus).
16J . L om eierus, Epimenides, sive de veterum gentilium lustrationibus syntagma, ed. 1
U ltra je c ti, 1681, ed. 2 Z u tp h a n ia e , 1700.
210 Miasma
g o d s .17 O n this evidence, E pim enides deserves the title
‘p u rifie r’ in the sense th a t he w as dealing w ith a ‘pollution’, but
n o t on th e basis o f the actu al ritu al m ethods th at he em ployed.
O n th e o th e r h a n d , it is pro b ab le th a t ‘p urification’ in the
n a rro w sense h ad alw ays form ed a p a rt o f the healer-seer’s
re p e rto ire . Before the p ro p itiato ry sacrifice, the A chaeans in
Iliad 1 a re told by C alch as to ‘w ash off th eir pollution an d throw
th e p o llu tio n in to the sea’.18 T h is clearly anticipates the
p ra c tic e o f th e purifiers o f On the sacred disease, who ‘b u ry some
o f (th e p o llu te d rem ain s) in the g ro u n d , cast som e in the sea,
arid c a rry o th ers to the m o u n tain s’.19 C onversely, a vestige of
d ia g n o stic p ro c ed u re survives even in the practice o f these
p u rifiers, since ‘I f the p a tie n t im itates a goat, or roars . . . they
sa y th e m o th e r o f th e gods is responsible’, a n d sim ilarly w ith a
w ide v ariety o f sy m p to m s.20 T h e classical purifiers should prob­
a b ly b e seen n o t as figures o f a q u ite different kind from the early
h ea le rs, b u t as heirs to a sm all portio n o f a divided patrim ony.
T h e g re a t em p ire o f the healer-seer w as fragm ented in the early
h isto rical period. M elam p u s a n d his descendants are m ythical
p re c u rso rs o f those g re at m antic families th a t played a n im ­
p o rta n t role in P eloponnesian history in the archaic age and
b e y o n d . M a n y points o f co n tact can be found betw een these
w a n d e rin g a risto c ra tic o p p o rtu n ists o f hered itary skills — the
Ia m id s, K ly tiad s, a n d T elliads - a n d the M elam podids, from
w h o m som e o f th em in fact claim ed d escent.21 O f the historical
p ro p h e ts , how ever, m iraculous cures are not recorded; their
s p h e re o f ac tio n w as above all m ilitary. O f M elam p u s’ healing
fu n ctio n s, m an y h a d obviously been ab sorbed by secular m edi­
cine. In so far as diseases still req u ired diagnosis in religious
term s, oracles such as th a t o f D elphi w ere able to offer it. M uch
17 D .L . 1.110. Ja c o b y , com m en tary on 457 FGrH, p. 310, regards the details as
a u th e n tic . F or E pim en id es as ‘healer-seer’ note Ar. Rhet. 1418a 2 3 -6 , Epim enides
p ra c tise d d iv in a tio n n ot a b o u t th e future b u t obscure events in the past, a n d 457 FGrH
T 4e, ‘he professed to purify people by rite from an y d am aging influence w hatever,
p hysical o r m en tal, a n d to state its cause" Both points are interesting even if they derive
from th e sp u rio u s w ritings.
18 //. 1.314. F or ‘h ealer-seer’ an d ‘purifier’ closely associated see Aesch. Eton, 62 f ,
Supp. 2 6 2 -7 .
19 Morb. Sacr. 1 4 8 .4 3 -6 J . , 1.42G .
20 146. 20 fT. J ., 1.33 If. G.
21 H d t. 7.221, P aus. 6.17.6. O n the characteristics o f such seers see I. Löffler, Die
Melampodie, M eisenheim , 1963, 1 1-29; an d on m ilitary prophecy P ritchett, iii, 47—90.
Disease, Bewitchment, and Purifiers 211
o f th e custom o f th e old healer-seers w as also, no doubt, diverted
b y th e increasin g ly im p o rta n t cults o f healing gods a n d heroes.
O n ly th e least re p u ta b le o f functions, the one th at m ost resem ­
b le d a m ag ical m an ip u latio n , rem ain ed for the purifier: B ut this
v ery expressive, m an ip u lativ e asp ect o f his tech n iq u e ensured,
p e rh a p s , th a t he w ould never entirely lack a clientèle. T h e
d ra m a o f p u rificatio n h ad a psychological appeal th a t in cu b a­
tion, p ra y er, o r th e o b servance Of a delicately b alanced dietetic
re g im e could scarcely rival.
I t h as often been said th a t th e fam e o f the healer-seers an d
p u rifiers is p ro o f o f th e obsessive anxiety th a t was felt ab o u t
p o llu tio n in th e a rch aic perio d .22 I t is scarcely justifiable, how ­
ever, to a ttrib u te to a rch aic G reeks en masse the m entality of
T h e o p h ra s tu s ’ S u p erstitio u s M an , w hose days are passed in a
p e rp e tu a l series o f p re cau tio n ary m easures against the con­
ta m in a tio n th a t th rea ten s him from every side. I f the archaic
G re ek in tim es o f affliction desired ‘pu rification’, th a t is testi­
m o n y to a cu ltu ra l idiom b u t n o t to a n obsession, since a rem edy
o f som e k ind is so u g h t by all afflicted societies. W e have no
ev id en ce th a t h ealer-seers w ere su m m oned in response to m ere
an x ieties. (U n fo rtu n ately , the precise circum stances in w hich
E p im en id es w as fetched from C rete to purify A thens are quite
u n c e rta in .23) T h e elim ination o f evils o f very various kinds, an d
by v ario u s m ean s, seem s to be seen in this period as a ‘purifica­
tio n ’, w ith o u t it necessarily being felt th a t they had been caused
by a pollu tio n . H eracles a n d T h eseu s, for instance, w ere seen as
h a v in g ‘p u rified ’ th e e a rth from th e m onsters an d brigands th a t
th e y slau g h tere d . T h o u g h it w as op en to a pious poet to explain
th a t in su ch a case e a rth h ad b ro u g h t forth these m onsters
‘th ro u g h th e p o llu tio n o f an cien t blo o d sh ed ’, in the p o p u lar
co n c ep tio n to call H eracles ‘p u rifier’ scarcely differed from
callin g him ‘a v e rte r o f evil’.24 P urification in this ra th e r broad
22 S ee w o rk s cited p. 209 n. 11 above, p articu larly G ernet, 120.
23 In one o f the versions in D .L . 1.110, a m ysterious plague afflicts A thens, and
E p im en id es identifies its cause in the C ylonian pollution. In Plut. Sol. 12 (from
A risto tle ), by co n trast, it is to deal w ith the C ylonian pollution, already a scandal th at
has led to th e trial a n d expulsion o f the guilty A lcm aeonids, th at he is sum m oned.
M o d e rn s ten d to follow P lu tarch , an d see th e purification as a sym bolic an ti-
A lcm aeo n id g estu re (e.g. Ja co b y , 4 0 f., W . G. F orrest, B C H 80 (1956), 3 9 -4 2 .) O n all
p ro b le m s co n cern in g E pim enides see Ja co b y , com m entary on 457 FGrH.
24 S oph. TV. 1012, E u r. H F 225, A pollod. 2.6.3 (T heseus); pious poet: Aesch. Supp.
2 6 2 - 7 ; alexikakos an d kathartës equivalent, A r. Vesp. 1043.
212 Miasma
sen se is not necessarily a m a tte r o f ritu al cleansing (in the case
o f H e racle s it obviously was not): E pim enides is supposed to
h a v e u sed sacrifice, T h a le ta s m usical therapy, the priests of
D io n y su s ecstatic ritu a l.25 A nd, if the w andering healer-seer’s
role is d e te rm in e d n o t by novel fears o f pollution, b u t by the
age-old need for healing, it becom es im plausible to see him as a
n ew co m er in th e a rc h a ic period. T h e scholarly trad itio n w hich
in sists th a t he is has no su p p o rt in the an cien t evidence, a n d has
to ig n o re C alch a s a n d tre a t M elam pus, d ista n t an cesto r o f a
c h a ra c te r in th e Odyssey, repeatedly alluded to in H esiodic
p o e try , as a m ythological latecom er.26
T h e p u rifica tio n offered by the purifiers o f the fifth a n d fourth
ce n tu rie s is o f a n a rro w e r kind. T h ey are called purifiers
b e c a u se they rem ove disease by a kind o f w ashing. W e have
a lre a d y seen th a t a purification in this n arro w sense is found in
th e Iliad, a n d th ere is fu rth e r evidence th a t the conception of
d isease as so m eth in g to b e w ashed o r pu rg ed aw ay was deeply
e m b e d d e d . H e alin g springs, for instance, sacred to H eracles,
A rte m is, o r th e N y m p h s, w ere w idespread.27 T h e sick person
b a th e d in th em , a n d th o u g h th eir sym bolic efficacy m ust have
b ee n b ase d o n feelings o f relaxation —the springs w ere norm ally
h o t — as well as purification, the, idea th a t their w aters carried
a w ay diseases w as certain ly àîso present. I t was th o u g h t th at
skin diseases in p a rtic u la r, by an obvious assim ilation to
o rd in a ry d irt on the skin, could be treated in this way. Pausanias
d e sc rib e s how th e victim o f leprosy, after an o in tin g the diseased
p a rts o f his body, w ould swim across the A nigrus m arsh; the
im p u rity re m a in e d in th e w ater, a n d h e em erged clean.28
A c co rd in g to one tra d itio n , M elam pus h ad w ashed the Pro^tids
25 O n th e h ealin g p ae a n see L . D eu b n er, NeueJahrb, 43 (1919), 385—406. I t h as long
been recognized th a t in one tra d itio n M elam pus ‘purified’ the P roetids by a D ionysiae
p u rs u it ritu a l: A pollod. 2.2.2.. rf. Paus. 2.7.8., B urkert, U N 189-200. O n D ionysiae
p u rific a tio n see C h . lv,.
26 So N ilsson, GGÆ 616, K . H an ell, Megarische Studien, L u n d , 1934, 99.
27 Cf. G in o u v ès, 361—73. B u tJ . H . C roon, Mnemos. 204 (1967), 225—46, argues th at
n a tu r a l h o t sp rin g s w ere n ot exploited for healing before ab o u t the 4th century.
28 P a u s. 5.5.11., cf. S tra b o 8.3.19.347 init. O n the w ashing-off o f skin diseases
cf. G in o u v ès, 370 η. 2, referring inter alia to H d t. 4.90, H ippoc. Epid. 5.9. (5.208 bottom
L ittré ), P liny, H N 31.11. N ie. Alex. 253 speaks o f ‘p o llution’ o f the skin; ointm ents used
a g a in s t sc ab ‘purify ’ it, D ioscorides Mat. Med. 1.1.3 p. 7.10 W ellm ann, 2. 163
p. 228.4, cf. 227.17 W ellm ann.
Disease, Bewitchment, and Purifiers 213
w ith w a te r from th e sam e sp rin g A n ig ru s,29 w hile an o th er
lo c a te d th e cu re in th e tem ple o f A rtem is H em era a t Lousoi
(‘w a sh in g s’), in th e precin ct o f w hich a pool for b ath in g has
b ee n u n co v e red .30 P urification p lay ed little p a rt in th e tem ple
m ed icin e o f th e classical period, b u t it has been arg u ed th at
h e a lin g cu lts first o f Apollo, a n d su b seq u en tly o f Asclepius,
co m m o n ly grew u p a t the site o f a sacred sp ring.31
T h e re is also th e evidence o f th e H ip p o cratic corpus. I t has
com e to be recognized th at H ip p o cratic m edicine is in m any
re sp e c ts a co n tin u a tio n o f trad itio n al practices a n d beliefs.32
T h is is p ro b a b ly tru e not m erely o f th e occasional im aginary
a ilm e n t th a t th e H ip p o cratic d o cto r still knew how to tackle
(w a n d e rin g w om b o r excess o f black bile), b u t also o f a large
n u m b e r o f th e form s o f tre a tm e n t th a t he h ad a t his disposal.
T h e c e n tra l im p o rta n c e in the H ip p o cratic corpus of katharsis is
th ere fo re m ost in terestin g .33 T h e body is a co n tain er whose
p u rity is n a tu ra lly m ain tain ed by periodic spontaneous ‘p u rifi­
c a tio n s ’ (excretion, m en stru atio n , a n d th e like). H ealth is the
b a la n c e o f th e h u m o u rs or vital principles present in the body.
W h e n one o f th em develops in excess, disease occurs, an d an
artificially in d u ced p u rification o f the peccan t m atter becomes
necessary . A lth o u g h this is achieved th ro u g h a p u rg ativ e d ru g
19 S tra b o 8.3.19.346 end.
30 B a tc h . 1 1 .9 5 -1 1 0 , cf. S tep h . Byz. s.v. Lousoi, an d p. 230 n. 131 below. O n the pool
see G in o u v ès, 383, ritm g / Ö A I 4 (1901), 15—18, a n d on this tem ple Stiglitz, 101—5.
31 G inouv ès, 3 2 7 t., 349 f., R .M a r tin an d H. M etzger, La Religion grecque, Paris. 1976,
C h . 3. B a th in g before in cu b atio n in A sclepieia, as before other incubation, was
re q u ire d (G inouvès, 3 5 2 -7 ; se e e sp . Ar. Plut. 656—8, X en. Mem. 3.13.3) but is distinct
from th e actu al h ealin g process. Prelim inary b ath in g w as required in o th er incubation
cults w ith healing functions (Paus. 9.39.7, T rophonius; X en. Mem. 3.13.3, A m phiaraus),
b u t a ro le for b a th in g in th e healing process is explicitly attested only for the cult of
P o d alirio s in A p u lia, L ycoph. Alex. 1050 ff. w ith schol. ( = T im aeus 566 FGrH fr. 56):
cf. G inouvès, 3 4 4 -9 . T h e re is considerable archaeological evidence, how ever, th at from
a b o u t th e 4 th cen tu ry ac tu a l h y d ro th erap y developed in Asclepieia: see G inouvès,
3 5 7 -6 1 , M a rtin /M e tz g e r, loc. cit. F or a cure in w hich stig m ata are wiped off see SIG 3
1 1 6 8 .4 7 -5 5 , cf. 5 5 -6 8 .
32 See e.g. O . T em k in , Isis 44 (1953), 2 1 3 -2 5 ; K udlien, passim·, idem , Clio Medica 3
( 1968), 3 0 5 -3 6 ; Lloyd, 3 9 -4 5 .
33 F irs t stu d ied in th is reg ard by O . T em k in , ‘B eiträge zur archaischen M edizin’,
Kyklos, Jahrbuchf . Geschichte und Philosophie der Medizin, 3 (1930), 9 0 -1 3 5 (cited by A rtelt
a n d G oltz: non vidi); cf. YV. A rtelt, Studien zur Geschichte der Begriffe ‘Heilmittel’ und ‘G iß ’
(Studien zur Geschichte der Medizin, ed. K. Sudhoff, 23), Leipzig, 1937, 4 9 -6 0 , 8 9 -9 1 ;
M o u lin ier, 1 5 8-68; G oltz, 2 8 3 -6 .
214 Miasma
(pharmakon —itself a w ord o f im p o rta n t extra-m edical connota­
tio n s ),34 m edical kalharsis entails m ore th a n the sim ple em pty­
in g o f the digestive tract. T h e h u m o u rs are situ ated no t in the
bow els b u t in the w hole body, a n d kalharsis affects flesh and
veins as well as the digestive organs;35 localized purifiers can be
a d m in iste re d to th e nose, head, a n d o th er regions, an d the effect
even o f the d ru g s ap p lied externally to w ounds is one o f purifi­
c a tio n .36 As a result, m edical texts contain expressions th at
strik in g ly recall th e lan g u ag e o f pollution. Phrases such as ‘such
bo d ies as a re im p u re ’, o r ‘it is beneficial for such patients, if they
a p p e a r u n p u rified , to cleanse th eir heads an d the rest o f their
b o d ie s,’37 w ith th e ir im plication o f a general bodily p u rity th at
is m o re th a n cleanness, sound initially like a sim ple transference
o f th e religious conception. W hen the p atien t, a n d no t ju s t a
p a r t o f him , is object o f the verb kathaiw,iS m edical a n d religious
p u rific a tio n are d istin g u ish ed only by the accom panying dative
o f in stru m e n t: M elam p u s purifies the P roetids w ith sulphur,
w h ile H ip p o c ra te s purifies the C oans w ith diu retic drugs. It is
p la u sib le th a t th e negative expression akatharsia, unpurified
m a tte r, filth, ac tu a lly en tered the language o f pollution, w here
it is co m m o n in th e fourth cen tu ry ,39 from the m edical side. T his
co n c ep tio n o f disease is certainly not a late developm ent in the
H ip p o c ra tic tra d itio n , as two texts from the corpus speak of
p u rg a tiv e d ru g s as a prim itive tech n iq u e in co n trast to the
‘m ore d o cto rly ’ m eth o d o f dietetics.40
O th e r H ip p o c ra tic trea tm e n ts, too, resem ble the m ethods of
th e purifier. Processes o f ‘w iping o ff’, fum igation, a n d localized
d re n c h in g h av e obvious c a th a rtic parallels, an d cataplasm s
b a se d on b arley g ro ats recall the b ra n m ashes som etim es used
34 A rtelt, op. cit. F or the religious use o f έλατήριος, which som etim es qualifies
ψ άρμακον, see L S J.
35 A rte lt, op. cit., 75 f.
36 A rte lt, 58 f., 55.
37 Aph. 2.10 (4.472 L .), A ff. 20 (6.230 L.).
38 e.g. Epid. 6.1.5 (5.268 L.) ονρητικο'ισι καθαίρειν, Loc. Horn. 28 (6.322 L.) καί
λοντροΐσι κάθαιρε.
39 D em . 19.199,21.199,25.63; for the H ip p o cratic use see G oltz, 284 n. 189, LSJ s.v.
άκαθαρσία.
40 De arte 6 (6 .8 -1 0 L .), AcutA (2.226 L.): cf. 1. M . Lonie, Medical History 21 (1977),
2 3 5 - 6 0 . S uch d ru g s clearly fall w ithin T em k in ’s conception o f ‘leechcraft’: see Isis 44
(1953), 219 f.
Disease, Bewitchment, and Purifiers 215
to w ipe off p o llu tio n .41 O n e o f the com m onest H ip p o cratic
p re scrip tio n s is a h o t b ath , still a form o f purification even
th o u g h th e ritu a l purifier w ould norm ally use cold w ater.42
T h e occasional re q u irem en t o f ab sten tio n from b ath s also has
an a lo g u e s in th e religious sp h ere.43 F o r th e doctors, o f course,
a n ab so lu te difference o f kind sep arates th eir m ethods from
th o se o f th e religious healer, a n d th e original family likeness is
sim p ly n o t perceived. I t is w ith o u t any sense o f incongruity th a t
th e a u th o r o f On the sacred disease, hav in g dism issed the purifiers,
goes o n to ex plain how th e origin o f epilepsy is a defective
‘p u rific a tio n ’ o f th e b ra in o f the still u n b o rn child, a n d he can
even fau lt his rivals for not su b m ittin g th eir p atien ts to m edical
tre a tm e n t by h o t b a th s .44 B ut this only shows how far H ip p o ­
c ra tic m ed icin e h a d ad v an ced from its origins; an d w hatever
d o cto rs m ay h av e said, p atien ts a t an unconscious level no
d o u b t co n tin u ed to perceive a n d resp o n d to the sim ilarities.45
T h e relatio n b etw een these tw o forms o f purification, by rite
a n d by m edicine, is a delicate one to define. T o see the one as a
sec u la r tran sp o sitio n o f the o th er w ould m ake it seem secondary,
w h e reas th ere is, in fact, n o th in g adv an ced ab o u t the use of
p u rg a tiv e d ru g s. R ath er, the two m ethods both derive from an
u n d iffe ren tiated ideal of p u rity , physical an d m etaphysical,
n ecessary b o th for h ea lth an d for p ro p er relations w ith the gods.
(T h u s in som e cu ltu res the purge is a p rep aratio n for ritu al
ac tiv ity .46) T h e tw o m ethods com e closest to convergence in the
tre a tm e n t o f m ad n ess. T h o u g h this long rem ained subject to
ritu a l p u rificatio n , an o th e r p o p u la r form o f treatm en t, this too
in v e n te d by M elam p u s for the d au g h ters of P roetus,47 was
41 W ip in g off: G oltz, 219 f. F um igation: ibid., 2 3 1 -7 . For fum igation w ith su lp h u r
a n d a s p h a lt, as in ritu a l purification, see Morb. 3.10 (7 .1 3 0 L .), Nat. Mul. 26 (7.342 L.).
D ren ch in g : G oltz, 2 2 1 -4 . C atap lasm s: ibid., 213 f. (on w iping off w ith b ran m ash see
p. 231 below ). O n ‘w iping off w ith m u d ’ in later m edicine see G raf, 106 n. 60.
42 G oltz, 2 1 7 -2 0 , G inouvès, 367 f.; see e.g. Loc. Horn. 28 (6.322 L.) λουτροΐσι κάθαιρε.
43 e.g. M ul. 1.66 (8.136 L .), Morb. 2.67(7.102 L.); the purifiers have the sam e rule,
Morb. Sacr. 140. 13 J . , 1.12 G ., as do T ro p h o n iu s, Paus. 9.39.5, and Pythagoreans
(B u rk e rt, L S 199 n. 34).
44 1 5 4 - 6 J ., 5 G ., 142.34J . , 1.21 G.
43 C f. Pl. Crat. 4 0 5 a - b on ήκάθαρσις καί ο'ίκαθαρμοί καί κατά τήν Ιατρικήν καϊ κατά τήν
μ α ντική ν w hich, he claim s, all serve the sam e end o f m aking m an ‘p u re in both body and
s o u l’.
46 L a n a ta , 53.
47 A pollod. 2.2.2, T h e o p h r. Hist. PI. 9.10.4: on hellebore see R E 8.163-70.
216 Miasma
p u rg in g by hellebore. Psychologically, the difference between
th e tw o m eth o d s lay m erely in w h e th e r the m adness was trans­
fe rre d in to th e purifying m aterials a n d carried off w ith them , or
sw e p t aw ay th ro u g h the body’s ow n channels.48 A pleasing
p ro o f th a t hellebore w as seen not pharm acologically but, in a
v ery g en e ral sense, as a ‘cure o f evil’ is the fact m entioned by
T h e o p h ra s tu s th a t ‘people use it to purify th eir houses a n d their
flocks, c h a n tin g som e kind o f ch a rm over it, a n d for a great
n u m b e r o f o th e r jo b s .’49 A house has no digestive system ; from
its use as a pu rg e, hellebore has becom e a full purifying agent.
T h e c a th a rtic m edicine o f the fifth a n d fourth centuries per­
p e tu a te s , it seem s, deep-rooted p o p u lar conceptions. I f we turn,
h ow ever, to co n sid er how the clients o f the purifier interpreted
th e process to w hich they w ere subm itting themselves, a paradox
a t o n ce arises. P urification assim ilates disease to d irt th at can be
w a sh e d off: A sclepius stretches o u t his gentle h an d a n d ‘wipes
o ff’ d iseases,50 a n d leprosy becom es an unclean excrescence on
th e skin th a t th e w aters o f a special stream will carry aw ay. But
a lth o u g h pu rificatio n th u s seems to be a kind o f m echanical
te c h n iq u e , it is in the tre a tm e n t o f ‘divine’ diseases th a t it is
a p p lie d .
T h e a u th o r o f On the sacred disease points ou t the difficulty: if
d ise a se com es from th e gods, the p ro p e r trea tm e n t is prayer,
sacrifice, a n d su p p licatio n ; p urification in these circum stances
is ‘m o st im pious a n d m ost godless’, one o f a series of practices
w h ich w ickedly im ply th a t m ortal techniques can constrain the
gods. ‘B u t I d o n o t believe th a t a m a n ’s body is polluted by a
god, th e c o rru p te st o f things by the p u re st, b u t th a t even ifit has
b een p o llu te d by so m eth in g else, the god w ould cleanse and
p u rify it, ra th e r th a n p olluting it.’51 R ites designed to w ash
a w ay d ivine an g e r in th e m an n er criticized are extrem ely com ­
m o n in the religions o f th e w orld,52 b u t th a t is no answ er to the
H ip p o c ra tic a u th o r’s objection.
48 See S im on, 317, n. 34. 49 T h e o p h r. H ist. PI. 9.10.4.
50 H ero d as 4.17 f.: cl. the ‘sc rap in g ’ o r ‘cleansing’ off of old age, H om . II. 9.446,
Nostoi, fr. 6 O C T H o m er v, p. 141, Aesch. fr. 45. S tigm ata are transferred to a bandage
in th e w o n d e r-c u re S /6 '3 1168.47-65. 51 1 4 4 - 1 5 0 J ., 1 .2 8 -4 5 G .
52 Cf. L évy-B ruhl, C h . 8., ‘D efilem ent a n d P urification’. T h e sam e sym bolism , not
su rp risin g ly , ap p e a rs in th e com pulsive acts o f neurotics: ‘O ccasionally patients w ith a
co m p u lsio n neurosis can m ake all th eir scruples d isa p p ear by b ath in g o r changing their
clo th es, “ b a d feelings” being conceived o f as d irt th at can be w ashed aw ay.’ (O.
F en ich el, The Psychoanalytic Theory o f Neurosis, L ondon, 1946, 289).
Disease, Bewitchment, and Purifiers 217
A d istin c tio n m ig h t be d ra w n betw een the sym ptom s o f the
d isease a n d its cause, p u rification being in ten d ed to tre a t the
fo rm er only. T h u s , after the p lag u e in th e Iliad, purification is a
p re lim in a ry to sacrifice, while b ath in g in the A nigrus m arsh to
c u re lep ro sy is p reced ed by p ra y er to th e N ym phs.53 From the
p o in t o f view o f th e p atien ts m entioned in On the sacred disease,
how ever, a d istin c tio n of this kind w ould probably be an over­
ra tio n a liz a tio n . I t is th e sensation of being in the g rip o f an
in v asiv e su p e rn a tu ra l force th a t is, for them , the p ollution.54
W h a t u n ites d irtin ess a n d divine in tervention, an d m akes the
o n e som ehow eq u iv ale n t to the o th er, is th a t both are external
in tru sio n s u p o n th e in tegrity o f th e body. T h e body th a t is
itself, free from all ou tsid e interference, is clean; an unw elcom e
in c u rsio n o f an y kin d d irties it, b u t an incursion as m ysterious
a n d su p e rn o rm a l as epilepsy is also divine. T h e epileptic
p a tie n ts do n o t s ta rt from th e prem iss th a t the divine pollutes,
w h ich w ould no d o u b t have shocked them no less th an the
slig h tly san ctim o n io u s H ip p o cratic; it is only in this restricted
a n d te m p o ra ry co n tex t th a t the gods becom e unclean. T his is
o n e o f th e an o m alies in trad itio n al belief th a t philosophy was to
re m e d y by th e p o stu late o f im p u re dem ons acting as agents of
th e divine vengeance, to w hom the rites o f purification and
ex p u lsio n a re a d d re sse d .55 Even in the earlier period, however,
it w as m o re com m on to envisage kêres, or daimones, perched upon
th e good th in g s in life an d polluting them than to identify the
p o llu tin g p o w er w ith an actual n am ed god, as h appens in On the
sacred disease.56
I t w ould n o t be surp risin g to learn th at diseases w hich
re q u ire d tre a tm e n t by purification w ere them selves the product
o f p o llu tin g acts o r conditions. T h is ap p ears to be the case in
som e cu ltu re s w h ich p ractise c a th a rtic m edicine; for certain
53 H o r n .//. 1 .3 14-317, P aus. 5.5.11.
54 D irt a n d div in e a n g e r a re close to eq u atio n in Soph. Aj. 655 f., i shall go to the
sh o re, ως α ν λύ μ α θ ’ άγνίσας έμά/μήνιν βαρείαν έξαλεύσωμαι fl« iç;(Eitrem , Opferritus, 121
η. 2). F or th e id ea o f katharmoi being addressed to sp iritual beings cf. the ‘P yth ag o rean ’
view in D. L. 8.32, w hich m ay reflect early attitu d es d espite an undeniable Platonic
influ en ce (N ock, ii, 601, W . B urkert, Gnomon 36 (1964), 564).
55 C h ry sip p u s ap. P lut. Quaesi. Rom. 51, cf. H erter, Dämonen, 6 8 -7 5 . T h is was the
B a b y lo n ian co nception, G oltz, 1-14.
56 F o r hires see e.g. PI. Leg. 937d; H erter, Dämonen, 5 4 -6 ; the όίος πράκτωρ o f Aesch.
Supp. 646—50 is an in te rm e d ia te figure.
218
Miasma
B a n tu peoples, for instance, consum ption and leprosy were
c a u s e d by th e pollutions o f b irth an d d e a th .57 In Greece, m ad­
n ess o f course m ig h t be d u e to the blood on a m u rd erer’s hands,
a n d th e re is also evidence th a t skin diseases could be traced
b ac k to pollutions. In a hellenistic story, the people o f Delos
in c u rre d a lep ro u s disease w hen they p erm itted a burial on the
sa c re d island, a n d the pollution th a t th rea ten ed O restes should
he fail to avenge his fath er w ould have taken the sam e form .58
T h u s th e affliction th a t w as a pollution in ap p e ara n ce (unclean­
ness o n th e skin) w as also in terp re ted as one aitiologically.
P o llu tio n on D elos seem s to have been suspected as a cause of
th e h isto rical g re a t p lague a t A thens, an d we find a rationaliza­
tio n o f these beliefs59 in the H ip p o cratic d octrine th a t plagues
a re cau sed by miasmata in the air. T h e R om an view th a t step­
p in g o n a n im p u re object causes m adness o r im potence is also
v e ry likely to go back to G reece.60 B ut a diagnosis o f this kind
w as n o t essential in o rd e r to a tte m p t a cath artic cure, as the
p u rifiers o f On the sacred disease do not seem to have offered one.
T h e sy m p to m s them selves, the violation o f the body’s integrity,
w e re th e p ollution to be cleansed, w ith o u t any antecedent
p o llu tio n being re q u ired to explain it.
It is n a tu ra l to ask w h a t relation there is, if any, betw een
th ese p o llu tin g conditions an d the infectious diseases of m odern
life.61 W e ca n n o t assum e, m erely because they required purifi­
c a tio n , th a t they m u st have been seen as contagious. T h e
c o n ta m in a tin g co n tac t th a t had to be cleansed was prim arily
th a t b etw een the victim a n d the god a n d not th at betw een him
a n d his fellow m en. In G reek p o p u lar belief, there seem to be
tw o k in d s o f contagious condition, n eith er closely related to
m o d e rn infections. O n the one h an d , there are pollutions such
as th o se o f b irth , d e a th , a n d blood-guilt th a t are com m unicable
57 L cv y-B ruhl, 232.
58 M a d n ess: see p. 129 n. 94. D elians: (Aeschines) Epistle 1.2. O restes: A esch. Cho.
278—82. Skin disease is also inflicted as a p u n ish m en t for religious offences th at are not
spec ifically pollutions: the P roetids, Hes. fr. 133 (cf. R oscher in R M L 3.458 —thus the
co n n e c tio n w ith ritu a l m asking, B urkert, H N 190 f., GR 170 f., is unnecessary); (Plut.)
Fluv. 21.4.
59 A th e n ia n plague: see p. 276 below. H ipp. Flat. 5, 6 (6.96,98 L.).
60 H ö r. Ars. P. 471, P etr. Sat. 134. See too X en. E phes. 5 .7 .7 -9 .
61 Cf. ( ). T em k in , ‘A n H istorical Analysis o f the C oncept o f Infection’, in the
collective w ork Studies in Intellectual History, B altim ore, 1953, 123-47, rep rin ted in his
The Double Face oj Janus, B altim ore, 1977, 456—71.
Disease, Bewitchment, and Purifiers 219
ac co rd in g to specified principles an d d em an d the form al seclu­
sion o f affected persons. O n the o th er, there are a series of
u n d e sira b le q u alities an d conditions th a t can be ‘w iped off’ on
p eople a n d w ith w hich one m ay be ‘filled’ - folly, im m orality,
b a d luck, a n d th e like.62 T h e contagiousness of bad luck often
a p p e a rs in com edy. ‘W ho goes there?’ ‘An unlucky m an .’ ‘K eep
to y o u rse lf th e n .’63 Ill-om ened w ords and prophecies often p ro ­
voke a sim ilar reactio n .64 A gainst contam in atio n s o f this kind
o n e can p ro tec t o neself by m ere words. ‘M ay it tu rn ag ain st
y o u r ow n h e a d ’, or by the sim plest o f all purifications, ‘spitting
o u t’ th e p o llu tio n .65 I t would be w rong to see the th re a t of
co n tag io n in all these cases as a m ere m etaphor. W e h ear, for
in stan ce , o f u n co n q u e red troops w ho w ere unw illing to be
jo in e d in o ne division w ith th eir defeated com rades, an d a
E u rip id e a n T h ese u s w arns o f the d a n g e r o f m arrying into an
u n fo rtu n a te h o u seh o ld .66 It seem s to be in this latter sense th a t
p o llu tin g diseases w ere contagious. People threw stones at
m a d m e n a n d m ig h t sp it at th e sight of a m ad m an or epileptic,
b u t these w ere p ro tections ag ain st som ething re p u g n an t and
frig h ten in g ra th e r th a n ag ain st a m edical infection or a formally
d efin e d p o llu tio n . N one o f the m any preserved sacred laws
in c lu d e th e d iseased am o n g th e polluted persons banned from
e n te rin g a tem p le.67 It is not clear th a t diseases ever truly
b e c a m e infectious in an y o th er sense th an this in G reek thought.
G reeks w ere p ractically aw are, in tim e o f plague, th at the
d isease could be co n tracted by co n tact,68 b u t in p o p u lar per-
62 ‘W ip in g o ff’: A r. Ach. 843, Eur. Bacch. 344. ‘Filling’: Ar. Nub. 1023, Dem . 20.28,
X e n . Lac. Pol. 14.4 (th is w ord is also used w ith reference to actu al pollutions. A nt. Tetr.
1 a 10, A eschin 2.88).
63 A r. Ach. 1018 f., Nub. 1263.
64 e.g. A r. Pax 1063, E u r. Hec. 1276; cf. Ar. Ach. 833, Pax 651, Lys. 506, Dem .
18.290,19.130,54.16.
65 S ee p reced in g note. F or sp ittin g see e.g. E ur. Hec. 1276, Gow on T heocr. 6.39,
p. 108 n. 9 on th e m u rd erer.
66 X en . //*//.1.2.15; E u r. Supp. 220—8 (cf. for contagious luck ibid., 591, an d for a
‘sta in o f m isfo rtu n e’ Soph. O T S W ).
67 S tones: A r. Av. 524 f. S pitting: T h eo p h r. Char. 16.15, cf. Plaut. Capt. 550, Pliny, H N
10.69,28.35. S acred laws: W äch ter, 43. T h e view som etim es expressed th at m adm en
w ere form ally excluded from tem ples seems to be based on m isinterpretation o f Ar./U·.
524 f. C o n tra s t, for form al seclusion in Persia, H d t. 1.138.1.
68 T h u c . 2.51.4—6 (som e aw areness o f contagiousness was clearly general - note the
reference to ‘fear’ o f ten d in g the sick); SIG 3 943.7 -1 0 ;? Soph. O T 181. For Rom e see
Thes. Ling. Lat. s.v. contagium, a w ord m ost com m only (and perhaps originally) applied
to infections am o n g sheep (w hich Greeks too will have observed).
220 Miasma
ce p tio n this m ay have been no m ore th an an acute instance o f
th e contagiousness o f m isfortune. Even T hucydides has no
o th e r lan g u ag e w ith w hich to describe infection th an th a t of
b ein g ‘filled w ith ’ th e disease, an expression com m only applied
to contagions o f a different kind. I t is w orth considering the
p o ssib ility th a t the H ip p o cratic doctors ignored the principle o f
infectiousness69 because they saw belief in it as m ere super­
stitio n . I t is th e S uperstitious M an o fT h e o p h rastu s w ho spits a t
sig h t o f a n epileptic.
C a th a rtic m edicine aim s to restore the sense o f personal
w holeness th a t has been d istu rb ed by a ttac k from outside. Seen
in this w ay, illness is no t a discrete phenom enon, b u t one o f a set
o f d a n g e ro u s in tru sio n s u p o n the norm al tenor of life. T h e w ord
nosos itse lf is no t confined to disease b u t covers a wide variety of
‘b a d th in g s’,70 a n d we h a v e ju s t seen th a t b ad luck, for instance,
m u st be avoided a n d b a d news sp a t out. T h e sources do not
allow us to estab lish the full range o f ‘b ad things’ against w hich
p u rifica tio n could be em ployed, b u t several can be nam ed. A
‘d iv in e d re a m ’ th a t p o rten d e d ill m ight be w ashed aw ay, sp at
o u t, o r p u rg e d in o th e r ways; a p articu larly serious case would
re q u ire sacrifice to th e gods o f aversion.71 Evil om ens could also
be tre a te d by purification, b u t the evidence here is surprisingly
sc a n ty .72 I t is p erh ap s characteristic o f G reek in contrast to
^ S e e J . C . F. Poole an d A. J. H olladay, CQ n.s. 29 (1979), 295—9.
70 e.g. H orn. Od. 15.407 1'., H es. Theog. 527, Pind. Pylh. 4. 293; cf. G. Preiser, Allgemeine
Krankheitsbezeichmngen im Corpus Hippocraticum (Ars M edica I I .5), Berlin, 1976,89-104.
O n ‘b a d n e s s’ see e.g. H orn. Od. 5.397,17.384,22.481. B ut on the lim itations o f this kind
o f a rg u m e n t see G . Lew is, Knowledge o j Illness in a Sepik Society, L ondon, 1975, 142 f.,
355 f. (n o one ever takes to th eir bed in response to the disease o f poverty, an d so o n ).
71 W ashing: A r. Ran. 1340, A p. Rhod. 4.670 f. Purification: Plut. De Superst. I66a.
S acrifice/lib atio n s an d p ray ers, to th e Iheoi apotropaioi, or Apollo, o r the pow er (perhaps
a d e a d m an ) w hose a n g e r the d re a m portended: A esch. Cho. 5 2 3 -5 (dead m an), Pers.
2 0 1 - 4 , 2 1 6 - 9 (apotropaioi), 219 f. (earth an d the d ead ), Soph. El. 405 -2 7 (dead m an),
6 3 4 - 5 9 (A pollo), X en. Symp. 4.33 (apotropaioi) , M en. Dysc. 4 0 9 -1 7 (P an). (For offerings
a fte r a fav o u rab le d re a m cf. X en. Cyr. 8 .7 .2 -3 ). P rayer, to sun, Zeus etc. after
fav o u ra b le d ream s, to apotropaioi, earth , an d heroes after unfavourable, H ippoc. Viet.
4.89,90 (6 .6 5 2 ,6 5 6 -8 L .). (F o r p rayers cf. M oschus 2.27, 4.123, C Q 32 (1982), 233 f.).
C o n s u lta tio n o f the d re a m in terp reters for advice w hich god to propitiate: E ur. Hec.
8 7 - 9 , T h e o p h r. Char. 16.11, cf. ‘M agnes’ fr. 4. D eclaration o f the dream to the open air:
S oph. E l. 424 f., E u r. I T 42 f. S p ittin g out: Aesch. Ag. 980. S tatem ent th at ‘I banish the
d r e a m ’: E u r. Hec. 72, 97. O n th e apotropaioi see Nock, ii, 599—602, w ith references; later
ev id en ce on d re a m p ro cu ra tio n in H ead lam ’s note on H erodas 8.11. Sim ilar beliefs
a p p e a r, thinly ratio n alized , in H ipp. Viet. 4: se eesp . §87—8 (6.642 L.).
72 P lu t. Alex. 57.3, 75; T h e o p h r. Char. 16.14.
Disease, Bewitchment, and Purifiers 221
R o m a n religion to view p o rten ts not as m onstrosities requiring
ritu a l b a n is h m e n t b u t as signs for w hich in terp retatio n is
n ecessary .73 T h e ap p ro p ria te response to these signs, once
in te rp re te d , will often be a n action on a practical level* such as
th e a b a n d o n m e n t o f a ca m p a ig n .74 T h e distinction is not
a b so lu te , since th e im pulse sim ply to elim inate the ab n o rm al is
also found in G reece. M o n stro u s b irth s an d o th er abom inations
w ere so m etim es b u rn t on ‘wild w ood’ o r the wood o f fig-trees
(w orthless m a te ria l ch aracteristically being chosen for the
d isp o sal o f a p o llu ted o b ject).75 B ut this, too, is a kind o f concern
th a t, in c o n tra st to th e conspicuous R o m an obsession, scarcely
p e n e tra te s o u r sources.
Love, in la te r an tiq u ity , was a condition the luckless suitor
m ig h t seek to get clear o f by purification, a n d it w ould not be
rid icu lo u s to p o stu late classical G reek p reced en ts.76 Plato urges
th e m a n d riv en by sacrilegious im pulses to tu rn to the rites of
e x p u lsio n .77 P articu larly in terestin g for the view of purification
as a n a tte m p t to resto re the p ersonality to its norm al state after
a n alien in cu rsio n is a passage o f th e Cratylus78 in w hich Socrates
p layfully sp eaks o f th e passion for etym ology as a ‘w isdom ’
w h ich h as ‘fallen u p o n him ’ sud d en ly from an unknow n source,
p ro b a b ly th ro u g h co n tact w ith E u th y p h ro , who in his ‘in sp ira­
tio n ’ h a d filled n o t ju s t S ocrates’ ears b u t also his soul w ith this
‘s u p e rn a tu ra l w isd o m ’. H e suggests th a t for the day he a n d his
in te rlo c u to rs sh o u ld exploit the alien inspiration, b u t on the
m o rro w ‘expel a n d purge it’ th ro u g h the offices o f w hatever
p rie s t o r so p h ist w as best at perform ing purifications o f this
73 e.g. P lut. Per. 6.2.
74 C f. th e m aterial in P ritch e tt, iii, C hs. 3 - 4 . O n prodigies see R. Bloch, Les Prodiges
dans l ’antiquité classique, P aris, 1 9 6 3 ,9 -4 2 , w ith bibliography.
75 (D ion. H al.) Rhet. 9.10, p. 309 U sener-R aderm acher, on E u ripides’ Melanippe
(ch ild ren bo rn to a cow ); Phrynichus Praeparatio Sophistica, ed. 1. de Borries, Leipzig,
1911, p. 15.12 = Anecd. Bekk. 10.26 (τά τερατώδη τήν φύσιν)·, Diod. Sic. 32.12.11.'.
(h e rm a p h ro d ite s); see too T heocr. 24.89—92 (snakes sent against inlant H eracles);
L ycoph. Alex. 1 1 5 5 -9 + schol. on 1155, citing T im aeu s 566 FGrH fr. 146 (Locrian
m aidens); ? T zetzes Chil. 5.735 (b u t cf. G eb h ard , 3 f.) (scapegoats). B urning on
fig-w ood: com ic po et ap. Dio C hrys. 33.63 (K ock, CGF, iii, p. 398), L ucian, Alex. 47.
B u rn in g o f katharmata a t crossroads: E upolis, fr. 120. T hrow ing of m onstrous births
o v er o n e ’s sh o u ld er is p erh ap s im plied by Eur. Andr. 293 f.
76 T ib . 1.2.59, N em es. Eel. 4 :6 2 -7 ; contra, O v. Rem. Am. 260.
77 Leg. 854b.
78 PI. Cra. 396c-e.
222 Miasma
k ind. T h is is a clear indication th a t purification counters such
d isru p tio n s o f the in d iv id u a l’s norm al personality as are felt to
h av e th e ir origin o u tsid e the individual himself.
It is n a tu ra l in these term s th a t bew itchm ent, the ‘im ported’
evil,79 should have been seen as a pollution. W hen E uripides’
P h a e d ra m akes h er fam ous statem en t th at, though her hands
a re clean, h er m ind is polluted, the nurse thinks at once of
sorcery, a n d we know from On the sacred disease th a t the person
a tta c k e d in this w ay m ight be purified by blood.80 It w ould be
in trig u in g to know in w h a t circum stances bew itchm ent of an
in d iv id u a l o r house w as liable to be diagnosed, bu t on this the
so u rces offer very little g u id an ce.81 T h e o p h ra stu s’ S uperstitious
M a n w as co n stan tly purifying his house, on the grounds th at
H e c a te h a d been conjured ag ain st it.82 O n e of the sorcerer’s
m e th o d s, it seem s, was to constrain the goddess by m agical
m e a n s to a tta c k his v ictim ’s person a n d hom e.
W e h av e a frag m en t of· an invocation o f H ecate which,
th o u g h its co n tex t is u n certain , gives a vivid hint of the kind of
co n c ep tio n w ith w hich the sorcerer m u st have w orked.83 It is
w ritte n in a rh y th m ic-so u n d in g D oric prose th a t suggests
S ophrori, a n d is th a t p a r t o f an invocation w hich describes the
g o d ’s location, pow ers, o r condition before the actual request is
m ad e. B ut in stead o f ‘w h eth er you are in x o r y’ we find
‘w h e th e r you have com e hasten in g from a hanging, o r from
g rin d in g to d e a th a w om an in childbed, or from ranging am ong
79 W itch craft o r spell as έπακτός: E ur. Hipp. 318, Inscr. Crel. 2.xix. 7.20, S G D I 3545
(b u t on th e la tte r two cf. P. M aas, Hesperia 13 (1944), 36 f.), and below, p. 348.
80 E u r. Hipp. 317 I'., H ip p . Morb. Sacr. 148.38 J ., 1.40 G. Cf. S uda, Photius, s.v.
π ερ ικ α θ α ίρω ν. άναλύων τόν πεφαρμακενμένον ή τον γεγοητενμένον. F or the use of
φαρμακεύω = bew itch cf. PI. Leg. 932e-933e; it derives from an original undilfer-
e n lia te d concept o [pharmaka as forces o perating invisibly an d m ysteriously lor good or
evil. It is u n clear w h a t form the ‘destructive d ru g s' feared by the people ol I eos m ight
tak e (M /L 30 A 1 -5 , cf. L atte , H R 68 n. 18); in PI. Leg. 845e the drugs used to dam age
w ells a re p resu m ab ly poisons ra th e r than spells (cf. T h u c. 2.48.2), but a religious
p u rificatio n p rescrib ed by th e e x e g e te rem ains necessary. Fum igation eures fascinatum
animal in iate veterin ary texts, V egetius, Mulomedicina, 3.12.1; C laudii H erm eri Mulo­
medicina Chironis, cd. IC. O d e r, Leipzig, 1901, 497, p. 163.
81 N ote, how ever, Kur. Andr. 15 7 1, (w om an's sterility). O th e r plausible occasions (cf.
th e defixiones) w ould be disease, unexpected failure in an im p o rtan t enterprise, or a run
o f b a d luck; an d o f course objects suggesting m agical attack m ight be seen (PI. Leg.
933 b ).
82 T h e o p h r. Char. 16.7. F or w hat m ay be a sp ectacu lar case see p. 348 below.
83 A p. P lut, de SupersI. 170b; on the text see W ilam ow itz, Griechisches Lesebuch'3, Berlin,
1936, i.336, ii.2 10 f.; R. H erzog, Hess. Blatt, f . Volkskunde 25 (1926), 219 n. 4. ■
Disease, Bewitchment, and Purifiers 223
co rp ses’, a n d th ere follow references to several o th er forms of
p o llu tio n . T h is re m a rk ab le language surely belongs to som eone
w h o se aim it w as to exploit the g o ddess’s pollution for sham eful
e n d s .84 L ove-m agic w ould be one possible context;85 conjuring
a g a in s t an en em y an o th er. T h e g o ddess’s pow er to h arm was
exp ressed in h er im p u rity .86
I t is very likely th a t an exorcism of H ecate, as perform ed by
T h e o p h r a s tu s ’ S u p erstitio u s M an , was represented at least
o n ce in lite ra tu re . S o p h ro n com posed a m im e w ith a title the
m o st obvious tra n sla tio n o f w hich is ‘T h e wom en who claim
th a t they are d riv in g o u t th e goddess’.87 (O f the altern ativ e
e x p la n a tio n s th a t h ave been offered, som e are linguistically
im possib le, som e sim ply m uch less plausible. ) Some idea of how
th e expulsion m ay have been achieved is given by a papyrus
fra g m e n t w hich p ro b ab ly belongs to this m im e.88 An officiant
84 T h is seem s to em erge from the exam ples o f ritu al loidoria collected by S. I .itrem ,
Sjm b. Osl. 2 (1924), 43 (T., cf. ibid., 12 (1933), 23 f., 21 (1941), 48 f. If correct, this
conclusio n excludes the generally accepted ascription o f the piece to the m im e ‘T h e
W o m en w h o claim . . cf. K.. K eren y i, Riv.Fil. 13 (1935), 10. P. L egrand, R E A 36
( 1934), 2 5 -3 1 , arg u es on o th e r grounds th at S ophron w rote several m agical m im es.
85 C f. th e loidoria T h eo cr. 2 .1 2 -1 6 . F o r pollution in m agic see T heocr. 5.121 w ith
C o w ’s note; p o llu tio n an d m agical attack , O rp h . Lith. 591 (585);
86 Schol. T h eo cr. 2 .11/12 records, im m ediately after a Sophron citatio n , an elegant
little folk-tale th a t exp lain s H ecate’s association w ith d eath and every form o f pollution.
H e r a bore Z eus a d a u g h te r n am ed "Αγγελος, th e S yracusan form o f A rtem is/H ecate
(cf. H esy ch s.v. άγγελος). 'Α γγελος stole h er m o th er’s magic m yrrh an d gave it to
E u ro p a . H e ra , furious, p u rsu ed her d au g h ter, b u t Ά γγελ ο ς fled first to the house o f a
new m o th er, th en in to a funeral procession, w here the O lym pian H era natu rally could
n o t follow her. Z eus in stru cted the C ab iri to purify Ά γγελος. They did so a t the
A c h e ru sia n m arsh , b u t this m ean t th a t she belonged for ever to the ch thonian world.
T h e sto ry alm o st certain ly derives from S op h ro n or A pollodorus' com m entary on him;
for differen t views see W ilam ow itz, Hermes 34 (1899), 2 0 6 -9 = Kl. Sehr, iv, 4 8 -5 1 ;
G . K aib el, CG F 161 an d Hermes 34 (1899), 319; O . ('ru siu s, Neue Jahrb. 25 (1910),
8 6 - 9 0 ; K.. L atte , Philol. 88 ( 1933), 263 = Kl. Sehr. 497.
87 S o p h ro n , fr. 3 - 9 in K aib el CGF. D oxography in O livieri, Frammenti della comedia
greca e del mimo, ii2, N ap les, 1947, 68 f. C om plete scepticism in Page, CLP 329; A. S. F.
G ow , Theocritus, C a m b rid g e, 1950, ii, 33. T h e m ost plausible altern ativ e is th a t of'
R. W ü n sch , J a h rb f. Klass. Phil. Suppl. 27 (1902), 1 11-22, and L atte. Philol. 88 (1933),
263 = Kl. Sehr. 497, ‘W om en who claim the goddess is riding o u t.’ B ut elaunô and
c o m p o u n d s a re rep eated ly used of ritual expulsion: see S. Eitrem , Symb. Osl. 12 (1933),
1 If., an d a d d A esch. Cho. 967 f., LSCG 56.1, L ucian, Philops. 16, O rp h . Lith. 596 (590),
Carmen de viribus herbarum, ed. E. H eitsch (Die griechischen Dichterfragmente der römischen
Kaiserzeit, ii, G ö ttin g en , 1964), 172,177; cf. G RBS 22 ( 1981 ), 284 n. 3.
88 Page, G L P 328 w ith bibliography; vital for the ritu al details S. E itrem , S m b . Osl.
12 (1933), 1 0 -2 9 ; K. L atte, Philol. 88 (1933), 2 5 9 -6 4 , 4 6 7 -9 = Kl. Sehr. 4 9 2 -8 , and
G ow . Theocritus, ii, 34.
224 Miasma
a sse m b les a series o f m aterials —salt, laurel, a puppy, asphalt, a
to rc h — th a t w ere com m only used in purifications, in o rd e r to
‘box a g a in st the goddess’. T h e fragm ent breaks off a t the vital
m o m e n t b u t, if it did p o rtra y an expulsion, it looks as if the
g o d d ess w as first p ro p itiated by the offering o f a m eal;89 this will
th e n h av e been carried out, an d the goddess w ith it. T h e m ost
in te re stin g asp ect is th e reservation on the p a rt o f the m imo-
g ra p h e r th a t his title im plies: he will no t endorse the w om en’s
c laim to b e ‘d riv in g o u t the goddess’. T h eir haste in assum ing
b e w itc h m e n t, th e ir folly in a tte m p tin g to constrain the gods by
m ag ical m eans, the im piety o f supposing th a t gods pollute
m e n :90 one o f these, p erh ap s m ore th a n one, m ay have been the
ta rg e t o f S o p h ro n ’s irony. It w as, how ever, to keep H ecate away
th a t pious A th e n ia n s carried out m eals for h er to the crossroads
ea ch m o n th .91
A b o u t fu rth e r contexts for purification we can only speculate.
I t w ould be in trig u in g to know w h eth er b ad luck a n d poverty,
for in stan ce , w ere diseases for w hich it w ould have m ade sense
to try su ch a cu re.92 A nim als as well as m en w ere often purified
a g a in s t harm fu l influences, b u t here too precise details escape
u s.93
I t re m a in s to consider briefly the techniques th a t the purifiers
u se d .94 S om e practices will be included w hich are found in such
*9 L ines 1 7 -1 8 . L a tte co m p ared the A rval B rothers’ offering to the M a ter L arum ,
th ro w n o u t d ow n th e hill th ro u g h the tem ple doors (D essau, IL S 9522 II 23 f. with
n otes; L a tte , R R 92). H e c a te ’s m eals obviously reflect the sam e idea, an d see p. 347
below .
90 H a ste : cf. M en. fr. 97, T h e o p h r. Char. 16.6-7. Folly: PI. Leg. 909b, M en. fr. 210.
Im p ie ty : p. 2 1 6 above.
91 p. 30 n. 65 above.
92 N o te the p a ra tra g ic line A r. Pax. 1250ώ δνσκάθαρτε όαιμον, ώςμ'άπώλεσας. P articu­
larly in trig u in g is Morb. Sacr. 148.38 J ., 1.40 G ., they purify epileptics ώσπερ μίασμά τι
έχοντας ή άλάστορας ή πεφαρμαγμένους ϋπ' ανθρώπων. (T h e ru n o f the sentence makes
οάάστορας object o f καθαίρονοι ra th e r th an έχοντας, an d thus hum an not dem onic.) It
in d ic a te s th a t being a n alastör is a condition a n individual m ight acknow ledge in
him self, a n d n ot ju s t a ta u n t h urled by enem ies: cf. only Aesch. Eum. 236 (itself
p ro b le m a tic ), an d p erh a p s th e cu lt o f Zeus A lastoros (p. 139 n. 143 above). In w hat
circ u m sta n c e s w ould one a d m it to being an alastör?
93 e.g. T h e o p h r. Hist. PI. 9.10.4, Diod. 3.58.2, O rp h . Lith. 208-218, p. 222 n. 80
ab o v e. A nim als receive o th e r purifications too, e.g. T h eo p h r. Hist. PI. 9.8.4 (purge), PI.
Leg. 735b.
94 Cf. R ohde, 5 8 8 -9 0 ; Stengel, 155—70; B ouché-L ecterq in D ar-Sagl. s.v. Lustratio',
B u rk e rt, GR 129-32; E itrem , Opferritus, passim. Im p o rta n t sources are the passages of
D ip h ilu s a n d M e n a n d e r cited above, p. 207, th e S o p h ro n m im e (p. 223 above), Theocr.
Disease, Bewitchment, and Purifiers 225
co n tex ts as in itiato ry cults or th e p u rificatio n o f priests and
tem p les, even th o u g h they are n o t a ttested as actu al m ethods of
h ealin g . T h is is p a rtly a n expository conveniende, b u t finds its
ex cu se in the v ery extensive overlap th a t does exist betw een the
p u rifica tio n s p ra ctise d in these different contexts. T h is overlap
is a n im p o rta n t facto r in the ap p e al th a t cath artic m edicine
exercised. I t w orked n o t m erely by assim ilating disease to dirt,
b u t also in d irectly by exploiting all the positive value assigned
to p u rifica tio n as a form o f actio n in a wholly religious context.
T h e non-specific ch a ra c te r o f rites w as noted by D urkheim ;95
o n ce th e ir p restig e is established, they ten d to be em ployed very
w id ely o u tsid e th e ir original context. C o m m union is taken at
w ed d in g s a n d fun erals as well as actu al com m union services,
j u s t as no an c ie n t cerem ony was com plete without, sacrifice;
m ass can be specially celebrated for healing, while a H in d u ,
w e ak en e d by m a ln u tritio n a n d an aem ia, m ay seek to recover
his s tre n g th by ritu a l b ath in g .96 In a sim ilar w ay the purifier
b enefits from th e techniques, a n d the prestige, o f the priest.
T h o u g h e la b o ra te stage directions seem m ore ch aracteristic
o f th e an c ie n t n e a r east, there is som e evidence for the sym bolic
ex p lo itatio n o f sp ace in G reek ritu als o f this kind. It is said th at
rite s o f ex p iatio n a n d purification w ere norm ally perform ed
facin g ea st.97 W e h av e alread y seen the im p o rtan ce o f sym bolic
en c irclem en t. L u stra l w ater w as d istrib u ted to the ring o f
p a rtic ip a n ts before sacrifice, a n d purificatory anim als were
c a rrie d ro u n d every A th en ian place o f m eeting; M an tin ean s
w e re su p p o sed on one occasion to have taken anim als aro u n d
th e ir e n tire te rrito ry .98 W hen an individual h u m an was initi­
a te d o r p urified, he w as seated subm issively in the m iddle, and
th e officiants perform ed w h atev er ritu al was ap p ro p riate
a ro u n d h im .99 V erb s like ‘purify in a circle’ (perikathaim) w ere
2 4.88—100, P lu t, de Superst. 166a, 168d, L ucian, N ec.l, C lem . Al. Strom. 7.4.26.2—3, vol.
iii, p. 19 St. (red wool, salt, torches, squills, su lp h u r), idem , Protr. 1.10.2, vol. i, p. 10
St. (lau rel leaves an d fillets).
95 Elementary Forms, 385 f.
96 R e ad , 71.
97 Schol. Soph. O C 477, cf. O rp h . Lith. 210.
98 p p . 20 f ; M a n tin e an s, Polyb. 4.21.9; cf. fu rth er Lucian, Philops. 12, Paus. 9.22.2,
th e R o m an (an d Ig uvian) Amburbia, Ambarvalia etc. (L atte, R R 411'.); m uch m ore in
E itre m , Beiträge, ii, 1—19.
99 M en. Phasm. 5 0 -6 .
226 Miasma
used even w hen th ere w as no ac tu a l encirclem ent.*00 B ut the
sam e effect could be achieved by different m eans. W hen the
M a c e d o n ia n a n d B oeotian arm ies w ere purified, the dog victim
w as n o t c a rried ro u n d them , b u t divided into two halves
th ro u g h w hich they m arch ed , the severed anim al creating w hat
h as b een called a n ‘ab so rp tiv e zone’.101 W h a t m attered was to
c re a te sym bolic co n tac t betw een the person w ho was to be
p u rified a n d th e c a th a rtic objects. A nother possibility is seen in
a ritu a l d escrib ed b u t surely not invented by V alerius Flaccus:
tw o priests ca rried p a rts of anim als in opposite directions
th ro u g h the m iddle o f the A rgonauts, touching the heroes w ith
th e m as they w e n t.102 T h e scapegoat w as w hipped on the
g e n ita ls w ith c a th a rtic p lants, while in a w ide variety o f initia­
tio n s th e v irtu e o f the sacred objects was tran sm itted to the
se a te d c a n d id a te by sim ply holding them over his h e a d .103
A m o n g ag en ts o f purification, the m ost w idely used a n d m ost
b asic w as w a te r.104 N a tu ra l though this seems, there are dif­
ferences as well as sim ilarities betw een secular an d religious
clean sin g . L u stra l w a te r h ad to be pure, an d draw n from a
flow ing so u rc e ;105 so too, if possible, w ater for ordinary
w a sh in g .106 B ut no w asherw om an w ould think of com bining the
w a te rs o f three, five, seven, or fourteen different springs to
rem o v e even the d eep est s ta in .107 T h is was a distinctively
religious source o f pow er. P articu lar springs were especially
fa v o u red for p urifications, a n d the m ost prized ca th a rtic w ater
w as th a t o f th e salt-stain ed sea: ‘the sea w ashes aw ay all evils
100 Cf. R E S u p p l.6.149—5 1 (Pfister) for a list o f the lustral ^ m -com pounds. Sim ilarly
R o m a n circumferre.
101 p. 22 above.
102 Argon. 3.439—443; cf. P. Boyancé, R E I. 13 (1935), 107—36.
103 S cap eg o at: H ip p o n ax , fr. 10. Sacred objects: G. S chneider-H errm ann, Antike
Kunst 13 (1970), 5 2 -7 0 .
104 Sec E itrcm , Opferritus, 7 6 -1 3 2 ; G inouvès, P art 3, passim.
105 A esch. Eum. 452, E u r. El. 794, Hipp. 653 w ith B arrett; J . S. R üsten, Z P E 45
( 1982), 284 n. 3. F or R om e see Borner on O v. Fast. 2.35.
106 E u r. Hipp. 123 I'., H orn. Od. 6.85—7.
107 3 S prings, I'heophr. Char. 16.2 ( if E. K . B orthw ick is right, Eranosdi (1966), 106),
M e n . Phasm. 55; 5 springs, E m pedocles B 143; 7, Ap. R hod. 3.860, P hilinna papyrus in
J H S 62 (1942), 36; 14, S u d a s.v. cuiö όις επτά κυμάτων (m u rd erer’s clothes); m ore in
R ohde, 589. F or rep eated w ashings in the sam e spring see Borthw ick, loc. cit., 108.
Disease, Bewitchment, and Purifiers 227
fro m am o n g m e n ’. 108 R eligious w a te r could be fortified; w hen
sa lt w as p u t in this m ay have been sim ply a way o f creating
artific ia l sea -w ate r (th o u g h salt w as a purifying agent in its own
rig h t), b u t o th e r ad d itiv es too a re fo u n d .109 Above all, lu stratio n
differs from w ash in g in its m a n n e r o f application. In rite, there
is a difference in d eg ree b u t n o t in kind betw een sim ple sp rink­
ling a n d to tal im m ersio n . R eligious w a te r is potentially effective
in even th e tin iest q u an tity ; certain crim es, on the o th er han d ,
n o t all th e rivers on e a rth could w ash aw ay .110
G reek s so m etim es spoke o f ‘purifying fire’. 111 T h e elem ent
w as in itself alw ays bright, never stain ed , and th rough the
h e a rth a n d sacrifice it h ad pow erful sacral associations.
T o rc h e s w ere an in d isp en sab le p a rt o f m any cerem onies, an d ,
sw u n g vigorously, they could purify a room or a m a n .112
N o rm ally , how ever, sharp -sm ellin g substances were ad d ed to
th e fire w hen p u rificatio n w as needed. T h e sm oke an d pu n g en t
o d o u r o f su lp h u r, lap p in g ro u n d the polluted object an d p en­
e tra tin g its every p a rt, rendered vividly perceptible th e desired
effect.113 A lread y in H o m er su lp h u r was a ‘cure for bad th in g s’,
a n d th e p u rifier held it in h o n o u r th ro u g h o u t a n tiq u ity .114 H e
108 P a rtic u la r sp rings: e.g. P aus. 2.17.1, cf. E itrem , Opferritus, 84; M oulinier, 71. Sea:
E u r. I T 1193, cf. 1039, S oph. Aj. 655, p. 283 on E leusinian mvstai, D iphilus, fr. 126,
T h e o p h r. Char. 16.13, A p. R h o d . 4.663, Plut. Quaest. Grace. 40, 301 a, LSCG 97 A 15, 151
B 23, ia m b i. VP 153, E itrem , Opferritus, 335 f. and exhaustively W ach sm u th . 2 1 9 -2 3 .
S ea’s p u rity , A esch. Pers. 578, D .L.8.35.
109 M en . Phasma 55 (salt a n d lentils), T h eo cr. 24.97 (salt), H esych. s.v. χερνιβείον
(sa lt a n d b arley g ro ats): cf. E itrem , Beiträge, iii, 8 f., idem , Opferritus, 86. Purifying salt:
S o p h ro n a n d C lem en t cited p. 224 n. 94, schol. Ar. Nub. 1237; E itrem , Opferritus, 323 ff.;
R E s.v. Salz, 2093 f. T h e p o in t o f lentils an d barley gro ats as additives is unclear; b u t cf.
P lu t. Quaest. Graec. 46, 302b.
110 C f. G . B ach elard , L ’Eau et les rêves, P aris, 1942, 193-4: ‘La m eilleure preuve de
cette pu issan ce in tim e, c ’est q u ’elle a p p a rtie n t à ch aq u e goutte d u liquide . . . pour
l’im a g in atio n m atérielle, la su b stan ce valorisée peu t agir, même en q u an tité infime, sur
un e très g ran d e m asse d ’au tre s sub stan ces.’ See too E itrem , Opferritus, 126; ineffaçable
crim es e.g. S oph. 0 7 Ί 2 2 7 f.
111 E u r. Hel. 869, H F 9 3 T , IA 1112, 1471. F o r a pseudo-m edical use see Plut. De Is. et
Os. 383d. F o r Rom e see Borner on Ov. Fast. 4. 727. T h e funeral fire, o f course, purged off
th e im p u rity o f m o rtality : R ohde, 49 n. 41, 334 n. 127, Anth. Pal. 7.49.
112 E u r. / 7 Ί 2 2 4 f., the T o rre Nova sarco p h ag u s (p. 285 below); cf. M . V assits, Die
Fackel in Kultus und Kunst der Griechen, diss. M un ich , 1899, 6 -8 .
113 Penetrat ad viscerum omnes recessus, ac curat saepius loca, quae potiones non potuerunt curare,
V egetius, Mulomediâna, 3.12.1.
114 H o m . Od. 22.481, cf. //.16.228, E ur. Hel. 866, alm ost all th e texts cited a b o v e, p.
224 n. 94; R E s.v. Schwefel, 7 9 8 -9 (B lüm ner); Böm er on O v. Fast. 4.739. O n c a th a r tic
fum igation see E itrem , Opferritus, 2 4 1 -5 0 . B urkert, Grazer Beiträge 4 (1975), 77,
su g g e sts a n o rig in al con n ectio n betw een καθαιρώ and Sem itic k tr, ‘r a iic h e rn ’.
228 Miasma
estee m ed it, o f course, n o t for its ac tu a l disinfectant pow ers, but
b ec au se its d ry acrid sm oke w as sym bolically fit to com bat the
d a m p ro tten n e ss o f im p u rity ;115 pitch w as som etim es b u rn t for
th e sam e p u rp o se .116 F or offerings to the gods, by contrast,
sw eet-sm elling su b stan ces w ere chosen.117
‘W a te r is b est, a n d gold shines o u t like blazing fire’, says
P in d a r; this m etallic hom ologue o f the two uncontam inated
elem e n ts is in fact, like them , a purifier, although the exact
m ech a n ism o f a ‘purification by gold’ is now here specified (it
w as p e rh a p s by sp rin k lin g o f w ater from a gold vessel).118 A nd it
is su re ly as th e p u re st form o f vegetable m atter, a title it claims
for its e lf in C a llim a c h u s’ Iambus,1*9 th a t the laurel expels and
clean ses evil. T h e evidence for this function is a b u n d a n t in
R o m a n sources, a n d no t negligible in G reek; A pollo’s priest
B ra n c h u s cleansed the M ilesians from plague by sprinkling
th e m w ith w a te r from a laurel b o u g h ,120 a n d the sam e m ethod
seem s to be a ttrib u te d to Apollo him self on two vases showing
th e p u rifica tio n o f O re ste s.121 In o th e r contexts too, laurel has
p o w ers for g o o d .122 I t seem s to derive its pu rity not directly from
its p h y sical p ro p erties, nor from its place w ithin the general
115 Cf. C . R. H allpike, The Foundations o f Primitive Thought, O xford, 1979, 160 for such
p a tte r n s o f sy n a esth etic association.
116 D ip h ilu s, fr. 126, S o phron in Page, G LP, p. 330; R E s.v. Asphalt, 1728 f.
1,7 F r. P fister in R E s.v. Rauchopfer, 284.
118 P in d 0 1 .1.1. P urification by gold: LSC G 154 A 29,30,44; B 2,6,15,26;? LSC G 156 A
15; E u r. I T 1216 (w here ed ito rs c o rru p t χρνσώ to πνρσώ)\ Iam bi. VP 153. Sprinkling
from a go ld en vessel, E u r. Ion 434 f. O n ap o tro p aic gold see E itrem , Opferritus, 192-7;
‘p u rify in g ’ bronze, A pollodorus 244 F G rH fr. 110 (banged against eclipses).
1,9 F r. 194. 3 7 -4 4 ; cf. A rtem id. 4.57 p. 282.1 Pack, Pliny, H N 15.135. L aurel in
f u n e ra ry contexts (Pliny, H N 16.239, T zetzes ad Lyc. Alex. 42, AJ A 11 (1907), 72) is
e x c e p tio n a l. O n laurel see M . B. O gle, A JP 31 (1910), 287—311; G o w o n T heocr. 2. 1;
A m a n d ry , 126—34; B örner on O v . Fast. 1.339; K . L em bach, Die Pflanzen bei Theokrit,
H e id e lb e rg , 1970, 5 7 -6 1 .
120 C a llim . fr. 194. 2 8 -3 1 w ith PfeifTer.
121 J H S 89 (1969), P lates 3.3,4.5; A pollo an d A rtem is hold laurel boughs even when
th e p u rificatio n is by p ig’s blood (ibid., P lates 2.1—2,3.4,4.6; cf. M elam pus on the
ca m e o R M L 3.3009). A L u can ian vase in Berlin p erh ap s show s a purification by laurel
(Archäologische Zeitung n. f. 1 (1847), Fig. 7, A. D. T re n d a ll, The redfigured vases o f Lucania,
Campania and Sicily, C ::fo rd , 1967, 150 n. 854). Ion sw ept A pollo’s tem ple w ith a laurel
b ro o m , E u r. Ion 80, 103, 113 if.; those w ho left a de a th house sprinkled them selves w ith
w a te r from a laurel b ra n c h , schol. E ur. Ale. 98, cf. Servius on Aen 6.230. L aurel-
s p rin k lin g in m agical p a p y ri, e.g. P G M 5. 200, S. E itrem , Gnomon 4 (1928), 194 f. For
la u re l used in fum igation, a com m on practice a t R om e (e.g. Pliny, H N 15.135, 138), I
k now for G reece only P lu t. De Pyth. or. 397a.
122 T h e o p h r. Char. 16.2, D .L .4.57; O gle, op. cit., 295 f., 307 ff; R ohde, 198 η. 95; G ow
o n T h e o c r. 2.1.
Disease, Bewitchment, and Purifiers 229
classificatory schem e o f p lan ts, b u t from its an cien t statu s, the
origin of w hich we can scarcely d eterm in e , as the sacred tree par
excellence, d e a r to all the O ly m p ian gods and especially to
A p o llo ,123 m ark o f h o n o u r assigned to those who, like pro p h ets
or poets, are them selves d ear to th e gods. O live b ra n ch es124 an d
w ool fillets125 occasionally a p p e a r as purifiers for sim ilar
reasons. (S uch p rio r sacral significance is the kind o f factor th at,
as h a s been p o in ted o u t, m uch com plicates a sim ple stru ctu ral
a n a ly s is .126) A p u re a n d purifying anim al, like the four-eyed dog
o f Z o ro a stria n ism , seem s not to be av ailab le to set alongside the
p u re m in eral a n d v eg e ta b le.127
W h ile th e processes considered so far dissolve pollution
th ro u g h co n tac t w ith th e p u rest forms o f m atter, others transfer
it in to a b so rp tiv e substan ces, not especially pu re in them selves
a n d p e rh a p s even th e opposite, w hich are then ostentatiously
d isp o sed o f.128 T h e co n tra st betw een the two m ethods, how ­
ever, is a p u rely form al one, as b oth w ere norm ally com bined in
th e sam e cerem ony, a n d even on a form al level not absolute,
since at least one o f th e p u re su b stances, w ater, is itself con­
ta m in a te d by th e d irt it w ashes aw ay .129 T h e passage o f On the
sacred disease th a t b est illu strates th e disposal o f these offscour­
ings h as alre ad y been quoted: ‘T h ey b u ry some o f them in the
g ro u n d , th ey th ro w som e in th e sea, and others they carry off to
th e m o u n tain s w h ere nobody can see or tread on th em .’ (H e
o m its th e com m on expedient o f sending them to the cross­
ro a d s .130) Local trad itio n in th e Peloponnese knew , the spot
s23 A m a n d ry , 127; O gle, 305 f.
124 T h e o c r. 24.98, O rp h . Lith. 214 f. T h o u g h opposed to the laurel as ch th o n ian to
O ly m p ia n , a n d th u s asso ciated w ith funerals (C allim . fr. 194. 40—56), it was united
w ith it a g ain st o th e r trees as sacred ag ain st profane (C allim . fr. 194. 101 if., Pliny, H N
15.135, cf. M u rr, 40—8; Diels, 119—21), and could thus be exploited for purification.
125 T h e o c r. 24.98, N em esianus Eel. 4.63, Clem . Al. Strom. 7.4.26.2, vol. 4, p. 19 St. For
wool as a n alexipharmakon cf. J . Pley, De lanae in antiquorum ritibus usu, R G W 11.2,G iessen,
1911, 80—94; E itrem , Opferritus, 380—6; G ow on T heocr. 2.2. For its high social statu s
see E m ped o cles B 112.6; Pley, 6 8 -7 9 . 126 Cf. p. 365.
127 F our-ey ed dog: Boyce, C h . 6. N ote, how ever, the (?sacred) fish held over initiands
on vases discussed by G . S ch n eid er-H errm an n , Antike Kunst 13 (1970), 5 2 -7 0 .
128 R u d h a rd t, 165.
129 A n d for th e th ro w in g aw ay o flu stral torches see C lau d . Cons. Hon. 329 f.
130 148.44 fT. J . , 1.42 G .: cf. H orn. //.1.314, Ap. Rhod. 4.710, Paus. 2.31.8, 8.41.2; on
th e cro ssro ad s, p. 30 n. 65; sacred laws restricting w here katharmata m ight be throw n
o u t, L S C G 108; IG I3 257 = L S S 4 (d an g er o f step p in g on one, Petron. Sat. 134). O nly
d e s p e ra te p ersons o r desp erad o es would eat such rem ains, p. 30 n. 35. For th eir power
to ‘take u p ’ evil from the purified person seeesp. Clem . Al. Strom. 7.4.26. 1, vol. 3, p. 19 St.
230 Miasma
w h e re M elam p u s h ad b u ried the offscourings of the Proetids, or
co u ld p o in t out th e spring into w hich he threw them , thus
c o n ta m in a tin g it.131 T h e verb ekpempd, ‘send o u t’, norm ally
a p p lie d to h u m an s, is som etim es used o f the disposal o f the
p o llu te d re m ain s, as though there w ere som ething slightly
a n im a te ab o u t th e m .132 T h e purifier w ould em phasize sep ara­
tion from th em by ‘throw ing them over his shoulder’, and
‘w a lk in g aw ay w ith o u t looking b ack’.133
M o st pow erful am o n g these rites of absorptive purification
w as th a t by blood sacrifice, practised for healing by the
p u rifiers o f epilepsy a n d also, according to a south Italian vase
o f th e fo u rth century, by M elam p u s.134 T h e sym bolism o f this
ritu a l is considered elsew here.135 It h ad a v aria n t form, ‘(purify­
ing) a ro u n d by p u p p y ’, in w hich the m ost despised o f anim als
w as used to receive the c a n d id a te ’s im p u rity .136 T h e com ­
m o n est su b sta n c e into w hich evil was transferred, by a process
th a t is now here m ade explicit, was the egg.137 It was perhaps
b ec au se the egg w as a com m on offering to the dead, a n d thus
‘food for co rp ses’, th a t it w as suitable for this u se.138 In m u rd er
p u rifica tio n s, a n d p erh ap s in o th er contexts, the candidate
p lace d his foot on a woollen fleece w'hich absorbed his
im p u rity .139 S ym bolically even m ore d irect w as the technique of
‘w ip in g o ff’ the evil th ro u g h sm earing w ith a clinging substance
131 Pharmaka b uried in ag o ra at Sicyon: Bulletin Epigraphique 69 (1956), 110, 72 (1959),
157. Kalharmata th ro w n into A nigrus m arsh: P aus. 5.5.10; throw n into fountain at
L ousoi - w hence all w ho d rin k from it hate w ine —O v. Met. 15.322-8; Heldensage, 247
n. 4.
132 A esch. Cho. 98, cf. A elian, VH 14.7 έξηλαννετο τής Σπάρτης ώς τά τών νοσονντων
κ α θά ρσ ια , a n d th e ‘sen d in g aw ay ’ o f evils, by a m erely verbal act, to d istan t regions in
apopompë ( cf. S o p h . O T 1 9 4 - 7, Hymn. Orph. 11.23, 14.14, 36.16, 71.11 ; the m any studies
o f ( ). W einreich on apopompë a re listed by H . H erter, Dämonen, 47 n. 12).
133 A csch. Cho. 98, E u r. Andr. 293 f.; R ohde, 325 n. 104; A. S. Pease on Cic. Div. 1.49;
G ow on T h e o c r. 24.96; Börner on O v. Fast. 5. 439.
134 T h e C a n ic a ttin i crater: see m ost recently Antike Kunst 13 (1970), 67, Fig. 1, with
references to o th e r p o rtray als. T h e en passant in terp retatio n o f the C anicattini crater by
K. I*auglotz, Die Kunst der Westgriechen, M unich, 1963, 25, as a n initiation scene fails to
e x p la in the u n m istak ab le A rtem is image.
135 A pp en d ix 6.
136 T h e o p h r. Char. 16.14, P lut. Quaest. Rom. 68, 280b—c; ?cf. Sophron in Page, GLP,
p. 330.
137 See p. 30 n. 65; also An/. u. Chr. 6 ( 1940-50), 5 7 -6 0 . Stengel, 162 speaks o f ‘wiping
oil *w ith egg-yolks, b u t it is clear from Clem . Al. Strom. 7.4.26.3, and L ucian, cited p. 30
η- 65, th a t a lte r use th e ca th a rtic eggs w ere still edible.
138 N ilsson, Op. Sel. i, 3—20.
139 See A ppendix 6.
Disease, Bewitchment, and Purifiers 231
w h ic h w as th e n w ash ed off, b ringing the pollution w ith it. M u d
w as an obvious m a te ria l to choose, since it em phasized the new
sta te o f p u rity by th e g reatest possible co n tra st.140 T h e use o f a
b ra n m ash for th e sam e p urpose is less easy to ex p lain :141
T h e use o f th e lau rel has alread y been discussed. O th e r
p la n ts som etim es describ ed as ‘purifiers’ p erhaps owe th a t title
chiefly to th eir function as alexipharmaka, ‘averters o f d ru g s/
b e w itc h m e n t’, ra th e r th a n to any specific use in p u rificatio n s.142
T h e se alexipharmaka seem to be a com plex class — the strong
sm ell o f th e b u ck th o rn is no less effective th a n the sanctity o f the
la u re l —a n d to discuss them here w ould lead too far afield. O n e
p la n t, how ever, th a t is rep eated ly m entioned as an actual
p u rify in g ag en t is th e sq u ill.143 It seems safe to infer th a t the
p la n t used by th e purifier was the ‘E p im en id ean ’ squill, which,
says T h eo p h rastu s,- ‘gets its nam e from its use’. 144 T h e squill
w as used as a w h ip in scapegoat an d o th er ritu a ls,145 b u t its
a p p lic a tio n in p urifications, now here clearly in d icated ,146 was
n o t necessarily th e sam e. In th e scapegoat ritual, it is associated
w ith d esp ised w ild p la n ts ,147 a n d is elsew here som etim es
sp o k en o f as co n tem p tib le, inedible, even d ead ly .148 It is tem pt-
140 D em . 18.259, G raf, 106.
141 D em . 18.259, ά π ο μ ά ττω ν τφ π η λφ καί τοίςπιτνροις. T h e prospermeia (hapax: b u t cf.
panspermia) o f C o s (LSC G 154 A 29,3 0 , 44; B 2, 6,15,26; LSCG 156 A 15 (restored)) was
p re su m a b ly sim ilar. W iping off w ith μαγίόες, cheese or bran cakes also offered to
T ro p h o n iu s a n d H ecate, is attested by H esych. s.v. μαγίόες-, μαγμόν; cf. Soph. fr. 734
w ith P earson an d R a d t, A th. 149c. T h is p erh ap s perm its the inference th a t here too the
ab sorptiv e su b stan ce is ‘corpse food’: for panspermia offered to the dead see D eubner,
1 Γ2 (C h y tro i).
142 e.g. b u ck th o rn : see R ohde, 198 n. 95; M u rr, 104-6; fig: Cook, 2.ii. 1103; R E 1.55 f ;
M u rr, 3 1 -5 ; G e b h a rd , 69 f.; R ohde, 590; Ιερά βοτάνη o r περιστερεών: R E 1.55; οχίνος:
C ra tin u s , fr. 232, A r. fr. 255, A m eipsias, fr. 25 (w ith asparagus) - I do not u nderstand
w hy in these cases LSJ a n d oth ers take σχίνος to be a squill. T rallian s used vetch in
pu rificatio n s, P lut. Quaest. Graer.46,302b.
143 In katharmoi T h eo p h r. Char. 16.14, D iphilus, fr. 126, Lucian, Nec. 7, Dio C hrys.
48.17, A rtem id . 3.50, p. 225. 13 Pack. H u n g a t do o r or buried under sill as an
alexipharmakon: T h e o p h r. Hist.PI. 7.13.4, D ioscorides, Mat.Med. 2.171.4, p. 239.11 \V.
144 T h e o p h r. Hist.PI. 7.12.1. Squills are norm ally identified as urginea maritima-,
W . T h ise lto n D yer, how ever, in the index to the Loeb Hist.PL, m akes the E pim enidean
sq u ill ornithogalum pyrenaicum.
145 H ipp o n ax , fr. 6, T heocr. 7.107 f. + schol. A ‘squill-battle’ in Inscr. Prien. 112.91,95.
146 Possibly lo r fum igation in C alp. Sic. 5.79. B u rk e rts H ittite parallel, GR 131,
w ren ch es th e squill aw ay from its quite com plex web o f G reek associations, unjustifi­
a b ly . 147 T zetzes, Chit. 5.736.
148 J . N . B rem m er, ‘S cap eg o at-R itu als in A ncient Greece", H SC P 87 (1983), citing
inter alia T heog. 537, A rtem id . 3.50, p. 225.11 Pack, Suda s.v. skilla; they grow on
to m b s, T h eo cr. 5.121.
232 Miasma Disease, Bewitchment, and Purifiers 233
ing, therefore, to see it as the vegetable equivalent o f the im pure
re g u la tio n s h av e close parallels in form , a n d often in detail, in
p u p p y , a d ish o n o u rab le p la n t ap p ro p riately used in a ritual
th e ab ste n tio n s (hagneiai) req u ired o f p artic ip a n ts in p artic u la r
a p p lie d to polluted persons. T h e difficulty, how ever, in in ter­
cults, a n d in th e m in u te regulations for daily living th a t are best
p re tin g the sym bolism o f n a tu ra l species lies in the com plexity
re p re se n te d for us by ‘H esio d ’ a n d th e P yth ag o rean symbola, b u t
o f th e ir possible uses. T h o u g h the squill m ight be aesthetically c e rta in ly d erive u ltim ately from p o p u la r belief. T h e purifiers of
a n d g astro n o m ically despised, the druggist and the horticul- epilepsy, therefore, differed from sim p ler purifiers in digging
tu ra lis t esteem ed it. It h a d a w ide variety o f m edical uses,149 an d d e e p e r th a n th em into the resources o f traditio n al religion.
w as believed (correctly, it is said) to foster the grow th o f seeds E x p lo itin g th ese resources, they p u t to g eth er for th eir patien ts a
a n d shoots p la n te d in the su rro u n d in g so il.'50 It stayed alive for
d istin c tiv e w ay o f life to follow. T h o u g h their m aterials and
a re m a rk a b le len g th o f tim e w hen d u g up, a n d T h eo p h rastu s
e x p la n a tio n s w ere religious, it is plau sib le th a t in doing so they
m ak es this exceptional vitality the reason for its use as an
w ere, consciously o r unconsciously, m im icking the special ways
‘a v e rte r o f spells’. T h e E p im en id ean squill w as actually the one o f life th a t non-religious healers o f the period were prescribing.
e d ib le fo rm ,151 w hich suggests th a t it was the positive qualities T h a t, a t least, is th e charge b ro u g h t ag ain st them by their
o f th e p la n t th a t th e purifier sought to exploit. B ut w ithout
H ip p o c ra tic critic; th eir abstinences, according to him , are ju st
b ein g ab le to see a n d h e a r him m an ip u late the m agic p la n t we d ie te tic p re scrip tio n s in disguise.155
can only guess a t its significance.
A th eo retical issue of im p o rtan ce is raised by the use of
T h e s e v ario u s.tech n iq u es, w hich have h ad to be sep arated in
u n c le a n m ateria ls (blood, m u d ) in som e of these rites. O n e of
d esc rip tio n , te n d e d to be freely com bined in actual use.
th e tru e o b serv atio n s o u t o f w hich the d o ctrin e o f the
D ip h ilu s ’ M elam p u s em ploys torch, squill, pitch, sulphur, an d
‘a m b ig u ity o f th e sac red ’ was b u ilt u p was th at, in some
s e a -w a te r all together, a n d o th er texts show a sim ilar profusion.
c u ltu re s, in som e contexts, pollution acquires positive powers;
T h e rites w ere acco m p an ied by in can tatio n s w hich probably
th e im p u re, n o rm ally shunned, becom es ‘sacred ’ in the sense
c o m p rised form ulas o f transference - ‘m ay th e evil pass into this
t h a t it is m ark e d o u t as pow erful in co n trast to the non-polluted
eg g ’ - a n d analo g y —‘as I w ash off this m ud, so m a y .. .’- a s well
o b jects o f fam iliar u se.156 O bscen e o r blasphem ous language is a
as m o re m ysterious m a tte rs .152 Expressions som etim es occur
co m m o n p lace exam ple. C ertain curren ts in the p opular
w h ich suggest th a t an in can ta tio n could be a ‘purification’ in
m ed icin e o f la te r an tiq u ity m ade conscious use o f the powers of
itse lf.153
p o llu tio n . A lth o u g h bodily w astes w ere p erhaps used as a
T h e purifiers o f On the sacred disease deserve special m ention,
materia medica sim ply because they w ere th o u g h t to have specific
b u t as th e ir m eth o d s have been well studied in detail it can be
v irtu e s like an y o th e r su b sta n c e ,157 there were au th o rities who
q u ite s u m m a ry .154 In ad d itio n to the actu al purifications, they
explicitly reco m m en d ed the u n speakable. ‘(In the trea tm e n t of
s u b je c te d th e ir p a tie n ts to various ab sten tio n s (from bathing,
fevers) D em o critu s says pollution is needed, for instance blood
a n d from p a rtic u la r form s of food), a n d rules o f life (such as not
g u ilt (?), m en stru al blood, th e flesh of sacred birds o r forbidden
to w e a r black, o r ‘p u t foot on foot a n d h an d on h a n d ’). T hese
a n im a ls given as food, a n d d ra u g h ts o f blood.’158 H on o u rab le
R o m a n a u th o rs record w ith revulsion the belief th a t hu m an
149 See G e b h a rd , 69 n. 28; R E 1.67 f.; ibid., 3 A 5 2 2 -6 ; K.. L em bach, Die Pflanzen bei
Theokrit, H eidelberg, 1970, 6 3 - 5 ; Dioscorides, M at.Med. 2. 171; Pliny, H N 20.97-101.
155 Morb.Sacr. 1 4 2 - 4 .6 J ., 1 .1 2 -2 3 G.
150 'I'h e o p h r. Hist. PI. 2.5.5, 7.13.4; R E 1.67; ibid., 3 A 523 f.
156 Cf. S tein er, 66, D ouglas, C h. 10.
151 T h e o p h r. Hist. PL 7.12.1.
157 R. M u th , Träger der Lebenskraft, Ausscheidungen des Organismus im Volksglauben der
152 Cf. H ip p o cr. Morb. Sacr. 138.10, 140.13, 148.34 J ., 1.4, 12, 39 G .; Arist. fr. 496 '
Antike. V ien n a. 1954.passim.
Rose; C a llim . F r. 194.30; cf. Boyancé, 37. F or surviving incantations see R. Heim ,
138 T h e o d o ru s P riscianus, Physica, p. 2 5 1 .2 -5 Rose. O n the pseudo-D em ocritean
‘I n c a n ta m e n ta m agica g raeca —l a t i n a Jahrb. f . klass. Phil. Suppl. 19 (1893), 463—576
lite ra tu re see R E S up p l. 4.219—23; on Bolus (its m ost im portant representative),
a n d F r. P fister in R E S uppl. 4 s.v. Epode.
b ib lio g ra p h y in Oxford Classical Dictionary2, s.v. Bolus; index t o j . Bidez/F. C um o n l, Les
153 A rist. fr. 496, D iod. 3 .5 8 .2 -3 . 154 L a n a ta , passim; cf. Lloyd, 37 f.
Mages hellenisés, P aris, 1938.
234 Miasma
blood, sm eared on the lips o r d ru n k hot from a glad iato r’s fresh
w o u n d s, could cure epilepsy.159 T h e abom ination, transferred
to th e blood o f m arty rs a n d executed crim inals, is said to have
co n tin u e d u ntil recen t tim e s.160 O th e r cures for epilepsy were
w a te r d ru n k from a m u rd ered m a n ’s skull (A rtem on), the flesh
o f a b ea st slain by the sam e w eapon as h ad killed a m an
( O r p h e u s a n d A rc h elau s’), g o at’s m eat roasted on a funeral
p y re (‘th e m a g i’) . 161 B ut it w ould p robably be w rong to father
su c h a ttitu d e s on the purifiers o f On the sacred disease. A lthough
p u rific a tio n by blood w as certainly a confrontation w ith the
horrific, it was a co n frontation licensed, in o th er contexts, by
tra d itio n a l religion, a n d it w as not the defiling pow er o f blood in
itse lf th a t m ade the cerem ony effective; the blood was a token o f
th e p o llu tio n th a t w as to be rem oved.162 S im ilar cerem onies,
like th e cleansing o f the P roetids according to D iphilus, did not
use p o llu tin g ag en ts a t all. In the early period, it is only for
h a rm fu l m agic th a t w e find im p u rity being sought ou t and
e x p lo ite d .163
A s w e no ted initially, the purifier was an object o f contem pt
to th e en lig h ten ed by the fourth century. Before intellectuals,
his m eth o d s could no t well be defended. Intellectuals, however,
co u ld sim ply be ignored. T hese p ractices continued; the
p u rifie r re ap p ea rs (as an old w om an) in R om an love elegy, and
in th e fo u rth ce n tu ry o f o u r era still th re a te n e d to lure w avering
C h ris tia n s into s u p e rstitio n .164
159 e.g. P liny, H N 2 S A .
160 F. J . D ölger, Vorträge der Bibl. Warburg 1923/4, 196-214, esp. 2U4 (F.; A bt, 199
(273), η. 10.
161 P liny, H N 28.8,34,226. M ore bloody epilepsy cures in O . T em kin, The Falling
Sickness2, B altim ore, 1971, 22 f., a n d on the use ofb lo o d cf. E itrem , Opferritus, 4 4 1 -7 .
162 See A pp en d ix 6.
163 p. 222 an d p. 223 n. 85. I am not convinced by the counter-instances o fV e rn a n t,
Socié'é, 137 I. R itual obscenity, of course, w hich does occur, is a related phenom enon.
164 C f. Constitutiones Apostolorum 8.32.11: the μάγος, έπαοιδός . . . περιάμματα ποιών,
περικαθαίρω ν to be excluded from com m union until reform ed; Concilium Ancyranum,
C a n o n 24, in C . |. H elele, Histoire des Concites, tr. H . L e d e rq , Paris, 1907- , i.324:5
y e a rs p en a n c e tor those w ho in troduce m agicians into th eir houses έπΙ άνενρέσει
φαρμακείων<ή και καθάρσει; the tem p tatio n to infidelity cited by A ugustine: ‘Sed ecce
a d s ta t vicinus et am icu s et an cilla . . . ceram vel ovum m anibus ferens et dixit “ F ac hoc
e t sa lv u s eris. Q u id prolongas tu a m aegritudinem ?” ’ (Revue Benedictine 54 (1938), p. 8.
121). F or th e use o f baptism as a rite o f exorcism see K . T h ra e d e in R A C 7.76 ff.
8
DIV IN E VENGEANCE AND DISEASE
A claim su ch as ‘Y o u ’ll go m ad if you en ter th at p recin ct’ has
tw o levels o f significance. 11 is on th e one h an d a w ay o f insisting
on th e san c tity o f th e g ods’ p ro p erty , a th rea t in su p p o rt o f a
v alu e ra th e r th a n a hypothesis ab o u t the causes o f disease. T h e
system o f tab o o is not, as it has seem ed to som e observers, the
p ro d u c t o f a cu ltu ra l neurosis, b u t a way in w hich ‘attitu d e s to
v alu es a re ex pressed in term s o f d a n g e r’. 1 O n the o th e r h an d , it
is n a tu ra l th a t, if a case o f m adness occurs in a society w here
s u c h th re a ts a re rife, b reach o f th e rule should be suspected as
its cause. In th e previous ch ap ters, we have surveyed a large
n u m b e r o f religious dangers, b u t prim arily in th eir role as
san c tio n s. I t rem ain s to consider to w h a t extent they were
a p p e a le d to in ex p lan atio n o f actu al afflictions. W here they
w ere no t a p p lied , it will obviously be necessary to pay a tten tio n
to th e a lte rn a tiv e ex p lan atio n s th a t su p p lan ted them .
U n fo rtu n ately , th e pro g ram m e here outlined is forbiddingly
v ast. T h e re is no special area o f experience to w hich the
o p e ra tio n s o f destin y , luck, or divine an g e r are confined, and,
a lth o u g h in ex p licab le h ap penings are liable to be especially
‘d iv in e ’, th ere a re m an y instances o f the gods’ will being worked
o u t th ro u g h events th a t G reeks could easily in terp ret in h u m an
term s. A m a n ’s s ta n d in g in relation to th e gods or destiny will
affect, a m o n g o th e r things, his h ealth , w ealth, length of days,
p ro c rea tiv e pow ers, success in farm ing, business or politics,
m a rita l or p a re n ta l fortunes. In w h a t follows, the restricted case
o f resp o n ses to disease will, for convenience, alone be con­
sid ere d . As it h ap p e n s, the evidence is scarcely available
th ro u g h w hich to consider some o f the o ther situations; our
so u rces do n o t reveal th e inm ost feelings of a father whose sons
h a v e d ied on th e verge o f m anhood, of a wife unable to produce
h eirs for h e r h u sb a n d , o f any o rd in ary individual all o f whose
1 S tein er, 21.
236 M iasma
p ro jec ts go inexplicably aw ry. B ut the question could certainly
be co n fro n ted m ore g enerally th an will be a ttem p ted here.
S erious disease is a n affliction th a t w renches the p a tie n t’s life
o u t o f its cu sto m ary unreflective course. F rom an in dependent
ag e n t, m a ste r o f his ow n affairs, he has becom e the prey o f
e x te rn a l forces he ca n n o t u n d ersta n d . H e wakes up one m o rn ­
in g ill; b u t the previous d ay he h ad been well, and he has not
e a te n o r d ru n k a n y th in g un u su al, o r changed his habits in any
w ay. ‘By d ay a n d by n ig h t diseases o f them selves (automatoi)
com e u p o n m an , a n d do him h arm , silently; for cunning Zeus
took o u t th e ir voice’, says H esiod.2 T h is brings o ut three crucial
facts a b o u t diseases, th a t they are uncontrollable, inexplicable,
a n d hateful. T h e q u estio n o f why they occur is not p erh ap s of
g re a t in te re st in itself (no one ever consulted an oracle to
e sta b lish the cause o f a disease safely overcom e) ; b u t d u rin g the
affliction a n e x p lan a tio n th a t will p erm it control becom es
v itally im p o rta n t.
Illn ess becom es com prehensible w hen it ceases to be a
ra n d o m event. In J u d a is m , as fam ous incidents in the gospels
show , it is sin, w h e th e r ritu a l o r m oral, th a t causes disease,* and
this id ea seem s to h av e been w idespread in the A ncient N ear
E a s t.4 T h e co rrect p ro ced u re for the sick m an o r his friends is to
d ia g n o se th e re le v an t sin, confess it publicly, a n d m ake a
sacrifice o f ato n em e n t. I f th e evil continues, fu rth er sins m u st be
e sta b lish e d (p e rh a p s w ith the help o f an oracle) an d publicly
co nfessed.5 T h is link o f sin w ith disease is found th ro u g h o u t the
w o rld .6 A m ong v ario u s N ilotic peoples, for instance, a sin, such
as in cest, a n d th e disease th a t it is believed to cause m ay bear
th e sam e n a m e .7 I f such an offence is com m itted, sacrifice will
p ro b a b ly be m ad e im m ediately to p revent the onset o f the
2 Op. 1 0 2 -4 .
3 See M a rk 2 :1 -1 2 (M a tth e w 9 :1 -8 , L uke 5 :1 8 -2 6 ), J o h n 5 :1 -1 5 , 9 .1 -2 , Epistle of
J a m e s 5 :1 4 -1 5 . In th e O ld T e sta m e n t e.g. Psalm 39,41:4, 103:3, 1 0 7 :1 7 -2 0 .1 have not
seen YV. v. S ieb en th al, Krankheit als Folge der Sünde, H anover, 1950.
4 See L a tte , K l. Sehr. 32 n. 42, G oltz, 7—10, an d a t length R. Pettazzoni, La Confessione
dei peccati, B ologna, 1 9 2 9 -3 5 , vols, ii and iii.
5 Λ good exam ple o f this search for the responsible sin is the H ittite ‘P rayers o f K ing
M u rsilis in a tim e o f p la g u e ’, Ancient Near Eastern Texts,3 ed. J . B. P ritchard, Princeton,
1969, 3 9 4 -6 : cf. R . P ettazzo n i in Occident and Orient, Studies in Honour o f M . Gaster,
L o n d o n , 1 9 3 6 ,4 6 7 -7 1 .
6 See R . P ettazzoni, op. cit., vol. i.
7 E v an s—P ritc h a rd , 184, L ie n h a rd t, 284, B uxton, 194.
Divine Vengeance and Disease 237
disease. W h e n illness actu ally occurs, th e diagnostic procedure
th a t follows is likely to identify ‘sin ’ as the cause, th o u g h o th er
p o ssib ilities exist.8 In all cases, diseases com e for definite
reaso n s. T h o se sen t by spirits can only be cured by religious
m ean s, th ro u g h sacrifice, though this is not to say th at m edi­
cines sh o u ld be ignored. O n the co n trary , everything possible
sh o u ld be d o n e on the p ractical level; b u t a final cure will
d e p e n d on G od. S u ch a set o f beliefs serves both to explain the
a p p a re n tly ra n d o m affliction, a n d also to indicate a practical
co u rse o f action: th ere is a god or sp irit to be appeased. T h e
psychological im p o rtan ce o f p u ttin g th e disease in professional
h a n d s (w h e th er d o cto r o r diviner) a n d treating it positively
(w h e th e r by p rescrip tio n o r sacrifice) is very great, as the
effectiveness o f the p lacebo p roves.9
F o r G reece, sch o lars have noted the supposed daem onic or
d iv in e origin o f v arious illnesses.10 B ut the fact th a t a god is
re sp o n sib le for disease does not reveal m uch ab o u t it. H as it
b een sen t as a p u n ish m en t, or in caprice? O r is the idea of the
g o d -se n t d isease sim ply a w ay o f expressing h u m an incom pre­
hensio n ? T h e su b ject does n o t seem to have interested
h isto ria n s o f d iv in e ju stic e a n d h u m an responsibility.11 Yet it
allow s a n in te re stin g confrontation o f religious theory and
th e ra p e u tic p ractice. M oralists m ay have seen disease as a
p u n ish m e n t; w ere religious a n d m agical cures based on the
s a m e belief?
F o r su ch a n investigation, the existence o f ‘scientific’ G reek
m ed icin e p resen ts a com plication. It is not o f course th at all
G reek s a t all tim es accep ted a n atu ra l acco u n t o f the causation
o f disease. E arly poets an d m yths reveal pre-H ip p o cratic con­
d itio n s, a n d b eh in d th e im posing edifice o f H ipp o cratic ratio n ­
alism we can alw ays detect sp irits w ho p u t less faith th an the
d o c to rs in th e d elicate d ietetic balancing o f the body’s
8 F o r a n acco u n t o f the diagnostic procedure in one case sec L ien h ard t, f>8-()2.
9 O n the reassu rin g function o f consultation see U n a M aclean, Magical Medicine,
L o n d o n , 1971, C h. 1; on th e ‘placebo eflfect’ T h o m as, 248 f.
10 Cf. L a n a ta , esp. 2 8 -3 9 ; E delstein, A M 2 1 9 -2 4 (rightly restrictive); H ertei,
Dämonen; L loyd, 29 n. 98.
11 B ut cf. L a n a ta , 28—39 an d F. K udlien, ‘E arly G reek P rim itive Medicine*. Clio
Medica 3 ( 1968), 305—36. \ \ . R. H alliday, ‘Som e Notes on the T reatm en t o f Disease in
A n tiq u ity ,’ in Greek Poetry and Life (Essays presented to Gilbert Murray), O xford, 1936,
277—94, does n o t d iscuss the theological problem .
 Miasma Divine Vengeance and Disease 239
238
on ly co n v in cin g w h ere it is obviously successful, b u t H ippo­
h u m o u rs .12 O n a sim ple practical level, it is unlikely th a t
c ra tic m edicine can never have enjoyed em pirical confirm ation
scientifically tra in e d doctors ever p en e trated far into the rural
o f th a t kind. Its history in later a n tiq u ity shows how fragile and
a re a s o f G re e c e .13 T h e problem is not so m uch th at rationalism
ill-p ro tected is a m aterialist m edicine w ith o u t effective p h a rm a ­
su p p re sse d religious m edicine, as th a t it deprived it of its voice.
ce u tic al s u p p o rt.19 E ven in w estern society today, w here most
A p a rt from d o cu m en ts concerning th e cu lt o f Asclepius, such
p eo p le h av e form ally ad o p ted a scientific view o f the causation
p ra c tic e s a re m ostly m entioned by those unsym pathetic to
o f disease, research has show n th a t, u n d er the stress o f really
th em , like th e polem ical a u th o r o f On the sacred disease, o r sim ply
serio u s illness, alm ost everybody reverts a t least in p a rt to
p assed over in silence. In this sense, H ip p o cratic m edicine is an
d ifferen t form s o f ex p lan atio n . C h ild ren blam e th eir illnesses on
o b sta c le to the p re sen t inquiry. B ut it is also in itself a proble­
p e tty thefts, o r p lay in g too h ard , o r th eir p a re n ts’ unkindness;
m atic p h en o m en o n th a t d em an d s ex p lan a tio n .14 A m aterialist
a d u lts on th e ir ow n im p rudence, th eir unsatisfactory personal
m ed icin e, in a w orld w here science is pow erless to prove its
re la tio n sh ip s, th eir m oral faults. I t is only in term s o f w h a t is
p o stu la te s, has no m ore claim to p o p u lar su p p o rt th a n the
h u m a n ly sig n ifican t th a t h u m an suffering becomes truly com ­
psychologically m ore satisfying a rts o f the diviner. W ith o u t
p re h e n sib le .20
su c h proof, its theories can only be a kind o f dogm a, even for the
T h e w ays in w hich H ip p o cratic m edicine achieved p lau si­
p h y sician him self. T h e w eakness o f n a tu ra l aitiology can be well
b ility h av e recen tly been b rilliantly an aly zed .21 Its origins in
seen from th e fate o f w estern m edicine w hen in troduced into
G re ek folk-m edicine, th e co n tin u in g resem blance o f m any o f its
tra d itio n a l societies. A com m on reaction seems to be to dis­
m eth o d s to th o se o f religious healers, an d the persuasive skills of
tin g u ish ‘E u ro p e a n ’ o r ‘d o cto r’ diseases - acu te conditions and
its p ra c titio n e rs em erge as im p o rta n t factors. A possible
o th e rs for w hich w estern pharm acology is conspicuously suc­
o b sta c le w ould h ave been a strongly held theological doctrine,
cessful - from ‘n a tiv e ’ diseases, ‘diseases the doctors d o n ’t
o f th e k ind a lre ad y m en tio n ed , th a t disease is the consequence
k n o w ’, only to be tre a te d by trad itio n al, perh ap s m agico-
o f sin. T h e re is, o f course, no d o u b t th a t diseases were on
religious m e th o d s.15 T h e d o cto r m ay be assim ilated to the
occasio n theologically explained by the Greeks, ju s t as d eath
h e rb a list, in c o n tra st to the diviner, as one w ho treats sym ptom s
w as, a n d th ere w as certainly n o th in g re p u g n an t to G reek
w ith o u t seeking o u t the underlying cau se.16 W estern p re­
th o u g h t in th e id ea o f divine re trib u tio n taking this form. M ost
ten sio n s to tre a t by n a tu ra l m eans clearly su p ern atu ral psycho­
o f th e im p o rta n t m y th s can be analysed by a schem a o f crim e
logical con d itio n s a re often viewed w ith polite scepticism ;17 in
a n d p u n is h m e n t,22 a n d though th e p u n ish m en t is com m only
g e n e ra l, even w hen E u ro p ean rem edies are used, native
d e a th , diseases too a re found;23 m ad n ess in p artic u la r often
m eth o d s a re ap p lied sim u ltan eo u sly .18 E uropean m edicine is
o c c u rs .24 Lysias, in his speech ag ain st C inesias, m entioned the
12 C f. L a n a ta . 15 f., 7 1 -6 . im p io u s d in in g club, th e kakodaimonistai, to w hich C inesias and
13 I h e ease lor a sh o rtag e o f doctors is p ut by L. C ohn-H aft, The Public Physicians o f
his frien d s belonged. T h e o th er m em bers, he says, h ad all died,
Ancient Greece, M a ssach u setts, 1956,23—31.
14 Cf. Thom as, C h . 7, esp. 226, 245. F or the English peasant, G alenic prescriptions
” See E delstein, A M 231—5 ; 0 . T em kin, The Falling Sickness1, Baltim ore, 1971,23—7.
w ere no m ore ratio n al th a n the m ethods o f the herbalist o r conjurer. 20 See R. H . B lum , The Management o f the Doctor Patient Relationship, New York, 1960,
15 S ee e.g. D. B. Je llille an d F. ]. B ennell, Journal o f Pediatrics 57 (1960), 252; J . B.
6 3 - 5 . F or the rem ark ab le diversity o f ex p lanation in a m odern Greek peasant com-
L o u d o n . The Health Education Journal 15 (1957), 98; R. H . an d E. Blum , Health and
m u n ity -en v iro n m en t, stress, em otional d istu rb an ce, hostile spirits, evil eye, sorcery,
Healing in Rural Greece, S tan fo rd , 1965, index s.v. illnesses ‘which doctors don Vknow
ritu a l a n d m oral failings —see R. H . an d E. Blum , Health and Healing in Rural Greece,
16 S ee M . G elfan d , The Central African journal o f Medicine, 1 (1955), 125.
S ta n fo rd , 1965, C h . 9.
17 N ot w ith o u t reason, as ‘Prim itive psychotherapy, in particular, can com pare
21 L lo y d , C h s. 1 a n d 2. Cf. p. 213 above.
fav o u ra b ly with its m odern rivals’ ( Thomas, 245 w ith references); cf. C . L évi-Strauss,
22 See V ickers, C h. 5, esp. 2 5 2 -5 .
Structural Anthropology (tra n s . C . Ja c o b so n ), L ondon, 1968, C h . 10, ‘T h e effectiveness o f
23 See W . R oscher, Rh. Mus. 53 (1898), 169-204, an d in R M L s.v. nosoi.
sy m b o ls’. 24 See J . M a ttes, Der Wahnsinn im griechischen Mythos und in der Dichtung bis zum Drama
18 S ee R ead, C h . 4; U . M a clean , Magical Medicine, L ondon, 1971 (a detailed study of
desfü n ften Jahrhunderts, H eidelberg, 1970, esp. 5 0 -2 .
th e p h en o m en o n in Ib a d a n ).
240 M iasma
‘as o n e m ig h t expect, being the m en they w ere’, b u t the gods
h a d re serv ed C inesias him self for a w orse fate, as an exam ple to
m a n k in d : ‘T o die o r fall ill in the n orm al w ay is com m on to us
all; b u t to live as C inesias has done for so m any years, to be
alw a y s dy in g a n d yet never m anage to die —th at is kept only for
th o se w hose crim es have been like his.’25 Such is Lysias’
diag n o sis o f the causes o f chronic illness. B ut though it is one
po ssib le in te rp re ta tio n , th ere is no p ro o f th a t it was the usual
o n e, n o r th a t the victim him self w ould have looked for an
e x p la n a tio n o f this kind. M yth, m oral principle, a n d rhetoric
n eed to be tested ag a in st the actu al responses o f the afflicted.
B efore the fifth cen tu ry , evidence is very scanty, a n d it would
b e ra s h to build m u ch on a few passing allusions in H o m er and
H esio d . B u t for w h a t it is w orth these suggest a view o f disease
th a t is fatalistic ra th e r th a n m oral. T h e Cyclopes,, supposing
P o ly p h e m u s’ cries to be d u e to an acu te in tern al disease, com ­
m en t: ‘T h e re ’s no escaping a disease sent by Zeus; so pray to
y o u r fa th e r P oseid o n .’26 W hen they speak o f the disease as sent
by Z eu s, they do no t seem to be thin k in g o f a punishm ent; this
so u n d s like a n am o ra l Z eus w ho d istrib u te s ‘good a n d b ad to
ea c h m an , as he w ishes’. O f the cause o f the disease, no m ore is
said . As to its tre a tm e n t, there is no h in t o f a diagnostic process
to e sta b lish w hy it has been sent. P olyphem us is recom m ended
to invoke th e aid o f a god w ho will on personal grounds be well
d isp o sed to him , his father Poseidon. O n e o ther H om eric
p a ssa g e is sim ilar in im plication. A sim ile com pares O dysseus’
d e lig h t a t sig h tin g d ry la n d to the delight felt by children whose
fa th e r has long lain to rm en ted by illness, ‘a n d a hateful daim on
h a s a tta c k e d h im ’, w hen the gods finally cure him o f his suffer­
in g s.27 ‘A n d a hateful d aim o n attack ed . . . ’ is an expression th at
co u ld be used o f a n y m isfortune,23 a n d one th at, even if it does
n o t exclude, certain ly does not encourage a theological ju s ti­
ficatio n . F ro m this atta c k by a ‘h atefu l’ (not a ju st) agent, the
v ictim ;s rescued by th e gods. I t is the sam e picture o f divine
fa v o u r o p p o sed to m alevolent disease.
25 F r. 53 T h a lh e im = 5 G ern et, ap. A th. 552b. A sim ilar interpretation o f chronic
d ise ase, D iod. 16.61.3. O n the special h o rro r o f long disease see K udlien, 106-24.
26 Od. 9.411 f. B ut D odds, 67 an d M attes, op cit., 31, suppose th a t the Cyclopes take
P o ly p h em u s to be m ad.
27 Od. 5 .3 9 4 -7 . Stugeros o f disease also II. 13.670.
28 Cf. Od. 10.64.
Divine Vengeance and Disease 241
H esio d know s th e origin o f this bane. T h ere was a tim e, he
says, w h en m a n lived free from all evil, labour, a n d disease.
T h e n P a n d o ra ’s box was op en ed , a n d now: ‘T h e land is full o f
evils, a n d so is th e sea .’ T h e re follows the passage ab o u t the
activ ities o f diseases th a t has alread y been quo ted . T h ey roam
a t larg e, u n ac co u n tab le a n d irratio n al, controlled, it seems,
n e ith e r by god n o r m an. T h ey are free agents, an d agents of
evil.29 I t h as b een claim ed th a t in this H esiodic passage we see a
m o ra liz a tio n o f th e concept o f disease, in th a t it was a crime,
P ro m e th e u s ’ defiance o f th e will o f Zeus, th a t b ro u g h t them
u p o n m a n for th e first tim e.30 B ut th e m y th ’s ethos is th a t of the
ju s t-s o story, n o t th e theodicy; w h a t G reek ever seriously
th o u g h t o f referring his troubles to P ro m eth eu s’ crim e?31 O ne
h a s only to co n tra st the en d u rin g explanatory, justificatory
p o w e r o f th e Je w ish m yth o f the Fall to app reciate this.32 T h e
P ro m e th e u s /P a n d o ra m yth em phasizes the irreversibility and
in e sc a p a b ility o f certain ills; b u t far from m oralizing the indi­
v id u a l’s sufferings it im plies m u ch random ness in th eir d istri­
b u tio n .
I t is p e rh a p s h a rd to believe th a t this is the full story. Divine
a n g e r is a u b iq u ito u s them e in H o m er,33 an d individuals may,
w h e n in tro u b le, su sp ect th a t they have ‘sinned ag a in st’ a god.34
T h e re w as m ateria l here for th e healer-seer to w ork w ith. T h e
sam e m ay be said o f H esiod. H is general philosophy would
fav o u r th e a tte m p t to explain disease as a consequence o f crime.
In p a rtic u la r, it is easy to suppose th a t the p enalty aw aiting
th o se w ho in frin g ed the rules o f conduct n ear the end of the
Works and Days m ay have taken this form .35 A nyone who
ob serv ed scru p u lo u sly rules such as these, or th eir P ythagorean
d e sc e n d a n ts, will surely have been inclined to seek a religious or
29 Op. 9 0 -1 0 4 .
30 F. K u d lien , Clio Medica 3 ( 196B), 315 f.
31 Cf. L ie n h a rd t, 33—7, 53—5 on a sim ilar, essentially non-m oral Fall m yth; idem , in
International Encyclopaedia o f the Social Sciences, New York, 1968, s.v. Theology (Primitive),
c itin g N ad el ‘T h e only problem in N upe theology is the actu al pow er o f evil, not its
o rig in .’
32 F o r a tellin g ex am p le see J . K . C am p b ell, ‘H onour and the D evil', in Honour and
Shame, the values o f mediterranean society, ed. J . G . Peristiany, L ondon, 1965, esp. 152-9
(‘th e sin s o f A d a m ’).
33 J . Irm sc h e r, Götterzom bei Homer, diss. Berlin, 1949.
34 p. 201 above.
35 Op. 706—64; early evidence, even if not genuine Hesiod.
 242 Miasma
a t least a m agical explanation for his personal m isfortune.
U n fo rtu n a te ly , it is im possible to ad vance beyond such general
pro b a b ilities.
T h e re follows a chronological gulf. E m pedocles’ ‘’Purifica­
tions', M u s a e u s ’ ‘Cures fo r disease'31 have perished. T he next
su b s ta n tia l evidence com es from H erodotus, an a u th o r w ith a
re p u ta tio n for pious credulity a n d insistent m oralizing. D espite
this, he can speak o f serious illnesses a n d su dden deaths as
a p p a re n tly n a tu ra l events.38 A ‘wise w a rn er’ can num ber
diseases a m o n g those am oral trials im posed by a jealous god
th a t re n d e r d e a th a sw eet refuge for m o rtals.39 T h e in te rp re ta ­
tio n o f disease as p u n ish m en t does, certainly, also occur; thus
th e h o rrib le fate o f P heretim a, w ho seethed w ith w orm s while
still alive, show s th a t the gods resent excessive severity in
re v en g e.40 S im ilar beliefs were held, according to H erodotus, by
th e G reek w orld a t large. In the case o f C leom enes, it w as the
g e n e ra l opinion, th a t he w ent m ad because he co rru p ted the
P y th ia , b u t the A th en ian s referred to his devastation o f the
E le u sin ia n p re cin ct a n d the A rgives to a sim ilar offence against
o n e o f th e ir sacred groves.41 T h a t division shows the excessive
n e a tn e ss w hich h as b ro u g h t H e ro d o tu s’ source indications in
g e n e ra l u n d e r suspicion,42 b u t even on the m ost sceptical view
h e a sc rib ed opinions to his inform ants th a t he believed they
m ig h t have held. H e m entions the intrig u in g case o f O tanes,
w ho resettled d e p o p u la te d Sam os ‘because o f a dream -vision
a n d a disease th a t afflicted him in the genitals’.43 Physiological
a n d religious ex p lan atio n s o f disease are twice presented as
a lte rn a tiv e s a n d , interestingly, H ero d o tu s him self ju d g es dif­
feren tly in th e tw o cases. H e cannot accept the S p a rta n s’
37 A r. Ran. 1033.
38 e.g. 1.161, 7.117.1. O n H ero d o tu s’ com plex attitu d e to divine a n d natural
c a u sa tio n cT. L loyd, 30 f.
39 7.46.3.
40 4.205
41 6 .7 5 .3 ,8 4 .
42 D. F ehling, Die Quellenangaben bei Herodot, Berlin, 1971, passim. In 1.105.4
H e ro d o tu s ascribes to th e S cythians the explanation o f th eir ‘female disease’ as a
c o n seq u en ce o f sacking the a n c ie n t tem ple o f A phrodite O u ra n ia a t Askalon. W . R.
H a llid a y , A B S A 1 7 (1 9 1 0 -1 1 ), 9 5 -1 0 2 , long ago pointed o ut how unlikely this was to be
a tru e S cythian doctrine. O th e r religious explanations o f disease attrib u te d to
foreigners: 1.138, 2.111.2.
43 3.149.
Divine Vengeance and Disease 243
n a tu ra l a c c o u n t o f th e m ad n ess th a t destroyed as g reat a sinner
as king C leo m en es, b u t thinks it ‘not im p lau sib le’ th a t
C a m b y se s’ m ad n ess was a co n sequence of congenital epilepsy
ra th e r th a n o f a crim e ag ain st A pis.44 It is noticeable th a t the
choice h ere is betw een in te rp re tin g disease as a god-sent
p u n is h m e n t a n d as a n a tu ra l event. In a fragm ent o f E uripides,
by c o n tra st, w h ere ‘divine’ diseases are co n trasted w ith those
th a t a re ‘self-chosen’,45 the disease’s divinity seems to consist
m erely in its in escap ab ility , like H o m er’s ‘disease o f Z eu s’. W e
sh all see too th a t th ere w ere o th e r ways in’ w hich in the fifth
c e n tu ry a d isease could com e from th e gods w ith o u t being a
p u n ish m e n t.
O n e p ro b lem in ju d g in g H e ro d o tu s’ evidence is to know
w h e th e r th e diseases o f o rd in ary people had the sam e causes as
th o se o f th e g re at. T h e sam e problem applies to tragedy.
N e ith e r trag e d y n o r Io n ian history can be dism issed, on the
m o ral level, as m ere rom ance. O n th e o th er h an d , it is obvious
th a t th e fo rtu n es o f a C roesus h a d a high dignity an d signifi­
c a n c e d e n ie d to o rd in ary people, ju s t as the hom e life o f m ost
A th e n ia n fam ilies w as not m u ch like th a t o f the Pelopids. T h e
lives o f kings w ho exercised a decisive influence on th e course of
h isto ry w ere easily ab so rb ed into th e sch em ata o f crim e and
p u n is h m e n t ch a rac te ristic o f m yth. W ould H erodotus, sm itten
by a ch ro n ic disease, have believed his ow n affliction to be
eq u a lly rich in significance?
A re c e n tly p u b lish ed com ic fragm ent, probably from A risto­
p h a n e s ’ Heroes, provides in terestin g evidence, in a context th at
is n o t elev ated o r m ythological, for the m oral in terp retatio n of
d isease .46 T h e ch o ru s o f H eroes here an n ounce th at: ‘W e are the
g u a rd ia n s o f good things a n d ill; we w atch out for the unjust, for
ro b b e rs a n d footpads, a n d send them d is e a s e s -s p le e n , coughs,
d ro p sy , c a ta rrh , scab, gout, m adness, lichens, swellings, ague,
fever. T h a t ’s w h a t we give to thieves . . A n o th er comic
44 6.84, 3.33. M a d n ess from physical causes is an accepted fact in X en. Mem. 3.12.6.
PI. Phdr. 265a d istin g u ish es m adness caused by h u m an disease from divine m adness.
P hysiological m ad n ess also Leg. 9 3 4 c -d , Ti. 86b; cf. the obscure rep o rt o f Empedocles
A 98. T h e ‘black bile’ theory o f m adness, com m on in the late 5th century, m ay well
h ave ea rlie r roots: cf. p. 246 n, 61.
45 F r. 292; p ro b ab ly , it is true, m etap h o rical diseases. See too Soph. Aj. 184 I.
46 A r. fr. 58 in C . A u stin , Comicorum Graecorum Fragmenta in Papyris Reperta. Berlin,
1973.
244 M iasma
fra g m e n t p e rh a p s testifies to the sam e belief: ‘I com m itted an
offence ag a in st a h ero .’47 T h e new fragm ent provides welcome
s u p p o rt for the idea th a t the heroes in G reece play the p a rt
assig n ed in o th e r religions to the ancestors. W hile the dead in
g en e ral, except for those disp atch ed violently, seldom seem to
in te rv e n e in th e affairs o f the living,48 the heroes are constantly
active. H ere, as in a passage o f H esiod, they exercise m oral
su p erv isio n over the co nduct o f m en o f the present d ay like true
an c e sto rs.49 O n the o th e r h an d , the heroes also h ad a re p u tatio n
as tro u b le-m ak ers, beings liable to attack for slight reason or
n o n e;50 a n d it seem s likely th a t the A ristophanes fragm ent
a tte s ts a b elief th a t aiso existed in a less m oral form.
T h e div ersity o f possible explanations is clear from a set of
fifth -c en tu ry texts th a t tre a t the causes o f m ental disturbance.
T h e purifiers o f On the sacred disease tried to diagnose the deity
re sp o n sib le for each p a tie n t’s affliction, b u t their m ethods were
p u re ly e x t e r n a l : I f th e p atien t im itate a goat, if he roar, or suffer
convulsions in th e rig h t side, they say th a t the M o th er o f the
G o d s is to blam e. I f he u tte r a piercing a n d loud cry, they liken
h im to a h orse a n d blam e P oseidon.’51 T h ere was ap p a ren tly no
q u e stio n o f seeking a cause for the an g er o f the god in question,
n o r o f ap p e a sin g him by sacrifice. T h e pow ers identified by the
p u rifiers as senders o f epilepsy - the M o th er o f the Gods,
P oseidon, E nodia, A pollo Nom ios, Ares, H ecate, a n d the
H ero es - do not, as a group, seem strongly involved w ith the
m o ral o rd e r. T h e m alicious attack s o f H ecate w ere a terro r to
th e su p erstitio u s, a n d if they h ad a cause, it was not the guilt of
th e v ictim b u t th e conjuring o f a sorcerer.52 E nodia w as o f the
sa m e c h a ra c te r.53 A pollo N om ios p robably represented the
47 A r. fr. 692a.
48 E ven P lato refers to the specific case o f the biaiothanatoi to establish the general
p o in t, Leg. 9 2 6 e -9 2 7 a , cf. 865e.
49 Op. 1 21-6.
50 A r. Av. 149 0 -3 w ith schol., M en. fr. 394, B abrius 63, A th. 461c, p. 272 n. 73 below;
cf. A. B relich, Glieroigreci, Rom e, 1958, 226 ff., an d H erter, Dämonen, 56.
51 H ippoc. Morb. Sacr 146. 21 ff. J ., 1.33 ff. G ., trans. Jo n es. O n the gods identified cf.
L a n a ta , 39 η. 94.
52 A bove, p. 222. F or la te r an tiq u ity see Fr. Pfister, Wochenschrift f. klassische Philologie,
29 (1912), 7 5 3 -8 .
33 E n o d ia exists in T h essaly as an in dependent chthonic goddess (W ilam ow itz,
Glaube, i, 17 0 -2 ; T . K ra u s, Hekate, H eidelberg, 1960, 7 7 -8 3 ) ; in A ttica she merges into
H e c a te (E u r. Hel. 569 f., Soph. fr. 535), or P ersephonc-H ecate (Eur. Ion. 1048, Soph.
Ant. 1199 f.). She is p atro n ess o f ‘attack s’ by night or day (Eur. Ion. 1048-50).
Divine Vengeance and Disease 245
sam e a m o ra l m en ace o f th e o p en a ir as Pan - the su d d en terrors
o f h erd s, th e m id d ay m adness o f m en. O f the M o th er m ore will
be said below . A fam ous ch orus in Sophocles rep resen ts Ares
as foe to m a n .54 T h e significance o f th e H eroes is, as we have
seen, am b ig u o u s. A lm ost all th e gods specified by the purifiers,
therefo re, h a d special associations w ith m en tal d iso rd er or
o th e r afflictions, w hich they sen t for reasons unco n n ected w ith
m o rality .
A sim ilar p ic tu re em erges from a passage in th e Hippolytus,
w h e re th e ch o ru s sp ecu late on P h a e d ra ’s m ysterious w asting
disease, w h ich they see as a form o f m adness. ‘A re you w an d er­
in g seized, prin cess, by P an or H e cate or the holy C ory b an tes or
th e m o u n ta in m o th er?’55 T h ese gods differ little in ch aracter
from th o se o f th e list in ‘H ip p o c ra te s’. Living on the fringes of
th e O ly m p ia n w orld, they lack its involvem ent w ith m orality.
T h e y seize th e ir victim s; they do n o t punish them . An attack
m a y be asc rib ed to ‘P a n ’s a n g e r’,56 b u t th at anger, if explained
a t all, h as m otives th a t are frivolous.57 A nyone can fall in to the
p o w e r o f gods like th e C o ry b an tes, an d the only cure is to
c e le b ra te th e ir rites. T h u s th e ch o ru s’s first suggestion implies
no offence a t all on P h a e d ra ’s p a rt. T h ey go on to w onder,
a lm o st in th e sam e b re ath , w h e th e r P h aed ra has om itted an
offering to D icty n n a. T h ey show no aw areness of having passed
from o n e level o f ex p lan atio n to an o th er, a n d in the next two
sta n z a s sp ec u la te on n a tu ra l causes, both psychological and
p h y sical, for th e sam e affliction. L a te r in the play, the nurse
co n sid ers w h e th e r P h a e d ra ’s frenzy m ight be d ue to the stain of
bloo d o r to w itch craft.58 T h is prag m atism , a n d receptivity to
differing a n d possibly co n trad icto ry theories o f disease and
m eth o d s o f trea tm e n t, is p erhaps characteristic of folk-medicine.
T h e re is a sim ilar passage in Sophocles’ Ajax, w here the
34 O T 190 -2 0 2 .
55 141 fT., B a rre tt’s tran slatio n . F or the controversy over entheos see B urkert, GR 178
n. 1.
56 A fainting-fit, E u r. Med. 1172, delirious terro r (E ur.) Rhes. 36.
37 See G ow o n T h eo critu s 1.15. R. H erbig, Pan, F rankfurt, 1949, 18 f., stresses that
th e co n cep tio n o f ‘p a n ic ’ derives from the b ehaviour o f anim als; it rem ains am oral
w h en tran sferred to m en. T h is aspect is fully treated by W . H. Roscher, ‘E p hialtes’
(Abh. Sächs. Ges. Hiss. 20.2, Leipzig, 1900), 6 6 -8 4 ; P an’s am oralitv is clear from the
eq u a tio n w ith E p h ialtes. See too B orgeaud, 137-75.
38 3 1 6 -1 9 .
246 M iasma
ch o ru s co n sid er the h ero ’s m adness. It m ight, they feel, be a
p u n is h m e n t se n t by A rtem is o r Ares for om itted thank-offer-
ings. O f P an a n d th e C o ry b an tes they say nothing. L ate r in the
p lay it tu rn s o u t th a t the oifended goddess is A thena, an d the
offence n o t a m ere failure in cult, b u t a w anton over-valuation of
h u m a n s tre n g th ag a in st divine, a classic insult to the gods.59
T h is highly m oral story is very far from On the sacred disease, an d
th e tw o texts m ay be taken as representing extrem e possibilities.
T h e re is no reaso n to do u b t th a t a n offence against the gods
co u ld be seen as causing disease, b u t Sophocles m ay have been
influenced by th e claim s o f tragic dignity in ignoring all o ther
p ossible diagnoses. A ristophanes m entions three religious
tre a tm e n ts for m ad n ess - purification, the C orybantie rites, an d
in c u b a tio n .60 N one o f them obviously entails the identification
a n d a p p e a se m e n t o f a p u nishing god. I f seers existed who
c o n d u c te d cures along those lines, they have left no trace in our
sources. T h e everyday expressions for ‘you’re off your h ea d ’
th a t are found in com edy tre a t m adness as a p ro d u ct either of
in ex p licab le d aim o n ic intervention, or o f an excess of black
b ile .61
T h e o p p o sitio n w hich has begun to em erge betw een m adness
as p u n ish m e n t for ritu al o r m oral offences, an d m adness as
seizu re by capricious, am oral spirits, finds parallels am ong
m a n y peoples. T h ro u g h o u t the w orld there are cultures th a t
a ttr ib u te disease, p articu larly m en tal disease, to possession by
sp irits. A re c u rre n t p a tte rn can be traced w hereby, alongside
th e c e n tral deities (often ancestors) w ho send affliction as
p u n ish m e n t, ce rta in perip h eral spirits, perhaps o f foreign
o rig in , are also active, a n d choose th eir victim s regardless of
m o rality . S uch sp irits attac k people w hose position in society is
as p e rip h e ra l as th e ir own, m en w ith o u t statu s an d , above all,
w om en. T h e ‘c u re ’, in so far as it is possible at all, c h a rac te r­
istically involves lifelong devotion to the cult o f the possessing
59 1 7 2 -8 6 , 7 5 6 -7 7 .
60 Vesp. 118—24.
61 T h e c o n tra st in th is respect w ith higher genres is noted by A. O 'B rien M oore,
Madness in Ancient Literature, diss. P rinceton, 1922, published W eim ar, 1924, 10 f.
D a im o n ic in terv en tio n : p. 248 n. 67 below. Black bile: often in A ristophanes (e.g. Nub.
833, A v. 14, Pax 66), an d M en an d er. T h is is probably in origin a popular, not a
scientific in te rp re ta tio n (see K udlien, 7 7 -8 8 ) a n d need not be later th an the religious
one.
Divine Vengeance and Disease 247
sp irit, a cu lt th a t will p ro b a b ly take an ecstatic form, an d be
c e le b ra te d by th e co m m u n ity o f p a s t victims. T h o u g h posses­
sion is a t first seen as a n affliction, ag ain st w hich help is needed,
th ese cults u n d o u b te d ly have a clan d estin e significance from
w h ich th e victim s of th e sp irit gain psychological benefit. T h e
h ea lin g cu lt becom es a p erso n al religion for persons sh u t out
from th e c e n tra l m o rality cults. It offers w om en a religious
ex p erien ce, a sp h ere o f in terest, a n d an identity, each of them
o p p o sed to th e typical fem ale role. B ut if possession is to be
in te rp re te d as a n in escapable affliction, the possessing spirit
sh o u ld be g u id ed by caprice. Such a cu lt cannot be based on the
a d m itte d g u ilt o f its m em bers. T h e taran tism o f so u th ern Italy,
sp irit-p o ssessio n m ed iated by an event as am oral as th e bite o f a
sp id er, is a c h a rac te ristic ex am p le.62
W e know o f one ‘foreign’ sp irit w ho afflicted G reek w om en in
this w ay: D ionysus. T h e social significance o f m aenadism , a
form o f b e h a v io u r not originally sh ared by bo th sexes b u t
p erfo rm ed by o n e in defiance o f th e other, has ten d ed to be
u n d e re s tim a te d .63 By th e fifth century, however, spontaneous
possession w as no longer a ttrib u te d to D ionysus b u t to powers
su c h as th e C o ry b a n te s.64 T h e re is no evidence th a t th eir rites
w ere especially celeb rated eith er by wom en or by a p articu lar
social class, b u t in o th er im p o rta n t respects they conform to the
p a tte r n o f p erip h e ral healin g cults. T h e C o ry b an tes them ­
selves, sen d ers o f m adness, also cure it; th eir healing m ethods
a re h o m o eo p ath ic, by ecstasy; they are explicitly foreign, and
th ey h av e no in tere st in m orality. T h e ir cult is not attested
before th e fifth ce n tu ry ,65 b u t this kind o f in terp re tatio n of
m e n ta l d is tu rb a n c e is unlikely to be a novelty. D ionysus was
th e ir p red ecesso r, a n d th e indigenous P an, also a sender of
m ad n ess, w as no g u a rd ia n o f m orality. S im ilar conceptions
a p p e a r elsew here in p o p u la r th o u g h t. In the language of'H om er
th e re a re h in ts o f th e view o f m adness as due to daim onic
62 O n all this see I. M . Lewis, Ecstatic Religion, H arm ondsw orth, 1971, esp. C h. 3 and
79—85. O n ta ra n tism , ib id ., 88—92, an d E. He M artin o , La Terra del rimorso, M ilan. 1961.
43 B u t see S im on, 2 4 2 -5 7 .
64 D o d d s, 7 7 -8 0 w ith b ibliography. PI. Euthyd. 277d im plies particip atio n of the
w ell-b o rn , as D o d d s notes.
65 E. R. D o d d s, Harvard Theological Review 33 (1940), 171-4; the ‘m o th er’ (G reek or
P hry g ian ?) is a lre a d y a h ealer in Pind. Pyth. 3. 7 6 -9 ; cf. Burkert, GR 277.
‘2 48 M iasma
in te rv e n tio n ,66 a n d th a t view is preserved in the use o f the verb
daimoniö, th o u g h we can n o t be sure how literally the idea
ex p ressed in it w as u n d ersto o d .67 A striking n u m b er o f diseases
h a v e a n im a l nam es o r nicknam es: fox, lion, crab, a n d m any
m o re .68 N ig h tm are s can be traced to E phialtes, the ‘leaper o n ’,
a n d su d d e n attac k s o f fever in the night to sim ilar strangling
d e m o n s.69 S uch language reveals a t the least an im m ediate
p e rc e p tio n o f disease th a t is am oral, although the beast or
d e m o n w ith w hich one struggles could o f course also be seen as
th e a g e n t o f a n avenging god.
I t is obviously relev an t to consider the m oral stance o f
G re e c e ’s countless healing gods a n d heroes.70 U nfortunately,
th o u g h we can trac e the existence o f healing cults from early
tim es, w e know little o f the ideas a n d expectations th a t patients
b ro u g h t to them . F or one cult, how ever, o f the fourth century
w e h av e d etailed evidence. T h e inscription recording the
m ira c u lo u s cures o f A sclepius a t E p id au ru s is obviously a docu­
m e n t th a t req u ires discreet h andling, b u t the ad m ix tu re o f
five-year preg n an cies a n d the like does n o t entirely disqualify it
as evidence. Pia fraus, to be effective, m u st be rooted in the
fam iliar; it glorifies the god by rep resen tin g the hopes and
d re a m s o f every w o rsh ip p er as achieved fact.71 In c u b atio n is a
p a rtic u la rly revealing technique, because w h at the p atien t
d eriv es from th e experience will correspond to w h at he him self
66 D o d d s, 67.
67 e.g. A r. Plut. 372, 501, M en. Dysc. 88, fr. 127; cf. A. O ’B rien M oore, op. cit., 14—18.
M a d n e ss as a being ‘stru c k ’ persists too, e.g. A r. Vesp. 947, M en. Dysc. 311, Perikeiromeni
496 w ith S an d b a c h , B orgeaud, 183 f. Physiological a n d daim onic are nicely conflated
in M en. Epit. 880 f., w here black bile ‘falls o n ’ you.
68 See YV. H . R oscher, Rh. Mus. 53 (1878), 173, 180 n. 5; also A. Riess, Rh. Mus. 49
( 1894), 181. F o r sem i-personified disease see A r. Nub. 243 w ith D over’s note, Soph.
Phil. 758 f. w ith J e b b . T h e claim th at sim ilar ideas still shim m er through in the
v o c a b u la ry o f the H ip p o c ra tic corpus is not proven: see G. Preiser, Allgemeine
Krankheitsbezeichnungen im Corpus Hippocraticum (A rs M edica I I .5), Berlin, 1976, 6 0 -3 ;
G o ltz, 2 7 2 -4 .
69 See S o p hron, frr. 68, 70; A r. Vesp. 1037 fF. w ith schol.; W . H . Roscher, ‘E phialtes’
(Abh. Sachs. Ges. IViss., 20.2, Leipzig, 1900), 4 8 -5 6 ; for the m edical litera tu re denying
d e m o n ic n a tu re o f such seizures ibid., 108—15.
70 A scholarly acc o u n t o fG ree k healing gods in general seem s not to exist. F or A ttica,
F. K u tsc h , Attische Heilgötter und Heilheroen, R G W 12.3, Giessen, 1913. Som e general
in d ic a tio n s in N ilsson, G G R (index s.v. Heilgötter), an d the ch ap ter ‘D isease and
C a la m ity ’ in W . H . D. R ouse, Greek Votive Offerings, C am bridge, 1902.
71 O n the sta tu s o f th e tem ple inscription as a w itness see D odds, 112 f.; idem ,
Progress, 169—71.
Divine Vengeance and Disease 249
b rin g s to it. T h e g o d ’s diagnosis a n d cure of his disease depend
on his ow n co n cep tio n o f how a divine healer o u g h t to operate.
G iv en this, it is strik in g to observe the indifference of the god to
th e o rigin o f th e disease th a t is to b e cured. H e ap p ears in a
d re a m a n d e ith er perform s an a c t o f m iraculous healing a t once,
o r in d icates th e course th e p atien t m u st follow on w aking to be
cu red . W h a t he does n o t do is to suggest a p ast offence against
a n o th e r god th a t m ig h t have caused the disease. A p a rt from a
few late a n e c d o te s,72 th e g o d ’s concern for h u m an ju stic e did
n o t ex ten d b eyond his ow n p erq u isites.73· T o questions of
m o rality h e h a d in g eneral th e professional indifference of the
tru e d o c to r.74
T h e p arallel w ith the d o cto r is im p o rtan t. M any people seem
to h av e m a d e th e jo u rn e y to E p id au ru s not as an altern ativ e to
m ed ical tre a tm e n t, b u t once such trea tm e n t h ad failed.7S C o­
o p e ra tio n b etw een doctors a n d priests of A sclepius is not
d e m o n s tra b le ,76 b u t n o r is hostility; and, though th ere are
im p o rta n t differences betw een scientific and tem ple m edicine,
th e re a re also im p o rta n t sim ilarities.77 T h e divine physician
w as ex p ected to m ake p rescriptions th a t w ere the paradoxical
rev erse o f n o rm al h u m an th erap y , a n d did not disdain the use of
m ag ical d ru g s. B u t he w as n o t req u ired to suggest a different,
essen tially religious aitiology o f diseases. H is was secular
m ed icin e, as u n d ersto o d by th e laym an, with an injection of
s u p e r n a tu ra l pow er. I t is therefore only p artially correct to see
th e triu m p h a n t rise o f the A sclepius cult as a sym ptom of
g ro w in g irra tio n a lism .78 T h e genuine achievem ents and
p ro g ra m m a tic aspirations o f H ippocratic medicine had aroused
72 T e stim o n ia 394, 395, 397, 517 in E delstein’s collection.
73 M iracu lo u s p u n ish m en t o f those who m ock A sclepius himself: S /C 3 1168, cures vii
a n d xxxvi; IG I V 2 123, cu re xlvii (cf. SE G xi 423 w ith bibliography). For confession in
this co n tex t (excep tio n al for a G reek) see too /G IV 2 123.67,91. O n the ‘S tralw u n d er’ cf.
O . YVeinreich, Antike Heiligungswunder, R G W S . l , Giessen, 1909, 5 5 -6 2 .
7< E d elstein , ii, 180.
75 E d elstein , A M 245.
76 See L. G o h n -H aft, The Public Physicians o f Ancient Greece, M assach u setts, 1956,
26—31, a n d S. M . Sherwin-YVhite, Ancient Cos, Hypomnemata 51, G öttingen, 1978,275—8,
criticizin g H erzog; also L loyd, 48 n. 209. Edelstein, A M 244 f. also believ ed th at doctors
m ig h t im p licitly refer cases to tem ples.
77 Cf. D o d d s, 115 f., Lloyd, 40 I. P aradox, e.g. testim onia 317.8, 408 in E delstein’s
collection. D ru g s, D o d d s, 115.
78 D o d d s, 193.
 M iasma
250
larg e ex p ectatio n s as to the possibility o f curing all forms of
disease, ex p ectatio n s w hich, n atu rally , it was in no position to
fulfil. T o satisfy them , p o p u lar im agination created, in the
h u m a n d o c to r’s im age, a divine do cto r whose m agic powers
allow ed him to m ake real the exaggerated claim s of rational
m edicine. T h o se w ho u n d erw en t incubation dream ed not of
a n g ry gods b u t o f skilful surgery a n d subtly balanced regimen.
D e lp h i m u st often have been consulted ab o u t disease.79 O ne
in sta n c e re co rd ed in H erodotus is th a t o f the L ydian king
A ly attes, w ho fell ill after accidentally b u rn in g dow n the tem ple
o f A th e n a A ssêsiê a t M iletus. T h e disease refused to clear up,
a n d so A ly attes d ecided to consult the oracle. T he P ythia would
n o t give him a n an sw er u ntil he reb u ilt A th e n a ’s tem ple. H e did
so, a n d recovered a t once.80 T h is is the classic p a tte rn o f a
disease caused by a religious offence, diagnosed by a religious
specialist, a n d cu red by expiation o r restitution. T h e very neat­
ness o f th e p a tte rn , indeed, exposes the story to suspicion.81
M o re reliab le evidence is available for D odona, because some of
th e lead tab lets on w hich requests to the oracle w ere w ritten
h av e su rv iv ed .82 U n fo rtu n ately th eir d atin g is insecure, an d
n o n e o f th em seem s to be very early. A typical exam ple runs:
‘N ik o k ra te ia asks w hich god she should sacrifice to in order to
fare b e tte r a n d be free o f h er disease.’83 T h ere are several such
re q u e sts, som e w ith the exhaustive form ula ‘w hich god o r hero
o r d aim o n should x sacrifice to’.84 It w ould be interesting to
know w h a t w as th e usual answ er. P erhaps the priests identified
a god to w hom th e in q u ire r h ad not sacrificed recently, an d the
o ra cle th e n suggested an offering to h im .85 P erhaps it sim ply
n a m e d a h ea lin g pow er.
I t is w o rth m en tio n in g som e factors th a t do not ap p e ar or
o n ly seldom a p p e a r in G reece as causes o f disease. V ery trivial
ritu a l in fractio n s are not attested. Possibly such subjects were
79 C f. L S C G 83. 12 f., health and preservation as tw in concerns o f A pollo’s oracle a t
C o ro p e.
80 1 .1 9 -2 2 .
81 F o n ten ro se, 301. Solid D elphic evidence is q u ite lacking.
82 T h e ‘o ld er ones are in S G D I 1557-1598 (H offm ann); recent discoveries SEG xiii
397, XV 385—409, xix 426—432, xxiii 4 7 4 -6 , xxiv 454.
83 S G D I 1561 B = S IG 3 1161.
84 S G D I 1564, 1566, 1582, 1587.
85 Cf. th e pro ced u re o f the m an tis Eukleides in X en. An. 7 .8.1-6.
Divine Vengeance and Disease 251
too und ig n ified to p en e trate o u r sources; p erh ap s the Greeks
w ere u n w illing seriously to believe th a t a god m ight send
d isease as p u n ish m e n t for en terin g a shrine in a d irty robe. It
h a s a lre a d y been noted th a t, alth o u g h pollutions can cause
p h y sical d iso rd e r,86 this is n o t an idea th at receives strong
em p h asis. N o r is th ere m u ch evidence in this connection for the
evil eye o r sorcery, th o u g h the silence o f the sources here m ay be
d ecep tiv e. P resu m ab ly th e b ew itch m en t th a t the purifier p ro­
fessed to cu re 87 could m anifest itself in disease. I f we think
fu rth e r o f th e defixiones, a few o f w hich specify the disease they
a re su p p o sed to cau se,88 w ith P la to ’s account o f the fear such
m eth o d s in s p ire d ,89 it seem s alm ost inevitable th at h u m an
m alice m u st often have been diagnosed, or a t least suspected, as
th e cau se o f a p a rtic u la r m isfortune. M ore th an this, un fo rtu n ­
ately , it seem s im possible to say.
N on-scientific attitu d e s to disease in G reece are, it seems, too
d iv erse to b e covered by a sim ple form ula. A recent stu d y by a
m ed ical h isto rian has tried to illu strate the transform ation of
early G reek society from sh am e cu ltu re to guilt cu ltu re by its
c h a n g in g u n d e rsta n d in g o f d isease.90 A sham e culture, it is said,
sees d isease eith er as ra n d o m evil, inexplicable fatality, o r as a
g o d ’s revenge for a n affront to his ow n honour, b u t in neither
case as a p u n ish m e n t for m oral evil. T h e m oral view is taken as
th e d efin in g ch a rac te ristic o f a g u ilt culture, an d it is also said to
86 p. 218.
87 p. 222.
88 W ü n sch , n n . 7 7 - 8 (im potence); sherd published by N ilsson, GGR HOI. Άριστίωνι
έπιτίθημι τεταρταίον ές ",Ά ιόα; later instances S t G iv. 47, E. Z iebarth, ‘Neue
V erfiu ch u n g stafeln au s A ttika. Boiotien u n d E u b o ia’, Sitz. Preuss. Ak. Berl. 33 (1934),
n. 24.
” Leg. 933a—b. C h a rm s cause childlessness an d aversion, Eur. Aniir. 155—(i0; evil eve
kills, A p. R h o d . 4.1669 IT'. O n early G reek m agic see T . H oplner, R E s.v. Mageia. 303
(H o m e r); A b t. 9 5 -1 0 0 (tragedy an d com edy); H oplner, loc. cit., 384 (trials of
sorceresses).
90 F. K u d lien , ‘E arly G reek P rim itive M edicine’, Clio Medica 3 ( 1968),.305-336, esp.
317. ( It is becau se o f this sch o lar’s deserved au th o rity th at 1 take issue w ith him
explicitly .) O n the w hole question, it is interesting to note th at, though psycho-
d v n am ically sh am e an d guilt can perh ap s be distinguished, the differentiation o f whole
cu ltu re s acco rd in g to these criteria has proved very problem atic. Sham e is or can he
in te rn a lly felt no less th a n guilt (G. Piers a n d M. B. Singer, Shame and Guilt, a
psychoanalytical and cultural study, Springfield, Illinois, 1953, esp. 48 If.). A H om eric hero
is co n strain ed by aidôs to observe certain social rules no less th an to assert his own
ag o n istic p re-em in en ce. A gainst a sh a rp distinction see Lloyd-Jones. 24—6, Dover, 220
n. 3: b u t no te th e co m m en t of J. G ould, CR n.s. 28 (1978), ‘287.
252 M iasma
be th e d o m in atin g a ttitu d e of the fifth-century Greek. But, as we
h a v e seen, the view o f disease as ran d o m affliction still persists
in th e fifth cen tu ry , a n d in the fourth century, in the practice o f
A sclepieia, seem s to triu m p h . O n e obstacle to a full m oraliza-
tio n o f disease w as th a t it m ight involve an adm ission o f w rong­
d o in g m ost unw elcom e to the p a tie n t him self.91 A G reek would
sca rce ly h av e cared to proclaim publicly th a t he w as suffering
th e consequences o f his perju ry or m altrea tm e n t of a guest. His
n e ig h b o u r’s affliction, indeed, w as no d o u b t a richly deserved
d iv in e p u n ish m e n t, b u t his ow n w as a random event, the
p ro d u c t o f m alice o r sorcery. T h e one confession th a t it was
re a d ily a c c e p ta b le to m ake w as th a t o f a ritual om ission, a
fo rg o tten sacrifice p erh ap s;92 the cloud of forgetfulness could
en v elo p th e m ost h o n o u ra b le o f m en in unp red ictab le w ays.93
A ja x ’s S a la m in ia n sailors loyally ascribed his m adness to an
offence o f this k in d .94 In the d ark hours w hen A lexander was
c o n su m e d w ith, g u ilt for the m u rd e r o f C leitus, tactful seers
p o in te d o u t th a t th e king h ad sacrificed to the D ioscuri on a day
tra d itio n a lly reserved for D ionysus: the d ru n k en m u rd er was
b u t th e w o u n d ed g o d ’s savage revenge.95 Even the idea o f
in h e rite d p u n ish m e n t can acq u ire a new significance in this
p ersp ectiv e, as a w ay o f evading personal guilt.96
I t w ould, o f course, be rash to deny th a t a G reek could, w ithin
him self, co n n e ct his crim es a n d his sufferings, w hatever he m ay
h a v e m a in ta in e d before the w orld. T h e w ord g roup su rro u n d ­
in g enthumios, ‘o n o n e ’s m in d ’, is o f im portance here.97 T hese
w o rd s can be ap p lied to an y object o f anxious th o u g h t,98 b u t in
m o st surviving instances they have a specialized reference to
91 Cf. esp. C a m p b e ll, 325. Also R. H. an d E. Blum , Health and Healing in Rural Greece,
S ta n fo rd , 1965, 127: ‘In a cu ltu re w here m ain tain in g philotimo requires th at a m an
re m a in blam eless, th e p e asan t does not a ttrib u te his sufferings to his ow n sinfulness.’
92 Cf. R. H . a n d E. B lum , loc. cit.: ‘although they do attrib u te illness to their ritual
failures, such failures do n ot im ply a personal m oral transgression.'
93 P ind. 01. 7.45,
94 See above, p. 246; so too the chorus in Hippolytus, 141 ff. icf. p. 245).
95 A rr. Anab. 4.9.5.
96 N o te the co n tex t o f Soph. OC 964 f.; cf. Boyce, 107, a n d j . K. C am pbell, cited
p. 241 n. 32.
97 See D odds, 55 n. 46, referring to W ilam ow itz on E ur. H F 722 an d W . H. P. H atch,
H S C P 19 (1908), 172—5 (H a tc h collects the instances o f enthumios b ut ignores r e l a t e
uses o i enthumeomai an d enthumêma).
98 e.g. I^om . Od. 13.421, Soph. 07"739, Track. 109, Eur./o«. 1347.
Divine Vengeance and Disease 253
religious scru p les or anxiety. I f a m u rd erer is unjustly
a c q u itte d , a sp eak er in th e Tetralogies points out, the m urdered
m an becom es enthumios for th e ju ro rs, whose d u ty it was to
aven g e h im .99 X erxes for an unknow n reason o rdered the
A th e n ia n exiles to m ake sacrifice on the acropolis; perhaps,
suggests H e ro d o tu s, th e fact th a t he h ad b u rn t dow n the shrine
was an enthumion for h im .100 An act o f potential religious dan g er
m ig h t becom e a n enthumion a t o n c e ,101 o r in tim es o f trouble the
victim s m ig h t p o n d er th eir afflictions a n d connect them
m en tally w ith p ast offences.102 An enthumion is not a pan g of
conscience, alth o u g h the English concept is often helpful in
tra n s la tin g it; w hereas conscience is guilt over bad actions,
reg ard less o f co n seq u en ces,103 a n enthumion is the anxious antici­
p a tio n o f evil as a result not m erely o f an act b u t even of an
o cc u rre n ce, su ch as a b ad o m e n .104 B ut though not confined to
th e m o ral sp h ere, th e w ord could certainly be used in connec­
tion w ith th e ex p ectatio n o f evil in consequence o f evil deeds.
I n its specialized sense, enthumios is not attested before the
fifth ce n tu ry , b u t the experience it denotes is certainly m uch
old er. W h e n ev er a c h a rac te r in H o m er rejects a p a rtic u la r form
o f b e h a v io u r th ro u g h religious scruples, he im plies that, were he
to p erfo rm it, it w ould th en be ‘on his m in d ’.10S W hen the
99 A n t. Tetr. 1 γ 10, 2 a 2, ô 9; cf. D em o critu s’ use ο ί'έγκάρόιον, B 262.
100 8.54.
101 S cru p les co n cern in g a future act, E ur. H F 722; cf. Soph. OC 292.
102 T h u c . 5.16.1. In 5.32.1 an d 7.18.2 enlhumeisthai is constructed w ith the present
m isfo rtu n es a s object.
103 Cf. D over, 220—3, w ho is cautious ab o u t recognizing allusions to conscience in
the m o ral sense. F o r th e litera tu re on conscience (disappointing) see the bibliography
to M . C lass, Gewissensregungen in der griechischen Tragödie, Spudasmala 3, H ildesheim , 1964.
104 H d t. 2. 175, T h u c . 7.50.4.
105 e.g. H orn. II. 6.266 f., Od. 14.406. A ttem p ts to generate religious scruples, II.
22.358, Od. 11.73. T h ese obvious rem arks are d irected against L atte 's characterization
o f H o m eric m a n , A R W 20 (1920/1), 258 = Kl. Sehr. 6: ‘E rst d as U nheil weckt in ihm
d a s E m p fin d en , sich vergangen zu h a b e n .’ T h e re never was such a m an. T h e enthumion
is also seen as a later developm ent by G ern et, Antiphon, 135 n. 1 and Dodds. 55 n. 46. 1
c a n n o t acce p t e ith e r D o d d s’s fu rth er claim that: ‘T h e specific usage is confined to this
perio d ; it v an ish ed , as W ilam ow itz says, w ith the decline o f the old beliefs, whose
psychological co rre la te it w as.’ Since VVilamowitz's Heracles w as published, new
evid en ce h a s been found (or th e w o rd ’s survival, and it may be rash to brush it aside as
‘a rc h a iz in g ’. See L S S 64 (T h aso s, late 5th or early 4th c.): if anyone ignores these
fu n eral reg u latio n s, ένθυμιστόν α ν τώεστω; LSCG 154 A 14 (Cos, first h alf of 3rd c.): if
a n y o n e ignores th e sacred law , ένθύμιον αντοϊςώ ς άσε\βήσασιν; LSS 72 A 5 (Thasos, 1st
c.): ένθυμιστόν for an y o n e w ho fails to pay tem ple dues. (L ater exam ples SIG 3 1184.7,
254 M iasma
c o m p an io n s o f O dysseus do in fact transgress a basic taboo by
sla u g h te rin g th e ca ttle o f the sun, they are plunged into guilty
a n x ie ty .106 W h a t m ay be new in the late fifth century is an
explicit aw aren ess o f the m echanism o f religious scruple, an d
a w illingness to speak openly o f this private condition.107
A n tip h o n in a re m a rk ab le passage describes the m ental effects
o f g u ilt a n d innocence. T h e re is no g reater com fort for a defen­
d a n t, he says, th a n the know ledge th a t he has com m itted no
crim e o r im piety. T h a t know ledge can sustain him in extrem e
bodily w eakness. B ut the guilty m a n ’s case is opposite. ‘His
sp irit fails him w hile his body is still strong, because it thinks
th a t this (th e illness? the trial?) has com e upon him as a p u n ish ­
m e n t for his crim es’.108 E uripides goes a step further in reducing
th e E rinyes th a t a tta c k O restes to his consciousness o f the dire
a c t h e h as p erfo rm ed .109
T h e co n c ep tu al fram ew ork for a religion o f confession there­
fore existed. In practice, how ever, it m ade little headw ay
a g a in s t the d o m in a n t ethic o f ‘tu rn in g the fair side o u tw ard s’. 110
As a co n trast, it is interesting to consider certain Lydo-Phrygian
in scrip tio n s o f th e second a n d th ird centuries A D , 111 w ritten in a
Inscr. Cos 319.) Cf. LSC G 130 (A stypalaea, 3rd c.): if anyone disobeys, αντώ έν νφ
έσσείται. T h e re is also the fact th at syneidêsis seem s to have been a concept o f popular
m orality (W ilam ow itz, Glaube, ii, 386). X en o p h o n ’s th reat th at ‘the goddess will take
ra re o f ’ violators o f his sacred law am o u n ts to the sam e thing (X en. An. 5.3.13).
A n o th er in sc rip tio n al th re a t, th a t the violator will have him self to blam e, is broader,
since the u n d e sira b le consequences are not necessarily su p ern atu ral, though they may
be. as in SIC? 1236 (on this form ula cf. A. W ilhelm , Sitz. Wien. 224.1 (1946),
21 = AkademieschriJ'ten zur griechischen Inschriftenkunde iii, Leipzig, 1974, 159; L. Robert,
Etudes Anatoliennes, Paris, 1937, 415 f.; specifically religious applications of this or
c o m p a ra b le expressions are already found in the 5th c., Ar. Ran. 630, E ur. Med. 1055).
N o te too th e ch a ra c te ristic language o f ‘good hopes’, above, p. 175 n. 173. Plut, de Pyth.
or. 404a nicely illu strates the enthumion.
,0" Od. 1 2.340-51.
107 In ru ra l G reece today, filolimo is ‘largely concerned with the protective conceal­
m en t o f every th in g internalised in a person o r society’, M . Herzfeld, Man 15 (1980),
346.
108 5.93, cf. 6.1.4. an d the texts in S tobaeus 3.24 περί τob οννεώότος. F or an instance
see P lut. Dion 56.2.
109 Kur. Or. 396.
' 10 P ind. Pyth. 3.83. Confession is un-G reek, L atte, Kl. Sehr. 32 n. 42 (b u t for a special
case see p. 249 n. 73).
111 C ollected a n d stu d ied in the valuable dissertation o f P. Steinleitner, Die Beicht im
Zusammenhange mit der sakralen Rechtspflege in der Antike, Leipzig, 1913. A dd SE G iv
6 4 7 - 5 2 , xxviii 910, 913 f.; Monumenta Asiae Minoris Antiqua iv, ed. W . H . Buckler et nt..
Divine Vengeance and Disease 255
form o f G reek b u t revealing a very un-G reek religious clim ate,
in w h ich disease was a d irect p u n ish m en t sent by a specific god
for sin, a n d th e p rin cip al h ealing tech n iq u e was to identify and
confess th a t sin. O n ce cured, th e p a tie n t was req u ired to set up
a tablet recording his transgression an d its punishm ent. O ccasion­
ally a s in n e r m ig h t a n ticip ate th e delays o f divine revenge by a
sp o n ta n e o u s confession.112 F rom these tablets we h ear of m en
p ay in g for th e ir offences not only in th eir own persons, b u t also
in th e p erso n o f a son, d au g h ter, relative, a n d even cow .113
T h e se sins a re alm ost all ritu al offences o f som e kind: the
ac c id e n ta l c u ttin g o f sacred wood, failure to fulfil a vow, en ter­
ing a p re c in c t in d irty clothes o r a state of ritu al im p u rity .114
T h e y are, how ever, also m oral offences in the sense th a t they
im p ly c o n te m p t for the sacred. T h e p en iten t com m only ends his
in scrip tio n w ith solem n advice to all to ‘take the stele as a
w a rn in g a n d n o t despise th e g o d .’
H o w d ifferen t is th e m essage incu lcated by these inscriptions
from th a t o f th e tem ple record o f Epidaurus! H ere th a t p er­
vasive u n ease som etim es su p p o sed characteristic of archaic
G reece does in d eed seem to be p re sen t.115 T h e G reek by
c o n tra s t co uld experience m isfortune in the form of disease
w ith o u t n ecessarily search in g his conscience w ith anxiety for a
po ssib le cause. D eity in different G reek au th o rs, som etim es in
th e sam e a u th o r, seems to o p erate at different levels: it guards
th e m o ral o rd e r, rew ard in g th e good a n d p unishing the bad; it
u p h o ld s th e form al rights o f gods against m en; as fate o r the
in s c ru ta b le d ivine will it m akes occurrences inevitable; and it
re p re se n ts th e ra n d o m m alicious elem ent in the universe th at
cau ses th e good to suffer an d the b ad to prosper. T hese levels
M a n c h e ste r 1933, n n . 2 7 9 -9 0 . Cf. F. C u m o n t, Les Religions orientales dans le paganisme
romain4, P aris, 1929, 218 n. 40; R. R eitzenstein, Die hellenistischen Mysterienreligionen3,
L eipzig, 1927, 137 fl'.; R. Pettazzoni, Essays on the History o f Religions {Numen supplem ent
1), L eid en , 1 9 5 4 ,5 5 -6 7 .
1,2 e.g. M A M A iv 285.
113 S tein leitn er, nn. 3,7,17,33 (M A M A iv 286).
1.4 Irrelev an ce o f in tention: S teinleitner nn. 11,14,16. R itual im p u rity is especially
c o m m o n , ib id ., n n . 24,26, M A M A iv 283,285,288,289, b u t m oral offences are also
p ossible, e.g. S tein leitn er, n. 29 (perjury). Steinleitner, n. 16, referring to an offence
co m m itted π α ώ ίο ν ώ ν , suggests th a t the search for causes could go back very far.
1.5 O f course, th e G reek w ho ‘tu rn ed th e fair side o u tw ard ’ m ay often have accused
h im self in w ard ly m ore b itterly th an these o stentatious confessors o f footling infrac­
tio n s, b u t th e po in t a b o u t the publicly accepted response to disease in Greece rem ains.
256 M iasma
co rresp o n d to as m an y n a tu ra l h u m an attitudes: the dem and
th a t th e n a tu ra l o rd e r should conform to the m oral order; the
n eed for a u to m a tic sanctions g u ard in g the restrictions th at
d iv id e god from m ortal; fatalistic acceptance o f events; a n d the
d isc o u ra g e d p ercep tio n o f cosm ic injustice. T hese four attitudes
d e te rm in e th e G reeks’ religious explanations o f disease. T hey
seem to coexist a t all periods, a n d it w ould be h a rd to d etect a
sig n ifican t shift in em phasis betw een them . T h e choice o f in ter­
p re ta tio n in a given case will be pragm atic. T h e obviously guilty
m a n ’s disease is seen as p u n ish m en t, certainly by his enemies,
p e rh a p s by h im self too; b u t the institu tio n alizatio n o f guilt in
confession, th e d o g m atic definition o f illness as a consequence
