1a|||aaet)p

P
MONHQC:
PLP1Lh
MAS:AH310DV01
Oxford Introductions to Language Study
Seres Editor H.G.Widdowson
Pragma
l
ICS
Geor
g
e Yule
OXFORD UNIVERSITY PRESS
OXOR
MNIV8SIT E8SS
Grat Cndon Steet, Oxor OX2bDI
Oxor Univerit Press ua deparent of the Unerit of Oxor.
It fer te Univerits objecve of excellence i rsea. scolhp.
and educton by publishing wordwde i
Oxord Ne York
Aucd Cpe Tow Dar es Salaam Hong Kong C
Kuala Lumpu Madrd Melbue Mexco Ct Nairbi
New Delhi Shangai Tpei Tornto
Wt ofces in
Arentna Au Br Ce Cech Republc Fnce Grece
Guatemala Hung Itly Japan Poland Porgal Singapor
Sout Kora Swterlad 1d Trke Ue Veta
OXFO8D ad OXFO8DENGLISHMrgsterd tde Kof
Oxord Unverit Press in te ad in cerin oter countes
GOxord Unerit Pss 1b
Te mor rgt of te autor hae been assered
Database rght Oxord Unverit Pss (maker)
Fmpublsbed 1b
2D11 2D10 2O 2DD6
20 19 T6 1

Tb 15 14 T
Nomumomphotmg
rgt rseed. No pa of Upublcton may Dreproduced.
stord ma rtevl s or tted. ay for or by ay mea,
wthout te pror psion i wtng of Oxord Unverit Prss.
or as exrssl perted b ,or under Uaged wth the apprprate
reprgpbc rgt orton. Enques conce rprducon
outide te scope of te 8Wsbould be sent to te ELT Rgt Deparent.
Oxor Univeit Prss, at te addss abve
You must not cate Ub k in any oter bindng or cver
and you must impose Use conditon on ay acquirer
Ay websites rferd to in Upublicton Min te public doma ad
tei addresses Mprvde by Oxr Univerit Pss for inoraton only.
Oxord Univerit Pss dany rsponsibilt for te content
5BN.601@2D
leset b We 21lmited. Brtol
Prted in L
-'
00I00IS
Preface XI
SECTI ON 1:
Surey 1
1 Defnitions and backround
Syntax, scmantics, and pragmatics q
Rcgularity q
1hc pragmatics wastcbaskct 6
Z Deixis and distance
Pcrson dcixis to
Spatial dcixis 1z
1cmporal dcixis 1q
Dcixis and grammar 1j
Reference and inference
Rctcrcntial and attributivc uscs t8
amcs and rctcrcnts 1y
1hc rolc ot co-tcxt zt
Anaphoric rctcrcncc zz
4 Presupposition and entailment
Prcsupposition z6
1ypcs ot prcsupposition z;
1hc pro|cction problcm j o
Ordcrcd cntailmcnts j j
Cooperation and implicature
1hc coopcrativc principlc j 6
Hcdgcs j8
Conversational implicature
qo
Generalized conversational implicatures qo
Scalar implicatures
qt
Particularized conversational implicatures qz
Properties of conversational implicatures qq
Conventional implicatures qj
6 Speech acs and events
Speech acts q8
IFIDs q

Felicity conditions jo
The performative hypothesis j t
Speech act classifcation j
Direct and indirect speech acts jq
Speech events j6
7 Politeness and intercion
Politeness 6o
Face wants 6t
Negative and positive face 6t
Self and other: say nothing 6z
Say something: off and on record 6
Positive and negative politeness 6q
Strategies 6j
Pre-sequences 6y
8 Conversation and preference strcture
Conversation analysis yt
Pauses, overlaps, and backchannels yz
Conversational style y6
Adj acency pairs ;6
Preference structure y8
8 Discoure and culture
Discourse analysis 8
Coherence 8q
Background knowledge 8j
Cultural schemata 8;
Cross-cultural pragmatics 8;
SCTION

Z
Readings t
SCT1O N
References t t;
SCTION q
Glossary tz;
f0BC0
Purpose
What | usthcaton mght thcrc bc tor a scrics ot introductons to
|anguagc study! Attcr al|, |ngustcs s alrcady wcll scrvcd wth
ntroductory tcxts. cxpostons and cxplanatons whch arc com-
prchcnsivc and authoritatvc and cxccl|cnt in thcir way. Gcncrally
spcakng, howcvcr, thcr way s thc csscntally acadcmic onc ot
provdng a dctalcd ntaton nto thc dscplnc ot lingustcs,
and thcy tcnd to bc lcngthy and tcchnical. appropriatcly so, gvcn
thcr purposc. ut thcy can bc qutc daunting to thc novicc. 1hcrc
s also a nccd tor a morc gcncral and gradual ntroducton to
languagc. transtona| tcxts whch wll casc pcoplc nto an undcr-

standing ot complcx idcas. 1hs scrcs ot ntroductons s dcsgncd
to scrvc ths nccd.
1hcir purposc, thcrctorc, s not to supplant but to support thc
morc acadcmcally

orcntcd ntroductons to lngustcs. to
prcparc thc conccptual ground. 1hcy arc bascd on thc bclict that
it is an advantagc to havc a broad map ot thc tcrran skctchcd out
bctorc onc considcrs ts morc spcchc tcaturcs on a smallcr scalc, a
gcncral contcxt n rctcrcncc to whch thc dcta| makcs scnsc. lt s
somctmcs thc casc that studcnts arc ntroduccd to dcta| wthout
t bcng madc clcar what t s a dctal of. Clcarly, a gcncral undcr-
standing ot dcas is not suthccnt. thcrc nccds to bc closcr scrutny.
ut cqually, closc scrutny can bc myopc and mcaninglcss unlcss
t s rclatcd to thc |argcr vcw. Indccd, t can bc sad that thc
prccondtion ot morc partcular cnqury is an awarcncss ot what,
n gcncral, thc partculars arc about. 1his scrcs s dcsgncd to
providc this largc-scalc vicw ot dittcrcnt arcas ot languagc study.
As such t can scrvc as a prc|mnary to and prccondton torj thc
PREFACE XI
morc spccihc and spccializcd cnquiry which studcnts ot linguist-
ics arc rcquircd to undcrtakc.
ut thc scrics is not only intcndcd to bc hclptul to such stu-
dcnts. 1hcrc arc many pcoplc who takc an intcrcst in languagc
without bcing acadcmically cngagcd in linguistics per see Such
pcoplc may rccognizc thc importancc ot undcrstanding languagc
tor thcir own lincs ot cnquiry, or tor thcir own practical purposcs,
or quitc simply tor making thcm awarc ot somcthing which
hgurcs so ccntrally in thcir cvcryday livcs. !t linguistics has rcvcal-
ing and rclcvant things to say about languagc, thcn this should
prcsumably not bc a privilcgcd rcvclation, but onc acccssiblc to
pcoplc othcr than linguists. 1hcsc books havc bccn so dcsigncd as
to accommodatc thcsc broadcr intcrcsts too. thcy arc mcant to bc
introductions to languagc morc gcncrally as wcll as to linguistics
as a disciplinc.
Design
1hc books in thc scrics arc al| cut to thc samc basic pattcrn. 1hcrc
arc tour parts. Survcy, Rcadings, Rctcrcnccs, and Glossary.
Suey
1his is a summary ovcrvicw ot thc main tcaturcs ot thc arca ot
languagc study conccrncd. its scopc and principlcs ot cnquiry, its
basic conccrns and kcy conccpts. 1hcsc arc cxprcsscd and
cxplaincd i n ways which arc intcndcd to makc thcm as acccssiblc
as possiblc to pcoplc who havc no prir knowlcdgc or cxpcrtisc in
thc sub|cct. 1hc Survcy is writtcn to bc rcadablc and is uncluttcrcd
by thc customary scholarly rctcrcnccs. !n this scnsc, it is simplc.
ut it is not simplistic. Lack ot spccialist cxpcrtisc docs not imply
an inability to undcrstand or cvaluatc idcas. lgnorancc mcans
lack ot knowlcdgc, not lack ot intclligcncc. 1hc Survcy, thcrctorc,
is mcant to bc challcnging. !t draws a map ot thc sub| cct arca in
such a way as to stimulatc thought, and to invitc a critical parti-
cipation in thc cxploration ot idcas. 1his kind ot conccptual
cartography has its dangcrs ot coursc. thc sclcction ot what is
signihcant, and thc manncr ot its rcprcscntation will not bc to thc
liking ot cvcrybody, particular|y not, pcrhaps, to somc ot thosc
insidc thc disciplinc. ut thcsc survcys arc writtcn in thc bclict
XII PREFACE
that there must be an alternative to a technical account on the one
hand and an idiot's guide on the other if linguistics is to be made
relevant to people in the wider world.
Readngs
Some people will be content to read, and perhaps re-read, the
summary Survey. Others will want to pursue the subj ect and so
will use the Survey as the preliminary for more detailed study. The
Readings provide the necessary transition. For here the reader is
presented with texts extracted from the specialist literature. The
purpose of these readings is quite different from the Survey. It is to
get readers to focus on the specifcs of what is said and how it is
said in these source texts. Questions are provided to further this
purpose: they are designed to direct attention to points in each
text, how they compare across texts, and how they deal with the
issues discussed in the Survey. The idea is to give readers an initial
familiarity with the more specialist idiom of the linguistics liter
ature, where the issues might not be so readily accessible, and to
encourage them into close, critical readig.
References
One way of moving into more detailed study is through the
. Readings. Another is through the annotated References in the
third section of each book. Here there is a selection of works
( books and articles) for further reading. Accompanying com
ments indicate how these deal in more detail with the issues dis
cussed in the different chapters of the survey.
Glossar
Certain terms in the Survey appear in bold. These are terms used
in a special or technical sense in the discipline. Their meanings are
made clear in the discussion, but they are also explained in the
Glossary at the end of each book. The Glossary is cross
referenced to the Survey, and therefore serves at the same time as
an index. This enables readers to locate the term and what it
signifes in the more general discussion, thereby, in effect, using
the Survey as a summary work of reference.
PREFACE XIII
Use
The series has been.designed so as to be fexible in use. Each title is
separate and self-contained, with only the basic format in
common. The four" sections of the format, as described here, can
be drawn upon and combined in different ways, as required by
the needs, or interests, of different readers. Some may be content
with the Survey and the Glossary and'may not want to follow up
the suggested references. Some may not wish to venture into the
Readings. Again, the Survey might be considered as appropriate
preliminary reading for a course in applied linguistics or teacher
education, and the Readings more appropriate for seminar dis
cussion during the course. In short, the.notion of an introduction
will mean different things to different people, but in all cases the
concern is to provide access to specialist knowledge and stimulate
an awareness of its signifcance. This series as a whole has been
designed to provi
d
e this access and promote this awareness in
respect to diferent areas of language study.
H.G. WI DDOWS ON
XIV PREFACE
Definitions and background
Pragmatics is concerned with the study of meaning as commun
icated by a speaker (or writer) and interpreted by a listener (or
reader). It has, consequently, more to do with the analysis of what
people mean by their utterances than what the words or phrases
in those utterances might mean by themselves. Pragmatics is the
study of speaker meaning.
This type of study necessarily involves the interpretation of
what people mean in a particular context and how the context
influences what is said. It requires a consideration of how speakers
organize what they want to say in accordance with who they're
talking to, where, when, and under what circumstances.
Pragmatics is the study of contextual meaning.
This approach also necessarily explores how listeners can make
inferences about what is said in order to arrive at an interpreta- ,
tion of the speaker's intended meaning. This type of study
explores how a great deal of what is unsaid is recognized as part
of what is communicated. We might say that it is the investigation
of invisible meaning. Pragmatics is the study of how more gets
communicated than is said.
This perspective then raises the question of what determines the
choice between the said and the unsaid. The basic answer is tied to the '
notion of distance. Closeness, whether it is physical , social, or con
ceptual, implies shared experience. On the assumption of how close
or distant the listener is, speakers determine how much needs to be
said. Pragmatics is the study of the expression of relative distance.
These are the four areas that pragmatics is concerned with. To
understand how it got to be that way, we have to briefly review its
relationship with other areas of linguistic analysis.
DEFINITIONS AND BACKGROUND 3
Syntax, semantics, and pragmatics
One traditional distinction in language analysis contrasts prag
matics with syntax and semantics. Syntax i s the study of the
relationships between linguistic forms, how they are arranged iR
sequence, and which sequences are well-formed. This type of
study generally takes place without considering any world of ref
erence or any user of the forms. Semantics is the study of the
relationships between linguistic forms and entities in the" world;
that is, how words literally connect to things. Semantic analysis
al so attempts to establish the relationships between verbal
descriptions and states of affairs in the world as accurate (true) or
not, regardless of who produces that description.
Pragmatics is the study of the relationships between linguistic
forms and the users of those forms. In this three-part distinction,
only pragmatics allows humans into the analysis. The advantage
of studying language via pragmatics is that one can talk about
people's intended meanings, their assumptions, their purposes or
goals, and the kinds of actions ( for example, requests) that they
are performing when they speak. The big disadvantage is that all
these very human concepts are exremely difcult to analyze in a
consistent and obj ective way. Tw. o friends having a conversation
may imply some things and infer some others without providing
any clear linguistic evidence that we can point to as the explicit
source of 'the meaning' of what was communicated. Example [ I]
i s just such a probematic case. I heard the speakers, I knew what
they said, but I had no idea what was communicated.
[I] Her: So-did you?
Him: Hey-who wouldn't?
Thus, pragmatics is appealing because it's about how people
make sense of each other linguistically, but it can be a frustrating
area of study because it requires us to make sense of people and
what they have in mind.
Regularity
Luckily, people tend to behave in fairly regular ways when it
comes to using language. Some of that regularity derives from the
fact that people are members of social groups and follow general
4 SURVEY
patterns of behavior expected witin the group. Within a familiar
social 'group, we normally fnd it easy to be polite and say appro
priate things. In a new, unfamiliar social setting, we are often
unsure about what to say and worry that we might say the wrong
. thing.
When I frst lived in Saudi Arabia, I tended to answer questions
in Arabic about my health (the equivalent of 'How are you? ' ) with
the equivalent of my familiar routine responses of ' Okay' or
'

Fine'. However, I eventually noticed that when I asked a similar

question, people generally answered with a phrase that had the
literal meaning of 'Praise to God'. I soon learne' d to use the new
expression, wanting to be pragmatically appropriate in that con
text. My frst type of answer wasn't 'wrong' ( my vocabulary and
pronunciation weren't inaccurate) , but it did convey the meaning
that I was a social outsider who answered in an unexpected way.
In other words, more was being communicated than was being
said. Initially I did not know that: I had learned some linguistic
forms in the language without learning the pragmatics of how
those forms are used in a regular pattern by social isiders.
Another source of regularity in language use derives from the
fact that most people within a linguistic community have similar
basic experiences of the world and share a lot of non-linguistic
knowledge. Let's say that, in the middle of a conversation, I men
tion the information in [2].
[2] I found an old bicycle lying on the ground. The chain was
rusted and the tires were fat.
You are unlikely to ask why a chain and some tires were suddenly
being mentioned. I can normally assume that you will make
the' inference that if X is a bicycle, then'X has a chain and tires
(and many other regular parts). Because of this type of assump
tion, it would be pragmatically odd for me to have expressed [2]
as [
3
].
[
3
] I found an old bicycle. A bicycle has a chain. The chain was
rusted. A bicycle also has tires. The tires were fat.
You would perhaps think that more ws being communicated
than was being said and that you were being treated as someone
with no basic knowledge (i.e. as stupid) . Once again, nothing in
DEFI NITI ONS AND BACKGROUND 5
the use of the linguistic forms is inaccurate, but getting the prag
matics wrong might be ofensive.
The types of regularities j ust described are extremely simple
examples of language in use which are largely ignored. by most
linguistic analyses. To understand why it has become the province
of pragmatics to investigate these, and many other aspects of
ordinary language in use, we need to take 3 brief historical look at
how things got to be the way they are.
Ihepragmatlcswastebasket
For a long period in the study of language, there has been a very
stong interest in formal systems of analysis, ofen derived fom
mathema
t
ics and logic. The emphasis has been on discovering
some of the abstract principles that lie at the very core of lan
guage. By placing the investigtion of the abstract, potentially
universal, features of language in the center of their work tables,
linguists and philosophers of language tended to push any notes
they had on everyday language use to the edges. As the tables got
crowded, many of those notes on ordinary language in use began
to be knocked off and ended up in the wastebasket. That
overfowing wastebasket has become the source of much of what
will be discussed in the following pages. It is worth remembering
that the contents of that wastebasket were not originally or
ganized under a single category. Tey were defned negatively, as
the stuff that wasn't easily handled within the formal systems of
analysis. Consequently, in order to understand some of the mater
ial
"
that we're going to pull out of the wastebasket, we really have
to look at how it got there.
The tables upon which many linguists and philosophers of lan
guage worked were devoted to the analysis of language structure.
Consider the sentence in [4].
[4] The duck ran up to Mary and licked her
A syntactic approach to this sentence would be concerned with
the rules that determine the correct structure and exclude any
incorrect orderings such as *'Up duck Mary to the ran'. Syntactic
analysis would also be required to show that there is a missing ele
ment {'and _licked her'}
-
before the verb 'licked' and to e
x
plicate
6 SURVEY
the rules that allow that empty slot, or accept the pronoun 'it' in
that position. However, those working on syntax would have
thought it totally irrelevant if you tried to say that ducks don't do
that and maybe the speaker had meant to say ' dog' . Indeed, from
a purely syntactic perspective, a sentence like ' The bottle of
ketchup ran up to Mary' is just as well-formed as [4].
Over on the semantics side of the table, however, there would
have been concern. An entity labelled ' duck' has a meaning
feature ( animate) whereas a 'bottle of ketchup' would be (non
animate) . Since a verb like 'ran up to' requires something animate
as its subj ect, the word ' duck' is okay, but not a ' bottle of
ketchup' .
Semantics is also concerned with the truth-conditions of
propositions expressed in sentences. These propositions generally
correspond to the basic literal meaning of a simple clause and are
conventionally represented by the letters p, q, and r. Let's say that
the underlying meaning relationship being expressed in 'The duck
ran up to Mary' is the proposition p, and in 'the duck licked
Mary', it is the proposition q. These two propositions are j oined
by the logical connector symbol for conj unction, & ( called
' ampersand' ) . Thus, the propositional representation of the sen
tence in [4] is as in [ 5].
[5] p & q
If P is true and q is true, then p & q is true. If either p or q is not
true ( i. e. false) , then the conj unction of p & q is necessarily false.
This type of analysis is used extensively in formal semantics.
Unfortunately, in this type of analysis, whenever p & q is true,
it logically follows that q & P is true. Notice that q & p, in this
particular case, would have to be expressed as in [6] .
[ 6] The duck licked Mary and ran up to her.
In the everyday world of language use, this state of affairs is not
identical to the original situation described in [4] . There is a
sequence of two events being described and we expect that
sequence, in terms of occurrence, to be refected in the order of
mention .
.
If P involves some action and q involves another action, we
have an overwhelming tendency to interpret the conjunction
DEFI NITIONS AND BACKGROUND 7
' and' , not as logcal &, but as thc scqucntal cxprcsson ' and
thcn' . 1hs s anothcr cxamplc ot morc bcng communcatcd than
s sad. Wc mght proposc that thcrc s a rcgular prncplc ot lan-
guagc usc whch can bc statcd as n [;] .
[;] lntcrprct ordcr ot mcnton as a rcflccton ot ordcr ot occur-
rcncc.
What s cxprcsscd n [;] s not a rulc ot synta or scmantcs. lt
sn't

