The House of Trimalchio Gilbert Bagnani The American Journal of Philology, Vol. 75, No. 1. (1954), pp. 16-39.

Stable URL: http://links.jstor.org/sici?sici=0002-9475%281954%2975%3A1%3C16%3ATHOT%3E2.0.CO%3B2-M The American Journal of Philology is currently published by The Johns Hopkins University Press.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/jhup.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org Tue Oct 2 10:43:39 2007

T H E H O U S E O F TRIRIALCRIO. Erery archaeologist knows that the recoristruction of any ancient building. even when extensive remains are still preserved, is no easy task. The attempts to recorer the plan and appearance of Pliny's Tillas or Varro's Aviary show only too clearly the insufficiency of eyen the most detailed verbal descriptions. T o t r y to reconstruct a building that never existed escept in the imagination of a novelist vould seem to be a complete vaste of time. Petronius lras rnade no attempt at giving his readers a n exact description of Trimalchio's house, escept as a setting for Trin~alchiohimself, and as an illustration of his character. Xo no\-elist: ancient or modernt has been greatly concerned as t o whether the houses in which his characters live are structurally possible or convenient. Only recently, svith the popularity of the detective novel, does the actual dispositiorl of the house heconle important t o author and reader, but I feel quite certain that no architect 15-ould e7-er pass lnost of the elaborate plans that writers of such stories so kindly provide for their
Cf. TS7arren Hunting Smitll, Srchitectzi~cin Ertglbsh Fietion (Yale Studzes 27, Enqlisli, L Y S S I I I [ S e n Haven, 1'3341 i, 1). 3. The author of this raluable n o r k has unfortunately restricted liiiuself to only " three main types of architectural setting: the purely decorative t r p e , used for ornament and ' local colottr '; tlie structural type performing a service in the narrative: and the emotional type, esplaininp the reactions of tlie characters and a r o ~ ~ s i n g feelings of the readers " (13. 5 ) . He ignores the the type used a s a means of characterisation, and thns omits all mention of such authors a s Trollope and S ~ ~ r t e e sYet . llow admirably does the dec;cription of Ullathorne Coart colnplete the character sketches of Wilfrid Thorne and his sister, or t h a t of Gatllerutn Castle prepare the reader for the Duke of Omnium! Surtees, iil a series of thumbnail sketches, passes i n revielv the whole domestic architecture of the ~ h i r e s ,from IYnodma~lsterne t o I'udclingpotc Bower; the Iiouses; and their interior decoration, are all quite tvpical of tlleir liind, and at the same time typical of their owners. Dumas uses t l ~ esame technique in Le Comte de So?riccr.isto: the descriptions of the Ildtel dc la Rue dza Helder, the petit 7;ersailles rue M P s l a ~ ,a nd the pretentious mansion of the ChatcssCe d'dwtin tell us more about their owners than any elaborate psychological description.

THE' HOUHE OF T R I Y A L C H I O .

17

Yet, though Petronius is certainly not describing any real house, he has a certain type of house i n mind, a type of house that was familiar to his readers, a type of house that they would associate with a person such as Trimalchio. Petronius is an extremely subtle author; as a s a t i r ~ s t ,if he can be called a satirist, he obtains his effects, not by the ranting and exaggeration that become so wearisome in Juvenal, but rather by the delicate juxtaposition of incongruous details. This house of Trimalchio's is a real house, it is the kind of house most people know, and many own, b u t no one, except Trimalchio, would have i t quite that way. It should therefore be recognizable as a characteristic type of house of the time of Xero, and this type of house is not to be sought for in Rome itself. As I have tried to show elsen~here,?Petrollius is not writing of the upper classes of the capital, he is describing "moeurs de province," a provincial life, that of the Campania, with which most of his readers were only too familiar. It is therefore in the houses of Pompeii and Herculaneum that we must seek the model of Trimalchio's house.3 Both these towns were not of any great importance and therefore, since the graeca urbs is almost certainly P ~ t e o l i we , ~ must not expect to find exact parallels, but, making allon~ancesfor this difference in scale, i t is in Pompeii and Herculaneum that we must loolr. And the parallels are even closer than might be expected. Trimalchio has not built his whole house. he has simply enlarged a smaller one (77. 4) : ut scitis, cusuc emt, nunc ten~plzinzest : "you remember. i t used to be a kind of hotdog stand, now it's a temple." At this any owner of one of the great Villas of the
G. Bagnani, " And passing rich . . ." in Studies i n Itonour of Gilbert Korzoood (Toronto, 1952), pp. 218-23. The House of Trimalchio has been briefly studied by P . Harsh i n Memoirs American Academy in Ronle, XI1 (1935), pp. 49-.50 and A. Maiuri, La Cena di Trimalchione (h'aples, 19453, pp. 243-5. 4 N a i u r i , Cena, pp. 5-14 and La Parola del Passato, I11 (1048), pp. 106-8, argues convincingly for Puteoli; BIarmorale, La Questione Petroniana (Bari, 1948), p. 129, tends t o prefer Seapolis. = W . B, Sedg~vick's suggestion, C . R . , XXXIX (1925), p. 118, t h a t cusuc may have some connexion with the Persian Kushk "kiosk," is n o t unlikely: it has been accepted by Marmorale, Cena, p. 174; contra Harsh, op. cit., p. 50. The exact meaning of templum does not concern us here

18

GILBERT BAGNANI.

Campanla must have burst out laughing. Trimalchio's " temple " is not, elen by the standards of Pornpeii and Herculaneum, very large, certainly it is no palace.= It contains four dining-rooms, twenty bedroorns (cubicula), two marble porticos, three libraries; and, upstairs, the two master suites, besides a wing for guests; some fifty rooms in all, 1e.s than in the Villa dei 3 1 i ~ t e r i . ~ 1word mlght be said on the four d~ning-rooms. One of these was probably an open-air one as in so many houses in P ~ m p e i i , ~ where the Casa del Fauno has exactly the same nunlber of dining-rooms. These mould seem, from our point of view, a very large number, but an examination of the houses at Ponlpeii shows that a plurality of triclinia is the rule, not the exception. Even the smallest houses have two triclinza. one large and one small. The reason for this 1s clearly hinted at by lTarro when describing the habits of the ancient Romans: ad focum hieme ac frigoribus ce~tifnbanf,nestiuo tempore in loco propatulo, rure i n chorte, in urbe in fabulino. . . . I n cold weather one dined by the fire. In summer one tricd to find a cool spot. K i t h the dexelopment of the simple Italic house under the influence of EIellenistic comfort, one no longer dined in the atrium-liitchenliving room; special dining-rooms mere introduced. This, however, created its own problems; the " Pompeian " house must have beer1 one of the coldest machines for living in ever d e s i p e d . Winter dining-rooms had to be made as small as possible. so that the gur,sts should. llke cattle in a stable, pro\-ide their own heat. Tn surnnler such a room would hecome intolerable, and a larger t ~ - i c l i ? ~ i ~ r r and. r z . if possible. a n open-air one, mould have to be p r o ~ i d e d . Alnotller reason for the necessity of various triclinia mav be found in the general arrangements of a n c i ~ n t banquets. that remind us of the Jlad Hatter's tea-party. Tlltse formal banquets, in any social class, were verv elaborate and intolerably protracted, at least from our point of ~ i e n *as . they
Cf. JIaiuri, Cenn, p. 244. 9 . 3laiuri. T.a T'zlla dei 3listert (2nd ed , 1947), p. 42, calculates some t\\erit)-t\\o room, on the round floor during t h e Augustan age, while hy 79 he t11ii1l.s it llad about ninety looms in all. On the open-air triclinia a t Pompeli we Pietro Sorano, I t ~ i c l i n i a7l' aperto d t I'ompei in P o m p e z a ~ z a (Xaple., 1950), pp 228-310, listing thirt) -1inie I;nol\n e \am~1le5. T'arro, Dr T z t a Poplrlz X o m a n i , I , apud Konius Narcellus, 83, 21 11; 117, 16 (1~111clsa) ).