o f sin, ritu a l o r m oral, is lacking. T o som e extent the idea of
d isease as a ra n d o m event w as in h e re n t in G reek thought, and
th is p o p u la r a ttitu d e h ad a negative relevance to th e success of
H ip p o c ra tic m edicine, in th a t m aterialist m odes o f e x p lan a tio n .
w ere n o t opp o sed by theological prejudice.
9
PURIFYING THE CITY
It w as n o ted earlier th a t th ere w as, in G reek belief, no such
th in g as n o n -co n tag io u s religious d a n g e r.1 Som e d angers were
m o re co m m o n ly seen as co m m u n icab le by contact, w hile others
ra th e r th re a te n e d the guilty p a rty ’s descendants; b u t th e differ­
ence w as one o f degree ra th e r th a n o f kind. Every m em ber of
a n y co m m u n ity , therefore, in principle lived u n d e r th rea t of
suffering for his n eig h b o u rs’ offences. T h e ways in w hich divine
a n g e r a g a in st a co m m u n ity could be expressed w ere diverse. A t
th e b e g in n in g o f Sophocles’ Oedipus Tyrannus, T h eb es is afflicted
in th re e w ays - th e crops have failed, w om en an d anim als
c a n n o t b rin g fo rth th eir young, an d plague is raging. T his is a
ty p ical situ a tio n th a t constan tly recurs b oth in m yth a n d in the
G re ek s’ ow n p ercep tio n o f historical reality. T h e n am e for this
w hole com plex o f d isasters is loimos, w hich is th u s m uch b roader
th a n ‘p la g u e ’ by w hich it is com m only ren d ere d .2 It could be
called d o w n ag a in st the v io lator o f an oath, o r its opposite
b eso u g h t in p ra y e r.3 B u t th o u g h divine an g er was typically
ex p ressed th ro u g h d istu rb an ce s o f the n atu ra l o rd er such as
these (sto rm s too are often m en tio n ed ),4 it m ight also be the
u ltim a te cau se o f events readily explicable on th e h u m an level.
C ivil strife a n d m ilitary failure are com m only associated w ith
loimos; th e S p a rta n setbacks in th e A rchidam ian w ar had a
religious o rig in , a n d th e an g er o f Z eus X enios ag ain st Paris was
fulfilled th ro u g h th e G reek expedition against T ro y .5 M oving
passag es in H esiod a n d A eschylus contrast afflictions o f all
' p. 10.
2 See M . D elco u rt, Stérilités mystérieuses et naissances maléfiques, Liege, 1938, C h. 1.
3 O a th : above, p. 191 η. 3. P rayer: A esch. Supp. 6 5 9 -9 7 , Eum. 9 0 7 -9 , 9 3 7 -4 8 ,
9 5 6 -6 7 . F o r plag u e as ‘d aim o n ic’ or ‘god -sen t’ see e.g. T h u c. 2.64.2; Polyb. 36. 17;
so u rces cited in E ro tian , p. 108. 1 6 -1 9 N achm anson an d G alen, Comm, in Hippoc. Prog.
1.4, C o rp u s M ed ico ru m G raeco ru m 5.9.2, p. 206. Plague not controllable by medical
m ean s: e.g. T h u c . 2.47.4. C ro p diseases too are ‘from Z eus’, (X en.) Ath. Pol. 2.6.
4 Cf. p. 279, an d W ach sm u th , 224 n. 746.
5 T h u c . 7.18.2, A esch. Ag. 6 9 9 -7 0 5 .
258 Miasma
th ese kinds w ith a n ideal picture o f the good life th a t a righteous
co m m u n ity m ay enjoy.6
O n e form o f p urification th a t the im perilled com m unity
co u ld u n d erg o w as th a t by expulsion o f a ‘scap eg o at’ —in Greek
he w as called e ith e r a ‘m edicine’ (pharmakos) or ‘offscouring’
(katharma), a n d th e rite ’s explicit purpose was to ‘purify the
city ’. S cap eg o at ritu als w ere m entioned earlier in connection
w ith p erio d ic festivals o f renew al, b u t the sources state th a t they
co u ld also be perform ed in response to a specific crisis.7 As their
sy m b o lism a n d significance have been well stu d ied in detail of
late, a fu rth e r analysis w ould be superfluous h ere.8 T h ey seem
a n a rc h a ic fe atu re in the religion o f the classical period, to the
p re o c c u p a tio n s o f w hich it is h ard to relate them directly.
In d ire c t reflections a n d co n tin u atio n s o f the sam e m entality,
how ever, m ay well a p p e a r in classical institutions a n d forms of
b e h a v io u r,9 a n d to investigate this intriguing possibility it will
be necessary to consider, briefly a n d partially, the conceptions
th a t a re asso ciated w ith the rituals.
T h e fu n d a m e n ta l idea is obviously th a t o f ‘one h ea d ’ (or
r a th e r two, in m o st cases) ‘for m a n y ’, b u t there is am biguity as
to w ho th e o n e sh o u ld be. In practice, it was som e m iserable
c re a tu re —physically repulsive, a condem ned crim inal, a beggar
— w ho could be forced into the role o r w ould even accept it
v o lu n ta rily in re tu rn for the prelim inary feeding th a t it b rought
w ith it. {T he b est evidence now indicates th a t the scapegoat
w as n o t k ille d .10) A itiologically, how ever, the pharmakos is not
m erely a w re tch b u t also a villain; the cerem ony com m em orates
th e p u n ish m en t o f one Pharm akos, who, detected stealing Apollo’s
6 H es. Op. 2 2 5 -4 7 (cl. C allirn. Dian. 1 2 2-35), A esch. Supp. 6 5 6 -7 0 9 , Eum. 902-87.
7 See p. 24 above.
8 B u rk ert, SH., C h . 3; G R 139—42; J . N. B rem m er, ‘Scapegoat R ituals in Ancient
G re e c e ’, H S C P 87 (1983). It is not clear w h eth er the obscure notice o f Hesych. s.v.
φ αρμακή, ή χύτρα , ην ήτοίμαζον τοίς καθαίρονσι τάς πόλεις, relates to this or some other
ritu a l. M a n n h a rd t's in terp retatio n o f the scapegoat as an em bodim ent o f fertility who
m u st h im self be cleansed is criticized, after D eu b n er, 194-7, by B urkert and Brem m er.
B ut th e p arallel betw een the treatm en t o f the scapegoat and contem porary m agical
c u res for im potence (H ip p o n ax , Irr. 78, 92, w ith M . L. W est, Studies in Greek Elegy and
Iambus, B erlin, 1974, 144 f.) suggests th a t this w as a com plex ritual in w hich the idea of
‘p u rify in g th e sc a p e g o a t’ coexisted w ith th a t o f ‘purifying the city’ by expelling him.
9 Cf. V e rn a n t, Tragédie, 116-31; m ost im p o rta n t lor w hat follows.
111 Diëgêsis 2.39 f. to C allim . fr. 90.
Purifying the City 259
sac red cu p s, w as stoned to d e a th by th e com panions o f A chil­
les.11 By this conception the pharmakos ceases to be a m ere
vehicle on to w hich, like th e original scapegoat o f the O ld
T estam en t, th e ills of the co m m u n ity are loaded by a m ech an i­
cal process o f transference* a n d becom es in stead , th rough his
crim e, th e a c tu a l cause of w h atev er affliction is being suffered.
A ccordingly, to exile A ndocides will m ean, says his opponent,
a t once ‘sen d in g o u t a pharmakos’’ a n d ‘getting rid o f an offender
a g a in st th e g o d s’ (alitèrios);12 ‘offscouring’ is a loaded in su lt.13
A q u ite different elem ent is in tro d u ced in the m any legends
w hich m ak e m ilitary success or th e safety of a city d ep en d en t on
the sacrifice o r v o lu n tary self-oblation o f a person o f especially
high v alu e —th e fairest virgin in the land, the king’s d au g h ter, or
even th e king h im self.14 T his m ight in origin be a q u ite distinct
co n cep tio n , since d e a th ra th e r th a n expulsion is here essential;
b u t, if so, a co n tam in atio n o f th e two forms seems early to have
o cc u rre d . L ate sources speak o f virgin sacrifice as a
‘p u rific a tio n ’, 15 a n d a hellenistic ro m an cer in tro d u ced the sac­
rifice o f tw o h an d so m e young m en into the account of
E p im e n id e s’ fam ous cleansing o f A th en s.16 M ore im portantly,
H e ro d o tu s tells how the A ch aean s h ad once been on the point
o f ‘m ak in g A th am as (th eir king) a purification for the country
a n d sacrificin g h im ’. 17 T h e lan g u ag e is significant; A tham as is
a n a n im a te ‘p u rificatio n ’ ju s t as the scapegoat is an an im ate
‘m e d ic in e ’. By a final tw ist th e p erson of high social value m ay
cease to be an in n o cen t o b latio n a n d becom e instead the pol­
lu te d cau se o f his n a tio n ’s affliction. T his m ay have been the
case w ith A th a m a s in the legend w hich H ero d o tu s refers to,
since he h a d ‘p lo tte d th e d e a th o f P h rix u s’, an d th ere is p erhaps
a reflection o f su ch m odes o f th o u g h t in the Oedipus Tyrannus ,18 A
11 Istro s, 334 F G rH fr. 50; pro b ab ly ailion for a festival o th er th an th e A ttic, cf.
Jaco b y , a d loc.
12 (Lys.) 6.53.
13 LSJ s.vv. κάθαρμα, φαρμακός.
14 See S ch w en n , 121—39.
15 S eneca, A g. 163, cf. Tro. 634 f.; Achilles T a tiu s 3.12.1, 3.16.3, 3.19.3, 5.18.4.
16 N ean th es o f C yzicus, 84 FG rH fr. 16, ap. A th. 6 0 2 c -d (cf. D .L . 1.110); declared a
fiction by P olem on cited in A th ., ibid. (a fact often neglected in m odern works).
17 H d t. 7.197.3.
18 See esp. J . P. V e rn a n t, loc. cit., also B urkert, S H 65. For P entheus, D odds on F.ur,
Batch. 963.
260 Miasma
c le a r m y th ic al exam p le is L ycurgus, w ho b ro u g h t b arrenness to
th e la n d o f the E d o n ian s by his offences against D ionysus. His
su b jects, in stru c te d by an oracle, p u t him to d e a th .19
O n e w ay o f re la tin g these conceptions to historical behaviour
w o u ld be to look a t the different categories to w hich ritual and
m y th ic a l scap eg o ats belonged, a n d consider to w h a t extent it
w as n a tu ra l to seek w ithin them n o n -ritu al scapegoats, persons
to b e b lam ed for the m isfortunes o f the com m unity. An
a p p ro a c h as general as this m ay o f course obscure im p o rtan t
differences, since it is m ore in terestin g a n d m ore surprising if a
g e n e ra l is b lam ed for crop failure th a n for failure on the field of
b a ttle . I t will therefore be necessary to consider not merely
w h a t categories o f p erson are identified as th reaten in g or cor­
ru p tin g presences, b u t also in w h a t circum stances a n d by w hat
m e a n s they a re felt to w ork th eir h arm .
A n obvious dichotom y am o n g the scapegoats o f m yth and
ritu a l is th a t b etw een the socially elevated a n d debased. (In the
leg en d s o f O e d ip u s a n d C o d ru s, it has been noted, the one is
tra n sfo rm e d in to the o th e r.)20 I t is tem p tin g to see here two
conflicting diagnoses o f the causes o f public m isfortune, corrup­
tio n o r in co m p eten c e on high, a n d subversion o r envy a t the
b o tto m . T h e tension betw een these diagnoses is perhaps
reflected th ro u g h tw o fam iliar ch aracters, the p o rtray al of
w h o m h a s b een th o u g h t to be influenced by the figure o f the
sca p eg o at. A esop w as ugly a n d a slave; in one version he was
p u t to d e a th o n a false charge o f stealing sacred vessels (like the
o rig in al P h a rm a k o s).21 T h ersites w as base-born a n d deform ed;
he d ied a t th e h a n d s o f A chilles (th e com panions o f Achilles
killed P h a rm a k o s), possibly, in one varian t, for the sam e crim e
o f pilfering tem p le p late.22 B oth figures have relations o f a
d istin c tiv e kin d w ith th eir social superiors. T h ersite s’ essential
ac tiv ity is to ‘q u a rre l w ith the kings’:23 he is indeed a kind of
‘’ A pollod. 3.5.1.
20 e.g. B u rk ert, S H 65.
21 See A. W iechers, Aesop in Delphi, M eisenheim , 1961, 3 1 -4 2 for A esop as scapegoat;
a n d now F. R. A d rad o s, Quademi Urbinati di Cultura Classica, n.s. 1 (1979), 9 3-112.
22 T h e evidence for the last point is a vase ap p aren tly illustrating the Achilles
Thersitoklonos o fC h a ire m o n : s e e j. M . P aton, AJA 12 (1908), 406—16 (b u t for a different
in te rp re ta tio n C . R o b e rt, Archaeologischt Hermeneutik, Berlin, 1919, 278—86: further
references in Lexicon Iconographicum Mythologiae Classicae i, Z ürich, 1981, 171). U sener
h a d a lre a d y identified P harm akos an d T h ersites (Kl. Sehr, iv, Leipzig, 1913, 2 3 9-59).
23II. 2.214.
Purifying the City 261
e m b o d im e n t o f ‘g ru d g e ’ o r ‘en v y ’, a pow er th a t was probably
asso ciated , in a n obscure w ay, w ith th e ideology o f the
sc a p e g o a t.24 In H o m er, th e lo u d -m o u th ed cu r is silenced and
h u m ilia te d - b u t n o t before he h as u ttered a n u m b e r o f criti­
cism s o f his co m m an d er-in -ch ief th a t strike hom e. In th e Aesop
legen d , g u ilt is even m ore effectively tu rn ed back ag ain st the
pow erful; th e accu satio n ag ain st A esop is false, a n d his d eath,
c a u sed by th e au th o rities, brings d isaster upon th e land. Aesop
w as, o f course, in v en to r o f th e literary genre th ro u g h w hich the
w eak co uld tactfu lly b u t firm ly ad m o n ish the m ig h ty .25 T hese
tw o figures in v ite us to co n sider the dichotom y betw een noble
a n d d eb a sed scap eg o ats n o t in m erely stru ctu ra l term s - kings
a n d b eg g ars coincide because b o th a re outside the norm - b u t in
term s o f d eb a te ; is th e real villain T h ersites, o r A gam em non?
W e begin w ith T h ersites. As the G reeks did for ritual
p u rp o se s, so m an y cu ltu res h ave in b itte r earn est recruited their
sca p eg o ats am o n g despised sections o f the com m unity or o u t­
siders. T h is is th e m en tality th a t dictates the w itch -h u n t, or the
po g ro m . In G reece, how ever (or a t least in A thens), significant
exp ressio n s o f this a ttitu d e are h ard to find. Envy th reaten ed
th e fo rtu n a te , b o th on a p ra g m a tic a n d m agical level, b u t we do
n o t find th e p o o rer classes being persecuted for perform ing
so rc ery a g a in st th e pow erful. W om en, a su ppressed class, were
to som e e x ten t th re a te n in g ,26 b u t m etics an d slaves seem to
h av e evoked co n tem p t ra th e r th a n fear. N eith er group,
p e rh a p s , w as en o u g h o f a unity to be truly form idable either
p ra c tic a lly o r in th e im agination. (T h e S p artan s, by contrast,
lived in p e rp e tu a l fear o f th eir helots, an d expressed this fear by
m u rd e rin g th em in d ividually an d , on one horrific occasion, en
masse\ they h ad , how ever, good g rounds for this disquiet, and
th e re is no evidence th a t th e helots becam e an im aginative
te rro r on an y o th e r level th a n th a t on which they w ere a real
th re a t.27) In te n se suspicion is found only in connection w ith one
s u b -g ro u p o f m etics, w ho controlled, for th eir ow n profit, a
24 P lu t. De mul. vir. 252e (baskanos) w ith P arth . Amat. Narr. 9.5 (T hargelia): cf.
B u rk ert, S H 72 f.
25 K . M eu li, Herkunft und Wesen der Fabel, Basle, 1 9 5 4 = Ges.Sehr, ii, 731-56.
26 p. 101 above.
27 F ear: e.g. T h u c . 4.80.3, A rist. Pol. 1269a 3 8 -9 . M urder: T h u c. 4 .8 0 .3 -4 , Isoc.
12.181, A rist. fr. 538. Cf. D. M. Lewis, Sparta an d .Persia, Leiden, 1977, 2 7 -9 ;
P. C a rtled g e, Sparta and Lakonia, a Regional History, L ondon, 1979, 176 f.
262 Miasma
d elica te a n d cru cial a rea o f A th en ian life. M ore corn-dealers
hav e been co n d em n ed to d ea th , m ain tain s the Lysianic speech
d ire c te d a g a in st them , th a n m em bers o f any o th er profession.
In th e p a rtic u la r case to w hich the speech relates, p o p u lar fury
h a d b een so stro n g th a t they cam e close to being executed
w ith o u t trial. ‘T h e ir in tere sts’, Lysias points out, ‘are the oppo­
site o f those o f o th e r m en. F or they m ake their biggest profits
w h e n new s o f som e d isaster has reached the city a n d they can
sell th e ir corn d e a r.’ A nd so ‘they look w ith jo y on your
afflictio n s.’28 T h is arg u m e n t from lack o f com m on interest is
often used by the orators to cast an op p o n en t as a n internal
en e m y o f th e s ta te .29 T h e corn-dealer is clearly on the way here
to beco m in g a Je w ish m erch an t; b u t he is a special case, and
th e re is no suggestion th a t even he th reaten s the general well­
b ein g by a n y m ore arcan e m ethods th a n hoarding, ru m o u r­
m o n g erin g , a n d price-fixing, o r from any m ore sinister m otive
th a n greed. T h e re w as no·religious divergence betw een A then­
ian s a n d co rn -d ealers to transform the difference o f interest into
a difference o f fu n d a m e n tal value. O n ly once are m etics as such
kn o w n to h av e com e u n d e r system atic attack, an d , though
ideology m ay h av e lurked in the background, the prim ary
m o tiv e o f th e th irty ty ran ts in 404 was to benefit from rich and
easy pickings in a tim e o f financial straits.30
T h e d a n g e r th a t d em an d e d co n stan t vigilance w as not so
m u c h th a t o f a tta c k from below as infiltration. Shortly after the
ex p u lsio n o f th e P eisistratids, citizens who w ere ‘im pure in
d e s c e n t’ w ere rejected .31 (T h e tim ing, of course, suggests th at
th is w as a p u rifica tio n from ty ran n y as well as a cleansing o f the
c itizen body.) P ericles’ law o f 451/0, excluding the children of
n o n -A th e n ia n m others, ren d ered the citizen body, in principle,
a sealed a n d im p en etra b le u n it.32 Penalties for infiltration were
savage, a n d it is clear from com edy a n d oratory th a t the possi­
b ility w as o n e th a t was constantly p resen t in m an y people’s
m in d s. T h e lan g u ag e of ‘p u rity ’ is som etim es found in this
28 L ysias 22.20,2,13 f. O n the speech cf. R. Seager, Histona 15 (1966), 172-84.
29 Lys. 27.9,29.10, fr. 1. 1 95-200 G ernet, A ndoc. 2.2 f., Dem . 18.198; cf. R. Seager,
o p. cit., 1 8 0 -2 . T h is is the trouble w ith the helots, A rist. Pol. 1269a 3 8 -9 .
30 X en . Hell. 2.3.21; cf. W hitehead, 155.
51 A rist. Ath. Pol. 13.5; cf. W hitehead, 143.
32 W h iteh ead , 149—51.
Purifying the City 263
c o n te x t,33 b u t th ere w ere fu rth e r gro u n d s for this grow ing exclu­
siveness in th e real ad v an tag es th a t attach ed to citizenship in a
p ro sp e ro u s state. I t w as, for in stan ce, a gift of corn from the king
o f E g y p t for d istrib u tio n am o n g th e citizens th a t in 445/4
pro v o k ed a revision o f the rolls, a n d th e expulsion of num erous
im p o sto rs.34 P rovided, how ever, th a t the dividing wall o f
privilege betw een citizen an d o u tsider was well g u arded, A then­
ians do n o t seem to have been greatly concerned a b o u t w hat
w e n t on o u tsid e it. i t w as left to the philosopher to w orry ab out
th e c o n ta m in a tio n o f citizen m orale by contact w ith th e values
o f fo reig n ers.35 A t S p a rta things seem to have been different,
sin ce X e n o p h o n explains th e periodic expulsions o f outsiders as
a d evice to p re v en t the city ‘catch in g sloppy ways from
fo reig n ers’;36 b u t, even here, the th re a t posed by the outsider
w as p ra ctically conceived an d , as S p artan history showed,
realistically.
F o r th e rich in A thens, a stro n g er th rea t from below was
e m b o d ied in th e person o f the sycophant. F or a reborn
T h e rsite s, this w ould surely have been the n atu ra l profession,
a n d m an y a m em b er o f an A th en ian pro p ertied family would
d o u b tless h ave enjoyed th e o p p o rtu n ity to drive a sycophant
figure b ey o n d th e b o u n d ary w ith stones. T his is p erhaps the
level o f feeling to w hich th e T h irty T y ra n ts ap p ealed w ith their
in te n tio n o f ‘purifying the city from the u n ju st’ (a category
id en tified by th eir su p p o rters w ith sy co p h an ts).37 T h e notion of
pu rify in g th e city by the expulsion o f som e d isruptive elem ent
(ekkathairo) is o n e th a t is qu ite com m only found: possible targets
for this tre a tm e n t are luxury, bribery, persons w ith no visible
m ean s o f su p p o rt, ‘co rru p ters o f y o u th ’, an d even, under
ty ra n n y , ‘th e b est citizens’.38 T h e purge, however, is not a form
o f b e h a v io u r confined to societies th a t practise the ritual expul-
33 A rist. Ath. Pol. 13.5, Dem . 57.55.
34 P h ilo ch o ru s. 328 FG rH fr. 119, Plut. Per. 37.4. Such distrib u tio n s am ong citizens,
fam iliar from H d t. 7.144.1, w ere a reg u lar arch aic institution: see L atte, ‘Kollektiv-
besitz u n d S ta a tssc h a tz in G riech en lan d ’, Nachr. Gött. 1945/8 (1948), 6 4 -7 5 = Kl. Sehr
2 9 4 -3 1 2 .
35 PI. Leg. 9 4 9 e-9 5 0 a.
36 X en . Lac. Pol. 14.4.
37 L ysias 12.5, cf. X en . Hell. 2.3.38. It is tem p tin g to try to connect the sycophant
(‘fig-show er’) d irectly w ith the fig-w earing scapegoat: but how?
38 Pl. Resp. 399e, D in arch u s 2.5, D iphilus, fr. 32.17, Pl. Euthphr. 2d, Resp. 567c; cf.
Com. Nov. Incert. Auct. fr. 214 (3.449 K ock) r à μνσαρά ταϋτα θρέμματ' έκόιωκτέον.
264 Miasma
sion o f scapegoats, a n d , in a d etailed discussion o f the ‘purifica­
tio n o f th e city ’, P lato seem s to have o th e r m odels in m ind - the
h e rd s m a n w ho purifies his flock by sorting healthy anim als
from d iseased , th e d o cto r w ho adm inisters purgative drugs.39
P la to does, th o u g h , locate the source o f danger at th e bottom of
th e social scale: ‘h av e-n o ts’ w ho, th ro u g h starvation, clam our
for th e p ro p e rty o f the ‘haves’ are a ‘disease’ w hich can be
p u rg e d by th e d isp a tc h o f w h a t is ‘euphem istically’ term ed a
colony.
O f th e o th e r ideas th a t em erged in connection w ith the ritual
a n d m y th ic al scapegoats, tw o can be passed over briefly. T he
o n e w h ich m akes th e scapegoat a n ‘offender against the gods’
in te rp re ts d isa ste r as the consequence of religious offences by
in d iv id u a l m em b ers o f the com m unity; this conception’s rele­
v an c e to a c tu a l b eh a v io u r will be considered later. T h e sacrifice
o f th e king’s in n o cen t son o r d a u g h te r lacks, not surprisingly,
close h isto rical equivalents. T h e n earest ap p ro ach is perhaps to
be found in H e ro d o tu s’ story o f how the S p artan s, suffering
from th e w ra th o f T a lth y b iu s for the m u rd er o f D a riu s’ heralds,
ask ed in p u b lic assem bly ‘if an y o f the L acedaim onians was
w illing to die for S p a rta ’ by being sen t u p to X erxes for p unish­
m en t; tw o S p a rtia te s, ‘well endow ed by n a tu re a n d in the first
ra n k for w e a lth ’, volu n teered .40 I f the story w ere true, it would
p ro v id e th e m o st sp ec ta cu lar evidence in all G reek history for
self-p u n ish m en t as a form o f religious expiation, since the afflic­
tio n w as m erely th e inability to sacrifice successfully, while the
c u re w as th e loss o f two S p artiate lives. U nfortunately, even if
th e b ro a d o u tlin e o f the story is considered reliable, the reason
for sen d in g th e tw o m en u p to X erxes need not have been the
o n e re c o rd e d by H ero d o tu s. W e a re therefore left w ith no secure
evid en ce th a t th e sacrifice o f the innocent was an y th in g m ore
th a n a tra d itio n a l legendary motif. B ut the m o tif is no t rendered
m ean in g less by being literally unrealistic, a n d p a r t o f w hat a
sto ry like th a t o f Ip h ig en e ia’s sacrifice seems to convey is th a t
o b lig a tio n as w ell as privilege is co n cen trated a ro u n d the person
39 Leg. 7 3 5 a -7 3 6 c (cf. Resp. 501a). In PI. Euthphr. 2d the im age is from weeding. O n
co lo n izatio n as sc ap eg o at expulsion see B urkert, GR 142.
40 7.134.2. T h e oracle w hich fortified L eonidas to accept d e a th a t T herm opylae
(H d t. 7 .2 2 0 .3 -4 ) is generally regarded as a forgery.
Purifying the City 265
o f th e lead er. I f responsibility is th u s placed upon th e shoulders
o f th e c o m m an d e r, it is obviously likely th at guilt too will be his.
W e com e finally to the figure of th e guilty king. H e has often
been co n sid ered in a retrospective light, as a d escen d an t of
F ra z e r’s m agical king, b u t sh o u ld p erh ap s also be seen as a
fo reru n n er of, for instance, D em osthenes, the ‘com m on polluting
d e m o n o f all G reece’. F rom every period o f G reek history there
is ev id en ce for th e co n c en tratio n o f blam e upon the figure of the
leader. O fte n , o f course, th e blam e relates to the specific sphere
o f activ ity , u su ally m ilitary, w ith w hich th e leader is m ost
co n cern ed , b u t w e can still find b ad w eath er being caused by an
o ra to r’s im p iety n e a r the end o f the fourth century. T h e special
influ en ce th a t th e person in a u th o rity exercises over h u m an
affairs ex ten d s also to th e w orkings o f n ature.
Som e m ythological evidence for tu rn in g ag ain st the king has
a lre a d y b een m en tio n ed . A clear exam ple is P lu ta rc h ’s story
th a t, in stru c te d by a n oracle, th e A enianes once stoned their
king to en d a d ro u g h t.41 A h in t th at this m ay once have been a
co m m o n resp o n se to m isfortune comes from a q u estion p u t to
th e k in g ’s son T efefnachus in th e Odyssey. ‘Is it in obedience to a
“ voice o f g o d ” th a t the people h ate your fam ily?’42 T h o u g h
som e c o m m en tato rs in te rp re t the ‘voice of g o d ’ as m erely a
‘m y sterio u sly in sp ired m ovem ent o f feeling’, it seems more
n a tu ra l to follow th e scholia a n d take it as an oracle. R ejection of
th e king in a tim e o f affliction is a logical counterpoise to the
belief, a tte s te d in a w ell-know n passage o f the Odyssey, th at
p ro sp e rity too d ep e n d s upon him :43 ‘A god-fearing king, who,
ru lin g o v er a large a n d m ighty people, m ain tain s straig h t
ju stic e , a n d th e d a rk ea rth bears corn and barley, an d the trees
a re w eighed dow n w ith fruit, an d th e flocks give b irth unfail­
ingly, a n d th e sea produces fish, because o f his good rule, and
th e p eo p le p ro sp e r.’ A belief o f this kind seems to be, in p a rt at
least, a k in d o f m o ral lever for use by subjects ag ain st their ruler.
T h is is ce rtain ly how it is deployed in H esiod’s fam ous diptych
41 Quaest. Graec. 26,297c. D ifferent from cases m entioned already in th at there is no
in d ic a tio n o f th e k in g ’s guilt.
42 3.215.
43 1 9 .1 0 9 -1 4 ; cf. VV. S p ey er, Jahrbuch f . Antike und Christentum 22 (1979), 3 0 -9 .
266 Miasma
o f th e ju s t a n d u n ju st city.44 J u stic e brings health, healthy
ch ild re n , th riv in g crops a n d anim als, calm seas, an d m ilitary
success; in ju stice the opposite. T h e w arning th a t ‘a whole city
often suffers from one b ad m a n ’ applies in principle to any
citizen; it w as in d eed true a t this d ate on the m ost pragm atic
level, since w hole com m unities w ere liable to reprisal strikes by
th e ir n eig h b o u rs for the offences o f individual m em bers.45 It is
clear, how ever, th a t the injustice w hich H esiod w ishes really to
p re s e n t as th re a te n in g the general welfare is th a t o f the kings.
E ven on the purely p ractical level th a t has ju s t been m entioned,
it was only th e crim es o f the m ighty th at really threatened
co m m u n a l w ell-being: if a T ro ja n com m oner h ad carried H elen
h o m e from S p a rta , he w ould o f course have been h an d ed over to
th e aven g in g G reeks to avoid w ar. W ith the powerful, however,
it w as different: ‘I fear lest the city be overcom e along w ith the
k in g s’; quidquid delirant reges, plectuntur Achivi.*6 T h is special
p o te n tia l o f th e king extends also to the m etaphysical level. H e
ca rries a n d em bodies the w elfare o f his people (and cannot
th erefo re be deform ed in body).47 His relation to the divine is
u n iq u e . H e is ‘from Z eu s’; gods m ay interfere w ith the natural
o rd e r by sen d in g a th u n d e rc la p to h o n o u r him ; even his dream s
hav e a m e an in g no t sh ared by those o f the com m oner.48 T he
converse, how ever, is th a t his crim es too have unique signi­
ficance on the religious p lane, a n d his em inence is in this respect
perilo u s. I f he m akes him self u n p o p u la r by injustice, his sub­
je c ts will know w here to lay the blam e w hen disaster occurs.
T h e p lag u e in Iliad 1 is caused by a crim e of the com m ander-in-
c h ie f (once ag ain , an y b o d y else w ould have been forced to hand
44 Op. 2 2 5 -4 7 .
45 L atte , R E s .w Σ Υ Λ Α Ν = Kl. Sehr. 4 1 6 -2 0 . T h e institution survived in certain forms
in to the hellenistic period.
46 A esch. Sept. 764 f., H o r. Epist. 1.2.14. O n ly kings can sin really effectively, PI. Gorg.
5 2 5 d -e .
47 X en. Hell. 3.3.3 (A gesilaus), Paus. 7.2.1 (M edon), also H d t. 4.161.1; cf. J . N.
B rem m er, ‘M ed o n , T h e C ase o f the Bodily B lem ished king’, in Perennitas·. Studi in Onore
di Angelo Brelich, R om e, 1 9 8 0 ,6 7 -7 6 . Cf. still* E kphan to s’ the P ythagorean, dem anding
ab so lu te p u rity in a king, Stob. 4.7.64 p. 273. 12 fT. H ense ( = TheslefF, p. 80. 15 ff).
J" H es. Theog. 96; H orn. II. 11.45 f.; II. 2. 7 9 -8 3 (cf. A rtem id. 1.2 p. 9.19 1Γ. Pack,
D odds, Progress, 178 n. 1; a priestess’s d ream s too are significant, Aeschin. 2. 10, and
D em o sth en es claim s his to be, A eschin. 3.77,219). I t is to the king th at b ad omens
p o rten d h arm , A rr. Anab. 4.4.3—9. L ater m aterial on gods and kings in N isbet/H ub-
b a r d ’s note on H o r. Carm. 1.12.50; cf. Soph. Phil. 139 f., X en. Hiero 8.5.
Purifying the City 267
back C h ry seis im m ed iately ). H a d he persisted in his refusal to
re tu rn th e girl, th e G reek arm y m ight have been forced to turn
to stoning.
By th e histo rical period, th e situ atio n h ad been in im p o rtan t
resp ects tran sfo rm ed . No ‘g o d -n u rtu re d ’ kings rem ain ed except
in S p a rta , a n d th e w ings even o f these w ere clipped. M ore direct
w ays o f disp o sin g o f u n p o p u la r co m m an d ers w ere now avail­
a b le th a n by accu sin g th em o f causing the crops to fail. As a
resu lt, th o u g h th e expulsion o f h ig h -ranking scapegoats was
e n d e m ic in th e society o f fifth- a n d fo u rth -cen tu ry G reece, the
v ictim w as n o t usually accused o f w orking h arm by arcan e or
im p o ssib le m ean s. A general m ay indeed fritter aw ay a n o p ­
p o rtu n ity o r lead an arm y to d isaster th rough folly, cow ardice,
o r co rru p tio n . T h e irratio n al elem ent in these proceedings was
n o n e th e less fully evident to m an y contem poraries. It is already
th e c o m p lain t o f one o f th e first com m anders w ho addresses us
in th e fifth ce n tu ry , A eschylus’ Eteocles: ‘Should we fare well,
god is responsible. B ut i f - m ay it not h ap p en - disaster befalls,
E teocles alo n e will be blam ed th ro u g h o u t the city.’49 T h e trials
o f A th e n ia n gen erals a n d S p a rta n kings are a leitm otif in the
h isto ry o f th e perio d , a n d th e ph en o m en o n seem s to have been
p a n -H e lle n ic .S0 D isaster was constan tly traced back to those
m aleficen t b u t invisible pow ers, b rib ery an d treachery. Unlike
w itc h c ra ft in th e sev en teen th cen tu ry , these two pow ers did
w ork real d am ag e , an d th ere are doubtless good stru ctu ral
re aso n s w hy they posed such a th re a t to G reek states. T hey are
no n e th e less, alo n g w ith conspiracy, the w itches o f classical
G reek so ciety .51
49 A esch. Sept. 4 - 6 .
so G . E. M . d e Ste C ro ix , The Origins o f the Peloponnesian War, L ondon, 1972, 350-3;
P ritc h e tt, ii, 4—33. T h e m o nthly o ath to govern legally w hich S p artan kings were
r e q u ire d to sw ear (X en. Lac. Pol. 15.7) is also revealing in this connection. For another
form o f tu rn in g a g ain st th e powerful note the pogrom s of P ythagoreans in the 5th
c e n tu ry (ex act d etails are u n fortunately not available): B urkert, L S \ 15.
51 L ysias 28 a n d 29 are instru ctiv e in this regard. W e here find an initial assum ption
(o u r c o m m an d ers steal o u r m oney) being defended against em pirical refutation (the
m oney he w as su p p o sed to have em bezzled w as n o t found am ong the property of
E rgocles after his execution) by an u n d em o n strated subsidiary hypothesis (it was
a p p ro p ria te d in tu rn by E rgocles’ associate P hilocrates). T h is process is fam iliar to
a n th ro p o lo g ists from the stu d y o f w itchcraft beliefs (the spell failed because o f a
co u n ter-sp ell). O f course, P hilocrates might be guilty: K .J . D over. Lysias and the Corpus
Lysiacum. B erkeley, 1972, 72.
268 Miasma
T h o u g h m o st o f the faults w ith w hich the powerful were
c h a rg e d now related to th eir functions, there also persisted a
co n c ep tio n o f the m ag istrate as a sym bolic vehicle of his
p e o p le ’s w elfare, w hich defects o f various kinds could
je o p a rd iz e . T h e m ag istrate, like the priest, was required to be
p h y sically in tact; m an y m en in A thens w ere ‘im p u re in body’,
b u t th e co m m u n ity w as only en d an g ered if one o f them held
office; it w as far w orse to have a m an who h ad parodied the
m y steries as g en eral th a n m erely serving in the ran k s.52 W hen a
p o liticia n d escrib es his o p p o n en t as the ‘polluting dem on o f the
c ity ’, h e is p rim arily den o u n cin g his o p p o n en t’s policies an d
th e ir p ra c tic a l consequences, b u t also seeking to suggest that,
w ith su ch a n im p u re rogue in charge, afflictions o f every kind
a re likely to follow. A eschines rep eated ly uses openly religious
la n g u a g e to re p re se n t D em osthenes as a pollution perilous to
g en e ral w elfare. D em osthenes is the cause of every m isfortune;
h e b rin g s c a ta stro p h e to all he associates w ith; he is ‘the pollut­
in g d e m o n o f G re ece’, a n d should be ‘cast beyond the frontiers’,
o r ‘sen t aw ay (apopempo) as the com m on disaster o f the G reeks’.53
T h e ‘lu c k ’ o f th e people d ep en d s on the character, o r m ere luck,
o f its lead er; it th u s becom es w o rth w hile for D em osthenes to
a rg u e w h e th e r he him self o r A eschines is the luckier m an. W ith
th is em p h asis o n fortune, a ch aracteristic fourth-century note
in tru d e s , b u t th e fram ew ork rem ains the ancient conception of
th e c o m m u n ity ’s m agical dep en d en ce upon the leader.54 O u t­
side A th en s, som e saw D ionysius as the ‘polluting dem on of
S icily’. I t w as n o ted th a t b attles w ere won in his absence, bu t
lost in his p re sen ce .55 T w o attack s on D em osthenes neatly bring
52 Lys. 24.13; p p. 97 a n d 169 above. A bsolute requirem ent for those holding public
office to be free from o th er taints, (Lys.) 6.4, Lys. 26.8, A nt. 6.45 f.
53 3.57,1 14 (b ecau se o f his association w ith polluted A m phissians), 131, 157 f., 253.
Cf. D em . 18.159,296, D in arch u s 1.77, an d for the idea o f ‘a country’s p o llu ter’ already
E u r. Or. 1584, S oph. O C 788, E upolis, fr. 120.
54 D em o sth en es unlucky: A eschin. 3.157 f., D in arch u s 1.31 (bad luck contagious),
41,74,77,91,92. L u ck ier th a n A eschines: D em . 18.252 ff. Dem . also points out, 18.255,
th a t his ow n p u n y luck could not dam age th a t o f the city - a nice parallel to the debate
a b o u t m o rta ls p o llu tin g gods, above, p. 145. H e is said none the less to have feared his
ow n luck, P lut. Dem. 21.3. O n ‘luck’ in the period see e.g. Lys. 30.18, Dem . 1.1,2.22
(lin k ed w ith ‘good will o f gods’, cf. p. 14 n. 60), 4.12, 20.110, A eschin. 2.51, Plut. Tim.
passim, esp. 16.1, 21.5, 30.7, 3 6 .6 -7 . Cf. X en . Cyr. 4.1.24, 7.2.24 (charism a o f divine
d e s c e n t). O n co n tag io u s luck see p. 219 above.
55 T im a e u s, 566 F G rH fr. 29, Diod. 14. 6 9 .1 -3 .
Purifying the City 269
o u t th e d ep e n d en ce o f these them es on trad itio n al concepts.
A eschines a c tu a lly q uotes th e passage o f H esiod w hich was
d iscu ssed earlier: w hen th e poet spoke of one m an bringing
affliction o n m an y , A eschines com m ents, it was creatu res like
D em o sth en es th a t h e h ad in m ind. D in arch u s urges the A th en ­
ians to ‘p u t th e affairs o f th e city u n d e r b etter om ens, by turn in g
the d isasters u p o n these lead ers’: he m ight be the voice o f the
o racle u rg in g th e A enianes to stone th eir king.56 T h e com ic poet
I'h ilip p id es even w ent back to blam ing d istu rb an ces in the
w e a th e r u p o n a p o litician ’s crim es. ‘Stratocles, w ho m ade the
acro p o lis in to a tavern, w ho lodged whores w ith the virgin
A th e n a , b ecau se o f w hom the frost scorched the vines, because
o f w hose im p iety th e goddess’ ro b e was cleft in the m iddle, he
w ho assig n ed to m en the ho n o u rs o f the gods’.57 I t is p erhaps
n o t a co in cid en ce th a t this reversion to the H om eric and
H esio d ic co n cep tio n in its m ost m agical form relates to a figure
w ho w as, like th e arch aic kings, h a rd to assail on a direct hu m an
level.58
I t h as so m etim es been suggested th a t ostracism is a kind of
exp u lsio n o f th e scapegoat in secularized form .59 T h e institution
seem s, how ever, to have been functional, if singular; an d it is
n o t clear th a t its sym bolic a n d expressive significance is
sufficiently im p o rta n t in co n tra st to its purely practical effect to
m ak e su ch a n e x p lan atio n a p p ro p riate. T h e original m otivation
h a s been m u ch discussed, b u t th e d an g er th at it w as designed to
m eet, w h e th e r ty ra n n y o r a p araly sin g clash o f rival leaders,
w as ce rtain ly p o litical.60 I t ap p ears as less o f a collective ritu al if
w e believe th e re p o rt th a t has recently been uncovered in a
V a tic a n gnom ologium th a t the vote was initially in ten d ed to be
con fin ed to th e council.61 If, how ever, ostracism is to be
m en tio n ed in this connection a t all, it should obviously be
56 A eschin. 2 .1 5 8 ,3 .1 3 4 -6 ; D in arch u s 1.29.
57 F r. 2 5 .2 -7 , ap. P lut. Dem. 12.7,26.5.
58 S trato cles w as th e tool o f D em etrius Poliorcetes. Sim ilarly, a defixio ag ain st Kas-
s a n d e r an d his circle h as now been found, Ath. M itt. 95 (1980), 230.
59 V e rn a n t, Tragédie, 1 2 4 -6 , developing unpub lish ed ideas o fG ern et.
60 R ecen t co n trib u tio n s: e.g. G. R. S tan to n , J H S 90 (1970), 180—3; J . J . Keaney,
Historia 1 9 (1 9 7 0 ), 1-1 1 ; R. T hom sen, The Origin o f Ostracism, C openhagen, 1972; A. j .
H o llad ay , Greece and Rome 252 (1978), 184-90.
61 J . J . K ean ey and A. E. R aubitschek, A J P 9 3 (1972), 8 7 -9 1 ; cf. G. A. L ehm ann,
Z P E W (1981), 8 5 -9 9 .
270 Miasma
co n n e cted w ith the scapegoat king ra th e r th a n th e scapegoat
beg g ar. O n a n y view, th ere lies behind the institution some
su ch th o u g h t as S olon’s: ‘It is th ro u g h big m en th a t the city is
d e stro y e d ’; in being tu rn ed ag ain st the low w retch H yperbolus,
o stra c ism suffered an abuse, a n d was a b a n d o n ed .62
S om e ju stific a tio n for seeing som e connection betw een the
o strac iz ed politician a n d the scapegoat has p erh ap s been pro­
vid ed by th e a c tu a l o strac a discovered d u rin g this century. T h e
q uite u n ex p e cted n u m b e r o f ca n d id ates th a t they have revealed
has show n how freely individual A thenians exploited the in­
stitu tio n to give v en t to th eir ow n feelings as to w hich powerful
figure th e sta te could best be rid of.63 From the angry an d
v enom ous m essages som etim es ad d ed , it has becom e clear th at
th e q u estio n , ‘W h ich of o u r politicians poses the greatest th reat
to sta b le g o v ern m en t?’ was no t sharply distinguished from
‘W h ich o f o u r politicians is the greatest rogue?’ ‘T h is ostracon
says th a t o f all the cursed prytanes X a n th ip p o s does m ost w rong
(?).’64 A g a in st ‘tra ito rs ’ the ostracon becomes the w ritten equi­
v a le n t o f th e p u b lic curse.65 Religious factors could certainly
p lay th e ir p a r t in encouraging the feeling th a t the state would
fare b e tte r if a p a rtic u la r individual w ere ou t o f it. Several o f the
still u n p u b lish e d o strac a from the C eram eicus are said to allude
to th e A lcm aeonid pollution, a n d one to associate T hem istocles
w ith a curse h ith e rto unknow n.66 T h e X a n th ip p o s ostracon ju st
q u o te d uses one o f the strongest term s denoting a religious
offender, alitêros (th e exact construction is unfortunately un­
clea r). A n o b scu re ostracon n am in g A risteides has been in ter­
p re te d as accu sin g him o f an offence against su p p lian ts.67 M ost
in tere stin g ly , the trad itio n found in an cien t sources th a tC im o n
w as o strac iz ed because o f his incestuous relations w ith Elpinice
now finds s u p p o rt in the m essage urging him to ‘clear out,
ta k in g his siste r w ith h im ’.68 T h o u g h a certain sardonic h um our
62 Solon, fr. 9.3. A buse: T h u c. 8.73.3, Plato C om icus, fr. 187 ap. P lut. Nie. 11.6-7.
63 See R. T h o m sen , op. cit., 70—80.
64 M /L , p. 42, w ith discussion.
65 T raitors: M /L , p. 42. T h e re is now t o o ‘R allias the M ede’ (but note the reservation
o f D . M . L ew is, Z P E 14 (1974), 3). C ursing o f traitors: p. 193 above.
66 H . M a ttin g ly , The University o f Leeds Review 14 (1971), 285—/.
67 M /L , p. 42.
68 H. M a ttin g ly , op. cit., 284.
Purifying the City 271
is u n m istak ab le, it could disguise real feeling. T h e sexual scan-
<lal will ce rtain ly n o t have caused C im o n ’s ostracism , b u t m ight
have h elp ed to focus ind ig n atio n ag ain st the d iscredited leader.
T h e offences o f th e m an y o b scu rer figures ag ain st w hom occa-
ional o stra c a w ere in scribed m ay well have been as m uch
m oral a n d social as political.
This su rvey o f v arious forms o f blam e-throw ing has taken us
I,ir from th e o rig in al situ atio n o f the stricken com m unity seek­
ing a cu re for its ills. T o this it is tim e to retu rn , in o rd er to apply
(o it th e k ind o f analysis a tte m p te d for the diseases o f the
in d iv id u a l in th e previous ch ap ter, by considering the
diagn o ses th a t w ere offered a n d rem edies ad o p ted in specific
cases o f affliction. Som e b u t not all o f the m aterial th a t has been
m en tio n ed in relatio n to th e scapegoat is also relevant here. In
I lie p reced in g discussion, ‘scap eg o ats’ w ere in clu d ed w ho were
a ccu sed o f cau sin g h arm by fam iliar h u m an m ethods. H ere,
how ever, it is w ith specifically religious diagnoses of d isaster
(h a t we are concerned.
A difficulty arises at once ov er evidence. A lthough th e p a t­
te rn o f tran sg ressio n leading to com m unal affliction is u b iq u it­
ous in m ythology, aitiology, a n d legendary history, secure
histo rical evidence for the religious in terp retatio n o f public
d is a s te r is sp arse. T h u cy d id es, for instance, says n o th in g of
w h a t w as said o r do n e on this level a t A thens d u rin g the great
p lagu e, a lth o u g h religious diagnoses m ust certainly have been
p re sen ted . O f th e D elphic oracles, well over fifty, th a t p u rp o rt
to h av e been u tte re d in such circum stances, only one is con­
sid ere d ce rtain ly a u th e n tic by th e latest critic, a n d th at was
given in th e th ird cen tu ry a d . 69 T h e verdict m ay be severe, but
the n u m b er th a t have m uch chance o f being genuine are certainly
very sm all. T h is sta te o f the evidence, however, is p erh ap s not
as serio u s a n o b stacle as it m ig h t ap p ear. H erodotus believed
th a t th e A gyllaeans in cu rred p lag u e by stoning a ban d of
P h o cae an survivors to d eath . Even ifincorrect, the beliefis good
eviden ce for a possible in terp re tatio n o f public d isaster in the
seco n d h a lf o f th e fifth century. N ot all the quasi-historical and
leg en d a ry in stan ces can claim the sam e value as evidence as
this, p a rtic u la rly w hen the plague is in troduced to explain an
69 F o n ten ro se, 442.
272 Miasma
existing ritu al; b tu th ere are enough early stories th at are not
n arro w ly aitiolog ical to suggest the kind o f explanation for
p u b lic m isfo rtu n e th a t m ight have seem ed plausible. T he p a t­
te rn s th a t em erge from the legendary a n d quasi-historical
m a te ria l c a n th en be co m p ared w ith the reliably historical
evidence.
In one p a tte rn , d isaster serves m erely as a stim ulus from the
gods to som e form o f cu ltu ra l change. Several stories in
H e ro d o tu s a re o f this type. C rop-failure induces the Epidau-
ria n s to set up im ages o f D am ia a n d A uxesia, d ro u g h t the
T h e ra n s to colonize C yrene, m ilitary setbacks the S partans to
b rin g ho m e th e bones of O re stes.70 S im ilar stories cam e in time
to ex p lain th e brin g in g hom e o f the bones o f T heseus, Hesiod,
H e c to r, a n d Pelops. T h e re is no suggestion th a t the Epidau-
ria n s w ere cu lp ab le in having no im ages o f D am ia a n d Auxesia,
b u t th e tim e h ad com e for them to set som e up, as it had for the
T h e ra n s to colonize C yrene. T h e affliction was a n adm onition
ra th e r th a n a pu lish m e n t.71
In a n o th e r p a tte rn , the ex p lan atio n lay in a neglect th a t was
in som e d egree culp ab le o f p ro p e r cu lt for a p a rtic u la r god or
h ero . P lag u e forced the L ocrians to resum e the fam ous tribute;
S p a rta n girls w ere born m isshapen because th eir city paid no
sufficient h o n o u r to A phrodite; th e P higaleans w ere punished
for failing to resto re a n im age o f D em eter b u rn t in a fire, the
S icyonians for allow ing two divine im ages to rem ain incom ­
p le te .72 S everal striking stories refer to failure to pay proper cult
to a h eroized O lym pic v icto r.73 M ythologically, O ineus
70 H d t. 5.82.1, 4.151.1, 1.67.2. C o m p arab le cases are 'Parke/W orm ell, nn.
179,223,237, th e aitiological n. 569, a n d the aitia for A thenian D em eter festivals,
P a rk e /W o rm e ll, ii, p. 79 on n. 169, L ycurgus, fr. 8 2 - 5 Blass. O th e r aitiological
p la g u e s /d ro u g h ts , Et. Mag. 252. 11 (D aitis, in E phesus), P /W 559 (B ouphonia), Paus.
8. 2 8 .5 -6 , A pollod. 2.5.11 (Busiris) an d m any o f the following. P opulation m ovem ents
e x p la in e d by p la g u e /c ro p failure: P /W , nn. 305,402,453;477 FG rH fr. 8, a n d cf. H dt.
7.171 (p o p u la tio n change). Plague prevents a prem atu re population m ovem ent, Apol­
lod. 2.8.2.
71 P ark e/W o rm ell, n n. 113,207,409,563. Cf. P lut. Cim. 19.5 for a cult of Cim on
in stitu te d έ ν λ ο ψ φ καί γήςάφορίφ .
72 P/VV, nn. 331,554,493,28. Cl. nn. 45 5 ,4 8 5 -7 .
13 P /W , nn. 388—91, cf. 118. For the hero's m alice cf. P/W , n. 392, P aus. 9.38.5; for
th e ste rn n e ss o f heroes' revenge against insult, A. Brelich, Gli eroi greet, Rome, 1958,226
if., citin g the Anagyrasios daimön (D iogenian 1.25, S uda s.v.), the hero o f T em esa
(P /W , n. 392), P rotesilaus an d A rgus (Paus. 3 .4 .5 -6 ), M inos (H dt. 7.169). O n the
story type cf. J . F ontenrose, ‘T h e H ero as A th lete’, California Studies in Classical Antiquity
1 (1968), 7 3 -1 0 4 .
Purifying the City 273
b ro u g h t th e b o ar ag a in st C aly d o n ia by forgetting a sacrifice,
a n d th e first assu m p tio n of th e G reeks in Iliad Book 1 is th at
Apollo h as sen t th e plague in an g e r for an unoffered hecatom b
or a d isre g a rd e d vow .74
Positive affronts to the gods are o f course all the m ore likely to
lead to d isaster. T h e story o f L aom edon can be taken as rep­
re sen tativ e o f a com m on m ythological p a tte rn .75 L aom edon
d ep riv e d A pollo a n d Poseidon o f th eir wages for fortifying Troy.
A pollo sen t p lag u e, Poseidon a sea-m onster; an oracle told
I ,ao m ed o n th a t release could be secured by exposing his daugh-
(er to th e m o n ster. In this type th e original im pious act is
perfo rm ed by th e king or a m em b er o f his family; disaster strikes
(he c o m m u n ity as a whole, b u t is ab a te d by an act of ren u n cia­
tion o r self-sacrifice on th e p a rt again of a m em b er of the royal
house. W e are back, o f course, w ith the scapegoat king. O c ca­
sionally, a p a rtic u la rly im pious king involves his people in final
d e s tru c tio n .76 A ffronts to gods by com m oners norm ally lead to
th e ir d irec t p u n ish m en t, b u t in o ne or two stories even they can
ca u se co m m u n al disaster: C o m aeth o and M elanippus bring
p lag u e by co p u latin g in a sacred precinct, an d the rape of
C a s s a n d ra by A jax provokes the storm s th at wreck the Greek
fleet.77 In th e case o f A uge too, w ho causes crop-failure by
b ea rin g h e r b ab y in A th en a’s p recinct, it seems m ore relevant
th a t she is a priestess o f A th en a th a n the king’s d a u g h te r.78
In th e q u asi-h isto rical ra th e r th a n legendary m aterial the
affro n t to th e gods norm ally involves a killing on sacred ground.
T h e re a re several story p attern s w hich m ake m u rd er a source of
d isa ste r. O n e is th a t o f th e killing, usually in civil w ar, in
d efian ce o f san c tu a ry . G uilt is norm ally ascribed to a whole
people, o r a ty ra n t.79 In the stories th a t blam e plague on the
killing o f a n in d iv id u al on profane ground, the victim is alm ost
in v aria b ly th e son o f a god, especially d ear to a god, a priest, or
74 H o r n .//. 9. 533 ff., 1.65.
75 A pollod. 2.5.9; for th e p a tte rn cf. ib id ., 2.4.3 (Cassiepeia an d A ndrom eda), 3.5.1
( L y cu rg u s), a n d A g am em n o n in Iliad 1. Also p erhaps Auge, cf. below.
76 H es. fr. 30. 16—19 (Salm oneus).
77 P au s 7 .1 9 .4 -6 (P/VV, n. 556, b u t aitiological); Ajax, p. 185 n. 225 above (affects a
fleet, n o t a co u n try ).
78 A pollod. 2.7.4, 3.9.1. T h e fact o f being priestess sufficiently enhances the outrage
in itself.
79 P/VV, nn. 27,74,75: on the type see F ontenrose, 76 f. M u rd er of am bassadors
cau sed the fall of S y b aris, Phylarchus, 81 FGrH fr. 45.
274 Miasma
fulfilling a m ission pleasing to the O lym pians; norm ally too
su ch stories explain the foundation o f a cult or tem ple, and are
ad hoc creatio n s no t involving su b stan tial figures o f mythology.
A ty p ical exam ple is the m u rd er o f K arnos, origin of the
K a rn e ia n g am es.80 H ero d o tu s has tw o stories set in historical
tim es th a t conform roughly to this p attern . T h e people of
A p o llo n ia b lin d ed E venius for allow ing wolves to attack the
sa c re d h e rd s o f th e sun w hich he was guarding. A nim als and
la n d im m ed iately becam e sterile, a n d Apollo told the Apol-
loniates to p ay E venius any com pensation he chose; Apollo
h im self w ould give him a g re ater gift. Apollo’s gift was prophecy
a n d E venius becam e fam ous as a seer throughout Greece. T he
D e lp h ia n s w ho executed A esop on a tru m p ed -u p charge suf­
fered te rrib ly u n til a t last they found som eone willing to receive
c o m p e n sa tio n for his d e a th .81 A esop, the poet, was obviously
d e a r to th e gods, w hile E venius founded a fam ous m antic
fam ily.
A n o th e r source o f plague is the m assacre o f particularly
defenceless victim s by the d o m in atin g section of the com m u­
nity. W h e n th e A gyllaeans stoned som e P hocaean refugees, all
living c re a tu re s passing the site o f the crim e becam e twisted.
E xcessive savagery against the helpless distorts the natural
o rd e r in th e sam e way in the legend o f the L em nians who
m u rd e re d th eir A ttic wives a n d offspring; an d the m u rd er of
ch ild re n often leads to plague in aitiological stories.82 F or disas­
te r d u e to th e sim ple m u rd er o f one individual by another, on
th e o th e r h a n d , th e evidence is rem arkably sparse. L aius’ death
b rin g s p lag u e in Sophocles, b u t th at, as we saw, is a special case;
b o th p artie s are kings, a n d one the father o f th e other. T he
m u rd e r o f S ty m p h a lu s by Pelops caused d ro u g h t throughout
G reece, b u t h ere too the victim w as a king.83 T h e plague in
80 P /\V , n n. 291—3. A fflictions follow the d eath s o f O rpheus (P/YV 376), Linus (386),
th e Point sen t by A pollo (387), Scephrus (566), an d a μητραγύρτης (572). So too for
v a rio u s in tro d u c e rs o f th e vine or wine (542,544,551). T h e d eath o fC h a rila (570) is an
ex cep tio n ally b ald aition.
81 H d t. 9.93—4, 2.134.4. Cf. the plague th a t struck A thens for the killing of
A n drogeos, A pollod. 3.15.8. In 90 FGrH fr. 45 (from X anthus?) d earth strikes Lydia
b ecau se o f a m u rd e r com m itted by the reigning king’s father.
82 H d t. 1.1 6 7 .1 -2 , 6.139.1; cf. P /W , nn. 130,199,385.
83 A pollod. 3.12.6. R illin g a king especially p ortentous, Horn. Od. 16.401 f., Soph.
O T 257.
Purifying the City 275
I’sophis d u e to A lcm aeon’s presence is probably E u ripides’
inven tio n .84
A final cau se o f plague, little attested , is th e sacking o f a city
I>.ii ticu larly d e a r to a god o r h ero .85
T h e reliab ly h isto rical ex p lan atio n s for m isfortune fall, with
i « i tain obvious m odifications a n d lim itations, into sim ilar p a t­
terns. A lth o u g h th e evidence for religious explanations o f p la­
gue is slight, we know a certain am o u n t ab o u t religious and
m agical responses. T h e quasi-leg en d ary plagues th a t lead to
die in stalla tio n o f new cults find th eir historical co rrelate in
i nits o r tem p les founded in response to affliction. T h e tem ples
öl A pollo H e lp e r a t B assae a n d P an R eleaser a t T ro ezen were
said to be thank-offerings for help given d u rin g the g reat pla­
g u e ,86 a n d it has been p lausibly suggested th a t the in tro d u c­
tion o f A sclep iu s’ cu lt to A thens was a reaction to the sam e
ev e n t.87 T w o D elp h ic responses th a t m ight be g enuine relate to
il. The A th en ian s w ere told to set up an im age of Apollo, the
( lleo n aean s to sacrifice a billy g o at to the rising su n .88 T h e sun
p ro b a b ly received th e offerings because o f p lag u e’s sym bolic
co n n ectio n w ith blazing h ea t,89 w hich would m ake this a semi-
m ag ical rem ed y . T h e sam e kind o f am oral m an ip u latio n a p ­
pears in th e C y re n a e a n custom o f countering the onset of plague
I>y sacrificing a red goat to Apollo A verter before the gates.90
A pollo o f course w as th e p re-em in en t av erter o f plague th rough­
o u t a n tiq u ity . H e m ig h t also be its sender, as in Iliad 1, b u t often
en o u g h his reaso n s for anger, if his anger was suspected at all,
m u st h av e been obscure, a n d the appeal to him sim ply an
84 A pollod. 3.7.5, cf. A p p en d ix 7 below s.v. A lcm aeon. P /W , n. 398, a hellenistic
ro m an ce, is a n insig n ifican t exception.
85 P /W , n. 169. Cf. P /W , n. 305.
86 P aus. 8.41. 7 -9 ,2 .3 2 .6 .
87 See e.g. A. B urford, The Greek Temple Builders at Epidaurus, Liverpool, 1969, 20 f. See
too D. M . Lew is, A B S A 55 (1960). 193 f., on IG I 3 130.
88 P aus. 1.3.4, 10.11.5 (P/W ', nn. 125, 158). Even Fontenrose, 330, seems disposed in
th e ir favour.
89 S oph. O T 27, 176, 191.
90 L S S 115 A 4—7, cf. A p pendix 2. F or fu rth er sem i-m agical techniques see Paus.
2.34.2 (M e th a n a ): tw o halves o f a sacrificed cock are carried aro u n d a vineyard to
cre a te a m ag ic circle ag ain st d am ag in g winds; D .L.8.60: Empedocles catches plague-
b e arin g etesians in ass-skin bags; Plut. Quaesi. Conv. 694 a - b : sacrifice o f black bull to
Βονβρω στις a t S m y rn a (th e bull perh ap s em bodying the Βονβρωστις)·, Paus. 9.22.1:
H erm es av erts a p lag u e at T a n a g ra by carry in g a ram around the walls (aition for a
s ta tu e o f H erm es K rio p h o ro s); Paus. 2.13.6: the Phliasians set up a bronze goat as
pro tectio n for th eir vines ag ain st the ‘g o at’ star.
276 Miasma
a tte m p t to secure the aid o f the relevant divine specialist.91
T h o u g h rem edies based on m agical m anipulation, or supplica­
tio n o f a h ea lin g deity, could no d o u b t coexist w ith a diagnosis
in term s o f religious guilt, it is interesting th a t the attested
p ra c tic a l responses should be on the level o f sacrificing a billy
g o at, n o t ‘d riv in g o u t the p o llu tio n ’. T h e A th en ian purification
o f D elos in 426/5 is a p artia l exception, if we accept D iodorus’
very p lau sib le view th a t this was provoked by the plague;92 but,
even here, th e pollution th a t was identified as a cause was
m erely a ritu a l offence against the god specifically associated
w ith th e disease.
M o st o f th e actu al explanations o f o th er forms o f disaster
h a v e a lre a d y been m entioned in o th er contexts, an d can be
listed su m m arily . T h e y fall alm ost w ithout exception into the
ca te g o ry o f violation o f divine rights. W here m ythologically the
g u ilty p a rty is th e king o r a m em ber o f his family, in the
h isto rical in stan ce s blam e lies w ith the com m unity as a whole,
s u b s ta n tia l p o rtio n s of it, or its representatives. I f the sum m on­
in g o f E p im en id es to A thens was indeed provoked by plague,
th is w as d u e to a m assacre o f su p p lian ts by m agistrates.93 T he
d e v a sta tio n o f A thens in 480 w as pro b ab ly caused by A thenian
in v o lv em en t in th e b u rn in g o f the tem ple of C ybele a t S ardis.94
Im p ie ty a g a in st a tem ple o f Poseidon provoked the tidal wave
th a t sw ep t aw ay a P ersian b atta lio n .95 T h e great S p artan earth ­
q u a k e w as p u n ish m e n t for the m assacre o f helot suppliants,
w h ile they suffered setbacks d u rin g the A rchidam ian w ar b e­
ca u se they en te re d upon it in violation o f their oath. T h a t at
least w as o n e view a t S parta; o th ers detected pollution on the
91 C f. F. G . VVelcker, ‘Seuchen von A pollon’, Kleine Schriften iii, Bonn, 1850, 33—45
(b u t on th e coins o f S elinus discussed there see A. H . Lloyd, The Numismatic Chronicle 15s
(1935), 73—93); O . YVeinreich, Ath. M itt. 38 (1913), 6 2 -7 2 = Ausgewählte Schriften i,
A m ste rd a m , 1969, 197-206; Nilsson, GF 174 (Apollo H ekatom baios); L. D eubner,
Neue Jahrb. 43 (1919), 3 8 5 -4 0 6 ; A. Severyns, Recherches sur la Chrestomathie de Proclos ii,
P aris, 1938, 128 (h ealin g p aean); Nilsson, GGR 541. Apollo can still send evil in the
classical p eriod (cf. the E ry th rae p aean , J . U . Powell, Collectanea Alexandrina, Oxford,
1925, 140), b u t in g eneral is invoked as av erter rath er than sender o f plague (see e.g.
S oph. 0 7 '2 0 3 —15, w here the plague is blam ed on Ares; and W einreich, op. cit.)
92 D iod. 1 2 .5 8 .6 -7 . T h u c. 3 .1 0 4 .1 -2 says nothing o f this m otive, b u t his chronology
is co m p atib le w ith it, cf. 3.87.
93 p. 211 above.
94 H d t. 5. 102.1. T h e connection is an inference from 7.133.2, d raw n e.g. by Stein on
7.133.
95 H d t. 8.129.
Purifying the City 277
llirone, since king P leisto an ax ’s re tu rn , it was said, had been
Im »cured by b rib in g th e P y th ia .96 A th en ian fortunes by co n trast
w ere b o u n d u p w ith the condition o f Delos; it was p erh ap s the
plague th a t first forced them to purify it, b u t w hen they ex­
ten d ed th e pu rificatio n to the p o in t o f expelling the in h ab itan ts,
m ilitary failu re followed, a n d th e D elians were restored.97 T h e
A cginetans w ere expelled from th eir island beca.use of a m u rd er
on sacred g ro u n d .98 Religious diagnoses becom e ra rer in the
p o st-H e ro d o te a n period (alth o u g h several o f the m ost in terest­
ing com e in fact from T h u cy d id es). T h e sense o f a precise and
d irec t link b etw een crim e a n d p unishm ent, w hich allowed
I lero d o tu s to discuss carefully w hich o f several offences a mis-
Ibrtu n e w as cau sed by,99 was p ro b ab ly giving way in the fourth
ce n tu ry , even am o n g the religious, to vaguer an d th u s less
p o ten t n o tions. W e h ear none th e less th at S p artan and T h e b a n
m isfortunes in the first h alf o f the cen tu ry were due to violation
ol o ath s, w hile H elice an d B oura w ere o b literated by a tidal
w ave b ecau se they d isru p ted o th er com m unities’ devotions to
P o seid o n .100 T h e S p artan defeat a t L eu ctra was d u e to an
a n c ie n t rap e, w hich led to suicide; th e culprits w ere individual
S p a rta n s , b u t stories th at rep resen t S p artan s as sexual o p pres­
sors o f d e p e n d e n t peoples are com m on, and seem to reflect on
th e S p a rta n s as a co m m u n ity .101 T h e d istan t cause was anyw ay
not sufficient to overcom e entirely the S p artan in stin ct to blam e
a king; m en rem em b ered now the oracle th at w arned against a
lam e kingship, in defiance o f w hich A gesilaus had been ap ­
p o in te d .102 A p leth o ra o f an cien t crim es, finally, was adduced to
ex p lain the d estru c tio n o f T h eb es by A lex an d er.103 It is intrigu­
ing th a t th e sam e event re b o u n d ed against the destroyer; com-
96 I h u e. 1.128.1, 7.18.2, 5.16.1. F or the last cf. 90 FGrH fr. 44.7 (from X anthus?),
d ro u g h t afflicts L y d ia u n d er usu rp in g king.
97 T huc. 5.1, 32.1. I'he A th en ian s also tu rn ed against Pericles in response to the
p lagu e, a n d in d eed b lam ed him for it, b u t n ot, to o u r knowledge, on an y su p ern atu ral
level (T h u c . 2 .5 9 .1 -2 , 6 5 .1 -3 , Plut. Per. 3 4 .3 -3 5 ).
98 H d t. 6.91.
99 H d t. 6.84.3, 7.133.2.
100 p. 188 n. 244 an d p. 176n. 181 above.
101 L eu ctra: p. 198 above. Stories: P lut. Cim. 6 .4 -6 , Narr. Am. 773f-774a.
102 P lu t. Ages. 30.1.
103 A rr. Anab. 1.9.7. - m edism , d estru ctio n o f P lataea, proposal to destroy Athens.
( )n delay ed p u n ish m en t o f states cf. Isoc. 8.120.
278
Miasma
m u n a l responsibility here ap p ro p riately gives way once again to
th a t o f th e k in g .104
F ro m all this evidence there em erges a conclusion th at m ust
be su rp risin g to anyo ne w hose p ictu re o f com m unal pollution is
based o n th e openin g o f Oedipus Tyrannus. W h a t is com m only
a n d loosely referred to as collective religious responsibility has
tw o th eo re tic ally d istin ct form s. O n e m akes the crim e or im pur­
ity o f a n y in d iv id u a l m em ber o f a com m unity a d an g er to the
w hole, w hile by the o ther it is for the offences o f its representa­
tives o r m a ste rs th a t the citizen body suffers.105 T h e form er is
th è d o c trin e o f the Tetralogies, w hich ascribe crop-failure to the
p re sen ce o f a single unpu n ish ed m u rd erer in the sta te ;106 it also
a p p e a rs c o n sta n tly in connection w ith sea voyages. T h e latter,
how ever, h as tu rn e d ou t to be the basis not m erely for the
c o n c e p tio n o f the scapegoat king, b u t also for the religious
in te rp re ta tio n o f com m unal d isaster in general. V irtually no
in sta n c e h a s em erged from eith er legendary or historical m ate­
rial o f a collective m isfortune blam ed upon an ordinary indi­
v id u a l’s offences. T his is a m erely em pirical observation - in
p rin c ip le , as the Tetralogies show, a cat m ay p ollute a king as well
as look a t him - b u t not the less interesting for th at. F or this
p re d o m in a n c e o f pollution from above, two explanations m ay
be suggested. O n e relates to the size o f com m unities an d to
synoecism . T h e re is, perh ap s, som ething ridiculous in the idea
o f a social g ro u p as large as classical A thens being punished for
th e offence o f a n individual m em ber, unknow n to virtually all of
its m a n y th o u s a n d in h ab itan ts. In relation to a ‘city’ o f the size
en v isag e d by H esiod, on the o th er h an d , the proposition a p ­
p e a rs m ore reasonable; an d the sm aller the com m unity be­
com es, th e m ore reasonable it ap p ears. T h u s it is upon the o ther
m em b ers of restricted an d clearly defined social groups (those
sh a rin g a ship, o r a sacrifice) th a t pollution is m ost likely to
w ork its effects. If we h ad access to the d eliberations o f an
afflicted dem e, o r p h ra try , we m ight find in terp retatio n s under
co n sid e ra tio n o f ju s t the kind th a t we miss for A thens as a
104 K p h ip p u s, 126 F G rH fr. 3, Plut. Alex. 13.4.
105 Cf. D. D a u b c , Studies in Biblical Law , C am b rid g e, 19+7, C h. 4; also Douglas, 100:
'I n g en eral, w e can distinguish beliefs w hich hold th at all m en are equally involved with
th e universe from beliefs in the special cosm ic pow ers o f selected individuals.’
106 A bove, p. 129; cf. too PI. Leg. 910b.
Purifying the City 279
whole. I t is in terestin g th a t tw o tow ns w hich do envisage an
o rd in a ry in d iv id u a l’s co n d u ct affecting the w eath er (one, it is
I rue, only in P lu ta rc h ’s day) are b oth of com paratively sm all
size. T h e p eople o f D o d o n a asked th eir god: ‘Is it because o f
som e h u m a n ’s im p u rity th a t we are suffering this storm ?’107
P lu tarch tells how it w as n o rm al p ractice ‘w hen an earth trem or
o r d ro u g h t o r o th e r p o rte n t h a d o ccurred, for th e T an a g rae an s
to in v estig ate a n d take a lot o f tro u b le ab o u t finding out
w h e th e r a w o m an h ad ap p ro a c h e d the place (shrine of a
‘w o m an -h atin g ’ hero) w ithout being d etected.’108 Full collective
resp o n sib ility , therefore, p erh ap s properly belongs to life before
synoecism . W e saw , how ever, th a t even H esiod, m an o f Ascra,
envisages chiefly th e injustice o f th e kings as th reaten in g the
n a tu ra l o rd e r, a n d this leads to th e second explanation. T h e
ind iv id u al is held in check by a tig h t m esh of h u m an sanctions.
The king is not; n o r is the com m unity. S u p ern atu ral constraints
.ire th erefo re im posed upon th e king (his subjects will rem ind
him o f th e m ). In th e sam e w ay, divine p u n ish m en t forces the
co m m u n ity to a d h e re to its ow n general ideals, how ever free
a n d te m p te d it m ig h t be to violate them in a p a rtic u la r case.
There w as no one to avenge th e rights of the helot suppliants,
slain by th e S p a rta n s in violation o f san ctu ary , except Poseidon.
T h e S p a rta n s , how ever, acknow ledged th at he h ad done so.
T h is id eal o f co m m u n al m oral responsibility is vividly con­
veyed in tw o passages in A eschylus. In the Supplices, the d au g h ­
ters o f D a n a u s invoke upon A rgos the characteristic blessings -
th riv in g crops, successful b irth s, freedom from disease an d civil
strife —th a t m ark th e ju s t an d prosperous city. A sim ilar prayer
for A th e n s is u tte re d by th e E u m en id es.109 B oth occur at
m o m en ts w h en pollution, w hose counterpoise they are, has ju st
b een a v e rted , a n d in bo th cases th e pollution w ould have been
p u b lic n o t m erely in effect b u t also in origin. Pelasgus points out
to th e D a n a id s th a t ‘you are n o t sitting at the h ea rth of my
p alac e (b u t a t p u b lic altars) . . . the (d an g er of) pollution
e x ten d s to th e w hole city .. .’,10 a n d it is in d em ocratic assem bly
107 S E G x ix 427.
Quaest. Graec. 40, 300f. N ote too the story in Ael. VH 8.5: a fleet is held in h arbour
by c o n tra ry w inds, a n d seers declare th a t pollution is the cause.
109 Supp. 6 5 9 -7 0 9 , Eum. 9 0 2 -8 7 .
110 365 f.
280 Miasma
th a t th e ir adm ission to the city is agreed. T h e anger o f the
E u m en id es, w hich A th en a w ith difficulty allays, extends to the
A th e n ia n s as a w hole, because O restes has been ac q u itted by a
co u rt re p re se n tin g the en tire citizen body. T h e th rea t o f com ­
m u n a l affliction is the price a t w hich, in A eschylus’ im pressive
vision, th e possibility o fa truly com m unal w ell-being is secured.
A final o b serv atio n reverts to scapegoats; or ra th e r, the a p p a ­
re n t ab sen ce o f scapegoats in situations o f the kind th at have
b een discussed. E agerly though persons guilty o f disaster on the
h u m a n level w ere sought out, a n d possible though it was
th eo re tic ally to ascribe su p e rn a tu ra l afflictions, too, to guilty
h u m a n s, a ce rtain fatalism is in fact a p p a ren t. T h is is not a
fatalism o f inaction, since supplications to the gods were of
co u rse m ad e, b u t a cut-off p o in t in the search for a m oral and
religious ex p lan a tio n o f h u m a n m isfortune, a willingness to
a c c e p t a c e rtain ran d o m n ess in ‘acts o f G od’. In a late fiction,
D e lp h i urges a plague-stricken city to seek o u t ‘the im pious
o n e ’;111 in th e Oedipus Tyrannus, too, the oracle institutes a hunt.
In a n y G reek com m unity th ere w ere no d o u b t im pious indi­
v id u als en o u g h , lurking u n d etected or unprosecuted, w ho
could h av e been soug ht o u t in tim e o f crisis a n d expelled. In
h isto rical p ractice, how ever, oracles seem to have been m ore
likely to u rg e com m unities to set u p a statu e of Apollo than to
d riv e o u t th e im pious one. I f the inquiries, ju s t m entioned, of
th e D o d o n a e a n s a n d T a n a g ra e a n s h ad revealed pollution as
th e cau se o f d istu rb a n c e , som e kind o f h u n t for the guilty party
w ould p e rh a p s have been started ; b u t it seems equally possible
th a t th e re cip ien ts w ould m erely have used the inform ation to
d e te rm in e th e a p p ro p ria te form o f expiatory sacrifice. A then­
ian s w ere in m an y contexts urged vehem ently to ‘punish the
g u ilty ’; b u t Pericles in tim e o f plague was p erh ap s m erely
re ite ra tin g a n accep ted ideal in encouraging them to accept
‘w h a t cam e from th e gods’ w ith fo rtitu d e .112
111 (P lu t.) Parallela Minora 310b.
1,2 T h u c . 2.64.2.
10
PURITY AND SALVATION
S ocrates in th e Phaedo is m ad e to u phold a rad ical dualism of
m ind a n d body. T h e philo so p h er can n o t b u t welcom e d eath,
1»ecause th e n he will be ab le to achieve th at sp iritu al contem pla-
I ion o f sp iritu a l reality w hich he has, indeed, alw ays aspired to,
b u t w hich th e body w ith its incessan t dem ands a n d deceptions
lias b a rre d him from attain in g . T h e sep aratio n of body an d soul
in d e a th does n o t in itself lead to su ch a vision. R eality is pure
a n d u n a d u lte ra te d , b u t d u rin g its sojourn in the body the soul
m ay a c q u ire a ta in t o f corporeality th a t is not dissolved even in
d e a th . I f he is to enjoy the heavenly vision, the philosopher,
w hile still alive, m u st seek to escape this tain t by resisting the
in telle ctu al a n d em otional d em an d s o f life in the body. T his
purp o sefu l asceticism is rep eated ly expressed in the language of
p u rificatio n . T h e soul th a t has sh u n n ed the body and tu rn ed in
u p o n itself escapes ‘p u re ’ at d eath , dragging n othing corporeal
w ith it; b u t th e soul th a t has been a slave of sensation dep arts
‘p o llu te d ’, so enm eshed in th e physical th at it can still be seen as
it flits, a sp ectre, aro u n d the place w here its body lies b u rie d .1
A gain a n d a g a in S ocrates speaks in the sam e way of the need to
a p p ro a c h d e a th w ith a soul ‘p u rified ’ from bodily desire.2 Such
la n g u ag e is by no m eans unknow n elsew here in P lato,3 b u t in its
c o n c e n tra tio n h ere is a distinctive feature of the Phaedo.
P lato is h a lf playfully p resen tin g ab n o rm al doctrine in a
fam iliar guise. T h e tru ism ‘Religious law forbids the im pure to
' 8 0 d -8 1 d .
2 6 5 e - 6 9 d , 8 0 d -8 3 e , 1 0 8 a-c, 113d (post mortem p u nishm ent as a catharsis), 114c (oi
(fiXoaocfiq. ίκανώς καθηράμενοι).
3 Cra. 404a, Resp. 496d, 611 c—d, Leg. 716e άκάθαρτοςγάρ τήν ψ νχήνö γε κακός, καθαρός
ô i ο ενάντιος, πα ρά δ ί μιαρού δώρα οντε &νόρ' άγαθόν οΰτε θεόν εστιν π ο τί τό γε όρθόν
όέχεσθαι. μ ά τη ν ούν περί θεονς ό πολύς έστι πόνος τοιςάνο σ ίο ις.. .-(mental pollution 777d,
872a, Soph. 230e), Tht. 177a, κάθαροις τής ψ υχής Soph. 227c: cf. X en . Symp. 1.4. άνδράοιν
ίκκεκαθαρμένοις τά ς ψνχάς. T h e place o f p u rity in P lato ’s th ought cannot be considered
here: cf. H . P eris, Lexicon der Platonischen Begriffe, Bern, 1973, 284—8; M oulinier.
3 2 3 -4 1 0 ; H . J . S tukey, The Conception o f Purity in Plato, diss. California, 1935 (non vidi);
A .J . F estugiere, Contemplation et vie contemplative selon Platon1, Paris, 1950, 123-56.
282 Miasma
to u ch th e p u re ’ is applied to the necessary conditions for con­
te m p la tio n o f u n a d u lte ra te d reality. P urification becom es the
s e p a ra tio n o f th e soul from the body, and, in place o f w ater,
eggs, a n d th e blood o f pigs, its agents are self-restraint, justice,
co u rag e, a n d intellectual activity itself.4 T he d octrine th a t Plato
h as su b jecte d to this idiosyncratic transposition seems to be
m o re specific th a n the norm al requirem ent o f every Greek
tem p le th a t the w o rsh ip p er should ap p ro ach the gods in a state
o f p u rity . P lato is not referring to a tem porary p re p ara tio n for
ritu a l ac tiv ity b u t to a way o f life whose aim is purification.5
T h is p u rity is so u g h t as a w ay o f salvation; w h at m atters is a
p u re d e a th , for w hich a p u re life is only a prep aratio n . T hrough
P lato we d e te c t cults or doctrines th a t a ttrib u te d to katharmos a
d efin ite eschatological im portance.
H e h im self indicates in one passage the source o f his im agery.6
H e has ju s t a rg u e d th a t the m oral virtues, truly understood, are
a form o f p u rificatio n from th a t anxious w eighing o f pains and
p leasu res w hich is generally m istaken for virtue; he goes on to
su g g est half-ironically th a t ‘those who established our rites’
w ere h in tin g a t this w hen they claim ed th a t in H ades the
u n in itia te d w ould lie in m ud, b u t those w ho w ere ‘purified and
in itia te d ’ w ould live there w ith the gods. It seem s clear that
P la to is referrin g here to doctrines associated w ith the Eleusi­
n ia n m y steries,7 though he m ay have had o th er rites in m ind
too. It has, how ever, recently been show n th at, by the end o f the
fifth c e n tu ry a t the latest, the public p a rt of the ‘E leusinian’
p ro m ise w as ex p o u n d ed in O r p h i c ’ poem s.8 T his m eans th at
a n y d o c trin e referred to by P lato in an E leusinian context m ay
be O rp h ic in origin. B ut w here one speaks o f O rp h e u s one
c a n n o t keep silent ab o u t P ythagoras. T o discover the original
co n n e ctio n b etw een purification an d salvation it is necessary in
fact to co n sid er all the cults th a t m ade prom ises ab o u t the after­
life to th e ir ad h e ren ts. It will also be necessary to define the
m e a n in g o f ‘p u rifica tio n ’ in this context, an d the techniques by
w h ich it w as achieved. W h at was such purification supposed to
4 67b, 67c, 69c.
5 P. B oyancé, R E G 54 (1941), 164 n. 3, cf. B urkert, L S 213, against Festugière, op.
c it., 1 2 3 -8 .
6 69c; b u t not 67c, on w hich s e e j. V. Luce, CR n.s. 1 (1951), 6 6 -7 .
7 N o te ημίν, άμνητος. D oxography, G raf, 100 n. 30.
8 G raf, passim, esp. 139—50.
Purity and Salvation 283
rll'ace? Sins o f th e presen t life, sins o f a previous in carn atio n ,
an c estral sins, o r sim ply an unexplained accu m u latio n o f im ­
purity? D id it take the form o f a single ritu al release, lasting
asceticism , or, as in Plato, m oral purification? W h a t connection
existed b etw een th e fam iliar everyday cleansings o f G reek re­
ligious life a n d th e saving purification? W holly clear answ ers,
I hough, are n o t to be hoped for. I t is p erhaps the very im preci­
sion o f th e co n cep ts involved th a t m akes ‘p u rity ’ a n d ‘purifica-
I io n ’ th e p o te n t religious m etap h o rs th a t they are.
W e b eg in w ith th e m ysteries o f Eleusis. T h e evidence for
purificatio n s, ab stin en ces, a n d requirem ents o f p u rity in this
co n tex t is extensive an d v aried .9 A solem n p ro clam atio n by the
h ie ro p h a n t excluded from th e rites all those w ho w ere ‘im pure
in h an d s o r in co m p reh en sib le in speech’. T h ree days before the
a c tu a l procession to Eleusis, all the candidates w ent down to
the sea to b ath e. It m ay well have been d u rin g the ensuing
perio d th a t they w ere su b ject to restrictions: avoidance of
n a tu ra l p o llu tio n s, o f certain foods, a n d p ro bably also, although
(his is n o t explicitly attested , o f sexual contact. O n one day at
least they w ere req u ired to keep to their houses — a form of
p re p a ra tio n for ritu a l activity th a t is h ard to parallel in Greek
religion. A t som e stage, too, they fasted, alth o u g h the occasion
a n d d u ra tio n o f this fast are u n certain . T h e ‘m ystic pig’ which
each c a n d id a te sacrificed ‘on b eh a lf of h im s e lf10- a n individual
re la tio n th a t is ag ain distinctive - d id not serve for purification
in a n y stric t sense, as its flesh w as e a te n ,11 b u t there w ere further
9 Cf. G in o u v ès, 376 if. Proclam ation: F oucart, 311. Bathing: B urkert, H N 2 S 5 n. 9.
R estrictio n s: A rb esm an n , 76 f., cf. A ppendix 4. N atu ral pollutions: P orph. Absl. 4.16 p.
255.6. E leu sin ian hagneiai p erh ap s lasted three days, Ar. Pax 151, 162 f., cf. L atte, Kl.
Sehr. 26. K eep in g to houses: A rist. Ath. Pol. 56.4. Fast: R ichardson on Hymn. Horn. Cer.
47. 10 Schol. A r. Ach. 747 —not ‘instead o f him self’, as B urkert renders, H N 285.
11 A r. Ran. 338; on th e m ystic pig cf. B urkert, H N 284 w ith references, on its function
M o u lin ier, 126—9, G inouvès, 376 n. 7. F or the inedibility o f purificatory sacrifices see
e.g. A p. R hod. 4.710, L S S 38 A 32,? LSA 79.19, p. 30 n. 65. T he pig katharmos at the
A n d a n ia m y steries (L S C G 6 5 .6 6 -8 ) cleansed the locale, not the m ystai; it is distin­
gu ish ed from th e sacrifice offered υπϊρ τονς πρωτομνστας (ibid.). T h e Eleusinian pig at
m ost co u ld have been a katharmos in the loose sense o f p. 1 0 n .4 2 . PI. Resp. 378a speaks of
it as a sacrifice, stressin g its cheapness. It is som etim es inferred from the reference to ol
σπλαγχνεϋοντες in th e fragm ent o f E u p atrid laws for the purification of suppliants that
the c a th a rtic a n im al could be eaten (356 F G rH fr. 1). W hile th at is not im possible,
n o th in g show s th a t ol σπλαγχνεϋοντες ate the sam e anim al as was used for purification.
Possibly, after the form al purification, representatives o f the state adm itted the sup­
p lia n t by sh a rin g a fresh sacrifice w ith him , ju s t as children were ad m itted to the
p h ra try th ro u g h sacrifice.
284 Miasma
ritu a l w ashings to be perform ed on the road to Eleusis an d on
a rriv a l th e re ;12 w e h ear o f a ‘w ater-m an: the purifier at
E leu sis’. 13 T h ese final cleansings w ere m erely the culm ination
o f a long series o f p re p a ra tio n s for the great revelation. In the
classical perio d , no one could be ad m itted to the greater
m ysteries a t E leusis in the a u tu m n w ithout first having been
in itia te d in the lesser m ysteries a t A grai in the spring. U nfortu­
n ately , very little is know n ab o u t the cerem onies a t A grai. T he
claim w hich is found in late sources th at they w ere ‘as it w ere a
p re lim in a ry p urification for the g reater m ysteries’14 does not
in d ic a te a n y th in g a b o u t the co n ten t o f the rites, as it is m erely a
w ay o f saying th a t the one is a p re p ara tio n for the other; other
w riters o f th e C h ristia n period use different m etap h o rs to m ake
th e sam e p o in t.15 N or can we infer any th in g from a m ention of
‘p u rifica tio n in th e Ilissus a t the lesser m ysteries’, as this is very
likely to h av e b een a m ere p re lim in ary .16 It is m ore im p o rtan t
th a t in o n e tra d itio n these rites are said to have been founded in
o rd e r to cleanse H eracles from the killing o f the C e n ta u rs.17
In te rp re te d literally, this w ould suggest th a t a specific ritual o f
p u rifica tio n from blood-guilt w as perform ed at A grai, but, de­
sp ite a S a m o th ra c ia n an alogy,18 this seems im plausible in a cult
th a t excluded those ‘w ith im p u re h a n d s’, a n d it has generally
b een felt th a t th e H eracles story is an explanation for a m ore
12 W ash in g a t R heitoi: H eysch. s.v. ‘Ρειτοί. S tone m aidens with lu stral w ater outside
telestenon: M ylonas, 202. W ater-carriers built into propylaea: H . H ö rm an n , Die
Inneren Propyläen von Eleusis, Berlin a n d Leipzig, 1932, 43 f. F or lustral stoups in the
E leusinion a t A thens see (Lys.) 6.52.
13 H esych. s.v. ύδρανός (not attested epigraphically), cf. M ylonas, 236 n. 61.
E. S im on, A ik. M itt. 6 9/70 (1 9 5 4 -5 ), 45 ff. an d , independently, N . H im m elm ann-
W ild sc h ü tz , Theoleptos, M a rb u rg , 1957, 21—2 w ith n. 69, have shown th at the so-called
‘K o re as H y d ra n o s’ (e.g. M ylonas, Fig.70), basis o f so m uch discussion o f Eleusinian
‘b a p tism ’, h as n o th in g to do w ith lustration: cf. C. Picard, R H R 154 (1958), 129-45.
14 Schol. A r. Plut. 845, cf. C lem . Al. Strom. 4.3.1, p. 249.8 St.
15 e.g. C lem . Al. Strom. 5.70.7, p. 373.23 St. (cited D eubner, 70 n. 10), where
p u rificatio n p recedes th e lesser m ysteries, w hich im p art ‘teaching’ in preparation for
the p u re experience o f the g reater (the distinction from A ristotle, fr.15). For further
m etap h o rical references see Lobeck, 188 note h. Radical do u b ts ab o u t the neo-
P la to n ists’ know ledge o f Eleusis in K . D ow den, R H R 197 (1980), 4 0 9 -2 7 .
16 P olyaenus, Strat. 5.17.1.
17 D iod. 4.14.3; for the purification cf. Plut. Thes. 30.5, Apollod. 2.5.12. O n H eracles
a t E leusis see m ost recently N. R obertson, Hermes 108 ( 1980), 274-99.
18 H esych. s.v. ΚοΙης. (F or use o f the ‘fleece o f Z eus’ by th e d a d u c h see Sud. s.v. Διάς
κώόιον.) F or the connection o f the C a b iri w ith purification see schol. T heocr. 2.11/12;
m ig h t the S a m o th ra c ia n ‘confession’ (B urkert, GR 423 n. 34) have led up to it?
Purity and Salvation 285
gen eral p u rificatio n w hich all in itiates h ad to undergo. Such a
ι Me seem s to be illu strated by the w ell-know n reliefs19 which
■.how a veiled a n d seated H eracles, behind w hom stands a
priestess, h o ld in g a w innow ing-fan over his head or a torch at
his side. A scene to the left p o rtra y s D em eter en th ro n ed in
.plcndour. H eracles seems to be undergoing a ritu al of subm is­
sive ‘s ittin g ’ o f a kind th a t is com m on in initiations and for
w hich ce rta in acts o f th e grieving D em eter in the Homeric Hymn
provide a p ro to ty p e .20 F orm ally this is a purification — the
m ystic to rc h ’ cleanses th e sittin g m an, an d the purificatory
‘llcece o f Z e u s’ is also to be seen on th e re lie fs -b u t its expressive
lorce clearly derives largely from the sym bolism o f ad m ittin g a
( a n d id a te to a new statu s by raising him up from his hum ble
postu re. (T h e sym bolism o f th e m u rd er purification cerem ony
was sim ilar, a n d so the two could be assim ilated in aitiology.)
W e find therefore a t A grai (if the association of this rite with
A grai is in d eed co rrect21) an im p o rta n t rite of purification and
in d u ctio n , th e beg in n in g for th e in itiate of the cycle th at in the
fully dev elo p ed form o f th e m ysteries was only com pleted some
eig h teen m o n th s later. B ut th e cerem onies at Agrai, in addition
to th is p ro sp ectiv e p u rification, d oubtless had su b stan tial con­
ten t o f th e ir ow n. T h ey w ere, one source reports, ‘an im itation of
Ihe ev en ts co n cern in g D io n y su s’.22
It h as seem ed w o rth while to illustrate fairly fully, in this
c e le b rate d case, th e fastings, abstinences, an d cleansings th at
preced ed a m ajo r sacral act. Such p rep aratio n s are recorded
w h erev er a rite re q u ired the w o rsh ip p er’s deep psychological
in v o lv em en t.23 B ut it is im p o rta n t to stress th at they were not
m o re th a n p re p ara tio n s. T h e y w ere not directed ag ain st any
19 R ep ro d u ced e.g. M ylonas, Figs. 83, 84; D eubner, Fig. 7; el*. Antike Kunst 13 (1970),
64—6. B ib lio g rap h y in R ichardson, 2 1 1 -1 3 , Burkert, H N 294 Γ, Antike Kunst, loc. eit.
20 Clf. B u rk ert, H N 294 n. 10, 296 n. 16, also Dem . 18. 259 (note ‘raising u p ’); Hymn
Hom. Cer. 1 9 2 -6 . O n the analogy w ith m u rd er purification see A ppendix 6.
21 B u rk ert, H N 296 denies the connection w ith Agrai because in Hymn. Hom. Cer. the
cerem o n y belongs to Eleusis. B ut the specific reference of the H eracles aition to Agrai
d e m a n d s ex p lan atio n ; an d for the possibility th at the scenes on the Agrai frieze
dep icted H eracles’ initiatio n see M öbius, cited by Nilsson, GGR 668 n. 10. Possibly,
w hen A grai an d Eleusis m ysteries were connected as lesser and greater, the prelim inary
rite w as tran sferred to A grai to em phasize the link. If dissociated from Agrai, the rite
w ould h av e to be assigned to ‘l’in itiatio n p réalab le', cf. P. Roussel, B C H 54 (1930),
5 1 -7 4 .
22 S tep h . Byz. s.v. "Αγρα, cf. G raf, 6 6 -7 8
23 Cf. p. 20 η. 9.
286 Miasma
d o c trin a lly specified pollution; they could be revealed - even
th e solem n sittin g cerem ony - to outsiders through sculpture or
p o etry . T h e in itia te could not proceed to the revelation w ithout
th e m , b u t they did not in them selves contribute anything to his
sa lv a tio n .24 T h e y did, it is true, perhaps acquire a special
sy m b o lic im p o rta n c e in the eschatology o f the cult. T hose not
in itia te d w ere co n d em n ed to lie in the underw orld in m ud; this
m ig h t h av e been because they w ere ‘unpurified’.25 A nother
p u n is h m e n t th a t th rea ten ed them was eternal w ater-carrying;
p e rh a p s th ey w ere conceived as trying, in vain, to fetch the
w a te r for th e purificatory b a th th a t they never took.26 B ut the
p u n ish m e n ts need not be in terp re ted in this way, and even if
th ey are, this m eans only th a t the om itted purification, for the
sak e o f a vivid im age, becam e the sym bol of the om itted initia­
tion as a w hole. (T h e p u n ish m en ts seem änyw ay to have en­
te re d th e E leu sin ian eschatology from ou tsid e.27) All the
so u rces insist th a t the salvation o f the initiate depended not on
p u rity , a m ere prelim in ary , b u t on w h a t he saw a n d heard on
th e n ig h t o f B oedrom ion 20 in the g reat hall o f initiation.
W h e n P lu ta rc h w ished to assu re his wife th a t life did not finish
w ith ph y sical d e a th , he rem inded h er o f the ‘tokens’ not of
E leusis, b u t o f th e m ysteries o f D ionysus.28 D efinite proof th at
esch ato lo g ical hopes could already atta c h to cults o f Dionysus
in th e classical period was finally provided in 1974 by the
p u b lic a tio n o f a gold leaf from H ip p o n iu m , d atin g from the end
o f th e fifth ce n tu ry , w hich declares th a t the p a th to felicity in the
24 So rightly F o u cart, 289.
25 PI. Resp. 3 6 3 c -d ( ‘M usaeus and his so n ’), Phd. 69c, D.L. 6.39, G raf, 103-7 (who
co n sid ers a specific connection with the rite o f ‘w iping off w ith m u d ').
26 PI. Resp. 3 6 3 c -d , Grg. 4 9 3 a -b , P aus. 10.31.9, G raf, 107-120; on the artistic
ev id en ce m ost recently E. K euls, The l l ’ater-Carriers in Hades, A m sterdam , 1974, 3 4-41,
H 3 - 103 (w ith a novel in terp retatio n ). A t m ost it w as by secondary ad ap tatio n th at the
w a te r-c a rry in g w as related to an om itted purification. In origin, it is ju s t a lorm
of fru stra te d activ ity , o f a kind ch aracteristic o f underw orld punishm ents
( I - R a d erm ach e r, Rh. Mus. 63 (1908), 535 Π., G raf, 118n. 118) ; the w ater-carriers are
not filling a b a th , b u t ‘po u rin g into a pierced p ith o s’ (proverbially futile), an d for the
m ain E leusinian purification the initiate did not fetch w ater but w ent dow n to the sea to
b a th e .
27 C f. PI. Grg. 493a ('som e Sicilian o r Ita lia n ’), an d , on the m ud, G raf, 107.
28 Cons, ad Uxor. 61 Id.
Purity and Salvation 287
111<·i life is o ne tro d d e n by ‘in itiates an d b acch an ts’.29 D ionysus
i lin c fo re d e m a n d s a place in this discussion, p articu larly as he
r. .i god w h o in m o d ern , th o u g h n o t ancient, descriptions is
i ilicu d u b b e d a ‘p u rifier’.30 In considering him , it will be neces-
..II y to tak e a c c o u n t o f th e d iversity o f forms in w hich he was
win s h ip p e d .31 H is place in th e official religious a n d even civic
Iilc o f th e city w as as g re a t as th a t o f any o th er god; it is a
i cvcaling d etail th a t in A th en s, in th e sacred m arriag e at the
A u th csteria, he received as a b ride th e wife o f the archon basileus
him self. B u t he w as also th e god who in m yth cam e from
a b ro a d , in d efian ce of the local king, a n d led aw ay the w om en to
i ci kless revelry in th e m ou n tain s. A n d while even m aenadism ,
lor all its su b versive ch aracter, by the fifth cen tu ry belonged to
e stab lish ed religion, th ere also existed unofficial b ands of in­
itiates o f D io n y su s B acchius w ho roam ed the streets o f Greek
Iow ns in ecstasy. Such initiatio n s w ere open to m en (in public
w orship, by c o n tra st, th ere was no place for ecstatic m ales) and
i lie cu lt o f this unofficial D ionysus w as alread y im p o rtan t by the
lilth ce n tu ry th ro u g h o u t the G reek w orld.32 T h e god honoured
in su ch diverse w ays was o f coprse the sam e D ionysus, dif­
feren tiatio n b eing in tro d u ced at m ost by the ad d ition o f an
e p ith e t, a n d th e u n d erly in g unity is expressed in E urip id es’
liacchae, a p lay w hich constan tly cuts across the divisions. It is
clear, n o n e th e less, th a t the so u th ern Italian in itiate’s hopes
a n d fears for th e afterlife w ere n o t necessarily sh ared by the
A th en ian farm er, d ru n k a n d h ap p y a t the A nthesteria.
O f p u rifica tio n or abstin en ce in the o rdinary civic cult, v irtu ­
ally n o th in g is know n. T h e fourteen m atrons who atten d ed the
archon basileus’ wife before her sacred m arriage w ith Dionysus
h ad to sw ear th a t they w ere p u re from intercourse a n d o th er
29 SE G xxvi 1139, cf. m ost recently S.G . Cole, GRBS 21 (1980), ‘2 2 3 -3 8 . O n
D ionysiae scenes in fu n erary contexts cf. Cole, op. cit., 237, B urkert, GR 438 f.
D io n y su s’ early con n ectio n w ith O rp h ism has been confirm ed by the new evidence
from O lb ia , o n w h ich see F. T innefeld, Z P E 38 (1980), 6 7 -7 1 ; W . B urkert, ‘Neue
F u n d e z u r O rp h ik ’, Informationen zum altsprachlichen Unterricht 2 (1980), 36—8; M .L.
W est, Z P E 45 (1982), 17-2 9 .
30 e.g. F arn eil, iv, 300, P. Boyancé, R E A 40 (1938), 171.
31 C f. e.g. M . D etien n e, in Orfismo, 56, 228.
32 H d t. 4.79 (S cy th ia), A r. Lys. I, Ran. 357, Pl. Phd. 69c, LSS 120 (C um ae); an d on the
Ihiasos o f A n acreo n see YV. S later, Phoenix 32 (1978), 185-94. O n hellenistic m aenadism
see now A. H en rich s, H S C P 82 (1978), 121-60.
288 Miasma
p o llu tin g c o n tacts, b u t it is the n a tu re o f the ritu al ra th e r th an of
th e god th a t im poses this re q u irem en t.33 O n the p urity o f the
M a e n a d , a p assag e in the Bacchae offers som e inform ation, b u t it
is h a rd to in te rp re t. It com es in the parodos, w hich in this play
seem s to reflect th e form o f a cult h y m n .34 ‘Blessed is he who —
h a p p y m a n - u n d e rsta n d in g the rites o f the gods is p u re in life
a n d en ters into th e spirit o f the revel band, dancing in the
m o u n ta in s w ith holy p u rificatio n s.’ T h e first u n certain ty con­
cern s the ‘holy p u rifications’; it arises p artly because, in
E u rip id e s ’ lyric m an n er, th eir syntactical relation to the rest of
th e sen ten c e is very loose. A stray item o f evidence attests the
u n su rp ris in g fact th a t w om en m ight w ash in p re p a ra tio n lor
D io n y siae rites,35 b u t, if the reference here is to prelim inary
p h y sical p u rificatio n , the conjunction ‘dancing in the m o u n ­
tain s w ith holy p u rificatio n s’ is puzzling. A nother possibility is
th a t m o u n ta in d an c in g is itself the ‘holy purification’. In m yth,
th e m a e n a d s are. freed by celebrating the rites o f D ionysus from
th e m ad n ess th a t has been caused by rejecting h im .36 O n an
ev e ry d ay level, too, D ionysiae revelry ‘breaks the rope of heavy
c a re s ’.37 I t is likely th a t the C o ry b an tie rites, w hich sim ilarly
cu re d m e n ta l d istu rb a n c e by hom oeopathic m eans, could be
sp o k en o f as a ‘p u rifica tio n ’;38 a n d o bservation o f ph en o m en a of
this k in d form ed th e basis o f A risto tle’s fam ous c a th a rtic theory
33 A bove, p. 85.
34 72—7. In a d d itio n to th e com m entaries see A .J . Festugière, Eranos (1956), 72 ff.
( = Etudes de religion grecque et hellenistique, P aris, 1972, 66 ff.).
35 P au s. 9.20.4.
36 C f. B oyancé, 64—73, M oulinier, 116—18. B ut in the historical period there is, pace
B oyancé, little evidence for a healing D ionysus (D odds, 95 n. 87). His title ‘doctor’
(A th . 1. 22e, 3 6 a—b) he owes to the therap eu tic value o f wine.
37 P ind. fr. 248.
38 E. H o w ald , Hermes 54 (1919), 200 d isp u ted it; b ut I assum e w ith I. M . Linforth
(Univ. Cal. Publ. in Class. Phil. 13 (1944—50), 163—72) th at the h o m o eo p ath ickatharmoi
a n d teletai th a t release from m adness o f Pl. Phdr. 244e are B acchic/C orybantic. Cf.
C ro iss a n t, 66. I t is, how ever, possible th at the expression ‘purifications an d initiations’
really does refer to tw o stages, so th a t the ecstatic dancing is distinct from the katharmos.
PI. Euthyd. 277d a tte s ts for the C orybantes th e often cath artic rite o f thronösis; LSA 23.8
sp e ak s o f ‘w a sh in g ’ the can d id ate. Schol. P ind. Pyth. 3.139b refers to ‘the m o th er’ as
κα θά ρτρια τής μανίας; in D iod 3.58.2 m ore generally C ybele invented purifications for
sick an im a ls a n d children. T o appeal to au th o rs w ho w ere them selves directly or
in d ire c tly influenced by th e fam ous A ristotelian theory o f katharsis is sim ply m isleading
(as e.g. Serv. a d Georg. 1.166, 2.389 Liberi patris sacra ad purgationem animae pertinebant, on
w h ich see R. T u rc a n , R H R 158 (1960), 129—44; o r the passage o f A ristid. Q u in t, cited
by D o d d s, 95 n. 87).
Purity and Salvation 289
<>f trag ed y , a lth o u g h it w as in term s o f m edical p u rification th at
lie in te rp re te d th e m .39 I f this view o f the holy purifications is
correct, th ey are a release from an xiety or m adness ra th e r th an
from g u ilt, a n d im m ed iate psychological w ell-being is m ore
likely to be th eir aim th a n a b etter lot in the afterlife. O n the
m ore literal view, o f course, they are reduced to m ere p re­
lim inaries.
T h e ex p ression ‘is p u re in life’ is less am biguous, b u t m ore
surp risin g . T e m p o ra ry rules o f p u rity in p re p ara tio n for specific
ritu als a re co m m o n in G reek religion, an d it w ould not be
stran g e to find som e im posed u p o n th e m aenad. A special ‘life’,
by co n tra st, is th e d istinctive m ark o f the esoteric O rp h ic an d
P y th a g o rea n m ovem ents. A fam ous fragm ent o f E u ripides’
(Cretans h as som etim es been com p ared , in w hich the chorus of
in itiates ex p lain how they have ‘led a pu re life’ from the tim e
they w ere ‘co n secrated a n d called b ac ch an ts’. ‘I w ear clothes
all o f w hite, a n d sh u n th e b irth (?and d eath) o f m ortals; tom bs I
do n o t a p p ro a c h , a n d I g u ard ag ain st eating food th a t comes
from living c re a tu re s.’40 T h a t indeed is ‘p u rity in life’, b u t it is
h a rd to believe th a t th e o rd in ary m a e n a d -A le x a n d e r’s m other,
as it m ig h t be - observed such restrictions. Even a less rigorous
regim e can scarcely be reconciled w ith the general outlook of
the ch o ru s o f th e Bacchae, for w hom true religion an d true
w isdom a re to av oid excessive asp iratio n s an d th e exaggerated
su b tletie s o f in tellect, an d , accep tin g the values o f the sim ple
m an , to relish th e in n o cen t pleasures of this life.41 D istinctive
rules o f life, by co n trast, sep a rate the w orshipper from the
sim p le m an , a n d d o n o t seem to be observed in G reek culture
excep t as a m ean s to a n eschatological end, in w hich E urip id es’
m a e n a d s show elsew here no in terest. T h e ideal o f ‘p u rity in life’
can be re d u ced to th e fam iliar, if it is in terp reted as a general
av o id an c e o f offence ag ain st th e gods ra th e r th a n the observ­
a n c e o f specific ritu al p ro h ib itio n s,42 b u t the verb th a t is used,
hagisteuö, n o rm ally has a precise ap p licatio n .43 It looks as if
E u rip id e s h as h ere derived a tin t in his p o rtra it o f the Bacchae
39 A rist. Pol. 1342a 7 -1 1 , cf. C roissant, 74 1Γ.
40 F r. 79 A ustin.
41 3 7 0 -4 3 2 .
42 Cf. Soph. O T 864 f.
43 In a D ionysiae context again (D em .) 59.78.
290 Miasma
from in itia to ry cults, w hich did , perhaps, foster eschatological
hopes, ra th e r th a n from trad itio n al m aenadism .44 If this is so,
th e possibility arises th at the purifications too h ad a signi­
ficance b eyond th a t o f m ental release.45
W ith p riv a te in itiato ry cults o f D ionysus we have reached or
a t least com e n e a r to the w orld of the gold tablets. (T hese are
leaves o f gold, inscribed w ith verses ab o u t the afterlife, that
h av e been found in w h at are obviously the graves o f initiates o f a
cu lt.) B ut since the tablets also show in crucial respects the
influence o f O r p h ic o r P ythagorean teaching, these two m ove­
m en ts will have to be considered before reverting to the O rphic—
B acchic rites. W ell-know n problem s o f m ethod at once present
th em selves. N o su re criterion exists for distinguishing early and
late elem e n ts w ithin P ythagorean a n d O rp h ic beliefs, nor for
d ra w in g a clea r line o f d em arcatio n betw een the two schools.
O n th e second question, the trad itio n al tug-of-w ar betw een
p a n -O rp h is m a n d p an -P y th ag o rean ism has given w ay of late to
a re co g n itio n th a t coincidences betw een the two doctrines are
p ro b a b ly m ore im p o rta n t th a n divergences. Ion o f Chios could
a sc rib e a n O rp h ic poem to P ythagoras, an d H erodotus prob­
ab ly said th a t O rp h ic rites w ere really P ythagorean.46 But
sim ila rity does n o t m ean identity, a n d a fu rth er com plication
44 Cf. B oyance, 83 η. 1, J . P. G uepin, The Tragic Paradox, A m sterdam , 1968, 234 -6 ;
A. H en rich s, Z P E 4 (1969), 238 n. 54.
45 F o r co m p leten ess' sake, a few m ore scraps o f evidence for katharmoi in the cult of
D ionysus sh o u ld be m entioned here. ( 1) There w as a proverb Λ έρνη κακών, which was
v ario u sly ex p lain ed (w hat seem s to be the tru e origin, 'a bottom less pit o f evils’, was
m issed). T h ese ex p lan atio n s show either th at purifications w ere perform ed in the
L e rn a e a n L ake (S tra b o 8.6.8, p. 371), o r th a t ofl'scourings w ere throw n into it
(A postolius 10.57, Z enobius 4.86 etc.). As th e L ernaean lake was stagnant, full of
w ater-sn ak es, a n d treacherous to the sw im m er (F razer on Paus. 2.37.5) it is m ore likely
to h av e been a recep tacle for katharmata th an place o fkatharmoi. If this is right, there is no
reaso n to co n n ect these katharmata w ith the m ystery o f D ionysus’ anodos w hich seems to
h av e been celeb rated th e re (P lut. D els, et Os. 364f, Nilsson, G F 2 8 8 -9 0 ). (2) T h e chorus
in the Antigone call on D ionysus to com e καθαραίω ποόί to rescue the city (1144). T his
rev eals n o th in g a b o u t the cult o f D ionysus. T h e chorus, needing purification, turn to
th e ir c ity ’s g reatest god (cf. Soph. O T 210) to supply it: a Pylian would have invoked
P oseidon. (3) E ven if D ionysus’ leap into th e sea a t Horn. 11. 6 .1 3 5 -6 does reflect an
in itia to ry rite o f im m ersion (H . Je a n m a ire , C.ouroi et Courètes, Lille, 1939, 336), its
c h a r a c te r is not one o f purification. (O n leaps into the sea cf. G inouvès, 41 7 ff.) (4) T he
g re a t p u rifier M e lam p u s had Dionysiae connections (H dt. 2.49.2, Paus. 1.43.5). (5)
F o r a later period note SE G xxviii 841.3.
46 D .L . 8.8, H d t. 2.81 (long text); cf. G u th rie, O CR 216—21, D odds 171, n. 95,
N ilsson, Gnomon 28 ( 1956), 21, B urkert, L S 125—32, G raf, 92—4. A fine form ulation in
B u rk ert. G R 445.
Purity and Salvation 291
,ii iscs from th e fact th at Em pedocles, m ost tangible p ro p o n en t
ul a d o ctrin e o f salvation, w as a m an of in d ep en d en t th o u g h t
nid im ag in atio n . W e do not know w h eth er he invented a cruc-
i , i I notion like th a t o f th e crim e o f the daimôn, or w here he
deriv ed it from . H e will here be associated w ith O rp h ism , since
lli.it ex p iatio n o f g u ilt on w hich he insists is b e tte r attested as an
( )rphic th a n as a P y th ag o rean preoccupation; b u t it w ould be
lit lie less p lau sib le to cite E m pedocles as p ro o f o f the im port-
,i nee o f g u ilt in early P y th ag o rean sensibility.47
1’y th ag o ra s su b jected his followers to a code o f restrictions
un iq u e in G reek life. T h is it w as th a t m ade such a profound
im pression on all o utsiders, a n d proved irresistible to comic
poets. H e also ta u g h t a d o ctrin e of m etem psychosis; its exact
iorm is irreco v erab le, b u t it m u st h ave allow ed som e scope for
I lie in d iv id u a l’s co n d u ct in this life to influence th e form of his
next in c a rn a tio n , o r the ‘P y th ag o rean life’ w ould lose m ost of its
p o in t.48 In d e e d , w hen Ion o f C hios says o f Pherecydes, ‘T h u s,
l<>r his m an lin ess a n d decency, he is enjoying a p leasan t life even
after his d e a th - if P ythagoras is truly a sage’, th e distinctive
d o c trin e o f m etem psychosis has d isap p eared from view, and
P y th a g o rea n ism seem s to be seen m erely as a way o f securing a
h a p p ie r p o rtio n in a conventional afterlife, m uch like initiation
i n th e m y steries.49 O n th e o th er h an d , we can n o t be sure to w hat
ex ten t th e th o u g h t o f last things really was a daily concern of the
P y th ag o rean s; a n d it is clear th a t m any o f the rules th a t m ade
up th e P y th a g o rea n life h ad alread y existed in som e form in
G reek c u ltu re , a n d w ere at m ost reapplied by P ythagoras for
esch ato lo g ical ends.
T h e collection o f rules n e a r th e end of the Works and Days is a
p a rtic u la rly re le v an t p arallel.50 A p a rt from th eir im plications
lor P y th a g o ras, it is w orth p au sin g over them because o f their
in trin sic in te re st for o u r them e. T h e H esiodic, like the P ythago­
re an , ru les pro v id e guid an ce on trivial and undignified areas of
d aily existence: n ail-cu ttin g , w ashing, excretion. B oth are often
47 D o d d s, 169 n. 81, B urkert, L S 133 n. 72 and Z u n tz, 265, however, associate him
w ith P v th ag o ras: cf. M . L. W est, Early Greek Philosophy and the Orient, O xford, 1971,
2 3 3 -5 .
■“ B u rk ert, L S 1 3 3 -5 .
49 F r. 30 W est, ap. D .L . 1.120, ( X the fragm ent o fA risto p h o n in D.L. 8.38 ( = 58
D ie ls /K ra n z E 3).
50 7 2 4 -5 9 .
292 Miasma
cast in a d istin ctiv e form, a rule followed by a c u rt explanation
w h ich is usually a w arn in g o f danger: ‘D o n ’t eat o r w ash from
u n c o n se c ra te d pots: a pen alty follows on th a t’ (H esiod), ‘D on’t
tu rn ro u n d a t a boundary: the Erinyes are behind you’
(P y th a g o ra s).51 T h e sam e form ap p e ars later in o th er m agical
w arn in g s: ‘D o n ’t w ear a black robe: for black belongs to d e a th ’
(ep ilep sy p u rifiers), ‘G a th e r th e fruit o f the w ild rose from
w in d w a rd ; oth erw ise there is d a n g e r to the eyes’ (herbalists).52
A n o th e r im p o rta n t sim ilarity is th a t the H esiodic rules are not
iso lated su p erstitio n s b u t are g rouped together as a unified
g u id e to co n d u ct; th us th ere exists a ‘H esiodic life’, ru d i­
m e n ta ry p re c u rso r (on the trad itio n al chronology) o f the
P y th a g o re a n a n d O rp h ic lives. T h e re even ap p e ars in H esiod
th e figure o f the ‘godlike m a n ’ —not, it is true, in a very godlike
p o s tu re .53 T h e o rd in a ry individual can, it is im plied, approach
th e co n d itio n o f th e godlike m an by obedience to the rules. A
k in d o f goal is therefore presen ted , even though there is no
in d ic a tio n o f th e ad v an tag es enjoyed by the godlike m an, an d
n o h in t th a t they relate to an y w orld b u t this.
I f th e verses a re H esio d ’s, P ythagoras was p erp etu atin g a n d
d ev e lo p in g a very an c ie n t trad itio n . T h e very features, how ­
ever, th a t m ak e th e com parison m ost interesting - the presence
in H e sio d o f a ‘life’, a n d a ‘divine m a n ’ —do perh ap s bring the
a sc rip tio n in to d o u b t. T h e re is no difficulty in supposing th at
th e p ro p h e t o f w ork, piety, a n d ju stic e should also have felt
scru p le s a b o u t u rin a tin g w hile facing the su n ,54 b u t it w ould be
s u rp risin g for him to see the avoidance o f this kind o f thing as
th e d istin c tiv e m ark o f a ‘divine m a n ’. T h ese rules find parallels
in sac red books o f th e E ast, the Laws o f Manu, 55 for instance, and
th e a reas o f co ncern th a t they reveal - sexuality, w ashing,
b od ily functions, p u rity o f kitchen utensils —are com m onplace
in m a n y an th ro p o lo g ica l discussions o f pollution. Closely com ­
p a ra b le evidence from the classical period is h ard to find. It
51 Iam b i. Prolr. 21.
52 H ippoc. Morb. Sacr. 142.23 J., 1.17 G. (th e explanations in this section o f Morb.
Sacr. seem to rep re sen t a curious blend o f the original m agical sanctions and rationaliz­
ing glosses by th e H ip p o cratic); T h eo p h r. Hist. PI. 9.8.5. T h e form is parodied in Plato
C o m icu s, fr. 173.
53 731. W est in te rp re ts θείος here as = θεουόής, b ut adm its this to be unique.
54 W est’s in tro d u c to ry note to Op. 724—59.
55 Cf. esp. 4 .4 5 -5 0 , referred to by W est on 7 2 7 -3 2 , 757.
Purity and Salvation 293
w ould be m ost im p o rta n t to know w h eth er a lively interest in
s u c h rules w as endem ic in a rch aic G reek p easan t life, as seems
lo follow if th e ascrip tio n to H esiod is accepted, o r som ething
i ultiv ated in restricted circles as a form o f differentiation from
I lie u n g odlike h erd . U n fo rtu n ately , au th o rsh ip , d ate, an d social
co n tex t o f th e verses rem ain m ost u n ce rtain .56
S everal o f th e H esiodic rules have alread y been m entioned in
o ilier contex ts. T h ey begin w ith a w arning ag ain st pouring
libatio n s w ith u n w ash ed h an d s; th ere are also regulations to
pro tec t th e h ea rth fire from co n tam in atio n by sexuality, the
m ale from co n tam in atio n by the female, and to p rev en t d an g e r­
ous c o n ta c t betw een d e a th a n d p ro creatio n .57 N ine lines are
I’iven u p to ru les a b o u t u rin atio n an d excretion. T h ese were, of
course, im p u re activities, a n d it is plausible th at m ost o f the
I lesio d ic p rin cip les were observed by properly train e d Greeks,
b u t no o th e r texts show the sam e em p h atic a n d explicit pre­
occ u p atio n . C a u tio n is req u ired w ith o th er bodily offscourings
too: ‘D o n o t cu t th e dry from th e green (i.e. finger nails) at a
festival.’58 H e re it is the sacred occasion th at d em an d s respect,
b u t the sun, hearthfire, rivers an d springs also require p articu ­
la r p ro te c tio n from th e bodily processes. T h is concern for the
p u rity o f th e elem ents p erh ap s suggests Persian religious sensi­
bility, for in stan ce, ra th e r th a n G reek, and external influence is
no t to be excluded; b u t once again it is the tone a n d em phasis
th a t cau se su rp rise ra th e r th an the fu n d am en tal value, since fire
is, for all G reeks, an especially p u re elem ent, rivers are divine,
a n d sp rin g s m u st be g u ard ed ag ain st various forms o f con­
ta m in a tio n .59 T h e claim th a t ‘T h e gods are angry w ith the m an
w ho crosses a riv er “ unw ashed in badness an d in h a n d s” , and
give him sufferings afterw ard s’, is particu larly striking.60 W ash ­
ing before crossing a river is otherw ise u n attested in G reece, but
56 T h e m ain linguistic difficulty is in 726, cf. W est ad loc.
57 Cf. pp. 76, 103, 53 above.
58 7 42 Π, also P y th ag o rean , cf'. W est ad loc. For later antiq u ity see Petron. Sal. 104-5:
h a ir a n d nails sh o u ld never be cut a t sea, except durin g a storm (w hen, clearly,
p o llu tio n acq u ires healing power).
59 Cf. L S S 4,50; LSC G 152; IG X II 5. 569; SE G xiii 521. 180-202; PI. Leg. 8 4 5 d -e;
ab o v e, p. 230 n. 131 ; P aus. 3.25.8. C h ry sip p u s praised H esiod's rule, Plut, de Stoic. Rep.
1045a. P ersian resp ect for rivers, H d t. 1.138.2; the rule ol Hes. Op. 739 contravenes it,
as W est notes.
60 740 f. A su b jectiv e sense for κακό της ( = κακός είναι) seems inescapable here (but
see Z u n tz 2 2 9 -2 3 2 ).
294 Miasma
is a logical enough re q u irem en t g ra n te d the two prem isses th at
rivers a re divine, a n d th a t gods should be ap p ro ach ed in purity.
T h e id ea o f bein g ‘unw ashed in badness a n d in h a n d s’, by
c o n tra s t, is sin g u lar, a n d no t in expression alone. O n the one
h a n d , it in clu d es m oral badness o f som e kind w ithin the concept
o f po llu tio n ; on th e o ther, it seem s to treat such badness as
effaceable by w ashing. T h is is a purification from guilt like th at
p re a c h e d by E m pedocles, though w ithout eschatological im pli­
catio n s. H e re too the ascription to H esiod, if accepted, has
su rp risin g im plications.
O th e r rules in the little collection relate not to avoidance of
d irt b u t to th e o rd e rin g o f experience in a m ore general sense.
T h e re is a w arn in g , for instance, against eating from uncon­
se c ra te d pots. B etw een these rules an d those th a t m ore speci­
fically co n cern p u rity no distinction is draw n; ‘G od is indignant
a t ’ all su ch offences alike, o r they are ‘not good’. I t has accord­
ingly b een claim ed, since the characteristic vocabulary o f pollu­
tion is ab se n t, th a t the concept itself is alien to ‘H esiod’.62 But,
th o u g h he speaks o f a broad set o f inauspicious acts ra th e r than
a s h a rp ly defined category o f miasmata, several o f these in au ­
sp icio u s a c ts closely resem ble the miasmata o f classical times,
w hile th e id ea o f m etaphysical ta in t is p resen t in the threat
posed by bodily em issions to fire a n d the sun. C om plicated
p u rific a tio n s a re not p rescribed, b u t, as we saw, w ater is cre­
d ite d w ith the pow er o f rem oving badness as well as dirt. By
m an y c rite ria th e H esiodic rules show m ore sensitivity to the
th re a t o f p o llu tio n th a n do classical authors. ‘D o not expose
y o u rse lf w hen you are stained w ith seed before your h e a rth ’, he
w a rn s. F o r this incom patibility betw een sexuality an d the pure
fire th e re is a possible parallel in H ipponax, b u t none later;63 by
th e fifth ce n tu ry , G reeks h ad pro b ab ly ceased to be troubled
a b o u t c o n ta m in a tio n o f this kind.
T h e P y th a g o rea n symbola o r acousmata64 resem ble the
62 R o h d e, 317 n. 70.
63 p. 77 above.
64 F. B oehm , De symbolis Pythagoreis, diss. Berlin, 1905. (I quote som e o f the sym bols
in w h a t follows by th eir n u m b er in B oehm ’s collection.) C om prehensive bibliography
a n d m asterly discussion in B urkert, L S 166-912. O n the au th en ticity o f individual
sym bols n o te B u rk e rt’s form ula, 188: ‘It is like a gravel pile; there is no pebble of which
w e ca n say th a t it m ust be prim itive rock, b ut any single one m ay b e .’
Purity and Salvation 295
I lesiodic code o f life in co n tain in g a n u m b er o f rules o f purity,
w ith o u t recognizing a distinctive category of pollutions am ong
i lie various d an g e rs a g ain st w hich they w arn. T h e range o f their
i on cerns is very wide. Som e w a rn ag ain st m agical dangers: ‘Do
not w ear a rin g ’; ‘D o not stir th e fire w ith a knife’; ‘Do not step
over a b ro o m ’; ‘W h en you get u p in the m orning, erase the
m arks o f y o u r b o d y on th e b e d .’65 Som e p ro tect from those
spirits a n d d em o n s w ho w ere u b iq u ito u s in the P ythagorean
world. ‘D o n o t pick u p scraps th a t fall from the table; they
belong to th e h ero es.’66 T o a p p ro a c h the gods in the right
con d itio n a n d th e rig h t w ay w as im p o rtan t; tem ples should be
en tered in clean clothing, barefoot, an d from the rig h t.67 Above
.ill th ere w as a concern w ith p artitio n , w ith not confounding
m an a n d god, d ead an d living, sacred an d profane. ‘Pour
libatio n s from th e edge o f the c u p ’; th e ex planation given, ‘so
(h a t m en a n d gods m ay not d rin k from the sam e p a r t’, is no
( lou b t co rrect.68 ‘D o not cu t y o u r h a ir o r your nails a t a festival’;
nails a n d h a ir are d ead m atter, a n d th eir cutting suits a funeral
m ore th a n a feast.69 F u n ctio n s are not to be confused: ‘Do not
ea t from a c h a ir’ (b u t from a tab le), ‘Do not w ipe a chair w ith a
to rc h ’ (an a n c ie n t torch was ra th e r like a broom ), ‘D o n ’t use
ce d ar, lau rel, m yrtle, cypress or oak to cleanse your body or
clean y o u r teeth : they are for h o n o u rin g the gods.’70
A n th ro p o lo g ical evidence shows th a t ap p a ren tly trivial rules
o f co n d u c t m ay assum e sta rtlin g im p o rtan ce because they de­
rive from prin cip les th a t are essential to a p a rtic u la r society’s
o rd e rin g o f th e w orld. H aw aiian s are disgusted by the E uro­
p ean h a b it o f lying now on a n d now un d er the sam e blanket,
becau se it tran sg resses th e fu n d am en tal opposition of the above
a n d th e below .71 T h e case o f a n Eskim o girl, who was banished
for p ersisten tly eatin g su m m er foods in w inter, has become
65 N n. 2 2 ,3 3 ,3 1 ,3 4 Boehm .
66 N. 19. O n P y th ag o rean dem onology cf. B urkert, Gnomon 36 (1964), 5 6 3 -7
67 N n. 1 -3 .
68 N . 8, Iam b i. VP 84, cf. F '-m . II. 16. 2 2 5 -7 . Cf. D.L. 8.34, sacred fish not to be
e a te n , ‘for m en a n d gods should not have th e sam e privileges any m ore th an m asters
a n d slav es.’
69 N . 49.
70 N n. 38, 36, 28.
71 L év i-S trau ss, 144 f. (w ith fu rth er exam ples), who speaks o f ‘m eticulous rigour in
the p ractical ap p licatio n o f a logical sy stem ’.
296 Miasma
n o to rio u s.72 P arallels like these m ay help us to see the Pythago­
re a n rules, too, as deriv in g from principles of order. T h e symbola,
how ever, are no t the norm s o f a w hole society, b u t the refine­
m e n ts a d o p te d by a restricted group; as their nam e, ‘tokens,
p a ssw o rd s’, indicates, they m ark off m em bers of the group from
o u tsid ers. Som e o f the rules o f p u rity seem to have this d if­
fe re n tia tin g function. ‘D o n ’t d ip your fingers in a lustral w ater
sto u p o r w a sh in a b a th house; it’s not certain if the o th er people
w ho use th em are p u re .’73 T h u s trad itio n al conceptions of ritual
a n d physical p u rity are rejected, in p u rity ’s own nam e. Im ages
o f th e gods w ere no t to be w orn by Pythagoreans, because they
m ig h t be b ro u g h t into co n tact w ith polluting objects;74 bu t to
w e a r a rin g w ith a g o d ’s p o rtra it was probably in conventional
term s a n a c t o f piety. T h e rule ‘Do not kill (even) a flea in a
te m p le ’75 is a n oblique rep ro ach to the traditions th a t p re­
scrib e d th e sacrifice o f far nobler anim als on sacred ground.
O n e source states, not im plausibly, th a t P ythagoreans avoided
all c o n ta c t w ith b irth a n d d eath . If so, they regarded
th em selves, like priests, as too godlike to en d u re even those
c o n ta c ts w ith n a tu ra l processes th a t the fact o f being hum an,
w ith m o rtal friends a n d kin, w ould norm ally im pose.76 A bout
sex u a lity th e symbola are surprisingly silent, bu t a strong trad i­
tion cred its P y th ag o ras w ith insisting, am id the loose-living
G reeks o f Italy , on the value o f reciprocal m arital fidelity.77
T h is re actio n ag a in st trad itio n al religion could have two
form s w hich, th o u g h a p p a re n tly opposite in intention, served
th e sam e end. O n e w as th a t o f o u trig h t rejection.78 By their
refusal to e a t a n im al flesh, the P ythagoreans (w hatever the
a ttitu d e o f th e m a ste r him self) isolated them selves from central
in stitu tio n s o f social a n d even political life. T h e o th er was the
re a p p lic a tio n o f trad itio n al elem ents in a way th a t transform ed
th e ir m e a n in g .79 W e find this in p a rtic u la r in connection w ith
72 M . D ouglas, Implicit Meanings, L ondon, 1975, 244.
73 N n . 44—5 B oehm , Iam bi. V P 83.
74 N . 9.
75 N . 6.
76 A lex an d er P olyhistor in D .L . 8.33. Cf. p. 52 above, and E ur. Cretans fr. 79 A ustin.
77 See B u rk ert, 178 n. 94, also D .L . 8.21.
78 CIf. D . S a b b a tu c c i, Saggio sulmisticismogreco, R om e, 1965,69—83; M . D étienne, ‘Les
ch em in s d e le déviance: O rp h ism e, D ionysism e et Pythagorism e’, in Orfismo, 4 9 -7 9 .
79 B u rk ert, L S 190 f., co m p arin g C alvinism . Cf. T u rn e r, 92—5.
Purity and Salvation 297
i lie dietetic reg u latio n s. T h o u g h details are in d o u b t, it is alm ost
i η tain th a t P y th ag o ras m u st h ave know n restrictions attach ed
i n p a rtic u la r existing cults,80 a n d he seem s to have followed not
only th eir form b u t also th e ir content: ‘P y thagoras told his
li il lowers to a b s ta in from . . . a n d everything else th at people
i m id u ctin g sacred rites tell th e w orshippers to av o id .’81 I
w ould be difficult to find any single food th at was definitely first
loi b id d en by P y th ag o ras. W h a t was ap p a ren tly his innovation,
.Hid a d ra s tic one, w as to ch an g e tem p o rary ab stinence, con­
fined to th e p erio d preced in g a ritu al act, into p erm a n en t rules
o f life on w h ich salv atio n d ep en d ed . T h e sam e is tru e to some
e x te n t even o f th e rejection o f an im al sacrifice, since bloodless
i ii Its a n d a lta rs h ad alw ays existed, and the trad itio n al ritual
itself in sisted th a t sacrifice w as a crim e, alth o u g h a necessary
one. A v o id an ce o f n a tu ra l pollutions, too, was m erely the exten­
sion to a w hole co m m u n ity o f beh av io u r th at was probably
I ra d itio n ally p rescrib ed for priests.
A b stin en ce o f various kinds was obviously integral to the
P y th a g o rea n w ay o f life. W h e th e r its goal w ould have been
conceived a n d spoken o f as precisely a ‘pu rification’ is less clear.
P o rp h y ry a n d Iam b lich u s in th eir biographies do indeed pre­
se n t p u rifica tio n as th e key to salvation, the h u b aro u n d which
all P y th a g o ra s’ religious an d philosophical interests revolved,82
b u t th ese n eo -P y th ag o rean s w ere also neo-Platonists, an d the
Phaedo h as decisively affected their whole conception o f the
m aster. S im ilar d o u b ts, except th a t the co rru p tin g influence is
h ere A risto tle ra th e r th a n P lato, a tta c h to the rep o rt th at ‘He
believed m u sic . . . could m ake a great co n tribution to health.
H e m a d e a very serious use o f this form of purification (th at was
his exp ressio n for m usical m ed icin e).’83 It is likely enough that
80 See A p p en d ix 4.
81 D .L . 8.33.
82 P o rp h . VP 12,45, an d passages cited in Boyancé, 8(i n. 3; Iam bi. Γ Ρ 3 Ι . (iii. 70. 74.
228, ‘H ip p a rc h u s’ in Iam b i. VP 7 5 -8 . Also unacceptable as ev idence are the purified
a n d u n p u rified souls o f A lex an d er Polyhistor in D .L . 8.31 - the next sentence betrays
P lato n ic influence (p. 217 n. 54). P ythagoras dem an d s purity ‘both ol'body a n d soul' in
D iod. 10.9.6.
83 Iam b i. VP 110, 68; cf. A ristoxenus, fr. 26 VVehrli. Porphyry speaks of musical
th e ra p y , b u t w ith o u t th e term kalharsis, VP 30, 3 2 -3 . For A ristotle's inlhioncc on
A ristoxenus see M . Pohlenz, Die Griechische Tragödie2, G öttingen, 1954. ii. 195 f. As lor
th e b io g rap h ers, neo-P latonism from the tim e o f P lutarch knew, though il could not
en tirely acce p t, th e A risto telian theory o (kalharsis: I. B yw ater. Aristotle on Ihe A n of
Poetry, O x fo rd , 1909, 157—9, C roissant, 113—34.
298 Miasma
P y th a g o ra s used th e m ystical pow er o f harm ony to cure both
b o d y a n d m in d ,84 b u t if he really an ticip ated A ristotle in seeing
this process as a purification, it is surprising th a t Plato said
n o th in g o f a th eo ry th a t w ould have been relevant to his rejec­
tion o f th e a r ts .85 Such a p urification w ould anyw ay, as de­
scrib e d , h av e no eschatological significance. M ore tem pting,
b ec au se it helps to reconcile P y th a g o ras’ religious a n d scientific
co n cern s, is the theory th a t he saw intellectual activity as a form
o f p u rifica tio n from the ties o f body, a m ental catharsis directly
beneficial to the destin y o f the soul.86 T h e idea o f philosophy as
d e a th to this w orld, m em orably expounded by P lato in the
Phaedo, w ould th en be P ythagorean. I f this is correct, P ythago­
re a n p u rifica tio n w as a high m etaphysical thing. T h a t is not
im possible; b u t it is not clear th a t Em pedocles, for instance,
a n o th e r scien tist w ho was also a m ystic, looked on his scientific
ac tiv ities in this lig h t.87
S ecure evidence th a t P ythagoras saw purification as the way
o f salv atio n is o f course provided if Em pedocles is enlisted in the
school. T h is, as w e noted, is a very uncertain issue. It is,
c e rta in ly , p lau sib le th a t freedom from the pollution o f anim al
sacrifice w as m u ch talked o f by P ythagoreans, as by E m pedo­
cles, as a necessary condition o f prosperous reincarnation. O ne
tra d itio n even claim ed th a t the m aster sh u n n ed butchers and
h u n te rs .88 T h e d o ctrin es th a t m an is bad, pleasure a n evil, an d
‘W e a re here to be p u n ish ed ’ are attested as P ythagorean,
th o u g h n o t in early sources, b u t there is no evidence th a t a
specific o rig in al pollution w as identified from w hich m an re-
84 Cf. L. D eu b n er, NeueJahrb. 43 (1919), 388—90 on the m edical p aean , Boyancé 35—8
on epodes; on the ‘scientific' adoption o f such m ethods Edelstein, A M 235 f., D odds, 80.
85 M . P ohlcnz, Gölt. Nachr. 1920, 172 f. = Kleine Schriften ii, H ildesheim , 1965, 466 f.;
contra, E. H ow ald, Hermes bl· ( 1919), 187-207; F. W ehrli, M H 8 (1951), 3 6 -6 2 esp. 56 (Γ.
= Theoria und Humanitas, Z ü rich , 1972, 177-206.
86 Λ. D öring, Archiv J . Geschichte der Philosophie 5 (1892), 505; cf. recently G uthrie,
H G P i, 199, 204 f. C riticism in B urkert, L S 211 —13; an d cf. H . B. G ottschalk, Heraclides
oj Pontus, O x fo rd , 1980, 23—33. T h is, n atu rally , is the neo-Platonist interpretation: see
e.g. P o rp h . VP 46. F or a m odern parallel cf. E dm und Gosse, Father and Son, ed.
J . H e p b u rn , O x fo rd , 1974, 7: P hilip G osse, FRS and P lym outh B rother, valued
scientific stu d y p artly because it ‘kept the stu d e n t “ o ut of the w orld” ’.
87 E. H ussey, how ever, The Presocratics L ondon, 1972, 71 points to B 110.
88 But on the p roblem o f P ythagoras’ ow n a ttitu d e to m eat-eating see Burkert, L S
180—2, G u th rie , H G P i, 187—95; against original full vegetarianism Nilsson, H T R 28
(1935), 206 = Op. Sei. ii, 657. Butchers: E udoxus in Porph. VP 7.
Purity and Salvation 299
<|uired p u rific a tio n .89 O n e p o in t th a t seems clear am id the
gen eral u n c e rta in ty is th a t P y th ag o ras offered his followers no
sh o rt cu ts th ro u g h rites o f lu stratio n . Special kinds o f physical
purificatio n a re now here m en tio n ed , and all the sources agree
Ih a t it w as ad h eren ce to a w hole w ay of life th a t m ade a
I’y th ag o re an . T h e w atchw ord o f th a t w ay o f life is p erhaps as
likely to h av e been ‘p iety ’ o r ‘h arm o n y ’ as ‘p u rity ’.90
W ith E m pedocles, O rp h ism , an d the gold tab lets the idea of
deliverance th ro u g h purification becomes inescapable. Em pe­
docles’ g reat religious poem w as entitled Katharmoi·, purifications
w ere th e m ain concern o f the O rp h e u s-in itia to r m entioned by
Plato, a n d th e soul assures P ersep h o n e through the gold plates
th a t it h as en tered th e u n d erw o rld in p u rity .91 In each case
specific pollutions are envisaged th at require cleansing. Em pe­
docles h im self is a daimön, b an ish ed from O lym pus for ‘staining
his d e a r lim bs w ith blo o d sh ed ’; th e h u m an race as a w hole has
lallen from a v eg etarian golden age, w hen ‘this w as the greatest
p o llu tio n am o n g m en, to w rench o u t (an a n im al’s life) an d eat
its stro n g lim b s’, a n d now defiles itself daily w ith an im al sac­
rifices th a t are, becau se o f m etem psychosis, acts o f m u rd er and
ca n n ib a lism .92 O rp h ic poetry too p erh ap s m ade vegetarianism
the d istin c tiv e m ark of the m ythical golden age.93 It broke
fu rth e r w ith H esiodic trad itio n in offering an explicit account o f
how th e h u m a n race cam e into being —if we accept, as we surely
now m u st, th a t th e m yth o f th e T ita n s ’ crim e a n d the b irth of
m a n from th ese ‘u n rig h teo u s an cesto rs’ is no hellenistic in-
89 Ia m b i. VP 82,85 (am o n g th e symbola).
90 R ites: only D .L . 8.33. Ion m akes a better lot in the P ythagorean afterlife a
con seq u en ce o f ‘m an lin ess an d aidös' (Ir. 30 W est), A ristophon o f ‘piety’ (D .L. 8.38).
P lato does, it is tru e, use the concept o f katharsis very widely in a context th at reeks of
so u th e rn Italy , Sop/ι. 2‘2 6 b -2 3 1 e, cf. W’ehrli, op. cit. But in the Pythagorean table of
op p o sed q u alities, A rist. Metaph. 986a24 If. (58 D iels/K ran z B 5), p u re/im p u re does not
a p p e a r.
91 Pl. Resp. 364e; gold tab lets A 1 -3 in Z u n tz ’s edition (Persephone, O xford, 1971, 277
If.). F o r gold tab lets p u b lish ed after Z u n tz see SE G xxvi 1139, xxvii 226 Aw.
92 B 115, 128, 1 3 6 -7 .
93 PI. Leg. 782c (not a strict proof; G u th rie, OGR 198 is too confident). T h e progres­
sive acco u n t (O F 292) o f native m a n ’s cannibalism g radually m itigated presum ably
belongs to a q u ite se p a ra te trad itio n w ith no place for a golden age (pace B. Gatz.
Weltalter, goldene Zeit und sinnverwandte Vorstellungen, Spudasmata 16, H ildesheim , 1967,
167). D icaea rch u s an d T h e o p h ra stu s told o f a vegetarian golden age, cf. G atz, op. cit.,
i .·><; f.
300 Miasma
v e n tio n .94 T h e c h a ra c te r o f the p rim al crim e, w hich was a n act
o f ca n n ib a lism , suggested the cure: rejection of th a t further
c a n n ib a lism w hich every anim al sacrifice entailed.95 T h e exist­
ence o f m a n k in d h a d h ith erto been a prim e fact o f experience
th a t could n o t be im agined otherw ise, an d so required no
serious ex p lan a tio n ; the O rp h ic anthropogony, by presenting
m a n as a n im m o rta l lapsed th ro u g h crim e, offered at the sam e
tim e th e possibility o f re d em p tio n .96 T h e gold tablets, w hich
sh o u ld p ro b a b ly be classed as O rp h ic texts,97 also testify to the
in itia te ’s hopes o f achieving divinity through expiation o f guilt.
‘I h av e p a id the p en a lty for u n ju st d eed s’, the initiate declares
in ho p e a n d confidence to the q u een of the underw orld.98
In these contexts, therefore, purification has a new signi­
ficance. W h e re E leusinian purification was sim ply the norm al
p re p a ra tio n for a solem n ritual, D ionysiae perh ap s a liberation
from m en tal stra in or d istu rb an ce , P ythagorean possibly p a rt of
a m ore g en eral concern for harm ony, the purifications of E m pe­
docles a n d O rp h e u s h ad a specific eschatological m eaning,
b ec au se th ey released the soul from a b u rd e n o f personal or
in h e rite d g u ilt.99 Legal notions w ere a n atu ral vehicle for con­
ce p tio n s o f this kind. E m pedocles is in exile for m u rd er, while
P e rse p h o n e ‘accepts com p en satio n ’ from m ortals for her
a n c ie n t g rief.100
E n c a se m e n t in flesh was in itself a pun ish m en t, b u t during
th is im p riso n m e n t fu rth er p urifications w ere necessary in order
94 Cl'. D o d d s, 155 f., G raf, 6 6 -7 8 , B urkcrt, G R 442 f. O n the recent transform ation of
o u r know ledge o f O rp h ism see B urkert, ‘N eue F unde zur O rp h ik ’, Informationen zum
altsprachlichen Unterricht 2 ( 1980), 2 7 -4 2 .
95 Cf. M . D etien n e, Dionysos mis à mort, Paris, 1977, C h. 4, w ho ingeniously interprets
th e T itan s’ crim e as a deliberately negative ‘origin of sacrifice’ m yth, a m odel of
c u ltu ra l regression.
96 Cf. D. S a b b a tu c c i, Saggio sul misticismo greco, Rom e, 1965, 116-26; Nilsson, H T R 28
( 1935), 224 f. = Op. Sei. ii, 677.
97 W . B u rk ert, ‘Le lam in ette auree: d a O rfeo a L am p o n e’, in Orfismo, 8 1 -1 0 4 , esp. 87
I.. 95; cf. S. G . C ole, G R IiS 21 (1980), 2 2 3 -3 8 . N ote too M . S chm idt, in Orfismo, 112-17,
o n a so u th Ita lia n a m p h o ra o f 330/20 show ing O rp h eu s facing a d ead m an w ho holds a
scroll (c o m p a ra b le to a gold tablet?); also the arg u m en t o f Boyancé, 78, th at epic verses
like those o f th e gold tab lets can scarcely have been attrib u te d to an y o th er poet than
O rp h e u s.
98 [ ab let Λ 3, p. 305 Z untz.
99 F or crim e an d exp iatio n in O rp h ism cf. PI. Cra. 400c, Resp. 364b—e, Arist. fr. 60,
O F 232, O rp h . Hymn 3 7 .7 -8 ; in P ythagoreanism ?, p. 298 above. Λ sim ilar atm osphere
in F u r. fr. 912.
100 P ind. fr. 133.
Purity and Salvation 301
lo escap e from ‘th e d ire cycle o f d eep g rie f (in c a rn a tio n ).101 T he
m ost im p o rta n t m eth o d for E m pedocles was vegetarianism ,
since, as we h av e seen, to ea t an im al flesh was cannibalism .
V ery little else, u n fo rtu n ately , is know n of his way o f salvation.
I Ie urged his followers to sh u n beans, a n d the laurel, possibly
because they w ere staging p o in ts for h u m an so u ls.102 H e may
have ad v o c ated a n d p ractised a rule of life as strict as the
P y th ag o rean , a n d ad m in istered or u ndergone rituals o f p u ri­
fication, b u t o f all this no evidence surv ives. O n e fragm ent
speaks o f d ra w in g w ater from five springs, obviously for
c a th a rtic p u rp o s e s,103 but, th o u g h this m ay have been a recipe,
it is eq u ally possible th a t the context was ra th e r ‘this is a
p ollu tio n th a t even the m ost elab o rate purification cannot
c u re .’ O n e C h ristia n source states, a little am biguously, th at he
urg ed his followers to show ‘self-m astery over intercourse with
w o m en ’, on th e g ro u n d s th a t it was a division ra th e r th an a
unio n a n d fu rth e red the dead ly work o f strife.104 It is, o f course,
p lau sib le th a t th e ascetic m ovem ents should have enjoined
sex u al re s tra in t o f som e kind, p articularly as periodic ab sti­
n en ce w as a p a rt o f priestly life. T h ere are hints, suggestiv e
th o u g h n o t conclusive, th at O rp h ism in p artic u la r was hostile
to sex u ality , or a t least to th e influence o f the female upon the
m ale; O rp h e u s w as to rn to pieces by the wom en o f T h race, and
it is p ro b a b ly because o f his professions o f ch astity that
E u rip id e s ’ H ip p o ly tu s is accused by his father of a hypocritical
e n ta n g le m e n t in O rp h ic rites.105 N othing is said, however, of
a b stin e n c e in the p o rtra it o f th e in itiate in the Cretans, a n d it is
h a rd to believe th a t a call to full sexual renunciation, if E m pe­
docles h a d m ad e one, should have provoked so little com m ent
in an tiq u ity . M o d eratio n an d self-control were pro b ab ly all that
he p re a c h e d .106
101 F ablet A 1.5 Z u n tz.
102 B 1 4 0 - 1. Souls a n d laurel: fr. 127. Souls and beans: hexam eters in schol. T . Horn.
//. 13.589 = T hesleli, 159 fr. 6.
103 B 143.
104 H ip p o l. Haer. 7 .2 9 -3 0 , q u o tin g B 1 10, I 15; c f. Dodds, 155. Sexual differentiation
is a p ro d u c t o f th e m ore general differentiation w orked by strife: the god o f B 29. 13 \ is
sexless. T h is d o ctrin e, incidentally, seem s to form a bridge betw een Km pedocles' two
poem s.
105 9 5 2 -4 . Cf. M . D etienne, in Orfismo, 7 0 -9 .
106 Cretans, fr. 79 A u stin . R estrain t, not abstinence, is all th at εγκράτεια (Hippolytus*
w ord ) in sexual m a tte rs entails, cf. LSJ s.v. εγκρατής, III.
 Miasma
302
T h e p rim e m ode o f O rp h ic salvation was the O rp h ic life. Its
a tte s te d co m p o n en ts are vegetarianism , abstinence from beans
a n d eggs, a n d b u rial in lin en ;108 we should perhaps ad d avoid­
an c e o f n a tu ra l p o llu tio n s109 an d , for the reasons ju s t noted,
som e d egree o f sexual renunciation. O rphism , how ever, in­
volved ritu a l as w ell as a w ay o f life. Ecstatic D ionysiae in itia­
tion, in p a rtic u la r, seems to have been adopted a n d given an
eschatological m ean in g th a t w as originally alien to it.110 I t was
chosen p a rtly , p erh ap s , because it h ad always been a ‘purifica­
tio n ’, th o u g h in a different sense, b u t m ore im portantly because
it w as a socially a n d psychologically ab n o rm al form of religious
ac tio n , w ell suited to serve as the vehicle of a new m essage, and
th e in tro d u c tio n to a n exotic w ay o f life.
T w o p ro b lem s are posed by the descriptions th a t we have of
th ese O rp h ic ‘p u rificatio n s’. O n e is w hether they are purifica­
tions m erely in the bro ad sense - a rite o f w hatever form the aim
o f w hich w as release from evil — o r involve an actual ritual
cleansing. T h e o th er, m uch m ore im p o rtan t, is w hether there
existed a p o p u la r O rp h ism in w hich the rite w as not m erely an
in tro d u c tio n to th e O rp h ic life or an elem ent in it, b u t a substi­
tu te for it as a m eans of salvation.
In th e Cretans, the in itiate’s p u re an d vegetarian life is
p arad o x ically in a u g u ra te d by the ch aracteristic D ionysiae rite
o f ‘ea tin g raw (flesh)’. T h e rite is here an introduction to the life.
By re d u c in g th e in itiate to b estiality as a p re p ara tio n for purity,
it em p h asizes the tran sfo rm atio n th a t he is to undergo. It is
u n fo rtu n a te ly u n ce rtain w h a t reality, if any, lies behind this
im a g in a tiv e p o rtra y al set in the fabulous land o f C re te .111 Less
exotic a n d less d em an d in g ideals are suggested by Dem os-
108 Kur. Hipp. 952 I'., PI. Leg. 782c; Ο/' Γ.219, F.291, Plut. Quaest. Cone. 635e; H dt.
2.81.2. I t is o f course plausible th at fu rth er P ythagorean dietetic rules were also O rphic
(see e.g. th e late L SA 84). B urkert, GR 448 refers to a ban on wine b u t cites no source.
109 A n inference from E ur. Cretans, lr. 79 A ustin, an d D.L. 8.33 (Pythagoras).
110 Cf. B u rk ert, in Orfismo, 92. M ain texts: H d t. 2.81, E ur. Hipp. 953 f., LSS 120, and
th e new evidence from H ip p o n iu m an d O lb ia, cf. p. 287 n. 29.
111 Fr. 79 A ustin. E xhaustive discussion an d bibliography in W . F auth, R E s.v.
Zagreus, 2 2 2 6 -3 1 , 2 2 4 3 -5 7 . The trad itio n o f vegetarianism in association w ith the
C re ta n K o u rètès is found elsew here too (P orph. Abst. 2.21); it authentic, it doubtless
relates to th eir role as gods o f initiation, since alim entary rules in connection with
in itia tio n are com m onplace.
Purity and Salvation 303
(hones’ a c c o u n t o f th e rites celeb rated by Aeschines a n d his
m other:
When you became a man you read out the books for your mother, as
xlic performed the initiations, and helped her in other ways, by n ight...
purifying the initiates, wiping them off with mud and bran, and as you
i .lised them from the purification telling them to say ‘I ’ve escaped the
had, I’ve found the better’ . . . and by day leading those fine revel
hands through the streets.112
T h ese rites seem to have been ad d ressed to S abazius rath er
(han D ionysus, b u t he w as a sim ilar god o f ecstasy. W hereas in
o th e r texts it is som etim es a rg u ab le th at dancing, or some
c o m p arab le activ ity , is itself seen as a p u rificatio n ,113 th e two
things are h ere clearly d istinguished. T h ey are p ro b ab ly stages
in the sam e in itiatio n , an d it is p e rh a p s m ore n a tu ra l to see the
p u rificatio n as a p re p a ra tio n for th e d ance ra th e r th a n vice
versa. E ven so, it clearly h ad in d ep en d e n t significance and
cllicacy. A fter receiving it, th e ca n d id a te was a t once urged to
pro claim th a t he h a d ‘escaped th e b a d ’. T h e form ula is vague,
p e rh a p s d elib erately so, a n d need n o t im ply eschatological
hopes; b u t th e books th a t A eschines read o u t w ere probably
( )rp h ic (w h a t else could they h ave been?), a n d are likely to have
c o n ta in e d prom ises o f this kind. H ere, therefore, we have,
u n u su ally , clear evidence for a rite o f deliverance th at can
re aso n ab ly be seen as O rp h ic, a n d th at took the form of a
p h y sical pu rificatio n .
Few o th e r texts a re so precise. An obscure sentence in the
Laws refers to B acchic dances, in w hich the p articip an ts im itate
d ru n k e n N y m p h s, P an , Silens, a n d Satyrs, as form ing p a rt of
c e rta in p u rificatio n s an d initiations. It is not clear w h eth er an
a c tu a l p u rificatio n preceded the dances, nor w h eth er the whole
rite w as in te n d e d to benefit the can d id ate in this life, the next
life, o r in b o th .114 A ccording to A d eim antus in the Republic,
w a n d e rin g priests w ent to the doors o f the rich, an d persuaded
th em to ex p iate th e ir ow n crim es or those o f th eir ancestors by
112 1 8 .2 5 9 -6 0 , cf. 19.199, 249, 281. See for details the com m entary on Dem. 18ο(Ή .
W ankel, H eid elb erg , 1976, ii, 1132 fT. (w ith his ad d en d u m Z P E 34 (1979), 79 f. on LSA
23.11). Iam b i. M yst. 3.10 locates S ab aziu s’ efficacy in ‘Bacchic dances, sp iritual
pu rificatio n s a n d release from ancient gu ilt’.
113 See p. 288 n . 38; sam e d o u b t in PI. Leg. 815c.
114 815c.
 Purity and Salvation 305
304 Miasma
I’y lh a g o ra s,119 th ere w ere certain ly altars w here no living
a n inexpensive, playful ritual, conducted in accordance with
Ix ings m ig h t be offered. B ut it is rig h t to em phasize th a t the
books o f M u sae u s an d O rp h e u s, a n d so assure them selves o f a
NV«»thesis o f these elem ents in to a life o f p erp etu al religious
blessed lot in the life to com e.115 O f the contents o f the playful
concern is so m eth in g alm ost wholly new. D espite its p artial
ritu a l no details are given; purifications a n d B acchic dancing
preced en ts, th e O rp h ic a n d E m p edoclean revaluation o f sac-
p ro b a b ly p lay ed a p art, a n d p erh ap s too a n im itation of the
11lice is p a rtic u la rly startlin g . C an n ib alism was, for G reeks, one
so u l’s p o sth u m o u s jo u rn e y .116 P lato gives no hint, any m ore
ni those ex trem e pollutions, often im agined, though never ex­
th a n D e m o sth en es in the passage cited earlier, th a t such rites
perienced - like p arricid e, or incest w ith the m o th er - which
w ere th e beginning, for the in itiate, o f a new way o f life. T hese en
served to define by co n trast th e p ro p e r h u m an co n d itio n .120
passant re m a rk s by co n tem p tu o u s w itnesses are, o f course, u n re­
hm p ed o cles a n d O rp h e u s now declared it to be in h eren t ii.
liab le evidence. A n allusion in the Hippolytus shows th a t the link
tra d itio n a l c u lt’s m ost sacred act. Psychological factors have
o f B acchic dances, O rp h ic books, an d vegetarianism was familiar
been invoked in ex p lan atio n , the arch aic G reek’s growing
in fifth -cen tu ry A th e n s.117 It is none the less plausible th at
Im rden o f an x iety a n d g u ilt.121 B ut it is h ard to know how
p u rifiers d id exist w ho w ould offer their clients salvation for the
im p o rta n t w ere su ch feelings in the tem p eram en t of, for in­
cost o f a ritu a l, w ith o u t insisting on the uncom fortable require­
stance, E m pedocles, th e divine m an w ho controlled the w eather
m en ts o f a n O rp h ic life.118 I t is interesting th a t P lato speaks of
.iiid w alked a m o n g his fellows as ‘a deathless god, no longer a
release, n o t from m etem psychotic o r T itan ic guilt, b u t from the
m o rta l’. P erh a p s em phasis should ra th e r be placed on the ways
crim es o f a n individual o r his ancestors. T h u s w ere exotic
in w hich these m ovem ents rejected o r reversed m any o f Greek
jrietaphysical speculations tailored to suit the conceptions of
.society’s m ost cherished values. T h e m otivation for such rejec­
co n v e n tio n a l G reek m orality. I f O rp h ic an d P ythagorean ideas
tion, how ever, is obscure; it could scarcely becom e plain w ith­
w ere in d eed ‘a d ro p o f alien blood in the veins o f the G reeks’, we
o u t a d etailed know ledge, th a t will pro b ab ly never be achieved,
see h e re one w ay in w hich th a t d ro p could be assim ilated into
o f th e social en v iro n m en t in w hich the m ovem ents h ad their
th e b lo o d stream o f G reek c u ltu re w ithout changing its funda­ ■ * 122
origin.
m e n ta l ch a ra c te r.
T w o claim s th a t are m ost relev an t to o ur them e have been
N o n e th e less, in its prim e, the O rp h ic /P y th ag o re an move­
m ad e a b o u t G reek asceticism , a n d th e age in w hich it em erged.
m e n t w as th e only u n q u estio n ab le novelty in the history of
( )n e is th a t G reek religion w as now on the road to becom ing,
a rc h a ic G reek religion. F or m ost o f its elem ents parallels can be
like H in d u ism o r Z o ro astrian ism , a religion o f lu strations and
fo u n d elsew here in G reek culture. T h e E leusinian an d other
cerem o n ial p u rity . T h e o th er, closely connected, is th at purity
m ysteries ta u g h t the need to take th o u g h t for the afterlife; m ost
ra th e r th a n ju s tic e was the m eans to salv atio n .123 It should be
o f th e rules o f life can be illu stra te d from cult or superstition; the
re m e m b ered , how ever, th a t G reek religion h ad alw ays been a
id ea o f p u n ish m e n t for an cestral guilt, a n d o f a tain ted race
religion o f lu stratio n s; the a u th o r in whom the act o f w ashing is
w hose m em b ers w ere lured into new crim e, was deeply em ­
m ost ch a rg ed w ith m eaning is H om er. New applications the
b e d d e d in m ythology; even th o u g h it is unclear w hether
idea o f p u rity certain ly received, b u t it is not clear th a t physical
v e g e ta ria n ism was ever system atically practised before
lu stra tio n g ain ed greatly in im p o rtan ce in these m ovem ents,