a rulc at all. lt s a pragmatc prncplc which wc trcqucntly
usc to makc scnsc ot what wc hcar and rcad, but whch wc can
gnorc t it docsn't apply n somc stuatons.
1hcrc arc many othcr prncplcs ot ths typc whch wll bc
cxplorcd n thc tollowng chaptcrs. ln Chaptcr 2 wc will start
with a rcally splc prncplc. thc morc two spcakcrs havc n com-
mon, thc lcss languagc thcy'll nccd to usc to dcntty tamlar
thngs. 1hs prncplc accounts tor thc trcqucnt usc ot words lkc
'ths' and 'that' to rctcr to thngs n a sharcd physcal contcxt tor
cxamplc, ' Would you lkc ths or that ! ' . Explorng ths basic
aspcct ot languagc n usc s thc study ot dcxs.
8 S URVEY
2
Deixis and distance
Deixis is a technical term (from Greek) for one of the most basic
things we do with utterances. It means 'pointing' via language.
Any linguistic form used to accomplish this 'pointing' is called a
deictic expression. When you notice a strange object and ask,
'What's that?', you are using a deictic expression ('that') to indic
ate something in the immediate context. Deictic expressions are
also sometimes called indexicals. They are among the frst forms
to be spoken by very young children and can be used to indicate
people via person deixis ('me', 'you'), or location via spatial deixis
('here', 'there'), or time via temporal deixis ('now', 'then'). All these
expressions depend, for their interpretation, on the speaker and
hearer sharing the same context. Indeed, deictic expressions have
their most basic uses in face-to-face spoken interaction where
utterances such as [1] are easily understood by the people present,
but may need a translation for someone not right there.
[11 I'll put this here.
(Of course, you understood that Jim was telling Anne that he was
about to put an extra house key in one of the kitchen drawers.)
Deixis is clearly a form of referring that is tied to the speaker's
context, with the most basic distinction between deictic expres
sions being 'near speaker' versus 'away from speaker'. In English,
the 'near speaker', or proximal terms, are 'this', 'here', 'now'. The
'away from speaker', or distal terms, are 'that', 'there', 'then'.
Proximal terms are typically interpreted in terms of the speaker's
location, or the deictic center, so that 'now' is generally under
stood as referring to some point or period in time that has the time
of the speaker's utterance at its center. Distal terms can simply
DEIXIS AND DISTANCE 9
ndcatc ' away trom spcakcr', but, n somc languagcs, can bc uscd
to dstngush bctwccn 'ncar addrcsscc' and ' away trom both
spcakcr and addrcsscc' . 1hus, n |apancsc, thc translaton ot thc
pronoun ' that' wl dstngush bctwccn 'that ncar addrcsscc'
'sore' and ' that dstant trom both spcakcr and addrcsscc' 'are'
wth a thrd tcrm bcng uscd tor thc proxmal 'ths ncar spcakcr'
'kore' .
Person deixis
1hc dstncton | ust dcscrbcd nvolvcs pcrson dcxs, wth thc
spcakcr ( 'I' ) and thc addrcsscc 'you' mcntoncd. 1hc smplcty
ot thcsc torms dsguscs thc complcxty ot thcr usc. 1o lcarn thcsc
dcctc cxprcssons, wc havc o dscovcr that cach pcrson n a con-
vcrsaton shhs trom bcng 'l' to bcng 'you' constantly. All young
chldrcn go through a stagc n thcr lcarnng whcrc ths dstncton
sccms problcmatc and thcy say thngs lkc 'Rcad you a story'
nstcad ot ' mc' whcn handng ovcr a tavortc book.
Pcrson dcxs clcarly opcratcs on a basc thrcc-part dvson,
cxcmplhcd by thc pronouns tor hrst pcrson ('1'), sccond pcrson
'you' , and thrd pcrson 'hc', ' shc', or 't' . ln many languagcs
thcsc dcctc catcgorcs ot spcakcr, addrcsscc, and othcr s arc
claboratcd wth markcrs ot rclatvc socal status tor cxamplc,
addrcsscc wth hghcr stams vcrsus addrcsscc wth lowcr status).
Exprcssons whch ndcatc hghcr status arc dcscrbcd as hon
orifcs. 1hc dscusson ot thc crcumstanccs whch lcad to thc
chocc ot onc ot thcsc torms rathcr than anothcr s somctmcs
dcscrbcd as social deixis.
A tarly wcll-known cxamplc ot a socal contrast cncodcd
wthn pcrson dcxs s thc dstncton bctwccn torms uscd tor a
tamlar vcrsus a non-tamlar addrcsscc n somc languagcs. 1hs
s known as thc TN distincion, trom thc Frcnch torms 'tu' taml-
ar and 'vous' non-tamlar , and s tound n many languagcs
ncluding Gcrman ('du/Sie') and Spanish ('tu/Usted' ) . 1hc chocc
ot onc torm wll ccrtanly communcatc somcthng not drcctly
said about thc spcakcr's vcw ot hs or hcr rclationshp wth thc
addrcsscc. ln thosc socal contcxts whcrc ndvduals typcally
mark dstnctons bctwccn thc socal status ot thc spcakcr and
addrcsscc, thc hghcr, oldcr, and morc powcrtul spcakcr wll tcnd
to S URVEY
to usc thc 'tu' vcrsion to a |owcr, youngcr, and lcss powcrtul
addrcsscc, and bc addrcsscd by thc 'vous' torm in rcturn. Whcn
socia| changc is taking p|acc, as tor cxamp|c in modcrn Spain,
whcrc a young busincsswoman highcr cconomic status is ta|k-
ing to hcr o|dcr clcaning |ady lowcr cconomic status , how do
thcy addrcss cach othcr! I am told that thc agc distinction rcmains
morc powcrtul than thc cconomic distinction and thc o| dcr
woman uscs 'tu' and thc youngcr uscs ' U sted' .
1hc Spanish non-tami|iar vcrsion (' Usted' ) is historica||y
rclatcd to a torm which was uscd to rctcr to ncithcr hrst pcrson
spcakcr nor sccond pcrson addrcsscc , but to third pcrson
somc othcr . In dcictic tcrms, third pcrson is not a dircct parti-
cipant in basic l-you intcraction and, bcing an outsidcr, is ncccs-
sarily morc distant. 1hird pcrson pronouns arc conscqucnt|y
dista| torms in tcrms ot pcrson dcixis. Using a third pcrson torm,
whcrc a sccond pcrson torm wou|d bc possib|c, is onc way ot
communicating distancc and non-tamiliarity . 1his can bc donc
in English tor an ironic or humorous purposc as whcn onc pcrson,
who's vcry busy in thc kitchcn, addrcsscs anothcr, who's bcing
vcry |azy, as in jz] .
jz] Wou|d his highncss |ikc somc cottcc!
1hc distancc associatcd with third pcrson torms is also uscd to
makc potcntial accusations tor cxamp|c, 'you didn't clcan up'
|css dircct, as in j j a. ] , or to makc a potcntia||y pcrsona| issuc sccm
likc an impcrsonal onc, bascd on a gcncral rulc, as in jj b. ] .
j j ] a. Somcbody didn't c|can up attcr himsclt.
b. Each pcrson has to c|can up attcr him or hcrsc|t.
Ot coursc, thc spcakcr can statc such gcncra| ' ru|cs' as applying to
thc spcakcr plus othcr s , by using thc hrst pcrson p|ura| 'wc' , as
in j{] .
j{] Wc clcan up attcr oursclvcs around hcrc.
1hcrc is, in English, a potcntia| ambiguity in such uscs which
al|ows two dittcrcnt intcrprctations. 1hcrc is an exclusive 'we'
spcakcr plus othcrs , cxcluding addrcsscc and an inclusive 'we'
spcakcr and addrcsscc inc|udcd . Somc languagcs grammaticizc
this distinction tor cxamp|c, Fi| ian has 'keimami' tor cxc|usivc
hrst pcrson p|ura| and 'keda' tor inclusivc hrst pcrson p|ural .
DEI XI S AND DI S TANCE
ln English, thc ambiguity prcscnt in j{] providcs a subtlc opportu-
nity tor a hcarcr to dccidc what was communicatcd. Eithcr thc
marcr dccidcs that hc or s
_
c is a mcmbcr ot thc group to whom
thc rulc applics i. c. an addrcsscc) or an outsidcr to whom thc rulc
docs not apply i.c. not an addrcsscc). ln this casc thc hcarcr gcts
to dccidc thc kind ot ' morc' that is bcing communicatcd.
1hc inclusivc-cxclusivc distinction may also bc notcd in thc
dittcrcncc bctwccn saying 'Lct's go' to somc tricnds) and ' Lct us .
go' to somconc who has capturcd thc spcakcr and tricnds). 1hc
action ot going is inclusivc in thc hrst, but cxclusivc in thc sccond.
Spatial deixis
1hc conccpt ot distancc alrcady mcntioncd is clcarly rclcvant to
spatial dcixis, whcrc thc rclativc location ot pcoplc and things is
bcing indicatcd. Contcmporary English makcs usc ot only two
advcrbs, 'hcrc' and 'thcrc', tor thc basic distinction, but in oldcr
tcxts and in somc dialccts, a much largcr sct ot dcictic cxprcssions
can bc tound. Although 'yondcr' morc distant trom spcakcr is
still uscd, words likc 'hithcr' to this placc) and 'thcncc' trom that
placc) now sound archaic. 1hcsc last two advcrbs includc thc
mcaning ot motion toward or away trom thc spcakcr. Somc vcrbs
ot motion, such as 'comc' and 'go' , rctain a dcictic scnsc whcn
thcy arc uscd to mark movcmcnt toward thc spcakcr ' Comc to
bcd| ') or away trom thc spcakcr ' Go to bcd| ' .
Onc vcrsion ot thc conccpt ot motion toward spcakcr i . c.
bccoming visiblc), sccms to bc thc hrst dcictic mcaning lcarncd by
childrcn and charactcrizcs thcir usc ot words likc 'this' and ' hcrc'
can bc sccn). 1hcy arc distinct hom 'that' and 'thcrc' which arc
associatcd with things that movc out ot thc child's visual spacc
( can no longcr bc sccn .
ln considcring spatial dcixis, howcvcr, i t is important to
rcmcmbcr that location trom thc spcakcr's pcrspcctivc can bc
hxcd mcntally as wcll as physically. Spcakcrs tcmporarily away
trom thcir homc location will ottcn continuc to usc 'hcrc' to mcan
thc physically distant homc location, as it thcy wcrc still in that
location. Spcakcrs also sccm to bc ablc to pro| cct thcmsclvcs into
othcr locations prior to actually bcing in thosc locations, as whcn
thcy say ' l' ll comc latcr' ( movcmcnt to addrcsscc's location).
1 2 S URVEY
1his s somctimcs dcscribcd as deictic projection and wc makc
morc usc ot its possibilitics as morc tcchnology allows us to
manipulatc location. lt 'hcrc' mcans thc placc ot thc spcakcr's
uttcrancc and 'now' mcans thc timc ot thc spcakcr's uttcrancc ,
thcn an uttcrancc such as j y] should bc nonscnsc.
j y] l am not hcrc now.
Howcvcr, l can say j y] into thc rccordcr ot a tclcphonc answcring
machinc, pro| ccting that thc 'now' will apply to any timc somc-
onc trics to call mc, and not to whcn l actually rccord
thc words . lndccd, rccording j y ] is a kind ot dramatic pcr-
tormancc tor a tuturc audicncc in which l pro| cct my prcscncc to
bc in thc rcquircd location. A similar dcictic pro| cction is accom-
plishcd via dramatic pcrtorancc whcn l usc dircct spccch to
rcprcscnt thc pcrson, location, and tcclings ot somconc or somc-
thing clsc. For cxamplc, l could bc tclling you about a visit to a pct
storc, as in j 6] .
j6] l was looking at this littlc puppy in a cagc with such a sad
look on its tacc. It was likc, 'Oh, l' m so unhappy hcrc, will
you sct mc trcc!'
1hc ' hcrc' ot thc cagc i s not thc actual physical location ot thc
pcrson uttcring thc words thc spcakcr , but is instcad thc loca-
tion ot that pcrson pcrtorming in thc rolc ot thc puppy.
lt may bc that thc truly pragmatic basis ot spatial dcixis is actu-
ally psychological distance. Physically closc ob| ccts will tcnd to bc
trcatcd by thc spcakcr as psychologically closc. Also, somcthing
that is physically distant will gcncrally bc trcatcd as psycho-
logically distant tor cxamplc, 'that man ovcr thcrc' . Howcvcr, a
spcakcr may also wish to mark somcthing that is physically closc
tor cxamplc, a pcrtumc bcing snittcd by thc spcakcr) as psycho-
logically distant 'l don't likc that' . In this analysis, a word likc
'that' docs not havc a hxcd i. c. scmantic mcaning, instcad, it is
' invcstcd' with mcaning in a contcxt by a spcakcr.
Similar psychological proccsscs sccm to bc at work in our dis-
tinctions bctwccn proximal and distal cxprcssions uscd to mark
tcmporal dcixis.
DEIXI S AND DI STANCE 1
l


Temporal deixis
Wc havc alrcady notcd thc usc ot thc proxmal torm 'now' as ndc-
atng both thc tmc concdng wth thc spcakcr's uttcrancc and
thc tmc ot thc spcakcr's vocc bcng hcard thc hcarcr's 'now' . ln
contrast to ' now' , thc dstal cxprcsson 'thcn' applcs to both past
[;a. ] and tuturc j;b. ] tmc rclatvc to thc spcakcr's prcscnt tmc.
[;] a. ovcmbcr zznd, t6j ! l was n Scotland thcn.
b. Dnncr at 8. j o on Saturday! Okay, l'll scc you thcn.
lt s worth notng that wc also usc claboratc systcms ot non-
dcctc tcmporal rctcrcncc such as calcndar tmc datcs, as n
[;a. ] } and clock tmc hours, as n j;b. ] . Howcvcr, thcsc torms ot
tcmporal rctcrcncc arc lcarncd a lot latcr than thc dcctc cxprcs-
sons lkc 'ycstcrday' , 'tomorrow', 'today', 'tonght' , ' ncxt wcck' ,
' last wcck' , 'ths wcck' . All thcsc cxprcssons dcpcnd tor thcr
ntcrprctaton on knowng thc rclcvant uttcrancc tmc. lt wc don't
know thc uttcrancc . c. scrbblng tmc ot a notc, as n j8] , on an
othcc door, wc won't know t wc havc a short or a long wat
ahcad.
j8] ack n an hour.
Smlarly, t wc rcturn thc ncxt day to a bar that dsplays thc noticc
n j] , thcn wc wll stll bc dcctcally} onc day carly tor thc trcc
drnk.
j] Frcc ccr 1omorrow.
1hc psychologcal bass ot tcmporal dcxs sccms to bc smlar
to that ot spatal dcxs. Wc can trcat tcmporal cvcnts as ob| ccts
that movc toward us nto vcw or away trom us out ot vicw .
Onc mctaphor uscd n Englsh is ot cvcnts comng toward thc
spcakcr trom thc tuturc tor cxamplc, 'thc comng wcck' , ' thc
approachng ycar' and gong away trom thc spcakcr to thc past
tor cxamplc, 'n days gonc by' , 'thc past wcck' . Wc also sccm to
trcat thc ncar or mmcdatc tuturc as bcng closc to uttcrancc tmc
by usng thc proxmal dcctic 'ths' , as n 'ths comng wcckcnd'
or 'ths comng) 1hursday' .
Onc basc but ohcn unrccognzcd typc ot tcmporal dcxs n
Englsh s n thc chocc ot vcrb tcnsc. Whcrcas othcr languagcs
havc many dttcrcnt torms ot thc vcrb as dttcrcnt tcnscs, Englsh
|
1q S URVEY
l
.J
1