T H E H O U B E OF T R Z M A L C H Z O .

19

still are in many parts of the Mediterranean world. Trimalchio's dinner begins at the usual hour after the bath, about two or three o'clock p. m.,1 and continues well into the night, for they hear the cock crow.ll Since there were no plates, all the remnants were thrown on the floor, that after a short time will have looked like a garbage heap. Every norr and then slaves came in and swept up the debris,12 but i t was often better to break off for a short time and resume the feasting in another clean dining-room. The four dining-rooms of Trimalchio are not exceptional or unnecessary. The real peculiarity of the house of Trimalchio, as compared with the houses of Pompeii and Herculaneum, is that the front door did not lead into the atrium, large or small, but into a great portico or peristyle, normally to be found behind the house. Moreover, no mention is made of either an atrium or a tablinum, and consequently most scholars have denied their existence.13 It would appear incredible that Trimalchio who, as an Oriental freedman, is keen to pose as more Roman than the Romans,14 should not have had the most characteristically Roman part of the house, and true enough we find an atriensis. .lfter walking half round the peristyle Encolpius says (29, 9 ) interrogare ego atriensenz coepi, quas i n medio picturas haberent. I n medio can
l oJ. Carcopino, Life in Ancient Rome (ed. Rowell, 1941), p. 264; t h e exact time would depend on the season of t h e year, which is difficult t o establish with cet.tainty. The probabilities would seem t o be either late autumn or early spring,-the question is discussed by Marmorale, Questione, pp. 107-17-but i s of no importance for our present investigation. Sat., 74, 1. Even immediately after the gustatio, Sat., 34, 3. The filth of a n unslvept dining-room, asaroton oecon, was immortalized in a mosaic pavement by the Pergamene a r t i s t Sosus, Pliny, N.H., XXXVI, 184; cf. the commentaries by J e s Blake and Sellers, The Elder Pliny's Chapters on the History of Art (London, 18961, p. 224; S. Ferri, Plinio il Vecchio (Rome, 194B), p. 277; G. E. Rizzo, La Pittura EllenisticoRomana (Milan, 19291, p. 42; Helbig-Bmelung, Fuhrer ( 3 r d ed., 1913), 11, p. 49. l3 Most recently RIaiuri, Cena, pp. 244 and 158. Friedlaender, Cena (2nd ed., 1906), p. 217, rightly sees t h a t t h e atriensis is standing a t t h e entrance t o the atrium, b u t is misled by Biicheler's emendation of the i n precario of 30, 9 into i n atrio. Cf. Maiuri's penetrating remarks, Cena, pp. 187-8, on 53, 13.

20

GILBERT B A B N A N I .

only refer to the inaln reccptlon roonis i n the centre of the house, that is to say the atrium, fnblinunl, alld alac. Encolpius, fillding that the llfe of Trinlalchio had been painted on the walls of the peristyle, rlaturally monders what 011 earth Trimalchio has put in the r~losti ~ n p o r t a n troonis in the house, those i ? rnedio. ~ It follorrs that an alrium and a irrbZiltum eslst, though the guests are not admitted to them. The reason for this 1s not f a r to seek: Trimalchio 1s no fool and is under no illusions as to the kind o coming to the house. I I e is not likely to of people ~ h are leabe 111\ bcst reception roonis open to the ~nspection of any llght-fingprpd prowler. The entrance to the nfriu?n fronl the portico 1s closed by cultains, or by a wooden partition such as the one that l ~ n sbeen found a t Herculaneum, and in front of i t the n t r i e ~ ~ s has i s heen btationed to see to i t t h a t none of the guests " accidentally " stray into the rooms in medio.15 T h a t this 1s the functlon of the ntriensis, and that his orders are not to move fro111 the entrance to the atrium. is proved by the may he suddenly reappears when Encolpins and friends are trying t o wander of? (7.3. 8 ) . And the a t ~ i e ~ i sis i s not alone; ~t is a neat touch that after all the guests are assembled the watch-dog is set to guard the interior. The only rc.al differellee therefore between the Rouse of T r i m a l c h ~ oand the regular Pompeian house is the presence of this great portico b t t v r e n the front door and the entrance to the crtrizl~tl. Tllr other nlarble portlco will be the great p ~ r l s t p l e behincl the houce, n-ith garden. sumnlcr triclinium. and fountains, and so the l i r m g quarter< \ ~ o u l dbe, as the French \ ~ o u l dsap, entre cour e t inrdin. It is this cntranci. portico. n ~ t estate h offices off it I R and dccorated with the allegory of Trimalchio's career,
The sentence t h a t follons the anqner of the atri~lzsis t o Fncolpias30, 1, no?^ Eicebat tnzultactamt consid?i.are-is hopelessly corrupt and after i t there is a short lacuna. All commentators and translators have follox\eil Heinsius iu interpreting t l ~ e general meaning a s " w e had not the time to r ~ a m i n e etc." , I should plefer t o take the ?zo?z lzcebat in i t s ab?olutc' meaning of a definlte prol~ibitlon, and, if n e accept ('arcopino's genial emendatiou multiczanz ( R e v , ittides ancienl~es,1940, pp. 393 f f . ) , t o translate ..\ye \\ere not allowed to cvamine this ltaleidoscope . . . ." There is no reason t o suppose with llarmoralc, Cetia, p. 11, t h a t the atriensls was shon ing thcm round. l o The precarittm of the steward-see below-certainly opened 0x1 t h a t p a r t of the portico along which they had come, for they know where

T H E H O U R E OF T R I M A L C H I O .