115 364b—e. ‘P lay ’ also o f C o ry b an tie ritual, PI. Euthyd. 277d. In OF T 208 an
O rp h e o te le st holds a tym panon; cf. too th e G u ro b papyrus, OF 31.
116 Pl. Phd. 108a, cf. G u th rie, OGR 176; on the ritual behind the gold plates see
fu rth e r G u th rie , ‘2 0 7 -1 5 , B urkert, Gnomon 46 (1974), 326 f., idem, in Orfismo, 95-100.
117 953 f.
118 B ut for rejection o f the dichotom y betw een ‘au th en tic’ a n d ‘degenerate’ O rphism
see B oyancé, 9 - 3 1 , idem , R E G 55 (1942), 2 1 7 -3 5 .
119 Cf. p. 3 0 2 n . 111. A ltars: D .L. 8.13, Paus. 1.26.5, cf. T huc. 1.126.6.
120 M . D etien n e, Dionysos mis à mort, Paris, 1977, 140-5.
121 D o d d s, 151 f., cf. Boyancé, R E A 40 (1938), 169.
122 S p ecu latio n by F. M . G o rn fo rd ,C Q 16 (1922), 140. Burkert, G R 416 em phasizes
individ u alism . F o r one factor see p. 143 above.
123 R o h d e, 302; D odds, 154.
306 Miasma
except in th e cults th a t su b stitu ted in stan tan eo u s purification
for a w ay o f life; a n d there is certainly no sign th a t p urity was
b ecom ing a d o m in a n t idiom to w hich all o th er forms o f evalua­
tion w ere su b o rd in a te d . As for ju stice, it was, as Plato knew,
alw ays possible to in te rp re t even orthodox G reek religion as if
th e gods w ere sw ayed by ritu a l m ore th an righteousness. A
fo u rth -c e n tu ry o ra to r could tu rn to O rp h eu s, ra th e r than
H esiod, for th e id ea th a t ju stic e, seated by the throne o f Zeus,
keeps w a tc h over the offences o f m ankind; a n d it was O rpheus,
a c co rd in g to A risto p h an es, w ho ‘ta u g h t us rites an d to refrain
from m u rd e r’. 124 O rp h ic poem s are likely to have contained the
sam e b len d o f m oral a n d cerem onial precepts as did the teach­
ing o f P y th ag o ras.
T h e p re o ccu p a tio n o f these m ovem ents w ith eschatology
w as, o f course, u n ch a rac te ristic o f G reek culture. Purification
could p e r h ap s be ann exed as a m eans of im proving one’s condi­
tion in m ore im m ed iate term s. M an y cults seem to have offered
it. T h e su p p o sed b ap tism in th e rites o fC o ty to m ay be based on
a m is u n d e rsta n d in g ,125 b u t a sacred law from the Peiraeus, for
in stan ce , re stric ts the rig h t o f w om en to ‘perform purifications’
124 (D em .) 25.11, A r. Ran. 1032, on w hich see G raf, 34 fi. Cf. (w ith Boyancé, 24) the
d ife w arn in g s o f ‘those w ho have taken a n interest in such things in connection with
in itia tio n s’, ‘p riests o f o ld’, a b o u t the fate o f the kin-killer, in PI. Leg. 8 7 0 d -e , 8 7 2 d -
873a. T h e progressive acco u n t o f civilization too [OF 292, cf. G raf, 161 f.) upholds
ju s tic e a g a in s t force. N ilsson even w rote, H T R 28 (1935), 228 = Op. Sei. ii, 680, ‘His
(H e sio d ’s) cra v in g for ju stic e becam e the leading principle o f O rp h ism .’ O n the
re latio n o f ritu a l a n d m orality cf. D over, 264 f., Boyancé, R E G 55 (1942), 222.
125 See on C o ty to th e im p o rta n t stu d y o f S. S rebrny in Mélanges Franz Cumont,
B russels, 1936, 423—47, sum m arized by Nilsson, GGR 835 f. T h e view th a t ritual
b a th in g h a d a n im p o rta n t place in these rites depends on linking the title o f E upolis’
Baptai (fr. 68—89) w ith his victim A lcibiades’ supposed revenge; E upolis had shown
A le. ritu a lly b a p tiz e d in th e sea, the infuriated Ale. ‘b ap tized ’ the poet by drow ning. If
th is is rig h t, are w e to regard baptai as (a) a nam e com m only applied to ad herents of
K o ty to ; o r (b) o n e invented opprobriously by Eupolis? (a) would be strange. R itual
b a th in g w as a p relim in ary to m any cult cerem onies, and it is h ard to see why it should
h av e received this special prom inence in th e rites o f K otyto, whose m ain co n ten t was
e c static d a n c in g by tran sv estite choirs (see S rebrny, loc. cit.). A s for (b), w hat is
c o n te m p tib le a b o u t ritu a l bathing? It w as p a rt o f the E leusinian cult. O n e should note
fu rth e r th a t baptai m ean s ‘d ip p e rs’, not ‘people initiated by d ip p in g ’. A q u ite different
in te rp re ta tio n w as proposed by A. M eineke, Historia Critica Comicorum Graecorum,
B erlin, 1839, 123 (follow ing earlier critics). E upolis’ play w as an attack on effeminacy,
a n d he n o ted th a t dy ein g the h a ir (baptesthai: M en. fr. 303.4, N icolaus C om icus 1.33)
w as a c h a ra c te ristic o f luxurious, effem inate youth. Even on this view, th e story of
A lcib iad es’ revenge, w ith a p u n on baptö, is not incom prehensible.
Purity and Salvation 307
in th e T h e sm o p h o re io n .126 T h e H esiodic ‘divine m a n ’, who
.«•cms to o b serve rules o f p u rity w ith o u t th o u g h t for an y future
Iilr, h as a fo u rth -c e n tu ry successor in the deisidaimön o r S uper­
c ilio u s M a n o f T h e o p h ra s tu s .127 T h e dan g er o f pollution is
never far from his th o u g h t. F irst th in g in the m orning he washes
Iiis h a n d s (p e rh a p s from th ree sp rin g s),128 an d sprinkles his
body w ith lu stra l w ater; for th e rest o f the day he protects
liim self by chew ing laurel. H e co n stan tly has his hom e purified,
su p p o sin g th a t H e cate has been conjured ag ain st it. Like a
p riest, b u t unlike a good citizen, he declines all co n tact with
b irth , d e a th , a n d tom bs. H e seeks o u t the Orpheotelestai every
m o n th , a n d re p eated ly undergoes ab lu tio n in the sea. T h e m ere
sight o f so m e p o o r w retch eatin g th e m eals o f H ecate (?)129
req u ires a n e la b o ra te ritu al w ashing; nor is this enough, b u t a
|>riestess m u st be su m m o n ed to perform a blood purification
loo. N o th in g suggests th a t all this activity has an y m ore d istan t
o r h ig h e r aim th a n th e im m ed iate ap p easem en t o f his persistent
unease. E v en th e rites o f O rp h e u s have becom e ju s t one of
m an y devices for this pu rp o se. It is a p iq u a n t coincidence
th a t w e sh o u ld ow e this disdainful d escription o f a life th at in
c e rta in resp ects closely resem bles the P y thagorean to th at
T h eo p h rastu s w ho, in his g reat lost w ork On Piety, tran sm itted
(he O rp h ic /P y th a g o re a n ideal o f vegetarianism to neo-
P lato n ism , w h ere th e figures o f the deisidaimön a n d the godlike
m a n w ere o nce ag ain to co n verge.130
126 L S C G 36.5.
127 T h e re is a b u n d a n t com m entary; see, besides th e editions o f H . S teinm etz and
R. G . U ssh er, H . B olkestein, Theophrasts Charakter der Deisidaimonia, R G W 21.2, Gies­
sen, 1929; N ilsson, GG R 796 Γ. (w ith the im p o rtan t observation th at this deisidaimön
belongs n o t to the low er classes, b u t to the bourgeois world fam iliar from New
C o m ed y ); co n v in cin g treatm en t o f som e tex tu al problem s, K . Borthw ick, Eranos 64
(1966), 1 0 6 -1 9 .
128 See B orthw ick, op. cit.
129 See B orthw ick, op. cit.
Iî0 M a rin u s ’ life o f P ro d u s 18, p. 160.33 Boisson. ( = O F 'Γ.239) νϋκιωρ τε κα ίμεθ'
ημέραν άποτροπα ΐς καί περιρραντηρίοις καί τοίς άλλοις καθαρμοίς χρώμενος, ο τί μ ίν Ορ-
φικοίς. ότε όέ ΧαλόαίκοΙζ.