has only two basc torms, thc prcscnt as n j oa. ] , and thc past as
in j ob. ] .
j to] a. l lvc hcr now.
b. l lvcd thcrc thcn.
1hc prcscnt tcnsc s thc proxmal torm and thc past tcnsc is thc
distal torm. Somcthng havng takcn placc n thc past, as n j t ta. ] ,
i s typcally trcatcd as dstant trom thc spcakcr's currcnt stuaton.
Pcrha ps lcss obvously, somcthng that is trcatcd as cxtrcmcly
unlkcly or mpossblc) trom thc spcakcr's currcnt situaton s
also markcd va thc dstal past tcnsc) torm, as n j t t b. ] .
j t t] a. l could swm whcn l was a chld).
b. l could bc in Hawaii it l had a lot ot moncy).
1hc past tcnsc s always uscd n Englsh n thosc i[-clauscs that
mark cvcnts prcscntcd by thc spcakcr as not bcing closc to prcscnt
rcalty as n j t z] .
j t z] a. lt l had a yacht, . . .
b. lt l was rch, . . .
Mcthcr ot thc dcas cxprcsscd n j t z] arc to bc trcatcd as havng
happcncd n past tmc. 1hcy arc prcscntcd as dcctcally dstant
trom thc spcakcr's currcnt stuation. So distant, ndccd, that thcy
actualIy communcatc thc ncgatvc wc ntcr that thc spcakcr has
no yacht and s not rch).
ln ordcr to undcrstand many English conditional constructions
ncluding thosc ot thc torm 'Had l known sooncr . . . '), wc havc to
rccognzc that, n tcmporal dcxis, thc rcmotc or dstal torm can
bc uscd to communcatc not only distancc trom currcnt tmc, but
also dstancc trom currcnt rcality or tacts.
Oeixis and grammar
1hc basc distnctons prcscntcd so tar tor pcrson, spatial, and
tcmporal dcxs can all bc sccn at work n onc ot thc most com-
mon structural dstnctions madc in Englsh grammarthat
bctwccn dircct and ndrcct or rcportcd) spccch. As alrcady
dcscri bcd, thc dcictic cxprcssions tor pcrson ' you' , pl acc
'hcrc' ), and tmc 'ths cvcnng') can all bc intcrprctcd wthn thc
samc contcxt as thc spcakcr who uttcrs j t j a. ] .
DEIXI S AND DI S TANCE t y
[ t] a. Arc you planning to bc hcrc this cvcning!
b. ! askcd hcr it shc was planning to bc thcrc that
cvcnng.
Whcn thc contcxt shihs, as tor cxamplc in j t b. ] , to onc in which
! rcport thc prcvious uttcrancc, thcn thc prcvious uttcrancc is
markcd dcictically as rclativc to thc circumstanccs ot asking.
otc that thc proximal torms prcscntcd in j ta. ] havc shihcd to
thc corrcsponding distal torms in [t j b. ] . 1his vcry rcgular dittcr-
cncc in English rcportcd discoursc marks a distinction bctwccn
thc 'ncar spcakcr' mcaning ot

dircct spccch and thc 'away trom

spcakcr' mcaning ot indircct spccch. 1hc proximal dcictic torms
ot a dircct spccch rcporting communicatc, ottcn dramatically, a
scnsc ot bcing in thc samc contcxt as thc uttcrancc. 1hc distal
dcictic torms ot indircct spccch rcporting makc thc original
spccch cvcnt sccm morc rcmotc.
!t should not bc a surprisc to lcarn that dcictic cxprcssions wcrc
all to bc tound in thc pragmatics wastcbaskct. 1hcir intcrprcta-
tion dcpcnds on thc contcxt, thc spcakcr's intcntion, and thcy
cxprcss rclativc distancc. Givcn thcir small sizc and cxtrcmcly
widc rangc ot possiblc uscs, dcictic cxprcssions always communic-
atc much morc than is said.
t 6 S URVEY
3
Reference and inference
Throughout the preceding discussion of deixis, there was an
assumption that the use of words to refer to people and things
was a relatively straightforward matter. It is indeed fairly easy for
people to do, but it is rather diffcult to explain how they do it. We
do know that words themselves don't refer to anything. People
refer. We might best think of reference as an act in which a
speaker, or writer, uses linguistic forms to enable a listener, or
reader, to identify something.
Those linguistic forms are referring expressions, which can be
proper nouns (for example, 'Shakespeare', 'Cathy Revuelto',
'Hawaii'), noun phrases which are defnite (for example, 'the
author', 'the singer', 'the island'), or indefnite (for example, 'a
man', 'a woman', 'a beautiful place'), and pronouns (for example,
'he', 'her', 'it', 'them'). The choice of one type of referring expres
sion rather than another seems to be based, to a large extent, on
what the speaker assumes the listener already knows. In shared
visual contexts, those pronouns that function as deictic expres
sions (for example, 'Take this'; 'Look at him!') may be suffcient
for successful reference, but where identifcation seems more
diffcult, more elaborate noun phrases may be used (for example,
'Remember the old foreign guy with the funny hat?').
Reference, then, is clearly tied to the speaker's goals (for ex
ample, to identify something) and the speaker's beliefs (i.e. can
the listener be expected to know that particular something?) in
the use of language. For successful reference to occur, we must
also recognize the role of inference. Because there is no direct rela
tionship between entities and words, the listener's task is to infer
correctly which entity the speaker intends to identify by using a
REFERENCE AND INFERENCE 1
7
particular referring expression. It is not un usual for people to
want to refer to some entity or person without knowing exactly
which 'name' would be the best word to use. We can even use
vague expressions
.
( for eample, 'the blue thing', 'that icky stuff',
'01' what's his name', 'the thingamajig' ) , relying on the listener's
ability to infer what referent we have in mind. Speakers even
invent names. There was one man who delivered packages to our .
offce whose 'real' name I didn't know, but whose identity I could
infer when the secretary referred to him as in [ I] .
[ I] Mister Aftershave i s late today.
The example in [ I] may serve to illustrate that reference is not
based on an objectively correct (versus incorrect) naming, but on
some locally successfl (versus unsuccessfl) choice of expression.
We might also note fom example [I] that successful reference
is necessarily collaborative, with both the speaker and the listener
having a role in thinking about what the other has in mind.
Referential and atributive uses
It is important to recognize that not all referring expressions hve
identifable physical referents. Indefnite noun phrases can be
used to identify a physically present entity as in [2a. ] , but they can
also be used to describe entities that are assumed to exist, but are
unknown, as in [2b. ] , or entities that, as far as we know, don't
exist [2C. ] .
[2] a. There's a man waiting for you.
b. He wants to marry a woman with lots of money.
c. We'd love to fnd a nine-foot-tall basketball player.
The expression in [2b. ], 'a woman with lots of money', can desig
nate an entity that is known to the speaker only in terms of its
descriptive properties. The word 'a' could be replaced by 'any' in
this case. This is sometimes called an atribute use, meaning 'who
ever/whatever fts the description' . It would be distinct from a refer
enial use whereby I actually have a person in mind and, instead of
using her name or some other description, I choose the expression
i [2b. ] , perhaps because I think you'd be more interested in hear
ing that this woman has lots of money than that she has a name.
1 8 S URVEY
A similar distinction can bc tound with dchnitc noun phrascs.
During a ncws rcport on a mystcrious dcath, thc rcportcr may say
[ j] without knowing tor surc it thcrc is a pcrson who could bc thc
rctcrcnt ot thc dchnitc cxprcssion 'thc killcr'. 1his would bc an
attributivc usc i . c. 'whocvcr did thc killing' , bascd on thc
spcakcr's assumption that a rctcrcnt must cxist.
[j ] 1hcrc was no sign ot thc ki|lcr.
Howcvcr, it a particular individua| had bccn idcntihcd as having
donc thc ki|ling and had bccn chascd into a bui|ding, but cscapcd,
thcn uttcring thc scntcncc in [j ] about that individual would bc a
rctcrcntial usc, bascd on thc spcakcr's knowlcdgc that a rctcrcnt
docs cxist.
Thc point ot this distinction is that cxprcssions thcmsc|vcs can-
not bc trcatcd as having rctcrcncc as is ottcn assumcd in scmantic
trcatmcnts , but arc, or arc not, 'invcstcd' with rctcrcntial tunc-
tion in a contcxt by a spcakcr or writcr. Spcakcrs ottcn invitc us to
assumc, via attributivc uscs, that wc can idcntity what thcy'rc
ta|king about, cvcn whcn thc cntity or individual dcscribcd may
not cxist, as in [ zc. ] . Somc othcr tamous mcmbcrs ot that group
arc thc tooth tairy and Santa Claus.
Names and referents
1hc vcrsion ot rctcrcncc bcing prcscntcd hcrc is onc in which thcrc
is a basic 'intcntion-to-idcnti' and a 'rccognition-ot-intcntion'
co|laboration at work. 1his proccss nccd not only work bctwccn
onc spcakcr and onc |istcncr, it appcars to work, in tcrms ot con-
vcntion, bctwccn all mcmbcrs ot a community who sharc a com-
mon languagc and culturc. 1hat is, thcrc is a convcntion that
ccrtain rctcrring cxprcssions wi|l bc uscd to idcnti ccrtain cntitics
on a rcgular basis. !t is our dai|y cxpcricncc ot thc succcsstu| op-
cration ot this convcntion that may causc us to assumc that rctcr-
ring cxprcssions can on|y dcsignatc vcry spccihc cntitics. 1his
assumption may lcad us to think that a namc or propcr noun |ikc
'Shakcspcarc' can only bc uscd to idcnti onc spccihc pcrson, and
an cxprcssion containing a common noun, such as 'thc chccsc
sandwich', can only bc uscd to idcnti a spccihc thing. 1his bclict
is mistakcn. A tru|y pragmatic vicw ot rctcrcncc a|lows us to scc
REFERENCE AND I NFERENCE 1y
how a pcrson can bc idcntihcd via thc cxprcssion, 'thc chccsc sand-
wich', and a thing can bc idcntihcd via thc namc, ' Shakcspcarc' .
or cxamplc, it would not bc strangc tor onc studcnt to ask
anothcr thc qucstion in [qa. ] and rcccivc thc rcply in [qb. ] .
[q] a. Can ! borrow your Shakcspcarc!
b. Ycah, it's ovcr thcrc on thc tablc.
Givcn thc contcxt | ust crcatcd, thc intcndcd rctcrcnt and thc
intcrrcd rctcrcnt would not bc a pcrson, but probably a book
noticc thc pronoun 'it' .
! n a rcstaurant, onc waitcr brings out an ordcr ot tood tor
anothcr waitcr and asks him j ya. ] and hcars j y b.] in rcply.
j y] a. Whcrc's thc chccsc sandwich sitting!
b. Hc's ovcr thcrc by thc window.
Givcn thc contcxt, thc rctcrcnt bcing idcntihcd is not a thing, but
a pcrson noticc thc pronoun 'hc' .
1hc cxamplcs i n [q] and [ y] may allow us to scc morc clcarly
how rctcrcncc actually works. 1hc Shakcspcarc cxamplc in jq]
suggcsts that thcrc is a convcntional and potcntially culturc-
spccihc sct ot cntitics that can bc idcntitcd by thc usc ot a writcr's
namc. Lct us call thcm 'things thc writcr produccd' . 1his would
allow us to makc scnsc ot thc scntcnccs in j 6] .
j 6] a. Shakcspcarc takcs up thc wholc bottom shclt.
b. Wc'rc going to scc Shakcspcarc in London.
c. I hatcd Shakcspcarc at school.
Obviously, this convcntion docs not only apply to writcrs, but
also to artists [ya. ] , composcrs [yb. ], musicians jyc.] , and many
othcr produccrs ot ob| ccts.
[y] a. Picasso's on thc tar wall.
b. 1hc ncw Mozart is bcttcr valuc than thc ach.
c. My Rolling Stoncs is missing.
1hcrc appcars to bc a pragmatic connecion bctwccn propcr namcs
and ob| ccts that will bc convcntionally associatcd, within a socio-
culturally dchncd community, with thosc namcs. Using a propcr
namc rctcrcntially to idcntiq any such ob| cct invitcs thc listcncr
to makc thc cxpcctcd intcrcncc tor cxamplc, trom namc ot
writcr to book by writcr and thcrcby show himsclt or hcrsclt
2C S URVEY
to bc a mcmbcr ot thc samc community as thc spcakcr. ln such
cascs, it is rathcr obvious that morc is bcing communicatcd than
is said.
1hc naturc ot rctcrcncc intcrprctation | ust dcscribcd is also
what allows rcadcrs to makc scnsc ot ncwspapcr hcadlincs using
namcs ot countrics, as cxcmplihcd in [ &a. ] whcrc thc rctcrcnt is to
bc undcrstood as a socccr tcam, not as a govcrnmcnt, and in [ 8b. ]
whcrc i t i s to bc undcrstood as a govcrnmcnt, not as a socccr
tcam.
[ 8] a. razil wins World Cup.
b. |apan wins hrst round ot tradc talks.
The role of co-tex
ln many ot thc prcccding cxamplcs, our ability to idcntity
intcndcd rctcrcnts has actually dcpcndcd on morc than our
undcrstanding ot thc rctcrring cxprcssion. lt has bccn aidcd by thc
linguistic matcrial, or co-tex, accompanying thc rctcrring cxprcs-
sion. Whcn [ 8a. ] appcarcd as a hcadlinc, 'razil' was a rctcrring
cxprcssion and 'wins Wor|d Cup' was part ot thc co-tcxt thc rcst
ot thc ncwspapcr was morc co-tcxt} . 1hc co-tcxt clcarly limits thc
rangc ot possiblc intcrprctations wc might havc tor a word likc
'razil' . lt is conscqucntly mislcading to think ot rctcrcncc bcing
undcrstood solcly in tcrms ot our ability to idcntity rctcrcnts via
thc rctcrring cxprcssion. 1hc rctcrring cxprcssion actually pro-
vidcs a range of reference, that is, a numbcr ot possiblc rctcrcnts.
Kcturning to a prcvious cxamplc, wc can show that, whi|c thc
phrasc 'thc chccsc sandwich' stays thc samc, thc dittcrcnt co-tcxts
in [a. ] and [b. ] lcad to a dittcrcnt typc ot intcrprctation in cach
casc i. c. 'tood' in [a. ] and 'pcrson' in [b. ] } .
[] a. 1hc chccsc sandwich is madc with whitc brcad.
b. 1hc chccsc sandwich lctt without paying.
Ot coursc, co-tcxt is | ust a linguistic part ot thc cnvironmcnt in
which a rcfcrring cxprcssion is uscd. 1hc physical cnvironmcnt, or
contex, is pcrhaps morc casily rccognizcd as having a powcrtul
impact on how rctcrring cxprcssions arc to bc intcrprctcd. 1hc phys-
ical contcxt ot a rcstaurant, and pcrhaps cvcn thc spccch convcn-
tions ot thosc who work thcrc, may bc crucial to thc intcrprctation
REFERENCE AND I NFERENCE zt
ot j b. ] . Similar|y, it is uscm| to know that a hospita| is thc contcxt
tor j toa. ] , a dcntist's omcc tor j tob. ] , and a hotc| rcccption tor
j toc.] .
j to] a. 1hc hc
_
rt-attack mustn't bc movcd.
b. Your tcn-thirty | ust canccllcd.
c. A couplc ot rooms havc comp|aincd about thc hcat.
1hc cxamp|cs in j to] providc somc support tor an ana|ysis ot
rctcrcncc that dcpcnds on |ocal contcxt and thc loca| know|cdgc
ot thc participants. lt may crucia|ly dcpcnd on tamiliarity with thc
|ocal socio-cultura| convcntions as thc bass tor intcrcncc tor
cxamplc, it a pcrson is in a hospital with an i|lncss, thcn hc or shc
can bc idcntihcd by urscs via thc namc ot thc i||ncss} . 1hcsc con-
vcntions may dittcr substantia||y trom onc social group to
anothcr and may bc markcd dittcrcntly trom onc languagc to
anothcr. Rctcrcncc, thcn, is not simply a rc|ationship bctwccn thc
mcaning ot a word or phrasc and an ob| cct or pcrson in thc wor|d.
lt is a socia| act, in which thc spcakcr assumcs that thc word or
phrasc choscn to idcntity an ob| cct or pcrson will bc intcrprctcd
as thc spcakcr intcndcd.
Anaphoric reference
1hc prcccding discussion has bccn conccrncd with singlc acts ot
rctcrcncc. ln most ot our ta|k and writing, howcvcr, wc havc to
kccp track ot who or what wc arc ta|king aboutfor morc than onc
scntcncc at a timc. Ahcr thc initial introduction ot somc cntity,
spcakcrs wil| usc various cxprcssions to maintain rcfcrcncc, as in
j t t ] .
j t t ] ln thc hlm, a man and a woman wcrc trying to wash a cat.
1hc man was ho|ding thc cat whi|c thc woman pourcd
watcr on it. Hc said somcthing to hcr and thcy startcd
|aughing.
ln English, initial rctcrcncc, or introductory mcntion, is ottcn
indchnitc ' a man' , ' a woman', ' a cat' . In [t t] thc dchnitc noun
phrascs [ 'thc man', 'thc cat', 'thc oman' j and thc pronouns 'it',
'hc' , 'hcr', 'thcy' } arc cxamp|cs ot subscqucnt rctcrcncc to a|rcady
introduccd rctcrcnts, gcncra||y known as anphoric rctcrcncc, or
zz S URVEY
anaphora. In technical terms, the second or subsequent expres
sion is the anaphor and the initial expression is the antecedent.
It is tempting to think of anaphoric reference as a process of
continuing to identif exactly the same entity as denoted by the
antecedent. In many cases, that assumption makes little difference
to the interpretation, but in those cases where some change or
effect is described, the anaphoric reference must be interpreted
differently. In example [ I 2] from a recipe, the initial referring
expression 'six potatoes' identifes something different from the
anaphoric pronoun 'them' which must be interpreted as 'the six
peeled and sliced potatoes'.
[ 1 2] Peel and slice six potatoes. Put them in cold salted water.
There is also a reversal of the antecedent-anaphor pattern some
times found at the beginning of stories, as in example [1 3 ] .
[ 1 3] I turned the corner and almost stepped on it. There was a
large snake in the middle of the path.
Note that the pronoun 'ie is used frst and is diffcult to interpret
until the full noun phrase is presented in the next line. This pat
tern is technically known as cataphora, and is much less common
than ana phora.
There is a range of expressions which are used for anaphoric
reference in English. The most typical forms are pronouns, such
as 'it' in [ I 4a. ] , but defnite noun phrases are also used, for ex
ample, 'the slices' in [ 14b. ] .
[ 14] a. Peel an onion and slice it.
h. Drop the slices into hot oil.
c. Cook for three minutes.
When the interpretation requires us to identify an entity, as in
' Cook ( ? ) for three minutes', in [ 14C. ] , and no linguistic expres
sion is present, it is called zero anaphora, or ellipsis. The use of zero
anaphora as a means of maintaining reference clearly creates an
expectation that the listener will be able to infer who or what the
speaker intends to identify. It is also another 0 bvious case of more
being communicated than is said.
The listener is also expected to make more specifc types of
inference when the anaphoric expressions don't seem to be lin
guistically connected to their antecedents. This point was noted in
REFERENCE AND I NFERENCE 23
Chaptcr t with thc ' bicyclc' cxamplc, and is turthcr illustratcd
in j t y ] .
j t y ] a. l | ust rcntcd a housc. 1hc kitchcn is rcally big.
b. Wc had Chardonnay with dinncr. 1hc winc was thc
bcst part.
c. 1hc bus camc on timc, but hc didn't stop.
Making scnsc ot j t ya. ] rcquircs an intcrcncc i. c. it x is a housc,
thcn x has a kitchcn to makc thc anaphoric conncction. Such
intcrcnccs dcpcnd on assumcd kn