21

that he has added to the old smaller house. Since i t does not strike Encolpius as in any way remarkable or curious, we must assume that this was a common way of amplifying and modernizing an old house. JThile there is no similar example i n Herculanenm or Pompeii itself, just outside the latter town the Villa dei 3Iisteri offers exactly the same arrangement of a quadriportico added in front of the atrium, and a closer observation v i l l show that between the house of Trimalchio and this large and once fashionable dwelling there are very close resemblances. Of course this resemblance is with the Tills as i t was during the reign of Augustus, that is to say the period of its greatest prosperity, and not with the Tilla as it v a s a t the time of the eruption, when i t was being turned into a farm house. If we read the description in the Cenn with the original plan of the Tilla dei JIisteri before us we shall, I hope, be able to clarify many details in the Cenn itself.17 Trimalchio l e a ~ e s the Baths in his litter, accompanied by all his suite, running footmen, musicians, deliciae in a bath-chair, etc., and goes along a t a good rate. H i s guests, true " meridionnli," saunter along in a much more leisurely fashion and are f a r behind by the time Trinialchio and his cortkge had reached the house. One of the cursores must ha^-e r u n on allead to warn the porter to unbolt the great door to let the master in. As we see in the great doors at Pompeii,18 this must hare been quite a business, needing se~-era1 men, but the bolts mere at last withdrawn and the great flaps creaked on their bronze hinges. The
i t is. I n most houses, even in the Villa dei Jlisteri, the offices are in the front p a r t of the houqe. I n the great Villa of BosI S Jlaiuri, Villa, pp. 41-2 and Tav. B. coreale, of the same period, the entrance led directly into the quadriportzcus, Barnabei, Ida Vzlla Pompeiana di P. Fannzo ginistore (Rome, 1 9 0 1 ) , Tav. 2 ; P. TV. Lehmann, Ronlalz SOall Paintings from Boscoreale (Cambridge, Jlass., 1 9 5 3 ) , p. 4, though in this case there does not seem t o h a l e been a n atrium uriless the small court marked 15 on the plan is really a n atvium Corinthium. For these coloririades I prefer to use Petronius' own term porticus, without dealing with the vezata quaestio concerning the terms porticus and peristyle on uhich see hlaiuri, " Portico e Peristilio" in L a Parola del Passato, I ( 1 9 4 6 1 , pp. 306-22. Is For the appearance of the main door of the Pompeian house see V. Spinazzola, Pompei alla luce degli Scavi Suovi d i Via dell' Abbondanea (Rome, 1963 j, especially I, pp. 317-34.

22

GILBERT BAGNANI.

moment the master was inside the door i t was at once closed and bolted agam: the main door mas only used by "litter folk," not by riff-raff such as was coming to dinner. The door itself was not directly on the street, but set back from it with a spacious vestibule in front. On either side of this vestibule were masonry benches covered with rugs or mattresses for the porter, his friends, and the hangers-on of Trimalchio. On one side of this vestibule opened the celln perbonn of the porter: the pride with which Trimalchio mentions it ('i?,4) shows that he mas responsible for the arrangements of this part of the house and confirms our conjecture that it mas he who added the portico in front of the atrium. On the other side of the vestibule was the little corridor used by pedestrian visitors, which we find in the Villa dei llisteri and in most of the more sumptuous houses in Pompeii. By the time Encolpius and the gang had arrived all traces of Trimalchio's adljentus had disappeared : the scene was the normal one presented by the exterior of any large house in a provincial town. The great door vas closed and the porter was sitting on his bench shelling peas for his dinner.lg This scene must ha^-e been as common 111 ancient Rome as that of a bawnb eating bread and onions with his cronies on a nznsfnba in modern Cairo. But in quch cases it nas unliliely that the porter was in full 11rerv-prnsinntus, cerasi~lo succinctus cingtclo-and shelling the peas into a silver basin. The talking magpie is probably a touch of the same kind: not long prerious1;v there had been quite a craze for them in Rome, but they nere still a novelty in the pro~lnces.'~ The entrance for pedestrians is, as usual. on the right-hand
I s For the benches along the n a l l s of the ~ e s t i b u l e cf. the photographs in Maiuri, Vzlla, figs. 9 and 10 on pp. 45-6: fig. 10 shows the " pedestrians' entrance." The porter is shelling peas for his olvu supper; he is not, so to speak, helping out the cooks, a s Sage. Satiricon, p. 150, seems t o sup1)ose. The point is t h a t in Trimalchio's house eleri the cf. 37, 8. sla~es dine off s i l ~ plate, e~ 20 Friedlaeniler, Cttta (2nd ed.i, p. 214, is quite right-pace Marmorale, Ccna. p. 8-in citing Pliny, H. S . , 5,78, nuper et adhuc tamen r a r a ab ripennino a d urbem versus cerni coepere picaruna genera, quae longa insiglzcs cauda variae appellantur. The term pica varia is clearly equivalent t o a magpie from the Apennines.

T I I E HOUBE OF T R I M A L C H I O .

23

side of the vestibule, opposite to the porter's lodge. Encolpius, whose curiosity is insatiable, does not turn a t once into it, but explores the vestibule still further. On the wall between the sinistram door of the porter's cella and the main door-ad intmntibus-is the painting of a watch-dog that terrifies him, so much so that he falls back and almost trips on the step. Two such representations of watch-dogs have been found in Pompeii, but in both cases as floor-mosaics: 21 is the painting of his watch-dog on the wall another of Trimalchio's eccentricities? Our friends now pass through the entrance corridor into the portico where the grex cursorurn is being exercised by their trainer. This part of the house is the more public part, where the estate offices and counting rooms are; still, no one but Trimalchio would allow such an activity in this part of the house and, of course, this is the point that Petronius is making. It is not an indication of the extraordinary size of the portico itself; i t would not need to be larger than that of the Villa dei Nisteri. I n the portico Encolpius, who takes a keen interest in painting, a t once starts to examine the decorations which are of a singular kind: a half realistic, half allegorical, account of the life and career of Trimalchio himself. We hare no comparable examples of such decoration in any house of this period that has been preserved, although historical painting was nell established. The subjects of the extant wall-paintings are usually drawn from mythical history or from the great mythological cycles,22 and it may well be that Petronius is suggesting that Tri~nalchio viewed his own life as a kind of epic cycle, a kind of " Trimalchioneia" that sllould be commemorated in the style used to cornmeinorate the lives of the ancient he roe^.'^ At the same time many of the scenes depicted on the walls of the portico can be paralleled by representations on tomb monuments, where allusions to the life of the deceased are fitting and proper, and the point of the joke map lie in this incongruity. How like Trimalchio to decorate the main portico with a typically funerary decoration !
I n the Casa del Poeta Tragico and the Casa di Paquio Proculo: cf. Maiuri, Cene, p. 154. 2 2 Cf. Rizzo, La Pittura, p. 33. 23 I am indebted to A. Maiuri for this suggestion.