11
SOM E SCENES FROM TRAGEDY1
F o r th e h isto rian o f religious beliefs tragedy provides, as was
n o ted in th e in tro d u ctio n , elusive evidence, in one sense, its
v alu e is u n iq u e, since, read arig h t, it offers insight into the
m in d s a n d feelings, a t a level o f intense seriousness, o f actual
A th e n ia n s, the trag ic poets them selves. T h e m ind of Aeschylus
is a m u c h solider historical reality th a n any synthetic
hy p o th esis a b o u t the A th en ian m ind; an d fundam ental beliefs
th a t, for various reasons, lie well below the surface of everyday
life m ay find expression th ro u g h literature. T h e concerns o f the
tra g e d ia n s are som etim es consigned to the m elancholy category
o f religious philosophy,2 b u t th a t is justified only in so far as
every believer is also a philosopher o f religion; there is no reason
to th in k th a t the o rd in ary A th e n ia n ’s relations w ith the gods
w ere m erely m agical, an d th a t the ju stice o f Zeus was a problem
left to theologians. O n t h t o th er h an d , w hen tragedy is asked to
p ro v id e historical inform ation on low er levels th an this, its
an sw ers becom e am biguous a n d h a rd to in terp ret, largely be­
cau se o f its settin g in the m ythical past. Several instances o f this
lack o f realism have already been encountered. T h e situation of
O e d ip u s, th e incestuous p arricid e, belongs to the world of
n ig h tm a re , no t everyday experience. T hough O re stes’ dilem m a
h a d o nce b een a real one, the em ergence of hom icide courts had
re m o v ed it from the level o f literal plausibility long before the
first trag e d y a b o u t him was w ritten. Even an O edipus left
d e s titu te by his sons could, in fifth-century A thens, have sought
re d ress from a m ag istrate. T h e plague a t the s ta rt of the Oedipus
1 CX V ickers, 145-56. G. R ichard, ‘L ’im p u reté contagieuse et la magie dans la
tra g é d ie g re c q u e ’, R E A 37 (1935), 301—21, is unhelplul.
2 N o m en tio n e.g. in E. R. D odds, ‘ I he Religion oi the O rd in a ry M an in Classical
G re e c e ’, in Progress, 140-55. But is it ‘religious philosophy’ w hen an Eskimo asks: ‘Why
m u st th e re be snow a n d storm s an d bad w eather lor hunting? W hy m ust the children ol
m y n e ig h b o u r sit s h iv e rin g . .. hungry? W hy m ust my old sister suffer pain a t the end of
h e r days? She has done no w rong th at we can see’ (cited in P. R adin, Primitive Religion,
L o n d o n , 1938, 54)?
Some scenesfrom Tragedy 309
I 'vi annus leads to a h u n t for the p o llu te r o f the city, b u t historical
parallels for su ch a m a n -h u n t a re h a rd to find. Even the family
t in.se o f th e trag e d ian s is a m e ta p h o r as m uch as a dogm a with
lived p ra c tic a l im plications.
T h is c h a p te r will try to m ake use of tragedy in a rath er
restricted w ay. O n the tragic stage we see the action, and
in teractio n , o f persons w ho a re them selves polluted o r are
< oiilro n ted by p o llu tio n in oth ers. N o o th er source offers evi­
dence o f th e sam e im m ediacy. Som e scenes o f this kind will be
•iiirveyed here, ra th e r unsystem atically, and w ith o u t su b o rd i­
n ation to a n y g en eral arg u m e n t. B ut th e assum ption th at pollu­
tion belief is o ne o f th e b izarre a n d im p en etrab le attitu d e s that
le n d e r trag e d y ‘d esp erately a lie n ’ will p erhaps be b ro u g h t into
d o u b t. A t least in th e trag e d ian s’ p resentation, it is the flexibil­
ity o f th e th in g a n d not its do g m atic rigidity th at causes
.urprise. T h is m u st be p artly a m a tte r of art, o f th e successful
a d a p ta tio n o f response to c h a ra c te r a n d situation; b u t this
a d a p ta tio n w ould n o t be possible if som e flexibility w ere not
in h e re n t in th e beliefitself.
W e m ay begin w ith those situ atio n s w here the infectiousness
o f p o llu tio n is e ith e r explicitly d en ied o r nobly disreg ard ed in
favour o f a h ig h er ideal. O restes w arn s Pylades o f the dangers in
a c tin g as his guide.
O r.: I t ’s d isg u stin g to to uch a sick m an.
Pyl. : N o t for m e to touch you.
O r.: B u t you m ig h t be infected by m y m adness.
Pyl.: So be it.
O r.: Y ou w o n ’t be afraid?
Pyl.: No. F e a r ru in s frien d sh ip .3
I'ylades does n o t den y b u t disregards the dangers in the act of
11iendship; we m ight com pare the action o f ‘those w ho laid some
Claim to v irtu e ’ in n u rsin g th eir friends du rin g th e A th en ian
p lag u e.4 T h ese u s reassures the sm itten H eracles in sim ilar
term s: ‘I ’m h a p p y to share suffering w ith you, ju s t as I once
sh ared p ro sp e rity .’ B ut he goes on to m ake a bolder claim.
H eracles asks ‘W h y have you uncovered my head before the
su n ?’ ‘Y ou a re a m ortal, a n d can n o t pollute the g ods’, T heseus
replies, a n d ad d s th a t from friend to friend no pollution can
3 E u r. Or. 7 9 2 -4 .
4 T h u c . 2.51.5.
310
Miasma
pass. I t is often a n d plausibly supposed th a t such form ulations
a re o f so p h istic o rigin.5 A sim ilar scene in Sophocles does not
en d in a sim ilar affirm ation: O ed ip u s is ab o u t to em brace
T h ese u s, th e n d raw s back. ‘B u t w h a t am I saying? How could I,
w re tch th a t I am , to u ch a m an in w hom no stain o f evil dwells?’6
T h is w as an o p p o rtu n ity for Sophocles’ Theseus m agnanim ously
to defy o r d en y th e reality o f pollution, b u t he did not take it.
T h o u g h tre a tin g O e d ip u s w ith all possible generosity, he kept
his d ista n c e . As w e saw earlier, Sophocles p u t the assertion ‘No
m o rta l can p o llu te the g ods’ into the m outh o f C reon at a
p a rtic u la rly u n h a p p y m om ent, w hen C reon h a d ju s t uttered
th e fearful blasp h em y : ‘N ot even if eagles carry scraps o f the
co rp se to Z e u s’ th ro n e will I consent to bury it for fear of
miasma.'’ T h e sophistic claim ap p e ars here as the last shred of
self-defence o f a d esp e rate m an, a n d stands condem ned.7 In the
Oedipus Tyrannus, O e d ip u s’ friends rem ind him o f the danger
th a t w ith his uncovered h ead he poses to the su n .8
I t m ay seem th a t we are confronted here w ith a sim ple
c o n tra s t b etw een the conventional piety o f Sophocles and
E u rip id e a n en lig h ten m en t. T h e re does seem to be a sense in
w hich, in E u rip id es, p ollution has lost its sting. His famous
in te rn a liz a tio n o f the E rinyes, by w hich they are reduced to
O re s te s ’ b ad conscience,9 w ould, if carried th ro u g h consist­
en tly , rem ove the need for o u tsid ers to g u ard against the th reat
o f e x te rn a l pollution from him . B ut we cannot sim ply detach
beliefs a b o u t pollution from th e w hole m oral fabric o f the plays.
F o r th e E u rip id e a n T heseus, it is m orally inconceivable th a t the
u n iv erse sh o u ld , th ro u g h pollution, set obstacles in the way of
frie n d sh ip , b u t he is no t concerned to deny the need for
p u rific a tio n ;10 C reon is defending a w anton violation o f divine
law s, a n d a t its logical extrem e his arg u m en t w ould m ean th at
even m u rd e re rs could en ter the tem ples at will. E veryw here in
G reek trag e d y , propositions th a t in them selves m ight deserve
serious co n sid eratio n are liable to grotesque an d unscrupulous
5 E u r. I ll· 1214—34. Cf. p. 145 above.
6 S oph. O C 1 132-4.
7 S oph. Ant. 1043 f., cf. above, p. 33.
* 1 4 2 4 -8 .
9 Kur. Or. 396. In his treatm en t o f the .Alcmaeon legend, however, he seems to have
exp lo ited real pollution, cf. A ppendix 7 s.v. A lcm aeon.
Some scenesfrom Tragedy 311
d isto rtio n : good arg u m e n ts are. n o t reserved for good m en .11
A nd, th o u g h S ophocles’ T h eseu s m ay not go to the length o f
em b ra cin g O e d ip u s, his b eh a v io u r th ro u g h o u t the play
proclaim s th a t, in th e m ag n an im o u s m an, h u m a n sym pathy
i lissolves th e fear o f pollution. As it tu rn s o u t, his nobility is also
I »rudential, since th e polluted O e d ip u s proves ‘a benefit to those
who received him , a n d a b an e to those w ho drove him o u t’. 12 In
p ractice, therefore, ‘th e re is no po llu tio n from friend to frien d .’
E ven in E u rip id es, by co n tra st, a superficially h u m an itarian
tlisreg ard for p o llu tio n m ay assu m e a dim m oral colour. In the
Orestes, H elen from th e h eig h t o f h er own good fortune com m is­
erates w ith h er siste r’s m u rd erers, explaining: ‘I ’m not polluted
by sp eak in g to you; I lay all th e blam e on P h o eb u s.’13 A t first
sight th a t m ig h t seem a h u m an e a n d rational insistence th at
p o llu tio n atta c h e s to the tru e g uilty party, a n d not his in­
v o lu n ta ry ag en t. B ut is this th e H elen we know? E lectra has
I>i tte r cau se to say o f h er in th e sam e scene: ‘S he’s still th e sam e
w o m a n .’14 E very o th e r p a rtic ip a n t in the play is clear th at
laying th e b lam e on Apollo does not vin d icate O restes.
T ynd areu s will n o t even ad d ress th e ‘m other-slaying snake’,
a n d is am a z e d th a t M en elau s sh ould do so.15 T o u n d erstan d
I le le n ’s a ttitu d e h ere it is p erh ap s legitim ate to refer to a famous
p assag e in th e Troades w here she justifies h er ow n crim e by the
p ow er o f A p h ro d ite .16 F or no serious G reek th in k er did divine
involvem ent ever exclude h u m an responsibility, an d only Helen
cou ld p re te n d th a t it did. H ow delightful life w ould be for th at
lad y if ted io u s people m ade less fuss ab o u t g u ilt a n d crime!
H e le n ’s ‘I ’m n o t p o lluted by speaking to you’ is sim ply an
ex p ressio n o f h e r glib m oral lax ity .17
" C . M . B ow ra, Sophoclean Tragedy, O xford, 1944, 108, though speaking ofL .reon’s
‘in fatu ate d elu sio n ’ notes, ‘In o th er circum stances his arg u m en t m ight carry w eight.’
S im ilarly K . R e in h a rd t, Sophokles3, F ran k fu rt, 1947, 98; ‘W as w äre d e r W ah n , wenn er
n ic h t m it d em Schein d e r W ah reit sich um gäbe?’ (H e brings o u t the superficial
p lau sib ility in his tran slatio n , ‘D er M ensch ist zu gering, G ott zu entw eihen.’) O n
D em o sth en es’ ex p lo itatio n o f a sim ilar arg u m en t see p. 268 n. 54.
12 0 C 9 2 f.
13 75 f.
14 129.
15 4 7 9 -8 1 .
16 9 4 8 -5 0 .
17 T h e D ioscuri use a sim ilar arg u m en t a t the end of the Electra (1293-7), w ithout, it
seem s, sim ilar m oral im plications; b u t the tone o f this p a rt o f the play is h ard to catch,
1 1. V'ickers, 5 6 4 -6 .
312 Miasma
I t is in te re stin g th a t H elen here identifies pollution with
g u ilt; h a v in g tran sferred blam e to Apollo, she regards O restes
as free too from pollution. T h e scope for d eb a te ab o u t the
re la tio n o f th e tw o things was larg e;18 and in the kind o f case
en v isag ed by th e trag ed ian s there w as no a u th o rity by appeal to
w h ich th e issue could be settled. In the case o f H elen, it is her
loose a ttitu d e to guilt ra th e r th a n h er willingness to equate it
w ith p o llu tio n th a t ap p e ars reprehensible. T h e tragedians do
n o t look on th e legalistic in te rp re ta tio n w ith sym pathy. C reo n ’s
claim th a t, by placing food in the cavern in w hich he incar­
c e rates A n tig o n e, he has m ade h im self ‘p u re in respect o f this
g irl’, 19 is n o t com m ended; an d , in the Iphigeneia in Tauris, T h o a s’
h o rro r a t O re ste s ’ pollution, a n d his attem p ts to evade it by
lu stra tio n s a n d m echanical protective devices, have an ironic
effect in a king w ho upholds the in stitu tio n o f h u m an sacrifice.20
A ccordingly, th e th re a t o f pollution does not, in tragedy,
n o rm ally im pose im peratives th a t override the dem ands of
o rd in a ry h u m a n feeling. In the Supplices, for instance, though it
is to avoid p o llu tio n th a t P elasgus accepts the supplication of
th e D a n a id s, A eschylus’ p re sen tatio n o f the plight o f the help­
less girls h as been such th a t the decision seems necessary in any
term s. In th e Oresteia, too, the Erinyes th a t seek to avenge
C ly ta e m n e stra a re upholders o f rights that, anyone would adm it,
a re g en u in e, even if not absolutely valid. P ollution plays, in­
d ee d , a n im p o rta n t p a rt in th a t im aginative re-creation of the
m o ral fo u n d a tio n o f existing institutions th at seems to be
c h a ra c te ristic o f A eschylus. M etics are, in origin, helpless
foreigners such as the D an aid s w ho have been accepted into the
s ta te th ro u g h a su p p licatio n th a t is backed by the th rea t of
p o llu tio n .21 W h e n legal trial replaces self-help, as in the
Eumenides, it is th e sam e th re a t th a t forces the ju ro rs to reach
th e ir verd icts ‘w ith reverence for th eir o ath s’.22
B ecause p o llu tio n a n d g uilt can be closely associated, the
im ag ery o f p o llu tio n m ay be used to express m oral revulsion.
T h is is so m eth in g th a t is com m onplace even in societies th at do
18 A bove, p. 111.
'* 889.
20 1 174 fl'.; for the irony cf. esp. 1194.
21 E. S chlesinger, Die griechische Asy tie, diss. G iessen, 1933, 3 8 -5 2 .(o n Aesch. Sufib.).
22 A bove, p. 126.
Some scenesfrom Tragedy 313
not m u ch fear infectious religious dangers, or p ractise rituals of
lu stratio n . T a in ts a n d co n tam in atio n s are u b iq u ito u s, for in-
·.lance, in E liz a b e th a n a n d J a c o b e a n tragedy. T h e ‘dam n ed
sp o t’ o f th e sleep-w alking scene in Macbeth is an obvious case;
M id d leto n a n d R ow ley’s Changeling has a p articu larly striking
exam p le in th e figure o f D e Flores, a m an whose physical
■ ip p e a ra n c e is as repulsive as his soul corrupt. O f him it is said:
h e’s so foul
O n e w o u ld sc a rc e to u ch him w ith a sw o rd he loved
A n d m a d e a c c o u n t of; so m o st d ea d ly venem ous,
H e w o u ld go n e a r to poison a n y w eap o n
T h a t sh o u ld d ra w blood on him ; on e m u st resolve
N ev er to use th a t sw ord a g a in in fight,
In w ay o f h o n e s t m a n h o o d , th a t strik es him;
S om e riv e r m u st d ev o u r’t, ’tw ere n o t fit
T h a t a n y m a n sh o u ld find it.
The w o m an he has co rru p ted says in penitence to her father:
O h , co m e n o t n e a - m e, sir; I sh all defile you.23
T h e sense o f co n tam in atio n has here obviously passed a long
w ay b e h in d th e m etap h o rical. Its source, how ever, is m oral
ho rror.
P o llu tio n as gu ilt, the avoidance o f pollution as m oral revul­
sion are b est seen in the Hippolytus, w here p u rity in all its senses
is o f su ch im p o rtan ce. In E u rip id es’ first play on the subject,
H ip p o ly tu s resp o n d ed to P h a e d ra ’s sham eful proposals by cov­
erin g his h ea d to avoid pollution. In the surviving treatm en t, he
ru sh es o u t in to th e p u re a ir to escape it, furiously forbids the
p a n d a r/n u rs e to to uch his robes, a n d swears th a t he will wash
his ea rs free o f th e co n tam in atin g w ords in a flowing strea m .24
T h is re actio n is not confined to the sensitive H ippolytus.
T h ese u s, believing the accu satio n ag ain st his son, refuses a t first
even to ad d ress his reproaches to him directly, a n d w hen his
p assio n driv es him to neglect the precau tio n declares ‘I have
involved m y self in p o llu tio n .’25 W ith the m ild ritu al im p u rity of
licit sex u al co n tac t, this miasma evidently has little to do. H ip ­
p o ly tu s a n d T h eseu s vent th eir repu g n an ce a t the w orst offence
23 5 .2 .1 5 -2 3 , 5.3.149, in th e text ol'N . \V. B aw cutt ( The Revels Plays), L ondon, 1958.
24 6 0 1 - 2 , 606, 6 5 3 -4 .
25 946. Cf. T y n d a re u s ’ lapse, Eur. Or. 526, co n trast 481.
314 Miasma
a son could co m m it ag a in st his fa th e r by treatin g it, in w ord and
even in d eed (H ip p o ly tu s will w ash o u t his ea rs),26 as a pollu­
tion. B ut T h e se u s in sh u n n in g H ippolytus is not protecting
h im self from d a n g e r b u t expressing m oral disgust by a form of
o strac ism fam iliar to us all. (T y n d areu s for the sam e reason
d ec la res th a t he w ould not speak to the adulterous H elen.)27
T h e tre a tm e n t o f sexual offences in the nineteenth-century
novel will p ro v id e closer parallels th a n do sacred laws regulat­
in g ritu a l p u rity . T h is is C hekhov, for instance, describing an
e n c o u n te r w hile b ath in g betw een a respectable m atro n w ith her
d a u g h te r a n d a w om an living in sin: ‘She (the m atron) stood
b etw e en N a d e z h d a a n d K a ty a , as if protecting h er daughter
from th e w a te r w hich lap p ed N a d e z h d a .’28 O r here from T rol­
lope is a m o th e r advising h er son, w hose fiancée has unw ittingly
form ed a frien d sh ip w ith an ad u ltero u s w om an: ‘B ut it does
seem to m e to be so very im p o rtan t! I f she h asn ’t got your letter,
you know , it w ould be so necessary th a t you should write again,
so th a t th e —th e —the co n tam in atio n should be stopped as soon
as p o ssib le.’29 T h e re is, therefore, a reality behind Sam uel
B u tle r’s satirica l picture o f a m istress expelling a n unchaste
se rv a n t on the in stan t: ‘W hen she tho u g h t o f the fearful con­
ta m in a tio n w hich E llen’s co n tin u ed presence even for a week
w ould occasion, she could not h esitate.’29“ T hese pollutions are
no m ere figures o f speech, b u t d em an d the m ost d rastic protec­
tive m easu res from those w ho com e into co n tact w ith them .
F rie n d sh ip s m u st be broken off, servants dism issed, whole
h o u seh o ld s (w here the ta in t occurs after m arriage) dissolved; in
Tess o f the D ’Urbervilles, Angel C lare p u ts h a lf the w orld between
h im se lf a n d his wife’s co n tam in atio n . T h eseu s’ response to
H ip p o ly tu s is no less extrem e. B ut in none o f these cases, of
course, is th ere a n y hope o f banishing the pollution w ith lustral
w a te r.291’
N a tu ra lly , th e th re a t o f pollution can have an im p o rtan t
26 (>33— l·, i f . A lex an d er A etolus, fr. 3.16 Powell, A d i. T a t. 6.12.3. The girl in
( I lieoc.) 27.3 ‘w ashes o il’ an d ‘spits o u t' unw elcom e kisses, l'h at does not m ean that
kisses pollute (cf. C a tu li. 99. 7 II'.).
27 Kur. Or. 520—1.
28 The Duel, in The Oxford Chekhov, vol. v, trans. R. H ingley, O xford, 1970, 157.
2e The Belton Estate, C h . 17.
2,u The ll'/tr oj A ll Flesh, C h . 38.
29|) Cl. O liv e r G o ld sm ith 's song ‘W hen lovely w om an stoops to lolly’.
Some scenesfrom Tragedy 315
I»lace in th e lan g u ag e o f m oral exhortation. As we noted,
IVlasgus in A eschylus is co n stan tly w arned o f th e d an g er in ­
volved in refusing th e D a n a id s’ su p p licatio n ,30 an d this is the
co n sid eratio n th a t finally sw ays him and his people. A n d ro ­
m ache finds M en elau s a n d H elen m uch less sensitive to such
I >lcas.31 In m o ral d en u n ciatio n , too, th e charge o f hav in g caused
I»ollution is com m on. It ten d s to be hotly denied, o r tu rn ed back
u po n th e accuser; th e altern ativ e strategy, o f acknow ledging
pollu tio n b u t d en y in g guilt, seem s n o t to occur. T h e encounter
betw een M e d e a a n d J a s o n after th e infanticide is a striking
instan ce. J a s o n exclaim s ag ain st his wife:.‘A nd do you d are to
look a t th e su n a n d earth , w hen y o u ’ve com m itted such a
crim e?’ a n d infers: ‘T h e gods have sent your avenging dem on
(,alastor) ag a in st m e.’ (P u n ish m en t for M ed ea’s fratricide has
com e ro u n d u p o n Ja so n . T h e d an g er is only perceived,
ch a rac te ristic ally , once d isaster h as already o cc u rre d .)32 T he
ch o ru s, d esp ite th eir p artia lity for M edea, h ad earlier called on
I lie sam e su n a n d e a rth to p re v en t so foul a pollution.33 B ut for
M ed e a herself, to a d m it pollution w ould be to ad m it guilt. In a
re m a rk a b le d ialo g u e she tu rn s back all J a s o n ’s accusations
u p o n him self:
J .: C h ild re n , w h a t an evil m o th er you had
M .: C h ild re n , y o u r fa th e r’s in fatu atio n destroyed you.
J .: B u t it w a sn ’t m y h an d th a t killed them .
M .: No; it was y o u r violence ag ain st me an d y o u r new m ar­
riag e . . .
J.: T h e y will p ollute you as avenging spirits (miastores).
M .: T h e gods know who sta rte d all these troubles.·’“
J a s o n m ay invoke th e ‘Erinys o f children an d ju stic e o f blood’
a g a in st M e d e a a n d call h er ‘polluted, child-killing lioness’, b u t
he can e x tra c t no adm ission o f guilt, rem orse, or pollution.
‘W h a t god listens to a ch eat a n d p erju rer like you?’, she asks in
d efia n ce.35
30 Supp. 366, 375, 385, 415,4 7 3 , 479, 619, 654 f.
31 E u r. Andr. 2 5 8 -6 0 , 335 IT.
32 13271'., 1333.
33 1 2 5 1 -6 0 , cf. 1268.
34 1 3 6 3 -6 , 1 3 7 1 -2 .
35 1389 f.; 1406 f., cf. 1393; 1391 f. F or a sim ilar interchange see E ur. Or. 1600 IT. For
recip ro c al accu satio n s o f pollution cf. T ra g . A desp. fr. 358 N auck = Soph. fr. 187 Radt;
T h u c . 1 .1 26.2-128.2.
316 Miasma
R ecognition bÿ a hero o f his ow n pollution does, o f course,
also occur. C reo n at the end o f the Antigone, w ho h ad earlier
in sisted on his ritu a l innocence tow ards A ntigone, em phasizes
his to ta l responsibility for the suicide o f his wife. F rom ‘In
re sp ect o f this m aid en , I am p u re ’, unconvincing as it w as, he is
re d u c e d to ‘I killed you, no-one else, it was F , a n d calls on his
a tte n d a n ts to lead him aw ay as a polluted being.36 T h e killer
w ho a d m its his guilt an d renounces his throne or goes into
v o lu n ta ry exile ‘acco rd in g to the law ’ is a com m on figure in
m y th a n d m y th ic al history,37 a n d tragedy has several harrow ­
in g scenes w h ere the hero confronts an d feels w ith boundless
a n g u ish his ow n pollution. A fter the m u rd er o f his children,
H e racle s needs no outside ad m o n itio n to hide his head from the
su n ; O e d ip u s w ith his ow n h an d s strikes o u t the eyes th a t had
seen w h a t they o u g h t not to have seen, an d even in his old age,
co n v in ced th o u g h he is th a t his crim es are no fault o f his own,
c a n n o t b rin g h im self to touch the spotless T heseus.38 But, as
alw ays, th e line betw een in tern a l g uilt a n d sham e before the
w o rld c a n n o t be sharply d ra w n .39 In se p arab le from the hero’s
p e rc e p tio n o f his ow n pollution is his know ledge th a t he will be
h en c efo rth a polluted being in the eyes o f the world. H eracles
ex p lain s his first suicidal im pulse as a w ay o f escaping the
‘d isg rac e th a t aw aits m e’; his do m in atin g em otion is one of
sh a m e , sh a m e above all th a t his ‘child-killing pollution’ should
b e seen by T heseus, his d earest friend.40 O edipus begs the
a tte n d a n ts to h ide him aw ay, o r kill him , or hurl him into the sea
‘w h e re you will never see m e a g a in ’.41 H eracles, O restes, and
O e d ip u s all im agine the contum ely an d rejection they will
suffer as p o llu ted exiles;42 from N eoptolem us an d C reon we see
th e w ay in w h ich ruthless enem ies could exploit th eir m is­
fo rtu n e s ag a in st th em .43 W hen heroes or their a tte n d a n ts say
th a t they p o llu te the sun, o r th a t earth itself will not receive
th e m , it is te m p tin g to see this rejection by the very elem ents as
36 889; 1317-46.
37 S e e p . 123 n. 77.
38 E u r. H F 1157 IT., cf. 1214 f.; Soph. O T \ 2 7 0 -4 , OC 1132-5.
35 C f. p. 251 n. 90.
40 E u r. H F 1152; 1160; 1156, 1199-1201.
41 S oph. 0 7 Ί 4 1 1 f., cf. 1436 f. ‘W here I m ay be seen and addressed by nobody’.
42 E u r. H F 1 2 81-90, El. 1195-7, Soph. O T 1380-3.
43 E u r. Andr. 9 7 7 -8 ; Soph. O C 9 4 1 -9 .
Some scenesfrom Tragedy 317
.m ex trem e ex tension of th eir exclusion from th e society o f
men. 44
H ere as elsew here we see a convergence betw een the conse-
( juences o f p o llu tio n an d o f d isg race.45 T h e S ophoclean Ajax in
his sh am e does n o t react very differently from O ed ip u s and
H eracles in th e ir p ollution. H e sp u rn s food an d drink, feels
1m ted by b o th gods a n d m en, could n o t look his fath er in the eye,
.ind dev o tes h im self to n ight because he is ‘u n w orthy to look
w ith p ro fit on a n y god or m a n ’.46 H elen, a disgraced w om an, is
re p ro a c h e d b ecau se ‘You show ed y o u r face u n d er the sam e sky
as y o u r h u s b a n d , you foul c re a tu re .’47 D em osthenes’ political
o p p o n e n ts, th o u g h they h ad b etray ed the G reek w orld to Philip
lor b rib es, ‘felt no sh am e before the sun nor th eir native land, on
w hich th ey sto o d . . Λ 48 B oth p o llu tio n an d disgrace should lead
to th e sam e ‘sh am e before th e s u n ’.
F o r th e victim , therefore, th e consequence ofh is pollution lies
not so m u ch in im m ed iate d a n g e r as in social stigm a. T heseus
gives H e racle s courage to live on by show ing him th a t he is not,
a lte r all, w holly cu t off from his fellow m en. W ith infinite
delicacy h e p ersu ad es H eracles to confront the ou tsid e world,
first passively by sight, th en by speech, and finally by actual
p h y sical c o n ta c t w ith one w ho is n o t po llu ted .49 W h a t disturbs
a n d d istan ce s th e m o d ern re a d e r in th e case b o th o f H eracles
a n d O e d ip u s is th e in ten sity o f th e pollution th a t em an ates from
a n u n in te n tio n a l act. C ertain ly , it expresses an im m ediate hor­
ro r th a t is w holly com prehensible, b u t it goes d eep er th an that
b ecau se it leaves a p erm a n en t stain. O edipus an d H eracles are
not, how ever, red u ced to the level o f rab id dogs w hich no sane
m a n w o u ld th in k o f ap p ro ach in g . T h e ir presence inspires u n ­
ease a n d revulsion at the th o u g h t o f the fearful acts which,
th o u g h un w ittin g ly , they have perp etrated ; b u t truly
m a g n a n im o u s figures are not d e b a rre d from helping them in
th e ir d istress. I t is p erh ap s not frivolous to point o u t th a t sim ilar
rev u lsio n , w ith sim ilar consequences, is far from unknow n in
44 E m pedocles B 1 15.9-12; Soph. O T \ 4 2 4 -8 ; E ur. H F 1295-8, Or. 822, Med. 1327 f.
(cf. 1251 f.), E l. 1177—9; the A lcm aeon legend.
45 Cf. p. 94 a n d p. 205.
46 324, 4 5 7 -8 , 4 6 2 -5 , 3 9 7 -4 0 0 .
47 E ur. Tro. 1 0 2 3 -4 .
48 D em . 19.267.
49 E ur. H F 121 4 -3 4 , 1398-1400.
318 Miasma
m o d e rn w estern society. U nease before p henom ena th at
th re a te n the o rd e r a n d n orm al assum ptions o f a given culture
c a n still o v erru le purely m oral form s o f assessm ent. Physical
illness is no longer sham eful o r dangerous a n d can be talked
a b o u t freely, b u t m en tal disease rem ains in m any circles dis­
g racefu l, u n m en tio n ab le, a n d th rea ten in g to the highest degree.
E x -m en ta l p a tie n ts are expected to find new hom es, a n d ex­
p e n d e x tra o rd in a ry energies in seeking to disguise their past.
S u ch p o llu tio n is m oreover em inently contagious. ‘T h e loyal
sp o u se o f th e m en tal p atien t, the d a u g h te r o f the ex-con, the
p a r e n t o f th e cripple, the friend o f the blind, the family of the
h a n g m a n , a re all obliged to sh are som e o f the d iscredit of the
stig m a tiz e d p erson to w hom they are re la ted .’50 Im m ediate
co n n e ctio n s are au to m atically affected; friends have the choice
w h e th e r to expose them selves to co n tam ination by m aintaining
th e asso ciatio n . T h eseu s a n d Pylades m agnanim ously risk pol­
lu tio n of th e ir ow n free choice, b u t no such decision is open to
th e c h ild re n o f O ed ip u s. It w ould be useless for them to shun
th e ir fa th e r’s presence, because the very blood th a t runs in their
veins is pollu ted ; by h an d lin g th em ,51 O edipus w orks no further
h a rm . T h e p o llu te d m a n ’s w orld is thus divided betw een an
inside circle th a t shares his stigm a an d society at large th at fears
a n d rejects it. ‘R eligion d em an d s th a t only relatives should see
a n d h e a r a m a n ’s affliction’, says C reo n .52 O restes well knows
th a t no o u tsid e r w ould offer him his d a u g h te r in m arriage,
a lth o u g h a relativ e m ig h t.53 Before his father, H eracles sim ply
la m e n ts his fate; his intense feeling o f exposure an d sham e
begins w hen T h eseu s arrives.54
It is a p p ro p ria te to end w ith a few rem arks on the Oedipus at
Colonus, a play th a t illustrates m ost o f the points a b o u t reactions
to p o llu tio n th a t have been discussed in this c h a p te r.55 (Even in
this very re stric te d field, S ophocles’ prim acy am ong the trage­
d ia n s in the p o rtra y al o f p lausible hu m an attitu d e s is unm is-
' E. C offm an. Stigma, N ew jerse y , 1963 (L ondon, 1968), 43, cf. 64. In both cultures,
we a re d e alin g w ith a ttitu d es, not legal disabilities.
51 S oph. <)T 1480 f. T h is pollution too is social, not legal (p. 205).
52 S oph. O T 1430 f. C onversely, love m akes the disgusting tolerable: Aesch. fr. 137,
και μήν, φιλώ γάρ, άβόέλνκτ' έμοί τάόε.
53 Kur. Andr. 975.
54 N ote the d ra m a tic όφθησόμεαθα o f 1155.
55 I am g ratefu l lor several points about OC to C hris M egone.
Some scenesfrom Tragedy 319
t.ikable.) O n ly u n d e r com pulsion, a n d only by a n oblique
ap p ro a c h , does O e d ip u s reveal to th e chorus o f m en o f C olonus
liis te rrib le id e n tity .56 T h e ir im m ed iate reaction is one of terror,
liorror, a n d , in consequence, irra tio n a l aggression: they have
I>cen ‘d eceiv ed ’ by O e d ip u s in to to leratin g his presence, a n d to
I>reak th e ir p ro m ise will be an ac t o fjustified ‘revenge’.57 As the
very voice o f O e d ip u s w ould a t this p o in t be terrib le to them ,
A ntigone tactfu lly intervenes. T h e chorus are softened b u t not
m oved. ‘K n o w , d a u g h te r o f O e d ip u s, th at we pity you a n d him
alike for his affliction. B ut w e fear th e consequences from the
gods, a n d c a n give no an sw er beyond w hat we have already
sa id .’58 T h e fear o f ‘consequences from the gods’ is an elem ent
not n o rm ally p re se n t in m o d ern responses to stigm atized
perso n s; b u t a b o u t th e concern th a t finds expression th rough
I his idiom th e re is n o th in g unfam iliar. O ed ip u s, however,
p ersu a d e s th em a t least to aw ait the verdict o f T h eseus. A little
later, th ey tu rn to O ed ip u s: ‘I t is terrib le to stir u p an evil th at
has lo n g lain q u iet, stran g er: b u t none the less I long to know’
th e sto ry o f his affliction.59 M an y h ave felt th a t th ere is some­
th in g h ea rtless in th e ch o ru s’s inquisition o f O ed ip u s, b u t it
does n o t lack psychological plausibility. I f th ere is to be any
sem b lan c e o f n o rm al in terco u rse betw een the tain ted person
a n d th e w orld, th e ta in t m u st be b ro u g h t out into th e open and
pu b licly acknow ledged. O th erw ise b o th parties a re constrained
by a n im p o ssib le u n ease.60
O e d ip u s ’ e x p lan a tio n concludes w ith a firm self-vindication:
‘P u re by th e law , unknow ing, did I come to th is.’61 Before
T h e se u s, o n his arriv al, no w ord o f justification is required;
T h e s e u s ’ ow n sufferings have ta u g h t him h u m an ity . Creon,
how ever, reveals th e sense in w hich O e d ip u s’ pollution is a
‘re p ro a c h ’ th a t can be exploited against him by a n enem y a t any
tim e. T h e re is no suggestion th a t C reon is him self frightened of
56 2 0 3 -2 3 ; cf. E. GofTman, op. cit., 143 on ‘disclosure etiq u ette’.
57 2 2 9 -3 6 .
58 2 5 4 -7 .
59 509 f.
60 Cf. E. G oflm an, op. cit., 143. T h e source o f a very apposite rem ark unfortunately
escapes me: ‘E in seltsam unglücklicher M ensch, und wenn er au ch schuldlos wäre, ist
a u f eine fü rch terlich e W eise gezeichnet. Seine G egenw art erreg t in allen, die ihn
g ew ah r w erd en , eine A rt von E ntsetzen. Je d e r will das U ngeheure ihm ansehen, was
ihm aulerlegC w ard ; je d e r ist neugierig u n d ängstlich zugleich.’
61 548.
 Some scenesfrom Tragedy 321
320 Miasma
is en v isaged for O e d ip u s th a n th e p ow er to co n tin u e helping his
O e d ip u s ’ ta in t, b u t he is h a p p y to declare th a t he ‘well knew friends a n d h a rm in g his enem ies from the grave. T h e suggestion
th a t A th en s w ould not receive a n im pure father-killer like is, how ever, m ad e in th e play, a n d is n o t to be entirely dism is­
th is ’.62 T o this ch arg e o f pollution O edipus responds w ith a sed, th a t in g ra n tin g him this pow er even the gods have in the
furiously w orded assertion o f his innocence.63 T h e contrast w ith en d ‘h a d som e c a re ’ for the m a n they involved in th e d irest of all
th e Oedipus Tyrannus has often been noted, w here a defence of p o llu tio n s.69
this kin d w as far from O e d ip u s’ m ind; it has som etim es been
su p p o se d th a t th e d o ctrin e o f pollution h ad undergone a m odi­
ficatio n in th e in terv en in g years, to take account o f motive. T his
w o u ld be a su rp risin g developm ent, since in respect of guilt, at
least, th e relevance o f in ten tio n h ad been well understood in
A th e n s since a t least the tim e o f D raco. In Oedipus Tyrannus, we
see th e first re actio n o f p assio n ate d isgust to a crim e whose very
o b jectiv e e n o rm ity leaves no place for ratio n al calculation of
g u ilt. L ong years have passed in Oedipus at Colonus, a n d O edipus
h as com e to term s w ith his deeds by clearly form ulating his own
in n o cen ce. H is self-abhorrence, though not destroyed, has been
g re a tly re d u ced , a n d so n atu ra lly also his sense o f personal
p o llu tio n . I t still persists, however; by a co n tra st o f beautiful
p lau sib ility , to C reon, who ta u n ts him w ith it, he m akes no
ad m issio n o f pollution, b u t before T heseus, his saviour, he feels
h im se lf im p u re .64
N e a r th e en d o f the play, w hen a th u n d erc lap sum m ons
O e d ip u s to his m iraculous d e a th , the chorus suppose for a
m o m e n t th a t they a re ab o u t to be p unished for associating w ith
a p o llu te d m a n .65 T h ey are w rong; T h ese u s’ h u m an ity to the
w a n d e re r, tru e to the A th en ian tradition, did indeed bring
‘benefit to th o se w ho received h im ’.66 A bout O ed ip u s him selfit
is less easy to be confident. T h e re is a d an g er o f describing in too
m ellow a n d h arm o n io u s term s the ending o f a play w hose hero
d ec la res th a t his ‘polluting dem o n will live on for ever’ in
T h e b e s ,67 a n d w ho shortly before his d ea th has condem ned his
sons to m u tu a l d estru c tio n . T h e idea th a t heroization can be a
‘c o m p e n s a tio n ’ for suffering is a tte ste d in P in d a r,68 b u t no m ore