wlcdgc which, as in j t yb. ] ,

may bc much morc spccihc i. c. Chardonnay i s a kind ot winc . ln
addition, thc intcrcncc can bc considcrcd so automatic tor somc
spcakcrs tor cxamplc, a bus has a drivcr , that thcy can go
straight to a pronoun tor anaphoric rcIcrcncc, as in jt yc. ] . ln this
cxamplc, notc that thc antcccdcnt 'thc bus ' and thc anaphor
' hc' arc not in grammatical agrccmcnt i . c. normally a bus
would bc ' it' . As pointcd out alrcady, succcsstul rctcrcncc docs
not dcpcnd on somc strictly litcral, or grammatically 'corrcct',
rclationship bctwccn thc propcrtics ot thc rctcrcnt and thc rctcr-
ring cxprcssion choscn. 1hc word ' sandwich' can idcntity a pcr-
son and thc pronoun 'hc' can bc an anaphor tor a thing. 1hc kcy
to making scnsc ot rctcrcncc is that pragmatic proccss whcrcby
spcakcrs sclcct linguistic cxprcssions with thc intcntion ot idcnti-
tying ccrtain cntitics and with thc assumption that listcncrs will
collaboratc and intcrprct thosc cxprcssions as thc spcakcr
intcndcd.
1hc social dimcnsion ot rctcrcncc may also bc ticd to thc cttcct
ot collaboration. 1hc immcdiatc rccognition ot an intcndcd rctcr-
cnt, cvcn whcn a minimal rctcrring cxprcssion tor cxamplc, a
pronoun is uscd, rcprcscnts somcthing sharcd, somcthing in
common, and hcncc social closcncss. Succcsstul rctcrcncc mcans
that an intcntion was rccognizcd, via intcrcncc, indicating a kind
ot sharcd knowlcdgc and hcncc social conncction. 1hc assump-
tion ot sharcd knowlcdgc is also crucially involvcd in thc study ot
prcsupposition.
zq S URVEY
l
t

4
Presupposition and entailment
In the preceding discussion of reference, there was an appeal to
the idea that speakers assume certain information is already
known by their listeners. Because it is treated as known, such
information will generally not be stated and consequently will
count as part of what is communicated but not said. The technical
terms presupposition and entailment are used to describe two dif
ferent aspects of this kind of information.
It is worth noting at the outset that presupposition and entail
ment were considered to be much more central to pragmatics in
the past than they are now. In more recent approaches, there has
been less interest in the type of technical discussion associated
with the logical analysis of these phenomena. Wi thout some
introduction to that type of analytic discussion, however, it
becomes very diffcult to understand how the current relationship
between semantics and pragmatics developed. Much of what fol
lows in this chapter is designed to illustrate the process of think
ing through a number of problems in the analysis of some aspects
of invisible meaning. Let's begin by defning our terms.
A presupposition is something the speaker assumes to be the case
prior to making an utterance. Speakers, not sentences, have pre
suppositions. An entailment is something that logically follows
from
what is asserted in the utterance. Sentences, not speakers,
have entailments.
We can identify some of the potentially assumed information
that would be associated with the utterance of [I].
[ I] Mary's brother bought three horses.
In producing the utterance in [ r], the speaker will normally be
PRESUPPOSITION AND ENTAILMENT 25

cxpcctcd to havc thc prcsuppo

itions that a pcrson cal|cd Mary

cxists and that shc has a brothcr. 1hc spcakcr may a|so ho|d thc
morc spccihc prcsuppositions that Mary has on|y onc brothcr and
that hc has a lot ot moncy. A|l ot thcsc prcsuppositions arc thc
spcakcr's and a|l ot thcm can bc wrong, in tact. 1hc scntcncc in j t]
will bc trcatcd as having thc cntai|mcnts that Mary's brothcr
bought somcthing, bought thrcc anima| s, bought two horscs,
bought onc horsc, and many othcr similar |ogica| conscqucnccs.
1hcsc cntailmcnts to|low trom thc scntcncc, rcgard|css ot
whcthcr thc spcakcr's bc|icts arc right or wrong, in tact. 1hcy arc
communicatcd without bcing said. ccausc ot its |ogica| naturc,
howcvcr, cntai|mcnt is not gcncra|ly discusscd as m
_
ch in con-
tcmporary pragmatics as thc morc spcakcr-dcpcndcnt notion ot
prcsupposition.
Presupposition
ln many discussions ot thc conccpt, prcsupposition is trcatcd as a
rc|ationship bctwccn two propositions. lt wc say that thc scntcncc
in j za. ] contains thc proposition p and thc scntcncc in [ zb. ] con-
tains thc proposition q, thcn, using thc symbo| to mcan ' prc-
supposcs', wc can rcprcscnt thc rc|ationship as in j zc. ] .
j z] a. Mary's dog is cutc. (= p)
b. Mary has a dog. (= q)
c. p q
lntcrcsting|y, whcn wc producc thc oppositc ot thc scntcncc in
[za. ] by ncgating it (= NOT p) , as in [j a. ] , wc hnd that thc rc|a-
tionship ot prcsupposition docsn't changc. 1hat is, thc samc
proposition q, rcpcatcd as j j b. ], continucs to bc prcsupposcd by
NOT p, as shown in j jc. ] .
j j] a. Mary's dog isn't cutc.
b. Mary has a dog.
c. NOT p q
( = NOT p)
( = q)
1his propcrty ot prcsupposition i s gcncra||y dcscribcd as con
stancy under negation. asica||y, it mcans that thc prcsupposition
ot a statcmcnt wi|| rcmain constant i. c. still truc} cvcn whcn that
statcmcnt is ncgatcd. As 3 turthcr cxamp|c, considcr a situation in
z6 SURVEY
which you disagrcc via a ncgativc, as in [ {b. ] } with somconc who
has a|rcady madc thc statcmcnt in [{a. ] .
[{] a. Evcrybody knows that |ohn is gay.
b. Evcrybody docsn't know that |ohn is gay.
c. |ohn is gay.
d. p q &NOT p q
( = p)
( = NOT p)
( = q)
Moticc that, although both spcakcrs disagrcc about thc validity ot
p i . c. thc statcmcnt in [{a. ] } , thcy both assumc thc truth ot q i. c.
[{c.] } in making thcir statcmcnts. 1hc proposition q, as shown in
[{d. ] , is prcsupposcd by both p and NOT p, rcmaining constant
undcr ncgation.
Types of presupposition
ln thc ana|ysis ot how spcakcrs' assumptions arc typically cxprcsscd,
prcsupposition has bccn associatcd with thc usc ot a |argc numbcr ot
words, phrascs, and structurcs. Wc shal| considcr thcsc |inguistic
torms hcrc as indicators ot potential presuppositions, which can on|y
bccomc actua| prcsuppositions in contcxts with spcakcrs.
As alrcady i||ustratcd in cxamp|cs [ t] to [ j] , thc posscssivc con-
struction in English is associatcd with a prcsupposition ot cx-
istcncc. 1hc existential presupposition is not on|y assumcd to bc
prcscnt in posscssivc constructions tor cxamp|c, 'your car' 7
'you havc a car' } , but morc gcncra||y in any dchnitc noun phrasc.
y using any ot thc cxprcssions in [ y] , thc spcakcr is assumcd to
bc committcd to thc cxistcncc ot thc cntitics namcd.
[ y] thc King ot Swcdcn, thc cat, thc gir| ncxt door,
thc Counting Crows
Wc shal| rcconsidcr thc basis ot cxistcntial prcsuppositions
|atcr, but hrst wc shou|d notc that thcrc was a dittcrcnt typc ot
prcsupposition prcscnt in [{] . ln [{j , thc vcrb 'know' occurs in a
structurc, 'Evcrybody knows that q', with q as thc prcsupposi-
tion. 1hc prcsupposcd intormation to||owing a vcrb |ikc 'know'
an bc trcatcd as a tact, and is dcscribcd as a factive presupposition;
A numbcr ot othcr vcrbs, such as 'rca|izc' in [6a. ] and 'rcgrct' in
[ 6b. ] , as wc|l as phrascs invo|ving ' bc' with 'awarc' [ 6c. ] , 'odd'
[6d. ] , and 'g|ad' j 6c. ] havc tactivc prcsuppositions.
PRES UPPOS. I TI ON AND ENTAI LMENT z;
[6] a. Shc didn't rcalizc hc was ill. ( Hc was ill
b. Wc rcgrct tclling him. ( Wc to|d him
c. ! wasn't awarc that shc was
marricd. ( Shc was marricd
d. lt isn't odd that hc |ch carly. ( Hc lch carly
c. !'m glad that it's ovcr. ( lt's ovcr}
1hcrc arc also a numbcr ot othcr torms which may bcst bc
trcatcd as thc sourcc ot lcxical prcsuppositions. Gcncra|ly spcak-
ing, in lexical presupposition, thc usc ot onc torm with its asscrtcd
mcaning is convcntionally intcrprctcd with thc prcsupposition
that anothcr non-asscrtcd mcaning is undcrstood. Each timc
you say that somconc 'managcd' to do somcthing, thc asscrtcd
mcaning is that thc pcrson succccdcd in somc way. Whcn you say
that somconc ' didn't managc' , thc asscrtcd mcaning is that thc
pcrson did not succccd. ln both cascs, howcvcr, thcrc is a prcsup-
position non-asscrtcd that thc pcrson 'tricd' to do that somc-
thing. So, ' managcd' is convcntionally intcrprctcd as asserting
'succccdcd' and presupposing 'tricd' . Othcr cxamplcs, involving
thc lcxical itcms, ' stop' , ' start' , and 'again' , arc prcscntcd, with
thcir prcsuppositions, in j;] .
[;] a. Hc stoppcd smoking.
b. 1hcy startcd complaining.
( Hc uscd to smokc
( 1hcy wcrcn't
complaining bctorc}
c. You'rc latc again. ( You wcrc |atc bctorc
Hthc casc ot lcxical prcsupposition, thc spcakcr's usc ot a particu-
lar cxprcssion is takcn to prcsupposc anothcr unstatcd conccpt,
whcrcas in thc casc ot a tactivc prcsupposition, thc usc ot a par-
ticu| ar cxprcssion is takcn to prcsupposc thc truth ot thc in-
tormation that is statcd ahcr it.
ln addition to prcsuppositions which arc associatcd with thc
usc ot ccrtain words and phrascs, thcrc arc also structural presup
positions. ln this casc, ccrtain scntcncc structurcs havc bccn ana-
yzcd as convcntionally and rcgularly prcsupposing that part ot
thc structurc is alrcady assumcd to bc truc. Wc might say that
spcakcrs can usc such structurcs to trcat intormation as prcsup-
poscd i. c. assumcd to bc truc} and hcncc to bc acccptcd as truc by
thc listcncr. For cxamp|c, thc ub-qucstion constructi on in
English, as shown in j 8a. ] and [ 8b. ] , is convcntiona||y intcrprctcd
z8 S URVEY
with thc prcsupposition that thc intormation attcr thc ub-torm
i. c. 'Whcn' and 'Whcrc' } is a|rcady known to bc thc casc.
j 8] a. Whcn did hc |cavc! ( 7 Hc lctt}
b. Whcrc did you buy thc bikc! ( 7 You bought thc bikc}
Thc typc ot prcsupposition illustratcd in j 8] can lcad listcncrs to
bc|icvc that thc intormation prcscntcd is

ncccssari|y truc, rathcr

than | ust thc prcsupposition ot thc pcrson asking thc qucstion.
For cxamplc, |ct's say that you wcrc standing at an intcrscction
onc cvcning. You didn't noticc whcthcr thc trathc signa| had
turncd to rcd bctorc a car wcnt through thc intcrscction. 1hc car
was immcdiatcly involvcd in a crash. You wcrc witncss to thc
crash and latcr you arc askcd thc qucstion in j] .
j] How tast was thc car going whn i t ran thc rcd |ight!
It you answcr thc qucstion as askcd |ust answcr thc qucstion| }
and cstimatc thc spccd ot thc car, thcn you would appcar to bc
acccpting thc truth ot thc prcsupposition i. c. 7 thc car ran thc
rcd light} . Such structura||y-bascd prcsuppositions may rcprcscnt
subt|c ways ot making intormation that thc spcakcr bclicvcs
appcar to bc what thc |istcncr shou|d bclicvc.
So tar, wc havc on|y considcrcd contcxts in which prcsupposi-
tions arc assumcd to bc truc. 1hcrc arc, howcvcr, cxamplcs ot non-
tactivc prcsuppositions associatcd wth a numbcr ot vcrbs in
Eng|ish. A non-fcte presupposition is onc that is assumcd not to bc
truc. Vcrbs likc 'drcam' , ' imaginc' , and 'prctcnd' , as shown in j to] ,
arc uscd with thc prcsupposition that what tol|ows i s not truc.
[ to] a. I drcamcd that I was rich. ( 7 l was not rich}
b. Wc imagincd wc wcrc in Hawaii.
(
Wc wcrc not
in Hawaii }
c. Hc prctcnds to bc i|l. ( 7 Hc is not i|l}
Wc havc a|rcady notcd, at thc cnd ot thc discussion ot dcixis, a
structurc that is intcrprctcd with a non-tactivc prcsupposition 'lt
I had a yacht, . . . ' } . Indccd, this typc ot structurc crcatcs a counter
factual presupposition, mcaning that what is prcsupposcd is not
on|y not truc, but is thc oppositc ot what is truc, or 'contrary to
facts' . A conditiona| structurc ot thc typc shown in j t t] , gcncra||y
ca|lcd a countcrtactua| conditional, prcsupposcs that th intorma-
tion in thc i[-clausc is not truc at thc timc ot uttcrancc.
PRES UPP OS I TI ON AND ENTAI LMENT z
. ,

I ,

l

' l
._f
j t t] lt you wcrc my tricnd, you would havc hclpcd mc.
( You arc not my tricnd
1hc cxistcncc ot non-tactivc prcsuppositions is part ot an intcr-
csting problcm tor thc analysis ot uttcranccs with complcx struc-
turcs, gcncrally known as 'thc pro| cction problcm' , to bc
cxplorcd in thc ncxt scction.
lndicators ot potcntial prcsuppositions discusscd so tar arc
summarizcd in 1ablc q. 1 .
Type Example
existential the X
factive I regret leaving
non-factive He pretended to be happy
lexical He managed to escape
structural When did she die?
counterfactual If I weren't ill,
TAB LE q. 1 Potential presuppositions
The projection problem
Presu pposition
X exists
1 left
He wasn't happy
He tried to escape
She died
1 am ill
1hcrc is a basic cxpcctation that thc prcsupposition ot a simplc
scntcncc will continuc to bc truc whcn that simplc scntcncc
bccomcs part ot a morc complcx scntcncc. 1his is onc vcrsion ot
thc gcncral idca that thc mcaning ot thc wholc scntcncc is a com-
bination ot thc mcaning ot its parts. Howcvcr, thc mcaning ot
somc prcsuppositions as 'parts' docsn't survivc to bccomc thc
mcaning ot somc complcx scntcnccs as 'wholcs' . 1his is known
as thc projection problem. ln cxamplc j t z] , wc arc going to scc
what happcns to thc prcsupposition q 'Kclly was ill' which is
assumcd to bc truc in thc simplc structurc ot j t zc. ] , but which
docs not ' pro| cct' into thc complcx structurc j t zh. ] . H ordcr to
tollow this typc ot analysis, wc havc to think ot a situation in
which a pcrson might say. 'l imagincd that Kclly was ill and
nobody rcalizcd that shc was ill. '
j t z] a. obody rcalizcd that Kclly was ill.
b. Kclly was ill.
c. p q
( = p)
( = q)
jC S URVEY
(At this point, the speaker uttering [ 1 2a.]
presupposes [1 2b. ] . )
d. I imagined that Kelly was ill.
e. Kelly was not ill.
f. r NOT q
(At this point, the speaker utering [ 1 2d.]
presupposes [1 2e.], the opposite of [ 12b.] . )
g. I imagined that Kelly was ill and nobody
realized that she was ill.
h. r & p NOT q
( = r)
( = NOT q)
( = r &p)
(At this point, afer combining r & p, the presupposi
tion q can no longer be assUmed to be true. )
In an example like [ 12] , the technical analysis may be straight
forward, but it ay be diffcult to think of a context in which
someone would talk like that. Perhaps example [ 1 3] will contex
tualize better. In an episode of a TV soap opera, two characters
have the dialog in [ 13] .
[ 13] Shirley: It's so sad. George regrets getting Mary pregnant.
Jean: But he didn't get her pregnant. We know that
now.
If we combine two of the utterances from [ 1 3 ] , we have the
sequence, 'George regrets getting Mary pregnant; but he didn't
get her pregnant' . Identifing the diferent propositions involved,
as in [ 14], we can see that the presupposition q in [ 14b. ] does not
survive as a presupposition of the combined utterances in [ 14e. ] .
[ 14] a. George regrets gtting Mary pregnant. ( = p)
b. George got Mary pregnant. ( = q)
c. p q
d. He didn't get her pregnant. ( = r)
e. George regrets getting Mary pregnant,
but he didn't get her pregnant. ( p & r)
f. p & r NOT q
One way to think about the whole sentence presented in [ 14e.] is
as an utterance by a person reporting what happened in the
soap opera that day. That person will not assume the presupposi
tion q (Le. that George got Mary pregnant) is true when uttering
[
I4e. ] .
PRES U .. POS ITI ON AND ENTAI LMENT 3 1