24

GILBERT BAGNAIIII

IIom are we to Imagine all this decoration? Are these scenes of the llfe of Trimalchlo painted on a small frleze running round the wall and formlng part of the architectural decoration. such as lve comnlonly find In the so-called second style? 2 4 Or are they " megalographles," occupying the main part of the wall, such as n e also find In the early Emplre ? 2 5 Unfortunately the text containing the description of the paintings (29, 3 - 7 ) , though preser7ed botli In the Tragurensls and the Leidensls, is not free from serious corruptions. I n both the descrlptlon begins wlth the IF-ords e m f alltern re~laliciuirl titulis pictulrz. Burnlailn '6 emended to <cum> tiiulis, conslderlng that the curn might have fallril out of 'tlie text o n l l ~ g t o the attraction of the preceding syllable -ctutri, and tlils emendation has been adopted bv all subsequellt erlltols elcept Ilfalurl." ~ v h ocoilsiders the addltion unnec?ssalj. It n ould qeem to me, h o n e ~ e r , t h a t there is a slight hut important difference in meaning between the t ~ v o readings. I f the malluscrl1)t tradition is correct, the ablatlre tltulis n ould n a t n r a l l ~ define not the I enaliczunl hut pzctum, and tlic meanlllg n o ~ l l dbe " a s l a ~ e ~narlcetwas p a i n t ~ dwith namcs ahoxi3 ~ t . "a practice that is e x t r e i n e l ~comnloil i n the ~vall-pa111 t i n p of this perlod. On the other hand. clr JU f i t ulis ~onld rc~ferto the nares In the slave marliet itqelf, a. though the full phrase Jverc " ceual~cizirrl ct irlancipin culn titulis." 1. e.. " a slaxc-niarlcet and the s l a ~ e s with t l l ~ i r price-tags round thelr " n l ~ i c h 1s tlic n a y all conin~cntators and translators. ~lccl\q S incc. riormull~.t i t u l u s can mean ~ n c l u d l n gSl:lluli. c x p l a ~ n~ t . botli " 1alx.l. tag. n o t ~ c e " and '. lnscr~ptlou. ~lanlc.'"~ both : it 1s tlierefore nececsary to ~ s a l n i l i e nlealilugi noulcl b~ posclhl~ the usage 01 Petronius h~nlsclf. A l l ~ a rfroill t tlii; passage and two otliel c\amples In TCTSF. f itlilt(s 1' used l o d e s c r i b ~a sliort of dedlcatol\- ~il'crlptioll on some kind of label (30. 3 ) . n o t l c ~ s sale (38. 1 0 and 1 6 ) . the descrlptlon on the labels of ~ ~ ~ l i c - j a r s (31. 6 : t h c ~ r ead the tziuli on the l ~ i f t n c i o ) and . the notices on the ccllne 111 the brotliel ('7. 3 ) . It nould seem therefore that
2 4 Case di Loreio Tiburtino and del Criptoportico i11 Ponlpeii and of the Farneqi~iaill Rome. 2 5 Botll i11 the S7111as of the AIysteries and of Boscoreale. 2 6 Burniann, Satyrtcon (Amsterdam, 1743), p. 143. 2 T e ~ l a11. , 03. 2 8 Fo~.cellilli, Lexicon, s. u.

T E E HOUBE OF T R I M A L C H I O .

25

Petronius uses the word to describe any kind of label, especially a commercial one, and thus Burmann's enlendation would be correct, as also the current interpretation " slaves with their labels round their necks." If such is the case, it would be impossible for any painter to paint legible inscriptions on labels round the necks of figures only a f e inches ~ high; the figures must have been at least half life size, and it follows that the walls of the portico are decorated with "megalographies" on the style of the famous ones in the Vllla dei Misteri. The slave market was painted on the entrance side of the portico and occupied all that half-wall from the entrance to the corner: at the corner. marking the change from one wall to the next, was the isolated figure of Trinlalchio " capillatus," with the caduceus in his hand. The next wall was devoted to his life and here we find inore textual difficulties. The Tragurensis has J1inerl;arjzie ducente Romam i~ztrabat, Scaliger's Leidensis has, however, tema vitabnt, with citabat added in the margin, while the Pithoean and Tornaesian editions have tenia intrabat. Most editors have avoided the difficulty by printing Ro~nnmintrnbat, ~ h i c h ,of coursp, gives such excellent sense that it becomes immediately suspect, as an emendation of the scribe of the Tragurensis. The whole passage seems corrupt, and it is indeed strange that if Ronza had been in the archetype, so familiar a word should have become corrupt." Marbach suggests lnoenia intrabat, which seems improbable on palaeographic grounds. Sage prints Romanz tensn intrabat, thus making the best of the various readingq, but hardly convincing, since he does not explain how Xon~amcould possibly have been omitted by Xcaliger's codices. Ilarmorale 30 points out that me have no evidence that Trinlalchio mas ever closely associated with Rome and suggests either Atella on palaeographic grounds, or that the scribes misunderstood the name of the graecn urbs. To this we can observe that if there is little evidence to connect Trimalchio with Rome, there is none at all to connect him xvith Atella. As
2 8 See the excellent discussion by Rlarmorale, Cena, p. 9 ; but since he considers the city to be Neapolis-Questione, p. 133-it is difficult to see how t h a t word could have given rise to the corruptions in the MSS. 01). cit., loc. cit. (n. 2 9 ) : in the whole of the Satiricon the only mention of Atella is in connexion with the mimes, 53, 13 and 68, 5.

26

GILBERT BAGh7ANI.

for the name of the graecu urbs i t is certainly either Puteoli or X e a p ~ l i s ,both ~ ~ names almost as familiar to copyists as that of Rome itself. Moreover one cannot understand why Trimalchio should be entering the grnecu urbs; as Marmorale very rightly points out the venuliciun~a t which Oriental slaves were being sold was probably held in the grueca. urbs itself, one of the great C'anlpanian markets for the trade with the East, and this scene, follo~vingas i t does the scene of the market, must depict the beginning of Trimalchio's life after he has been sold, not before. All scholars who have dealt with the passage seem t o have considered that the corruption lies in the single word Romam, but the evidence of the Leidensis ~vouldindicate that the corruption extends to the follo~vingverb. If we reject the reading Ronzanz intrabat me lllust reject the int,rabat no less than the Romanz. I n this case Sage's emendation may give a clue, and I venture to suggest that the rcading of the archetype was 31inerl;aque rlucente tema vehebatur, which, the final -ur being abbreviated, might give rise to the strange readings of the Leidensis. It is therefore not a scene of arrival, no joke on the adz.entus rlugusti which had not yet become a stereotyped subject i n art, but a scene of departure, the beginning of Trimalchio's new life after his sale. B e leaves naturally lilie a God, i n the carriage used by divinities, under the guidance and protection of Jlinerva. This Goddess is here selected as his patroness, not as representing craftiness or worldly ~visdom,~'but as the patroness of study and learning. Trimalchio is quite sincere in his gratitude to his patron for having given him a good educat i ~ n and , ~ ~the n e s t scene, how he learnt booli-keeping, sho~vs him putting this learning to use with the rcsult that he mas appointed disjlensator. With this we come to the next corner. On the other side of the main entrance and on the other opposite wall of the portico where the cursores are training, other scenes of the life of Trimalchio will have been represented, those t h a t dealt with his ~nanumissionand subsequent career under the patronage of Mercury. All these mill have occupied the surfaces of the wall betveen the doors leading to the offices, which will
31
82

3S

Cf. above.
SO Friedlaender, Sage, hlaiuri, Marmorale, ad loc.
Sat., 39, 4.

T H E HOUBE OF T R I M A L C H I O .