62 944 f.
63 9 6 0 -1 1 0 2 .
64 1 1 3 3 -5 . Cf. P. E asterling, Greece and Rome 24 (1977), 127.
65 1462-85.
66 A n oracle in schol. Soph. OC 57 p erh ap s im plies a n actu al T h e b a n defeat near
C o lo n u s, cf. J a c o b y on A ndrotion, 324 FG rH fr. 62. 67 788.
68 O l.l.77. I. M . L inforth, Univ. Cal. Publ. in Class. Phil. 14 (1950—2), 102, refers also
to P ind. Nem. 1 .6 9 -7 2 , E ur. Hipp. 1423.

EPILOGUE
T h is book has no t been a history; the evidence for significant
ch a n g e in a ttitu d e s to pollution is too sparse. If we look forward
briefly b eyond th e fourth century, we still find m ore evidence
for co n tin u ity th a n transform ation. O ne fam iliar figure does, it
is tru e, seem to d isap p ea r, th a t o f the polluted m u rd erer. Little
is know n a t all a b o u t the legal a n d social responses to hom icide
in this perio d , b u t it is pro b ab le th at, if pollution h ad been m uch
spoken of, it w ould in som e w ay have in tru d ed upon the
sources. T h e function o f ‘purification’, o r the restoration of
n o rm a lity by a positive a n d public act, had been taken over by
legal process, a n d it g rad u ally ceased to be necessary to think of
th e killer as significantly different from any o th er m alefactor
w hose offences w ere d ealt w ith by the co u rts.1 S om ething simi­
la r ca n p e rh a p s be observed in respect o f certain forms of
sacrilege. In h istorical tim es, cu ttin g sacred wood no longer
evokes a savage p u n ish m en t from the gods, as it does in m yth,
b u t a co m p arativ ely m odest fine.la T h e gods could afford to be
m o re len ien t because they now h a d precinct governors who
p ro v id ed effective p ractical protection for their groves.
in o th e r areas, however, change is h ard er to find. C hrysippus,
as w e h av e seen, criticized the ‘irra tio n a lity ’ o f rules forbidding
b irth , co p u latio n , a n d d e a th on sacred ground, an d sim ilar
feeling can alre ad y be found in E uripides;2 b u t this h ad no
in flu en ce o n cu lt practice. M ore significant perh ap s was the
fam ous co u p let inscribed in the fourth century above the portal
o f th e tem p le o f A sclepius a t E p id au ru s: ‘H e w ho goes inside the
sw eet sm elling tem ple m ust be p u re (hagnos). P urity is to have
1 But lor residual ritu a l concern see L SS 112, certain forms o f accidental killing
specifically d e c la re d pure; B C H ( 1978), 325, line 9; LSCG 55, earlier text (Sokolowski,
p. 108); an d the persistence o f the E leusinian proclam ation (p. 283).
'a A bove, p. 165. P recinct governors: J o rd a n , 2 3 -8 .
2 A bove, p. 34.
Epilogue 323
.in ho n est m in d (literally, to th in k hosia).’3 T h e ideal o f hosiä had
l<mg h a d a m oral dim ension th a t hagneia norm ally lacked. T h ere
was so m eth in g com ic ab o u t using hagneuô in the sense o f ‘be
ju st’;4 th o u g h S ophocles’ C reo n said th a t he was hagnos in
respect o f his niece A ntigone, w hom he was b u ry in g alive, he
could scarcely h av e claim ed to be hosiosf an d X en o p h o n brings
o u t th e c o n tra st w hen he says th a t the gods ‘take pleasure in
good acts {hosia) no less th a n in p u re offerings’.6 A b ro ad er
in te rp re ta tio n o f hagneia w as also possible (it ap p ears in
S oph o cles),7 b u t it is fair to say th a t the E p id au ria n couplet
unites tw o con cep ts th a t in trad itio n al usage w ere alw ays liable
lo be d ra w n a p a rt. T h is m o ralizatio n o f ritu al p u rity has obvi­
ous affinities w ith th e insistence by w riters of the fifth a n d fourth
cen tu ries th a t th e m odest offerings o f â pious disposition are
m ore w elcom e to th e gods th a n hecatom bs slain by the lawless
ric h .8 Katharos h a d begun to be used in the sam e kind o f way
ra th e r e arlier. I t h ad p ro b ab ly long been possible to say th at an
op en a n d straig h tfo rw ard m an h a d ‘a clean m in d ’,9 a n d it was
s ta n d a rd co lloquial G reek to d u b a villain miaros, ‘d irty ’.10
S lightly m o re specific ap p licatio n s a p p e a r a t the end o f th e fifth
ce n tu ry . A risto p h an e s’ m ystic choir, in a p arody o f ritual, bans
Irom its co m p an y all those w ho are ‘im p u re in th o u g h ts’; we
begin to h e a r o f people ‘purified in so u l’ a n d o f m inds th a t ‘have
a p o llu tio n ’. 11 T h e idea o f a p o llu ted m ind follows n atu rally
from th e specification in hom icide law th at ‘the p lan n er be
tre a te d in th e sam e way as th e m an w ho did it w ith his h a n d ’.12
T h is ten d en c y cu lm in ated in such form ulations as P lato ’s claim
th a t th e w icked h av e no access to th e gods, because ‘th e bad
m a n is im p u re in s o u l. . . an d neith er a good m an n or a god m ay
3 Ap. P o rp h . Abst. 2.19. N ock suggests (ii, 851 = H S C P 6 Î (1958), 418) th at it was the
rising A sclepius c u lt’s im itatio n o f th e well-know n D elphic tem ple precepts.
4 A lexis, fr. 15.6, cf. (less clear) Eupolis, Demes, 62 Page ( GLP, p. 212), 79 A ustin
(p. 89).
5 S oph. A nt. 889.
6 X en . Ages. 11.2.
7 O T 864.
" C l. sources cited in P o rp h . Abst. 2 .1 3 -2 0 ; also H d t. 1.50.1, E ur. fr. 327,946, PI. Leg.
955e, M en. fr. 683 (ifg en u in e), T h eo p h r. fr. 152 W im m er.
* T h eo g . 89, E u r. Med. 660; χαθαρώς — honestly, T heog. 198; p u re m ind o f m odest
w om an , P M G 901.
10A bove, p. 4. A lread y in Ale. 347.4 ?
' ‘Ar. Ran. 355; E u r. Hipp. 317, Or. 1604 (in fact hagnos) ; above, p. 281 n. 3.
12A ndoc. 1.94.

324 Miasma
rig h tly receive gifts from th e p o llu te d ’, or in the E picharm an
verse (o f u n c e rta in d ate): ‘Ify o u have a p u re m ind, you’re pure
in all y o u r b o d y .’13
T h e E p id a u ria n couplet enjoyed enorm ous p o pularity in the
follow ing centuries. C h ristia n w riters qu o ted it w ith apprecia­
tion, v aria tio n s on the sam e them e entered gnom ic lite ra tu re ,14
a n d m an y a sacred law co n tain ed the instru ctio n not to enter
unless ‘p u re not only in body b u t also in soul’.15 B ut its signi­
ficance is easily overestim ated. I t did not m ake m orality an
o b ject o f religious concern for the first time; it m erely as­
sim ilate d th e two entirely trad itio n al requirem ents o f hagneia
a n d .‘th in k in g hosia’. Pious G reeks m ay norm ally have seen the
tw o th in g s as d istin ct, b u t they h a d alw ays believed both to be
necessary. ‘H ow could I pray to Z eus’, asked Eum aeus, ‘if I
m u rd e re d m y guest?’16 M ore im portantly, it is q u ite m istaken
to see in th e co u p let a b reaching o f barriers, com parable to the
C h ris tia n d e c la ra tio n th a t all foods are pure. M orality m ight be
in c lu d e d w ith in the category o f purity, but it did not replace
th a t c a te g o ry ’s m ore trad itio n al content, any m ore th a n Philo’s
allegorical in te rp re ta tio n o f the M osaic dietetic laws exem pted
th e w o rsh ip p e r from observing them literally. It is clear that,
d e sp ite th e d o ctrin e o f the Phaedo, citizens o f P lato ’s M agnesia
w ould be su b jected to the fam iliar purifications a n d absti­
nences; b irth a n d d e a th w ere no m ore perm issible w ithin the
tem p le a t E p id a u ru s th a n in any o th er consecrated a re a ;17 the
m o ral in ju n ctio n s in sacred law s occu r am id a w elter o f require­
m e n ts for p u rity from birth, d e a th , intercourse, a n d the eating
o f m eat. C o n ta c t w ith E gyptian a n d oriental cults m eant that
th e H ellen istic p erio d saw n o t a decline b u t an increase in ritual
a b stin e n c e s, w hich w ere no t confined to m arginal superstition
b u t w ere tre a te d by a cu ltu red G reek such as P lu ta rch with
in te re st a n d respect. (D espite its initial im pulsion tow ards a
w holly m o ral view o f pollution, even C h ristian ity could not
p e rm a n e n tly sta n d ou t ag ain st w h a t was seen as the inherent
13 PI. Leg. 716 d —e; (E p ich arm u s), fr. 269.
14J . B ernays, Theophrastos’ Schrift über Frömmigkeit, Berlin, 1866, 77; Ps.-Phocylides
Sent. 228, Anth. Pal. 14.71, 74 (cf. Philol. 17 (1861), 551 ).
15 L S S 91.5, cf. 59.13; 82; 86.3; 1 08.6-7; LSC G 139.3-7; Clem . Λ1. Strom. +.22, p. 311
S t.; B C H 51 (1927), 120. T h e ideal is ascribed to Pythagoras in Diod 10.9.6.
16 H orn. Od. 14.406.
17 C f. SIC,3 1168.5; P aus. 2.27.1.
Epilogue 325
im p u rity of, p artic u la rly , th e fem ale body.18) It is tru e th at
c e rtain specifically m oral req u irem en ts were som etim es intro­
d u ced , b u t by assim ilation to the fo rm at o f the ritu al hagneia
Ibey lost m u ch o f th eir force; fornication was deem ed to pollute
m ore th a n leg itim ate in tercourse, b u t this only m ean t exclusion
from th e sh rin e for a few ex tra d ay s;19 ab o rtio n becam e a serious
pollution, b u t still one th a t the passage o f tim e could cure.20
( )nly th e re m a rk ab le p rescriptions o f a basically un-G reek
p riv ate cu lt ce n tre a t P h ilad elp h ia declared th at those who
tran sg ressed fu n d a m e n tal m oral laws were p erm an en tly unfit
10 w o rsh ip th e m ig h ty gods o f the shrin e.21 W ithin the m ain­
strea m o f G reek cu ltu re the E p id au ria n couplet altered
nothing.
C o n tin u ity has been one them e o f this book; div ersity is
.m o th er. A c o n stan t, p erh ap s an obsessive atte m p t has been
m ade to tra c e divergences a n d m ark o u t lines of differentiation.
I lie ju stific a tio n is th a t there has been a tendency in the past to
see p o llu tio n as a single hom ogeneous category ab o u t w hich
un q u alified g en eralizatio n s can be m ade: ‘Pollution was un­
know n to H o m e r’, o r ‘R itu al im p u rity has no relation to m oral
v alu es.’ T o su ch claim s one m ight well respond w ith the tire-
som e ‘B u t w h a t do you m ean by . . .?’ of the philosophers. A
g en eral th eo ry o f pollution m ay prove, as a goal, an ignisfatuus,
•is ‘d irtin e s s ’ is a n a tu ra l source of m etaphorical a n d sym bolic
expression th a t is liable to be exploited in an alm ost unlim ited
n u m b e r o f w ays; a n d in the G reek case it has com e to overlap
w ith ideas o f collective responsibility and divine an g er whose
I· igical o rig in is p erh ap s q u ite d istinct. But it m ay be interesting
m conclusion to relax this vigilance in discrim ination a n d re­
unite som e o f th e scattered pollutions (or near pollutions) by
i elatin g th em very generally to the norm s o f an ordered
<xistence.21“ By doing so we are once again rejecting th e idea
IMK. J .j o n k e r s , λ /nemos. 113 (1943), 1')()-(>(); G .K .M . de Ste C roix, The Class Struggle
m thr Ancient Greek World, L ondon, 1981, 109; cf. th e churching o f wom en.
,w See p. 75.
10 See A p p en d ix 3.
11 LSA 20.
J,a Cf. th e very interestin g attem p t, w hich anticipates M ary Douglas, o f
11 |c a n m a ire , R H R 145 (1954), 103. It begins: ‘L ’im pureté est ressentie lorsque des
• «intacts ou des rap p o rts ju g és an o rm au x s’établissent entre des ord res de choses qui
•In iv n it rester d istin c ts.’
326 Miasma
th a t a c u ltu re ’s beliefs ab o u t pollution derive from anxiety or a
sense o f guilt. T h e y are ra th e r by-products o f a n ideal o f order.
A first re q u ire m e n t is, it seem s, the veiling or repudiation of
w h a t is d isru p tiv ely o r disgracefully physical. Civilized life has
no p lace for those dying or being born, excreting, o r engaged in
sexuality. T h e philosophers w ho m ade it their ideal to ‘live in
a g re e m e n t w ith n a tu re ’ attac k ed c u ltu re’s p recepts in ju s t these
areas. R ules a g a in st dying, copulating, o r being born in sacred
p re cin cts, C h ry sip p u s poin ted out, divide us, irrationally, from
th e a n im a ls.22 H ero d o tu s, th o u g h evaluating it differently, had
seen th e rule a b o u t copulation in the sam e way, as a m ark o f the
civilized o r u n n a tu ra l life; it distinguished the nations he most
esteem ed , G reeks an d E gyptians, from the b ru te creation and
th e rest o f m a n k in d .23 A n o th er prerequisite for dignified,
o rd e re d existence, again connected w ith control o f the body, is
h e a lth . P a rtic u la rly alarm in g are the disruptions caused by
m ad n ess, w hich can lead to a com plete loss of control, an d by
skin disease, a co rru p tio n o f the body’s visible form. B ut if the
in d iv id u a l is subjected to external intrusion o f any kind,
th ro u g h th e a rts o f the sorcerer, purification is required. An
o rd e re d existence is obviously im possible if n a tu re breaks its
ow n rules. U n n a tu ra l occurrences such as m onstrous births
m ay, therefore, re q u ire p urification - although it is also possible
th a t they have been caused by the gods m erely to presage
e x cep tio n al events. D iet d em an d s no strict control; b u t it was
th e a b a n d o n m e n t o f cannibalism th a t m arked a decisive step
fo rw ard from the prim eval savagery, a n d m an differs from the
a n im a ls in n o t ea tin g d u n g .24 T o sleep w ith a blood-relation is a
m o n stro u s act; it is like m u rd erin g o n e’s father, or eating the
flesh o f a k in sm a n .25 H ere too opposition from the advocates of
n a tu re helps to define the civilized norm s. Z eno an d C hrysip­
p us ta u g h t th a t one should be p rep ared to sleep w ith one’s
m o th e r o r d a u g h te r, should circum stances d em an d it, as also to
e a t th e lim bs o f o n e ’s d ead p a re n ts.26
F u rth e r rules relate m ore specifically to social life. B oth sexes
22 Ap. Plut, de Stoic. Rep. 10441-1045a.
2J 2.64.
24 See A ppendix 4.
25 See p. 98.
26 S VF, i, nn. 2 5 3 -6 , iii, nn. 743-52.
Epilogue 327
m u st u p h o ld th e v irtu es th a t are distinctive for them . T h e m an
who accep ts a passive sexual role becom es th ereb y a w om an,
w hile th e w o m an w ho a b a n d o n s h er sham e is a m an, or a dog;
bo th lose th e ir rig h t to a place in com m unal life. Life in society is
based on th e prem iss th a t each individual m ust be accorded a
c e rta in m in im u m o f respect, a n d the concrete vehicle o f honour
is th e body. T o den y a corpse b u rial is, therefore, in norm al
circ u m sta n ces a d an g ero u s act, because it carries co n tem p t to a
p o in t a t w h ich sh ared existence becom es im possible. (Cynics, o f
course, care n o th in g for th e fate o f the corpse.26“) T h e most
violent a ssa u lt u p o n social o rd e r is th a t by m u rd er. Before the
in stitu tio n s o f th e classical city h ad developed, killing d istu rb ed
(he e q u ilib riu m betw een th e tw o families involved; by the fifth
c e n tu ry it h a d becom e a n o th er offence against a basic ru le o f life
in society. ‘M u tu a l slau g h ter’ w as now a characteristic of the
prim itiv e p ast, from w hich O rp h e u s by his gentle h arm o n y had
d ra w n civilized m an aw ay. P lato believed th at m a n ’s dangerous
a n im a l n a tu re w as revealed in d ream s not ju s t o f incest and
c a n n ib a lism b u t also o f m u rd er.27 O th e r obligations fell to the
in d iv id u al as a m em b er o f th e sm allest social group, the family,
a n d alm o st th e largest, the city. T h e basic needs o f b o th were
(he sam e, n u rtu re a n d p ro tectio n from attack. A nyone who
d e p riv e d his p a re n t or his city o f eith er was liable to a curse,
c o m p a ra b le in its effects to pollution, invested w ith the full
pow er o f rig h t.
F inally, in ad d ition to obligations tow ards kinsm en, unrelated
fam ilies, a n d o n e’s native land, there were the claim s o f the
m asters a n d arb ite rs of civilized life, the gods. T h e m ost crucial
in stitu tio n s th ro u g h w hich m en deal w ith one a n o th er —hospi­
tality, su p p licatio n , an d th e o ath - w ere u n d er th eir protection.
The savage C yclopes, who lived in no cities, ploughed no fields,
d ra n k no w ine, a n d ate h u m an flesh, also cared n o th in g for the
gods. R esp ect for th eir im ages, precincts, an d cerem onies was
the m àrk o f a m an fit to live in society,28 free from the disgusting
a n d bestial q u ality of bold sham elessness. T his was the true
sou rce o f ‘rev eren t p u rity in every w ord and d eed ’.
26a F o r M o sch io n (fr. 6. 3 0 -3 3 Snell) th e laws o f burial are an o th er cultural product,
a co n seq u en ce o f progress.
27 Resp. 5 7 1 c -d .
28 B ut n o t o f a C ynic, SV F , i, nn. 2 6 4 -7 , ‘A nacharsis’, Epistle 9 H ercher.

Appendix 1: The Greek for Taboo
T h e d istin c tiv e fe a tu re o f ‘ta b o o ’ is th a t it un ites th e sacred a n d the
u n c le a n w ith in th e single categ o ry o f th e forbidden. It is n ot su rp ris­
ing, th erefo re, th a t tab o o has often been m en tio n ed in connection
w ith th e ag-lhag- w o rd gro u p , w h ich seem s to co n tain w ords denoting
b o th sa c re d a n d p o llu te d .1 B y zan tin e scholars even believed th a t the
s a m e w o rd co u ld in different co n tex ts b ea r b oth m eanings; thus
C r a tin u s is sa id to h av e used hagios in the sense o f miaros.2 T h e
e x p la n a tio n th a t they offer for th e p h en o m en o n , ‘eu p h e m ism ’, is
u n p e rsu a siv e , a n d som e o f th e ev idence q u o ted by them sim ply
irre le v a n t; b u t th e theo ry itself o f the d o u b le value o f ag- a n d hag-
w o rd s is n o t a B y z an tin e in v en tio n ,3 an d th e scholars w ho form ulated
it w ill h av e b een a c q u a in te d w ith a far w ider ran g e o f evidence for
classica l u sag e th a n w e a re to d ay . E ven in th e su rv iv in g texts, agos
o n ce m e an s so m e th in g like ‘ex p iato ry offering’,4 a n d exagistos is
c e rta in ly u sed for bo th ‘u n to u c h a b ly sa c re d ’ an d ‘ac c u rse d ’;5 an
a d je c tiv e panagês too is found in post-classical texts w ith b o th positive
a n d n e g a tiv e sen ses.6 B ut, as w e h av e seen, th e ex p lan atio n o f the
a m b ig u ity lies n o t in a failure to d ifferen tiate the sacred an d the
u n c le a n , b u t in th e possibility o f a perilous p u n itiv e consecration.
U n to u c h a b ility , a n d hen ce ‘p o llu tio n ’, is a consequence o f such
p e rilo u s c o n se c ra tio n , b u t the m erely p o llu ted is n o t consecrated.
O n e a s p e c t o f tab o o , how ever, th a t o f u n to u c h ab le sanctity, is
c e rta in ly ex p re ssed th ro u g h som e w ords in the group. W e noted
e a rlie r th a t th is id e a is n o t conveyed a t all th ro u g h hieros, an d only
p a rtia lly in hagnos/ hagios·,1 b u t it is in se p arab le from the verbs hagizô,
enagizô, a n d kalhagizß, w hich a re used o f a con secratio n th a t alw ays
involves c o m p le te rem oval from th e h u m a n sphere. T h e y com m only
1 Cf. p. 6, an d for docum entation on w hat follows the im portant article ofG hantraine/
M a sso n cited there.
2 C ra tin u s, fr. 373. Cf. e.g. Pearson on Soph. fr. 689, M oulinier, 250—2.
3 F irst a tte ste d in H elladius ap. Phot. Bibl. 535a8. 'E uphem ism ' or anliphrasis is a
s ta n d a rd topic o f rheto rical handbooks.
4 S oph. Ant. 775 (puzzling), cf. fr. 689.
5 S oph. OC 1526; D em . 25.93, A eschin. 3.113 etc.; on the im p o rtan t but obscure
e p ig ra p h ic evidence see A. M . W oodw ard, Hesperia 25 (1956), 100 f, (cf. Hesperia 43
(1974), 177 n. 77).
6 L SJ s.v.; C h a n tra in e /M a ss o n , op. cit. 7 p. 151.
Appendix 1 329
I elcr to th e b u rn in g o f offerings in th e c u lt o f the d ead o r heroes, b u t
■.in also be a p p lie d to incense o r th ig h s b u r n t in O ly m p ia n sacrifice,
οι to lib a tio n s; th e essen tial p o in t is th e en tire d e s tru c tio n o f the
u llrrin g .8 T h e v erb hagnïzô h as a su rp risin g special sense in trag ed y
w hich is closely co m p a ra b le .9 N o rm ally it m eans ‘p u rify ’, a n d is a
m ain ly p o etic v a r ia n t o i' kathaird, u sed in sim ilar co n tex ts alth o u g h
never, o f course, o f sim p le n o n -relig io u s cleansing. B u t there a re a
n u m b e r o f p assag e s in w h ich hagnizö o r its co m p o u n d s govern as
o bjects su c h th in g s as sacrificial cakes, funerary offerings, or corpses.10
I hus w e find ex p ressio n s like τάν θανόντα θ'άγνίσαι an d
τάφφ τε κρύψαι και τα πάντ' έφαγνίσαι
a τοίς άρίστοις ’έ ρχεται κάτω νεκροις.11
\ \ Ί iere th e referen ce is to b u rn in g a corpse, the ren d e rin g ‘p u rify ’ is
p ossible, b ec au se o f th e ca th a rtic force o f fire;12 b u t it is m ore plausible
to see ev en th ese cases as p a rt o f the sam e g ro u p , a n d reg ard hagnizö as
m e a n in g ‘c o n s e c ra te ’ (by d e s tru c tio n ).13 O th e r verbs too - kathosid
a n d hagisteuo14 —a re occasio n ally used in th e sam e sense as enagizô.
N one o f th e m , it is in te restin g to n o te, n o rm ally expresses the m ore
m o d e ra te d e d ic a tio n o f a th in g o r person to the service o f the
( )ly rn p ian g o d s .15 In d ee d , alm o st w ith o u t exception the object o f
i o n se c ra tio n is n o t m erely d ec lared u n to u c h ab le b u t actu ally
d e stro y e d . T h e a c t of'enagizein does n o t, therefore, leave in the w orld a
series o f ta b o o e d o b je c ts .16
T h o u g h a satisfac to ry classical G reek w ord for ‘ta b o o ’ ca n n o t be
lo u n d , a v ery p la u sib le e q u iv ale n t for th e negative sta te o f noa, not-
KCf. C h a n tra in e /M a ss o n , op. cit.; the sense ‘consecrate (w ithout d estru ctio n )' that
they a n d LSJ a d m it for kathagizö is unnecessary - all the passages cited m ay, or m ust,
refer to b u rn in g . Hagizô is a p artial exception, being used o f the consecration o f altars
( L SJ); b u t only in high po etry , an d only, it seem s, o f the kind o f altars liable to receive
holocausts.
9 Cf. M o u lin ier, 279 f., W illiger, 48. LSJ is very inaccurate.
10 S oph. Ant. 196, 545, 1081; E ur. Ion 707, Supp. 1211; cf. Ap. R hod. 2.926. O f hum an
sacrifice, E u r. I T 705, H esych s.v. άγνίσαι (= E u r. fr. 314, Soph. fr. 116); and for
d estru ctiv e con secratio n cf. E ur. Ale. 76. άφαγνίζω = deconsecrate in Eur. Ale. 1146.
11 S oph. Ant. 545, 196 -7 .
12 R o h d e, 334 n. 127, w ith parallels; b u t note th at in Soph. Ant. 545 b u rning is not in
q u estio n .
13 By assim ilatio n o f hagnizö to hagizô?
14 A r. Plut. 661 (p a ra tra g ic ), E ur. I T 1320, T h eo p h r. ap. Porph. Abst. 2.27; Soph. Ant.
247.
15 O n hagizô, a very p artial exception, see above, kathosiô = ‘consecrate’ is post-
classical.
16 B ut th ere are έξάγιστα (see a b o v e ) a n d o f c o u r s e έναγή.
330 Miasma Appendix 1 331

ta b o o , is av a ila b le in th e co n c ep t o f hosiä.17 Hosios h as a basic sense of
‘p e rm itte d o r en jo in ed by th e g o d s’, ‘inoffensive o r pleasing to the
g o d s ’. In d iffe ren t co n tex ts it is c o n tra ste d to b o th th e sacred an d the
p o llu te d . G ro u n d , o r m oney, w hich is hosion is th a t w hich is not
sa c re d , sin ce w holly free use o f sa cred p ro p erty is n o t ‘p erm itted by
th e g o d s ’; on th e o th e r h a n d , hosios is often a v irtu a l synonym of
kalharos o r hagnos, ‘p u re ’, 18 since p o llu tio n is ‘offensive to the go d s’.
T h u s th e re seem s to exist an enclav e o f hosiä, safe n o rm ality , betw een
th e d a n g e ro u s ex tre m es o f sacred n ess an d p o llution. T h re a ts to this
n o rm a lity c a n com e from e ith e r side; hosiä an d its co m p o u n d s are
u se d , in a m o re g en e ral sense th a n kathairö,19 for th e resto ra tio n o f hosiä
b y th e rem o v a l o f th a t o b stacle to it w hich is p o llu tio n , w hile apho-
siausthai h a s a sp ecial ap p lic a tio n for th e fulfilm ent o f a religious
o b lig a tio n (a c o n s tra in t, th a t is, im posed by th e gods) neglect o f
w h ic h w o u ld be a n offence a g a in st hosiä.20 (B ecause unw elcom e obli­
g a tio n s w ere so m etim es ca rrie d o u t w ith n arro w legalism , it acq u ired
a f u rth e r sense o f ‘d o a th in g form ally o r p erfu n cto rily ’.21) In both
ca se s it is a q u e stio n o f p u ttin g o n eself in th e clear.
In p o st-classica l G reek this v erb aphosiousthai u n d erw en t a rem a rk ­
a b le d e v e lo p m e n t.22 T h e w ord does n o t seem to be a tte ste d betw een
th e fourth ce n tu ry a n d the Roman Antiquities o f Dionysius H alicarnassus.
It re a sse rts its e lf p a rtic u la rly in P lu ta rc h , w ith a w ide variety o f
m e a n in g s, m o st o f w hich co n tin u e classical usage o r can readily be
d e riv e d from it: to fulfil a religious o b lig atio n (often p erfu n cto rily ), to
fulfil a n y o b lig a tio n (often a g a in p e rfu n c to rily ), to resto re hosiä after a
p o llu tio n (it n o w h a s a n ac cu sativ e o f th e po llu tio n e x p ia te d ), or to
av o id a p o llu tio n . T h e re also em erges a specialized use, ren d e re d by
L SJ 'to esch ew o n religious g ro u n d s, hold in a b o m in a tio n ’. T w o
fea tu re s o f th is u sag e are p a rtic u la rly n o tew o rth y . O n e o f co u rse is the
stro n g e m p h a s is o n th e religious v alu e o f avoidance, w h ich is treated
as a m e an s o f p rese rv in g hosiä. T h e o th e r is th a t th e v erb expresses the
fact o f relig io u s a v o id an c e w ith o u t im p ly in g an y th in g a b o u t its
m o tiv e (th u s ‘esch ew on religious g ro u n d s’ is a b e tte r in te rp re ta tio n
lh a n ‘h o ld in a b o m in a tio n ’). T h e m a n w ho ab stain s from a food,
w h a te v e r his g ro u n d s, άφοσιοϋται the food;23 to ab a n d o n ac tio n on a
g iv en d a y b ec au se o f b ad o m en s is άφοσιώσασθαι την ημέραν, b u t
P h ilip , a fte r a sn a k e h ad been seen lying beside O ly m p ia s, sh u n n ed
i n te rc o u rse w ith h e r την ομιλίαν ώς κρείττονι σννονσης άφοαιούμενος24
H e re th e n w e h av e good G reek for ‘to tre a t as ta b o o ’ — b u t, m ost
in te re stin g ly , it is n o t a p rim itiv e su rv iv al b u t a H ellen istic dev elo p ­
m e n t.

17 See esp. M . H . van d e r Valk, Mnemos. 103 (1942), 113-40; H . Je an m aire, R EG 58

(1945), 66—89 (w ith the response o f van d e r V alk, ibid., 64 (1951), 4 17-22);
B enveniste, ii, 198—202. N ote th a t Je a n m a ire ’s concept o f desacralization is
in a p p lic a b le to the hosiä o f hum ans; lor them the relevant contrast is not οαιος/ίερός but
δοίος/άνόσιος (or εναγής).
18 e.g. A esch. Ag. 778, Cho. 378, Soph. OC 470, E ur. Ion 150, A ndoc. 1.96. O f course
th e relatio n o f ‘p u rity ’ to hosiä is one o f p a rt to whole. O n the breach o f hosiä by death
a n d m o u rn in g see p. 65 η. 110.
19 R u d h a rd t, 169, cf. e.g. PI. Euthphr. 4c, Leg. 873b. Dem . 23.73, 47.70. I f one ‘sins
a g a in s t the g o d s' by a n im pious speech, one should άφοσιοϋσθαι by recanting it, PI.
Phdr. 242c. W h en , as often, the verbs are used o f responses to m u rd er (p. 121), the
senses ‘p u rify ’ a n d ‘disch arg e an oblig atio n ’ (to the dead m an) becom e inseparable,
since the victim ’s rights are the source o f pollution.
20 H d t. 1.199.4,4.154.4,4.203.1 (here n o t ‘fulfil an obligation’, b u t ‘escape the threat
co n ta in e d in a n o ra c le ’); PI. Phd. 60e, 61b, Phlb. 12b. F or an active use, ‘1 p u t in the
cle a r (by fulfilm ent o f a n ob lig atio n )’ cf. A eschin. 3.120 τήν πόλιν τά πρός τοϋς θεούς
άφοαιώ , C lea rch u s, fr. 43a YYehrli, ap. A th. 516a (for the force here, m istaken by LSJ, cf.
H d t. 1.199.4); sim ilarly, b u t 'by rem oval o f a pollution’, PI. Euthphr. 4c, Leg. 873b.
21 See LSJ s.v. άψοοιόω, II.2.C, W . W yse on Isae. 7.38, and for an instance H dt.
4.154.4.
22 See \V J . T ersteg en , Eusebes en Hosios, diss. U trecht, 1941, esp. 167 f. 1 discussed
this, w ord m ore fully in m y O xford doctoral dissertation (1977), sam e title as this book, 23 C o n tra s t e.g. Plut. Quaest. Conv. 635e (respect), 67 0 f (distaste).
3 8 8 -9 2 . 24 P lu t. Caes. 64.5, Alex. 2.6.