'
"
' I

'
| |

l '

"

, '
. '
| I

A simple explanation for the fact that presuppositions don't
'proj ect' is- that they are destroyed by entailments. Remember that
an entailment is something that necessarily follows from what is
asserted. U example [ 1 3], Jean's uterance of 'he didn't get her
pregnant' actually entails 'George didn't get Mary pregnant' as a
logical consequence. Thus, when the person who watched the
soap opera tells you that 'George regrets getting Mary pregnant,
but he didn't get her pregnant', you have a presupposition g and
an entailment NOT g. The entailment (a necessary consequence
of what is said) is simply more powerfl than the presupposition
( an earlier assumption) .
The power of entailment can also be used to cancel existential
presuppositions. Normally we assume that when a person uses a
defnite description of the type 'the 7 (for example, 'the King of
England' ) , he or she presupposes the existence of the entity
described, as in the uterance of [ 15 a.]. Also, in any utterance of
the form `7doesn't exist', as in [I 5b.], there is an entaihnent that
there is no 7.But does the speaker of [ 1 5 b.] also still have the pre
suppositon of the existence of the entity described?
[ 1 5] a. The King, of England visited us.
b. The King of England doesn't exist!
Instead of thinking that a speaker who utters [ I 5 b.] simultan
eously believes that there is a King of England ( presupposition)
and that there is not a King of England ( entailment) , we recog
nize that the entailment is more powerful than the presupposi
tion. We abandon the existential presupposition.
As already emphasized, it may be best to t of all the types
of presuppositons illustated in Table 4. 1 as 'potental presupposi
tions' which only become actual presuppositons when intended
by speakers to be recognized as such within utterances. Speakers
can indeed indicate that the potential presupposition is not being
presented as a strong assumption. possessive constructions such
as 'his car' have a potential presupposition (i. e. he has a car)
which can be presented tentatively via expressions such as 'or
something' , as in [ 16].
[ 16] a. What's that guy doing in the parking lot?
h. He's looking for his car or something.
3 2 S URVEY
In j t 6b. ] , thc spcakcr is not committcd to thc prcsupposition hc
has a car} as an assumcd fact. lt is worth rcmcmbcring that it is
ncvcr thc word or phrasc that has a prcsupposition. On|y spcak-
crs can havc prcsuppositions.
Ordered entailments
Ccncra||y spcaking, cntai|mcnt is not a pragmatic conccpt i . c.
having to do with spcakcr mcaning}, but instcad is considcrcd a
purc|y |ogica| conccpt, symbo|izcd by 1 1-. Somc cxamp|cs ot
cntai|mcnt tor thc scntcncc in j t;] arc prcscntcd in j t &] .
j t ;] Rovcr chascd thrcc squirrc|s. p)
j t &] a. Somcthing chascd thrcc squirrc|s. ( q)
b. Rovcr did somcthing to thrcc squirrc|s. ( r)
c. Rovcr chascd thrcc of somcthing. ( s)
d. Somcthing happcncd. ( t)
In rcprcscnting thc rc|ationship of cntai|mcnt bctwccn j t;] and
jt &a. ] as p 1 1 - q, wc havc simp|y symbo|izcd a |ogica| con-
scqucncc. Lct us say that in uttcring thc scntcncc i n j t;] , thc
spcaker is ncccssari|y committcd to thc truth ot a vcry |argc num-
bcr of background entailments on|y somc ot which arc prcscntcd in
j t & a.-d. ] } . On any occasion ot uttcrancc j t ;] , howcvcr, thc
spcakcr wil| indicatc how thcsc cntailmcnts arc to bc ordcrcd.
That is, thc spcakcr wi|| communicatc, typica||y by strcss, which
cntai|mcnt is assumcd to bc in thc forcground, or morc important
for intcrprcting intcndcd mcaning, than any othcrs. For cxamp|c,
in uttcring j ta. ] , thc spcakcr indicatcs that thc foreground entail
ment, and hcncc hcr main assumption, is that Rovcr chascd a
ccrtain numbcr ot squirrc|s.
j t] a. Rovcr chascd THREE squirrc|s.
b. ROVER chascd thrcc squirrc|s.
In j tb. ] , thc tocus shitts to Rovcr, and thc main assumption is
that somcthing chascd thrcc squirrc|s. Onc function ot strcss in
Eng|ish is, in this approach, c|car|y

ticd to marking thc main

assumption ot thc spcakcr in producing an uttcrancc. As such, it
a||ows thc spcakcr to mark for thc |istcncr what thc focus of thc
mcssagc is, and what is bcing assumcd.
PRES UPPOS ITI ON AND ENTAI LMENT j j
A very similar function is exhibited by a structure called an 'it
cleft' construction in English, as shown in [20] .
[ 20] a. It was R
'
OVER that chased the squirrels.
h. It wasn't ME who took your money.
In both examples in [ 20] , the speaker , can communicate what
he or she believes the listener may already be thinking (i.e. the
foreground entailment) . In [ 20b. ] that foreground entailment
(someone took your money) is being made the shared knowledge
in order for the denial of personal responsibility to be made. The
utterance in [20h. ] can be used to attribute the foreground entail
ment to the listener( s} without actually stating it ( for example, as
a possible accusation) . It is one more example of more being com
municated than is said.
34 S URVEY
5
Cooperation and implicature
In much of the preceding discussion, we have assumed that
speakers and listeners involved in conversation are generally
cooperating with each other. For example, for reference to be suc
cessful, it was proposed that collaboration was a necessary factor.
In accepting speakers' presuppositions, listeners normally have to
assume that a speaker who says 'my car' really does have the car
that is mentioned and isn't trying to mislead the listener. This
sense of cooperation is simply one in which people having a con
versation are not normally assumed to be trying to confuse, trick,
or withhold relevant information from each other. In most cir
cumstances, this kind of cooperation is only the starting point for
making sense of what is said.
In the middle of their lunch hour, one woman asks another how
she likes the hamburger she is eating, and receives the answer
in [ I] .
[ I] A hamburger i s a hamburger.
From a purely logical perspective, the reply in [ I] seems to have
no communicative value since it expresses something completely
obvious. The example in [ I ] and other apparently pointless
expressions like 'business is business' or 'boys will be boys', are
called tautologies. If they are used in a conversation, clearly the
speaker intends to communicate more than is said.
When the listener hears the expression in [ I], she frst has to
assume that the speaker is being cooperative and intends to com
municate something. That something must be more than just
What the words mean. It is an additional conveyed meaning,
called an implicature. By stating [ I] , the speaker expects that the
COOPERATION AND IMPLICATURE 3 5
listccr wi|l bc ablc to work out, on thc basis ot what is a|rcady
known, thc imp|icaturc intcndcd in this contcxt.
Givcn thc opportunity to cvaluatc thc hamburgcr, thc spcakcr
ot j t ] has rcspondcd without an cvaluation, thus onc imp|icaturc
is that shc has no opinion, cithcr good or bad, to cxprcss.
Dcpcnding on othcr aspccts ot thc contcxt, additional imp|ic-
aturcs tor cxamp|c, thc spcakcr thinks all hamburgcrs arc thc
samc} might bc intcrrcd.
lmplicaturcs arc primary cxamp|cs ot morc bcing communicatcd
than is said, but in ordcr tor thcm to bc intcrprctcd, somc basic
coopcrativc princip|c must hrst bc assumcd to bc in opcration.
The cooperative principle
Considcr thc to||owing sccnario. 1hcrc is a woman sitting on a
park bcnch and a |argc dog |ying on thc ground in tront ot thc
bcnch. A man comcs along and sits down on thc bcnch.
j z] Man. Docs your dog bitc!
Woman. o.
1hc man rcachcs down to pct thc dog. 1hc dog
bitcs thc man's hand. }
Man. Ouch| Hcy| You said your dog docsn't bitc.
Woman. Hc docsn't. ut that's not my dog.
Onc ot thc prob|cms in this sccnario has to do with communica-
tion. Spccihcal|y, it sccms to bc a problcm causcd by thc man's
assumption that morc was communicatcd than was said. lt isn't a
prob|cm with prcsupposition bccausc thc assumption in 'your
dog' i. c. thc woman has a dog} is truc tor both spcakcrs. 1hc
prob|cm is thc man's assumpton that his qucstion 'Docs your dog
bitc!' and thc woman's answcr 'o' both app|y to thc dog in tront
ot thcm. From thc man's pcrspcctivc, thc woman's answcr pro-
vidcs |css intormation than cxpcctcd. ln othcr words, shc might
bc cxpcctcd to providc thc intormation statcd in thc last |inc. Ot
coursc, it shc had mcntioncd this intormation carlicr, thc story
wouldn't bc as tunny. For thc cvcnt to bc tunny, thc woman has to
givc lcss intormation than is cxpcctcd.
1hc conccpt ot thcrc bcing an cxpcctcd amount ot intormation
providcd in convcrsation is | ust onc aspcct ot thc morc gcncral
j 6 S URVEY
idca that pcoplc involvcd in a convcrsation will coopcratc with
cach othcr. Ot coursc, thc woman in [z] may actually bc indicat-
ing that shc docs not want to takc part in any coopcrativc intcrac-
tion with thc strangcr. ln most circumstanccs, thc assumption ot
coopcration is so pcrvasivc that it can bc statcd as a cooperative
principle ot convcrsation and claboratcd in tour sub-principlcs,
callcd maxims, as shown in 1ablc y . t .
The cooperative principle: Makc your convcrsational contribu-
tion such as is rcquircd, at thc stagc at which it occurs, by thc
acccptcd purposc or dircction ot thc talk cxchangc in which you
arc cngagcd.
Thc maxims
Quantity
t. Makc your contribution as intormativc as is rcquircd tor thc
currcnt purposcs ot thc cxchangcj .
z. Do not makc your contribution morc intormativc than is
rcquircd.
Quality 1y to makc your contribution onc that is truc.
t. Do not say what you bclicvc to bc talsc.
z. Do not say that tor which you lack adcquatc cvidcncc.
Relation c rclcvant.
Manner c pcrspicuous.
! Avoid obscurity ot cxprcssion.
z. Avoid ambiguity.
) . c brict avoid unncccssary prolixity} .
. c ordcrly.
TABLE y . 1 The cooperative principle (following Grice I97S)
lt is
important to rccognizc thcsc maxims as unstatcd assump-
tions wc havc in convcrsations. Wc assumc that pcoplc arc norm-
ally going to providc an appropriatc amount ot intormation
unlikc thc woman in [ z] , wc assumc that thcy arc tclling thc
truth, bcing rclcvant, and trying to bc as clcar as thcy can. ccausc
thcsc principlcs arc assumcd in normal intcraction, spcakcrs
rarcly mcntion thcm. Howcvcr, thcrc arc ccrtain kinds ot cxprcs-
sions spcakcrs usc to mark that thcy may bc in dangcr ot not tully
COOPERATI ON AND I MPLI CATURE j ;
,
I
! I
"
adhering to the principles. These kinds of expressions are called
hedges.
Hedges
The importance of the maxim of qualit for cooperative interac
tion in English may be best measured by the number of expres
sions we use to indicate that what we're saying may not be totally
accurate. The initial phrases in [3 a.-c. ] and the fnal phrase in
[3 d.] are notes to the listener regarding the accuracy of the main
statement.
[ 3] a. As far as I know, they're married.
b. I may be mistaken, but I thought I saw a wedding ring
on her fnger.
c. I'm not sure if this is right, but I heard it was a secret
ceremony in Hawaii.
d. He couldn't live without her, I guess.
The conversational context for the examples in [ 3] might be a
recent rumor involving a couple known to the speakers. Cautious
notes, or hedges, of this type can also be used to show that the
speaker is conscious of the quanti maxim, as in the initial phrases
in [4a.-. ] , produced in the course of a speaker's account of her
recent vacation.
[4] a. As you probably know, I am terrifed of bugs.
b. So, to cut a long story short, we grabbed our stuff and
ran.
c. I won't bore you with all the details, but it was an excit-

ing trip.
Markers tied to the expectation of relevance ( from the maxim of
relation) can be found in the middle of speakers' talk when they
say things like 'Oh, by the way' and go on to mention some poten
tially unconnected information during a conversation. Spe'akers
also seem to use expressions like 'anyway' , or 'well, anywa
'
y', to
indicate that they may have drifed into a discussion of some pos
sibly non-relevant material and want to stop. Some expressions
which may act as hedges on the expectation of relevance are
shown as the initial phrases in [ sa.-. ] , from an ofce meeting.
3 8 SURVEY
[ 5] a. I don't know if this is important, but some of the fles
are missing.
b. This may sound like a dumb question, but whose hand
writing is this ?
c. Not to change the subj ect, but is this related to the
budget?
The awareness of the expectations of manner may also lead
speakers to produce hedges of the type shown in the initial
phrases in [ 6a.-c. ] , heard during an account of a crash.
[6] a. This may be a bit confused, but I remember being in a
car.
b. I' m not sure if this makes sense, but the car had no
lights.
c. I don't know if this is clear at all, but I think the other
car was reversIng.
All of these examples of hedges are good indications that the
speakers are not only aware of the maxims, but that they want to
show that they are trying to observe them. Perhaps such forms
also communicate the speakers' concern that their listeners j udge
them to be cooperative conversational partners.
There are, however, some circumstances where speakers may
not follow the expectations of the cooperative principle. In court
rooms and classrooms, witnesses and students are ofen called
upon to tell people things which are already well-known to those
people (thereby violating the quantity maxim). Such specialized
institutional talk is clearly different from conversation.
However, even in conversation, a speaker may 'opt out' of the
maxim expectations by using expressions like 'No comment' or
'My lips are sealed' in response to a question. A interesting aspect
of such expressions is that, although they are typically not ' as
informative as is required' in the context, they are naturally inter
preted as communicating more than is said ( i.e. the speaker knows
the answer). This typical reaction ( i.e. there must be something
'special' here) of listeners to any apparent violation of the maxims
is actually the key to the notion of conversational implicature.
COOPERATI ON AND I MPLI CATURE 39
. 1
1
: i
, I i
. I
: '
Conversational implicature
The basic assumption in conversation is that, unless otherwise
indicated, the participants are adhering to the cooperative princi
ple and the maxims. In example [7] , Dexter may appear to be vio
lating the requirements of the quantity maxim.
[7] Charlene: I hope you brought the bread and the cheese.
. Dexter: Ah, I brought the bread.
Afer hearing Dexter's response in [7] , Charlene has to assume
that Dexter is cooperating and not totally unaware of the quan
tity maxim. But he didn't mention the cheese. If he had brought
the cheese, he would say so, because he would be adhering to the
quantity maxim. He must intend that she infer that what is not
mentioned was not brought. In this case, Dexter has conveyed
more than he said via a conversational implicature.
We can represent the structure of what was said, ith b
(= bread) and c ( = cheese) as in [ 8]. Using the symbol +> for an impli
cature, we can also represent the additional conveyed meaning.
[ 8] Charlene: b & c?
Dexter: b ( +> NOT c)
It i s important to note that i t i s speakers who communicate mean
ing via implicatures and it is listeners who recognize those com
municated meanings via inference. The inferences selected are
those which will preserve the assumption of cooperation.
Generalized conversational impl icatures
In the case of example [7] , particularly as represented in [ 8] , no
special background knowledge of the context of utterance is
required in order to make the necessary inferences. The same
process of calculating the implicature will take place if Doobie
asks Mary about inviting her friends Bella ( = b) and Cathy (= c) to
a party, as in [9a. ] , and gets the reply in [9b. ] . The context is dif
ferent from [7] , but the general process of identifying the implica
ture is the same as in [ 8].
[9] a. Doobie: Did you invite Bella and Cathy? ( b & c? )
b. Mary: I invited Bella. ( b +> NOT c)
40 S URVEY
Vhcn no spccia| know|cdgc is rcguircd in thc contcxt to ca|cu-
|atc thc additiona| convcycd mcaning, as in j;] to j] , it is ca||cd a
generalized conversational implicature. Onc common cxamp|c in
Eng|ish invo|vcs any phrasc with an indchnitc artic|c ot thc typc
'a/an 7,such as 'a gardcn' and 'a chi|d' as in j t o] . 1hcsc phrascs
arc typica||y intcrprctcd according to thc gcncra|izcd convcrsa-
tiona| mp|icaturc that. an 7~ not spcakcr's 7.
o] I was sitting in a gardcn onc day. A chi|d |ookcd ovcr thc
tcncc.
1hc imp|icaturcs in j to] , that thc gardcn and thc chi|d mcntioncd
arc not thc spcakcr's, arc ca|cu|atcd on thc princip|c that it thc
spcakcr was capab|c ot bcing morc spccihc i. c. morc intormativc,
to||owing thc quantity maxim} , thcn hc or shc wou|d havc said
'my gardcn' and 'my chi|d'.
A numbcr ot othcr gcncra|izcd convcrsationa| imp|icaturcs arc
common|y communicatcd on thc basis ot a sca|c ot va|ucs and arc
conscgucnt|y known as sca|ar imp|icaturcs.
Scalar implicatures
Ccrtain intormation is a|ways communicatcd by choosing a word
which cxprcsscs onc va|uc trom a sca|c ot va|ucs. 1his is particu-
|ar|y obvious in tcrms tor cxprcssing guantity, as shown in thc
sca|cs in [ t t] , whcrc tcrms arc |istcd trom thc highcst to

thc |ow-
cst va|uc.
j t t] a||, most, many, somc, tcw>
a|ways, ottcn, somctimcs>
Whcn producing an uttcrancc, a spcakcr sc|ccts thc word trom
thc sca|c which is thc most intormativc and truthtu| quantity and
gua|ity} in thc circumstanccs, as in j tz] .
j tz] I' m studying |inguistics and I'vc comp|ctcd somc ot thc
rcquircd courscs.
y choosing ' somc' in j t z] , thc spcakcr crcatcs an imp|icaturc
( "~ not a|| } . 1his is onc scalar imp|icaturc ot uttcring j z] . 1hc
basis of scalar |mp| |caIoto is that, whcn any torm in a sca|c is
asscrtcd, thc ncgativc ot a|| torms highcr on thc sca|c is imp|ic-
atcd. 1hc hrst sca|c in j t] had 'a||', 'most', and 'many' , highcr
COOPERATI ON AND I MPLI CATURE {t