2'1

thus have split them up into tableaux. The wall opposite the entrance wall is the one on which opens the door of the atrium, and the two great paintings of Mercury raising him to the tribunal and of the Fates spinning his golden thread will have been painted on either side of this entrance to the principal reception rooms. Encolpius has thus reached the corner of the portico formed by the right wall-since he came in to the right of the main door, I presume he will have continued walking to the right-and the wall of the atrium and of the apartments i r ~ medio. I n this corner he notes a great wooden cup-board-lararium similar to the one actually found at Her~ulaneum.~"t this period the Zararia are not necessarily in the atrium, but may be found in any part of the house that might be considered suitable. With the silver Lares and the golden pyxis containing-or said to contain-Trimalchio's first beard, is a little marble statue of Venus. This is somewhat surprising; we should expect Mercury, under whose protection Trimalchio has placed himself, and who was generally revered by successful freedmen. And why is it of marble? The Lares are of silver, the pyxis of gold, and we would expect the statuette to be of metal too, perhaps bronze. Some jolie, some incongruity is intended; Friedlaender's explanation, accepted by J l a r m ~ r a l e that , ~ ~ Trimalchio had a special devotion to Venus because his fortune was due in the first place to his being the deliciae of both his master and his mistress (75, 11) is hardly convincing. Trinlalchio never alludes to Venus in any way. I s her presence in the larariwm due to the hope that she may favour Trimalchio's own eroticism, which mas definitely frovned on by Fortunata (74, 8-9)? Venus is involied by Eumolpus in the story of the Perganlene Boy (85, 5 ) , and thus the image of Tenus in the lararium might be an indication of the manners of the master of the household. Encolpius looks across the portico, notices the cursores exercising in the other wing, and he will have seen, but unfortunately not noticed. the large piscina in the centre of the court. into which he and his friend will fall in attempting to escape (72, 7 ) . ISe then moves on and comes to the door of the atrium in front
s4Maiuri, Cena, p. I56 and Tav. IV.
55 Friedlaender, Cena, p. 217, and Slarmorale, Cena, p. 11.

28

GILBERT BAGNASI.

of which stands the atriensis. Encolpius, surprised a t the decoratlon of the portico, enquires what kind of paintings mere to be found in the most elegant part of the house, which he takes for granted he is not likely to see. The butler answers Iliada et Udyssian ac L a ~ u a t i sgladzatoriztn~ nLunzLs. Scenes from the Iliad and the Odyssey are 50 colninon 111 ancient paintlng as to requwe no comnlent. It 1s a different matter with the gladiaioriun~) I Z L L ~ ~ U S SO . sceues of the llfe of the amphitheatrc h a l e been used as decoration In any of the pllrate houses of Pompeii 01 of I:ome a t tlus perlod. The! arcJ, on the other hand, quite contnlon or1 funerary monument.. of magistrates and others who c ?tcunerurn, and 1 5 ho naturally deslre the most hare been ~c2iiot.r spectacular and cxpensire actlon of thelr llres to be commemorated for posteritj.jb Trlmalchio, ineligible as a freedman for an!- of the clvlc ~nag~stracies, has not been and coulcl never be a n edifor nluneris. A t this date to palnt 011 the walls of one's on n houqe scmeb Lrom one's o v a nluuus vould ha\ e been odd and in bad taste. to p a i ~ l t scenes from sonleone else's mas to be re all^ extraordinary. -\s I'ar as I h n o ~ the ~ earliest gladlatorial scpnes used to decorate a prllate house are those on the mosaics of the Yilla a t Zllten, of tllc end of the 1st century of this era." Jlnlurl's tentatlie sugge;tion that this Laenas was connected n-1t11 the C'. Pompeius. the patroll of Trimalchio, nus st be regretfully rejcjcted; 3' t h t ~ e1s 110 rcason to suppose that C. Po~xlpt~i ~.epl.csentq u~ a real person. but l l a l u r i 1s rlght i n I~i>liev~ilg that the Laertatis nlzlrizls Ti-as a real and nlenlorable e ~ e n t . ludi ~\-ouldbe remen~bered not i n e r e l ~by These ce7eb~i~r~irtzi Encolpius artd ('o., but by the readers of the Satiricon, t h a t is to bay l e r o and the court. r~ lrlnlalchio is clearly an arena " f a n "; not only has he such ol~ picturrs in 111; ,rir.izim and tnlrlir~umbut he has a r ~ l ~ sen-ice the fights of Hermeros and Petraites
of silrer cups decorated ~1~1th (5?, 3 ) . falllo~iigladiators ~ ~ inax h o nell h a ~ e been the heroes
of the Lrrfnrrfis rnunlls. S o w tl~,oughthere are no picti!res of
'

3 R Tlie onlv example of paintcd representations of rni?)ze)-a on a tomb by a l e t1lo.e on the tomb of C. T7e.itorius Priicus in Pompeii i l l u q t ~ a t e d G. Spano, 31ernorlr. .lceadernia d ' l t a l i a . Ser. T-11, 101. 111, pp. 2 3 i 315. 3'S huiigemma, I U o s a t c i cli B I I ~ P ? (llome, L 1 9 2 6 ) , p. 278. 3 6 > [ a i ~ ~ r(181111. i, p. 12, n. 1 ; l\l,~lmoraleis inclined t o accept the sugP. L ,11. gestion, C E ~ I (

T H E HOUNE OF T R I M A L C H I O .

29

gladiatorial life a t Pompeii, graffiti and caricatures of gladiators and bestiaries are very comlllon indeed, and from their position on the walls would seem to have been scratched by children and adolescent^.^^ It is therefore possible that a passion for gladlators was considered by the elegant and refined members of Xeronian society as not merely common, but downright childish, and that Trimalchio is represented as being both. The game of pila vhich he n-as engaged in a t the baths n-as certainly considered a childish game.40 Thus the effect of the IZiada et Odyssiar~; et Laenatis gludiutorium munus on a contemporar~ reader would be similar to that produced on us by the information that someone had framed photographs of the Sistine Nadonna and Joe Louis side by side i n his drawing room. And Petronius' must have been almost the last attempt to arrest the popularity of the amphitheatre; by the time of Domitian no one is ashamed to own his passion, and its life invades even the most elegant private houses. BS now they had all come to the entrance to the main triclinium. This vould appear to hare been situated on one side of the apartments in medio, and would have opened also on a small court ou the other side of which mere the kitchens and the servants' quarters.*l From the portico i t was approached through a small ante-chamber in which the procurator n-as receiving his accounts. On either side of the entrance were

fasces cum securibus fixi, quorum unam partem quasi embolum navis aelleum finiebat, irz quo erat scriptum: " C. Pompeio Trimalchioni, seviro dugustali, Cinnamus dispensator." Maiuri 4 2 is probably right in supposing these fasces of bronze, nailed
to a panel forming part of the trim of the door posts. The axes are technically reserred to magistrates with inzperium, but
S8RIaiuri, Pompei ed E,-colano (Padova, 1950), pp. 141-54. It is sincerely t o be hoped t h a t the author of this admirable essay will give us his projected Corpus of Pompeian caricatures. The graffiti of bestiaries on the columns of the Temple of Antoninus and Faustina in Rome (Ch. Huelsen, The Roman Forum [Rome, 2nd ed., 19091, p. 221) would appear t o have been scrawled by men, not children. do Bianca RIaiuri, in Pompeiana (Kaples, 1950), p. 106. The kitchens in all large houses are a s far away a s possible from the dining-rooms. b Z Cena, p. 157.