Appendix 2: The Cyrene Cathartic Law
T h e law is SEG ix 72 (cf. xx 717), LSS 115, S o lm sen /F raen k el4 39,
B uck 115. I t w as d iscov ered in 1922 in the R o m an b a th s a t C yrene,
w h e re it h a d been in c o rp o ra te d as a seat for b ath ers in the frigidarium,
a n d p u b lis h e d w ith a n extensive co m m en tary by S. F erri, Notiziario
Archaeologico del Ministero delle Colonie 4 (1927), 93—145. D etailed
re c o n sid e ra tio n s o f th e w hole w ere quickly offered by YVilamowitz,
Sitz. Preuss. Ak. Berl. 19 (1927), 155—76 (not in Kl. Sehr.), G. de
S a n c tis, Riv. Fil. n .s. 5 (1927), 1 8 5 -2 1 2 , A. V ogliano (helped by
P. M a a s ), Riv. Fil. n.s. 6 (1928), 2 5 5 -3 2 0 , K . L atte, A RW 26 (1928),
4 1 - 5 1 = Kl. Sehr. 1 1 2 -2 1 . T h ese co n trib u tio n s, to g eth er w ith notes
o n in d iv id u a l p assag e s by S chulze, R a d erm a ch e r, an d M aas (cited in
LSS a n d SEG), solved m ost o f th e p ro b lem s in th e d o cu m en t th a t
a p p e a r so lu b le. G . O liverio , La stele dei nuovi comandamenti e dei cereali
(D o c u m e n ti A n tic h i d ell’ A frica I ta lia n a 2), B ergam o, 1933, 7 -2 8 ,
35—84, re p u b lish e d the text, m ost u n reliab ly even th o u g h he was
w o rk in g from th e sto n e, w ith a n elab o rate , eccentric, an d often irrele­
v a n t c o m m e n ta ry , b u t excellent p lates. (T h e p lates suggest th a t his
re a d in g s m u st be tre a te d w ith ca u tio n .) G . L u zzatto , La Lex Cathartica
di Cirene, M ilan, 1936, did n o t re-exam ine the stone, an d w rote mostly
from th e p e rsp e ctiv e o f th e legal h isto ria n . T h ese w orks will be cited
by a u th o r ’s n a m e in w h a t follows. F ull b ib lio g rap h ies are av ailab le in
LSS a n d SEG ; a d d O . M asso n , Annuaire de l ’École pratique des Hautes
Études, IVesection, 102 (1 9 6 9 -7 0 ), 232 f. (linguistic notes).
T h e law occu p ies on e face a n d a b o u t tw o -th ird s o f an o th e r o f a
q u a d r a n g u la r stele; the th ird face b ears SEG ix 2, a list o f th e cities
th a t receiv ed g ra in su b v e n tio n s d u rin g th e fam ine o f 331 to 326, w hile
th e fo u rth w as u n w o rk ed a n d b lan k . D eb ate ab o u t th e chronological
r e la tio n of-the c a th a rtic law a n d th e corn su b v en tio n list has proved
in c o n clu siv e. F e rri p u t th e corn list a b o u t 320 an d the ca th a rtic law,
o n th e b asis o f le tte r form s, som e tw enty years later; sim ilarly
W ilam o w itz. D e S an ctis acknow ledged th a t th e le tte r form s o f the
c a th a r tic law a p p e a r younger, b u t ex p lain ed this by th e respective
ag e s o f the sto n e c u tte rs a n d arg u e d for th e c a th a rtic la w ’s priority. Cf.
to o O liv e rio , 10. O f the reaso n for th e sto n e ’s p u b licatio n n o th in g is
k n o w n . S om e h av e su p p o sed th a t it w as o riginally to p p ed by an o th er
ste le w h ich ex p la in e d th e circu m stan ces, as does th e p rea m b le to the
Appendix 2 333
fam ous F o u n d e rs ’ O a th o f C y ren e ( M /L 5 ). W e do n o t know , th ere­
fore, w h e th e r th e G y re n ae an s h ad only recently so u g h t A pollo’s
a p p ro v a l for th e ir c a th a rtic tra d itio n s, o r w h eth e r line A 1 allu d es to
th e m o re d is ta n t p ast.
I offer h ere a tra n sla tio n o f th is d o c u m e n t, w ith discussion of the
m o re im p o r ta n t u n c e rta in tie s th a t rela te to th e th em e ol'this book. No
w holly sa tisfa c to ry te x t is a v a ila b le (LSS a n d SEG a re d e p e n d e n t on
O liv e rio ); th e m o st p ru d e n t is th a t o f F raen k el (S o lm sen /F raen k el4
39), a n d I h av e a d o p te d this as th e b asis o f m y tran slatio n . A rep u b li­
c a tio n from th e sto n e w ould be w elcom e, b u t w ould p ro b ab ly not
larg ely affect o u r u n d e rsta n d in g . Som e co n tro l is av ailab le th ro u g h
( )liv e rio ’s p h o to g ra p h s. Sokolow ski pro v id es an extensive a p p a ra tu s
c ritic u s in LSS. M y tre a tm e n t is selective, an d on the in te rp re ta tio n of
indiv id u a l w o rd s I a ssu m e know ledge o f F ra en k e l’s helpful notes.
A
1—3 'A p o llo d e c re e d th a t (th e G y ren aean s) sh o u ld live in L ibya [?
lo r ev er] o b se rv in g p u rifica tio n s a n d ab stin en ces a n d [ ].’
In view o f G y re n e ’s w ell-know n co n tac ts w ith D elp h i, A pollo here is
su re ly th e D e lp h ia n a n d n ot a C y re n a e a n o ra c u la r A pollo; cf. M /L 5;
H d t. 4.150-8; id em , 4.161.1: C y re n a e a n s ask A pollo δντινα τρόπον
καταστησάμενοι κάλλιστα äv οίκέοιεν, D iod. 8.30: A rcesilaus 3 told by
D elp h i th a t G y re n e w as suffering th ro u g h divine an g e r because of
'd is re s p e c t for p iety to th e g o d s'. O n G y re n ae an respect for ‘Apollo
th e F o u n d e r’ see P. M . F ra ser, Ptolemaic Alexandria, O x fo rd , 1972, i,
788.
N o o th e r sa c re d law co n tain in g a code o f p u rity p resen ts itself as an
o r a c u la r resp o n se. P lato , how ever, envisages th e possibility: see p.
140 n. 147. F o r cita tio n o f w h a t ‘the god d ec reed ’ in o th e r sacred
m a tte rs see I G I 3 7, SIG 3 735.19 f., ib id ., 1158, cf. D em .. 21.51 - 3 . 11 has
b een u n iv e rsa lly reco g n ized th a t D elp h i ca n n o t have p rescrib ed the
c o n te n ts o f th e following code. T h e dialect is G yrenaean, an d allusions
to G y re n a e a n in stitu tio n s a n d cu sto m s are n u m ero u s. (W ilam ow itz
th o u g h t th a t a t le ast th e form χρειμένος in A 3, th e proem , was
D elp h ic, b u t ev en th is is not ce rtain , cf. Buck, §158.) P erh ap s the
G y re n a e a n s s u b m itte d to D elp h i for ap p ro v a l a code th a t they had
a lre a d y d ra fte d (W ilam o w itz); o r it m ay be th a t A pollo’s o racle was
co n fin ed to th e g e n e ra l in stru ctio n to ‘live in L ibya ob serv in g p u rifi­
c a tio n s ’, a n d th a t th e c o n ten t o f th e code does not even profess to be
p a r t o f th e resp o n se.
334 Miasma
The code itse lf is so m e w h at d iso rg an ized , a n d n o t com prehensive;
th u s in B 25 th e re is an allu sio n to d ea th -p o llu tio n , as to som ething
fam iliar, b u t its o p e ra tio n s a re n o w h ere reg u lated . T h e rules it
c o n ta in s a re d o u b tle ss very v ario u s in d a te ; b u t no p a rt o f it read s like
a v e rb a tim tra n s c rip t o f a tru ly a rc h a ic code o f rules.* (N ote for
in s ta n c e the v irtu a l absen ce o f w holly in co m p reh en sib le w ords.) It
p e rh a p s m a n ifests th e sam e k ind o f reto u ch ed arc h aism a s M /L 5, the
F o u n d e rs ’ O a th . T h e po ssibility th a t a m o re o rd erly code has been
a b b re v ia te d a n d u n in te llig en tly reo rg an ized is occasionally raised in
w h a t follows. T h e re a re inco n sisten cies b o th o f p h rasin g (cf. below on
δησεΐ/όησεΐται ) a n d d ia lec t (εκασσα/εκοΐσα , B 5, 7, cf. W ilam ow itz).
4 —7 ‘I f d ise ase [o r ] o r d e a th should com e ag a in st the
c o u n try o r the city, sacrifice in fro n t o f th e g ates [in front of] the shrine
o f av e rsio n (?) to A pollo th e A v erte r a red h e-g o at.’
T h e in te rp re ta tio n o f th e g en itiv e τώ άποτροπαίω in 6 is very u n ­
c e rta in . F e rri, follow ed by L u zz atto , took τό άποτρόπαιον as ‘th e evil
to be a v e r te d ’, w hence su p p le m e n ts like καθαρμόν, ‘a purification
from th e ev il’, άποτρόπαιος as an ad jectiv e bears this sense in post-
classical p ro se, as d oes άπότροπος in trag e d y (cf. L S J), b u t th e su b ­
sta n tiv a l use seem s im plausible. M ost scholars have followed V ogliano
in ta k in g τό άποτρόπαιον as a ph y sical o bject o u tsid e th e gates, a
s ta tu e , u n w o rk e d stone, o r a lta r d ed icated to th e god (so e.g.
W ila m o w itz , Glaube, i, 173n. 1), a n d in su p p le m e n tin g a p reposition,
‘in fro n t o f th e apotropaion'. T h is is m o re a ttractiv e, even though
p a ra lle ls for th e n o u n άποτρόπαιον are lacking.
R ed v ic tim s a re n o t often specified in sacred laws, a n d d o n o t seem
to h a v e h a d a n y fixed significance (cf. P. Stengel, Öpferbraiiche der
Griechen, L eipzig, 1910, 18 7 -9 0 , idem , Kultusaltertümer, 151 f.). L atte
su g g e ste d , c o m p a rin g R o m an festivals, th a t in this case th e red goat
w as a sy m b o lic e m b o d im e n t o f th e evil to be av erted , th e fiery plague;
cf. p. 275, a n d th e b lack bull b u rn t ‘for’ B oubrostis in S m y rn a, Plut.
Quaesi. Symp. 6 9 4 a—b. It certain ly seem s th a t, in a red goat, a rath e r
d is re p u ta b le a n im a l is d elib erately chosen, like the dogs sacrificed to
H e c a te ; g o ats a re sham eless, a n d red h a ir too is a m a rk o f sham eless­
ness a n d evil g e n e ra lly (E. W u n d e rlich , Die Bedeutung der roten Farbe im
Kultus der Griechen und Römer, R G W '20.1, G iessen, 1925, 6 6 -7 2 ). In
- 1 owe this p o in t to B ryan H ainsw orth. O n th e o th er h a n d , IG I 3 104, if a verbatim
tra n s c rip t o f D ra c o ’s code, attests considerable lucidity for a late 7th“century law; on us
style see G a g a rin , Drakon, C h. 8.
Appendix 2 335
l SS 116 A 3, also from C y ren e, LSCG 18 A 33, C 33, ib id ., 20 A 26,
A pollo A p o tro p a io s a g a in receives a go at, b u t in D em . 21.53 an ox
T h e sacrifice to A pollo ‘in front o f th e g a te s’ obviously relates to the
tfod’s fu n ctio n as o ne w ho sta n d s o u tsid e city gates (P ropylaios) or
house d o o rs (A guieus) a n d averts evil from them , a n d m o re generally
■.lands in fro n t’ o f th re a te n e d h u m a n s (P ro staterio s); cf. P reller/
R o b ert, i, 276 n. 1, Æ E 2.64, F arn ell, iv, 1 4 8 -5 2 . F or sta tu e s o f A pollo
o u tsid e th e w alls, firing his arro w s to av e rt p lague, see O : W einreich,
i ilcd p. 276 n. 91; W e in reic h ’s evidence is late, b u t cf. alre ad y Soph.
<)T 202—6. F o r a ‘H e c a te before th e g a te s’ see SIG 3 57 (LSA 50) 26,
',!!) Γ; also A esch. Sept. 164.
8 —10 ‘W o o d g ro w in g in a sacred are a. I f you pay th e god the price,
you ca n use th e w ood for sacred , p ro fan e a n d unclean p u rp o ses.’
S acre d p u rp o ses: sta tu e s, sacrificial fires. U n clean purposes: the
b u rn in g o f u n c le a n o b jects, especially corpses, p erh a p s too use in
<h th o n ic sacrifices. F u rth e r possibilities are suggested by the
P y th a g o re a n ru le, Ia m b i. VP 154, a g a in st using ced ar, lau rel, m yrtle,
o r cy p ress for clean sin g the b o d y o r th e teeth , since th ey should be
k ep t for h o n o u rin g th e gods.
O n th e p ro te c tio n o f sacred w ood see p. 165 above. T h e entirely
c o m m e rc ia l a p p ro a c h is u n u su al, b u t w ith o u t know ing the ch a racter
o f th e ‘w ood g ro w in g in a sacred a r e a ’ it is rash to d raw conclusions
a b o u t C y re n a e a n lib eralism (L u z za tto ). U nless ‘the g o d ’ is the
re le v a n t god in ea ch case, it looks as if this rule relates specifically to
I he s a n c tu a ry o f A pollo.
1 1 -1 5 ‘C o m in g from a w o m an a m an , if he has slep t w ith h er by
n ig h t, ca n sacrifice [w herever? w henever?] he wishes. I f he h as slept
w ith h e r by d ay , h e ca n , after w ash in g [ ] go w h erev er he
w ishes, ex cep t to \two lines missing]
See p p . 74 ff. on su ch rules. H ere in terco u rse by n ig h t req u ires no
p u rificatio n . It sh o u ld be em phasized th a t the sacrifice h ere m entioned
is n o t in te n d ed to efface the pollution o f intercourse, as Ferri, L uzzatto,
a n d Sokolow ski assu m e. T h is is n o t a tte ste d as a function o f sacrifice;
th e case en v isag ed is th a t o f a m an w ho Wants to sacrifice b u t has
rec en tly h a d in te rco u rse. R esto ratio n o f th e lim iting clause ‘except to
336 Miasma
. . is q u ite u n c e rta in . A specific sa n c tu a ry w as p ro b ab ly nam ed.
M a a s th o u g h t o f sh rin es in general, b u t such severity w ould be
u n p a ra lle le d .
1 6 —2 0 ‘T h e w o m an in ch ild b ed shall p ollute th e house. \gap\ she
sh a ll n o t p o llu te [th e p erso n w ho is o u tsid e the house(?)], unless he
com es in. A ny p erso n w ho is inside shall be polluted for th ree days,
b u t sh a ll n o t p o llu te an y o n e else, n ot w herever this perso n goes.’
I n 16 I d iv erg e from F ra en k e l’s text, r e a d in g λεχώι (nom inative); for
th e form see B uck, §111.5. T h e clue to the sectio n ’s articu latio n , as
V o g lia n o saw , is th e 0 ’ in 17, w hich is u n m istak a b le in th e photo­
g ra p h . T h e p rev io u s section becom es lucid if we accep t O liverio’s
έ ξ ό ρ ο φ ο ν , as in th e tra n sla tio n above. (F o r -Ö- after έξ- cf. E. Risch,
Wortbildung der homerischen Sprache2, Berlin, 1974, 225, 188.) B ut the
w o rd is n o t a tte s te d , an d th e ξ read by O liverio n o t visible on the
p h o to g ra p h . O th erw ise th e stone m u st have said so m eth in g like ‘she
s h a ll n o t p o llu te a roof, unless she com es u n d er it.’ B ut it seems
p re fe ra b le to m ak e the m obile p a rty , liable to com e u n d e r a roof,
so m e o n e o th e r th a n th e m o th er.
2 1 —5 ‘T h e re is hosiä in respect o f th e A k am a n tia for everybody, both
p u re a n d p ro fan e. E xcept from th e m an B a ttu s th e lead er an d the
T r ito p a te r e s a n d from O n y m asto s the D elp h ian , from anyw here else,
w h e re a m a n d ie d , th e re isn ’t hosiä for one w ho is pu re; in respect of
sh rin e s th e re is hosiä for e v e ry b o d y .’
A vexed section; an y tra n sla tio n is ten d en tio u s. T h e different possi­
b ilitie s a re b est ex pressed by, respectively, L a tte an d V o g lian o /M aas.
T w o m a in difficulties are th e re a d in g in 21, an d the articu la tio n of
2 2 - 3 . In 2 1 , Κ Α Μ Α Ν Τ ΙΩ Ν is c e rta in ,a n d a p reced in g trace is visible
w h ic h p ro b a b ly belongs to an A. W ilam ow itz in te rp re ted α( ϊ) κα
μ α ντίω ν, su p p o sin g ac cid en ta l om ission o f the i; L a tte ä κα μαντίων,
M a a s Ά κ α μ α ν τίω ν (from Α κα μ α ντίο ν, sh rin e o f the A kam antes: a neat
p a ra lle l in c o n stru c tio n to 2 5 ).W ilam ow itz tran slated : ‘I fth e re is hosiä
o f seers, th e re is it for everybody . . .’. Sim ilarly L atte: ‘W h atev er hosiä
o f seers th e re is, th e re is for e v ery b o d y ’, the p o in t being th a t co n su lta­
tio n of o racles w as only p erm itted a t ce rtain tim es. (L a tte suggested
th a t μ α ντίω ν m ig h t sta n d for μαντείω ν, o ra c u la r shrines, b u t V ogliano
Appendix 2 337
p o in ted o u t th a t μ α ντή ιω ν w ould b e expected.) B u t th e p o stu lated
■m lission o f th e v erb in b o th co n d itio n al (or relative) an d m ain clause
un-ms im p o ssib ly ab b re v ia te d , an d the expression ‘hosiä o f seers’ is
u n co n v in cin g . I f seers are rejected here, the in te rp re ta tio n o f the
lollow ing lines w ith reference to to m b -o racles (de S an ctis a n d others)
i ollapses. T h e A k a m a n te s, in tro d u c ed by M a a s’s in te rp re ta tio n , are
know n as re c ip ie n ts o f offerings in a sacred calen d ar from M a ra th o n ,
I.SC G 20 B 32; th e re as here they a p p e a r close to the T rito p a te re s, b u t
rt s L atte p o in ts o u t th e o rd e r o f offerings in the M a ra th o n text is by
i .ilen d a r, a n d so th e ju x ta p o sitio n need n ot be significant. O f their
II.1 lu re n o th in g c e rta in is know n; as ‘the u n tirin g o n es’ th ey m ig h t be
w inds (cf. L S J s.v. άκάμας, ά κάμα τος), b u t in the C y ren e law, if they
,ui- co rrec tly in tro d u c e d in to it, th ere is p erh a p s a c o n tra st w ith the
u n u s u a l use o f κάμνω = ‘d ie ’ in 24, w hich w ould m ake th em ‘u ndying
u n e s’ (cf. W ilam o w itz, Glaube, i, 309, n. 2). O n e o f th e A n ten o rid s,
w ho receiv ed cu lt a t G yrene, w as called A kam as, an d som e have
id en tified A k a m a n te s a n d A n ten o rid s (J. D efrad as, R E G 65 (1952),
299, G . C a p o v illa , Aegyptus42 (1962), 85); b u t, as V ogliano n o ted , it is
h a rd to see w h y A n te n o r’s second son sh o u ld have given his n am e to
ih cgens. T w o s u b s ta n tia l difficulties in M a a s ’s rea d in g w ere ind icated
by L a tte . (1) T h e p lu ral; d id th e G y ren aean s really h av e a series o f
■.hi ines o f th e A k am a n te s? A possible solution w ould be to suppose a
m a s o n ’s e rro r for Α καμάντω ν; or p e rh a p s Α κα μ α ντίο ν w as a generic
w ord in G y ren e m e an in g so m eth in g like ‘hero sh rin e’. (2) Some
c o n tra s t b etw een th e A k am a n tia a n d th e ίε ρ ά ο ΐ'25 will have to be
lo u n d , o r th e fo rm er provision could have been left to b e covered by
I lie la tte r. V o g lia n o suggests th a t th e A k am a n tia are ηρώα as opposed
Ι ο ιερ ά , b u t th a t in tu r n leaves th e c o n tra st betw een them an d lines
2 2 - 3 , w h ich also seem to tre a t ηρώα, obscure.
1’he a rtic u la tio n o f 2 2 - 4 d ep en d s on th e read in g a t the en d o f ‘23. If
.1 co n ju n c tio n ca n be in tro d u c ed there, they becom e a su b o rd in ate
( lau se q u alify in g 21, ‘th e re is hosiä for all . . . except th a t, from Battos
. o r an y w h e re else, w here a m an died, there is not hosiä for a pure
m a n .’ W ith o u t a conjunction, 2 2 - 4 becom es a n in dependent sentence,
w ith π λ ά ν m o d ify in g ά πα λλώ , as in the tran slatio n offered a t the start
o f th is sectio n . F o r κ α ί a t th e en d o f 23 th e re is no space; ή m ight seem
possible, b u t ac co rd in g to V o g lian o a n d O liverio, the only letter
c o m p a tib le w ith th e traces is t (a m istaken ad scrip t i o f a com m on
kind: see e.g., a t G yrene, six instan ces in SE G ix 4). I f this is correct,
th e tra n s la tio n offered ab o v e becom es inescapable. O n an y view, the
re la tio n b etw een out' άνθρώπω Βάττω an d άπαλλώ p resen ts a further
difficulty. It is g en erally agreed th a t ‘from the m an B a tto s’ m eans
‘from th e (to m b of) th e m an B atto s’. W e know from P in d ar o f his
to m b in th e m a rk e t-p la c e a t G yrene (Pyth. 5.93). E x cavation has
338 Miasma
rev e ale d tw o ro u n d tom bs in th e ag o ra , th e larg er o f w hich contained
tw o d is tin c t a lta rs ; W ilam o w itz acco rd in g ly assigned th e larg er tom b
to a jo in t c u lt o f ‘B atto s a n d th e T rito p a te re s’, the sm aller to the
m y ste rio u s ‘D elp h ic O n y m a s to s’ ( Kyrene, Berlin, 1928, 9 n. 1; cf.
O liv e rio , Fig. 12, F. C h a m o u x , Cyrène sous la monarchie des Battiades,
P aris, 1953, 1 3 2 ,2 8 5 -7 , w ith P late 7.1, an d fu rth er references in P. M.
F raser, Ptolemaic Alexandria, O xford, ii, 1097 n. 508). B u t the conjunc­
tio n o f to m b s a n d ‘an y o th e r p la ce w h ere a m an d ie d ’ is illogical, as
m e n d o n o t d ie in th e ir to m b s (unless heroic tom bs are envisaged as
bein g sited a t th e p la ce o f d e a th ). A reference to ac tu a l places o f death
( th o u g h a c c e p te d by L atte) seem s o u t o f place in th is context. W e are
p e rh a p s d e a lin g w ith a b rach y lo g y for ‘from an y o th e r place, w here is
b u rie d a m a n w h o h as d ie d ’.
O n th e T rito p a te re s Sokolow ski gives bibliography.
‘T h e r e is hosiä o f the sh rin es for ev ery b o d y ’ is norm ally taken to
m e a n (cf. W ilam o w itz): ‘It is hosion, religiously inoffensive, for every­
b o d y to a p p ro a c h th e sh rin e s’, everyone has free access to them . It is
in itia lly te m p tin g to in te rp re t 2 2 - 4 as in d ic atin g places w here it is not
hosion for a p u re p erso n to go. T h e lines w ould exclude ‘the p u re ’ from
to m b s, a n d th u s from h ero cult, w ith th e exception o f those tom bs
s itu a te d in th e ag o ra itself, w hich m u st have been generally recognized
as a n e x c ep tio n to th e n o rm al p rin cip le th a t tom bs pollute; cf. p. 42,
a n d on th e sim ila r restrictio n s im p o sed on C o a n p riests against
a p p ro a c h in g g rav es, o r a house o f d e a th , p. 52. T h e difficulty is th at
th e se lines, in c o n tra st to 21 a n d 25, speak not o f ‘hosiä o f’ b u t ‘hosiä
fro m ’: ‘F ro m a p la c e w here a m a n d ied th ere is n o t hosiä for a pure
m a n .’ M o st e d ito rs have sim ply glossed over this αυιό (V ogliano even
tr a n s la te s ‘al luogo dove u n o è m o rto . . . non è d a ta facoltà di
a c c o s ta rs i’), a n d if it is tak en serio u sly baia becom es vague: there is
n o t hosiä for a p u re perso n (com ing) from a to m b - to do w hat? (O nly
th o se w ho rea d μαντίων in 21 can p ro v id e an an sw er, cf. L a tte and
B uck).
A p o ssib le so lu tio n w ould be to in te rp re t baia n o t in term s of
freed o m o f access, b u t o f freedom o f co n su m p tio n o f sacrificial offer­
ings. Άκαμαντία a n d ιερά w ould b e ch a n g ed from places to offerings:
‘E v ery o n e m a y s h a re in offerings m ad e to the A k am an tes . . . th ere is
n o t th e rig h t o f e a tin g from th e to m b o f B attu s for a p u re m an . . .
ev e ry o n e m ay sh a re in offerings m a d e to th e go d s.’ T h is w ould find a
p a ra lle l in th e fu rth e r C o a n restric tio n for priests on παρ’ ήρωνα εσθεν
( LSCG 154 A 22, 156 A 8); an d on th e p u zzlin g p h ra se όσίη κρεάων in
Hymn Horn. Merc. 130 see H. J e a n m a ire , REG 58 (1945), 6 6 -8 9 , w ith
B en v en iste, ii, 1 98-202 . B u t the lack o f an y explicit reference to eating
is su rp risin g .
T h e referen ce to a class o f ‘th e p u re ’ is u n iq u e (as is the earlier
Appendix 2 339
d iv isio n o f p u re -p ro fa n e -u n c le a n ). T h e p u re m u st be p riests an d
o th e rs w h o a re , for w h ate v er reaso n , su b ject to te m p o rary hagneiai.
2 6 —31 ‘I f h e sacrifices u p o n the a lta r a victim w h ich it is not
<u sto m a ry to sacrifice, let h im rem ove th e rem ain in g fat (?) from the
a lta r a n d w ash it o ff a n d rem ove th e o th e r filth from th e sh rin e an d
i.tke aw a y th e ash es (?) from the a lta r a n d th e fire to a p u re spot, and
th en let h im w ash him self, p u rify th e sh rin e, sacrifice a full grow n
■mimal as p e n a lty , a n d th e n let h im sacrifice as is c u sto m a ry .’
< )n b re a c h o f relig io u s rules as a p o llu tio n see p. 144. N o te th a t in this
<ase th e illicit sacrifice po llu tes th e sacrificer as well as th e shrine.
12 ‘A m a n is b o u n d as far as his b ro th e rs ’ ch ild ren .’
This s ta n d s in iso latio n , se p a ra te d by paragraphi from w h a t precedes
.ind w h a t follow s. (T h e first paragraphes is u n m istak a b le on the
p h o to g rap h , a n d refutes attem p ts to m ake 32 run on from 31.) 6]κώχιμος
is th e o n ly su p p le m e n t th a t fits th e space. T h e g ro u p o f relatives
ex te n d in g to th e sons o f b ro th e rs is a fam iliar one (cf. L a tte ), b u t the
rea so n for its in tro d u c tio n h ere is very obscure. A law in D em . 43.58
(cf. H a rris o n , i, 128 n. 2) a p p a re n tly specifies th a t all th e heirs, an d
n o t ju s t th e d ire c t d e sc e n d a n ts, o f a m an w ho dies ow ing m oney to a
g o d s h o u ld b e atimoi u n til they p ay th e d eb t. T h a t suggests a p lausible
kind o f c o n te x t for o u r reg u latio n , especially in view o f w h a t follows;
b u t th e v ag u en ess a n d brev ity o f th e C y ren a ea n law m ake it seem
a lm o st like a fra g m e n t o f a fuller code.
33—72 ‘I f a g ro w n m a n is su b ject to a tith e, h av in g purified him self
w ith blo o d , h e sh all p urify the sh rin e; afte r being sold in th e m a rk e t­
p la ce for th e m o st th a t he is w o rth , he sh all first sacrifice as a p enalty
befo re th e tith e a fully g row n victim , n ot from th e tith e, a n d th e n he
sh a ll sacrifice th e tith e a n d carry it aw ay to a p u re spot; otherw ise, the
sa m e m e a su re s w ill b e necessary. E very o n e w ho sacrifices sh all b rin g
a vessel. I f a [boy] is p o llu ted unw illingly, it’s sufficient for h im to
p u rify h im s e lf a n d a p e n a lty is n ’t necessary. I f he is p o llu ted w illingly,
he sh a ll p u rify th e sh rin e a n d sacrifice first as a p en a lty a fully grow n
victim .
(43) I f p ro p e rty is su b ject to a tith e, he (the ow ner) shall assess the
340 Miasma
value of the property, purify the shrine and the property separately,
and then sacrifice first as a penalty a fully grown victim, not from the
tithe, and then sacrifice the tithe and carry it away to a pure spot.
Otherwise, the same measures will be necessary. From the property,
as long as it is subject to a tithe, no one shall make funerary offerings
nor shall he bring libations until he pays the tithe to the god. If he
brings libations or makes funerary offerings, after cleansing the
temple of Apollo he shall first sacrifice as a penalty, according to his
offence, a fully grown victim.
(53) If a man subject to a tithe dies, after they bury the man he (the
heir?) shall place whatever he likes on the tomb on the first day, but
nothing subsequent to that, until he pays the tithe to the god, and he
shall not sacrifice nor go to the tomb. They shall assess him (the dead
man) for the most that he was worth, being a partner to the god. After
purifying the temple of Apollo and the property separately, he (the
heir) having first sacrificed as a penalty a fully grown victim not from
the tithe, in front of the altar, shall sacrifice the tithe in front of the
altar and carry it away to a pure spot. Otherwise, the same measures
will be necessary.
(63) If a m an subject to a tithe dies and of the children who are left
some live and some die, having assessed the [dead children?] for the
most that they are worth he (the heir) shall purify the temple of Apollo
and the property separately, sacrifice first the penalty of the grown
m an before the altar, and then sacrifice the tithe before the altar. As
for the living descendant, having purified himself he shall purify the
shrine separately; after being sold in the market place, he shall
sacrifice the penalty of the grown man, a fully grown animal, and then
he shall sacrifice the tithe and carry it away to a pure spot. Otherwise,
the same measures will be necessary.’
Notes on the translation. ‘The same measures will be required’ is
expressed by either τών αυτών δησεί or δησεΐται. For the latter, other
renderings have sometimes been offered, but for impersonal ôehm
with a genitive see PI. Men. 79c, Dem. 18.145, LSJ s.v. όεΐ, III. In 40
and 41 μιάι is probably middle or passive, ‘incurs pollution’, in view of
the absence of an object and the apparently passive use of the related
future/aaaei in B 3 (contrast activeμιανεί in A 16), but an active sense
is perhaps not inconceivable, ‘pollutes (the shrine)’. The regulation,
1Everyone who sacrifices shall bring (take away?) a vessel’ (39), seems
misplaced here. T he force o f‘carrying away to the pure’ in 38, 46, 62,
71 is obscure. It can scarcely mean ‘(thereby) restore (things) to
purity’, as Vogliano suggests. The sacrifice ‘before the altar’ has
sometimes been thought to be especially appropriate to the case of a
tithed man who has died, but προβώμιος has no intrinsic funerary
application (for the word cf. Eur. Ion 376 and at Cyrene SEG ix 345),
Appendix 2 341
and the failure to specify this form of offering in earlier sections is
perhaps mere carelessness (YVilamowitz).
Various forms of tithing were familiar in Greece, several of them
especially associated with Apollo, who was όεκατηφόρος (RE s.v.
Apollo, 47). A part from the purely secular use of the tithe as a form of
tax or rent, there was the common practice of dedicating to a god a
tenth of first-fruits, plunder, or the product of any enterprise; aitio-
logical stories told of humans sent to Delphi as tithes by their con
querors or even, in time of plague or famine, by their own people, and
a penal tithing, of disputed character, was threatened against the
medizers in 479 (cf. How/Wells on Hdt. 7.137. 2, Parke/VVormell i,
51-5 , H. W. Parke, ‘Consecration to Apollo’, Hermathena 72 (1948),
82—114; also Diod. 11.65.5). Epigraphic evidence for the payment of
tithes to Apollo is quite exceptionally abundant at Cyrene: see SEG ix
68, 78, 80, 84, 87 f., 94, 100, 302-17, and (partly reproducing material
from SEG) nn. 35-42, 49, 133-42, 151 f., 248-52 of the Supplemento
Epigrafico Cirenaico (Annuario della Scuola Archeologica di Atene, n.s. 23-4
(1961-2), 219—375). Beyond the fact that these are payments by
individuals, the character of the tithe is impossible to determine, but
apparently, in contrast to our text, it is not merely sacrificed, but in
part at least goes to pay for the inscription, ‘X dedicates his tithe to
Apollo.’ For tithes paid from spoils at Cyrene see SEG ix 76 f., Suppl.
Epig. Cir. 132a. It does not seem that the institution of our inscription
corresponds exactly with any of the familiar forms. It is for us bafflingly
obscure, because the law assumes knowledge of the institution’s
general intent, and confines itself to procedural formalities and
special cases (although even in these respects, lor all its verbosity, it is
annoyingly unexplicit and incomplete). The form of tithe that is
assumed is individual, and not, as in the case of the Medizing cities,
collective; on the other hand, the reference to ‘penalties’ seems to
show that we are not dealing with the ordinary individual tithe
voluntarily offered, but with an obligation that is imposed as a
punishm ent. If the tithe is itself a punishment, the requirement of a
penal sacrifice in addition to it is perhaps surprising; but the possi­
bility of being subject to a tithe without being subject to a penal
sacrifice is not envisaged, which is hard to explain on the view that
voluntary thank-offerings are in question. It is perhaps conceivable
that the opening provision of 30 is a brachylogy for ‘Ifa grown man is
under a tithe (and incurs pollution while under it) ’ (cf. 40-2); but if
the stone really omits such vital specifications, it is beyond inter­
pretation.
342 M iasma
No direct indication is available about the offences through which
such tithing was incurred. The recurrent references to purification,
both of the offender and of the shrine, suggest that they were pollu­
tions of some nature; sacrilegious pollutions, which could include
alm ost any breach of religious rules, are the most likely kind (cf. pp.
144 ff. above). T he rules for the youth who ‘incurs pollution’ perhaps
support this view (but see below for M aas’s view). It is sometimes
suggested (e.g. by Wilamowitz) that the tithed man requires purifica­
tion because the position of sacred debtor (κοινός έών τωι θε&ι 58)
is, through the contagiousness of the sacred, intrinsically polluting.
But on this view one might rather expect the purification after the
tithe has been sacrificed; and it offers no convincing explanation for
the purification of the shrine (Wilamowitz suggests, implausibly, that
the mere presence of the sacred debtor pollutes it). It is not very likely,
though perhaps conceivable, that these purifications are not the
response to a specific pollution, but mere preparations for the solemn
act of sacrificing the tithe (for ‘sacrificing off’ a tithe cf. Xen. Ages.
1.34, idem .,Hell. 3.3.1,4.3.21).
T he section on the youth, 40-2, perhaps, as was noted, provides a
clue. In40, the choice of reading is between α]νηβος and ε]νηβος (the v
is certain; cf. schol. Theocr. 8.3, you are ανηβος till 15, ενηβος hence­
forth) . ανηβος gives a contrast with ήβατάς, 34; with ενηβος we are left
to wonder about the consequences of the pollution of an ανηβος.
According to Maas, the section has nothing to do with tithing, but
refers, with its contrast between involuntary and voluntary pollution,
to wet dream s and masturbation. (He sees this reference, and the
severity of the penalty, as an argument in favour of ενηβος.) It has
perhaps been misplaced here because of a desire, observable from
32—82, to divide the inscription into ten-line sections. The point about
the ten-line sections is correct (cf. Vogliano 289), and, in a context
that treats o f ‘tenths’, startling; is this conscious number-symbolism,
and if it is, what parallels are available at this religious level? The
sexual interpretation ofμιάι is linguistically plausible (cf. p. 76 n. 9),
but its implications here are too extraordinary to be accepted without
modification. Can we really imagine a Greek sacred law imposing
such penalties - indeed any penalties - on young men for such
offences in ordinary circumstances? If the sexual interpretation is
correct, we would have to assume some specific and restricted appli­
cat on (temple servants, boys preparing for a specific ritual, or the
like). O f this, however, the text offers no hint.
Vogliano thinks these boys are δεκατοί like the ήβαταί. The dif­
ference in their situation lies only in the concession made.for involun­
tary pollution. Although it is not stated, we understand that the youth
voluntarily polluted must also sacrifice a tithe (note the prefix in
Appendix 2 343
προθνσεί, 42). In favour of this it may be said that 33 seems to
envisage the possibility of δεκατοί who are not full grown. But it is
perhaps more plausible that the ανηβος has committed an offence
which would have rendered him δεκατός but for his age (cf. V.
Arangio-Ruiz, Persone efamiglia nel diritto deipapiri, Milan, 1930, 12 n.
2); as it is, he gets off more lightly. This would confirm that ‘tithing’ is
a consequence of pollution; but the character of the pollution would
remain unclear.
T he assessment of a tithed m an’s value by ‘selling him in the
m arket’ is an extraordinary and unparalleled procedure. The com­
m entators without exception assume the sale to have been a fictional
one; and if it is not, to whom do the remaining nine-tenths of the tithed
m an’s value fall? But it is hard to see how interested participation and
fair bidding at a fictional sale could be ensured. (It has been suggested
that P. Oxy. 716.18 ff provides a parallel for the assessment of a man’s
value by mock-sale, but there seems no difficulty about seeing the sale
there as genuine.)
T he restrictions of 48-53 cause surprise by their position. Latte
cited Aeschin. 3.21, a law forbidding officials who were υπεύθυνοι to
dedicate goods, and Gaius, Dig. 44.6.3, ‘rem de qua controversia est
prohibem ur in sacrum dedicare’. O ur text, however, is more limited,
referring only, it seems, to mortuary offerings. Ferri and Wilamowitz
took this as an extreme case, ‘not even for a pious duty, much less for
anything else’. But the rule, which names a penalty, is oddly specific if
so. It is tempting to suppose that it belongs somewhere in the follow­
ing section, on the obligations of the dead dekatos’ heir. If it is correctly
placed, the point is perhaps to protect Apollo’s goods from even
indirect contact with funerary pollution.
O ther serious difficulties, less relevant to this book’s theme, can
only be mentioned here without full discussion. In 33 ff. we hear of a
m an who is tithed, in 43 ff. of property that is tithed; the procedure in
the two cases is distinct. In 58- -9, however, the heir of a man tithed in
his person is required to purify the inherited property, presumably in
preparation for sacrificing a tenth of it to the god. Thus here the
personal tithe seems to extend to the property too. Probably, there­
fore, the true distinction is not, as 33—48 initially imply, between a
tithe on person and one on property, but between a tithe on person
plus property and on property alone (Vogliano; Luzzatto, however,
believes th at the tithe always covers both person and property.)
The possibility ‘if a man who is dekatos dies’ is envisaged twice (53,
63). In the second case he is imagined as having produced several
children, some still living and some now dead. What of the first?
Vogliano, alone among scholars who have explicitly considered the
problem , argued that the dekatos left a single child as heir. Commoner
344 M iasma
has been the view that 53 ff. treat the case of the dekatos who dies
without direct heirs. Its proponents (de Sanctis, Luzzatto, Koschaker,
Abh. Sächs. Ak. 42 (1934), 53-5, and particularly V. Arangio-Ruiz,
loc. cit.) point out that in 60-3, in contrast to 69-72, there is no talkol'
the heir assessing his own value and sacrificing a tithe of it, or
undergoing personal purification. The difference, they argue, implies
a qualitative distinction: the heres externus must pay the dead m an’s
tithe, but only direct descendants become polluted in their own
person. Thus the two sections treat extreme cases (no direct heirs/
various direct heirs, living and dead), on the basis of which proper
responses to interm ediate situations can be worked out, if they are not
obvious anyway. The argument is ingenious; but it is impossible to be
certain that the omission of a provision in a particular part of this law
proves it to have been inapplicable there.
Amid all this uncertainty, the positive information that emerges is
disappointingly slight. The most conspicuous feature is, perhaps, the
rigorous protection of the god’s rights. The condition of being dekatos
is, unless effaced, hereditary, and seems to extend to all the tithed
m an’s offspring, since the surviving son is required to pay tithes also
for his dead siblings. With the tithe, pollution too is inherited; the son
requires purification from his father’s taint. Here we have one sub­
stantial gain; this is virtually the only instance that can be quoted of
an inherited pollution that has recognized legal effects (cf. pp. 204 ff.,
and p. 185 on the ‘Gottesurteil von M antinea’).
73—82 (Fragmentary beyond restoration)
Wilamowitz remarked that the only certain fact about the content of
these lines is that it had nothing to do with either what preceded or
w hat followed; but even that negative conclusion is perhaps too
positive.
B passages illustrating such obligations have been collected by com­
2 —8 ‘. . . but she herself shall not be under the same roof as her
husband nor shall she incur pollution until she comes to Artemis. Any
woman who, without doing this, voluntarily incurs pollution, after
purifying the temple of Artemis shall sacrifice in addition as penalty a
full grown anim al, and then shall go to the sleeping chamber. But if
she incurs pollution involuntarily, she shall purify the shrine.’
Appendix 2 345
9 —14 ‘A bride m ust go down to the bride-room to Artemis, when­
ever she wishes at the Artemisia, but the sooner the better. Any
woman who does not go down [shall sacrifice in addition, or
(Calhoun) shall not sacrifice] to Artemis [what is customary at the
Artemisia]; not having gone down, [she shall purify the shrine] and
sacrifice in addition [a full grown animal as penalty.]
T he detailed logic of 9 -1 4 is quite uncertain. The repetition â όε καμή
κατένθηι . . . μή κατεληλενθνία looks almost like a product of con­
flation in drafting. Calhoun suggested his οϋ θυσεί in CP 29 (1934),
345 f.
15—23 ‘[A pregnant woman] shall go down to the bride-room to
Artemis . . . shall give to the bear (a priestess) the feet and head and
skin. If she does not go down before giving birth she shall go down
with a full grown animal. She who goes down shall observe purity on
the seventh and eighth and ninth, and she who has not gone down
shall observe purity on those days. But if she incurs pollution, she
shall purify herself, purify the shrine and sacrifice in addition as
penalty a full grown anim al.’
Ferri and Wilamowitz could make little of this section. De Sanctis and
M aas independently suggested that we are dealing with successive
stages in a w om an’s career, and the ritual obligations attendant on
them: 1-8 pre-m arital, 9-14 the new bride, 15-23 the expectant or
new mother. T he interpretation is almost certainly correct; indeed,
obscure though it is, this section illustrates as effectively as any text
the way in which it is through ritual performances that social change
is articulated and expressed. The performances are here required not
merely by custom but by an actual religious law. Before marriage the
girls must go to the ‘sleeping-room’ for the προνύμφιος ύπνος (cf.
Callim. fr. 75.2 with Pfeiffer), after it to a Nympheion in the precinct
of Artemis (on its probable identity see F. Chamoux, op. cit., 315-19;
its position gives special relevance to the verb ‘go down’). Various
m entators (see too L. Deubner, ‘Hochzeit und Opferkorb’, JD A I
(1925), 210-23): Suda s.v. άρκτος ή Βρανρωνίοις: έψηψίσαντο οϊ
'Αθηναίοι μή πρότερον σννοικίζεσθαι άνόρί παρθένον εί μή άρκτεύσειε
Xf] θεώ; Plut. Amat. Narr. 772b, Suda s.v. προτέλεια, pre-marriage
sacrifice to nymphs; schol. Theocr. 2.66, appeasement of Artemis by
those about to marry, or pregnant for the first time; Apostolius 10.96,
346 Miasma
Suda s.v. Λνσίζωνος γυνή, maidens before first intercourse dedicate
girdles to Artemis.
T he penalties and purifications in this section strongly recall those
of the tithed man. It becomes tempting to turn back to the dekatos and
try to interpret his condition in similar terms, as an obligation
incurred by young men at a particular stage in life rather than the
consequence of an offence. The temptation is strengthened by the fact
that δεκατεϋω could be used in Attic as an equivalent to άρκτεύω
(Didymus ap. Harpocration s.v. όεκατενειν)·, the world of the
C yrenaean girls recalls that of the Attic ‘bears of Artemis’, and we
even find in Cyrene a bear priestess (B 16, cf. SEG ix 13.12, Chamoux,
op. cit., 319). But it proves impossible to carry this interpretation
through. The dekatos can be of any age (he might die, leaving
children); and, in contrast to the girls of face B, no form of behaviour
seems to be available to him by which he will avoid the need for penal
sacrifice.
Sev eral details in the section are elusive. What, for instance, is the
pollution of 3 -8 , that may be incurred either voluntarily or involun­
tarily? M enstruation is involuntary only; intercourse may be either,
but one would expect μη εκοισα to mean ‘accidentally’ rather than
‘against her will’. If the pollution is indeed sexual, it is remarkable
that the act performed in private should make necessary a purifica­
tion of the temple of Artemis. In '2\,μιάι probably does refer to sexual
pollution, in view of the contrast \Ν\ύ\άγνενσεϊ in 19, and this supports
the sexual interpretation earlier, (άγνεΰω is not confined to sexual
purity, but without further specification this is the most natural
reference.) Unfortunately the point of this hagneia ‘on the seventh,
eighth and ninth’ is uncertain. Some commentators feel that it should
precede the ‘going down’ of 15, 18, and 19; accordingly Maas, by his
supplem ent in 15, located the ‘going down’ on the tenth. But ‘20 f.
imply rather strongly that the occasion for ^jping down (whether
performed or not) preceded the hagneia. Are the seventh, eighth, and
ninth perhaps the days leading up to the tenth-day ceremony alter
birth? (cf. p. 51).
2 4 —7 ‘Ifa woman throws out {i.e. miscarries), ifit is distinguishable
(i.e. if the foetus has recognizable form), they are polluted as from one
who has died, but ifit isn’t distinguishable, the house itself is polluted
as from a woman in childbed.’
O n the pollution of miscarriage see p. 50 n. 67. There is doubt about
the sense o f‘the house itself’ here: cf. p. 50, and G. M. Calhoun, CP 29
(1934), 345 f., whose reading αυτά in 26 I adopt.
Appendix 2 347
28 ‘O f Suppliants’
(a new heading in large letters)
T he suppliants are helpfully discussed b y j. Servais, BCH 84 (1960)
112-47.
2 9 - 3 9 ‘Suppliant from abroad (or, Visitant sent by spells). If a
(‘suppliant’) is sent to (or, against) the house, if (the householder)
knows, from whom he came to him, he shall name him by proclama­
tion for three days. If (the sender of the suppliant) has died in the land
or perished anywhere else, if (the householder) knows his name, he
shall make proclam ation by name, but if he doesn’t know his name (in
the form) “o man (anthröpos), whether you are a man or a woman” .
Having made male and female figurines either of wood or of earth he
shall entertain them and offer them a portion of everything. When you
have done what is customary (the change to secondperson appears random),
take the figurines and the portions to an unworked wood and deposit
them there.’
We have here a triangular relationship between a ‘suppliant’, the
m an to whose house the ‘suppliant’ was sent, and the sender of the
‘suppliant’. Little beyond this was clear when the text was first
published, but subsequently the second half at least has been con­
vincingly interpreted. Radermacher (Anz. Akad. Wien, 1927, 182 IT.)
and de Sanctis independently explained that the recipient of the
suppliant was required to propitiate the sender, and, since he could
not do so literally, acted symbolically instead, by forming figurines to
represent the sender, entertaining them to dinner with portions of
food and drink, and so establishing a magically effective bond of
guest-friendship. (On kolossoi, figurines, cf. M /L 5.44 (Cyrene), E.
Benveniste, Rev. Phil. 58 (1932), 118-35, G. Roux, REA 62 (1960),
5-40 , V ernant, Pensée, ii, 65-78). After the dinner the figurines and
the food offered to them were to be carried out to an unworked wood,
beyond the sphere of human activity, where their presence could do
harm to none. As parallels for such symbolic entertainment, lectisternia
and thèoxenia (not unknown at Cyrene, cf. J . Defradas, R E G 65 (1952),
282-301 on Pind. Pyth. 5. 83-6) can be quoted; Radermacher pointed
out that Byzantine sailors used to seek good passage by entertaining
an effigy of Saint Phokas to dinner, and Servais quoted magical
parallels (Papyri Graecae Magicae, ed. K. Preisendanz, Leipzig, 1928,
1.40, 86, 4.54-70). Even more closely relevant, because it combines
Ihe two aspects of propitiation and expulsion, is the familiar Greek
practice of sending out ‘meals for Hecate’ (p. 30 above).
348 Miasma
Several different situations are envisaged in the law —the sender
may be known or unknown, dead or alive - but the necessary ritual is
not spelt out in each case. It is unclear whether the symbolic enter­
tainm ent is always required, or only in the case specifically described
where the sender is dead and unknown. The real difficulty, however,
concerns the nature of the ‘suppliant and his sender. The common
assum ption is that he is a foreigner seeking incorporation in the
com m unity of Cyrene, and that the regulation reflects a time when
this could only be achieved by admission, as a suppliant, in a private
household. The ceremony described is the formal transfer of potestas
over the suppliant from the foreign sender to the Gyrenaean recipient.
To this interpretation there are serious objections. Rituals of this kind
relating to potestas are unattested in Greece; it is very hard to see why
the recipient should be ignorant of the name and sex of the sender,
when the suppliant himself could readily enlighten him, or why the
sender should remain so dangerous that his expulsion to an unworked
wood was required. A female sender of suppliants is also surprising. A
different approach was offered by H. J . Stukey, ‘The Gyrenaean
Hikesioi’, CP 32 (1937), 32-43. He pointed out that the έπι - com­
pounds, of which there are three in our passage, are typical of the
language of magical attack. For επακτός see p. 222 n. 79; for επαγωγή
see LSJ s.v. επαγωγή, 4b; for έπιπέμπω (Lys.) 6.20, Xen. Cyr. 8.7.18,
Pl. Cri. 46c, Dem. 24.121, LSJ s.v. έπιπέμπω, 2, and for later evidence
Fr. Pfister, Wochenschrift f . klassische Philologie 29 (1912), 753—8; for
έπηλυοία see Hymn Horn. Cer. 227 f., Merc. 37; and for the ‘house’ as
target of magical attack see Theophr. Char. 16.7, Orph. Hymn 37.7 f.,
on a higher level Aesch. Ag. 1188-90, and probably a Sophron mime
(see p. 223 above). Each of the έπι- compounds used in the inscription
may, certainly, bear a non-magical sense, but the collocation is
striking, and the further έπί in έπι τάν οικίαν strongly suggests that the
action is an aggressive one. Indeed, it is not cle^r that a non-aggressive
sense ο ι έπιπέμπω exists, except for that o f‘send in addition’ (which is
perhaps the force of SIC3 93.7, 273.24, cited by Servais). The ίκέσιος
επακτός, therefore, is not a human suppliant but a demon sent
against the house, as Hecate sometimes was, by an enemy. The
suggestion has not been taken seriously, chiefly, no doubt, because its
author rashly tried to transform the second and third suppliants, who
are palpably hum an, into further spirits. But the same idea had
occurred to M aas (Epidaurische Hymnen, Halle, 1933, 139, cf. Hesperia
13 (1944), 37 n. 4 = Kl. Sehr. 202 n. 4), and is very likely to be
correct.* It explains the characteristic language, the need for pro­
* O n first read in g the inscription, H ugh Lloyd-Jones independently thought of the
sa m e in te rp re ta tio n . I am grateful to him for persuading me th at it is right.
Appendix 2 349
pitiation and expulsion, and the possibility of a dead or unknown
sender. The householder may suspect a particular enemy of working
magic against him; or he may infer the fact of an attack from a series of
misfortunes, but not know whom to ascribe it to specifically. Victims
of murder, at least, could send out demons from the grave against
their killers (Xen. Cyr. 8.7.18, έπιπέμπω), and the possibility doubt­
less extended to other cases; Pherecrates, fr. 174 ολαγώςμε βασκαίνει
τεθνηκώς, may play with such ideas.
Maas saw a difficulty in the application of the term ίκέσιος to a
spirit; but this could be euphemism, and etymologically a suppliant is
anyway merely a ‘comer’ (for the connection of the suppliant and the
stranger see p. 181 above). A prostropaios may be an innocent human,
or he may be an avenging demon (p. 108 n. 13 above). Stranger
perhaps would be the intermingling of human and demonic suppli­
ants in the same law, but even this is not inconceivable; in each case,
an alien intrusion into the familiar world must be countered with due
ritual procedures. (The demonic interpretation could be maintained,
but the suppliant restored to humanity, by translating ίκέσιοςέπακτός
as ‘bewitched suppliant’; but this extension in the application of
επακτός is unattested.)
4 0 -4 9 ‘Second suppliant, initiated or not initiated, having taken
his seat at the public shrine. If an injunction is made, let him be
initiated at whatever price is enjôined. If an injunction is not made,
let him sacrifice fruits of the earth and a libation annually for ever.
But if he omits it (?: cf. Buck), twice as much next year. If a child
forgets and omits it, and an injunction is made to him, he shall pay
to the god whatever is told him when he consults the oracle, and
sacrifice, if he knows (where it is) on the ancestral tomb, and if not,
consult the oracle.’
T he ‘injunctions’ are presumably oracular, in view of the juxta­
position in 46 f.
Another regulation that is almost wholly obscure. Doubt centres on
the meaning οί τελέω, τελίσκω. A telesphoria had an important place in
the cult of several gods at Gyrene (cf. Servais, op. cit., 137 η. 1; add
SEG ix 65, 68 f., and Supplemento Epigrafico Cirenaico, 144—6, 252). It
seems normally to be a procession, but the word could also be used
with reference to initiation, Callim. Cer. 129. As in the case of the
tithes, it is hard not to suspect a connection between the cathartic law
and the institution revealed by the other epigraphic evidence; but
once again it is impossible to advance beyond suspicion. If the
350 Miasma Appendix 2 351

reference here is to initiation, it is of a kind otherwise unknown, since
it may entail permanent, and even hereditary, sacral obligations
(perhaps it does so in all cases; or Latte may be right that, ‘if an
injunction is m ade’, the initiate fulfilled his obligations by a once and
for all paym ent). The connection between supplication and initiation
is also unfamiliar. It has been thought that ‘initiation’ here confers
admission not merely to a sacral but also a social grouping, and that
the suppliant is a refugee seeking reception in Cyrene (Latte). But the
uncertainties are too many for speculation to be profitable. The most
useful discussion is that by Servais, op. cit.
5 0 —55 ‘T hird suppliant, a killer. He shall present the suppliant to
the [ ] cities(?) and three tribes. When he announces that (the
killer) has arrived as a suppliant, he (someone else?) shall seat him on
the threshold on a white fleece, [wash] and anoint him; and (they?)
shall go out into the public road, and all shall keep silent while they
(the killer and his sponsor?) are outside, obeying the announcer . . .
(fragmentary: there is a reference.to ‘sacrifices’).’
The words αντοφόνος, αύτοφόντης, αντοκτόνος, αύτοσφαγής are used in
tragedy of kin-murder, and occasionally of suicide. But that does not
settle the question of αντοφόνος in the inscription. Tragedy is much
occupied with kin-murder, and these words are used to indicate it
adjectivally, not independently, in a context that clearly determines
the meaning. Even in tragedy, αντοκτόνος once means ‘killing with
one’s own h an d ’ (Aesch. Ag. 1635), and that is perhaps the natural
origin to ascribe to the proper name Autophonos of Horn. II. 4.395. Of
the two αντο- compounds used of killing that have some non-poetic
existence, αύθεντης can mean ‘murderer’ (Soph. 0 7 Ί0 7 ), and αύτόχειρ
often has that sense, sometimes (PI. Leg. 872a, Xen. Hell. 6.4.35) but
not always with special emphasis on the actual physical performance
of the killing. (The weakening of sense from ‘killer with one’s own
h an d ’ to ‘killer’ is very natural.) With the possible exception of PI.
Resp. 615c, which is anyway of elevated style, it seems nowhere to
have the sense of ‘kin-killer’ in prose. Three renderings of αντοφόνος
are therefore possible: kin-killer, actual killer, killer. The most plaus­
ible, because the most general, is probably the third.
M asson, op. cit., questions whether -πολίαν in 51 can derive from
πόλις; but cf. άστνπολία, άλλοπολία.

The suppliant appears to have a sponsor, who ‘presents him as a

suppliant’ (άφικετενω, a new word, but cf. Ζευς άφίκτωρ, Aesch. Supp.
1) to the [ ] πολίαν καϊ τριφνλίαν. These bodies are otherwise
unknown (but for the Zeus Triphylios of Euhemerus see RE 10 A
347); the mention of a threshold in 52 suggests that the precise
reference here is to a building, or buildings, in which they met. The
significant point is that the triphylia must in some sense represent the
whole state; the purification of the murderer is thus a matter of public
concern. O n the further ritual details the commentary of Latte was
definitive. He noted the significance of the threshold (the murderer
may not yet enter), the fleece (cf. p. 373), and the ‘announcer’ (cf. Eur.
I T 1208—10, 1226, εκποδών ό’ανδώ πολίταις τοϋά’ έχειν μιάσματος,
and the heralds who preceded Roman flamens to stop artisans from
working in their presence). ‘Silence’ in this context normally belongs
to the killer himself (p. 371), but it can scarcely be applied to him
here, and the extension is very natural. In 52-3 I read νί]ζεν rather
than λενκ[ιμονί ]ζεν (Oliverio); a rule about dress is quite out of place at
this point, and the verb is ill-formed (Masson, op. cit.)
This suppliant is probably a refugee from abroad, since he requires
presentation to the triphylia. He is designated αντοφόνος, but the exact
force of this is uncertain (for discussion of the amo- compounds used
of killing cf. Fraenkel on Aesch. Ag. 1091, F. Zucker, Sitz. Leipz. 107
( 1962), n. 4, 22—4, and references in p. 122 n. 69 abo've on authentês).

Appendix 3:
Problems concerning ‘Enter pure trom . .
Requirements in Sacred Laws
( 1) Purity άπό λεχους.
In four sacred laws purity is required από λεχους. (In LSA 51.6 read
[τ]ε[κούση]ς with Fraser rather than λ[εχοϋ]ς.) In three there is formal
am biguity between the accentuation άπό λεχοϋς (λεχώ, woman in
childbed) and λεχους (λεχος, bed). In the fourth, LSS 115 B 27, λεχοϋς
is certain, as λεχος would give λεχεος in Cyrenaean. This unambigu­
ous Gyrenaean case is a strong argument in favour of reading λεχοϋς
throughout; so too is the analogy with the commoner way of expres­
sing the same regulation, άπό τεκούσης vel. sim. (cf. references in p. 50
n. 67 above).
Iϊ λέχους were read, it would still refer to childbirth, as intercourse
is spoken of differently in sacred laws (P. M. Fraser, ‘An Inscription
from Cos’, Bulletin de la Société Archaeologique d’Alexandrie 40 (1953),
35—62, at p. 45). There remains, however, a possibility of real differ­
ence in meaning between the two readings. With λεχοϋς, the rule
refers to those who come into contact with a new mother, but says
nothing explicit about the mother herself. With Λεχους, interpreted as
‘after (contact with) childbirth’, the laws might be taken as regulating
the access of:
( 1) the m other herself
(2) both the mother and those in contact with her
(3) those in contact with the mother.
O ption (3), of course, maizes λεχους effectively synonymous with
λεχοϋς; and it can practically be discounted, as it is clear from the
parallels cited earlier that the way to express the outsider’s ritual
status is άπό τεκονσης or λεχοϋς, just as ‘to approach a woman in
childbed’ is έπι λεχώϊεναι (Theophr. Char. 16.9).
In LSS 91.15, άπά λεχους y' λεχώ κα, λεχοϋς is clearly right: ‘After
contact with a woman lying in, 3 days; the woman herself, 21 days’.
For this way of expressing the contrast between the mother and others
cf. LSCG 124.5—6,7—8. Only in the implausible sense (3) could λεχους
be adm itted here. For LSCG 171.16-17, άπά λεχους καί έγ
δία(φθ)οράς (?)άμέραςδέκα, άπό γυναικόςτρεΐ[ς], the arguments were
well presented by Fraser in the first publication (op. cit.). The period
Appendix 3 353
of ten days would better suit the mother herself (cf. p. 52 n. 74), but is
not inconceivable for those who have contact with her; cf. LSS 54.5,
seven days, and perhaps the Arcadian text about to be discussed. The
shrine is intended for men, line 16, and it is to them that the purity
rules are addressed (αγνόν 15, άπό γνναικός 17). We are thus left
again with the possibility o fλέχους only in sense (3).
T he new sacred law from an Arcadian cult of Isis is the one where
λέχους has most plausibility (BCH 102 (1978), 325). Masculine and
feminine endings alternate in this text surprisingly, and the adjective
in the relevant rule is feminine, lines 5—6: άπό μέν λεχ[ο]υς έναταίαν.
In 8—9 we have άπό ôè των φυσικών (menstruation) εβδομαίαν. If the
latter, as seems almost inevitable, refers to the menstruating woman
herself, ought not the former to refer to the mother? (The intervening
regulation, 6 -8 , on miscarriage, is itself too ambiguous to help in
either direction.) This would impose the reading Λέχους, interpreted
in sense (1) or (2). Against this we may observe: (a) λεχους would be
quite isolated among sacred laws (but the possibility that the formula
was reapplied mistakenly here should be considered); (b) where the
ritual status of the mother is undeniably specified, it is not done in this
form; (c) the genitive after άπά is normally a specifically polluting
object or substance; (d) the feminine is explicable if we assume that
the law envisages the feminine birth-helpers. (This would mean that
birth-pollution is here conceived as a m atter o f‘touching’ rather than
‘entering the same roof, as the possibility of male pollution is not
considered.) It is, however, certainly more natural to refer the word to
the mother, if we accept that the following provision about menstrua­
tion concerns the woman herself only. Here, therefore, the internal
logic of the text on the whole supports λέχους-, but analogy still
strongly urges the claims of λεχοϋς.
(2) What constitutes contact in ‘Enter purefrom . . . ’laws?
W here purity from certain foods is required, or from intercourse,
there is obviously no difficulty. For death-pollution there are differen­
tial scales (above, p. 37 n. 17), but entering a house of death or
attending a funeral creates some pollution even without physical
contact with the corpse. For birth-pollution in the post-classical
period there are no conclusive arguments; men can incur it, but they
might touch the woman as well as enter the house. In the Cyrene
cathartic law, ‘entering the same roof had been the determinant
(Appendix 2). In Arcadia in the first century BC physical contact may
have been (above).
Special problems are presented by rules requiring purity from
abortion, m enstruation, defloration, and the like. (We have con­
sidered in the previous section the same problem in relation to birth-
pollution.) They could concern:
354 Miasma
(a) the woman only
(ib) the woman and those who touch her, during the period of
active contamination
(c) the woman and those who come into social contact with her,
during the same period.
In the case of menstruation, there are strong arguments for (a). In
LSCG 55.5, BC H 102 (1978), p. 325, line 9, feminine endings show
th at the person contaminated is a woman, and in LSS 119.13 a
m enstruation rule appears in the female but not the male section of a
law that appears to legislate for the two sexes successively (G.
Plaum ann, Ptolemais in Oberägypten, Leipzig, 1910,54-8). ΙηΖΛ^ΘΙ.Ιβ
a reference to menstruation should perhaps be restored; the rule,
whatever it was, was for women only. LSS 54, however, appears to be
addressed to men (it requires purity άπό γνναικός but not cut’ άνδρός)
but concludes άπό γυναικείων έναταίονς. Interpretation (c) is hard to
credit here, given the length of the exclusion, and even (b) surprising.
If the rule is indeed for men, we should perhaps adopt a moderated
form of (b) and understand ‘after intercourse during menstruation’.
But it is not inconceivable that women are envisaged; the masculine
would be by attraction, and the absence of a purity rule cbr ’ άνδρός an
omission due to the fact that the worshippers were primarily male.
In the classical period, abortion (spontaneous or contrived) cer­
tainly polluted according to principle (c) (p. 50 n. 67). There are
several later sacred laws for which (a) is impossible. LSS 119.5 is a rule
for men; unfortunately the period is lost, and we cannot see whether it
was forty days as for the mother in the same law (10). Where a
genitive of specification occurs, as in LSS91.11 άπö φθοράς γνναικός ή
κυνός ή δνου, and LS Λ 84.5, the rule is certainly not addressed to the
woman (or dog, or ass), or it would be phrased γυναίκα άπό φθοράς.
In Z.S'5'91.11 (c) is preferable to (b) - in other cases we cannot choose-
as there would be little occasion for manhandling the dog or donkey in
these circumstances. LSCG 171.17 seems to be addressed to men
(abov e), BC H 102 (1978), p. 325 lines 6—8 to women, but not neces­
sarily the mother alone (above), while LSS 54 is, as we have seen,
am biguous. LSCG 55.7 and 139.12 are quite indefinite. LSS 119.10 is
apparently addressed to women, but has been preceded by a rule for
men. Thus interpretation (a) is nowhere certain, while it is excluded
in several cases. This need not necessarily mean, however, that it is to
be excluded everywhere. It is quite plausible that at different times
an d in different places the way in which pollution was diffused should
have varied.
It is often thought that LSCG 124.5-6 (2nd c. B C ?) referred to
abortion, and specified 40 days impurity for the mother and 10 for
those in contact with her. If this were correct, we would see how the
Appendix 3 355
40-day term was extended in later laws from the mother herself to all
those involved. But, as E. Nardi notes (Eranion Marid.ak.is, cf. below,
63), the m other who has aborted would not normally be called à
τετόκοισα (contrast LSS 119.11 ). He suggests a reference to exposure
of the child, but the most polluted person should then be ‘he/she who
puts it out’ and not the mother. Stillbirth would be a possibility, but it
is not elsewhere envisaged separately in sacred laws. Perhaps the
argum ent from τετόκοισα is too nicely drawn.
About pollution ‘from a m aidenhead’ (LSCG 139.18, LSS 91.12)
there is no evidence; both parties were probably affected.
(3) Pollution άπά φθοράς vel sim.
(The relevant laws are printed and discussed by E. Nardi, ‘Antiche
prescrizioni greche di purità cultuale in tema d ’aborto’, Eranion in
honorem G. S. Maridakis, Athens, 1963, i, 43—85, with an addendum
in Studi in Onore di E. Volterra, Milan, 1971, i, 141—8.)
T he period of pollution for miscarriage in classical sacred laws
corresponds either to that for birth or death (p. 50 n. 67 above). In
later sacred laws the period leaps up to a typical 40 days (so LSS 54.6,
probably 91.11, 119.10, LSA 84.5, LSCG 55.7, LSCG 139.12 (from
‘abortive drugs’); BCH 102 (1978), p. 325 lines 6 -8 has 44 days,
LSCG 171.17 probably 10; the relevance o f LSCG 124.5—6 is unclear,
cf. above). Thus in most cases it far exceeds the periods specified after
a birth or death. Scholars have often thought that these regulations
concern ‘procured abortion’, and that the extended period of impur­
ity reflects a new ethical condemnation of the practice (cf. LSA 20.20;
there is disagreement whether this condemnation is indigenous or
im ported). The objection sometimes advanced (e.g. Ant. u. Chr. 4
( 1933—4), 18 f.) that these laws concern ritual impurity, which has no
connection with guilt, is not serious, as in just this kind of sacred law
the duration of sexual pollution is influenced by moral considerations.
T he real difficulty is that all the Greek words in question (φθορά,
διαφθορά, εκτρωσμός, and others) indicate merely the fact of the
expulsion of the foetus but not the cause (J. Ilberg, ARW\2> ( 1910), 3).
T hus it is hard to see the justification for saying that in a particular
law the reference is to miscarriage and not procured abortion, or vice
versa. We cannot say that φθορά in sacred laws had acquired a
restricted sense o f‘procured abortion’, as LSS 91.11 is a clear counter­
case: άπά φθοράς γνναικός ή κυνός η δνου ημε. μ'. Dogs and donkeys do
not have procured abortions, and the suggestion that φθορά here
m eans something quite different, sexual assault (Nock apud
Sokolowski ad loc., cf. Nardi, Studi Volterra, loc. cit.), is a desperate
one, as it goes against the regular use of the word in sacred laws - and,
in the case of the latter two victims, opens the door to a world of quite
unsuspected pollutions. The only one of the laws, therefore, that
356 M iasma
specifically concerns procured abortion is LSCG 139.12, where there
is explicit reference to abortive drugs; in all the other cases, any
wom an who entered within 40 days of an involuntary miscarriage
would be violating the rule. It is however plausible that it was moral
revulsion against procured abortion that rendered abortion of any
kind so im pure and threatening, and that the legislators had procured
abortion chiefly in mind. In two laws, significantly, abortion rules are
associated with rules that make exposure of the child a pollution (LSS
119.7, 14 days; LSA 84.3—4, 40 days).
Appendix 4: Animals and Food
T here was no category of impure animals in Greece. Aristotle ack­
nowledges that there are some which ‘we dislike looking a t’, but
designates them ‘most lacking in honour’ rather than unclean. The
distinction of animals in terms of honour appears in two further places
in Aristotle, and Isocrates says that Egyptians reverence ‘animals
despised among us’.1 (The classification by honour extends to plants,
as Callim achus’ fourth iambus shows.) Nor was there a category of
im pure food.2 A culture can, without recognizing such a category
explicitly, be strictly selective about what it regards as edible (dogs
are inedible for the Englishman, without being unclean); but the
Greeks were ready to eat more kinds of flesh than many peoples, to
judge from the Hippocratic writer On Regimen, who lists, as the
‘anim als that are eaten’, cattle, goats, pigs, sheep, donkeys, horses,
dogs, wild boar, deer, hares, foxes, and hedgehogs. For most of the
more surprising items in this list independent evidence is available,
although some of them turn out to be despised food that all but the
poor would avoid.3
Despite this, exclusion of a particular animal, or abstinence from a
food, might sometimes be required by religious rule.4 The obvious
example of the former concerns the dog. It was denied entry to the
Athenian acropolis, to Delos and other sacred islands, and, no doubt,
to many sacred places besides.5 In other respects, too, the dog’s status
was degraded. It was a symbol of shameless behaviour, and occupied
the most ignominious place in the sacrificial system, being exploited
in purifications and as an offering to the marginal Eileithyia, the
1 A rist. Poet. 1448b 12, De An. 404b 4, Part. An. 645a 15; tsoc. Bus. '26.
2 1 h is is n o t confuted by the fact th at there are special 'p u re loaves’ (H dt. 2.40.3,
A lexis, fr. 220. 10, A th. 149e; cf. R h inthon, fr. 3 K aibel) and 'p u re piglets’ (B runcau,
286 1.); the sense o f 'p u re ' here is anyw ay u ncertain.
3 H ippoc. Viet. 2.46 (6 .5 4 4 -6 L.), w hich refutes Porph. Abst. 1.14. Kor lox cf.
A n an iu s, fr. 5.5 W est, M nesim achus, fr. 4.49; dog, A r. Eq. 1399, H ippoc. Morb. Sacr.
142.18 J . , 1.14 G ., Alexis, fr. 220.4; ass, Ar. Eq. 1399, w ith R. A. N eil’s note, X en. A nab.
2.1.6 (d ire need), Pollux 9.48 (sold, interestingly, in a special place).
4 Cl. W äch ter, 76 -1 15 (m uch o f the m aterial irrelevant).
5 H. Scholz, Der Hund in der griechisch-römischen Magie und Religion, Berlin, 1937. 7 f.
(m ain text Plut. Quaest. Rom. 2 9 0 a -d ); ad d X en. Cyn. 5.25, LSS 112 IV B.
358 Miasma
dishonoured Ares, and the tainted Hecate.6 On the other hand, there
is no reason to think that any of these negative connotations attached
to the sacred dogs of Asclepius at Epidaurus, or those kept in other
temple precincts. (Even the epic tradition that knew the man-eating
dogs of the I l i a d also had place for Odysseus’ faithful Argos.) As we
have seen, there was no danger in eating dog flesh, and the practice
seems to have been common.
Abstention from foodstuffs is attested in various forms. There are a
certain num ber of local taboos on the consumption of particular
species, usually of no great economic significance;7 the Seriphians, for
instance, threw back any lobster they caught, saying they were
‘Perseus’ playthings’. O f permanent abstinence from particular foods
by devotees of particular cults there is no trace. We know only of a
few, not very rigorous, limitations imposed on priests.8 Temporary
abstinence, however, in preparation for specific ceremonies does seem
to be occasionally attested for early Greece, just as actual fasting is.
Late sources tell of a proclamation that was made to Eleusinian
initiates to abstain from certain foods, and something similar is
recorded about another festival of Demeter, the Haloa. Combination
of these sources gives as the forbidden foods: house-birds, beans,
pom egranates, apples, eggs, ‘egg-laying animals’, the meat of animals
that died naturally, and various kinds offish.9 The attestation is late,
but the A tthidographer Melanthius, writing at some date between
350 and ‘270 BC, mentioned one of the forbidden fishes, the red mullet,
in a work on the Eleusinian Mysteries, and it is hard to see why, unless
in connection with the ban.10 Restrictions almost certainly existed,
therefore, before the Roman period, although the list may well have
been extended. For a different cult, we have similar evidence, again
not incontestable but very strong, for the fifth century. A fragment of
C ratinus’ play about the incubation oracle of Trophonius runs: ‘And
not to eat any more the red-skinned Aixonian triglë, nor the Irygôn nor
the dread m e la n o u r o s '." The lines are quoted without context, but we
6 S cholz, op. cit., 14-2 2 ; cf. J. M . Redfield, Nature and Culture in the Iliad, Chicago,
1975, 193-202.
7 N ilsson, G G R 212 f.
8 C heese lor priestess o f A th en a Polias at .Athens, fish for some Poseidon priests: see
A rb e sm a n n , 72 f. E ur. Cretans h. 79 A ustin is questionable evidence.
9 Eleusis: A rb esm an n , 76 f. H aloa: schol. L ucian 280. 22 ΙΓ. R abe (cited D eubner,
(>1.5). I accept th e universal view th a t these are m erely tem porary restrictions (cf. the
reference to ‘those being in itiated ’, not 'those w ho have been in itiated ’, in Ael. NA 9.51,
65), b u t know no conclusive evidence; devoted individuals m ight have extended them.
10 326 F G rH fr. 2 ap. A th. 325c.
11 Fr. 221 (? cf. A ristophanes, fr. 23, from th e Amphiaraus). P urity a t L ebadeia:
G inouvès, 344 n. 4, an d generally A rb esm an n , 9 7 -1 0 2 on pre-m antic dietary
restrictions.
Appendix 4 359
know that the cult of Trophonius imposed several requirements of
purity, and two of the three fishes mentioned are among those banned
at Eleusis. There is a reference to ‘refraining from’ (particular foods)
‘for three days’ in Aristophanes’ Peace, although it is impossible to tell
what specific cult he is parodying.12
Even a fifth-century attestation leaves open, of course, the formal
possibility of Pythagorean influence. Pythagoreanism embodied a
considerable num ber of dietetic precepts, not for temporary observ­
ance only but as parts of a permanent way of life, and some have
supposed that this was the source of all the dietary rules of cult.13 It is
perhaps not very plausible that an Eleusis to which food restrictions
were alien should have adopted the eccentric regulations of a south
Italian sect that was best known to Athenians as a butt of comedy;
but, in view of the Orphic reinterpretation of the Eleusinian eschato­
logy, the possibility cannot quite be excluded. Pythagorean influence
on the cult of Trophonius seems at first sight even less likely, but here
too we know too little of the religious climate of fifth-century Lebadeia
to say absolutely, ‘This cannot be.’ The alternative hypothesis is that
Pythagoras adopted his rules, or many of them, from cult14or the fund
of popular magico-religious beliefs that lay behind the cult rules. As a
development it seems more natural that a temporary restriction
should have been taken over and extended by a sect that sought
especial sanctity than that rules, the point of which lay largely in
being permanently observed, should have been trimmed down for
cultic use in drastically attenuated form; but the Greeks seem later to
have performed ju st such trimming in their reception of oriental cults,
reducing perm anent abstention from fish, for instance, to a three-day
preparation for the festival.15 The hypothesis of borrowing from
philosophy is therefore irrefutable, though implausible. The same
may be said in the case of the purifiers of On the Sacred Disease, who told
their patients to abstain from various meats, fishes, and birds. 16They
do not seem to have mentioned the bean, abhorred of Pythagoras.
12 151, cl. 162—3, n oted by L atte, Kl. Sehr. 26. F or C hrysippus such rules a re p art of
tra d itio n a l religion, P lut, de Stoic. Rep. 1044f. T h e epigraphic evidence for food hagneiai
begin s only in the 2nd cen tu ry B C , an d initially in relation to oriental cults; t h e influence
of p o p u la r n eo -P y th ag o rean ism seem s subseq u en tly also to becom e p erceptible (A. D.
N ock, H S C P 63 (1958), 4 1 5 -2 1 = Nock ii, 8 4 7 -5 2 ). See LSCG 55, 95, 139; L S S 54, 59,
108; LSA 84; CR Acad. Inscr. 1916, 263 f.; Altertümer von Pergamon, viii. 3, ed. C. H abicht,
B erlin, 1969, p. 168; B C H 102 (1978), p. 325. References to food hagneiai in magical
p a p y ri in L a n a ta , 54, η. 177.
13 L obeck, 190, Z iehen, 150. O n the P yth ag o rean rules see B urkert, L S 180-5, M .
D etien n e, 'L a cuisine de P v th ag o re’, Archives de sociologie des religions, 29 ( 1970), 141-62.
14 D. L. 8.33.
15 See e.g. M en . fr. 754 in contrast to L SS 54.
16 H ip p o c . Morb. Sacr. 142. 1 6 ff.J., 1.13—16 G ., well discussed by L an ata, 5 3 - 60.
360 Miasma
Passages in comedy offer some idea of the non-gastronomic
grounds on which, in a secular context, Greeks might reject particular
foods. In several places large deep-sea fish are spurned because they
are ‘m an-eating’. 17 The objection is already implicitly present in
Homer, where fish are seen as uncanny creatures, hostile to man,
lurking in the depths ready to devour the flesh of shipwrecked
sailors.18 This is perhaps the reason why they are excluded from the
heroic diet (as also, in the main, from cult), even though it is clear
from the similes that fishing was familiar to Homer’s audience.
Though the argum ent can scarcely be pressed, it is interesting to note
this pre-Pythagorean evidence for an ideal diet that is very selective,
particularly as it is to species of fish that the earliest evidence for
abstinence at Eleusis relates. Another mark held against animals as
food was the practice o f‘eating excrement’.19 It was too common to be
an absolute disqualification, but it is perhaps not a coincidence that
the purifiers of On the Sacred Disease told their patients to abstain from
dog, pig, and goat (as well as deer), the three domestic animals that
were commonly charged with scatophagy. By consuming such ani­
mals, one becomes a vicarious ‘m an-eater’ and ‘dung-eater’ oneself.
H um an flesh and dung were, of course, the supremely impossible
foods for a man. Cannibalism is analogous to incest,20 while ‘dung-
eater’ is an expression used of a man who will stop at nothing, and
more loosely as one of those insults that derive, like ‘temple-robber’,
‘m urderer’, and ‘mother-sleeper’, from the most degraded or pollut­
ing acts.21 In their literal form these were pollutions of the imagina­
tion only, since no Greek was tempted by either diet; but, as we have
seen, the taint could attach vicariously to other foods. O ther objec­
tions that are brought against foodstuffs are that they are 'food for
corpses’ or ‘for Hecate’, and, sometimes in connection with this point,
that they are anaphrodisiac.22
Such explicit interpretations of the religious rules as are available,
none of them certainly early, are based on no consistent principle. It is
characteristic that Aelian, referring to two of the Eleusinian rules,
states that the initiate shuns the dogfish as unclean, because it gives
birth through its mouth, but spares the red mullet as a mark ofhonour
17 A n tip h an es, fr. 68.12, 129.6, cf. Alexis, fr. 76. 1 -4 ; m ost explicitly the didactic
A rc h e stra tu s ap. A th. 163d, 3 10e (pointing o ut th at all fishes are ‘m an-eating’ and not
som e only); cf. too P in d ar, fr. 306.
18 See H . F rän k el, Die homerischen Gleichnisse, G öttingen, 1921, 8 6 -8 , esp. 87 η. 2.
19 E p ic h arm u s, fr. 63, cf. (m ud) Philem on, fr. 79.19. T h e w riters on fish often allude
to ‘m u d -e a tin g '. F or scatophagous anim als (and m en) see J . H enderson, The Maculate
Muse, V ale, 1975, 192 -4 .
20 See p. 98.
21 See S a n d b a c h on M en. Sam. 550.
22 P lato C om icus, fr. 173.19, A ntiphanes, fr. 68.14, A m phis, fr. 20, E ubulus, fr. 1 I.
Appendix 4 361
lor its exceptional fertility, or its services to man.23 The same ambi­
guity is found in the Pythagorean tradition. It was generally agreed,
although detailed interpretations varied, that it Was as an abomina­
tion that Pythagoras banned the bean from his table.24 But the white
cock was sacred (to the sun or Men) and should be spared for that
reason.2S As to fish, on the one hand sacred fish (obviously those
sacred to the Olympians) were not to be eaten, ‘for men and gods
should not have the same privileges any more than masters and
slaves.’26 O n the other, one should abstain from two particular species
because they ‘belong to the chthonian gods’ (and were therefore
im pure).27 The participants in one of Plutarch’s table conversations
discuss Pythagoras’ rejection of fish, which they take to have been
total. Three possibilities are canvassed; he held fish in honour for
their silence; he regarded them, inhabitants of the deep, as wholly
alien to man; or he felt that man had no right to eat inoffensive
creatures that he neither tended nor fed.28 Iamblichus sums up the
am biguities of the tradition when he states explicitly that Pythagoras
banned such foods as were indigestible, or alien to the gods, or, on the
contrary, sacred to the gods and so worthy ofhonour, or, finally, liable
to interfere with the purity, moderation, or mantic powers of the
soul.29 The same ambivalence must have been already present within
the oldest form of explanation, that by myth. Among the forbidden
Eleusinian foods Demeter had good cause to abhor the pomegranate
because of its use to trap Persephone, but she herself was the ‘apple-
bringer’.30 Their sacrificial victims, too, gods either ‘loved’ or ‘hated’.
We are near once again to the ‘primitive confusion of the sacred and
unclean’. But, though the ancients might doubt whether a particular
forbidden food was sacred or unclean, they never supposed it to be
both; and nothing suggests that any of the forbidden foods were
invested with a fearsome and ambiguous sanctity.
To offer an alternative explanation of most of these restrictions is,
23 N A 9.51, 65. ‘H o n o u r’ lor m allet also in Plut. De soll. an. 983Γ. In Anth. Pal. 7.406
(T h e o d o rid a s 14 G o w /P ag e) the mystes ‘loves’ p o m egranate, apple, and m yrtle (with
obscen e d o u b le m ean in g ).
24 B u rk ert, L S 183—5, w ith references in 183 η. 124.
25 e.g. D. !.. 8.34, Iam b i. Protr. 21, B urkert, LS 172 n. 47.
26 D. L. 8.34.
27 Iam b i. VP 109, Protr. 21.
28 Quaest. Com. 8.8 7 2 8 c- 730f. For o th er d iscussions of' hagneiai see ibid., 669e—67 le
(am b ig u ity ag ain ), De Is. et Os. 352Γ-354b, J u lia n Or. 5 .173d- 177c (am biguous). T he
Je w ish exegetical trad itio n by co ntrast is clear th at forbidden foods are unclean: sec e.g.
Philo De Spec. Leg. 4 .1 0 0 -1 3 1 , De Agr. 130 ff.; the Letter o f Aris teas, 144 ff.; S. Stein, Studia
Pa trist ica 2 {Texte und Untersuchungen 64), 141 ff.
29 VP 106.
30 S IG 3 1122.6, P aus. 1.44.3. Pom egranate: Hymn. Horn. Cer. 372 w ith Richardson.
362 Miasma
perhaps, an impossible task. Their precise extent in the early period is
uncertain; in the case of the Pythagorean rules the evidence is, at first
sight, actually contradictory, and it is as though through a mist that
we see the milieu in which they were observed. If the various restric­
tions are but a part of a more elaborate web of symbolic associations
and oppositions, only this torn-off fragment remains. It is by chance
only that we know that pomegranates, banned to the Eleusinian
initiate, were eaten ritually at the Thesmophoria.31 Another area of
our ignorance is the archaic ethnozoology of the Greeks, their native
ways of classifying plants and animals. Aristotle cannot tell us what
associations a particular species or natural product may have had for
the contemporaries of Homer.
But a few conclusions seem possible. The early vegetarians refused
to eat animals on the grounds that this was a form of cannibalism.32
Thus a central institution of society was subverted by appeal to that
society’s own values, since for all Greeks human flesh was the most
impossible of foods. (The further step of repudiating bread, the staple
of normal diet, as a form of dung was reserved for Herodotus’ mythi­
cal Ethiopians.33) An Orphic poet assimilated bean-eating too to
cannibalism : ‘It is no better to eat beans than your father’s head.’34 As
we have seen, fish-eating could be seen as a vicarious form of the same
offence. Even on the more restricted interpretation of Pythagoras’
teaching, by which he banned ‘womb’, ‘heart’, and ‘brain’, the associ­
ation of these parts with vitality seems significant; to eat them is
life-destroying. (The same can perhaps be said of the Orphic ban on
eating eggs.35) The possibility of consuming a restricted set of sacri­
ficial animals was justified by the claim that into these no human soul
could pass.35 The general purport of the cultic rules is different, since
there is no question in this case of radical revaluation of accepted
norms. 11 seems, however, that some of the same forms of assessment
are at work here too. The fish that was most commonly banned was
the red mullet (triglê), which fits neatly into the pattern. It ‘delighted
in polluted things,’ and ‘would eat the corpse of a fish or a m an’.
Blood-coloured itself, it was sacred to the blood-eating goddess
H ecate.37 It seems a symbolic summation of all the negative
·" D eu b n er, 58.
32 p. 305.
3:1 H d t. 3. 22.4; cl. J . P. V ern an t, i n j . P. V ern an t and M. D etienne, La Cuisine du
sacrifice en pays grec, Paris, 1979,239-49.
34 O F fr. 291, cf. P orph. VP 43.
35 W om b, h e a rt, a n d brain: e.g. A rist. fr. 194, Iam bi. Protr. 21. Eggs: O F fr. 291.
36 Iam b i. V P 85, B urkert, L S 182.
37 P ollution, corpses: Ael. NA 2.41. A nd H ecate: A th. 325 a —b, cf. A ntiphanes, fr.
6 8 .14fl/>. A th. 358f, 313b, a n d N ausicrates, fr. 1 -2 ap. A th. 296a. H ecate an d blood: R E
s.v. Flekate. col. ‘2 776. T h e triglê was banned a t Eleusis an d the H aloa, by the purifiers of
Appendix 4 363
characteristics of the creatures of the deep. (It is therefore disconcert­
ing that Aelian explains the Eleusinian rule as a mark o f ‘honour’ for
its good qualities.38) O f the other five fishes39 certainly or probably
banned at Eleusis one, the dogfish, was a ‘man-eater’ (as well as
abnorm al in its way of giving birth);40 two were, according to Iam b­
lichus, sacred to the chthonians,41 and, even if that information is
unreliable, alluded to the colours of blood and death in their names,
melanouros and erythrinos\ and one was sacred to Hecate, although not
perhaps for any other reason than that it was cheap.42 At Lebadeia,
the m urderous sting-ray,43 with which Telegonus slew Odysseus,
seems to have been banned. But there remains one Eleusinian fish,
the karabos or crayfish (as also one Pythagorean), for which no such
explanation is available.44 And it does not seem that the ban on
pom egranates, apples, and house-birds relates directly to their
natural properties, whether actual or ascribed. The connections here
are rather with Demeter’s own powers and mythology - her role as
‘apple-bringer’, and the tricking of her daughter with a pomegranate
seed.45
T he ambiguity in the ancient interpretations, by which animals are
spared for both good and bad qualities, should perhaps encourage us
to look for a structural explanation, whereby two tabooed extremes
m ark out an area of the edible in the middle. In some societies, it has
Morb. Sacr., a t th e oracle o fT ro p h o n iu s, an d to the priestess of H era at Argos (Ael. N A
9.65). Cf. F. J . D ölger, Ichthys, M ü n ster, 1922, ii, 316—330.
38 A bove, p. 361 n. 23.
34 E x h au stiv e discussion in F. J . D ölger, op. cit., ii, 330—58.
40 A rc h e stra tu s ap. A th. 163d, 310e (on κύων; b u t the point m ust apply to all the
sh a rk s, cf. R. S trö m b erg , Studien zur Etymologie und Bildung der griechischen Fischnamen,
G ö teb o rg , 1943, 104); Ael. N A 9.65. Cf. D ’A rcy W . T hom pson, A Glossary o f Greek
Fishes, O x fo rd , 1947, s.v. γαλεός.
41 Iam b i. VP 109, Protr. 21 .
42 T h e mainis: M elan th iu s, 326 FGrH fr. 2 ap. A th. 325c (from which the ban is
in ferred ). C h e ap n ess o f mainis·. e.g. P herecrates, fr. 56.
43 Trygon, C ra tin u s, fr. 221: cf. D ’A rcy W . T h o m p so n , op. cit., s.v.
44 Karabos·. even F. J . D ölger, op. cit., w as baffled. P ythagorean fish:akalïphë (sea
a n e m o n e ), A rist. fr. 1 9 4 .11 w as an am biguous creatu re, half-plant an d h alf-anim al (see
D ’A rcy VV. T h o m p so n , op. cit., s.v.), b u t we do not know th at such things m attered to
P y th ag o ras. T h e a p p a re n t association w ith ‘w om b’ in Ar. Lys. 549 m ay be m ore
relev an t. T h e b an on grey m ullet an d eel in Morb. Sacr. can be explained in either
d ig estiv e (cf. H ippoc. Int. 12, 7.198.16 L., Viet. 2.48, 6.548.18 L.) or religious term s: for
eels a n d H e c a te see A r. Lys. 700—2; for bo th fish together in a sacred pool, Ath. 3 3 le,
a n d for th e special sta tu s o f grey m ullet as the ‘fasting fish’, the p o lar opposite o f triglê
(b u t no so u rce m akes the connection), A th. 307 ff., D ’Arcy W. T ho m p so n , op. cit., s.v.
kestreus. T h e ir presu m ed spontaneous generation could also be relevant (Arist. H A 569a
2 2 - 6 , 570a 3 -2 4 ).
45 ‘A p p le-b rin g er’; SIG 3 1122.6, Paus. 1.44.3.
364 M iasma
been argued, there is a correlation between an animal’s edibility and
its ‘social distance’ in relation to man.46 The scale of social distance
might in a typical case extend from house-animals and labouring
anim als (inedible), via the domesticated but non-labouring animals
(edible) and game animals (ambiguously edible), to the wild beasts
(inedible). This scale, it is suggested, may be subconsciously
perceived as analogous to that which determines permissible
m arriage-partners, who have to be sought in the middle area between
close kin and strangers. (For Greece, we have already noted the
connection between sexual crime and monstrous food.) Although the
problems involved are complex,47 such an approach might help to
interpret the Greek sacrificial system. The sacrificial animals come
from the middle range, and the less domesticated victims, in particu­
lar the goat, are to some extent favoured by the less domesticated
gods.48 There is nothing alien to Greek thought in explaining sacri­
ficial practice in terms of an anim al’s moral relation to man. A
Bouzygean curse is supposed to have threatened those who sacrificed
the ploughing ox (by his services to man, he is brought too close for
edibility); but the pig must be eaten, because he pays back his nurture
in no other way than with his meat.49 Particularly within Pythagorea-
nism such considerations have importance; goat and pig, on one view,
are fit for sacrifice because they interfere with agriculture, but ox and
sheep should be spared.50 If these animals are too close to man to be
eaten, fish by contrast are too distant. Domesticated animals ‘live
with m en’ (synanthwpeuo), and in neo-Pythagoreanism this proximity
becomes explicitly a reason for sparing them.51 But, promising
though the structural approach may appear for the sacrificial system
in general, it can scarcely interpret the specific Eleusinian and
Pythagorean restrictions that we have been considering, unless
num erous subsidiary structures are to be introduced. It cannot deal
with the spasmodic but precise character of the Eleusinian
restrictions.
The same is true of the notorious bean taboo. Ancient explanations
certainly present the bean as a bizarre, polluted, and structurally
46 See the articles o f L each, T a m b ia h (Ethnography ), and H alveson cited above, p. 61
n. 101.
47 Two are p ra g m a tic (cost an d p alatab ility ). T h ree relate to the ideology o f the
sacrificial act, w hich d em an d s w illing subm ission by a live anim al, bloodshed, and
division o f m eat betw een m en an d gods. W ild anim als, fish, and birds are thereby
excluded.
48 Cf. S. now, B C H (1965), 199 f.
49 Bouzyges: Ael. VH 5.14. Pig: P lato Com icus, fr. 28.
50 A ristoxenus, fr. 25, 29a YVehrli, in terp reted Ov. Met. 15. 110—15; M. Detienne,
Archives de sociologie des religions, 29 (1970), 141—62.
51 P orph. Abst. 1.14. O n fish cf. p. 361.
Appendix 4 365
am biguous product, associated with sex, the cycle of birth and death,
and Hades.52 But it is not clear what principles of classification could
have caused the bean and it alone to be left in this invidious position.
A new possibility has emerged with the recent discovery that, for
individuals suffering from a particular hereditary enzyme deficiency,
faba vicia is indeed a poison.53 The phenomenon, known as favism, has
been reported from Sicily, southern Italy, and Sardinia. Pragmatic
interpretations of religious rules are often misguided; but the indi­
vidual instance needs to be judged on its own merits. A medical
explanation of the bean taboo appeals to no sage legislator with a
mysterious insight into hygienic rules unknown in his own day, but to
an easily observable causal connection that has lead to the avoidance
of beans by peasants today in the areas where favism is prevalent. The
real uncertainty concerns the occurrence of favism in antiquity. If it
did occur, in an environment where magical dangers were rife, the
natural classification for it would surely have been as a magical
danger. The significant conclusion, however, if the favism hypothesis
were correct, would be not so much that the taboo had a sound
prudential origin, but that it persisted so long and insistently after this
was forgotten. The bean came to symbolize polluting and threatening
elements in the Pythagorean world-view, all the more strikingly,
perhaps, because of its apparent innocuousness. The world-view,
however, cannot explain the original choice of the bean.
An im portant recent study of religious symbols, and food taboos, in
New Guinea explicitly rejects structural models, and emphasizes the
variety of their origins (where these can be discerned at all) and
complexity of their connotations. Unlike the elements in a computer
language, it is pointed out, which only have meaning as part of a
system, a religious symbol can derive significance directly from the
reality to which it relates.54 This is surely the kind of approach which,
in its details, the Greek evidence demands. We need only add that
‘abstaining from’ â food is, in itself, an important mode of differentia­
tion from everyday life, whatever the food may be. The content of the
restriction, though unlikely to be wholly arbitrary, is in a sense less
im portant than its context. The rules are found where the individual
is required to shed his profane self (actual fasting is found in the same
contexts): as a preparation for initiation or incubation, and as part of
the perm anent abnormality of the Pythagorean life.
52 Cf. M . D etienne, op. cit., 153 f.
53 R. S. B ru m b au g h an d J . Schw artz, ‘Pythagoras and Beans: Λ M edical E xplana­
tio n ’, Classical World 73 (1980), 421 f.
54 B arth , passim, esp. 12,161, Chs. 20, 23; cf. Burkert, S H 48.