I I
I
than ' somc' . Givcn thc dchnition ot sca|ar implicaturc, it should
tol|ow that, in saying 'somc ot thc rcquircd courscs', thc spcakcr
also crcatcs othcr implicaturcs tor cxamplc, +> not most, +> not
many} .
lt thc spcakcr gocs on to dcscribc thosc linguistics courscs as in
j t j ] , thcn wc can idcnti| somc morc scalar implicaturcs.

j t j ] 1hcy'rc somctimcs rca||y intcrcsting.
y using ' somctimcs' in j t j ] , thc spcakcr communicatcs, via
implicaturc, thc ncgativc ot torms highcr on thc scalc ot trcqucncy
+> not always, +> not ohcn} .
1hcrc arc many scalar implicaturcs produccd by thc usc ot
cxprcssions that wc may not immcdiatcly considcr to bc part ot
any scalc. For cxamplc, thc uttcrancc ot j tqa. ] will bc intcrprctcd
as implicating ' +> not ccrtain' as a highcr valuc on thc scalc ot
' |ikclihood' and j tqb. ] ' +> not must' on a scalc ot 'ob|igation' and
' + > not trozcn' on a scalc ot 'co |dncss' .
j tq] a. lt's possiblc that thcy wcrc dclaycd.
b. 1his should bc storcd in a coo| placc.
Onc noticcablc tcaturc ot scalar implicaturcs is that whcn
spcakcrs corrcct thcmsc|vcs on somc dctai|, as in j t y] , thcy typ-
ica|ly canccl onc ot thc scalar implicaturcs.
j t y ] l got somc ot this | cwclry in Hong Kongum actua|ly
l think l got most ot it thcrc.

ln j t y] , thc spcakcr initia||y implicatcs '+> not most' by saying

' somc', but thcn corrccts hcrsclt by actua|ly asscrting 'most' . 1hat
hna| asscrtion is sti|l |ikc|y to bc intcrprctcd, howcvcr, with a
scalar implicaturc +> not all } .
Paricularized conversational i mplicatures
ln thc prcccding cxamplcs, thc imp|icaturcs havc bccn ca|culatcd
without spccial knowlcdgc ot any particular contcxt. Howcvcr,
most ot thc timc, our convcrsations takc p|acc in vcry spccihc con-
tcxts in which locally rccognizcd intcrcnccs arc assumcd. Such
intcrcnccs arc rcquircd to work out thc convcycd mcanings which
rcsult trom paicularized conversational implicatures. As an i||ustra-
tion, considcr cxamplc j t 6] , whcrc 1om's rcsponsc docs not
qz S URVEY
appear on the surface to adhere to relevance. ( A simply relevant
answer would be 'Yes' or 'No' . )
[ 1 6] Rick: Hey, coming to the wild party tonight?
Tom: My parents are visiting.
In order to make Tom's response relevant, Rick has to draw on
some assumed knowledge that one college student in this setting
expects another to have. Tom will be spending that evening with
his parents, and time spent with parents is quiet (consequently +>
Tom not at party) .
Because they are by far the most common, particularized con
versational implicatures are typically just called implicatures. A
further example, in which the speaker appears not to adhere to
(i.e. to ' fout' ) the maxim of manner, is presented in [ 1 7] .
[ 17] Ann: Where are you going with the dog?
Sam: To the V- E-T.
Ir the local context of these speakers, the dog is known to re
cogize the word 'vet', and to hate being taken there, so Sam pro
duces a more elaborate, spelled out (i .e. less brief) version of his
message, implicating that he doesn't want the dog to know the
answer to the question j ust asked.
In [ 1 8] , Leila has just walked into Mary's offce and noticed all
the work on her desk. Mary'S response seems to fout the maxim
of relevance.
[ 1 8] Leila: Whoa! Has your boss gone crazy?
Mary: Let's go get some coffee.
In order to preserve the assumption of cooperation, Leila will
have to infer some local reason ( for example, the boss may be
nearby) why Mary makes an apparently non-relevant remark.
The implicature here is essentially that Mary cannot answer the
question in that context.
In addition to these fairly prosaic examples of implicatures,
there are other more entertaining examples, as in [ 19] and [20] ,
where the responses initially appear to fout relevance.
[ 19] Bert: Do you like ice-cream?
Ernie: Is the Pope Catholic?
COOPERATI ON AND I MPLI CATURE 43
I ,
[20] Bert: Do vegetarians eat hamburgers ?
Ernie: Do chickens have lips ?
In [ 19] , Ernie's response does not provide a 'yes' or ' no' answer.
Bert must assume that Ernie is being cooperative, so he considers
Ernie's 'Pope' question and clearly the answer is ' Yes' . So, the
answer is known, but the nature of Ernie's response also impli
cates that the answer to the question was ' Obviously, yes! ' . An
additional conveyed meaning in such a case is that, because the
answer was so obvious, the question did not need to be asked.
Example [ 20] provides the same type of inferencing with' an
. answer ' Of course not! ' as part of the implicature.
Properies of conversational i mpl icatures
So far, all the implicatures we have considered have been situated
within conversation, with the inferences being made by people
who hear the utterances and attempt to maintain the assumption
of cooperative interaction. Because these implicatures are part
of what is communicated and not said, speakers can always
deny that they intended to communicate such meanings .
Conversational implicatures are deniable. They can be explicitly
denied ( or alternatively, reinforced) in different ways. To take a
simple example, there is a standard implicature associated with
stating a number, that the speaker means only that number, as
shown in [21] .
[21 ] You have won fve dollars! ( +> ONLY fve)
As shown in [22], however, it is quite easy for a speaker to sus
pend the implicature ( +> only) using the expression ' at least'
[22a. ] , or to cancel the implicature by adding further informa
tion, often fol lowing the expression 'in fact' [ 22b. ] , or to reinforce
the implicature with additional information, as in [22C. ] .
[22] a. You've won at least fve dollars!
b. You've won fve dollars, in fact, you've won ten!
c. You've won fve dollars, that's four more than one!
We have already noted with many of the previous examples
that implicatures can be calculated by the listeners via inference.
In terms of their defning properties, then, conversational
44 S URVEY
imp|icaturcs can bc ca|cu|atcd, suspcndcd, canccl|cd, and rcin-
forccd. onc ot thcsc propcrtics app|y to convcntiona| imp|icat-
urcs.
Conventi onal impl icatures
In contrast to a|| thc convcrsationa| imp|icaturcs discusscd so far,
conventional i mplicatures arc not bascd on thc coopcrativc prin-
cip|c or thc maxims. 1hcy don't havc to occur in convcrsation,
and thcy don't dcpcnd on spccia| contcxts for thcir intcrprctation.
Mot un|ikc |cxica| prcsuppositions, convcntiona| imp|icaturcs arc
associatcd with spccihc words and rcsu|t in additiona| convcycd
mcanings whcn thosc words arc uscd. 1hc Eng|ish con| unction
' but' is onc ot thcsc words. 1hc intcrprctation of any uttcrancc
of thc typc p but q wi|| bc bascd on thc con| unction p & q p|us
an imp|icaturc ot ' contrast' bctwccn thc intormation in p and
thc intormation in q. In j zj ] , thc tact that 'Mary suggcstcd bIack'
( p) is contrastcd, via thc convcntiona| imp|icaturc ot ' but' , with
my choosing whitc ( q) .
jzj] a. Mary suggcstcd b|ack, but I chosc whitc.

b. P & q ( +> P is in contrast to q)
Othcr Eng|ish words such as 'cvcn' and 'yct' a|so havc convcn-
tiona| imp|icaturcs. Whcn ' cvcn' is inc|udcd in any scntcncc
dcscribing an cvcnt, thcrc is an imp|icaturc ot 'contrary to cx-
pcctation' . 1hus, in jz{] thcrc arc two cvcnts rcportcd i . c. |ohn's
coming and |ohn's hc|ping with thc convcntiona| imp|icaturc of
'cvcn' adding a 'contrary to cxpcctation' intcrprctation of thosc
cvcnts.
jz{] a. Evcn |ohn camc to thc party.
b. Hc cvcn hc|pcd tidy up ahcrwards.
1hc convcntiona| imp|icaturc of 'yct' is that thc prcscnt situa-
tion is cxpcctcd to bc diffcrcnt, or pcrhaps thc oppositc, at a |atcr
timc. In uttcring thc statcmcnt in j za. ] , thc spcakcr produccs an
imp|icaturc that shc cxpccts thc statcmcnt 'Dcnnis is hcrc' ( p) to
bc truc |atcr, as indicatcd in j zb. ] .
j z] a. Dcnnis isn't hcrc yct. ( NOT p)
b. NOT P i s truc ( +> p cxpcctcd to bc truc |atcr
COOPERATI ON AND I MPLI CATURE {

lt may bc possiblc to trcat thc so-callcd dittcrcnt ' mcanings' ot
' and' in English discusscd in Chaptcr t as instanccs ot convcn-
ti

nal implicaturc in dittcrcnt structurcs. Whcn two statcmcnts

containing staticintormation arc | oincd by 'and', as in jz6a. ] , thc
implicaturc is simply ' in addition' or 'plus' . Whcn thc two statc-
mcnts contain dynamic, action-rclatcd intormation, as in j z6b. ] ,
thc implicaturc ot 'and' is ' and thcn' indicating scqucncc.
j z6] a. Ycstcrday, Mary was happy
and rcady to work. (p & q, p plus q)
b. Shc put on hcr clothcs and lch
thc housc. (p & q, q afer p)
ccausc ot thc dittcrcnt implicaturcs, thc two parts ot jz6a. ] can
bc rcvcrscd with littlc dittcrcncc in mcaning, but thcrc is a big
changc in mcaning it thc two parts ot jz6b. ] arc rcvcrscd.
For many linguists, thc notion ot ' implicaturc' is onc ot thc
ccntral conccpts in pragmatics. An implicaturc is ccrtainly a
primc cxamplc ot morc bcing communicatcd than is said. For
thosc samc linguists, anothcr ccntral conccpt in pragmatics is thc
obscrvation that uttcranccs pcrtorm actions, gcncrally known as
' spccch acts' .
{6 S URVEY
6
Speech acts and events
In attempting to express themselves, people do not only produce
utterances containing grammatical structures and words , they
perform actions via those utterances. If you work in a situation
where a boss has a great deal of power, then the boss's utterance
of the expression in [r] is more than just a statement.
[ r] You're fred.
The utterance in [r] can be used to perform the act of ending your
employment. However, the actions performed by utterances do
not have to be as dramatic or as unpleasant as in [ r]. The action
can be quite pleasant, as in the compliment performed by [2a.] ,
the acknowledgement of thanks in [2b.] , or the expression of sur
prise in [2C.].
[2] a. You're so delicious.
b. You're welcome.
c. You're crazy!
Actions performed via utterances are generally called speech
acts and, in English, are commonly given more specifc labels,
such
as apology, complaint , compliment , invitation, promise, or
request.
These descriptive terms for different kinds of speech acts apply
to the speaker's communicative intention in producing an utter
ance. The speaker normally expects that his or her communica
tive
intention will be recognized by the hearer. Both speaker and
hearer are usually helped in this process by the circumstances
surrounding the utterance. These circumstances, including other
utterances, are called the speech event. In many ways, it is the
S PEECH ACTS AND EVENTS
4
7
i ,
"
nature of the speech event that determines the interpretation of an
utterance as performing a particular speech act. On a wintry day,
the speaker reaches for a cup of tea, believing that it has been
freshly made, takes a sip, and produces the utterance in [3 ] . It is
likely to be interpreted as a complaint.
[ 3] This tea is really cold!
Changing the circumstances to a really hot summer's day with the
speaker being given a glass of iced tea by the hearer, taking a sip
and producing the utterance in [3] , it is likely to be interpreted as
praise. If the same utterance can be interpreted as two different
kinds of speech act, then obviously no simple one utterance to one
action correspondence will be possible. It also means that there is
more to the interpretation of a speech act than can be found in the
utterance alone.
Speech acts
On any occasion, the action performed by producing an utterance
will consist of three related acts. There is frst a locutionar act,
which is the basic act of utterance, or producing a meaningful lin
guistic expression. If you have difcult with actually forming the
sounds and words to create a meaningful utterance in a language
( for example, because it's foreign or you're tongue-tied) , then you
might fail to produce a locutionary act. Producing 'Aha mokofa'
in English will not normally count as a locutionary act, whereas
[4] will.
[4] I've j ust made some coffee.
Mostly we don't j ust produce well-formed uterances with no
purpose. We form an utterance with some kind of function i n
mind. This i s the second dimension, or the illocutionar ac. The
illocutionary act is performed via the communicative force of
an utterance. We might utter [4] to make a statement, an ofer, an
explanation, or for some other communicative purpose. This is
also generally known as the illocutionary force of the utterance.
We do not, of course, simply cr
e
ate an uterance with a func-_
tion without intending it to have an effect. This is the third dimen
sion, the perlocutionary act. Depending on the circumstances, you
48 SURVEY
will utter [4] on the assumption that the hearer will recognize the
effect you intended ( for example, to account for a wonderful
smell, or to get the hearer to drink some coffee) . This is also gener
ally known as the perlocutionary efect.
Of these three dimensions, the most discussed is illocutionary
force. Indeed, the term 'speech act' is generally interpreted quite
narrowly to mean only the illocutionary force of an utterance.
The illocutionary force of an utterance is what it 'counts as' . The
same locutionary act, as shown in [ 5a. ] , can count as a prediction
[ 5b. ] , a promise [ sc. ] , or a warning [ S d. ] . These different analyses
[Sb.-d. ] of the utterance in [ sa. ] represent different illocutionary
forces.
[ 5] a. I'll see you later. ( = A)
b. [ I predict that] A.
c. [ I promise you that] A.
d. [I warn you that] A.
One problem with the examples in [ 5] is that the same utterance
can potentially have quite different illocutionary forces ( for ex
ample, promise versus warning) . How can speakers assume that
the intended illocutionary force will be recognized by the hearer?
That question has been addressed by considering two things:
Illocutionary Force Indicating Devices and felicity conditions.
I FI Os
The
most obvious device for indicating the illocutionary force
(the I llocutionary Force I ndicating Device, or I FID) is an expression of
the type shown in [ 6] where there is a slot for a verb that explicitly
names the illocutionary act being performed. Such a verb can be
called a performative verb (V p ) .
[6] I (Vp) you that . . .
In the preceding examples, [ 5C. ,d. ], 'promise' and 'warn' would
be the performative verbs and, if stated, would be very clear
I FI Ds. Speakers do not always 'perform' their speech acts so
explicitly, but they sometimes describe the speech act being per
formed. Imagine the telephone conversation in [7] , between a
man trying to contact Mary, and Mary's friend.
S PEECH ACTS AND EVENTS
4
9
II
'1 1
1
.
\
i: l .
1( 1
' j
' I
[7] Him: Can 1 talk to Mary?
Her: No, she's not here.
Him: I'm asking you-an 1 talk to her?
Her: And I'm telling you-SHE' S NOT HERE!
In this scenario, each speaker has described, and drawn attention
to, the illocutionary force ( 'ask' and 'tell' ) of their utterances.
Most of the time, however, there is no performative verb men
tioned. Other IFI Ds which can be identifed are word order,
stress, and intonation, as shown in the different versions of the
same basic elements (Y-G) in [8] .
[8] a. You're going! [I tell you Y -G]
b. You're going? [I request confrmation about Y -G]
c. Are you going? [I ask you if Y -G]
While other devices, such as a lowered voice quality for a warn
ing or a threat, might be used to indicate illocutionary force, the
utterance also has to be produced under certain conventional
conditions to count as having the intended illocutionary force.
Felicity conditions
There are certain expected or appropriate circumstances, tech
nically known as felicity conditions, for the performance of a
speech act to be recognized as intended. For some clear cases,
such as [
9
] , the performance will be infelicitous (inappropriate) if
the speaker is not a specifc person in a special context (in this
case, a j udge in a courtroom) .
[9] 1 sentence you to six months in prison.
In everyday contexts among ordinary people, there are also pre
conditions on speech acts. There are general conditions on the par
ticipants,. for example, that they can understand the language
being used and that they are not play-acting or being nonsensical.
Then there are content conditions. For example, for both a promise
and a warning, the content of the utterance must be about a
future event. A furher content condition for a promise requires
that the future event will be a fture act of the speaker.
The preparator condiions for a promise are signifcantly differ
ent from those for a warning. When 1 promise to do something,
5 0 SURVEY
there are two preparatory conditions: frst, the event will not hap
pen by itself, and second, the event will have a benefcial effect.
When I utter a warning, there are the following preparatory condi
tions: it isn't clear that the hearer knows the event will occur, the
speaker does think the event will occur, and the event will not have
a benefcial effect. Related to these conditions is the Sincerit condi
tion that, for a promise, the speaker genuinely intends to carry out
the future action, and, for a warning, the speaker genuinely
believes that the future event will not have a benefcial effect.
Finally, there is the essential condition, which covers the fact
that by the act of uttering a promise, I thereby intend to create an
obligation to carry out the action as promised. In other words, the
utterance changes my state from non-obligation to obligation.
Similarly, with a warning, under the essential condition, the utter
ance changes my state from non-informing of a bad future event
to informing. This essential condition thus combines with a
specifcation of what must be in the utterance content, the con
text, and the speaker's intentions, in order for a specifc speech act
to be appropriately ( felicitously) performed.
The performative hypothesis
One way to think about the speech acts being performed via utter
ances is to assume that underlying every utterance (U) there is a
clause, similar to [ 6] presented earlier, containing a performative
verb (V p) which makes the illocutionary force explicit. This is
known as the performative hypothesis and the basic format of the
underlying clause is shown in [ 10] .
[ 10] I ( hereby) Vp you (that) U
In
this cl ause, the subj ect must be frst person singular ( ' 1' ) ,
followed by the adverb 'hereby' , indicating that the utterance
'counts as' an action by being uttered. There is also a perform
ative verb ( Vp) in the present tense and an indirect obj ect in sec
ond person singular ( 'you' ) . This underlying clause will always
make explicit, as in [ l I b. ] and [ I 2b. ] , what, in utterances such as
[ I Ia. ]
and [ I 2a. ] , is implicit.
[ I I] a. Clean up this mess !
b. I hereby order you that you clean up this mess.
S PEECH ACTS AND EVENTS 5 I
|
I
j t z] a. 1hc work was donc by Elainc and mysclt.
b. l hcrcby tc|l you that thc work was donc by E|ainc and
mysc|t.
Examplcs likc j t t b. ] and j tzb. ] normally without 'hcrcby' } , arc
uscd by spcakcrs as explicit performatives. Examplcs likc j t t a. ]
and j t za. ] arc i mplicit performatives, somctimcs callcd primary
performatives.
1hc advantagc ot this typc ot analysis is that it makcs clcar | ust
what clcmcnts arc involvcd in thc production and intcrprctation
ot uttcranccs. ln syntax, a rchcxivc pronoun likc 'mysclt' in j t z] }
rcquircs thc occurrcncc ot an antcccdcnt in this casc 'l' } within
thc samc scntcncc structurc. 1hc cxp|icit pcrtormativc in j t zb. ]
providcs thc 'l' c|cmcnt. Similar|y, whcn you say to somconc, 'Do
it yoursclt| ' , thc rchcxivc in 'yoursc|t' is madc possiblc by thc
antcccdcnt 'you' in thc cxplicit vcrsion ' l ordcr you that you do it
yoursclt' } . Anothcr advantagc is to show that somc advcrbs such
as ' honcstly', or advcrbial clauscs such as ' bccausc l may bc latc',
as shown in j t j ] , naturally attach to thc cxplicit pcrtormativc
c|ausc rathcr than thc implicit vcrsion.
j t j ] a . Honcstly, hc's a scoundrc|.
b. What timc is it, bccausc l may bc latc!
ln j t j a. ] , it is thc tc||ing part thc pcrtormativc vcrb} that is bcing
donc 'honcstly' and, in j t jb. ] , it is thc act ot asking thc pcrtorm-
ativc again} that is bcing | ustihcd by thc ' bccausc l may bc latc'
c|ausc.
1hcrc arc somc tcchnical disadvantagcs to thc pcrformativc
hypothcsis. For cxamplc, uttcring thc cxplicit pcrtormativc vcr-
sion ot a command j t tb. ] has a much morc scrious impact than
uttcring thc implicit vcrsion j t t a. ] . 1hc two vcrsions arc con-
scqucntly not cquivalcnt. lt is also dimcult to know cxactly what
thc pcrtormativc vcrb or vcrbs} might bc tor somc uttcranccs.
Although thc spcakcr and hcarcr might rccognizc thc uttcrancc in
j tqa. ] as an insult, it would bc vcry strangc to havc j tqb. ] as an
cxp|icit vcrsion.
j tq] a. You'rc dumbcr than a rock.
b. ! l hcrcby insult you that you'rc dumbcr than a rock.
1hc rcally practical problcm with any analysis bascd on idcnti-
z S URVEY
[ing cxp|icit pcrtormativcs is that, in princip|c, wc simp|y do not
know how many pcrtormativc vcrbs thcrc arc in any |anguagc.
lnstcad ot trying to |ist a|| thc possib|c cxp|icit pcrtormativcs, and
thcn distinguish among a|| ot thcm, somc morc gcncra|
c|assihcations ot typcs ot spccch acts arc usua||y uscd.
Speech act classification
Onc gcncra| c|assihcation systcm |ists hvc typcs ot gcncra| tunc-
tions pcrtormcd by spccch acts. dcc|arations, rcprcscntativcs,
cxprcssivcs, dircctivcs, and commissivcs.
Declarati ons arc thosc kinds ot spccch acts that changc thc
wor|d via thcir uttcrancc. As thc cxamp|cs in [ y] i||ustratc, thc
spcakcr has to havc a spccia| institutiona| ro|c, in a spccihc con-
tcxt, in ordcr to pcrform a dcc|aration appropriatc|y.
[ t y] a. Pricst. l now pronouncc you husband and witc.
b. Rctcrcc. You'rc outl
c. |ury Forcman. Wc hnd thc dctcndant gui|ty.
In using a dcc|aration, thc spcakcr changcs thc wor|d via words.
Representatives arc thosc kinds ot spccch acts that statc what
thc spcakcr bc|icvcs to bc thc casc or not. Statcmcnts ot fact,
asscrtions, conc|usions, and dcscriptions, as i||ustratcd in [t 6] ,
arc a|| cxamp|cs ot thc spcakcr rcprcscnting thc wor|d as hc or shc
bc|icvcs it is.
[ 6] a. 1hc carth is hat.
b. Chomsky didn't writc about pcanuts.
c. lt was a warm sunny day.
ln
using a rcprcscntativc, thc spcakcr makcs words ht thc wor|d
ot bc|ictj .
Epressives arc thosc kinds ot spccch acts that statc what thc
spcakcr tcc|s. 1hcy cxprcss psycho|ogica| st