30

GILBERT BAGWANI.

fa.sces with axes are frequent on the tombstones of minor rnagistrates, though, since these were outslde the pomerizlm of the city, the abuse was less flagrant than in the case of Trlmalchio. The real oddity was the way these two fasces ended in something that looked llke the enzbolum of a shlp and supported the ~nscrlptlon. This is the only example of the word ernbolum In Latin, and is obviously the not uncomrnon Greek word ippohos nhich usuallj means the " b e a k " of a b a t t l e s h ~ p . The ~ ~ use of rost~.uas an ornamental motlf is both ancient and common, and appears frequently 111 the decorative a r t of Pompeii, but its use In this particular instance is by no means clear. If ernbolum is really equlralent to r o s t r u m , it is d~fficultto understand why Encolplus, n h o is not a Greek or an Oriental, would use i t in preference to the extremely common Latin word. JIoreover it is still more difficult to see how the cylindrical fasces could end In a r o s i r ~ t mt h a t 1s a projecting and horizontal object." 4 suggest therefore that Encolpius uses the word to mean a decorative ship's stanchion or mooring ring, such as wele present in the S e m i barges.4a Thlb terminatlon to the fasces is odd enough but structurally possible, and the inscription could be engra7 ed on it, or be attached to the mooriilg rlng. And the natural and obvlous positlon would be on the top of the fasces, which " ended " in the e m b o l u m , so the emelldation i n l a m p a r t e m 46 1s not only unnecessarj but falsifies the sense. I n the following sentrnce, s u b ~ o d e m tztulo el lurprna b i l y c h n i s d e calnera lieladebat, Blumner 47 is undoubtcdly 1-1ght In consldeiing the lamp as also a gift of Cinnamus, bearing the same inscription; ~t mould ha~e been inscribed on a plaque inserted above the lamp itself on the cham from n hich i t hung from the ceiling.48 The inscription would h a w been arranged as folloas :
41 Cf. 12itldell and Scott's I,exicon, s. v.; tlie word can be either masculine or neuter, in the qense of rostra the masculine oi 2p. seems preferred. 4 4 Tlii? diffic~~lty had occurred to Sage, Satzrzcorz, p. 131. 4 6 C. l t o l e t t i in G. Ccelli, Ide A avi dz Semz (Rome, 1940), pp. 199 ff. *@By Lipsius, ~ x h o has been follo\\ed by practically all editors except Marmorale and Terzaghi. 47 Phtlolog~ts, J X X T I (19203, p. 333 ; accepted by Ernout and Xaiuri, rejected by Xlarmorale. It m u i t h a l e been ~ e r y like tile htlycl~litsfrom Stabiae reploduced i n Blkmner, Dze Romi.sclte~zP r i r a t a l t e r t ~ i m c r (Xunich, 1911), p. 137.

THE HOU8E OF TRZYALCHZO.

C .POMPEIO TRIMAI~CHIOKI VIVIRO .AVGVSTALI
CINNAMVS
DISPENSATOR

Before entering the friclinium proper our friends are asked to intercede for one of the slaves with the dispensator whom they find (30, 9 ) i n precario azlreos numerantem. The word precurium found in all the codices has been lengthily discussed. As a term for a room or a part of the house i t is found nowhere else in Latin. Biicheler emended the text to i n atrio, and has been followed by Ernout and Terzaghi. To this conjecture there are several objections: it is improbable palaeographically, there is no reason why the dispensator should be counting his money i n the atrium and many why he should not, Encolpius and Co. never go into the atrium or the central part of the house. F a r better is JIarbach's 4 9 suggestion precario = prooecario, possible both palaeographically and etymologically, and the meaning would be the room or vestibule i n front of the oecus, oecarium, and equi~walent,as hlarbach himself says, to pars prima triclinii. J1aiuri,jo however, points out that, as f a r as the terminology of the house is concerned, the language of i t all is strictly Latin, with no Greek neologisms. B u t the real and decisive objection to this conjecture is that this prooecarium would be the very room the guests are i n a t this moment, and the dispensator aureos nulnerans would be the same person previously described as the procurator rationes accipiens, which is absurd. The account given would seem to imply that they go a short way to find the dispensator, who, as the most important official of the household, has his own office. The alliteration in precal-io . . . deprecati sumus seems deliberate, and Sage, Maiuri, and Marmorale have rightly argued that precarium is the name of the apartment of the dispe~zsator,a n " oratory "-so Sage-where he receives petitions, a kind of " Court of Pleas." Since Encolpius knows where i t is, i t m u 4 open on the portico, on the side t h a t he has already visited, and its door would be most suitably situated between the scene of Trimalchio's book-keeping lessons and that of his appointment as dispensator.
A. Marbach, Wortbildung (Giessen, 1931 ) , p. 127.
Cena, p. 159.

50

32

G I L B E R T BAGA'AIVI.

Tlle triclir~iunz itself is not described in any way, but the description of the banquet allows us to infer something about it. I t was obviously a large though not necessarily a vast room and the couches xvere not against the walls since the s l a ~ e s seem to be able to walk behind them." It mas poscibly oecus Corynthius, such as ~ v e find in the Casa del Neleagro or in the Casa del Labirinto, with columns round three sides.52 I n the centre of the ceilings of some of the larger rooms of Pompeian houses there are rectangular imitation s1;glights; traces of one of these has been found for example in the Casa del 3 I e n a n d r 0 . ~ It ~ is possible that these are a h u r v i ~ a lof real skylights of wood. If such is the case, come could haye easily been mounted on wooden rails and been .lid open to a l l o ~ ~ and food to be let down on gifts the guests, a tricli that mas practised by Trimale1110 (60,1-4).54 Tlle only other part of the house t h a t the guests visit is the bath, ~ ~ h i c \vill h have opened on to the portico. It had been a bakery and thus the ovens could be used as the furnaces; " this is a f u r t h ~ r proof that this is the part of the house that Trimalchio had added to the o r l g ~ n a l one, the rlucleus of ~ ~ h i mar: c h tile apartment itz medio. It was dark and small, just like the one in the Tills dei I\Iisteri. The other rooms which Trinlalehio describes i n 7 7 , 4 were either on the upper floor or i n the back part of the house. The only point about mhicli there might be some doubt is the mention of the two marble porticos: is one of these the entrance portico in front of the afrit~rrz or a r t there
& l I n 31, 3 the mashing of the hands and the pedicure seem t o take place a t the same time: it is difficult t o see ha\\- the latter o l ~ r r a t i o n could be performed unless the slaves could go behind the conches. " Saglio in D. S., V, 1, p. 152; hlau-Kelsey, Ponlpcii (h'ew Tork, 1899 1, p. 259; 9. JIaiuri? " 1-isiolzi Italiche '' Pompei (Novara, 1928), pp, 59-60. The largest triclinium in Pompeii is the one in the Villa a t Porta JIarina, 6.00 hy 8.80 meters, wlrich also has a vestibule: Nainri, Pompei " Itinel-ari " ( 5 t h ed., 1049), p. 106. (Rome, 19331, p. 171. 63 31aiuri: L a Casa dcl Uc?ta~zdro 64 I t \\as a trick a s old as the time of Jietcllus Pius (Val. XIasimus, IX, 1, .5) b u t i t is referred to by Suetonius, Sera: 31, 2 and Seneca, Epist.,90, 15. It is probable t h a t only by the time of S e r o were the technical dificulties ~uccessfullysurmounted. \\'as this perhaps one of the elegantiae of Petronius himself? 6 6 The calida~iunz was built directly orer the ovens a s in t h e Casa del Criptoportico ( I l a i n r i , 3. S., 1033, p. 270) and the Casa del l l l e ~ ~ a n d r o p. 2 2 0 ) . (Maiuri, Casa del ~Wenandro,

T H E HOCNE OF TRIYALCHIO.