Appendix 5: The Ritual Status of the
Justified Killer at Athens
This is an issue that has been discussed several times,1but it is worth
reconsidering briefly, as there has been a tendency to confuse actual
laws, some perhaps dating back to Draco, and the interpretations that
are offered by the orators who quote them. Taken by themselves, the
laws reveal a historical development that is reasonably clear, even if
its significance is uncertain.
T here was no formal category o f ‘justified homicide’ at Athens. We
find instead a variety of situations in which it is stated that killing
should not be liable to sanctions.2 In certain circumstances ‘it is
permissible to kill’, or ‘he shall not go into exile, having killed’, or the
victim shall ‘die without compensation’.3 In none of these early laws is
anything said about the ritual status of the killer. Either the lawgivers
thought the position on purification self-evident, or they knew nothing
of pollution, or they felt that it lay outside the province of the law. The
surviving portions of Draco’s law are equally unrevealing.
The religious issue first emerges in the more expansive style of a
decree of the late fifth century. The psephism of Demophantus,
passed after the rule of the Four Hundred, prescribes that any sub-
verter of the democracy should be ‘an enemy of the Athenians and
should die without compensation’.4 T hat is the secular aspect; no
prosecution can be brought for such a killing. There follows the
specification that the killer is όσιος καί ευαγής. The Athenians as a
whole are required to ratify the decree by oath, swearing to kill all
subverters of democracy and to consider those who did so δσιον . . .
και προς θεών και δαιμόνων, ώς πολέμιον κτείναντα τών ’Αθηναίων. Here
as elsewhere a killing is rendered non-polluting by classifying the
victim as an enemy.5 Parts of the decree seem to be taken over from
earlier legislation against tyranny,6 but even if the prescriptions about
1 See M acD ow ell, Homicide, 1‘28 f.
2 See C a lh o u n , (>(>-71.
3 L aw s ap. D eni. 23.28, 53, (>0; A rist. Ath. Pol. 16.10; Dem . 9.44. O n the sense ofatimos
in th e two la tte r see p. 204. S im ilar form ula in S I G’3 194.
4 Ap. A ndoc. 1.96—8.
5 See p. 113 n. 37.
6 See M . O stw ald , Τ Α Ρ Α 86 (1955), 103-28.
Appendix 5 367
purity date back to, say, around 500, they still seem to mark a
development from the earliest-attested law against tyranny, which
says nothing about the ritual issue.7-
Demosthenes’ comments on the laws of justified homicide provide
evidence for fourth-century views. The law quoted in 23.53 refers to
three different categories of killing, and Demosthenes glosses the
prescription τούτων ενεκαμή φεϋγειν κτείναντα in three different ways;
τούτον ώρισεν (ό νομοθέτης) ούκ άόικεΐν; καί τούτον είναι καθαρόν,
(τούτον) άθωον ποιεί. This is stylistic variation; the three expressions
all gloss the same phrase o f ‘Draco’ and all mean the same thing. In
20.158, which still refers to justified homicide, we find καθαρόν είναι,
in 23.60 άθώον είναι; in 37.59s Demosthenes uses έπιόείξας μή
καθαρόν for ‘prove guilty’ of involuntary homicide. The existence of
the Delphinion court shows, argues Demosthenes, that the ancients
believed certain kinds of killing to be δσιον; he glosses the old law και
άτιμος τεθνάτω as a case where killing is ευαγές and the killer is
καθαρός.9 It is clear that for him ‘not liable to punishment’ and ‘pure’
are synonymous.
It has, however, been suggested that, when applied in the specific
sense of ‘not liable to punishm ent’, katharos no longer conveys any
information about ritual status.10 ‘Pure’ might, therefore, mean ‘not
punishable (but im pure)’. This is too paradoxical to be readily
accepted. It is true that, as early as the Tetralogies, katharos often needs
to be translated ‘innocent’;11 but the legal sense is dependent on the
ritual one, since in a context of m urder ‘pure’ entails ‘innocent’, and
there is no evidence that the meaning ‘innocent’ could survive dis­
sociated from ‘pure’. For Plato, certain forms of killing do require
purification even though not liable to legal sanctions; but in these
cases it is only after purification that the killer becomes katharos.11
Porphyry’s claim that all killings of whatever kind required purifica­
tion should not cause us to distort the natural meaning ofDemosthenes’
tex t.13 Individuals may have chosen to undergo purification on their
own account,14 particularly, perhaps, in the kind of cases for which
Plato made it obligatory; but there can have been no legal require­
m ent. It is interesting that Plato’s Euthyphro considers it self-evident
that a killing is only polluting ifit is unjustified.15
7 A rist. Ath. Pol. 16.10.
8 = 38.22.
9 23.74; 9.44.
10 D o x o g rap h y a n d criticism in J . W . H ew itt, ΤΑΡΑ 41 (1910), 9 9 -1 1 3 .
11 A n t. Telr. 3 a 1,4; <5 10,11. Cf. Soph. OC 548.
12 Leg. 865b, 869a.
13 Abst. 1.9.
14 M acD ow ell, Homicide, 128 f.
13 PI. Eulhphr. 4 b - c .
368 M iasma Appendix 5 369

By the late fourth century, even a law could subsume the legal Athenians seem recently to have established.22 This was no banish­
aspect of justified homicide under the religious. Eucrates’ law of 336 m ent but, for a foreigner, a remarkable privilege (cleruchs were all
against subverters of the democracy, based largely on earlier legisla­ A thenian citizens); the scholion is thus virtually the earliest evidence
tion, says only that anyone who kills such a subverter should be hosios. for the extraordinary honour enjoyed by tyrannicides in the Greek
It is no longer necessary to state that his victim dies ‘without com­ world.23
pensation’. Outside Athens, the same is true of a third century decree
from Teos, which states simply that anyone who kills a rebellious
garrison-com m ander shall be ‘not polluted (miaros)’.16
T he situation in the early period remains uncertain. One cannot
prove that purification was not required, but there is no firm evidence
to suggest that it was, and the laws were clearly not so interpreted in
the fourth century. One may wonder too in what sense purification
could have been obligatory, if there was no explicit requirement in the
law .17
A scholion on Demosthenes makes a claim that, if correct, would
invalidate the whole preceding argum ent.18 It tells how the Athenians
granted ‘citizenship and a gift’ to certain individuals who had killed
‘M yrrhine the daughter of Peisistratus’, but forced them to live in
Salamis, because ‘anyone who had killed in any circumstances was
not permitted to set foot in Attica.’ The incident seems to be unattested
elsewhere, and the scholion contains one evident mistake: Myrrhine
was wife of Hippias, and thus not daughter but daughter-in-law of
Peisistratus.19 T he circumstantial details, however, suggest that the
ultim ate source for the story might even have been an inscription. The
scholiast’s interpretation is based on a very strong claim about the
pollution created by justified killing. But it cannot be correct; to say
nothing of the evidence already discussed, Phrynichus’ assassins were
invited to live in Athens, and Xenophon contrasts the impurity of the
norm al m urderer with the honours paid to the tyrannicide.20 This
evidence is not decisive for attitudes early in the fifth century, but
from the time of Draco it had been quite normal for the involuntary
killer to ‘set foot in Attica’ after a period of exile. The details fit into
place, however, if we assume that M yrrhine’s assassins were not
‘ordered to reside in Salamis’ but received a plot of land there as their
‘gift’,21 one of the shares in the Salaminian cleruchy that the

16 S E G x ii 87.11; SE G xxvi 1306.23-6.

17 F or a y e a r’s p urificatory exile in early Sicyon in consequence ot (justified)
fra tric id e see Nie. D am . 90 FG rH fr. 61 - scarcely trustw orthy evidence.
18 Schol. P atm . D em . 23.71 (B C H 1 (1877), 138).
19 T h u c . 6.55.1 ; D avies, 450.
20 I G I 3 102. 30—2 (M /L 85), cf. Lys. 7.4, 13.70; X en. H iero^A —b. R ew ards for killers:
M /L 43, A r. Av. 1072—5, D em . 23. 119. N o pollution in killing a rogue,.Dem. 19.66 (cf. 22 See M eiggs a n d Lewis on M /L 14.
Anth. Pal. 7.230,433, 531 on the S p artan m other). 23 H . F riedei, Der Tyrannenmord in Gesetzgebung und Volksmeinung der Griechen, S tu ttg art,
21 Cf. H d t. 8.11.3. 1937. ‘N o nemesis' from gods for tyrannicide, T heog. 1181.

Appendix 6: The Ritual of Purification
from Homicide1
T he fullest account comes in the fourth book of Apollonius’
Argonautica - a valuable source, given its author’s antiquarian learn­
ing. Jason had m urdered Medea’s brother by treachery in a temple,
with M edea’s connivance, and the Erinys had seen the deed (452-
76). Jason m utilated the corpse, and thrice sucked out its blood and
spat it back ‘in the way that murderers expiate treacherous killings’
(476—8 1),2 but Zeus was outraged at the crime and resolved that after
purification by Circe they should suffer endless troubles before reach­
ing home (557—61). As the Argo sailed on, the mast itself, made of oak
from Dodona, announced the need for purification (580—8). On
reaching Circe’s island, the Argonauts found her on the shore cleans­
ing herself from a terrifying dream of blood (662-71); Jason and
M edea alone followed her back to her palace. Circe offered them
seats, but they rushed without a word to the hearth and sat down
there; Medea covered her face with her hands, Jason planted in the
ground the great sword with which he had slain Apsyrtus; neither
raised their eyes from the ground. Circe understood, and in respect for
Zeus of Suppliants ‘performed the sacrifice by which innocent sup­
pliants are cleansed, when they come to a person’s hearth’. Holding
up a sucking pig she cut its throat and sprinkled their hands with its
blood; then she poured offerings to Zeus of Purification, with invoca­
tions. This completed the purification itself, and Circe’s attendants
carried outside the polluted remnants (lumata). She herself remained
at the hearth and made burnt offerings and libations to appease the
Erinyes and Zeus, ‘whether they were stained by a stranger’s or
related blood’ (685-717). Only then did she raise Medea and Jason
from the hearth, give them seats, and ask their names and story.
T he priestess at the opening of the Eumenides discovers Orestes in
the same position of silent submission at the omphalos as Jason and
M edea at Circe’s hearth. His hands are still dripping with blood; in
one hand he holds the sword he killed with, in the other the sup­
p liant’s olive branch with a woollen fleece around it.3 Little that is
1 See recently G inouvès, 3 1 9 -2 5 , R. R. D yer, J H S 89 (1969), 3 8 -5 6 , Burkert, GR
1 3 7 -9 .
2 Cf. p. 133 n. 111.
3 4 0 -5 .
Appendix 6 371
clear emerges from the surviving fragments of the Athenian exegetic
rules for the ‘purification ofsuppliants’,4 but the Cyrene cathartic law
offers suggestive hints. Details are obscure, but we read of a threshold
which the polluted suppliant probably may not cross; he must be
seated there on a fleece, washed and anointed; when he is led out into
the public street, all those present must be silent, and apparently a
special herald goes ahead to warn passers-by of the coming pollution.5
T h at law is a mere fragment, but it is sufficient to show again,
outside poetry, the constrictions placed on the polluted man, the
danger that he is for normal people, and his helplessness. A purifica­
tion is a ritual dram a, more effective even than simple supplication.
T he polluted man is excluded from society, and in his appeal for
purification he expresses that isolation by silence and, perhaps, veiled
head.6 O ther participants and bystanders confirm his abnormal con­
dition. They too are silent, and may cover their heads.7 Not words but
symbolic actions tell the m urderer’s story and make his appeal. He
sits at the hearth or threshold; sitting itself, in ritual, expresses
submission, and no normal person chooses such a place.8 His part is
one of complete passivity, since he cannot purify himself. The ‘doer’ of
the killing has been reduced to mere ‘suffering’.
W âshing with water sometimes formed part of the rites themselves.
It is mentioned in the Cyrene law, and an obscure fragment of the
cathartic regulations of the Athenian Eupatridai,9 but it seems only
to have been a subsidiary. The central act, already criticized by
H eraclitus in a famous passage, was the cleansing of blood by blood.
‘Vainly they cleanse themselves with blood when they are polluted by
blood, as if a m an who had stepped into mud were to wash himself in
4.Jacoby, 16.
5 See p. 350.
6 Silence: A esch. Eum. 448, E ur. H F 1219, idem , fr. 1008 ap. schol. Aesch. Eum. 276,
A p p e n d ix 7 s.v. T elep h u s. V eiled head: E ur. / 7 Ί 2 0 7 ,///'1 2 1 4 0 '. (not d em onstrable for
th e cerem o n y itse lf). T h e silence in theory covers the entire period before purification
(A esch. Eum. 448, T elep h u s legend): an ideal th at, taken seriously, proves ridiculous
(A rist. Poet. 1460a 32).
7 B y stan d ers are silent in the C yrene inscription; so is Circe w hen she understands
th e situ atio n . N o te too E ur. I T 9 5 1. T h o as, Eur. / 7 Ί 2 1 8 (slightly different context),
covers his head.
8 O n -sittin g a n d th e h e a rth in supp licatio n s e e j . G ould, J H S 93 (1973), 9 5 -7 .
9 356 F G rH fr. 1 ap. A th. 410b (associated w ith blood). O vid, Fast. 2.45 f., chides the
an c ie n ts for su p p o sin g w ater could efface bloodshed. It was sufficient after b attle or
ju stifie d killing (H orn. II. 6 .2 6 6 -8 , Od. 22.478 f.), b u t H eraclitus, A eschylus, and
A pollo n iu s im ply th a t blood was necessary in cases o f m urder. O vid w as perhaps
m isled by trad itio n s associating p articu lar springs o r areas o f sea w ith the purification
o f fam o u s killers: e.g. P aus. 2.31.9, O vid, loc. cit., A m m . M arc. 22.16.3, G inouvès, 323,
A p p en d ix 7 s.v. A chilles, C ad m u s, H eracles. O n the Apolline purification by water
a n d lau rel p o stu lated by P. A m andry, Rev. Arch. 116 ( 1938), 19- 27, see the criticism s of
G in o u v ès a n d D yer, locc. citt.
372 Miasma
m u d .’10 Heraclitus was only emphasizing a paradox of which all who
thought about the rite were aware, and which seems to have been
essential to its meaning. The strangeness of washing blood with
blood, of purifying by defilement, is constantly underlined in other
references: ‘to wash away foul blood by blood’; ‘he washed the trace of
killing from my hand by slaughtering fresh blood upon it’; (you will
not be clean) ‘until the slaughter of a young animal, by a man who
purifies from the stain of blood, bloodies your hands’; ‘until Zeus
him self stains you with drops of pig’s blood’.11
The language o f ‘wiping out blood with blood’ is sometimes found
not in relation to the purification ceremony, but to actual vengeance
killing.12 The ritual has accordingly been seen as a substitution, the
pig dying in place of the murderer himself.13 The verbal parallel
suggests that this idea did hover in the background, but the details of
the cerertiony cannot be explained in these terms: a sucking pig, the
cheapest of offerings, or a lam b14 is a poor replacement for the life of a
m an, and the substitution theory ignores what is central to the rite,
the sprinkling of the anim al’s blood on the killer’s hands. It is more
plausible to see here merely one of several special applications of
Greek religion’s most powerful form of action, the killing of an
anim al;15 a comparable case, also involving the manipulation of
bloody rem nants, is the ritual that accompanied oaths. Purification
‘by blood’ often occurs where there is no question of purification ‘from
blood’. Temples, assemblies, and armies were regularly cleansed in
this way; so were priests who had contracted a pollution, the
mysterious tithed men of the Cyrene law, the ‘sixteen women’ and
Hellanodikai at Elis before any ceremony, and persons mad,
epileptic, or bewitched.16 The exact procedure is not clear in all these
cases, but vases that show the cleansing of the daughters of Proetus
suggest that actual sprinkling with the victim’s blood w as. not
10 B 5 (86 M arcovich).
" E u r. I T 1223 f., Stheneboea, prologue 25 v. A rnim , Aesch. Eum. 449 f., fr. 327. But
d e s p ite the ‘defiling’ th e process is a w ashing, Aesch. Eum. 281, E ur. I T 1224, 1338,
Sthen., loc. cit.; cf. G inouvès, 321.
12 e g- P f Leg. 872e—873a, Soph. O f 100; cf. R. H irzel, ‘Die T alio n ’, Philol. Suppt. 11
(1 9 0 7 -1 0 ), 4 0 5 -8 2 .
R o h d e, 296, D iels, 69 n. 2, 122, Stengel, 159 f., J . P. G uépin, The Tragic Paradox,
A m ste rd a m , 1968, 1 6 0 -7 . Contra, Schw enn 8 1 - 4 , Nilsson, GGR 104, R u d h a rd t, 166.
L a n g u a g e like th a t ot L ù C G 156 A 14 περιταμέσθω χοίρψ is irreconcilable w ith the
su b s titu tio n theory.
14 E u r. I T 1223. Still less explicable is the dog (p. 230 n. 136). T h ere is no explicit
G re e k testim ony for th e idea th a t the evil passes into the anim al: for R om e see Val.
F lacc. Arg. 3 .4 3 9 -4 3 (clearly im plicit), A ppian, B C 5.96.401, Serv. A uct. ad Virg. Aen.
2.140 (explicit).
15 N ilsson, G G R 106.
16 See p. 21 f., p. 30 n. 66, pp. 230 an d 339 ff.; Z.SCG 156 A 14, 157 A 2; P aus. 5.16.8.
Appendix 6 373
restricted to the purification of m urderers.17 Unless, therefore, we
suppose this specific form to have been transferred to other contexts
from that of m urder purification, the homoeopathic idea o f ‘washing
blood with blood’ is a secondary development. The original source of
power is the contact with blood, a repugnant, polluting substance, in
a controlled ritual context that renders the threat tolerable.18 For a
m urderer, this sanctification of pollution is particularly apposite.
Blood falls on his hands again, but this time it is not he who has struck
the blow. The original blood, profanely shed, clung to his hands; the
anim al blood, shed in ritual, may be wiped off or washed away, and
the bloody remains are readily disposed of.
It is clear from Apollonius and other evidence that the actual
purification was followed by rites of appeasement addressed to under­
world powers.19 The two aspects seem to be united in the symbolism
of the so-called ‘fleece of Zeus’,20 on which the candidate for purifica­
tion sometimes stood. He placed on it his left or inferior foot, which
suggests that it was a receptacle for his impurity;21 indeed, ancient
scholars were perhaps correct in deriving the verb apodiopompeisthai,
‘send away (pollution)’, from the fleece of Zeus. But since it came
from a ram sacrificed to Zeus Meilichios or Ktesios, it also brought
the candidate into symbolic contact with the god he sought to
appease.
As we have seen, a rite that recalls that of murder purification
seems to have formed part of initiation at Eleusis or Agrai. The sitting
posture, veiled head, silence, and passive submission of the candidate
are all the same; even the fleece of Zeus appears in the Eleusinian
context.22 It is generally agreed that the explanation lies in the
17 R eferences in p. 230 n. 134. B u rk ert h as accordingly suggested (G aisford lecture
h eld in O x fo rd , 9 M a r. 1982) d erivation from a B abylonian healing ritu al w hich uses a
p ig (R. C . T h o m p so n , The Devils and Evil Spirits ojBabylonia, London 1 903-4, ii, 1 6 -2 1 ).
B u t for the B ab y lo n ian rite g oats seem to have been used as com m only as pigs
(T h o m p so n , op. cit., 2 1 -3 7 ), an d th e essential sym bolism is q u ite d istin ct in the two
cases (‘w ash in g ’ w ith blood in G reece, laying p a rts o f the anim al on corresponding
p a rts o f the sick h u m a n b eing in the B abylonian text). 18 Cf. V ickers, 142 f.
19 e.g. A rctin u s, O C T H o m er V , p. 105.29, LSS ! 15 B 58, Dem. 23.72, Plut. Thes. 12.1,
V al. F lacc. Arg. 3.444—58 (cf. P. Boyancé, R E L 13 (1935), 107-36).
20 Cf. J . H a rriso n , Prolegomena to the Study o j Greek Religion2, C am bridge, 1908, 23i—7;
C o ok , i, 4 2 2 -8 ; B u rk ert, GR 87 (M y cen aean fleece of Zeus?). A com parison has
som etim es been m ad e w ith the an cien t R om an p ractice in cases of involuntary
h om icide. T h e killer gave the victim ’s relatives a ram qui pro se agatur, caedatur ( Festus, p.
476.20 L .). T h is is a clear case o f su b stitu tio n . C incius ap. Festus p. 470.21 L, says this
w as d o n e exemplo at . . . (Atheniensium Scaliger); G reek sources offer no support for
S caiig er’s resto ratio n . 21 H esych. s.v. Διύςκώ όιον.
22 p. 285 ab o v e. N ote how ever th a t th ere is no h in t th at the piglet on the Lovateili urn
is to be used for purificatio n . T h u s th e claim th at the rites o f m urder purification are in
o rig in sim p ly th o se used in any induction w ould o u tru n the evidence —though it would
n o t be a b su rd .
374 Miasma
common character of the two ceremonies as rites of passage. The
candidate at Eleusis is inducted into the society of the initiated;
homicide purification means the reacceptance of the killer into social
and religious life. The killer in the Cyrene inscription supplicates a
body that represents the whole of Cyrene.23 In Athenian law, the
involuntary homicide on return from exile was purified as a token of
reintegration into his old community.24 When Plato in the Laws
prescribed that the killer should be purified before going into exile, he
seems to have severed an organic link that he no longer understood
between purification and admission to a social group.25 A killer may
in theory be purified in one place and go to settle in another, but in the
predom inant mythical pattern he starts life anew at the place of
purification. Conversely, the man guilty of a particularly repugnant
crime may be refused purification, because the person supplicated is
not prepared to tolerate his continuing presence.26 (The alternative
response, of purifying and then expelling, seems to be found only in
Apollonius.27 O f course if, as in Apollonius and a passage in
H erodotus,28 the purifier was really expected to postpone inquiry
about the candidate’s identity until after the ceremony, it is hard to
see how he could avoid applying ‘the rites by which innocent sup­
pliants are cleansed’, as did Apollonius’ Circe, to the very guilty.
I here are obscurities here that we cannot resolve; but the point
rem ains that purification without reception is quite untypical.) It is
im portant to hold fast to this social context. Purification for murder
was not performed by chance persons, amateurs, peripheral seers,
disreputable magicians; ifit was, public opinion could have denied its
efficacy. In myth, the purifier is the man of wealth, position, and
responsibility; in history, a priest or official of the community,
perhaps representative of an ancient aristocratic family.29 Only when
these rites are considered outside their social context do they seem
wholly mechanical and amoral; in context, they have a logic which,
though not exactly that of morality, suits well enough their patron,
Zeus the arbiter of social life.
23 See p. 350 above.
24 D em . 23.72.
25 8<>5d, 866a.
26 M y th o f Ixion; E ur. Or. 429 f.; cf. p. 118 n. 58.
27 E xpulsion: 4.745.
28 1.35.
29 O n G yrene see text. 1n S am othrace, a special priest purified m urders: Hesych. s.v.
κοιής. 1 he E leusinian d a d u c h used the ‘fleece o f Z eus’, b ut it is not clear w hat for (Suda
s.v. Δ ιά ς κώόίον, cf. above, p. 285). A t A thens the E u p atrid exegetes (Jacoby 16)
su p e rv ised the ritu a l, an d the exêgëtai pythochrêstoi offered advice. T h e P hytalid gens
c o n tro lle d a n a lta r at w hich purifications seem to have been perform ed,'Plut. Thes. 12.1,
P au s. 1.37.4, cf. TöpfFer, 249 f. T h e role of the έγχντρίστριαι (p. 36 n. 15) will have been
su b o rd in a te .
Appendix 7:
Exile and Purification of the Killer in Greek Myth
This appendix collects evidence for acts of homicide in Greek myth
that lead to the exile, purification, or, occasionally, trial of the killer.
This means that certain classes of homicide are omitted:
( 1) Killings that lead to simple human revenge: the Hippocoontids
kill Licymnius and are killed in turn by Heracles (Apollod. '2.7.3),
Aegisthus kills Agamemnon and is killed by Orestes, and so on.
(2) Semi-justified homicide: a bad character is killed and no conse­
quences are reported —Neleus and Peleus kill (at the very altars) their
wicked stepm other (Apollod. 1.9.8), Oeneus kills his son or brother-
in-law Toxeus for jum ping over his wall (Apollod. 1.8.1, cf.
Heldensage, 86), Cycnus kills his wife who had slandered her stepson
(Apollod. Epit. 3.25, cf. RE 11. 2440 f., Heldensage, 387), Aepytus/
C resphontes kills the Aegisthus figure Polyphontes (Apollod. 2.8.5),
the sons of Antiope kill their wicked stepmother and found Thebes
(Apollod. 3.5.5), Cercyon kills his unchaste daughter Alope (Hyg.
Fab. 187).
(3) Only slightly distinct from the former, cases where no conse­
quences of killing are reported although these might have been ex­
pected: Aetolus slays his hosts (Apollod. 1.7.6), Apollo accidentally
kills Hyacinthus (Apollod. 1.3.3 with Frazer), the daughters of
Cocalus kill Minos (Apollod. Epit. 1.15, Heldensage, 367 f.).
(4) Aitiological myths of the killing of individuals particularly dear
to the gods (like Linus): cf. p. 274 n. 80.
(5) Killings where the perpetrator commits suicide (Themisto,
Deianeira) or suffers transformation (Procne, Harpalyce, etc.).
The fact that categories two and three can be omitted has, of
course, some negative significance; purification is not indispensable
to every myth about killing. But there is no consistency in this area
(again an interesting fact); in terms of moral justification, several
instances in the following list might well find a place in category two.
Even with these exclusions the following list makes no pretence to
comprehensiveness (though I have included all the cases known to
me). Especial emphasis has been placed on the function that these
killings have within a particular hero’s career. Vickers has recently
376 Miasma
protested against the tendency to dismiss as insignificant the
‘standard elements’ in Greek myths and concentrate on the search for
distinctive features; these standard elements - murder, sacrilege,
offences against the gods or family —are, he insists, the very key to
understanding the mythology as a whole (Vickers, Appendix 2). That
protest needed to be made. None the less, as most scholars since
Lobeck (Aglaophamus, 969) have recognized, and the following
analysis confirms, killing is constantly used as a purely structural
device in myths whose main concern is genealogical. The number of
m yths in which the moral implications of murder are of central
im portance is by contrast small (although a poet was, of course, free
at any time to transform a structural device into a main theme).
Achilles: (1) on Achilles and Thersites see p. 131 n. 102.
(2) in a raid from Troy, on Lesbos or in Miletus, Achilles killed the
local hero Tram belus, and, on learning him to be a son of Telamon,
lam ented him deeply (Istrus, 334 FGrHfr. 57, Parth. Amat. Nan. 26).
A m iraculous well was shown in Miletus where Achilles had purified
him self from this killing (139 FGrH fr. 6 ap. Ath. 43d).
Local legend of indeterminate date (RE 6 A 2129, Jacoby on Istrus,
loc. cit.).
Aeolus', in exile for a year for ‘a killing’ (hypothesis to Eur. Melanippe ή
σοφ ή, ap. H. v. Arnim, Supplementum Euripideum, Bonn, 1913, 25 f.)
Probably an invention by Euripides to allow Melanippe’s seduc­
tion in her father’s absence.
Aetolus: An Elean by birth, he killed Apis accidentally and fled to the
C ouretan country (65 FGrH fr. 1, Apollod. 1.7.6, schol. Pind. 01.
3.22c), which subsequently took his name, Aetolia.
T he Eleans were of Aetolian stock, through Oxylus (Pind. 01. 3.12,
Ephorus, 70 FGrH fr. 115, 122, RE 5.2380 f.). Aetolus, who in
Hecataeus was born and bred in Aetolia (1 FGrH fr. 15), was trans­
formed into an Elean to show that Oxylus in invading Elis was
reclaiming an ancient heritage, perhaps to make Aetolia derive from
Elis and not vice versa (cf. RE 1.1129, Heldensage, 281).
Agave\ exiled for Pentheus’ death, Eur. Bacch. 1330-92.
Appendix 7 377
Alcathous: (1) cf. s.v. Pelopids.
(2) when king of Megara, killed his own son Callipolis in a com­
prehensible fit of anger; was purified, without exile, by the
M elam podid Polyidus (Paus. 1.42.6, 43.5).
Local legend of uncertain but probably not early origin (cf. Jacoby
on 485 FGrH fr. 10 n. 45) uniting (1) M egara’s chief hero Alcathous
(2) the M egarian connection of Polyidus (3) an existing monument,
the ‘monument of Callipolis’.
Alcmaeon: The early tradition about the consequences of the
m atricide cannot be recovered with certainty (cf. M. Delcourt, Oreste
et Alcméon, Paris, 1959; Heldensage, 956 ff.). Asclepiades ofTragilus, a
w riter on tragic plots, offers two unusual features ( 12 FGrH fr. 29) : the
m atricide precedes the expedition of the Epigoni against Thebes (if we
accept, as we surely must on Proppian principles, that Alcmaeon on
this point obeyed his father’s behest), and when Alcmaeon goes mad
in consequence of the murder the gods themselves intervene to cure
him. Some have supposed this to be an earlier version (Heldensage, 956 f.,
following Bethe) than the more familiar one by which the matricide
followed the expedition, and Alcmaeon fled to Acarnania in conse­
quence, in search of a land invisible to the sun when the matricidal
blow was struck (Thuc. 2.102.5-6). An alternative explanation for
the migration is available: he had marched up in that direction to help
Diomedes recover his Aetolian heritage. Several scholars have as­
cribed this motivation to the epic Alcmaeonis, because the source for it,
Ephorus (70 FGrH fr. 123), went on to cite the poem (fr. 5 Kinkel =
Ephorus, fr. 124): so e.g. RE 1.1563, P. Friedlaender, Rh. Mus. 69
(1914), 330 f. = Studien zur Antiken Literatur und Kunst, Berlin, 1969,44.
T here is nothing compelling about this reconstruction; Asclepiades’
version could well derive from tragedy, and the Ephoran account of
the A carnania expedition looks like a rationalization, which had the
further advantage of explaining a mythographical problem, the
A carnanians’ absence from the Trojan expedition (Ephorus, loc.
c it.). A polluted, wandering Alcmaeon could therefore have appeared
in early epic (even Friedlaender, loc. cit. postulated him for Epigoni).
A Tyrrhenian am phora (570-60) shows a serpent rising from
Eriphyle’s corpse to pursue Alcmaeon (K. Schefold, Myth and Legend
in Early Greek Art, London, 1966, 80, Fig. 30).
For the elaborated Alcmaeon romance with various purifications,
relapses, and marriages see Apollod. 3.7.5-7. Much of this we owe to
Euripides, cf. Heldensage, 959 ff.

378 Miasma
Althaemenes: a C retan who, on learning that his father Catreus was
destined to die at the hand of one of his children, emigrated to Rhodes.
T here he killed his sister for her supposed unchastity, and unwittingly
killed his father too when he landed secretly on the island. On
realizing what he had done, he prayed and was swallowed into the
earth (Apollod. 3.2.2) or went out into the wilderness and there died
(Diod. 5.59.4).
Althaemenes was a Rhodian hero, credited with introducing the
worship of Zeus Atabyrios (Diod. 5.59.2). The tradition that made
him a Heraclid was probably secondary (Heldensage, 373). His two
m urder myths have no obvious aitiological meaning, but their date is
uncertain; Rohde thought them Hellenistic (cf. Heldensage, 371 η. 6).
Amphitryon: killed his father-in-law Electryon, accidentally (3 FGrH
fr. 13b, Apollod. 2.4.6) or in anger (Hes. Asp. 11,82; 3 FGrH fr. 13c),
and fled from Argos to Thebes, where he was received (Hes. Asp. 13, 3
FGrH fr. 13c) or purified (Apollod. 2.4.6, Hes. Asp. hypoth. D, E
Rzach) by Creon.
T here is some reason to think that Amphitryon was originally a
T heban (cf. Hes. Asp. 1 f.); when he was adopted into the Perseid
genealogy at Argos, the killing of'Electryon was necessary to take him
back to Thebes: Robert, Oidipus, ii, 40-2; P. Friedlaender, Herakles
(Philol. Untersuch. 19), Berlin, 1907, 47 f.; Heldensage, 605 ff.
Apollo'. ( 1) forced to serve Admetus for a year as penance for the
killing of the Cyclopes (or the Delphic dragon, 404 FGrH fr. 5). This
service was already known to the Hesiodic catalogue (fr. 54b-c, cf.
Eur. Ale. 1—7, 3 FGrH fr. 35a; P/R 270).
(2) after the slaying of the dragon at Delphi, Apollo fled to Tempe
or Crete (via Aegialeia, Paus. 2.7.7) for purification. The Homeric
hym n knows nothing of purification; the Tempe tradition derives
from an aitiological connection of uncertain date with the Septerion (?
first attested in the fourth century, Theopompus, 115 FGrH fr. 80, cf.
Parke/W orm ell, i, 14n. 12); the Cretan tradition might beolder (cf. p.
142) though one could argue that the very idea that the dragon’s
death required purification was created by the Septerion aition.
Archias: of C orinth killed his lover Actaeon (Plut. Amat. Narr. 772e-
773b, Diod. 8.10); Actaeon’s father committed suicide, plague fol­
lowed, Archias went into voluntary exile and founded Syracuse,
where another lover killed him. According to another tradition, the
Appendix 7 379
Bacchiads in general were responsible for the crime and were expelled
because of it (Alex. Aetolus, fr. 3.7-10, Coll. Al. p. 122; schol. Ap.
Rhod. 4.1212; cf. A. Andrewes, CQ 32 (1949), 70 f.).
A rationalized version of the Actaeon myth has been adopted to
explain political change or colonization.
Archelaus: a Temenid, who kills the Thracian king Cisseus, with every
justification, but on Apollo’s advice flees to Macedonia (Hyg. Fab.
219).
T he very hero seems to be a courtly invention, perhaps entirely
Euripides’, to please king Archelaus of Macedon (Heldensage, 669 f.).
Ares: killed Halirrhothius, Poseidon’s son, who was trying to rape
A res’ daughter Alcippe. Prosecuted by Poseidon, tried before the
twelve gods, and adjudged to have committed justified homicide
(Eur. El. 1258—63, implied by counter-etymology in Aesch. Eum.
685-90; cf. Apollod. 3.14.2 with Frazer).
A simple charter for the Areopagus and for the category ofjustified
homicide (killing of rapist as justified homicide, Dem. 23. 53), confus­
ing only in that such cases were in fact tried at the Delphinion.
H alirrhothius’ only other legend seems secondary (RE 7. 2270).
Athamas: killed his son Learchus in madness, was expelled from
Boeotia, and after wanderings settled at Halos or Athamantia in
Thessaly (Apollod. 1.9.2, cf. Jacoby on 4 FGrHfr. 126).
Infanticide is an element proper to the Athamas myth, but these
wanderings link the Boeotian and Thessalian Athamas legends and
provide an etymology for Halos (άλη; cf. Heldensage, 43 f., Jacoby, loc.
cit.).
Atreus: cf. s.v. Pelopids.
Bellerophon: killed his brother accidentally and went to Proetus to be
purified (Apollod. 2.3.1 with Frazer). In Homer his presence at
Proetus’ palace was unexplained; the killing and purification first
appear in Euripides’ Stheneboea (prologue 23-5 v. Arnim) and might
be his invention (cf. Heldensage, 181-3, esp. 183 n. 1).
380 Miasma
Cadmus: served Ares for a great year (8 years) in appeasement for the
killing of the dragon (Apollod. 3.4.2, containing 3 FGrH fr. 89; Ares’
anger against the Cadmaeans, Eur. Phoen. 931—5).
Jacoby on 3 FGrH fr. 89 and Latte, RE 10. 1464, regard the service
as ancient, a necessary preliminary to marriage with Ares’ daughter
H arm onia.
For Nonnus (5.4) Cadm us purified himself from the dragon’s blood
in Dirce.
Carnabas·. son of Triopas, the savage Perrhaebian prince, killed his
father and was honoured by the people for this liberation, but fled
none the less through his blood-guilt; he sailed to the Troas, was
purified by king Tros, and given land where he founded Zeleia (schol.
T and Eustath. ad Horn. II. 4.88).
O n e of the m any mythical reflections of Thessalian colonization in
this region’, R E 10. 1950; for various myths linking Triopas and
Triopids with east Greek colonization cf. R E 7 A 171.
Cephalus\ killed his wife Procris accidentally (383 FGrH fr. 2, Apollod.
3.15.1) or in passion (3 FGrH fr. 34), was arraigned by her father
Erechtheus before the Areopagus (4 F G r//fr. 169a, Apollod. 3.15.1,
cf. 334 FGrH fr. 14), and condemned to permanent exile. Pausanias
and Aristodemus (Paus. 1.37.6, 383 FGrHfr. 2) tell how he went in his
exile to Thebes; according to Aristodemus, supposedly deriving from
the epic cycle (Epigoni, fr. 2, O C T Homer v, p. 115 Allen; cf.
Heldensage, 162 η. 5), he was purified by the Cadmaeans and then
aided them with his miraculous dog against the Teumessian fox; his
involvement with the fox was elsewhere narrated independently of his
exile in Thebes (Apollod. 2.4.7).
It looks as if two substantive elements in Cephalus’ myth - the
Procris romance (already Horn. Od. 11.321), the fox episode - were
conveniently linked, perhaps already in the Epigoni, by the Theban
exile.
Copreus: a son of Pelops, killed Iphitus, fled to Mycenae, and was
purified by Eurystheus (Apollod. 2.5.1).
In Horn. II. 15.639 Copreus was Eurystheus’ herald, and properly
nam ed for that office. Impossible to say when and how he became a
Pelopid (R E 11.2.1364).
Appendix 7 381
Daedalus: m urdered his nephew Talos/Perdix in jealousy of his skills,
was condemned to death by the Areopagus, and fled to Crete (Apol­
lod. 3.15.8 with Frazer; the m urder already implicit, Soph.fr. 323).
M. Ventris and J . Chadwick, Documents in Mycenaean Greek2,
Cam bridge, 1973, n. 200 (Cnossus) shows that Daedalus is a Cretan,
appropriated by Athens for his intellectual sharpness and sent back to
his homeland through the m urder myth.
Danaids: after the m urder of their husbands they were purified by
A thena and Hermes at the command of Zeus (Apollod. 2.1.5, and no
other source). A. F. Garvie, Aeschylus’ Supplices: Play and Trilogy,
Cam bridge, 1969, 211—33, discusses whether this detail is Aeschy­
lean, inconclusively.
Epeigeus, Lycophron, Medon (Horn. II. 16.571; 15.430; 13.695): all
Hom eric warriors said to have fled their homes through murder.
T hey have no existence outside Homer; Homer makes them killers to
give them a touch of individuality at the point of death.
Heracles: (1) after the murder of his children, Heracles condemned
him self to exile, and was purified by Thespius; he consulted Delphi,
an d was told to serve Eurystheus (Apollod. 2.4.12). For Menecrates
(schol. Pind. Isthm. 4 .104g) he was purified by Sicalus, for Ap. Rhod.
4.539—41 by Nausithous the Phaeacian; several sources make the
m urder prelude to the service with Eurystheus (Moschus 4.13-16
36-45; 90 FGrH fr. 13; Diod. 4.11.1 -2 ).
Considered by Wilamowitz (ed. Eur. HF, Berlin, 1895, i, 87)
followed by P. Friedlaender (Herakles, Philol. Untersuch. 19, Berlin,
1907, 51) ‘ein Hilfsmotiv ohne innerliche Bedeutung’, whereby the
T hebans explained the activity o f ‘their’ Heracles in. Argos and .his
lack of Theban progeny. If so, this is a clear case of a structural device
th at grew into a central theme; but it is hard to accept so banal an
origin for the m otif (cf. Heldensage, 628).
(2) while feasting with his father-in-law Oeneus, accidentally kil­
led the cupbearer, a relative of his host, and though Oeneus was
prepared to forgive the involuntary crime chose ‘to undergo exile
according to the law’ and went to Trachis (Apollod. 2.7.6, cf. Frazer
ad. loc., Heldensage, 576 f.; first reference 4 FGrH fr. 2).
The story set Heracles on the road for his encounter with Nessus,
382 Miasma
illustrated his uncontrollable strength, and provided an aition for cult
of the variously named cupbearer (Heldensage, 576 f.).
(3) after the treacherous murder of Iphitus, Heracles was purified
a t Amyclae by one Deiphobus (Diod. 4.31.5, Apollod. 2.6.2), after
Neleus in Pylos had refused this service because of his friendship with
Ip h itu s’ father Eurytus (Diod., Apollod., loc. cit.; Heldensage, 537 η. 3;
cf. schol. Pind. 01. 9.43, 44c). Despite purification, Heracles’ illness
continued; he sought a cure at Delphi, where Apollo enjoined his sale
to O m phale and the payment of the proceeds of the sale to Eurytus or
Iphitus’ sons (Diod., Apollod., loc. cit.; in the probably older version
of 3 FGrH fr. 82b, Soph. Tr. 274-6, Zeus himself saw to Heracles’
sale).
T he m urder of Iphitus, known to Homer and older perhaps than
the sack of Oechalia, probably served originally to motivate the
servitude under O m phale (P. Friedlaender, op. cit., 73—8). The sack
of Pylos by Heracles is mentioned in Homer, but not its motive (II.
11.690). According to one prominent later tradition, it was due to
Neleus’ refusal to purify him from the m urder of Iphitus (cf. above),
but it is not safe to attribute this motive to Hesiod on the strength of
schol. AD Horn. II. 2. 333—5 (1.102.17 Dindorf), which concludes a
narrative along these lines with ‘Hesiod tells the story in the Cata­
logues'. Lobeck long ago warned that there might be conflation here
(Aglaophamus, 309), as there commonly is in the Homeric
m ythographic scholia (E. Schwartz, Jahrb. f. Klass. Phil. Suppl. 12
( 1881 ), 405 if.); we now have a Hesiodic fragment describing the sack
of Pylos (fr. 33), and it seems clear that the scholion referred to Hesiod
merely for the picturesque detail of Periclymenus’ transformations.
O th er explanations for the sack of Pylos were known (cf. the con­
troversy in the scholia to Horn. II. 11.690, Heldensage, 537).
(4) the Lesser Mysteries at Agrai were founded to purify Heracles
from the blood of the Centaurs: cf. p. 284.
(5) in his youth Heracles killed Linus, his lyre-teacher, in anger
because Linus had struck him. Heracles cited a law of Rhadamanthys
to show that this was justified homicide, and went free (Apollod. 2.4.9
with Frazer).
(6) Heracles cleansed himself from Cacus in a neighbouring
stream: Dion. Hal. Ant. Rom. 1.39.4.
(7) after killing the wicked sons of Proteus, he was purified by their
own father: Conon ‘26 FGrH para. 32.
Appendix 7 383
(8) while returning from Troy Heracles was blown to Cos, fought
with the Meropes and defeated them, was purified, and married ?
Chalciope (Plut. Quaesi. Graec. 58, 304c—e; for Heracles on Cos cf.
Horn. II. 14.255, Apollod. 2.7.1).
A Coan aition.
For other killings by Heracles, even that of the sons of Boreas (2
FGrH fr. 31, A p.R hod. 1.1300—6 with schol.), purifications seem not
to be recorded. He was, of course, the ‘justest of homicides’ (Peisan-
der of Rhodes fr. 10, p. 252 Kinkel).
Hippotes·. the Heraclid, killed the seer Carnus, mistaking him for a
m alevolent magician; plague afflicted the Heraclids, and Hippotes
was banished for ten years (Apollod. 2.8.3, Paus. 3.13.4). The killing
of C arnus followed by plague was an aition of common type for a
festival, the Carneia. The exile and wandering of the killer, not typical
of such aitia, explained the name of Hippotes’ son Aletes, the con­
queror of Corinth (/?£ 8.1923).
Hyettus: killed Molurus son of Arisbas flagrante delicto and fled from
Argos to king Orchomenus, who received him and gave him great
wealth, including the village now called Hyettus (Paus. 9.36.6-7,
citing Hes. fr. 257).
Hyettus, M olurus, and Arisbas are otherwise unknown (Arisbe is a
place name, R E 2.847) ; the name Hyettus itself belongs originally to a
place, not a person (RE 9.91, citing Kretschmer). A simple but early
aitiological invention.
Ixion: treacherously murdered his bride’s father, and as first shedder
of ‘kindred blood’ went mad. No one would purify him until Zeus was
moved to pity; he now attempted to seduce his benefactor’s wife, and
was condemned in punishment to spin endlessly on his wheel (for the
sources see R M L s.v. Ixion, Nephele·, first in Pind. Pyth. 2.32; Aesch.
Eum. 441, 718, idem, the Ixion trilogy; cf. Heldensage, 12-15).
A tightly knit myth of social crime, whose solar origins (Heldensage,
15), if genuine, were quickly forgotten.
Jason·, in flight from Colchis murdered the brother of Medea
(.Heldensage, 800—2, R E 2.285). It is very uncertain whether the puri­
fication from this crime described Ap. Rhod. 4.662—717 belongs to the
old legend (contrast Heldensage, 827, RE 15.37 pro·, R E 11.504, RML
384 Miasma
2.1202 f. contra). Itis true that Circe, sister of Aietes (Horn. Od. 10.37),
belongs rather to the Argonautica than the Odyssey, but her role will
surely have been to give advice about the voyage on the journey out,
not to perform purification on the return (K. Meuli, Odyssee und
Argonautika, Berlin, 1921, 95 ff., esp. 112-14, = Ges. Sehr., ii, 661 if.
esp. 672 f.).
Ja so n ’s participation in the funeral games for Pelias was recorded
(Paus. 5.17.10), which suggests, although it does not prove (cf. the
Hom eric O restes’ funeral feast for Aegisthus), that his revenge
against Pelias was unknown to the earliest tradition (Heldensage, 37).
W hen the m urder was added, Jason and Medea were expelled from
Iolcus or left it voluntarily, and travelled to Corcyra or Corinth
(.Heldensage, 869, R E 9. 767; cf. Eumelus, fr. 2, p. 188 Kinkel, for
M edea’s ancient Corinthian connection).
Leucippus: the Lycian accidentally slew his father and so was forced to
leave home; according to the Magnesian legend the god of Delphi
appointed him leader of a group of displaced Thessalians who were
seeking new homes, and he established the colony of Magnesia (Inscr.
Magn. 17.36 fl. = Parke/W'ormell, nn. 381—2; Parth. Amal. Nan. 5: cf.
R E 12.2264).
A typical oecist romance.
Lycus and Nycteus: fled from Hyria (?) to Thebes after killing Phlegyas
(Apollod. 3.5.5; on the text cf. Robert, Oedipus, i, 398).
Brings Nycteus’ daughter Antiope, a Hyrian in Hesiod (fr. 181), to
Thebes {RE 17.1511 f.).
Medea: her flight to Athens after the murder ofC reon’s children and
her own is familiar from Eur. Med. Pre-tragic tradition on the conse­
quences of the infanticide, and M edea’s presence at Athens, is hard to
recover (RE 15.46, ibid., Suppl. 13.1081). The history of the latter is
complicated by problems surrounding the interpretation of several
sets of vases; but C. Sourvinou-Inwood, Theseus as Son and Stepson,
London, 1979, has argued strongly for detecting an Athenian Medea
from early in the fifth century. If so, it will have been after accidental
infanticide (the version of Eumelus, cf. R E 15.42 f.) that she was
originally received in Athens. Purification only in Ov. Fast. 2.42.
Odysseus'. ( 1) some told how the kinsmen of the slain suitors submitted
Appendix 7 385
the issue between themselves and Odysseus to Neoptolemus as arbi­
trator, who sent him into banishment (Apollod. Epit. 7.40, cf. Plut.
Quaest. Graec. 14. 294c—d).
T he banishm ent explained those Aetolian connections ofOdysseus
that existed in other versions too; Neoptolemus’ arbitration seems
also to have served as aition in an Ithacan cult of Telemachus (cf.
W. R. Halliday on Plut., loc. cit.).
(2) killed Euryalus, his son by a Thesprotian princess; this may
have been brought into connection with his own death at Telegonus’
hands (cf. Pearson and Radt, introductions to Soph. Euryalus).
Oedipus·, the problems of the literary treatment are complex; see after
Robert, Oidipus, Nilsson’s review of Robert, Göll. Anz■ 184 (1922),
3 6-46 = Op. Sel. i. 335-48; idem, The Mycenaean Origin of Greek
Mythology, California, 1932, 102-12; L. Deubner, Oedipusprobleme,
Berl. Abh. 1942, n. 4; F. Wehrli, M H 14 (1957), 108-17 = Theoria und
Humanitas, Ziirich, 1972, 60—71;Jacoby on 3 FGrHir. 95, 16 LG rH fr.
10; M. Delcourt, Oedipe, ou la légende du conquérant, Liège, 1944, intro­
duction; L. W. Daly, R E Suppl. 7. 769-86; W. Burkert, ‘Seven against
T hebes’, in Ipoemi epici rapsodici non omerici e la tradizione orale, Padova,
1981, esp. 29-35.
For the Iliad, Oedipus perhaps died in battle and certainly received
funerary games (//. 23.679, Robert, Oidipus, i, 115; but note the
reservation of Burkert, op. cit.). The Odyssean Nekyia knows the
parricide and incest, but leaves Oedipus on the throne, suffering, and
says nothing of blindness (Od. 11. 271-80; blindness unknown?,
Robert, Oidipus, i, 112). According to Pherecydes, Oedipus con­
tracted two further marriages after the incestuous one (3 FGrH fr. 95;
cf. 16 FGrH fr. 10.8, Paus. 9.5.11), and it is almost certain, although
Robert disputed it, that the remarriage of the polluted Oedipus
already occurred in early epic (Jacoby on Pherecydes, loc. cit., Nils­
son, Op. Sel., i. 345, Deubner, op. cit., 27 ff., Wehrli, op. cit., 112 (65).
Even the third wife may have been in the Catalogue, cf. Hes. fr. 190. 13 ff.,
193. 1—8 with M /W ’s notes). If αφαρ in Horn. Od. 11.274 is taken in
its normal sense, the Nekyia poet too must have known though not
m entioned a further marriage (Paus. 9.5.11), unless we suppose,
implausibly, that he did not know of the great expedition against
Thebes, or did not connect it with a quarrel between the sons ot
O edipus (so Nilsson).
O n the character of these early traditions see above, p. 136. Later
poets rendered Oedipus coherent by gradual elimination of incongru­
ous elements. He seems already to be blind and in his sons’ power in

386 Miasma
the Thebais^h. 2,3, pp. 11 f. Kinkel, with the parody in schol. Soph.
OC 1375 (Robert, Oidipus, i, 171). In Pherecydes, a purificatory year
apparently preceded the second marriage (this might, of course, be
traditional). T he tradition by which ‘Jocasta’ is mother of Eteocles
and Polyneices, and lives on after the incest is discovered, is now
attested in the Lille Stesichorus, cf. ZPE 26 (1977), 7-36; it excludes
subsequent marriages. Exile first appears in Sophocles, hinted at in
O T (e.g. 1518), worked out in OC; but it does not obliterate the older
tradition whereby Oedipus stayed in Thebes (Eur. Phoen., cf. Soph.
OC 765-7). Oedipus is never chased by Erinyes (although he suffers
from those of a mother in Homer), or purified.
Orestes: on the Homeric treatment see p. 136. Other pre-tragic tradi­
tions are little known. Orestes’ act was perhaps sometimes mitigated
by being seen as self-defence against the axe-carrying Clytaemnestra
(M . Delcourt, Oreste et Alcméon, Paris, 1959, 26; on the axe see Lesky,
R E 18. 973 f.). The Erinyes first appear in Stesichorus, as does Apollo,
who gives Orestes a bow to ward them off (PMG 217). Apollo’s
support for Orestes need not exclude moral conflict or the need for
purification (RE 18. 977) any more than it does in Aeschylus. Some of
the local traditions about Orestes’ healing may date back to the sixth
century (Paus. 2.31.4 - purification —, ibid., 3.22.1, 8.34.2; later
traditions R E 18. 990 f.); so too that of his Arcadian residence (Eur.
El. 1273-5, Or. 1643-5), which it is natural to connect with his
Irenzied wanderings. Jacoby believed that the trial before the Areopa­
gus was Aeschylus’ invention (commentary on 323a FGrH fr. 1), but
more probably the poet’s innovation was only to make Orestes’ trial
the one for which the Areopagus was founded (RE 18.980 f.). His
aitiological association with the Choes (Eur. /T947^60) is likely to be
ancient, but does not have to be. For a possible representation of
O restes struggling with a snake (Erinys?) see P. Zancani Montuoro
and U. Zanotti-Bianco, Heraion alia Face del Sele, ii, Rome, 1954,
289—300 with Plates 46,89.
Aeschylus’ treatm ent is helpfully discussed by O. P. Taplin, The
Stagecraft of Aeschylus, Oxford, 1977, 381-4. The purification occurs in
three stages. The first is the physical rite performed at Delphi by
Apollo (Cho. 1059 f., Eum. 282 f.), who in this assumes the role that
normally belongs to a human purifier (therefore άνόρός, Eum. 449, is
not a problem). Aeschylus insinuates into our minds the fact that this
purification occurs, even though we do not see it on stage and cannot
identify a point during the action at which it could have been
performed. There follows a period of exile, during which Orestes’
pollution is ‘rubbed off by social intercourse and ‘purified’ by time,
Appendix 7 387
because the blood on his hands ‘falls asleep’ (Eum. 238, 280, 286).
Aeschylus may be hinting at the Peloponnesian traditions concerning
O restes (above), but the basic conception is the well-attested one by
which the killer’s exile is itself a form of purification (p. 114). The
cleansing at Delphi was not redundant because it permitted him to be
received by hosts abroad during this exile. He points out (Eum. 285,
cf. 238 f.) that he brought no harm to them - a reasonable point, as
purification did not always take (Apollod. 2.6.2,3.7.5; for the ‘proof
by safe contact’ cf. Ant. 5.82 f.). The final stage, which permitted him
finally to return to Argos, free from the Erinyes, was the trial before
the Areopagus.
There is nothing unusual in the combination of purification and
‘purificatory’ exile; it was the fate of any killer who was not to be
banned in perpetuity (MacDowell, Homicide, 120-5). The trial is, of
course, an intrusive element among mythological responses to
homicide, and even from a fifth-century perspective misplaced, since
it would normally precede exile and not follow it. Given the trial’s
actual position, however, the Erinyes in a sense correspond to the
victim ’s relatives, whose pardon was necessary before an involuntary
killer could return from exile. There may also have been a reversal in
the order of physical purification and purificatory exile, but it is not
clear at w hat stage a killer going into temporary exile would normally
have received purification. There is no reliable historical evidence,
and the mythological or semi-mythological instances of temporary
exile do not normally mention physical purification (Eur. Hipp. 34-7,
Or. 1643—5, 90 FGrH fr. 45, 61 ; Aeolus above; but note Poemander
below, purified abroad). Common sense suggests that he underwent
it on arrival abroad, as did the permanent exile; would he not
otherwise be too dangerous to associate with? It also seems to be
necessary, except in the abnormal conception of Plato (p. 374), that
purification should occur away from the scene of the crime; Achilles
sailed to Lesbos to be cleansed from the killing of Thersites, and then
at once returned. If this is correct, Orestes conforms to the normal
pattern. Demosthenes, however, seems to attest for Athens purifica­
tion on return (23.72). It is perhaps not impossible that a killer could
be purified more than once, but such a repetition is unattested; if we
reject this possibility, it will be necessary, in order to keep Eumenides
consistent with Athenian practice, either to postulate a change be­
tween the time of the play and of Demosthenes, or to suppose that the
killer already cleansed abroad was exempt from purification on re­
turn (cf. perhaps Eum. 235-43). There are irresoluble uncertainties
here; but on any view of them, it does not emerge that Aeschylus is
concerned, as is often supposed, to dispute the importance of ritual
purification or deny its efficacy. Though he presents it as merely one
388 Miasma
stage in the process by which the killer was prepared for return to his
home territory, it was, as we have seen, never anything else.
in Soph. El., notoriously, the question of pollution is not raised
explicitly. In Eur. El. 1250-75, Orestes is required to leave Argos,
undergo trial at Athens, and settle in Arcadia. In Eur. Or. 1643-60
the trial is to be preceded by a year of (purificatory) exile in Arcadia,
but he may then return to Argos.
Oxylus: who by capturing Elis recovered the ancient heritage of his
ancestor Aetolus (above), was in some versions in exile in Elis for
accidental homicide when he met the returning Heraclids and in
obedience to an oracle was chosen by them as leader (Apollod. 2.8.3
with Frazer; Paus. 5.3.7).
T he exile is a simple mechanism to put him in the Heraclids’ path.
Patroclus·, born in Opus, he killed a youth in anger over a game of
knuckle-bones and fled to Peleus with his father (Horn. II. 23. 84-90;
4 FGrH fr. 145; Apollod. 3.13.8).
Perhaps an attem pt to reconcile two traditional homes of Patroclus,
Phthia and Opus (RE 18.4.2275 f.).
Peleus·. (1) Peleus and/or Telamon murdered their brother Phocus
(first in Alcmaeonis, fr. 1, p. 76 Kinkel; the accidental tradition of Ap.
Rhod. 1.92 f., Diod. 4.72.6 is certainly secondary), and were expelled
by Aeacus from Aegina. Peleus went to Phthia, was purified by king
Eurytion, and m arried his daughter (3 FGrH fr. lb, Apollod. 3.13.1).
(In Ov. Met. 11. 409 he is purified from Phocus by Acastus.)
(2) in a boar-hunt (sometimes the Calydonian boar-hunt) he acci­
dentally killed Eurytion (Pind. fr. 48), fled to Acastus at Iolcus, and
was purified by him (Apollod. 3.13.2). Then followed the attem pt to
seduce him by Acastus’ wife, on whom he subsequently achieved
revenge (Apollod. 3.13.3,7: ? from Pherecydes, RE 19.278).
T he killings served to move Peleus, who though Thessalian had
acquired an Aeginetan father and other local connections, up and
down the Greek world (RE 19.274,277).
Peliades: the consequences of their involuntary parricide in the
earliest versions are uncertain; perhaps the question was not raised.
T he Chest of Cypselus showed them as spectators at the funerary
Appendix 7 389
games of Pelias, but this detail is commonly taken (e.g. R E 19. 309 f.)
to reflect an earlier legend in which Pelias did not die at their hands
(cf. above s.v. Jason). Hyg. Fab. 24 and Paus. 8.11.1 make them flee;
Palaephatus 40 (Mythographi Graeci, 3.2, ed. N. Festa, Leipzig, 1902)
has them expelled by Acastus (cf. RE 19.310). Eur. Med. 504 f.
(perhaps not decisive evidence) envisages them still in Iolcus; nothing
in Apollod. 1.9.27. Subsequently we hear that the Iolcians forgave
them their involuntary act, and young noblemen took them in m arr­
iage ‘as being pure of bloodshed’ (90 FGrH fr. 54). Another version
even made Jason himself act as matchmaker for them (Diod. 4.53.2 =
Dionysius Skytobrachion, 32 FGrH fr. 14). The happy ending is, no
doubt, ‘sentim ental invention’ (Heldensage, 869), even though its exact
source is hard to define; Dionys. Skyt., loc. cit., is believed to be
dependent on post-Euripidean tragedy (Jacoby ad loc.), but 90 FGrH
fr. 54 cannot come from quite the same source.
Pelopids: some or all, were expelled (along with Hippodameia, Paus.
6.20.7) by Pelops for murdering their half-brother Chrvsippus (4
FGrH fr. 157, Thuc. 1.9.2, RE 3.2498 f., Heldensage, 217-19). The
names commonly mentioned are Atreus and Thyestes; 485 FGrH fr.
10 adds Alcathous, and schol. Eur. Or. 4 envisages a general diaspora
of the Pelopids.
W hatever its origin, the story had useful consequences in interpret­
ing or creating Pelopid links throughout the Peloponnese (Heldensage,
218). It explained, for instance, how the Pelopid Alcathous was at
M egara to build its walls (already Theog. 773).
Pelops: (1) the plague that followed his murder of Stymphalus (Apol­
lod. 3.12.6; Heldensage, 74 n. 5) was aitiological.
(2) According to Apollod. Epit. 2.8-9 Pelops was cleansed from
the blood of Myrtilus by Hephaestus (the choice of god is unex­
plained) at Ocean. The detail seems ancient; it might have stood in
Pherecydes, for whom the killing of Myrtilus was justified response to
a gross offence (3 FGrH fr. 37b) and so perhaps fit to be cleansed by a
god (Heldensage, 214; RE 16.1154; Jacoby on Pherecyd., loc. cit.).
Jacoby suggests that this was originally a purification from the blood
of Oenomaus, not Myrtilus; but it was Myrtilus whom Pelops killed
with his own hands. Concurrent versions of the saga ignored Myrtilus
(RE 16.1152).
390 Miasma
Penthesilea: came to Troy and was purified by Priam after acciden­
tally killing her sister while hunting (Apollod. Epit. 5.1, cf. Quint.
Smyrn. 1.24 f.).
T he Aethiopis may, but need not, have used this mechanical device
to explain her presence at Troy; 4 FGrH fr. 149 has a quite different
explanation (cf. Heldensage, 1176 f., R E Suppi. 7.870).
Perseus·, when he finally fulfils the prophecy and kills his grandfather,
Acrisius of Argos, by an accidental discus cast, he is ‘ashamed’ to
accept his inheritance and exchanges it with Megapenthes king of
Tiryns; he settles in Tiryns and founds Mycenae (3 FGrH fr. 12; Paus.
2.15.4, 2.16.2-3; Apollod. 2.4.4).
Acrisius’ death is of course essential to the folk-tale motif; but
Perseus’ reaction to it is used to bring him into the proper aitiological
connection with Mycenae (Heldensage, 237).
Phalces: a Tem enid, kills his sister Hyrnetho unintentionally at
Epidaurus, and flees to Sicyon (Paus. 2.6.7, 2.28.3-7; Heldensage,
667).
Explains cult of Hyrnetho at Epidaurus, and continues dispersion
ofT em enids (q.v.).
Poemander. king of Tanagra, accidentally killed his own son, fled to
Chalcis, where he was purified by Elpenor, and apparently then
returned to T anagra (Plut. Quaest. Graec. 37, 299c-e; a new variant of
the killing in P. Oxy. 2463.6-20).
Telamon·, see s.v. Peleus. The m urder of Phocus took him back from
Aegina to Salamis, mythologically his original or at least early
adopted home (Heldensage, 1043-5, Jacoby on 3 FGrH fr. 60).
Telegonus: son of Odysseus and Circe, slays his father in ignorance
(firstin the Telegonia). Nowhere do we hear of exile or purification; on
the contrary, in the Telegonia and perhaps Sophocles, the story ended
with a double m arriage and a heroization (RE 6 A 315; Pearson’s
introduction to Soph. Όόυσσενς Άκανθοπλήξ).
Telephus: killed his uncles the Aleads and on Delphi’s instructions fled
Appendix 7 391
to Mysia (Hyg. Fab. 244.2, Corp. Paroem. Graec. ed. Leutsch/Schneide-
win, 1.412) - the famous journey which he conducted in strict observ­
ance of the killer’s silence (Arist. Poet. 1460 a 32, Amphis, fr. 30.7,
Alexis, fr. 178.3; that the same homicide is in question is not demonstr­
able, Heldensage, 1146, but very plausible). Telephus ultimately
achieved kingship in Mysia; there is no record of a purification.
T he traditions about Telephus and Auge are, chiefly through
im aginative elaboration by all the tragedians, remarkably involved;
the killing is only one of the ways in which the anomaly of the Mysian
king’s Tegean birth was resolved (Heldensage, 1138-60).
Temenids: murdered their father Temenus, jealous of the honour he
paid to his son-in-law Deiphontes; the army expelled the Temenids in
consequence and made Deiphontes king (90 FGrH fr. 30, Diod.
7.13.1, Paus. 2.19.1, Apollod. 2.8.5; probably from Ephorus, cf. 70
FGrH fr. 18).
T he legend as we find it seems to derive from Euripides’ Τημενίόαι.
It explains the dispersion of Temenids through the Peloponnese
(.Heldensage, 665 f.; cf. s.v. Pelopids).
Theoclymenus: was probably invented for his role in the Odyssey. To
bring a M elampodid to Ithaca a murder was necessary.
Theseus: (1) on reaching the Cephisus, he was purified by the
Phytalids at the altar of Zeus Meilichios from the blood of ‘various
robbers, and Sinis his relative through Pittheus’ (Plut. Thes. 12.1,
Paus. 1.37.4; cf. R E Suppl. 13.1080, citing the vase JH S 56 (1936), 77
with Plate 5).
Aition for the Zeus Meilichios cult, and the role of the Phytalids in
cults of Theseus (cf. Töpffer, 249 f.).
(2) for the killing ofhis relatives the Pallantids (RESuppl. 13.1091 f.)
he was either tried at the newly instituted Delphinium (e.g. Pollux
8.119); which recognized his plea ofjustification, or went into a year’s
exile at Troizen (Eur. Hipp. 33-7; Paus. 1.22.2). The trial is an aition
of uncertain age, while the exile was probably Euripides’ invention to
take Theseus to Troizen (Eur. Hipp., ed. W. S. Barrett, Oxford, 1964,
p. 33; Jacoby on 328 FGrH fr. 108).
Thyestes: see s.v. Pelopids.
392 Miasma
Tlepolemus: killed Licymnius accidentally or in anger (R E 6 A 1615),
and fled to Rhodes with a large band of followers (Horn. II. 2.653-70).
A simple aition for the foundation of Rhodes by a Heraclid.
Tydeus: expelled from Calydon for shedding ‘kindred blood’ (Soph,
fr. 799.3, Eur. Supp. 148, fr. 558.2; details vary, cf. Apollod. 1.8.5,Ä£7
A 1705, Robert Oidipus, i, 140 f.); fled to Argos, was purified by
A drastus, and married his daughter (3 FGrH fr. 122).
T ydeus’ migration was already known to Homer (II. 14.113 AT.),
though not there explained. The obvious motive was subsequently
supplied (Robert, loc. cit.; speculations on Tydeus’ original home,
Heldensage, 924 f.).
T he theme of servitude as a form of expiation for blood-guilt appears
in several of those stories: Apollo (1), Cadmus, Heracles (1) and (3).
C adm us was actually enslaved to the father of his victim, while the
price of Heracles’ sale was paid to his victim’s relatives. It is known in
some cultures for the killer to be taken into his victim’s family to work
in his place; Glotz, p. 173, suggested that we have in these myths the
faint reflection of such a custom in Greece. But specific explanations
can be thought of for this form of penalty in each case.
The list of those who, but for the need to transfer them from one
mythological homeland to another (or other aitiological reasons),
need never have killed is a long one. One may dispute individual
cases, but it seems to include Aetolus, Amphitryon, Bellerophon,
Daedalus, Hyettus, Lycus and Nycteus, Oxylus, Patroclus, Peleus
and Telamon, Penthesilea, Theoclymenus, Tydeus, the Pelopids and
Tem enids. The m otif of the killer who, perhaps after consulting
Delphi, founded a foreign colony, was a natural development of this:
cf. Archias, C arnabas, Leucippus, Tlepolemus, Triopas. In some
cases, murders that had an independent place in a hero’s story were
also exploited aitiologically: cf. Athamas, Cephalus, Perseus, and
Oedipus.
A final point that deserves emphasis is the lack of uniformity in the
consequences of particular forms of killing. This is clearest, perhaps,
from the contrast between two involuntary parricides, Oedipus and
Telegonus, but numerous smaller instances emerge from the preced­
ing catalogue. The task of exploiting these stories as historical evi­
dence is thereby much complicated.
Appendix 8:
Gods Particularly Concerned with Purity
It is commonly assumed that the status of Apollo and Artemis in this
respect was unique. As we have seen, not all the arguments are very
strong. Being born or dying was forbidden in all temple precincts, and
contact with birth or death seems to have made the affected person
‘im pure’, not ‘impure in respect of Apollo’ (see pp. 33 and 37 above).
The god who presided over purification from killing was Zeus (p. 139
above), and if Apollo assumed the role of the human purifier for
Orestes, Hephaestus did the same for Pelops, and ‘all the gods’ for
Alcmaeon (see Appendix 7). O n the other hand, several items of
evidence, though individually inconclusive, suggest in combination
that certain forms of purity were particularly although not exclusively
required by the Delian gods. The dramatic motif of the god who
departs to avoid contact with death, used more widely in the fourth
century, was initially applied by Euripides to Apollo and Artemis (see
p. 33 n. 3). It was the purity of Delos that caused particular concern to
the Athenians. Above all, Apollo and Artemis are the gods who
dom inate the cathartic law of Cyrene (see Appendix 2). They prob­
ably owe this special position to their role as senders and healers of
disease. This is the function of Apollo that appears at the start of the
Cyrene cathartic law, and disease could be viewed as a form of
pollution (see C hapter 7). The connection of thought becomes almost
explicit when the Athenians purify Delos in response to plague.
O ther gods were perhaps distinguished in terms of their concern for
purity. On a general level this was a difference between gods of the
upper and lower worlds, since Hecate and the heroes were impure.
(Presumably, therefore, it was not necessary to approach them in a
state of purity.) About more detailed discriminations we can only
guess. O n Cos we find the priests of, at least, Demeter and Zeus
subject to stringent rules (see p. 52). Dionysus was involved with
purifications, although of a special kind (see Chapter 10). But the
attitude of country deities to sexual purity was relaxed (see p. 76).
 INDEXES
P R IN C IP A L PA SSA G ES
Where several passages are covered by a heading in the General Index (e.g. ‘Plato on homicide ’) they
are not listed separately here. Nor are the sourcesJor Appendix 7.
References to this book are printed in italics. An Asterisk indicates discussion o f a textual problem.
A chilles T a tiu s (4.7.7) I02n. 115; (6.12.3)
314 η. 26
A elian:
Ν Α (9.51, 65) 360f . , 363
VH (5.14) 364 n. 49; (8.5) 129 n. 97,279
n. 108; (12.50) 142 n. 162
F r. ( I I ) 3 3 n. 3; {39) 2 3 n. 24; (44) 81
n. 32,179
A elius A ristid es (4 8 .31) 3 0 n. 66
A eneas T a c tic u s (22.17) 159
A eschines:
1 (19) 153 nn. 4 6 -7 ; (28) 197; (183) 94
n. 82, 96 n. 88
9. (8 7 -8 ) 126 n. 86; (1 3 3 -4 ) 155 n. 58
3 (77) 64 n. 109; ( 107-112) 164 n. 114;
(1 1 1 )7 9 / n. 3; (1 1 8 -9 ) 166 n. 127;
(1 3 2 -4 ) 173 n. 166
(A eschines) Epistles (1.2) 33η. 46,43η. 43,
218; (2.5) 197η. 40
A eschylus:
Ag. ( 1505—8) 201; (1 6 4 4 -5 ) 111; (1645)
145 η. 8
Cho. (98) 230 η. 132; (2 6 9 -9 6 ) HO;
(2 7 8 -8 2 ) 218; (4 7 2 - 4) 201; (635) 8
η. 35; (909) 123 η. 71; (923) 117 η. 55;
(9 6 7 -8 ) 2 2 3 η. 87; (1 0 5 5 -6 ) 129η. 94
Eum. (4 0 - 5) 370; (6 2 -3 ) 139 η. 140;
(236) 224 η. 92; (238) 118; (2 7 8 -9 6 )
129; (280) 118; (285) 129 η. 97; (286)
118; (304) 6'; (448) 371 η. 6; (449 f.)
372; (9 0 2 -8 7 ) 257f . , 279f .
Sept. '(4 -6 ) 267; (6 7 9 -8 2 ) 113η. 37;
(6 8 1 -2 ) 137η. 135; (7 6 4 -5 ) 266;
(859) 33 η. 3
Supp. (2 2 5 -6 ) 98; (2 6 2 -7 ) 211 η. 24;
(375) 146 η. 12; (4 5 9 -7 9 ) 185; (6 4 6 -
50) 217 η. 56; (6 5 6 - 709) 257f . , 279f .
Theori/lsthmiastae ( 2 9 - ‘i \ ) 8 4 n . 4 2
F r. (137) 3 / 8 η. 52; (327) 372; (354) 133
η. I l l
A lcaeu s (F r.346.4) 101 n. I l l
Alcmaeonis 377
A le x a n d e r A etolus ( F r .3 .16) 314 η. 26
A lexis F r. (15.6) 323 η. 4; (7 6 .1 -4 ) 360
η. 17·, ( 178.3) 391
A m p h is F r. (20) 360η. 22; (30.7) 391
‘A n a c h a rs is’ (Epistle 9) 327 η. 28
A n a x a n d rid e s (F r.39.10) 175η. 177
A n d o cid es (1.29) 179 η. 193; (1.94) 114
η. 41, 323; ( 1 .9 6 -8 ) 366; (4.33) 98
η. 98
A n d ro tio n , FG rH 324 (Fr.30) 161 η. 99,
166 η. 130
Anthologia Palatina (7.406) 361 η. 23;
(14.71, 74) 324η. 14
A n tip h a n e s F r. (68.12—14) 360 nn. 17,22;
(129.6) 360 η. 17
A n tip h o n :
5 ( 1 1 ) 122; (1 1 -1 2 ) 187η. 241; (8 2 -4 )
9 η. 39,109, 129,191 η. 4. 3»7; (87) 119
η. 62; (93) 254
6 ( 1 ) 254; (4) 119,254; (37) 38 η. 20;
( 4 5 - 6 ) 268 η. 52
Tetr. (1/3 9) 204 η. 84; (2 a 2, γ 7 -8 )
117
A p o llo d o ru s, FG rH 244 (Fr.89) 83η. 37
A p o llo d o ru s (3.5.1) 260; (3.12.6) 274
A pollo n iu s R ho d iu s, Argon. 4 (4 52-745)
3 7 0 -4 , 383f . ; (4 7 7 -9 ) 108; (479) 133
n. U l; (6 9 9 - 717) 108; (710) 283η. 11;
( 1669 ff.) 251 η. 89
A rch ed ik o s (F r.4) 97,99
A rc h e stra tu s (ap. A th. 163d, 31 Oe) 360
n. 17
A risto p h an es:
Ach. (44) 21
Av. (4 6 3 -4 ) 20; (5 2 4 -5 ) 219 n. 67;
( 1490—3) 244 n. 50
Eccl. (128) 21 n. 12; (647) 99 n. 101;
(1033) 35 n. 10
396
Indexes
A risto p h an es: (coni.)
Eq. (4 4 5 -8 ) 206; ( 128 0 -9 ) 99n. 101,101
n. 111
Lys. (1 8 1 -2 3 7 ) 85 n. 46; (549) 363n. 44;
(6 4 1 -7 ) 79/.; (7 0 0 -2 ) 363n. 44;
(7 4 2 -3 ) 33 n. 5; (9 1 1 -3 ) 75, 76n. 8;
(9 1 4 -5 ) 186n. 235; (1129 f.) 2 2 n. 18
Nub. (243) 248 n. 68; (1321-1450) 196
Pax ( 151, 1 6 2 -3 ) 283 n. 9,359; ( 1250)
224 n. 92
Plut. (21 ) 153n. 46; (6 5 6 -8 ) 21 3 ». 31;
(845) 180 n. 201
Ran. (338) 2 8 3 n. 11; (355) 323n. 11;
(366) 162 n. 101; (630) 25 3 n. 105;
(1032) 143 n. 164.306
Thesm. (330) 82 n. 35; (3 3 2 -6 7 ) 193f .
Vesp. ( 118-24) 208,246; (394) 162n. 101;
(1037 fi'.) 24 8 n. 69; ( 1043) 211 n. 24
F r. (58 A u stin ) 243f . ; (32 0 -1 ) 83 n. 36;
(G92a) 244
A ristotle:
A th. Pol. (13.5) 262 n. 31,263n.33;
(16.10) 194,366/.; (56.4) 283n. 9;
(57.4) 159 n. 85; (60.2) 165n. 120
Pol. ( 1329a 2 7 -3 4 ) 8 7 n. 5 3 ,9 7 n. 92,175
n. 177; ( 1335b 1 2-16) 49 n. 63; ( 1335b
2 4 - 6 ) 50 n. 67
( De somniis 459b 2 3 -4 6 0 a 23) 102; (Rhet.
1418a 2 3 - 6 ) 210 n. 17; (Poet. 1448b
12) 357; (Poel. 1460a 32) 391
F r. (60) 300n. 99; (101 ) 3 5 n . 12,64
n. 107; (194) 362n. 35,363 n. 44; (496)
232 n. 153; (611.10) 65 n. 110
A risto x en u s (ed. VVehrli) Fr. (25) 364
n. 50; (26) 29 7 n. 83; (29a) 364 n. 50
A rrian :
Anabasis (1.9.7) 277η. 103; (1.9.9) 168
η. 1 3 4 ,176 η. 179; (4.9.5) 252
( Cyn. 33) 113η. 37; (diss. Epict. 3.21.16)
88 η. 55
A th e n a e u s ( 4 6 e - f) 37 η. 17; (150a) 85
n .4 3 ; ( \ l \ e ) 157η. 69; (461 c) 244 η. 50
B a b riu s (63) 244 η. 50
B acchylides (1 1 .9 5 - 110) 209 η. 14,213
η. 30
C a llim ach u s:
(Dem. 1 3 0 -2 ) 49η. 64; (Jov. 11-13) 49
η. 64; (Pall.) 2 7 / ; (Epigr. 9 Pf.) 43
F r. (6 3 .9 -1 2 ) 82n. 35; (1 9 4 .2 8 -3 1 )2 0 5
n. 7,228; (1 9 4 .3 7 -4 4 )5 5 « . ll,5 3 n .8 0 ,
228; (1 9 4 .4 0 -5 6 ) 229 n. 124;
(194.101 II.) 229 n. 124
Carmen Priapeum (14) 78 n. 18
C e n so rin u s (De die natali 1 1.7) 48 n. 58,
52 n. 74
C h a ire m o n (Achilles Thersitoktonos) 260
n. 22
C h a rito n (1.5.5) 47 n. 54
C h ry sip p u s (ap. Plut, de Stoic. Rep. 1044f-
1045a) 34,293 n. 59,326,359 n. 12
C le id e m u s, F G rH 323 (Fr. 14) 36 n. 15
C le m e n s A lex an d rin u s, Strom. (4.19
p. 302.1 - 3 ) 8 2 n. 33; (4.22. p. 311 ) 324
n. 15; (7 .4 .2 6 .2 -3 , vol. iii, p. 19)
225-32.
Com. Adesp. (239 A ustin F r. 1) 183n. 213
Com. Nov. Adesp. (Fr.214) 263n. 38
C ra tin u s (F r.221) 358
C u rtiu s R ufus (10.9. \ \ ) 23η. 21
D a m o p h ilu s, F G rH 70 (Fr.96) 172
nn. 164-5.
D em o c ritu s (B 262) 253 n. 99
D em o sth en es (9.44) 366,367n. 9; ( 12.2-4)
188 n. 249; ( 18.159) 268 n. 53; ( 18.259)
231; (1 8 .2 5 9 -6 0 ) 303; (18.296) 268
n. 53; (19.267) 317 n. 48; (20.158) 125
n. 82,367; (21.16) 151 n. 39; (21.53)
155; (21.126) 176 n. 180; (21.180) 158
n. 74; (22.2) 123n. 72; (22.78) 88n. 55,
97; (23.28) 366-7; (23.53) 366-7;
(23.60) 366-7; (23.72) 114,116n. 49,
373 n. 19,374 n. 24,387; (23.73) 121
n. 66; (23. 74) 141 n. 156,366-7;
(24.29, 31) 157; (24.55) 175 n. 175;
(25.11 )306; (25.30) 204 n. 80; (25.61)
194 n. 17; (37.59) 108n. 10,367;
(4 3 .5 7 -8 ) 3 8 n. 21;.(43.62) 36n. 14, 70
n. 123; (47.70) 3 8 n. 20, 41 n. 36,121
n. 66; (49.66) 6 n. 23; (54.39) 21 n. 12:
(57.55) 263n. 33; (5 9 .7 2 -1 1 7 ) 97,178
n. 190; (59.78) 8 5 n. 45; (5 9 .8 5 -7 ) 94
n. 82, 95 nn. 84,87; (59.92) 97 n. 92;
(5 9 .1 1 6 -7 ) 178 n. 189; (60.30) 64
n. 108
D in a rc h u s (2.5) 263 n. 38
D io d o ru s S iculus (3 .5 8 .2 -3 ) 224 η. 93,232
η. 153,288η. 38; (4.14.3) 284; (10.9.6)
297 η. 82,324 η. 15; ( 10.30) 46 η. 48;
( 1 2 .5 8 .6 - Ί ) 276; ( \ 3 .8 6 A - 3 ) 3 9 η .24;
(1 5 .4 8 -9 ) 176η. 181; (15.49.6) 10
η. 44; (1 6 .2 3 -3 9 , 5 6 -6 4 ) 172-5;
(16.58.6) 168 η. 133; '(16.61.3) 240
η. 25; (17.64.3) 58 η. 94; (32.12.2) 221
η. 75
Indexes
D iogenes L a e rtiu s (1.110) 210 η. 17,211
η. 23; (8.31 ) 297η. 82; (8.32) 217η. 54;
(8.33) 296 η. 76,297,299 η. 90; (8.34)
295 η. 68,361; (8.38) 299 η. 90; (8.43)
82 η. 33; (9.43) 37 η. 17
D ip h ilu s F r. (32.17) 263 η. 38; (126) 207,
2 2 5 -3 2
D o n a tu s (on T e r. Andr. 483.3) 51 n. 69
‘E k p h a n to s’ (p. 80. 15 ff. T hesleff) 266
n. 47
E m p ed o cles (B 110) 298 n. 87; (B 115.9-
12) 317 n. 44
E p h ip p u s, FG rH 126 (F r.3) 163n. 109,278
n. 104
E p ic h a rm u s F r. (63) 360 n. 19; (269) 324
Epigoni (F r.ii A llen) 131 n. 102,380
E rin n a (v. 19 Page, G L P 488) 5 3 n. 80
E u b u lu s (Fr. 14) 360 n. 22
E u p o lis (Demes 31—2 Page) 191 n. 3;
(Fr. 120) 221 n. 75
E u rip id es:
Aie. ( 2 2 -3 ) 33 n. 3; (98 ff.) 3 5 n. 10;
( 1 1 4 3 -6 ) 37 n. 17; ( 1146) 329 n. 10
Andr. (1 5 5 -6 0 ) 251 n. 89; (2 5 8 -6 0 ) 315;
(2 9 3 -4 ) 221 n. 75,230 n. 133; (3 3 5 //)
315; (6 1 4 -5 ) 111 n. 25; (6 5 4 -9 ) 122;
(975) 318 n. 53; (9 7 4 -6 ) 205 n. 86;
(9 7 7 -8 ) 3 1 6 n. 43
Bacch. (7 2 -7 ) 288-90
El. (256) 86 n. 51; (654) 52 n. 74;
(1 1 2 4 -3 3 ) 52 n. 74; (1 1 7 7 -9 ) 317
n. 44; (1 1 9 5 -7 ) 3 1 6 n. 42;
(1 1 9 8 -1 2 0 0 ) 2 0 5 n. 86; ( 1 2 9 3 - 7 ) 3 //
n. 17; ( 1 3 5 0 -5 ) 9 « . 39
Hec. (345) 6 n. 25; ( 1276) 219 nn. 6 4 -5
Hel. (9 8 5 -7 ) 185 n. 228; (1430) 3 8 n. 20
Heracl. (71) 145n. 9 ,1 4 6 n. 12; (2 5 5 -6 )
185n. 226; (2 5 9 f.) 183n. 215; (264)
145n. 8 ,1 4 6 n. 12; (5 5 8 -9 ) 108n. 10
H F (225) 211 n. 24; (722) 253n. 101;
(757) 145n. 8,146; (922 ff.) 10 n. 42,
114; (9 2 8 -9 ) 20 n. 7; (9 6 6 -7 ) 129 n. 94;
(1214 ff.) 371 n. 6; (1219) 371 n. 6;
( 1232) 145 n. 7; ( 1258-62) 200 n. 60;
(1 3 6 1 )7 2 3 « . 71
Hipp. (7 3 -8 1 ) 164,190; (141 ff.) 245,
252; (316) 129 n. 94; (316—9) 245;
(317) 323n. 11; (3 1 7 -8 )2 2 2 ; (8 3 1 -3 )
201 n. 67; (9 5 2 -4 ) 301,302n. 108,304;
(1 3 7 9 -8 1 ) 199 n. 52, 201 n. 67; (1415)
192; (1437 ff.) 33,67; (1447-51)
108 n. 10
397
IA ( 9 3 8 - 4 7 ) '/ / / ; (1 1 9 1 -2 ) 123 n. 71
Ion (80) 228n. 121; (4 3 4 -5 ) 228n. 118;
(9 3 6 -9 ) 7 6 n. 8; ( 1 118) 145n. 8:
(1334) 113n.37
I T (3 8 0 -4 ) 34,37; (4 6 8 -9 ) 157n. 72;
(693) 2 05n. 89; (7 9 8 -9 ) /7 5 « . 178;
(9 4 9 -5 7 ) 195n. 25; (1 0 4 0 -1 ) 5 3 n. 79;
(1174 ff.) 312; (1 1 7 6 -7 )5 3 « . 79;
(1193) 227 n. 108; ( 1199-1201) 27
n. 50, 53 n. 79; (1207) 371 n. 6; (1208-
10) 350; (1216) *228n. 118; (1218)
*108n. 14,110, 371 «. 7; (1 2 2 3 -4 ) 372
nn. 11,14; (1226) 350; (1 2 2 6 -9 ) 49;
(1 4 6 2 -3 ) 9 0 n .6 6
Med. (6 0 7 -8 ) 1 9 7 /; (6 6 5 -8 1 ) 8 6 n. 49;
( 1055) 253 n. 105; ( 1251 - 6 8 ) 315;
( 1 3 27-1407) 315; ( 1327-8) 317n. 44;
(1333) 108 n. 14
Or. ( 7 5 -6 ) 311; (339) 129 n. 94; (396)
254,310; (429 f.) 374 n. 26; (4 7 9 -8 1 )
311; (5 0 0 -4 ) 137 n. 133; (515) 121 n.
66; (526) 313 n. 25; (5 8 0 -4 ) 110 n. 20;
(7 9 2 -4 ) 309; (793) 129n. 98; (822) 317
n. 44; (1 6 0 0 -4 ) / / / ; (1604) 323n. 11
Phoen. (9 4 4 -5 ) 81 n. 29; (1050) 98 n. 96
Supp. (2 2 0 -8 ) 205 n. 89,219 n. 66
Tro. (4 1 -2 ) 93 n. 79; (2 5 1 -8 ) 93 n. 77;
(501) 86 n. 51; (1 0 2 3 -4 ) 317 n. 47
(Aeolus) 98; (Anliope 80, Page) 65 n. 110;
(Auge) 3 4 n. 7; (Cretans, Fr.79,
A u stin ) 33 n. 2 ,3 9 n. 23,142 n. 162,
289,301-2; (Melanippe) 221 n. 75;
(Oedipus, F r.98, A u stin ) 183n. 215;
(Stheneboea, prologue 2 2 -5 , v.
A rn im ) 134 n. 120, 372, 379
F r. (82) 107 n. 6, 110 n. 19; (292) 243;
(368) 145n. 8; (645.4) 123n. 76; (662)
101 n. 109; (912) 300 n. 99; (1008) 371
n. 6
F e stu s (p. 470.21, p. 476.20 L.) 373n. 20
F G rH (356 F r. 1) 134 n. 119,141 n. 151,283
n. 11, 371 n. 9; (532 D 2) 27 n. 50,53
n. 79,122 n. 67,185 n. 228
H elio d o ru s Aeth. (1.2.7) 33n. 3; (10.4.5)
102 n. 112
H ero d o tu s:
1 (1 9 -2 2 ) 250; (35) 123 n. 77,134 n. 121,
374; (44) 134 n. 120; (64.2) 73; (91) 202
n. 69; (1 5 7 -6 0 ) 185; (182) 93 n. 77;
(198) 77
398 Indexes
H ero d o tu s: (coni.)
2 (64) 74,326; (81)290; (81.2) 302η. 108;
(86.2) 64 η. 108; (175) 253η. 104
3 (22.4) 362 η. 33\ (47.3) 185 η. 226;
(5 0 -5 3 ) 123 ηη. 71, 7 7 ,194 η. 17
4 (1 5 4 .4 ) 155η. 55; (161.1)260«. 47
6 (56) 7 ,1 9 2 -3 ; (5 8 .1)41 η. 33; (58.3)
65η. 110; (86) 187; (91) 10η. 42,184,
191 η. 1; (106.3) 154 η. 53,159 η. 82;
(121.1) 206η. 95; (1 3 4 -6 ) 179;
(134.2) 81 η. 32
7 (39.3) 22 η. 20; (1 3 3 -7 ) 188- (134.2)
191 η. 4, 264; (137.1) 17; (137.2) 200
η. 59; (141.2) 185 η. 228; ( 169) 272
η. /5 (1 7 1 ) 272 η. 70; (197) 203η. 73,
259; (206.1) 154 η. 53; (2 2 0 .3 -4 ) 264
η. 40; (231) 194 η. 17
8. (54) 253
9 ( 9 3 .1 - 3 ) 176; (1 1 6 -2 0 ) 75η. 3
H esiod:
Op. (9 0 -1 0 4 ) 241-, ( 1 02-4) 236; (1 2 1 -6 )
244; ( 2 2 5 -4 7 ) 257f . , 266; (3 3 6 -7 ) 149
η. 2 6 ,150η. 34; (7 0 4 -5 ) 103η. 118;
(7 0 6 -6 4 ) 241; (7 2 4 -5 9 ) 291-4;
(7 3 3 -4 ) 76; (7 3 5 -6 ) 70; (7 5 3 -5 ) 103
F r. (3 0 .1 6 -1 9 ) 273η. 76; (37.14) 209
η. 14; (133) 218 η. 58
Scut. (11) 122η. 68; (13) 135η. 125
H esy ch iu s s.v. (Ά λμ υρ ίό ες) 4 7 η. 52; (ίν
Π νθίφ γία α ι ) 162 η. 101; (Κοίης) 284
η. 18,374 η. 29; (παναγείς, παναγία) 90
η. 68; (περίατιον) 21 η. 15,38η. 20
H ip p o c ra te s (Flat. 5, 6 (6.96, 98 L.) )
3 η. 10, 218; (M orb. Sacr. 148.38 J .,
1 .4 0 G ,)224n.92; (Morb.Sacr. 148.55
J . , 1 .4 6 G .) 19; (Viel. 2.46 (6 .5 4 4 -6
L .) ) 357: see also General Index
H ip p o ly tu s (Haer. 7 .2 9 - 30) 301 n. 104
H ip p o n a x F r. (6) 231 n. 145; (78) 258n. 8;
(92) 208, 258n. 8; (104.20) 77
H o m er:
II. ( 1.65) 273; ( 1 .3 1 4 -7 ) 210 n. 18,217,
2 2 9 n. 130; (3 .5 6 - 7) 195; (4.160-2)
201; (4 .2 3 4 -9 ) 187n. 243; (6 .1 3 5 -6 )
290 n. 45; (6 .2 6 6 -8 ) 19; (16.228-30)
6 7 /.; (1 6 .6 6 6 -8 3 ) 67; (16 .7 9 5 -9 ) 68;
(1 8 .2 3 -5 ) 68; (21.75) 182n. 207;
(21.83) 201; (2 1 .2 1 8 -2 1 ) 66/.;
(22.213) 67 n. 116; (22.358) 70;
(22.402 f.) 68; (2 3 .3 9 -4 1 ) 6 8 n. 120;
(2 3 .4 4 -6 ) 68; (2 3 .4 9 -5 3 ) 5 9 / ;
(23.579 fl.) 187n. 242; (24 .3 3 -7 6 ) 70;
(24.480) *135n. 124; (2 4 .5 0 5 -6 ) 122
n. 68; (2 4 .5 9 2 -5 ) 134 n. 116
Od. (3.215) 265; (4 .3 7 7 -8 ) 201; (5.394-
7) 240; (9 .1 9 7 -2 0 1 ) 176; (9.4 1 1 -2 )
240; ( 11.73) 70; (1 2 .3 4 0 -5 1 ) 254;
(1 2 .3 7 4 -4 1 9 ) 176; (1 9 .109-14) 265;
(1 9 .3 9 5 -6 ) 186; (2 1 .2 5 8 -9 ) 158;
(2 2 .3 1 0 -2 9 ) 182n. 210; (22.481-94)
114 n. 39; (22.481) 227 n. 114
H o ra c e (Ars Ρ Λ 1 \ ) 2 1 8 n. 60
Hymn Horn. Cer. (1 9 2 -6 ) 285
H y p e rid e s (Euxen. 1 4 -1 7 ) 160
Ia m b lic h u s, VP (68) 297 η. 83; (82) 299
η. 89; (83) 2%; (84) 295; (85) 299 η. 89,
362 η. 36; (106) 361 η. 29; ( 109) 361
η. 2 7 ,363η. 41; (110) 297 η. 83; (Myst.
3.10) 303 η. 112; (Protr. 21 ) 361-3,
ηη. 27,35, 41
Io n o f C h io s (Fr.30 W .) 291,299 η. 90
Isa e u s (6.49 f.) 178η. 191
Istro s, F G r H 334 (Fr.50) 259
L ivy ( 4 0 .6 .1 -5 ) 22 η. 19
L u c ia n ( Tim. 17) 89 n. 65
L y cu rg u s, Leocr. (79) 186; (1 1 2 - 115)
45n. 47; ( 1 17) 206; (133) 118
L y sias (1.14) 6 5n. 110; (2.7) 145n. 7; (6.4)
268 n. 52; (6.53) 259; ( 12.5) 263;
(13.79) 194n. 17; (13. 7 9 -8 7 ) 114;
(24.13) 153n. 47, 268n. 52; (26.8) 268
n. 52; (31.31) 194 n. 16; (32.13) 187;
(F r.53 T h alh eim ) 159 n. 84,
170 n. 1 4 6 ,2 3 9 /
M e la n th iu s , F G rH 326 (F r.2) 358
M e n a n d e r (Asp. 9 7 - 8 ) 33 n. 3; (Asp. 216
K .)3 3 n .2 ; (Asfr. 4 6 6 - 7 )3 5 n . 10; (Epit.
440) 8 0 ,8 5 n. 46; (Epit. 749 f.) 82n. 35;
(Epit. 8 8 0 -1 ) 248 n. 67; (Phasma 5 0 -
6) 207, 225-32; (Fr.394) 244 n. 50;
(F r.754) 359 n. 15
N e a n th e s ofC yzicus, F G rH 84 (F r. 16) 259
N ico lau s o f D am ascus, F G rH 90 Fr.(45)
114 n. 42, 123n. 77,204 n. 85,275n. 81;
(47.10) 2 0 2 n. 69; (52) 159n. 87; (61 )
123 n. 77, 368 n. 17
Orphicorum Fragmenta (ed. O . K ern) (156)
142 η. 162; (232) 300 η. 99; (291 ) 302
η. 108,362; (292) 143 η. 164, 299 η. 93,
306η. 124; (Τ . 219)502«. 108; (Τ.239)
307 η. 130
O r p h i c ’ Lithica (208-18)' 224 η. 93; ( 2 10)
225η. 97; (214 f.) 229η. 124; (591
(585) ) 223 η. 85
Indexes
O v id (Fast. 2 .4 5 -6 ) 131 η. 103,371 η. 9; I V)n IN, illi 11 I m i ill I I I , | i| * | u i
(Fast. 5 .6 8 1 -2 ) 10 η. 46; (Met. 10. il iIti, (II ΙΙΙ,ι li| Ii*I ii h>4
4 3 4 -5 ) 82 η. 33; (Met. 1 5.110-5) 364 (lll'lr II II Im ) . ' I l ., / : MI „I ,ι ‘I /
η. 50; (M et. 1 5 .3 2 2 -8 ) 230η. 131 n. W; (Il I h I ' n MU III 111,1 ! Ί
(869<l) ll'l i i ' U d i ,I .
P a rth e n iu s, Amat. Narr. (9.5) 261 η. 24; 872(1 8 7 ln ) H U lt<l Hlh n Ι Ί
(14.5) 123 η. 77 (881(1 <·) I'M il I l'il 1,1 I I , . . ,
P au san ias: (926c 927») 'l i n i , , |
1 (3 .4 ) 275 η. 88; (37.4) 374 η. 29; (43.3) 2 2 2 n. 80; (9'Π.ι I.) |·η Μ ι Ί
71 η. 126 η. 56; (947b cl) / lu 10 ι ί i ,|i .,
2 ( 10.4) 8 8 η. 58,90 η. 6 7 ,9 2 η. 73; ( 14.1 ) n.80, 70n. 123; (9 l'li 'i iii.n v.,.,
88η. 55; (20.2) 124 η. 78; (2 2 .6 -7 ) 85 (956a) 52 n. 78
η. 44; (24.1 ) 93 η. 77; (27) 33 η. 5,324 Menex. (238b) 64 η. 10/1
η. 17; (29.10) 119η. 61; (32.6) 275η. 86 Phd. (5 8 a—c) 153n. 46; (Ιι/ι . ΙιΊι ι ‘Il '
4 ( 1 2 .6 )5 2 « . 75 n. 6; ( m a ) 304 η. 1H,
5 (5.10) 230 η. 131; (5.11 ) 212,217η. 53; Phdr. (244e) 288n. 38; (2(i.'m) .7 in II
( 13.3) 3 9 η. 25; (14.5) 2 7 η. 46; (16.8) Resp. (363c—d) 286; (36 Ib ι·) '101 / .
372η. 16; (27.10) 1 Π η .5 4 (364c) 202; (364e) 299, M ) n W,
6 (1 1 .6 ) 117 η. 54; (20.9) 85 η. 43 (399e) 263 n. 38; (567c) 263 η. VI,
7 (2.1) 266η. 47; (25.13) 88 η. 58 (501a) 264 n. 39; (5 7 1 c -d ) 9 8 n. 99,
8 ( 4 1 . 7 - 9 ) 275 η. 86 327
9 (27.6) 9 3 η. 76; (20.4) 288η. 35; (39.7) Soph. (2 2 6 b -2 3 1 e ) 2 99n. 90; (226.1) 18
213 η. 31; (39.5) 215 η. 43 P lu tarc h :
10 ( 11.5) 275 η. 88; (31.9) 286 η. 26 Ages. (3.9) 8 6 n. 50; (29.7) 4 3 n. 42;
P eisa n d er, F G rH 16 (Fr. 10) 199η. 55 (3 0 .1 )2 7 7 « . 102
P etro n iu s, Sat. ( 1 0 4 -5 ) 2 9 3 η. 58; (134) Aie. (1 8 -2 1 ) 168-70; (23.9) 8 6 n. 50;
218 η. 60 (29.5) 176n. 179; (3 4 .1 -2 )2 6
P h erecrates (F r. 174) 349 Alex. (11.12) 176 n. 179; (13.4) 163
P h ilem o n (F r.79.19) 360 η. 19 η. 109,278η. 104; (16.2) 155n. 55;
P h iletas (Epigr. 1.5 G /P ) 102n. 114 (57.3) 220n. 72; (75) 220 n. 72
P h ilip p id es (F r.2 5 .2 -7 ) 269 Arat. (5 3 .2 -4 ) 4 3 n. 40
P h ilo ch o ru s, F G rH 328 F r. (86) 31 n. 68; Arist. (20.4) 23; (20.6) 71 n. 125; (2 5 .1)
(155) 161 n. 9 9 ,1 6 6 n. 130; (190) 30 186 n. 235
n. 63 Cim. ( 4 .5 -7 ) 9 8 n. 98; ( 6 .4 -7 ) 107,129
P h y larch u s, F G rH 81 (Fr.45) 4 7 n .5 3 n. 94, 277 n. 101
P in d a r (Ol. 7.77) 3 2 0 n. 68; (Pyth. 3 .4 3 -4 ) Dem. (21.3) 268 n. 54
67; (Pyth. 3 .7 6 -9 ) 247n. 65; (Fr. 133) Demetr. (30.2) 171 n. 155
300 n. 100 Dion (56.2) 254 n. 108
P lato com icus F r. (28) 364 n. 49; (173)292 Lyc. (27.1) 71; (27.4) 3 6 n. 16
n.52; (173.19) 360 n. 22 Lys. ( 8 .4 -5 ) 187; (30.1) 183n. 214
P lato: Nie. ( 16.7) 171 n. 155
Cra. (3 9 6 c -e ) 221; (400c) 300 n. 99; Num. (9.11) 88 n. 58,92 nn. 73, 75
(4 0 5 a -b ) 2 1 5 n. 45; (405b) 139n. 140 Pel. (33.5) 43 n. 40; (33.8) 39 n. 27
Ep. (329b) 6; (3 5 6 d -3 5 7 a ) 159n. 85,175 Per. (30.3) 188 n. 249; (3 3 A - 2 ) 206 “
n. 177 Phoc. (2 8 .2 - 3) 158n. 75; (37.2) 158;
Euthphr. (2d) 263 n. 38; (3 e -4 d ) 119 (3 7 .3 -4 ) 4 7 n.5 2
n. 63, 121 n. 66; ( 4 b -c ) 367; (4c) 111 Sol. (12) 21 l n . 23
n. 21 Sull. (35.2) 4 0 n .2 9
Euthyd. (277d) 247 n. 6 4 ,2 8 8 n. 38 Thes. (12.1) 139η. 143,373η. 19,374n.29
Grg. (4 9 3 a -b ) 286 nn. 2 6 -7 Timol. (22.2) 3 9 n. 24; (3 0 .7 -9 ) 10n. 44;
Leg. (7 1 6 d -e ) 323/.; (7 29e-730a) 182; (39.3) 4 3 n. 40
(7 3 5 a-7 3 6 c ) 264; (759c) 9 7 n. 92,175 De Superst. ( 166a) 220 n. 71; ( 170b) 2 2 2 /
n. 177, 205; (759d) 87 n. 53,92 n. 75; Apophth. Lac. (223e 1 1) 2 08n. 5; (238d)
(782c) 299 n. 9 3 ,3 0 2 n. 108; (8 0 0 d -e) 41 n. 33
400 Indexes
P lu ta rc h : (cont.)
De mul. vir. (252e) 261 η. 24
Quaesi. Rom. (5 .2 6 4 f-2 6 5 a) 6 0 Λ. /00-,
(51. 276Γ—277a) 2 1 7 η. 55;
(6 8 .2 8 0 b -c ) 230η. 136; (68.280c) 30
η. 65; (85.284t ) SO«. 215; (111 ,290d) 22
n. 19; (111,2 9 0 a -d ) 357 n. 5
Quaest. Graec. (2.291 e —f) 95 n. 87;
(24.297a) 3 5 n. 10; (26.297c) 265;
(40.300f) 85 n. 44, 279; (46.302b) 231
n. 142; (54.303c) 20 8 n. 10
De Pylh. or. (397a) 2 2 8 n. 121; (403f)
84 n. 40,87 n. 5 4,92 n. 75; (404a) 253
n. 105
Cons, ad Uxor. (61 ld ) 286
Quaest. Conv. (635e) 302 n. 108; (655d)
7 8 n. 15; ( 6 9 4 a -b ) 334; (700e) 103
n. 116; (7 2 8 c-7 3 0 f) 361 n. 28
Praec. Reip. Ger. (814b) 21 n. 16
Quaest. Nat. (36) 95 n. 87
(P lu ta rc h ) (Cons, ad Apoll. 118 c - 119d) 40
n. 29; (Par. M in. 19a.310b) 98 n. 96,
280n. 111; (Am. Narr. 773c-774d) 198
n. 48,277n. 101; ( X Oral. 8 3 3 a -4 a ) 45
n. 47
P M G (895) 160
P ollux ( 1.35) *90 n. 68
P o ly aen u s (Strat. 5.1 7.1) 284
P o lybius (4 .2 1 .8 -9 ) 22 n. 17,225;
(23.10.17) 22 n. 19
P orp h y ry :
Abst. (1.9)567; (1.14) 356 n. 3,364 n. 51;
(2 .1 3 -2 0 ) 323 n. 8; (2.19) 323 n. 3;
(2.44) 180n. 199; (2.50) 102n. 112;
(4.16 p. 255.6) 28 3 n. 9
VP (1) 298 n. 88; (12) 297 n. 82; (45) 297
n. 82
P o sid ip p u s F r. (1.5—6) 99; (26.21 ) 8 9n. 65
P ro p e rtiu s (4 .8 .8 3 -6 ) 95 n. 87
Scholia (A r. Ach. 747) 283 n. 10; (Ar. Plut.
845) 284; (Schol. P atm . D em . 2 3 .7 1)
368; (A D H om . 11.2. 3 3 3 -5 ) 382;
( T H o m ./ / . 13.589) 301 n. 102;
(L u cian p. 112.5 R abe) 8 9 n. 65;
(L u cian p. 276.5) 82 n. 33; (Lucian
p. 279.21) 8 9 n. 65; (L ucian
p. 2 8 0 .1 6 - 17) 8 3 n. 39; (L ucian
p. 2 8 0 .2 2 fr.)JÖ 5n.9; (Soph. OC477)
225 n. 97; (Soph. OC 680) 83 n. 36;
(T h eo cr. 2.11/12) 22 3 n. 86
S en eca (Ag. 163) 259n. 15; (Tro. 6 3 4 -5 )
259 n. 15
S erv iu s(o n V irg. Georg. 1.166, 2.389) 2811
n. 38
S im o n id es (P M G 5 3 1 .3 -4 ) 43 n. 41
S olon F r. (9.3) 270; (1 3 .2 3 -3 2 ) 199
S ophocles:
A j. ( i 7 2 -8 6 ) 246,252; ( 1 84-5) 243n. 45;
(655 f.) 2 1 7 n. 54; (7 5 6 - 77) 246,252
Ant. (1 9 6 -7 ) 329; (256) 8 n . 34,192;
(545) 329; (775) 6n. 2 2 ,3 2 8 n. 4;
(7 7 5 -6 ) U l; (889) 312,316,323;
(9 9 9 -1 0 4 7 ) 33 ,4 4 ,6 5 /.; (1 0 4 3 -4 ) 145
n. 7,310; ( 1070-1 ) 62; ( 1144) 290
n. 45; ( 1317-46) 316
E l. (84) 35 n. 11; (434) 3 5 n .U
O C (292) 25 3 n. 101; (367 (Γ.) 204; (407)
123 n. 73; (4 6 6 -9 2 ) 10 n. 42; (466) 4,
146; (4 9 0 -2 ) 146,195n. 25; (548) U l,
124; (9 4 1 -9 ) 118n. 58,316n. 43; (964-
5) 201,252 n. 96; ( 1132- 5) 310,316 n.
38; (1 4 8 2 -4 ) 17
O T (181) 219 n. 68; (1 9 4 -7 ) 230 n. 132;
(2 0 2 -6 ) 335; (2 3 6 -7 5 ) 193f .;
(2 3 6 -4 1 )5 5 ; (2 6 9 -7 2 ) 191 n. 3; (313)
107; (656) 6; (833) 219n. 66; (864) 150
n. 34, 323 n. 7; (1 4 2 4 -8 ) 310; (1426) 8
n. 35; ( 1486 IT.) 205; ( 1492-1502) 205
n. 86
(Phil. 758 f.) 24 8 n. 68; (Tr. 1012) 211
n. 24; (Tr. 1201 f.) 192 n. 11 (Fr. 34)
208; (F r. 734) 231 n. 141
S o p h ro n (F r.68, 70) 248 n. 69
S tesich o ru s (Fr.223) 202n. 70
S tra b o (7.7.12) 9 3 n. 77; (8.3.19, p. 346)
2 1 3 n. 29; (8.6.8, p. 371 ) 290n. 45
SV F :
i ( 2 5 3 - 6 ) 326; (256) 100 n. 104; (264-7)
327 n. 28
iii (7 4 3 -5 2 ) 326; (743T 6) 100 n. 104
T a c itu s (Ann. 3.60) 183 η. 216
T h e o c ritu s (2 .1 2 -1 6 ) 223 η. 85; (5.121)
223 η. 85; (7 .1 0 7 -8 ) 231 η. 145;
(2 4 .8 8 -1 0 0 ) 225-32; (2 4 .8 9 -9 2 ) 221
η. 75; (27.5) 314 η. 26
T heodorus P riscianus (Physica
p. 2 5 1 .2 -5 )2 5 ?
I'heognis (7 3 1 -4 2 ) 200η. 59
T heophrastus:
Char. (16) 225-32,307; (16.7) 222;
( 16.9) 39 η. 23,51 η. 73; ( 16.14) 220 η.
72; (16.15) 219 η. 67
Hist. Pl. (7.12.1 ) 231-2'; (7.13.4) 231-2;
(9.8.5) 292; (9.8.7) 152-3; (9.10.4)
216,224 η. 93
Indexes
T h u cy d id es:
1 ( 126 .2 -3 5 .1 ) 183f . ; (139.2) 166η. 128
2 ( 1 3 .4 -5 ) 173; (1 7 .1 -2 ) 164 η. 115;
(5 1 .4 -6 ) 219 η. 68, 220; (52.3) 33 η. 5;
(64.2) 280
3 (56.2) 156; (58) 122η. 68; (65.1) 156;
(7 0 .4 -5 ) 165η. 123; (81.3) 185η. 227;
(104.1—2) 163 η. 107
4 (9 7 .2 -9 9 ) 44,190 η. 255; (97.3)
162 η. 102
5 ( 1 ) 203; (16.1) 253 η. 102; (32.1 ) 253
η. 102; (49.5) 175η. 176; (54.3) 155
η. 55
7 (18.2) 188,253 η. 102; (50.4) 253 η. 104
T im a e u s, F G rH 566 Fr. (29) 268 η. 55;
(56) 213η. 31; (101) 47η. 53; ( 146) 221
η. 75
Trag. Adesp. (F r.358) 315 η. 35
V aleriu s Flaccus, Arg.3 (4 3 9 -4 3 ) 226;
(4 44-458) 373 η. 19
X en o p h o n :
Ages. (5.7) 75η. 3; ( 1 1.2) 323
IN S C R IP T IO N S
Altertümer von Pergamon viii 3 (p. 168.11 —
14) 74 n. 4,359 n. 12
B C H (51, 1927, 120) 324 n. 15; (60, 1936,
182 f.) 161 n. 97; (102, 1978, 326) 37
n. 17,50 n. 67, 74 n. 4,102 n. 112,322
η. 1 ,3 5 3 -5 ,3 5 9 n. 12
B uck (17) 185; (64) = Z iehen 61
B ull. Epig. (69, 1956, η. 110) 230 n. 131
Chiron (11, 1 9 8 1 ,7 )7 9 5
C R Acad. Inscr. (1916, 263 f.) 359 n. 12
Der E id von Plataiai, ed. P. Siew ert (50-1 )
7,191
Die Inscrijten von llion (25.86) 3 n . 10
Epigraphica, ed. H . W . Pleket (i n. 43)
162 n. 104
Hesperia (11, 1942, p. 265 n. 51) 89 n. 65
401
An. (4.5.35) 5 2 n. 78,176; (5.3.13)255
n. 105; (5 .4 .3 3 -4 ) 76n. 7; (5 .7 .1 3 -3 5 )
2 2 /.; (5.7.35) 124 n. 78; (6 .4 .9 -1 3 ) 42
n. 38; ( 7 .8 .1 - 6 ) 2 5 0 n. 85
Hell.
1 (2.15) 219n. 66; (4.12) 26.158; (7.20)
4 7 n. 52; (7.22) 4 5 n. 47; (7.35) 194
n. 17
2 ( 3 . 2 \ ) 262n .30; (3 .2 3 ,2 6 ,5 1 ) 195n .24
3 ( 1.9) 196n. 27; (3.1 ) 6 5 n. 110; (3.3)266
n. 47; (4.11) 187 n. 243; (5.24) 44 n. 46
4 ( 4 .2 - 4 ) 159 n. 89; (5 .1 -2 ) 155n. 58;
(7 .2 -3 ) 155f .
5 ( 4 .1 ) 188
7 (4.34) 199
Lac. ( 9 .4 -6 ) 194 ». 17; (14.4) 219 n. 62,
263
Mem. (3.8.10) 162n. 103; (3.12.6) 243
n. 44; (3.13.3)2 1 3 n .3 1 ; (4 .4 .1 9 -2 3 )
100 n. 104
(Cyn. 5.25) 163η. 108,357η. 5; (Cyr.
8.7.18) 107,129 n. 94; (Hiero 4 .4 -5 )
129,368; (Symp. 1.4) 281 n. 3
(X en o p h o n ) Ath. (2.6) 2 57n. 3; (3 .2 -8 )
157 n. 68
Z eno: see S V F
Inscr. Cos (319) 253 n. 105
Inscr. Cret. (4.76) 38 n. 21; (4.146) = LSS
114
IG
I 3( 1 A 14) 2 7 n. 46; (6 C 48) 8 9 n. 62; (7)
26 nn. 40,42; (35) 89 n. 60; (45) 183
n. 216; (52 A 18 -2 2 ) 171; (7 8 .5 4 -7 )
164 n. 115,165 n. 121; (84) 161,162
n. 105; (1 0 2 .3 0 -2 ) 368; (104.20) 125;
(257) 229 η. 130,293η. 59
I I 2 (1035.10) 33 n. 5 ,1 6 2 n. 106; (1316)
89 n. 65; (1 6 3 5 .1 3 4 -4 0 ) 176 n. 181;
(1 6 7 2 .1 2 6 -7 ) 30 n. 66; (2342.31) <99
n. 59; (2501) 161 n. 97,162 n. 104;
(2874) 90 n. 66; (3462) 89 n. 64; (3512)
88 n. 55; (3606.15) 90 n. 68; (3607) 89
n. 65; (3629) 89 n. 62; (3725) 89 n. 65;
(4076) 89 n.62; (4851) 89 n. 62
I V 2 (123) 249 n. 73
X I I (5.569) 2 93n. 59; (5.593) = LSCG
97
402 Indexes Indexes 403