tcs and can bc statc-

mcnts ot p|casurc, pain, |ikcs, dis|ikcs, | oy, or sorrow. As i||us-
tr
atcd in j ;] , thcy can bc causcd by somcthing thc spcakcr docs
Othc hcarcr docs, but thcy arc about thc spcakcr's cxpcricncc.
[ t
;] a. l'm rca||y sorry|
b. Congratu|ations|
c. Oh, ycs, grcat, mmmm, ssahh|
S PEECH ACTS AND EVENTS

j
'

! ,
i

, /1

,
ln using an cxprcssivc, thc spcakcr makcs words ht thc world
ot tccling .
Directives arc thosc kinds ot spccch acts that spcakcrs usc to gct
somconc clsc to do somcthing. 1hcy cxprcss what thc spcakcr
wants. 1hcy arc commands, ordcrs, rcqucsts, suggcstions, and, as
illustratcd in jt 8] , thcy can bc positivc or ncgativc.
j t 8] a. Gimmc a cup ot cottcc. Makc it black.
b. Could you lcnd mc a pcn, plcasci
c. Don't touch that.
ln using a dircctivc, thc spcakcr attcmpts to makc thc world ht thc
words via thc hcarcr .
Com missives arc thosc kinds ot spccch acts that spcakcrs usc to
commit thcmsclvcs to somc tuturc action. 1hcy cxprcss what thc
spcakcr intcnds. 1hcy arc promiscs, thrcats, rctusals, plcdgcs,
and, as shown in j t ] , thcy can bc pcrtormcd by thc spcakcr
alonc, or by thc spcakcr as a mcmbcr ot a group.
j t ] a. l'll bc back.
b. l'm going to gct it right ncxt timc.
c. Wc will not do that.
ln using a commissivc, thc spcakcr undcrtakcs to makc thc world
ht thc words via thc spcakcr} .
1hcsc hvc gcncral tunctions ot spccch acts, with thcir kcy tca-
turcs, arc summarizcd in 1ablc 6. t .
Direct and indirect speech acts
A dihcrcnt approach to distinguishing typcs ot spccch acs can bc
madc on thc basis ot structurc. A tairly simplc structural distinction
bcwccn thrcc gcncral typcs ot spccch acts is providcd, in English,
by thc thrcc basic scntcncc typcs. As shown in jzo] , thcrc is an casily
rccognizcd rclationship bctwccn thc thrcc structural torms
dcclarativc, intcrrogativc, impcrativc and thc thrcc gcncral
communicativc tunctions statcmcnt, qucstion, commandrcqucst .
j zo] a. You wcar a scat bclt. dcclarativc
b. Do you wcar a scat bclt! intcrrogativc
c. Wcar a scat bclt| impcrativc
Whcncvcr thcrc is a dircct rclationship bctwccn a structurc and a
yq S URVEY
Spccch act typc Drccton of ht S spcakcr,
7 stuaton
Dcclaratons words changc thc world S causcs 7
Kcprcscntatvcs makc words ht thc world S bclcvcs 7
Exprcssvcs makc words ht thc world S tccls 7
Drcctvcs makc thc world ht words S wants 7
Commssvcs makc thc world ht words S ntcnds X
TABLE 6. 1 The fv,e general functions of speech acts (following
Searle :;)
functon, wc havc a direct speech act. Whcncvcr thcrc s an n-
drcct rclatonshp bctwccn a structurc and a functon, wc havc an
indirect speech act. 1hus, a dcclaratvc uscd to makc a statcmcnt s
a drcct spccch act, but a dcclaratvc uscd to makc a rcqucst s an
ndrcct spccch act. As llustratcd n [ zt] , thc uttcrancc n [ zta. ] s
a dcclaratvc. Whcn it s uscd to makc a statcmcnt, as paraphrascd
n [ ztb. ] , t s functonng as a drcct spccch act. Whcn t s uscd to
makc a commandrcqucst, as paraphrascd n [ztc. ] , t s functon-
ng as an ndrcct spccch act.

[ zt] a. lt's cold outsdc.
b. l hcrcby tcll you about thc wcathcr.
c. l hcrcby rcqucst ot you that you closc thc door.
Dffcrcnt structurcs can bc uscd to accomplsh thc samc basc
luncton, as n [ zz] , whcrc thc spcakcr wants thc addrcsscc not to
stand n lront of thc TV. 1hc basc functon of all thc uttcranccs n
[zz] s a command/rcqucst, but only thc mpcratvc structurc n
[zza. ] rcprcscnts a drcct spccch act. 1hc ntcrrogatvc structurc
n [ zzb. ] s not bcng uscd only as a qucston, hcncc t s an n-
drcct spccch act. 1hc dcclaratvc structurcs n [ zzc. ] and [ zzd. ]
arc also ndrcct rcqucsts.
[ zz] a. Movc out of thc wayl
b. Do you havc to stand n tront of thc T!
c. You'rc standng n front of thc T.
d. You'd makc a bcttcr door than a wndow.
Onc of thc most common typcs of ndrcct spccch act n
Lnglsh, as shown n [ zj] , has thc form of an ntcrrogatvc, but s
S PEECH ACTS AND EVENTS y y
'
I i ,
1
, : /' I .
not typically uscd to ask a qucstion i. c. wc don't cxpcct only an
answcr, wc cxpcct action . 1hc cxamplcs in j zj] arc normally
undcrstood as rcqucsts.
j zj] a. Could you pass thc salt!
b. Would you opcn this !
lndccd, thcrc is a typical pattcrn in English whcrcby asking a
qucstion about thc hcarcr's assumcd ability ' Can you!' , ' Could
you! ' or tuturc likclihood with rcgard to doing somcthing 'Will
you!' , 'Would you! ' normally counts as a rcqucst to actually do
that somcthing.
lndircct spccch acts arc gcncrally associatcd with grcatcr
politcncss in English than dircct spccch acts. ln ordcr to undcr-
stand why, wc havc to look at a biggcr picturc than | ust a singlc
uucrancc pcrtorming a singlc spccch act.
Speech events
Wc can trcat an indircct rcqucst tor cxamplc, thc uttcranccs in
jzj] as bcing a mattcr ot asking whcthcr thc ncccssary conditions
tor a rcqucst arc in placc. For cxamplc, a prcparatory condition is
that thc spcakcr assumcs thc hcarcr is ablc to, or CAN, pcrtorm thc
action. A contcnt condition conccrns mturc action, that thc hcarcr
WILL pcrtorm thc action. 1his pattcrn is illustratcd in j zq] .
jzq] lndircct rcqucsts
a. Contcnt Futurc act ot 'WILL you do X!'
condition
h. Prcparatory
condition
hcarcr
( hcarcr WILL
do X
Hcarcr i s ablc to 'CAN you do X!'
pcrtorm act
( hcarcr CAN
do X
c. Qucstioning a hcarcr-bascd condition tor making a
rcqucst rcsults in an indircct rcqucst.
1hcrc is a dchnitc dittcrcncc bctwccn asking somconc to do X and
asking somconc it thc prcconditions tor doing X arc in placc, as in
y 6 S URVEY
z{c. ] . Asking about prcconditions tcchnically docsn't count as
making a rcqucst, but docs allow thc hcarcr to rcact 'as t' thc
rcqucst had bccn madc. ccausc a rcqucst is an impositon by thc
spcakcr on thc hcarcr, it is bcttcr, n most socal circumstanccs, tor
thc spcakcr to avod a dircct imposition via a dircct rcqucst. Whcn
thc spcakcr asks about prcconditions, no drcct rcqucst s madc.
1hc prcccding discussion s csscntially about onc pcrson trying
to gct anothcr pcrson to do somcthing without risking rctusal or
causing otIcnsc. Howcvcr, this typc ot situation docs not consist
ot a singlc uttcrancc. It is a socal stuation nvolvng participants
who ncccssarily havc a socal rclationship ot somc kind, and who,
on a spccihc occasion, may havc particular goals.
Wc can look at thc sct ot uttcranccs produccd in ths kind ot situ-
ation as a speech event. A spccch cvcnt is an ctvty in whch par-
tcipants intcract via languagc in somc convcntional way to arrivc
at somc outcomc. It may includc an obvious ccntral spccch act,
such as 'I don't rcally lkc this' , as in a spccch cvcnt ot 'complain-
ng', but it will also includc othcr uucranccs lcadng up to and sub-
scqucntly rcacting to that ccntral acton. In most cascs, a 'rcqucst'
is not madc by mcans ot a singlc spccch act suddcnly uttcrcd.
Kcqucsting s typcally a spccch cvcnt, as llustratcd in [ zy] .
[zy] Him. Oh, Mary, I'm glad you'rc hcrc.
Hcr. What's up!
Him. I can't gct my computcr to work.
Hcr. Is t brokcn!
Him. I don't think so.
Hcr. What's it dong!
Hm. l don't know. I'm usclcss with computcrs.
Hcr. What knd is it!
Him. It's a Mac. Do you usc thcm!
Hcr. Ycah.
Him. Do you havc a minutc!
Hcr. Surc.
Him. Oh, grcat.
Thc cxtcndcd intcraction n [zy ] may bc callcd a ' rcqucsting'
spccch cvcnt wthout a ccntral spccch act of rcqucst. otcc that
thcrc is no actual rcqucst trom 'hm' to ' hcr' to do anything.
Vc might charactcrzc thc qucstion 'Do you havc a minutc! ' as a
S PEECH ACTS AND EVENTS y ;
' prc-rcqucst', allowing thc rcccivcr to say that shc's busy or that
shc has to bc somcwhcrc clsc. ln this contcxt, thc rcsponsc ' Surc'
is takcn to bc an acknowlcdgcmcnt not only ot having tmc avail-
ablc, but a willingncss to pcrtorm thc unstatcd action. 1hc ana-
lysis ot spccch cvcnts is clcarly anothcr way ot studying how morc
gcts communicatcd than is said.
1hc usctulncss ot spccch act analysis is in illustrating thc kinds
ot things wc can do with words and idcntifing somc ot thc con-
vcntional uttcrancc torms wc usc to pcrtorm spccihc actions.
Howcvcr, wc do nccd to look at morc cxtcndcd intcraction to
undcrstand how thosc actions arc carricd out and intcrprctcd
within spccch cvcnts.
5 8 S URVEY
7
Politeness and interaction
In much of the preceding discussion, the small-scale scenarios
presented to illustrate language in use have been populated by
people with virtually no social lives. Yet, much of what we say,
and a great deal of what we communicate, is determined by our
social relationships. A linguistic interaction is necessarily a social
interaction.
In order to make sense of what is said in an interaction, we have
to look at various factors which relate to social distance and
closeness. Some of these factors are established prior to an inter
action and hence are largely external factors. They typically
involve the relative status of the participants, based on social values
tied to such things as age and power. For example, speakers who
see themselves as lower status in English-speaking contexts tend
to mark social distance between themselves and higher status
speakers by using address forms that include a title and
a last name, but not the frst name (for example, Mrs Clinton,
Mr Adams, Dr Dang). We take part in a wide range of interac
tions (mostly with strangers) where the social distance deter
mined by external factors is dominant.
However, there are other factors, such as amount of imposition
or degree of friendliness, which are often negotiated during an
interaction. These are internal to the interaction and can result in
the initial social distance changing and being marked as less, or
more, during its course. This may result, for example, in partici
p
ants moving from a title-plus-last name to a frst-name basis
within the talk. These internal factors are typically more relevant
to participants whose social relationships are actually in the
process of being worked out within the interaction.
PO LITENES S AND INTERACTI ON
59
I
I , .
oth typcs ot tactors, cxtcrnal and intcrnal, havc an inhucncc
not only on what wc say, but also on how wc arc intcrprctcd. ln
many cascs, thc intcrprctation gocs bcyond what wc might havc
intcndcd to convcy and includcs cvaluations such as ' rudc' and
' inconsidcratc' , or ' considcratc' and 'thoughttul' . Rccognizing
thc impact ot such cvaluations makcs it vcry clcar that morc is
bcing communicatcd than is said. 1hc invcstigation ot that
impact is normally carricd out in tcrms ot politcncss.
Politeness
lt is possiblc to trcat politcncss as a hxcd conccpt, as in thc idca ot
' politc social bchavior', or ctiqucttc, within a culturc. lt is also
possiblc to spcci[ a numbcr ot dittcrcnt gcncral principlcs tor
bcing politc in social intcraction within a particular culmrc. Somc
ot thcsc might includc bcing tactml, gcncrous, modcst, and sym-
pathctic toward othcrs. Lct us assumc that participants in an
intcraction arc gcncrally awarc that such norms and principlcs
cxist in thc socicq at largc. Within an intcraction, howcvcr, thcrc
is a morc narrowly spccihcd typc ot politcncss at work. ln ordcr
to dcscribc it, wc nccd thc conccpt ot tacc.
As a tcchnical tcrm, fce mcans thc public sclt-imagc ot a pcr-
son. lt rctcrs to that cmotional and social scnsc ot sclt that cvcry-
onc has and cxpccts cvcryonc clsc to rccognizc. Politenes, in an
intcraction, can thcn bc dchncd as thc mcans cmploycd to show
awarcncss ot anothcr pcrson's tacc. ln this scnsc, politcncss can bc
accomplishcd in situations ot social distancc or closcncss.
Showing awarcncss tor anothcr pcrson's tacc whcn that othcr
sccms socially distant is ohcn dcscribcd in tcrms ot rcspcct or
dctcrcncc. Showing thc cquivalcnt awarcncss whcn thc othcr is
socially closc is ottcn dcscribcd in tcrms ot tricndlincss, cama-
radcric, or solidarity. 1hc hrst qpc might bc tound in a studcnt's
qucstion to his tcachcg shown as j za. ] , and a sccond typc in thc
tricnd's qucstion to thc samc individual, as in j t b. ] .
j t ] a. Excusc mc, Mr uckingham, but can l talk to you tor a
minutc!
b. Hcy, ucky, got a minutc!
lt tollows trom this typc ot approach that thcrc will bc dittcrcnt
6o S URVEY
L
kinds ot politcncss associatcd and markcd linguistically with thc
assumption ot rclativc social distancc or closcncss. ln most
English-spcaking contcxts, thc participants in an intcraction
ottcn havc to dctcrminc, as thcy spcak, thc rclativc social distancc
bctwccn thcm, and hcncc thcir 'tacc wants' .
Face wants
In this discussion, lct's assumc that thc participants involvcd in
intcractions arc not living in a contcxt which has crcatcd rigidly
hxcd social rclationships. Within thcir cvcryday social intcrac-
tions, pcoplc gcncrally bchavc as it thcir cxpcctations conccrning
thcir public sclt-imagc, or thcir face wants, will bc rcspcctcd. lt a
spcakcr says somcthing that rcprcscnts a thrcat to anothcr indi-
vidual's cxpcctations rcgarding sclt-imagc, it is dcscribcd as a face
threateni ng act. Altcrnativcly, givcn thc possibility that somc
action might bc intcrprctcd as a thrcat to anothcr's tacc, thc
spcakcr can say somcthing to lcsscn thc possiblc thrcat. 1his is
callcd a face saving act.
lmaginc a latc night sccnc, whcrc a young ncigbor is playing
his music vcry loud and an oldcr couplc arc trying to slccp. Onc ot
thcm, in j zj, proposcs a tacc thrcatcning act and thc othcr sug-
gcsts a tacc saving act.
j z] Him. l'm going to tcll him to stop that awtul noisc right
now|
Hcr. Pcrhaps you could | ust ask him it hc is going to stop
soon bccausc it's gctting a bit latc and pcoplc nccd to
gct to slccp.
ccausc it is gcncraly cxpcctcd that cach pcrson will attcmpt to
rcspcct thc tacc wants ot othcrs, thcrc arc many dittcrcnt ways ot
pcrtorming tacc saving acts.
Negative and positive face
Vhcn wc attcmpt to savc anothcr's tacc, wc can pay attcntion to
thcir ncgativc tacc wants or thcir positivc tacc wants. A pcrson's
negative face is thc nccd to bc indcpcndcnt, to havc trccdom of
action, and not to bc imposcd on by othcrs. 1hc word 'ncgativc'
POLI TENES S AND I NTERACTI ON 6t