33

two perisiyles behind the central block, as i n the Casa d ~ Fauno? l There is no example of porticos or colonnades i n marble in Pompeii, and this fact, together with the pride with which Trimalchio refers to them, wo~lldlead one to suppose that, had the entrance portico been of marble, Encolpius would have noticed it. On the other hand Maiuri 5"iglltly observes that the use of Oriental marbles must have been much commoner in Puteoli, the centre of the import trade from the East. than i n small rural communities such as Herculaneum and Pompeii. It is therefore possible that a marble portico did not strike Encolpius as anything really out of the ordinary, compared to the oddity of the decoration. From this examination of the House of Trimalchio one curious fact is at once apparent. For anyone living during the quinquennium 2L'eroni.s this typical house, so typical that Petronius assumes that his readers will immediately recognize it, is a n extremely old-fashioned one." JTe have seen its close resemblance to the lTilla dei Ilisteri, to the Villa dei 3listeri such as i t was i n the Augustan age. the time of its greatest prosperity. The paintings and the decorative systems are those of the second style, with bands of n a r r a t i ~ epaintings with inscriptions. and great " megalographies." This s t ~ l e had certainly ceased being fashionable in Rome by the time of C a l i g ~ l a .TTe ~ ~ can see an example of decoration in the height of contemporary fashion in the rooms belonging to Xero's D O M U S T rnnsitoria on the Palatine under the Flavian Palace. which are certainly earlier than the fire of 6.2 -1.D. and probably reflect the personal taste of I'etronius." The figures are small and the scenes, thougll drawn from the epic cycle, hare no inscriptions. The fourth style is launched by Sero's Golden House and Naiuri has shown that the most perfect example of this style, the Rouse of the Tettii, was executed after the earthquake of 63 A. D.'O Nothing could
Cena, p. 245. F. Marx in Xeue Jahrb., XXIII (1009), p. 552 had pointed o u t t h a t Petronius was describing the style of house t h a t mas fashionable in the second half of the 1 s t cent. B. C., but did n o t realize the importance of this fact. 6 8 C f . G. E. Rizzo, N o n u m e n t i della Pittura A n t i c e L e Pitture dell'A4ula Isiaca d i Caligola (Rome, 1936), especially pp. 32 and 38. '* Cf. Rizzo, La Pittura Ellenistico-Romana, p. 19 and P1. 32. A. hIaiuri, L'ultima fase edilizia di Pompei (1942), p. 112.
6e

67

34

GILBERT BABNANZ.

be more unlike the description of Trimalchio's wall decoration than either the third or the fourth styles. Though Encolpius does not refer to or cominent on this aspect of the decoration, he does comment unfarourahly on the smallness and darkness of the bath. Both Jlaiuri and Jlarmorale 61 interpret this as an indication of Trimalchio's plebeian vulgarity, the latter going as far as to say that a typical nouveau riche such as Trimalchio would spend vast sums on his reception rooms but neglect those that were for personal use. But in the Roman world the bath, as Trimalchlo hiinself proves, is actually part of the reception rooms; i t is not intended for the exclusively personal use of the owner. It is not that Trimalchio has tried to economize on his baths, he has sinlply built an old-fashioned one. Seneca in a TI-ell-known and practically contemporary letter to Lucilius (83,4-7) describes his visit to the villa of Scipio Africanus a t Liternum and his own surprise at finding a balneolum angustunz, tenebricosum-practically the same words as Encolpius'. He goes on to cay that such x a s the ancient fashion, ex consuetudine ccr~tiqun, ?zon videbntur inaioribus nostris cabdurn nisi obscurum, and contrasts this n i t h the luxury of modern private baths, especially those of freedmen. This fashion of wanting to "bathe and sun bathe " at the same time was certainly recent in Seneca's day, for the bath in the Villa dei Xisteri must not have been unllke Scipio's, angustum et tenebricosum, and was certainly by later standards far below the comfort, elegance and luxury of the r e 4 of the house. H o v are n e to elplain this curious fact? The explanation that would seen1 obvious is that Petronius has set the time of the novel in an earlier age, probably in that of ,lugustus. That Petronlus may have specified in some lost part of the work the name of the Emperor ill xvhose reign he supposes the action of the novel to take place is certainly possible. but there is no e\ideuce to suppolt such a ~ i e w and i t seems to ine improbable in the extreme. l f we place the action in any time earlier than K ~ r o ' sT T e find nulnerouq and glaring anachronisms and inconsisteucleq, and in any case i t 1s perfectly clear that the manners and custonlq are those of Petronlus' onn age. Since Be is not intending to satirize the Court, but nlerely to amuse it. no
alaiuri, C e x a ,
13.

214; Jlarmorale, Cena, p. 169.

T H E H O U S E OF TRZMALCHZO.

35

reasons of prudence, such as later influenced Juvenal, would suggest the need of placing a novel on contemporary provincial manners in an earlier reign. And who, a t Nero's court, would care to read about how the lower classes behaved in the reign of Augustus, or appreciate the research which the author had done to be so accurate as to certain particulars, but which had not caused him to avoid certain other outrageous blunders? Interest in historical novels, based on more or less accurate research, is nowhere evident earlier than the Romantic Movement, than Scott, Thackeray, Dumas, Victor Hugo. And whatever Petronius and his readers may have been, they were certainly not Romantics. The Satiricon reflects, accurately and consistently, the manners of the time of Nero. The character of Trimalchio's House is therefore an oddity rather than an anachronism: we know from Pompeii that most of the houses there still preserved their first and second style paintings at the time of the earthquake of 63 A. D. Since the whole purpose of the minute description of the house and its decoration is to define and illuminate the character of the owner, this oddity must, in the intention of Petronius, contribute some important detail to our understanding of Trimalchio himself. It certainly excludes the idea that in Trimalchio Petronius is intending to satirize the enormously wealthy Imperial freedmen, the ostentation and luxury of whose bathing establishments were censured in Seneca's letter. Indeed the purpose of Petronius may be that of drawing a neat distinction between town mice and country mice, between the behaviour and manners of the wealthy freedmen of Rome and of the wealthy freedmen of a provincial city. Personally, however, I prefer another explanation. Trimalchio is described (27, 1) as being a senex calvus. If Petronius is using the word in its strict meaning, he would be over sixty years old and the hypothesis that the soothsayer Serapa had In prophesied that he would reach a hundred is a t t r a c t i ~ e . ~ ~ this case Trimalchio would be in his seventieth year. On the other hand Petrorlius is more probably using it in its more general meaning, but even so it is difficult to imagine that Trimalchio is less than fifty-five at the youngest. Since the date
e2

Cf. Sat., 77, 2.