IG (coni.) (9 6 5 .1 5 -1 7 ) 161 n. 100; ( 1161 ) 250; P. Steinleitner, Die Beicht im Zusammenhange

n. 23; ( Π ) * 3 9 n. 23; (2 4 -3 2 ) 53; (2 4 -
( 1168.1 ) 33 n. 5; ( 1168.47-55) 213 n. mit der sakralen Rechtspflege in der Antike,
X I V (645.137) 162η. 106; (865) 5) 27 n. 50; (3 3 -6 ) 42 n. 37, 52 n. 77,
198 η. 46 185 n. 228 31; (1168. vii a n d xxxvi) 249 n. 73; L eip zig 1913 [passim) 254f .
(1184.7) 253 n. 105; (1218) =
156 (A 7 -1 6 ) 52; (A 8 -1 0 ) 39 n. 25: IA
LSCG 97; (1236) 253 n. 105 Z ie h e n (61) 74 n. 3,144 n. 3
L S A (12) 37 η. 17,50 η. 67, 74 η. 4; ( 16) 41 11 ) 3 7 n. 17; (A 1 2 -3 ) 51 n. 73; (A 13)
η. 34, 65η. 110,191 η. 4; (18) 37η. 17, 50 n .6 7 ; (A 14)372««. 13,16; (A 15) S o lm sen /F raen k el4 (5) 185; (3C) = L SS
74 η. 4- (20) 74 η. 4,325,355; (23.8) 288 228 n. 118,231 n. 141; (B 2 9 -3 5 ) 88 115
η. 38; (29) 37η. 17, 74 η. 4; (36.36) 30 n. 56
η. 66-, (42 Α) 84 η. 40; (51) 37η. 17,50 L S S (I) = IG I 3 1; (4) = 7 G I 3257; (24)
η. 67, 74 η. 4,352; (52 Β 10) 52 η. 74; 170 n. 149; (27) 170n. 149; (28) 83n. GREEK
(56.11) 139 η. 143; (6 1 .8 -9 ) 81 η. 32; 3 6 ,1 4 5 n. 6; (31 )37n. 17,145n. 6; (32)
(73) 97η. 9 2 ,175η. 177; (74) 170 8 3 n. 36,144 n. 5; (33) 83 n. 36,144 n. 5;
η. 149; ( 79) 27η. 5 0 ,88η. 5 6 ,283η. 11; άγίζω 328 f. καθαγίζω 328 Γ.
(38 A 32) 283n. 11; (50) 293 n. 59; (54)
(83) 33 η. 5; (84) 37η. 17, 302 η. 108, άγιος 147 η. 16, 329 καθαίρω 4, 227 η. 114
50 n. 67, 74 n. 4,102 n. 112,354-5,359
3 5 4 -6 ,3 5 9 η. 12 n. 12; (59) '74 n. 4,324 n. 15,359 n. 12; αγιστεΰω 289, 329 κάθαρμα 229 η. 130, 259
L S C Q (5) = IG I 3 78; ( 14) = /G 3 84; (63) 84 n. 40; (64) 43 n. 41,253 n. 105; άγνίζω 329 καθαρμός 4, 18
( 15) = ÏG I 3 7; (32.23 ff.) *161 η. 99, (65) 8 3 n. 37,139 n. 143; (69) 158n. 77; άγνίτης 135 η. 124 καθαρός 323, 367
163 η. 111; (32.58) 145 η. 6; (36.5) άγνός 12, 1 4 7 -5 1 ,3 2 3 κά θα ρσις (m edical) 55 η. 87, 213 f.
(72 A 5) 25 3 n. 105; (82) 324 n. 15;
306/.; (3 9 .2 3 -4 , 26) 27 η. 47, 30 η. 66; (86.3) 324 n. 15; (88) 85 n. 44; (91) 37 α γο ς 5 - 1 2 , 328 καθοσιώ 329
(47) 161 η. 9 7 ,162η. 104; (55) 37η. 17, n. 17,50 n. 67,52 n. 74, 74 n. 4,102 n. αίσχύνω 3 η. 8, 95 η. 84 κ α κό της 293 η. 60
74 η. 4,102 η. 112, 322 η. 1,354f . , 359 112,324 n. 15,354/.; ( 106) 37 αιτιώ μαι έμαντόν 253 η. 105 λαικά ζω 99 η. 101
η. 12; (56) 3 η. 10, 3 7 η. 17,112,223 άκαθαρσία 214 λοιμος 257
n. 17; ( 108) 74 n. 4,324 n. 15,359n. 12;
η. 87; (58.12 f.) 27η. 47; (60) 171 (112 ) 112,322 η. 1, 357 n. 5; ( 114) 21 άλάστω ρ 15, 109, 224 η. 92 λ ο υ τρ ό ν $ 5 η. 11
η. 155; (63.10) 81 η. 32; ( 6 5 .1 2 - 1 3 ) « n. 1 4,22 n. 19; (115) see below; (117) ά λιτήριος 109, 268, 270 λυμαίνομαι 195 η. 24
η. 62; (6 5 .1 6 -2 3 ) 83η. 36; (65.23) 52 άναπίμπλημι 219 f. μελαγχολώ 246 η. 6 1 , 248 η. 67
170 n. 149 (119) 37 n. 17,50 n. 67, 74 n.
η. 78; (65.37) 20η. 9; (6 5 .5 0 ,6 6 ,6 7 f.) άποόιοπομποϋμαι 29, 373 μιαίνω 3
4 ,102 η. 112,354-6; (120) 302
30 η. 66; (6 5 .6 6 -8 ) 30η. 66; (65.107- n. 110;( 133) 177 άπόνιμμα 36 η. 15 μια ρό ς 3 - 5
12) 20η. 9; (68) 83η. 36,144 η. 5; (76) άπ οτρόπα ιος 220 η. 71, 334 μία σ μα 3 f., 12 f.
115 Appendix2; also (A 1 -3 ) 140,393;
145 η. 6; (77 D 13) 52 η. 74; (78.15- άραίος 192 η. 11 νόσος 220
(A 4 - 7) 275; (A 1 6-20) 37«. 1 7 ,4 9 /,
21) 166η. 127; (79) 176η. 182; (82) 85 άρεστήριον 145 η. 6 ό ξυθύμια 30
54; (A 2 1 - 5 )3 9 n .2 5 ; ( B 2 4 - 7 ) 40,
η. 44; (83.40) 88 η. 56; (95.5) 74 η. 4, ανθέντης 122 ό ργά ς 164 η. 113
4 9 / , (B 50 ff.) 134,371,373n. 19
359 η. 12; (96.9) 85; (97) 34-41, 53,54 αντοφ όνος 350 f. οσία 338
η. 81,58,69; (97 A 2 8 - 9 ) *40η. 30; (97 M A M A (iv 2 7 9 - 90) 2 5 4 / άφαγνίζω 329 η. 10 όσιος 323, 330
Β 5) *38η. 22; ( 108) 229η. 130; (109) M ichel (5 2 4 C 1 )3 « . 10 άφοσιώ , - οϋμαι 121, 330 f. όσιώ 121,330
85 η. 44; (1 1 6 .2 2 -5 ) 170 η. 149; (124) M /L (1 3 .1 2 -1 4 ) 132 n. 107,180 n. 202,196 βάπτω 306 η. 125 παλαμναίος 108
36η. 15,37η. 17,50 η. 67,52η. 74, η. 78, βοηθώ τώ θεώ 165 η. 119 π α ν α γ ή ς 328
n. 29; (30) 193-5,222 n. 80
74 η. 4,85η. 44,354/.; ( 130) 253η. 105; όαιμονώ 246, 248 περικαθαίρω 222 η. 80, 225 f.
( 136) 145 η. 6,165 η. 121; ( 139) 37 έγχντρίσ τρ ια 36 η. 15, 374 η. 29 περιρραντήριον 19
S ch w y zer (272) 198n. 46; (412) = Ziehen
η. 17,74η. 4,324 η. 1 5 ,3 5 4 -6 ,359η. 12; 61; (661)185 εκθνμα, έκθνομαι 10 η. 42 περιστία ρχος 21
(149) 145 η. 5; (150 A 5) 165 η. 121; S E G (iv 64) 116 n. 46; (ix 72) = L SS 115; έκκαθαίρω 263 προσπερμεία 231 η. 141
( 151 A 4 2 - 4 ) 75η. 6 ,8 6 η. 48,94 η. 81; (xii 80) 89 n. 60; (xii 87) 204,368; (xiii έκμιαίνομαι 76 η. 9 προστρόπαιος 108
( 151 Β 23 ) 180 η. 198,227η. 108; ( 152) έλατήριος 214 η. 34 $ άκος 102 η. 113
5 2 1 .1 8 0 -2 0 2 ) 293 n. 59; (xix427) 139
145 η. 6,293 η. 59; ( 154) see below; έλαννω, έξελαύνω 223 η. 87 σκα το φ ά γο ς 360
n. 144,141,279; (xxiv 116) 89 n. 59;
ένα γής see άγος σννανθρω πενω 364
( 156) see below; ( 157 A 2) 372 n. '16; (xxv 447.6) 3 «. 10; (xxvi 121) = IG
(166.9) 175 n. 177; (1 7 1 .1 6 -1 7 ) 50 I I 21035; (xxvi 136.5 2 -4 ) 26». 42; έναγίζω 328 f. σ ννείό η σ ις2 5 3 η. 105
n. 67, 74 n. 4 ,3 5 2 - 5 ένθνμιος, ένθυμοϋμαι 252 Γ. τραγωόώ 15
(xxvi 1306.23-6) 3 n . 10, 195n. 23a,
1 5 4 A (1 4 )2 5 3 « . 105; (1 6 -1 8 ) 52 n. 76; έξά γιοτος 328 ύ ό ρ α νό ς284
368; (xxvi 1139) 287; (xxviii 421) =
έπα κτός 222 η. 79, 348 φαρμακεύω , φάρμακον 214, 222 η. 80
(2 1 -4 5 ) 52; (22, 37) 3 9 n. 25; (24, 39) B C H 102, 1978, 326; (xxviii 841.3)
50 n. 67,51 n. 73; (2 4 -6 ) 37 n. 17 (27) έπι- co m p o u n d s, o f m agic 348 φ αρμακός 24—6, 258 Γ.
290 n. 45
52 n. 78; (29, 30, 44) 228 n. 118,231 θνησείόια 52 η. 78, 358 φθορά 3 5 4 - 6
S G D I (1153) 7; (1 5 6 1 -1 5 8 7 ) 250; (5398)
n. 141; (39—41) 3 7 n. 17 θρόνωσις 285, 373 f. φοιβ- 139 η. 140
= L SC G 97
154 B ( 1- 1 6 ) 145 n. 6; ( 2 ,6 , 15, 26) 228 S IG 3 (360) 193,194 n. 16; (711) 26 n. 37; ιερομην ία 154—8 χέρνιψ 35 η. 11
n. 118,231 n. 141; ( 1 7-32) 38 n. 21,39 (9 4 3 .7 -1 0 ) 219n. 68; (963) 161 n. 97; ιερός 151 f.
404 Indexes
GENERAL
NamesJ'rom Appendix 7, which is arranged alphabetically, are not included.
A b a ris 209
Abaton 167
A b o rtio n 325, 3 5 4 -6
A chilles, kills T h ersites 130 f.
A chilles T a tiu s , virgin sacrifice in 259
n. 15
Acousmata, P y th ag o rean 294—6, 298
A d u ltery 75 n. 4, 9 4 - 7 , 325
A elian , on divine vengeance 179 n. 193
A eschines, on D em osthenes as pollution
268 I'.; religious a ttitu d e s in 14 n. 60,
16 n. 73, 128 n. 90, 187 n. 241
Speeches 1 a n d 2, o n sexual pollution
9 4 -7
A eschylus, an d institu tio n s 312;
co m m u n al m oral responsibility in
279 f.; on fam ily curses 199 n. 53; on
th e prosperous city 257 Γ.
Choephori, pollution th reaten in g
O re ste s in 110, 129
Eumenides, E rinyes in 107 f., 126, 196,
2 79, 312; purification o f O restes
139 f., 3 8 6 -8
Septem, fratricid e in 137
Supplices, expressions for pollution in 5
n. 21, 8, 9; incest in? 98 n. 99; threat
o f pollution in 185, 279, 312, 315
A esop, as scap eg o at 260; d e a th o f 274
Aethiopis, p urification in 131 n. 102,
1 3 8 -4 0 , 373 n. 19
A gesilaus, lam eness o f 277
A gis, king 86
Agora, b u rial in 42, 337 f'.; p u rity o f 1.9, 125
A g rai, m ysteries o f 284 f., 373 f.
A g ric u ltu re , a n d sexual p u rity 77
Aidos 189
A igeus 86
A k a m a n tia 336—8
A lcm aeon 124, 136,377
A lcm aeo n id ai 16 f., 131, 204, 206, 211
ii. 23, 270
A lem an , Parlheneion 1 80
Alêtrides 80
A lex an d er, m u rd e rs C leitus 252
A lta rs, m u rd e r a t 184 n. 223
A ly attes, d isease o f 250
A m p h ia ra u s, d e a th o f 43 n. 42;
purificatio n s in cult o f 213 n. 3 !,
359 n . I I; sacred land o f 160
Amphidromia 51
Anairesis, d en ial o f 44 n. 46
A n a th e m a 7 n. 30
Anchisteia 40
A n d o cid es, on events o f4 1 5 168-70;
religious attitu d e s in 16 n. 73
'Angelos' ( = H ecate) 223 n. 86
A n ig ru s, m a rsh 212 f.
A n im a ls, in G reek religion 35 7 -6 4 ;
‘d u n g -e a tin g ’ 360; sacred 176;
sacrificial, ran g e o f 364
‘A n n o u n c e r’, o f pollution 350
A n th e s te ria 39, 85, 287 f.
A n th ro p o g o n y , O rp h ic 299 f.
A n tip h o n , Speeches 1, 5, 6, religious
a rg u m e n ts in 119, 126 f., 254
Tetralogies, enthumëmata in 253; on
‘a c c id e n ts’ 117; pollution in 104-10,
127, 129 f., 278
Aphrodisia, o f m agistrates 85 n. 43
A pollo, a n d arch aic healers 209; and
in p lag u e 275 f.; a n d p urity 393; an d
S a rp e d o n 67; an d T h arg elia 25
A p o tro p aio s 334 f.; D elphinios 141 f.;
inassociable w ith g rief 33 n. 3, 67
nn. 114, 116; N om ios 244 f.; o f
D elp h i 138—43; purified 378
A p p les, a n d D em eter 361—3
A ra tu s o f Sicyon, burial o f 42
A reo p ag u s, m ythical origin of379, 386;
sessions on im p u re days 159;
supervision o f religion 118, 178
A res 85, 244 f., 358 . C ro p -fa ilu re 130, 257, 2 7 1 -5 ; see also
A rgives, devious 13, 155
A risto p h a n e s, contagious qualities in
219; expressions for m adness in 246
n. 61, 248 n. 67; on disease 243;
religious outlook o f 14
Lysistrata, L y sistrata an d M yrrhine in
89
Nubes, values in 189, 196
A risto tle, on parricid e 124; on tragic
kalharsis 288 f., 297 n. 83
A rm ies, purification o f 22 f., 226
A rrêphoroi 80
A rte m is, a n d brides 345 f.; an d purity
393; 'b e a rs ’ of80, 345 f.; H em era, at
L ousoi 213
A sclepieia, b ath in g a t 213 n. 31
Indexes
A sclep iu s, a n d irra tio n a lity 249; an d
m o rality 248 f.; b ro u g h t to A thens
275; see also E p id a u ru s
A te 16 n. 73
A th a m a s , as scap eg o at 259
A th letes, sexual ab stin en ce o f8 4 n. 42
A tim ia 19, 46, 9 4 - 6 , 197; as outlaw ry
194 f., 204; h ered itary 204, 339
B a b y lo n ian p u rificatio n 373 n. 20
B acis 209
B a p tim ism , in rite so fC o ty to ? 3 0 6 n . 125;
n o t a t E leusis 284 n. 13
B a rb a ria n s, purified ofT23
B arley -g ro ats 227
B a rth F. 363
B a th , after b irth 50; after funeral 36;
before ritu a l 20
B a th s, a b ste n tio n from 215; healing 212 f.
B a ttle, p u rificatio n after? 113 n. 37
B a ttu s 179, 3 3 6 - 8
B ean s 3 0 1 ,3 0 2 , 3 5 8 -6 5
‘B e a rs’: see A rtem is
B eck ett S. 50 n. 68
Bees, hostility to sexuality 77, 8 3 ,9 5
B estiality 355
B ew itch m en t 2 2 2 -4 , 251, 348 f.;
p u rificatio n from 222, 372
B irth , pollu tio n o f C h . 2, esp. 4 8 -5 2 ; 336,
353
B irth a n d d e a th , av o id an ce o f contact
w ith 33 f., 52 f., 289, 296, 302, 307
B irth d a y s, d u rin g T h a rg e lia 25
B lack bile, an d m ad n ess 2 4 6 ,2 4 8 n. 67
B lood, p u rificatio n by 230, 3 7 1 -3
Blood-feud 125
B lood-m oney 116, 131
Boulé, p u rificatio n o f 21
B oys, in ritu a l 81 n. ‘28
‘Boy from th e h e a rth ’ 81 n. 28
B ra n -m a sh 231
B rides, ritu a l d u ties o f 345 f.
B ron ze ‘2 28 n. 118
B u c k th o rn 231
B u rial, in agora 42; in tra m u ra l 7 0 -7 3 ; o f
p u rificato ry relics 229 f.; pollution o f
37 n. 17; refusal o f 45, 70, 170, 190,
195; rig h t to 44, 327
B u rn in g , o f p o llu tin g objects 221
B u tch e rs ‘2 98
B u tler S. 314
C a b iri 223 n. 86, ‘2 84 n. 18
C a m b y ses, m ad n ess o f 243
405
C a m u s A. 60
C a n n ib a lis m 305, 326, 360; m etap h o rical
362
C a s s a n d ra , an d A pollo 93; rap e o f 185,
202 f., 273
C a te g o ries, v iolation o f 62, 189
C h e k h o v , A. 314
C h ild re n , b u rial o f 41, 72; ritu a l roles o f
7 9 -8 1
C h ristia n ity , an d purificatio n s 234, 324 f.
C h ry sip p u s , on rules o f p u rity 34, 322,
326
C im o n , his incest w ith E lpinice 98, 270
C in e sia s 239 f.
C ir r h a e a n p lain 164, 166
C itiz e n sh ip , exclusivity o f 262 f.
C leisth en es 16
C leo m en es, m ad n ess o f 242
C lo th in g , p u rity o f 52, 68
C o d ru s 260
C o llectiv e resp o n sib ility , tw o form s 278
C o lo n iz a tio n , ‘p u rifies’ city 264 n. 39
C on fessio n 236 f., 249 n. 73, 254 f.
C o n fessio n -in scrip tio n s, L ydo-P hrygian
254 f.
C o n scien ce 2 5 2 -4
‘C o n s e c ra tio n ’, by d estru ctio n 328 f.;
p u n itiv e 6 - 1 2
C o n ta g io n , G reek view s o f2 1 8 -2 0
C o rp se s, dep riv ed o f b u rial 45—47; futility
o f p u n ish in g 45 n. 47; pollu tin g C h. 2
C o ry b a n tie rites 2 4 5 -7 ; as p urification
288 n. 38
C o s, ru les o f p u rity on 52 f., 393
C o ty to , b a p tism in rites of? 306 n. 125
C re te , an d pu rificatio n 142
Loimos
C ro ssro a d s, p u rificato ry rem ain s sent to
30 n. 65, 229
C ro w n 35 n. 12, 36, 145, 153, 176
C u n n ilin c tu s 99
C u rse s 7,186 n. 234, C h . 6; B ouzygean44,
192, 364; h ered itary 199-206;
h ered itary , tra g e d ia n s’ in te rp re ta ­
tion o f 200 f.; p a re n ta l 196 f.; pow er
o f cu rsin g effectively 192 f.; public
1 9 3 -6 ; spoken by in an im ate objects
198 n. 46
C y b ele, an d p u rification 245 f., 288 n. 38
C y n icism 325—7
C y p re ss 35 n. 10
C y re n e , ca th a rtic law A p p en d ix 2;
relatio n s w ith D elphi 333
406 Indexes
D a d u c h , a n d sexuality 89
Daidala, in P lataea 27 n. 51
D an cin g , as purification 212, 283, 303
D ays, im p u re 102 n. 113, 158 f.
D e a th , a n d p o llution C h. 2; 353; false
re p o rt o f 61; in battle, not polluting
42; o f good m en, not polluting 43
D eb ts to gods 175; hered itary 339, 344,
349 f.
Defixiones 191 n. 2, 198, 251, 269 n. 58
D efloration 75 n. 4, 355
D elos, A th en ian a ttitu d e s to its purity 17,
73, 276—7; no dogs on 357;
purificatio n s o f 33 n. 6, 73, 163, 203,
2 1 8 ,2 7 6 —7, 393; reg u lar purification
o f tem ples on 30
D elp h i, an d C yrene 333; a n d sacred laws
140, 333; a n d the g re a t plague 275;
a n d th e E leu sin ian orgas 161; its
e x p lan atio n s o f disease 250; and
o f p u b lic d isa ste r 2 7 1 -6 , 280;
influences d o ctrin e o f pollution?
138 -4 3
D em e, purified 38
D em e te r, an d p u rity 393; festivals o f
81 —3, 82 n. 33; T hesm ophoros, and
sexual p ro p riety 83 n. 36, 144 f.
D em o n s, an d p o llution 55, 107, 217; in
P y th ag o rean ism 295
D em osthenes, attack ed as a pollution 97
n. 93, 268 f.; on h is own luck 268
n. 54; on ju d ic ia l o a th s 187; on
M e g arian s 166 n. 130; religious
a ttitu d e s in 14 n. 60, 16 n. 73, 128
n. 90, 168 n. 133, 219 n. 64
Speech 22, on sexual pollution 9 4 -7
D esacralizatio n 179 f.
D escen t, ‘p u rity ’ o f 262
D ia g o ra s o f M elos 178n. 192
D ickens C . 18
D iet, G reek 357, 360
D in a rc h u s, on D em osthenes as a
p ollution 268 f.
D io d o ru s, on purification o fD elo s 276;
his source for T h ird S acred W ar 172
n. 165
D io n y siu s o f S yracuse 268
D ionysus, an d eschatology 286 f.; and
O rp h ism 287 n. 29; a n d purification
218, 2 8 6 -9 0 ; D io n y sia c/O rp h ic
ritu a l 3 0 2 -4 ; diverse forms ofhis cult
287; u n m en tio n ab le in funerary
co n tex t 64
D isa ste r, public, ex p lan atio n s o f ‘2 7 1 -8 0
D isease, cau sed by pollution 217 f.;
ch ro n ic 240; contagious 58, 219; in
m ythology 239; not form ally a
po llution 219; purification from
C h . 7; ratio n ale o f purification from
'216-8; religious explanations of
C h . 8;
see also E pilepsy, Im potence,
M a d n ess, Skin-disease
D iseases, a n im al nam es for 248
D o d d s E .R . 2, 9, 110
D o d o n a , responses 141, 250, 279
D ogs 357 f.
D o u g las M . 56, 6 1 ,6 3 , 179
D raco , hom icide law o f 115, 125
D re a m s, ritu a l responses to 219 n. 71;
significance of, determ in ed by
d re a m e r’s sta tu s 266 n. 48; w et 342
D u m o n t L. 63
D ü rk h e im E. 150 f., 225
E a rth q u a k e s 86, 276
E ggs, in p urifications 230; n ot eaten 302,
3 5 8 ,3 6 2
E g g -lay in g an im als 358
Eiresiônê 25
Eisangelia 195
Ekklësia, p urification o f 21
E le u sin ia n M ysteries, an d O rp h ism 282;
d ietetic restrictions before 3 5 8 -6 3 ;
eschatology o f 286; H eracles an d
284 f., 373 f.; initiates dedicate
clo th in g 180; p recin ct purified
30 n. 66; purification as aim of? 2 8 5 1’;
p urificatio n s before 2 8 3 -5 ; see also
A grai, D ad u ch , Hiereus Panagës,
H ie ro p h a n t, H iero p h an tid s,
M ysteries, P riestess o f D em eter and
K ore.
E lia d e M . 11
E m p ed o cles, Katharmoi 208 f., 242, 291,
2 9 9 -3 0 1 , 305
E n circlem en t, in purifications 225 f.
E n o d ia 244
Enthumion, survival o f the concept 253
n. 105
E p h ia ltes, dem o n 248
E p ic , early , purifications in 131 n. 102,
377, 380, 3 8 2 -4
E p ic u re a n s, purified o(F23
E p id a u ru s , inscription over tem ple
3 2 2 -5 ; tem ple record 248; see also
A sclepius
Indexes
E pilep sy , im p u rity used to cu re '234;
pu rifiers of, see H ip p o crates, Morb.
Sacr.
E p im en id es 142, 209 f., 211 η. 23, 259,
276
E rin y es 107, 109 η. 15, 196 η. 34; as
conscience 310; u n real in 4th
cen tu ry 14; see also A eschylus.
H o m er
E u n o sto s 85 n. 44
E u p o lis, Baptai 306 n. 125
E u rip id es, a ttitu d e to p ollution 310 f.;
criticism o f rules o fp u rity in 34,322;
E rinyes in 254, 310; fam ily curses in
199 n. 53; in terest in religious
p h en o m en a 9 1 ,9 3 , 1 6 4 ,2 8 8 -9 0 ;
m y thological innovations in 376,
377, 378, 391
Bacchae, D ionysus in 14; m aenadism in
2 8 8 -9 0
H F , H eracles’ p o llu tio n in 109, 309 f.,
3 1 6 -8
Helen, c h a ra c te r o f T h eo n o e 93 f.
Hippolytus, c h a ra c te r o f H ipp o ly tu s 75,
84 n. 42, 301 ; sexual pollution in 95
n. 8 4 ,3 1 3 f.
A4, m etap h o rical pollu tio n in 111
Ion, c h a ra c te r o f Ion in 91
Medea, d e b a te a b o u t p ollution in 315
Orestes, ascrip tio n o f p ollution in 111,
30 9 -1 1
Supplices, d e b a te ab o u t b u rial in 44
Cretans, p u rity o f in itiate in 289
E ven iu s o f A pollo n ia ‘2 74
E x co m m u n ica tio n , inform al 194
E x cretio n 162, 293
E xecu tio n , fo rb id d en d u rin g festivals 157
E xegetes, A th en ian 112, 131, 141;
A th en ian , th eir ca th a rtic rules 371,
374 n. 29
E xile, as p u rificatio n 114, 118, 386 f.; in
m yth A p p en d ix 7; v oluntary,
because o f p o llu tio n 123
E x p o su re o f child 356
F am ily , in o a th s a n d curses 186;
p u n ish m en t o f 186, 198-206
F a st, E leu sin ian 283; in m ourning
36 n. 16
F av ism 365
F ellatio 99
Festiv als, confined to m en 83—5; to
w om en 81—3; restrictio n s on profane
activities d u rin g 154—8; revolutions
407
d u rin g 159; su rp rise attack s durin g
156
F ig -trees 42 n. 37, 221
F ig u rin e s 347
F in es, p ay ab le to god 180
F ire , a n d sex u ality 77; ca th a rtic 227; new
23, 25, 35
F ish , a b ste n tio n from certain species
3 6 0 -3 ; ‘m an -eatin g ’ 360
F leece o fZ e u s 28 f., 230, 284 n. 18, 285,
350, 373
F o o d , p u rity req u ired to p re p a re 77 f., 80,
99
F o o d s, ab ste n tio n from 52 n. 78, 283,297,
3 5 7 -6 5 ; an cien t ex p lan atio n s lor
a b ste n tio n from 360 f ; ab sten tio n
from in m agic 359 n. 12
F o re ig n ers, co n tam in atio n by 263
‘F o rty d a y s’, an d G reek gynaecology 48,
52
F ra tric id e 137
F r a z e r J .G . 11
F u m ig atio n 215 n. 41, 227
F u n ctio n alism 59
F u n e ra ls, pollu tio n o f C h . 2
G a te s , gods o u tsid e 335
G e n erals, tria ls o f267
G in o u v ès R. 19 n. 4
‘G o d lik e m a n ’ 292
G o d s, accep t h u m b le offerings 323;
c a n n o t su ller p ollution 309; d ebts to
175; forgive 14; p articu larly
co n cern ed w ith p u rity 393; shun
pollu tio n 33, 37, 65; su ller pollution
145
G o flm an E. 318
G o ld , as purifier 228
Gold Leaves 286, 290 f., 2 9 9 -301
G o ld sm ith O . 314
G o rg ia s, Palamedes 127
G osse P. 298 η. 86
G rav e-cu lt, im p u re 38
G roves, sacred 164 ('., 322, 335; penalties
for olfences a g a in st sacred 165 n. 121
G u ilt, an d sh am e 251
h e re d ita ry 199-206; o f states 202 I.;
post-H o m eric? 201
p u rificatio n from 294, 300; sense of
254, 305
H a ir-c u ttin g 293, 295
H a lo a 83, 3 5 8 -6 3
H a rd y T . 314
408 Indexes
H a rv e st, p urity re q u ire d for 78
'H e a le r-se e r’ 209-1 1
H e a rth , p u rity /p u rific a tio n o f 2 ί , 38
η. 2 0 ,5 1 , 77, 293
H e a rth -te m p le 167
Hec a te , a n d dogs 358; exorcism o f 2 2 2 -4 ;
‘food lo r’ 360; im p u rity o f 222, 223
n. 86; m eals o f 30, 224, 307, 347;
sacred fish o f 362 f.
H elike an d B oura, d estru ctio n o f 176,277
H elleb o re 215 f.
H elo ts 261
H em erology 29
H e p h a e stu s, an d purification 389, 393
H e ra , b a th o f 27
H eracles, as ‘p u rifier’ 211; a t
A g rai/E leu sis 284 f., 373 f.; ‘w om an-
h a te r ’, priest o f 87; w om en excluded
from cults o f 84
H e ra c litu s, on purification by blood 371 f.
H erald 188
H e rb a lists 153, 292
H e rm a p h ro d ite s , b u rn t 221 n. 75
H e rm s, m u tilatio n o f 168 -7 0
H e ro -c u lt, im p u re 39, 18Ü
H e ro d o tu s, divine vengeance in 164
n. 117, 168 n . 133; on com m unal
afflictions 2 7 2 -8 ; on disease 242 f.
H ero es, send disease 243 f.
H e ro iz a tio n , as com pensation 320
H esio d , hem erology in 29; on disease 236,
241 ; on ju s t a n d u n ju st city 257 Γ,
2 6 5 f , 2 7 8 -9
Op. 7 2 4 - 5 9 2 9 1 - 4
Hiereus panagës 89 n. 62, 90 n. 68
Hiereiapanagia 90 n. 68
H ie ro p h a n t, a n d sexuality 87, 89
H in d u ism 32, 46 n. 51, 65, 225
‘H ip p o c ra te s’, Morb. Sacr., a ttitu d e of
a u th o r 207 f , 215 f., 233; m ethods of
th e purifiers attack ed in ‘2 07 f., 210,
215 n. 43, 217, 222, 230, 2 3 2 -4 , 244,
292, 359 f , 363 n. 44, 372
H ip p o c ra tic C o rp u s, b a th in g in 215; no
infection in 220; ‘p u rificatio n ’ in
2131'.
H ip p o c ra tic m edicine, a n d tem ple-
m edicine 249; its origins in p o p u lar
m edicine 213; success o f 238 f.
H ittite purifications 22, 231 n. 146
H o m er, disease in 240; divine anger in
241, 273; E rinyes an d curses in 133,
196 f.; hom icide in 130-7;
no pollution in? 9 ,6 6 - 7 0 , 130-43,
176, 189; plague in Iliad 1 176, 209 f.,
217, 266 f., 273,275; purifications in
19 f., 67 f., 114 η. 3 9 ,2 1 0 , 227,305;
religious scruples in 253 f.;
su p p lic a tio n in 181 f.
H o m icid e C h . 4; 322, 327; disasters
cau se d by pollution o f l 2 8 -3 0 ,2 7 3 f ;
in hellenistic period 322 n. 1; in
H o m e r 130 -7 ; in m ythology
A p p en d ix 7; purification from 114,
135, 350, A ppendix 7; rite of
p u rificatio n from A ppendix 6;
ritu a l sta tu s ofjustified hom icide
A p p en d ix 5
H o m o se x u ality 94
‘H o p e s’, good an d b a d 175
H o u se, destro y ed 194; object o f m agical
a tta c k 348
H u m ilia tio n s, p ublic 95 n. 87, 195 f.
H u n te rs, im p u re 298; sexual abstinence
by? 84 n. 42
H u n tin g , purification after 113 n. 37
H y p e rb o lu s 270
I a m b lic h u s, on d ietary rules 361; on
P y th ag o ras an d purification 297
Ig u v iu m , purifications a t 225 n. 98
I m ages, sa cre d 168
Im p ie ty , trials for 189
Im p o te n c e , purification from 208
In c a n ta tio n s 232, 298
In c e st 97 f., 100, 326
In c u b a tio n , b a th in g before 213 n. 31;
d ie ta ry restrictions before 358 f.
In su lt, form s o f 9 7 -1 0 0 , 132, 171,206,
258 f., 2 6 2,268, 360
Isla n d s, sacred 163
Iu lis, fu n era ry law s o f 3 4 -4 1 , 69
J a m e s \V. 57
J u r o r s , im perilled 126-8; th eir oath 187
n. 241
K a lly n te ria 26—8
Kanëphoroi 80
K eos: see Iulis
K in g , a s scap eg o at 259, 265; d ream s o f
266; p u b lic welfare d ep e n d e n t on
265, 274; w ithout blem ish 266
K in g s o f S p a rta , polluted 276 f.; trials o f
267
Indexes 409
K in sm a n , killing o f 122 f., 129, 133, 137, M a n u , L aw s o f 292
351; possibility o f p ro secuting 137 M a rria g e , ritu a l o b lig atio n s atta c h e d to
n. 133 345 f.
K isses, im p u re 99; w ash ed off314 n. 26 M a rria g e -b e d , p o llu tio n o f 95 n. 84
M a s tu rb a tio n 342
M ed ical m aterialism 57
L a m b , in purificatio n s 372 M e g acles 17
L a n d , sacred 1 6 0 -6 M e g a ra , an d A thens: see Orgas
L a u re l 228, 301 M e la m p o d id s 210 f.
L e a th e r 52 n. 78 M e la m p u s 2 0 7 -9 , 212 f., 215, 230, 290
L eg alistic devices, to avoid religious guilt η. 45
133, 154 f., 184, 186 f., 312 M e n stru a l blood, p ro p erties o f 102
Leges sacralae 7, 12 M e n stru a tio n 100—3, 354
L em n o s, fire-festival on 82 M e tic s 261 f.
L en tils 227 M id d le to n a n d Row ley, Changeling 313
L e rn a , lake o f 290 n. 45 M iltia d e s, im piety o f 179
L e u c tra 198, 2 0 2 ,2 7 7 M in d , p u re 323
L eviticu s 61 f. M isc a rria g e 50 n. 67, 346, 3 5 4 -6
L o ch ial bleed in g 55 M o d i J . J . 57
L o c ria n trib u te 202 f. M o n ste rs, b u rn t 221
Loimos, m ean in g o f 257; m ythological M o o n , a n d m e n stru a tio n 102 η. 113
ex p lan atio n s o f 2 7 1 -5 ; ritual M o sch io n , on b u rial 45 n. 47, 48 n. 56,
responses to 275 327 n. 26a
Love, p u rificatio n from 221 M o th e r, im p u rity after b irth 52 n. 74
L uck, co n tag io u s 219; o f lead er 268 M o th e r o f gods 244 f., 288 n. 38
L u stra l w ater, d istrib u ted before sacrifice M o u lin ie r L. 4 n. 13
20; sh a rin g o f2 2 ; sources o f 226;
M o u rn e rs, pu rificatio n o f 36
sto u p s for 19
M o u rn in g , a n d pollu tio n 64, 65 n. 110;
'L ycu rg u s o f S p a rta , fu n erary laws 71
fo rb id d en 43
L y cu rg u s, o ra to r, religious attitu d es in 16
M o u th , p u rity o f9 9
n. 73, 128 n. 90, 183 n. 215
M u d , p u rificatio n by 231; lying in, as
L y d ia n p u rificatio n s 134
u n d erw o rld p u n ish m en t 286
L y san d er, a n d o ath s 187
M u lle t, red 362 f.
L ysias, ‘h a tre d o f peo p le’ in 206; on
M u rd e r: see H o m icide
C in esias ‘2 39
M u sa e u s 242, 304
Speeches 1 a n d 12 128
M u sic, as p u rificatio n 212, 297 f.
Speech 6, religious attitu d e s in 16 n. 73,
M y rrh in e , wife o f H ip p ias, assassinated
179 n. 193
368
Speech 22 262
M y steries, p ro fan atio n o f in 415 168-70,
Speeches 28 a n d 29 267 n. 51
191 ; secrecy o f 177 f
L y sim ach e 88
N a il-c u ttin g 293, 295
M a d n ess, cau sed by m u rd e r 129, 218;
N u e r, m u rd er-p o llu tio n am o n g 120 f.
causes o f 2 4 3 -8 ; cu re of, as a
p u rificatio n 288; purification from
‘2 08, 215 f., 372 ;treatm en ts of246 O a th s 1 8 6 -8 ; in hom icide tria ls 126; of
M ag ic, p o llu tio n in 223 nn. 84 f. sexual p u rity 85
M agical rules, form o f292 O e d ip u s 199, 308, 385 f.; as scapegoat
M ag istracies, p u rity req u ired for 153 259; in H o m er 136, 385
n. 47, 268 O live-w ood 229
M a id en -ch o irs 80 O ly m p ia , w om en excluded from stad iu m
M a im a k te rio n 28 85
M a im o n id es M . 57 O m e n s , responses to 219—221
410 Indexes
O ra c le , consulted on religious ch an g e 161
n. 100; enjoins d e a th o f king 265;
en jo in s sexual ab stin e n ce 86
O re ste s 124, 308; in H o m e r 136;
pu rificatio n o f 139 n. 142, 3 3 6 -8
Orgas, sacred 161, 163 f., 166
O rp h is m , a n d E leusis 282 f.; an d
in h erite d guilt 201 f.; an d ju stic e 305;
a n d killing 143, 306; a n d purification
299—307; a n d P ythagoreanism
290 f.; its d ie ta ry rules 302, 362; its
ritu a l 3 0 2 -4 , 307
O s tra c is m 269 f.
O u tla w ry : see Atimia
P an , a n d m adness 245; co p u latio n in
precin cts o f 76
P a rd o n , rem oves killer’s pollution 108
P a rric id e 124
P a u sa n ia s, regent 107, 183
P a u sa n ia s, periegete, m agic in 275 n. 90
P e isistra tid s 206
P e isistra tu s, an d D elos 73
P elarg ik o n 164
P e n th e u s, as scapegoat 259 n. 18
P ericles, a n d A lcm aeonid p ollution 16,
206; his citizenship law 262 f.
P erju ry 10, 1861'., 199
P h e re tim a 242
P h ilip p id es, on S trato cles 269
P h o cian s, in T h ird S acred W ar 17 2 -5
P h re a tto , co u rt a t 119
P h ry n ic h u s, assassins o f 368
P h y talid s 374 n. 29
Pig, m ystic 283; purification by 30 n. 66,
283 n. 1 1 ,3 7 1 -3
P in d a r, an d pollution 16, 67
P itch 228
P lague, A th e n ia n 218, 275 f.
P lagues, caused by p o llu ted a ir 218; god-
se n t 257; sem i-m agical cures of275;
see also Loimos
P lan ts, purify in g 231
P lato, m en tal an d sp iritu al purification in
281 f., 323
Leges, on in h erite d g u ilt 205; on
offences ag ain st p a re n ts 196 f.; on
pu rificatio n from hom icide 374; on
th e p o llution of killing 107, 108 n. 10,
110—29, 137, 367; right o f burial
den ied in 45 47 n. 52; role o f
D elp h i in 140 f.
Phaedo, on purification 281 f., 324
P ie isto a n a x 2 7 7
P lu ta rc h , d efen d er o f religious traditions
29 f., 57, 324; on d ie ta ry rules 361
n. 28; on ‘tra g ic ’ history 15 n. 71;
‘ta b o o ’ in 330 f.
Timoleon, luck in 268 n. 54
P ly n te ria 26—8
P o llu tio n , a n d d irt 56; a n d disgrace 94,
205, 316 f.; an d divine a n g e r 9—11,
110, 146; a n d law 37 n. 17, 114-25;
a n d m o rality 34, 75, 94, 1 1 1 -4 , 117,
312, 325, 355, 367; a n d o rd er 325—7;
an th ro p o lo g ic al definitions o f 3,
6 1 - 4 ; its consequences social, not
legal 98, 205, 317 f.; conveys m oral
rev u lsio n 1 1 1 ,3 1 2 -4 ; definition of
2 - 1 1 , 96; em otional im plications o f
53; G reek in terp retatio n s o f 44, 55,
107; h ealin g pro p erties o f 233, 373;
h e re d ita ry 185, 2 0 4 -6 , 344; how
diffused 39 f , 49 f., 54, 110, 318,
353—5; how intensely feared 128,
211 ; in hellenistic period 3 2 2 -5 ; i n .
m ag ic 222 f.; in te rm itte n t concern
w ith 16, 315; invoked in curses 191;
n o t m en tio n ed or p resen t 4 2 - 4 , 128,
159 f.; ob ject o f d isp u te 111, C h. 11 ;
o f a b s tra c t values 3, 146; o f m ind
323; possibility of, denied 309;
p ra c tic a l effects o f 53, 205, 318;
‘sleep s’ 17; spoils m arriag e prospects
205, 318; sp read by relationship 40,
318; w ith o u t physical basis 8, 144 f.
P o m e g ra n a te 358, 362 f.
P o m p a ia 28
P o seid o n 85, 244
P ra x ie rg id a i 26
P re g n a n c y 48 f., 3 4 4 -6
P rie st o f H eracles a t T h esp iai 93 n. 76
P rie sts/p riestesse s, age an d m arita l
s ta tu s o f 8 7 -9 4 ; descent o f 97; d iet of
52, 238; exceptional atte n d a n c e at
fu n era ls 43, 53 n. 80; excluded from
h e ro -c u lt 39; inviolable 175; no
c o n ta c t w ith b irth an d d e a th 52 f ;
p u rity o f 175, 205; sexual abstinence
by 8 7 - 9 4
P riestess o f D em eter an d K ore, a t Eleusis
89
Priestesses, A th en ian , m arita l sta tu s of
8 8 -9 0
P ro c la m a tio n , ag ain st killer 125;
E le u sin ian 283
Indexes
P ro clu s 307 n. 130
P ro c rea tio n , im perilled by co n tact w ith
d e a th 53, 70
P rodigies, b u rn t 221
P ro etu s, d a u g h te rs of: see M elam p u s
P rophecy, p u rificatio n before 20
P ro p h etesses, sex u al sta tu s o f 93
P ro stitu te, im p u rity o f in terco u rse w ith
75 n. 4
P ro stitu tes, m ale 94, 95 n. 84
P ry ta n eu m , c o u rt o f 117
Pulvillus, H o ra tiu s 40 η. 29
P u p p y , in p u rificatio n s 30 n. 65, 230
P urges, m edical 2 1 3 -5
P urificatio n , an im al victim s used in
in ed ib le 283 η. 11; a s b ro ad term for
elim in atio n o f evils 211 f.; perform ed
facing east 225; tech n iq u es o f
2 2 4 -3 4 , A p p en d ix 6; see also Arm ies,
B a th , B attle, B ew itchm ent, Boute,
C o ry b a n tie rites, D ancing, D isease,
Ekklësia, Exile, G u ilt, H earth ,
H o m icid e, H u n tin g , Im potence,
L ove, M o u rn ers, M usic, Perjury,
P ro p h ecy , Sacrifice, Sacrilege, Sex,
S h ip y ard s, T em p les
P u rificato ry m aterials, disposal o f 229 f.
P urifiers, sta n d in g o f 2 0 7 -9 , 374
P u rity , o f m in d an d soul 281 f., 323
P y th ag o rean ism , an d O rp h ism 290 f.;
a n d p u rificatio n 2 9 0 -9 ; d ietary rules
in 296 f , 359, 361 f.; guilt in? 291,
298 f.
P y th ag o rean s, pogrom s o f 267 n. 50
P y th ia 93
R a p e 185
R ed sacrificial victim s 334
R elatives, p o llu ted 40, 58, 318
R itu a l om issions, b rin g d isaster 272 f.;
n o t sh a m in g 252
R itu a l rules, kinds o f 176 -8 ; violation o f
1 4 4 -6 , 1 7 6 -8
R ivers, resp ect for 293
R o m an p u rificatio n s/ru les o f pu rity 23
n. 24, 24, 65, 77, 225 n. 98
Roof, sh a rin g o f 122, 336
Row ley: see M id d leto n
S ab aziu s 303
S acred :« « A nim als, G roves, Im ages,
Islan d s, L an d , T rirem es
‘S acred law s’ 176 f.
411
S acred m arriag e 85, 287 f.
S acred W a r, th ird 166, 1 7 2 -5 ; fourth 166
S acred n ess, a n d agos 6; confused w ith
p o llu tio n ? 11, 159, 180, 233, 361;
m ean in g o f 1 5 0 -4 ; o f th e city 153,
193 f.
S acrifice, as p u rificatio n 10, 209 f ;
m u rd e r a t 159; o m itted , causes
d ise ase 252; p e n a l 339—46;
p u rificato ry , in ed ib le 283 n. 11
Sacrifice, h u m an , as pu rificatio n 259; of
k in g ’s d a u g h te r 264 f.; requires
v irg in victim 81, 259
S acrileg e C h. 5; causes p u b lic d isaster
272—4, 276—8; m ob responses to 196;
p u rificatio n after 144—6
Sacrum anniversarium Cereris 82 η. 33, 89
η. 65
S alam is, cleru ch y on 368
S alt 227
S a m o th ra c e , M y steries ο( 284 η. 18, 374
η. 29
S a n c tu a ry : see S u p p licatio n
S cap eg o ats, n o n -ritu al 260—71 ; ritual and
m ythological 2 4 -6 , 2 5 8 -6 0 ;
w h ip p ed 226
S ea, pollu tio n th ro w n into 230;
p u rificatio n in 226
S eers, m ockery o f 15
Self-defilem ent, in m o u rn in g 41, 68
S ep terio n 25 n. 30
S e rv itu d e , for killing 392
Sex, E m p e d o cles’ a ttitu d e to 301;
im p u rity o f C h. 3, 335 f.; O rp h ic
a ttitu d e to 301; pu rificatio n from 74;
P y th ag o rean a ttitu d e to 296;
‘u n n a tu ra l’ form s o f 98
S ex u al ab stin e n ce, before hunting? 84;
before m agic 91 n. 71; before
w arfare? 84; enjoined by oracle 86;
o f ath letes 84; o f h iero p h an t 87;
o f lay m en involved in ritu a l 85 I'.;
o f priests an d priestesses 8 6 -8 ;
provoked by p o rte n t 86
S h ak esp eare, Macbeth 313
S h am e, an d guilt 251
S h ip w reck 9, 17, 129
S h ip y a rd s, p u rificatio n o f 21 n. 14
Silence, o f hom icide 350, 371, 391
S in, a n d disease 236 f.
S ittin g , ritu als o f 285, 3 7 1, 373 f.
S kin-disease, caused by pollution 218;
pu rificatio n from 208, 212 1’
412 Indexes
S kira 82 n. 34
S lau g h ter, im p u re techniques? 52 n. 78
Solon, funerary legislation o f 34; 40; his
religious op tim ism 14 n. 60
S ophocles, on fam ily curses 199 n. 53
A jax, A jax’s sham e in 317; d e b a te on
b u ria l in 44
Antigone, C re o n ’s a ttitu d e to pollution
33, 310, 316; exposure o f corpse in
4 6 - 8 ; p ollution by corpse in 33, 44
OC, d e a th o f O ed ip u s 43 n . 42;
O e d ip u s’ pollution in 137, 318-21
O T , O e d ip u s’ pollution in 316—20;
p lag u e in 130, 1 4 1 ,2 5 7 ,2 7 8
S o p h ro n , invocation of'H ecate in 222 f.
The Women who claim . . . 223 f.
S orcery: see B ew itchm ent
S oul, p u rity o f 281 f., 323
S p a rta , expulsion o f foreigners 263;
in tra m u ra l b u rial in 71
S p a rta n religious attitu d e s 12 f., 43 n. 42,
155 f., 184, 188, 264, 276 f., 279
S p a rta n s an d rap e 277
‘S p ittin g -o u t’ pollution 108, 133 n. 11,
219
S p rin g s, healing 212 f ; pollution throw n
in to 230; rules p ro tectin g p u rity of
293 n. 59; special, used for ritual
p u rp o ses 51, 150, 227 n. 108
S quill 231 r.
S ta tu e s, w ashing an d b a th in g o f 27;
afte r p ollution 27 n. 50, 53
S te in e r F. 235
S tig m a 317 f.
S toicism , critical o f rules o f p u n tv 34,
326 f.
S to n in g 194
S to rm , d u e to pollution 257
S tra to c le s 269
S tu d y , as purification 298
S u b stitu tio n , in sacrifice 372, 373 n. 20
S uicide 42, 52, 198 n. 48; o f su p p lian ts
185
S ulla 40 n. 29
S u lp h u r 57 f., 227 f.
S u n , p o llu tio n o f 293, 310, 3161'.
‘S u p p lia n ts’, C y ren aea n 347—51
S u p p lic a tio n 146, 181 -6 ; a t tom b 152;
rejection of, as pollution 146
S y co p h an ts 263
Symbola, P y th ag o rean 2 9 4 -6
T a b o o 11 ; G reek for A ppendix 1
Indexes 413
T a lth y b iu s , w rath o f 17, 188, 191,264 U rin a tio n 162, 293 W a te r, in pu rificatio n s 226 f., 371; new 35
T a ra n tis m 247 W a te r-c a rry in g , in u n d erw o rld 286
Temeni, leasing o f 160 -3 ; p u rity of 161-3 V a n G en n ep A. 59 W ater-v essel, o u tsid e ho u se o f d e a th 35
T e m p le s, closed on im p u re days 26; V eg etarian ism , im p o rtan c e o f 304 f ; in W easel, in p u rifications? 21 n. 12
defilem ent o f 162; exclusion from , E m pedocles an d O rp h ism 299 f., ‘W ip in g off’ 215 n. 41, 231
a fte r im p u re contacts 37 n. 17, 50 302; in E u rip id es, Cretans 289, 302; W itch craft: see B ew itchm ent
n. 67, 52 n. 7 4 ,6 4 - 6 , 74 n. 4, 102 P y th ag o rean 296, 298, 362 W o m e n , d an g ero u s a n d d eb ilitatin g 84,
n. 112, 352—6 ,3 5 9 n. 12; exclusion V en g ean ce, divine: ag ain st cities 2 7 1 -8 0 ; 101, 261; im pure? 101
from , o f killer 119, 125, 185; an d disease C h . 8; delayed 175; W o o d , M rs. H . 46 n. 48
exclusion from , o f sexual offenders d elayed u n til d escen d an ts 199; W ool 229; sh u n n e d by O rp h ies 302
94; founded in response to plague form s o f 257; in stan tan e o u s 179;post
275; lo an s by 173; m u rd er in 185; no mortem 186; th ro u g h h u m an agency X a n th ip p u s , o straco n a g a in st 270
b irth , d e a th o r copulation in 33, 74; 165,194 X e n o p h o n , his sacred horse 176; on
p u rified 30, 53, 144 f., 3 3 9 -4 6 ; rape V ico G . B. 63 p o llu tio n o f killer 129; purifies arm y
in 185; siting o f 162 V irg in priestesses 90—3 23; religious attitu d e s in 16 n. 73, 168
T e m p le -ro b b e ry 170—5 V irg in s, in ritu a l 7 9 -8 1 ,8 1 n. 28 n. 133
T e o s, p u b lic curses in 193—5
T h a le ta s 209, 212 W arfare, o b stru cted by festivals 154-6; Z eu s A lasto ro s 224 n. 92; H ikesios 181 f.;
T h a rg e lia 25 f. religious ex p lan atio n s for failure in K ath arsio s 139
T h e b e s , sack o f 163, 168, 175,277 2 7 6 -8 Z o ro a stria n ism 3 2 ,4 6 n. 51, 5 7 ,6 5 , 229
T h em isto cles, accursed 270
T h e o p h ra s tu s , On Piety 307
The Superstitious M an 307
T h e rs ite s 130 f., 260
T h e sm o p h o ria 8 1 - 3 , 179; an d
m e n stru a tio n 102 n. 113
T h ir ty T y ra n ts , b u tch er m etics 262;
‘purify city ’ 263
T h o m a s K . 13
T hreshold 350
T hu cy d id es, festival truces in 154-6; on
ev en ts o f 415 168—70; on the plague
2 2 0 ,2 7 1 ; pollution in 1 ,8 , 13, 183 f.,
203; religious ex p lan atio n s o f public
d isa ste r in 276 f.
T im e , purifies 386 f.
T ita n s , crim e o f 299 f..
T ith in g 341; in C yrene 339—44
T o m b s, im p u re 38; re-used 39
T orches 227
'T ragedy, as evidence for religious
a ttitu d e s 13-1 5 , 308
T re a c h e ry 5 n. 18, 45 n. 47, 1 93-6, 206,
270
T rire m e s, sa cre d 153
Γ rito p a te re s 3 3 6 -8
T rollope A. 314
T ro p h o n iu s, purifications in cult of213 n.
31, 358 f.
T ru c e s , o f festivals 155 f.
T y lo r E. B. 55
T y ra n n ic id e s , h o n o u re d '368 f.
T y ranny, legislation ag ain st 366—8