36

GILBERT BAGNANI.

of the Celza is probably 60 il. l ) . or a few years earlier and Trlmalchlo came to Italy as a puel- cupillatus, he must have arriied somewhere round the turn of the century, and in any case 111 the relgli of Liugustus. H e was bought a t once by C. Pompelus, a prominent and ~vealthy person of that time, though there is no reason to suppose that any real person is meant, or elen that the Pcctlimonlum latzcla~zumnhich he left to Trimalchio ( 7 6 . 2 ) means that he was necessarily a senator. H e will certalnl~h a ~ e l l ~ e di n a large and elegant house decorated in the f a s h ~ o nof the time, that is to say in a house \er> similar t o the T7llla d(ll Rl~iteri. llnd tlni is the o i l l ~large and elegant hai house T r ~ n l a l c l l ~ o eTer Lnorvll. 111 the prolinces the dirislon ha betwoen tlie diffrre~ltstrata of Society ~ ~ 1 1 1\ e heen preserved muell more strlctl! than 111 T'rome itself. -1s the delicicte of his master he n 111 scarcely ha\ e been taken on I isits to othrr T lllas ; ~ v h e nhe begail to interest his mistress as well. he n as exlled to a dlstarlt country estate, and finally he n a s entirely taken u p by hls buslliess as disp(>nsnfor. AS a freedman he n a s excluded froni thc ioclety of the tonn. and he was hy no means rlch ! into ~ t 111s . clrcle n as entirely conlposed enough to b u ~ 111s n a of hls ow11 p:lrasltes and hangers-on. other freedmen in similar circun~stanceq. terr11)le ~ulparia11ssuch as H a h i m a s and the other s ~ r i r q .I17hat could he Iinow of rlcpance and the changes of fashion? IThen he 11~-made 111s plle he nlshes to reproduce the only p~eat house he ha, ei er bno-ii n , the house 111 TI-hichhe serred ~vherlhe came florrl Asla. that hnd struck his c11ildlsl-r imaginatlon as a mlraclp of comfort and elegancr. H i s house. oldfashioned, i n c o n ~ r n i e n t .~ i ~ t l e g a n t and . \ \ l t h a leallj torrlble ))nth I S simply the "~r-~sh-fulfillmeut " of the puPr crrprlltrtuc of fifty !ears before, and a clear p ~ o o fof the nice psychological 1ns1gl1t of o11t of the greatest and n-rost subtle of literary a r t ~ s t s .

E ~ c u ~ s O\us

TIlE

ENTRASCE TO

THE

HOUSE.

The entrance to the House of Trimalchio raises some perplexing problems ~vhicll should he discussed in detail. It is certainly ~ e r y curious that Encolpius does not describe his actual entry into the House itself. the actual crossing of the threshold: he from the ~ ~ p s t l h u l e wems to pass by a kind of osmosis into the entrance portico. This is strange in view of the elaborate

T H E HOUBE OF T R I M A L C H I O .

37

description of the entrance into the triclinium. If Trimalchio was so fussy as to the a a y guests entered his dining-room, he would surely have inlposed a n even more elaborate ritual for the entrance into the llouse itself. Our text of the Cena is notoriously unsatisfactory: many short lacunae are admitted by all editors, many others have been suggested, and i t is probable that others exist unsuspected. Encolpius, after his fright a t the painted watch-dog, continues et collegae mei quidem riserunt.

ego autenz collect0 spiritu non desliti t o t u ~ n parientem persequi. It is certainly not beyond the hounds of possibility t h a t a
sentence describing the actual entrance of Encolpius has fallen out between spiritu a n d non destiti. I have assunled that Petronius is not thinliing of any particular house, but merely of a certain type of house. The great majority of I ? o n ~ a n houses are laid out on a roughly symmetrical plan on either side of a main axis that runs through the entrance, the centre of the atriunz, a n d the centre of the tablinum. There are, of course, ~lunlerous exceptions due to the personal predilections of the owner or the nature of the ground, such as the House of Livia on the Palatine or the Tillas of Diomedes or of Boscoreale. But since the typical Roman house has such an axis, and since the phrase in nzedio would seem to indlcate that the principal reception roonls are on this axis, I have assumed that the main entrance to the House of Trimalchio is also on this axis, and therefore in the centre of orle of the four walls of the quadriporticus. This would seem to be confirmed by the account of the attempted escape ( 7 2 , 7 ) : when they reach the front door by ~r-hich they had entered the ~vatch-dogset u p such a howling was certainly t h a t Llscyltos stepped back into the piscina, ~r-hich in the centre of the portico and therefore ~r-ouldhe directly behind our friends. The word porticus is used in Latin and by Petronius to describe not merely a single colonnade, a stoa, hut also the whole qui~driporticusor peristyle. I n such a case it may be said to "begin " on either side of the entrance and t o " end," deficere, a t the point exactly opposite, where the two sections of the quadriporticus reunite. This I believe to be the meaning of in, deficiente porticu and I have therefore placed the tribunal scene on one side of the entrance to the atrium ; on the other

38

GILBERT BAGNANI.

side, pruesto, mere Fortune and the Three Fates. Haring thus reached the " end" of the portico, he 100lis back and remerllbers that he should mention the grex cursorum on the other side and the curious armarium in one of the corners he has passed. The watch-dog was certainly painted on the left wall of the vestibule and, given the constant use of litotes in vulgar speech (cf. Hofmann, limgang~sprache,~ pp. 147ff.), non longe ab ostiarii cella must mean "right next to the porter's lodge." I have therefore placed the fresco on the wall between the porter's door and the main ianua. I have also taken the totum parietem of the sentence 11011 destiti t. p. persequi as equivalent to totum porticus parietenz "the wall-surface of the whole portico," not merely of a single wing, a usage that is certainly not impossible in Latin (cf. Forcellini Facciolati, s. v. paries) and very common in Italian, " esarn~naitutto il muro del chiostro." Of course the interpretation of paries as a single wing of the portico ib also possible, and in that case, assuming there is no lacuna in the text, i t would appear to be the continuation of the wall of the vestibule on which the watch-dog was painted. All the frescoes mentioned would be on this wall and the entrance to the triclinium immediately after the nngulus at the end of this single wing of the portico. Encolpius mentions only one corner, but that may be only an afterthought, because it contained the nrnzariunz ; the others were not mentioned because they were not memorable. I n this case the main entrance would be at the extreme left corner of the quadriporticus. I n the 1-illa di Diomede on the Via dei Sepolcri a t Pornpeii the entrance is in the right-hand corner (plans of the Villa di Diornede in all handbooks and guides to Pompeii, including Naiuri's Itinerario, and, more accurately in A. Maiuri and R. Pane, La Casa di Loreio Tiburtino et la T'illn di Dionzede, ,110wumenti Italiarzi, 11, I [Rome, 19471, pls. IX-XITI, with sections and reconstructions), and would therefore seem to justify the possibility of such an arrangement. Rut the plan of this Villa is entirely abnormal, because of the nature of the ground and its position with respect to the road, and on these grounds I have come to the conclusion that it cannot be regarded as a typical plan. The position of the entrance in this Villa is due to the desire of the owner to have as direct an access as possible from

T H E HOUNE OF T R I M A L C H I O .

39

the front door to the great garden peristyle at a lower level. The presence of a larger summer triclinium in this garden would seem to confirm my suspicion that in a large house one might not want one's dinner guests to wander all over it before coming to the triclinium. If, notwithstanding these considerations and the difficulty of explaining how Ascyltos falls into the p k i n u , this arrangement be preferred, Petronius then must have some particular and abnormal house in mind, perhaps the house of Calvisius Sabinus. This seems to be improbable; he could hardly expect his readers to be familiar with the particular house of a particular nouveau riche. And he even more certainly could not expect that, some twenty centuries later, someone would be foolish enough to conduct this kind of investigation !