'

|
here doesn't mean 'bad', it's j ust the opposite pole from 'positive'.
A person's positive fce is the need to be accepted, even liked, by
others, to be treated as a member of the same group, and to know
that his or her wants are shared by others. U simple terms, neg
ative face is the need to be independent and positive face is the
need to be connected.
So, a face saving act which is oriented to the person's negative
face will tend to show deference, emphasize the importance of the
other's time or concerns, and even include an apology for the
imposition or interruption. This is also called negative politeness.
A face saving act which is concerned with the person's positive
face will tend to show solidarity, emphasize that both speakers
want the same thing, and that they have a common goal. This is
also called positive politeness.
Self and other: say nothing
One way to see the relevance of the relationship between these
politeness concepts and language use is to take a single speech
event and map. out the diferent interpretations associated with
different possible expressions used within that event. For exam
ple, you arrive at an important lecture, pull out your notebook to
take notes, but discover that you don't have anything to write
with. You tink that the person sitting next to you may provide
the solution. In this scenario, you are going to be 'Self', and the
person next to you is going to be 'Other'.
Your frst choice is whether to say something or not. You can,
of course, rUage in your bag, search rather ob:iously through
your pockets, go back into your bag, without uttering a word, but
with the vague intention that your problem will be recognized.
This 'say nothing' approach may or may not work, but if it does,
it's because the other ofers and not because the self asks, as in [3] .
[3] Self: (looks in bag)
Other: ( ofers pen) Here, use this.
Many people seem to prefer to have their needs recognized by
others without having to express those needs in language. When
those needs are recognized, as in [3] , then clearly more has been
communicated than was said.
62 SURVEY
Say something: of and on record
Evcn it you dccidc to say somcthing, you don't actually havc to
ask tor anything. You can pcrhaps attcr your scarch through
your bag simply producc a statcmcnt ot thc typc in [qa. ] or [qb. ] .
jq] a. Uh, l torgot my pcn.
b. Hmm, l wondcr whcrc l put my pcn.
Thcsc, and othcr similar typcs ot statcmcnt, arc not dircctly
addrcsscd to thc othcr. 1hc othcr can act as it thc statcmcnts havc
not cvcn bccn hcard. 1hcy arc tcchnical|y dcscribcd as bcing of
record. ln casual dcscriptions, thcy might bc rctcrrcd to as 'hints' .
Oncc again, an ott rccord statcmcnt may or may not succccd as a
mcans ot gctting a pcn , but it it docs, it will bc bccausc morc has
bccn communicatcd than was said.
In contrast to such ott rccord statcmcnts, you can dircctly
addrcss thc othcr as a mcans ot cxprcssing your nccds. 1hcsc
dircct addrcss torms arc tcchnical|y dcscribcd as bcing on record.
Thc most dircct approach, using impcrativc torms such as thosc
in [ y] , is known as bald on record. 1hc othcr pcrson is dircctly
askcd tor somcthing.
[ y] a. Givc mc a pcn.
b. Lcnd mc your pcn.
Thcsc bald on rccord torms may bc tollowcd by cxprcssions likc
'plcasc' and 'would you!' which scrvc to sohcn thc dcmand and
arc callcd mitigating devices.
It is tcmpting to cquatc thc bald on rccord approach with all
dircct command torms i.c. impcrativcs . 1his would bc mislcading
bccausc impcrativc torms arc ohcn uscd by closc tamiliars without
bcing intcrprctcd as commands. Examplcs would bc a tricnd ot-
tcring somcthing to cat, as in [6a. ], or trying to hclp you, as in [6b.] .
[6] a. Havc somc morc cakc.
b. Gimmc that wct umbrclla.
Lmcrgcncy situations also occasion thc usc ot dircct commands,
rcgardlcss ot who is bcing addrcsscd, as whcn dangcr prompts usc
ot thc cxprcssions in [;] .
j;] a. Don't touch that|
h. Gct out ot hcrc|
POLITENES S AND I NTERACTI ON 6j
'

I '
- ,; 1
0 .
I
, , I
1hcrc arc, conscqucntly, somc social circumstanccs whcrc using a
dircct command as a bald on rccord cxprcssion is considcrcd
appropriatc among social cquals.
Howcvcr, gcncrally spcaking, bald on rccord cxprcssions arc
associatcd with spccch cvcnts whcrc thc spcakcr assumcs that hc
or shc has powcr ovcr thc othcr tor cxamplc, in military con-
tcxts and can control thc othcr's bchavior with words. ln cvcry-
day intcraction bctwccn social cquals, such bal d on rccord
bchavior would potcntially rcprcscnt a thrcat to thc othcr's tacc
and would gcncrally bc avoidcd. Avoiding a tacc thrcatcning act
is accomplishcd by tacc saving acts which usc positivc or ncgativc
politcncss stratcgics.
Positive and negative politeness
A positive politeness strateg lcads thc rcqucstcr to appcal to a
common goal, and cvcn tricndship, via cxprcssions such as thosc
in j 8] .
j 8] a. How about lctting mc usc your pcn!
b. Hcy, buddy, l' d apprcciatc it it you' d lct mc usc your
pcn.
1hcsc on rccord cxprcssions do rcprcscnt a grcatcr risk tor thc
spcakcr ot suttcring a rcmsal and may bc prcccdcd by somc 'gct-
ting to know you' talk, ot thc kind prcscntcd in j] , dcsigncd to
cstablish thc ncccssary common ground tor this stratcgy.
j] Hi. How's it going! Okay it l sit hcrc! Wc must bc intcr-
cstcd in thc samc crazy stutt. You takc a lot ot notcs too,
huh! Say, do mc a big tavor and lct mc usc onc ot your
pcns.
Howcvcr, in most English-spcaking contcxs, a tacc saving act is
morc commonly pcrtormcd via a negative polieness strateg. 1hc
most typical torm uscd is a qucstion containing a modal vcrb such
as j toa. ] .
j to] a. Could you lcnd mc a pcn!
b. l'm sorry to bothcr you, but can l ask you tor a pcn or
somcthing!
c. l know you'rc busy, but might l ask you itcmit you
4. SURVEY
'
I
happcn to havc an cxtra pcn that I could, you knowch-
maybc borrow!
Using this stratcgy also rcsults in torms which contain cxprcs-
sions ot apology tor thc imposition, ot thc typc shown in j ob. ] .
Morc claboratc ncgativc politcncss work can somctimcs bc hcard
in cxtcndcd talk, ottcn with hcsitations, similar to that shown in
j toc. ] .
It is worth noting that ncgativc politcncss is typically cxprcsscd
via qucstions, cvcn qucstions that sccm to ask tor pcrmission to
ask a qucstion tor cxamplc, 'Might I ask . . . !' as in j toc. ] . On thc
surtacc, such qucstions prcscnt an opportunity tor thc othcr to
answcr in thc ncgativc to thc qucstion without thc samc rctusal
cttcct ot rcsponding with a ncgativc to a dircct, bald on rccord
impcrativc. Jhis distinction is an important motivation tor thc
distinction bctwccn dircct and indircct spccch acts, discusscd
alrcady.
Evcn morc rclcvant tor our conccrn with thc pragmatics ot lan-
guagc in usc, thc availability ot thc bald on rccord torm, as wcll as
ott rccord torms, mcans that thc usc ot a tacc-saving on rccord
torm rcprcscnts a signihcant choicc. 1hc choicc ot a typc ot
cxprcssion that is lcss dircct, potcntially lcss clcar, gcncrally
longcr, and with a morc complcx structurc mcans that thc spcakcr
is making a grcatcr cttort, in tcrms ot conccrn tor tacc i. c. po-
litcncss , than is nccdcd simply to gct thc basic mcssagc across
cthcicntly.
1hcsc obscrvations arc summarizcd in Figurc ;. t ovcrlcat.
Strategies
Thc tcndcncy to usc positivc poitcncss torms, cmphasizing closc-
ncss bctwccn spcakcr and hcarcr, can bc sccn as a solidarit strateg.
This may bc thc principal opcrating stratcgy among a wholc group
or it may bc an option uscd by an individual spcakcr on a particu-
lar
occasion. Linguistically, such a stratcgy will includc pcrsonal
intormation, usc ot nicknamcs, somctimcs cvcn abusivc tcrms
p
articular|y among malcs , and sharcd dialcct or slang cxprcs-
sions. Frcqucntly, a solidarity stratcgy will bc markcd via inclusivc
tcrms such as 'wc' and 'lct's', as in thc party invitation in j t t] .
POLITENES S AND I NTERACTI ON 6
How to get a pen from someone else
I
say something say nothing
'
on record
face saving act
( but search in bag)
off record
[ 'forgot my pen' )
bald on record
('Give me a pen' )

positive politeness negative politenesss
( 'How about letting me use your pen?' ) (' Could you lend me a pen?' )
FI G URE ; . t How to get a pen from someone else (following
Brown and Levinson I9B7)
j t t ] Comc on, lct's go to thc party. Evcryonc will bc thcrc.
Wc'll havc mn.

1hc tcndcncy to usc ncgativc politcncss torms, cmphasizing thc
hcarcr's right to trccdom, can bc sccn as a deference strateg. lt
can bc thc typical stratcgy ot a wholc group or | ust an option uscd
on a particular occasion. A dctcrcncc stratcgy is involvcd in what
is callcd ' tormal politcncss' . lt is impcrsonal, as it nothing is
sharcd, and can includc cxprcssions that rctcr to ncithcr thc
spcakcr nor thc hcarcr tor cxamplc, ' Customcrs may not smokc
hcrc, sir' . 1hc languagc associatcd with a dctcrcncc stratcgy
cmphasizcs thc

spcakcr's and thc hcarcr's indcpcndcncc, markcd

via an abscncc ot pcrsonal claims, as in j t z] , an altcrnativc vcr-
sion ot thc party invitation in j t t ] .
j tz] 1hcrc's going to bc a paq, it you can makc it. lt will bc
tun.
1hcsc gcncral qpcs ot stratcgics arc illustratcd hcrc via uttcr-
anccs which arc actually ccntral to thc spccch cvcnt tor cxamplc,
invitation . Facc saving bchavior, howcvcr, is ohcn at work wcll
bctorc such uttcranccs arc produccd, in thc torm ot prc-scqucnccs.
66 S URVEY
Pre-sequences
As alrcady suggcstcd, thc conccpt ot tacc saving may bc hclptul in
undcrstanding how participants in an intcraction incvitably
undcrstand morc than is said. 1hc basic assumption, trom thc pcr-
spcctivc ot politcncss, is that tacc is typically at risk whcn thc sclt
nccds to accomplish somcthing involving othcr. 1hc grcatcst risk
appcars to bc whcn thc othcr is put in a dimcult position. Onc way
ot avoiding risk is to providc an opportunity tor thc othcr to halt
thc potcntially risky act. For cxamplc, rathcr than simply makc a
rcqucst, spcakcrs will ohcn hrst producc what can bc dcscribcd as
a pre-reques. Wc alrcady notcd onc cxamplc in discussing spccch
cvcnts carlicr, at thc cnd ot Chaptcr 6. Anothcr is prcscntcd as j t j] ,
along with onc analysis ot thc structurc ot this intcraction.
j t j ] Hcr. Arc you busy! = prc-rcqucst
Him. ot rcally. = go ahcad
Hcr. Chcck ovcr this mcmo. (= rcqucst
Him. Okay. = acccpt
1hc advantagc ot thc prc-rcqucst clcmcnt is that it can bc
answcrcd cithcr with a 'go-ahcad' rcsponsc, as in j t j ] , or with a
'stop' rcsponsc, as in j tq] .
j tq] Him. Arc you busy!
Hcr. Oh, sorry.
= prc-rcqucst
= stop
1hc rcsponsc in j tq] allows thc spcakcr to avoid making a rcqucst
that cannot bc grantcd at thc timc. Undcrstanding that it is a
rcsponsc to a prc-rcqucst also allows us to intcrprct thc cxprcs-
sion 'sorry' , not only as an apology about bcing busy, but also as
an apology about bcing unablc to rcspond to thc anticipatcd
rcqucst.
1hcrc is, howcvcr, a gcncral pattcrn ot prc-rcqucsts actually
bcing trcatcd as rcqucsts and bcing rcspondcd to, as in j t ], with
thc unstatcd, hopcd tor action bcing pcrtormcd.
j t y] Hcr. Do you havc a sparc pcn!
Him. Hcrc. hands ovcr a pcn
1
his
' short-cut' proccss ot going trom prc-rcqucst to granting ot
rcqucst hclps cxplain thc litcral oddncss ot thc common pattcrn in
j
6] .
POLITENES S AND I NTERACTI ON 6;

"

i
j t 6] Hcr. Do you mind it l usc your phonc!
Him. Ycah, surc.
As a litcral rcsponsc, 'Ycah' or 'Ycah, surc' would bc thc cqui-
valcnt ot 'l do mind' and wouldn't count as allowing usc ot thc
phonc. Howcvcr, thcsc torms arc normally intcrprctcd as a posi-
tivc rcsponsc, not to thc prc-rcqucst, but to thc unstatcd rcqucst.
Prc-scqucnccs arc also commonly uscd in making invitations.
As illustratcd in j t;] , with a 'go ahcad' , and [ t 8] , with a 'stop' ,
invitcrs tcnd to ask a pre-invitation qucstion and rcccivcrs tcnd to
rccognizc thcir tunction.
[ t ;] Him. What arc you doing this
Friday!
Hcr. Hmm, nothing so tar.
Him. Comc ovcr tor dinncr.
Hcr. Lh, l'd likc that.
j t 8] Him. Arc you doing anything
latcr!
Hcr. Oh, ycah. usy, busy, busy.
Him. Oh, okay.
= prc-in(ation
( go ahcad
= invitation
= acccpt
= prc-invitation
( stop
= stop
Childrcn ohcn usc pre-announcement to chcck it thcir parcnts arc
willing to pay attcntion, asin cxamplc j t] .
j t] Child. Mom, gucss what
happcncd! = prc-announccmcnt
Mothcr. Silcncc
Child. Mom, you know = prc-announccmcnt
what!
Mothcr. ot right now, |acy,
l`m busy. = stop
ln cxamplc [ t] , thcrc arc two prc-announccmcnts, ncithcr ot
which rcccivcs a 'go-ahcad' . 1hc initial prc-announccmcnt is mct
with silcncc, which is gcncrally intcrprctcd as a 'stop' . 1hc child's
sccond attcmpt must bc bascd on an intcrprctation that thc parcnt
did not hcar thc hrst attcmpt. 1h hnal rcsponsc has to bc intcr-
prctcd as a 'stop' , but noticcably it is cxprcsscd, in tacc-saving
tcrms, as a postponcmcnt.
1hroughout this discussion ot politcncss in intcraction, wc
havc bccn assuming a wcll-known and casily rccognizablc
68 S URVEY
structure for the interaction. That structure must now be
analyzed because it i s our comfortabl e familiarity with its
regularity that allows a great deal to be communicated that is
never said.
POLITENESS AND I NTERACTI ON 69
L
8
Conversation and preference
structure
The previous chapter focused on aspects of social awareness which
can have an impact on what gets communicated by what is said
during an interaction. The term 'interaction' could actually apply
to a very large number of quite different social encounters. For '
example, a teacher talking to students in a classroom is one kind of
interaction; others include a doctor talking to a patient in a clinic,
or individuals taking part in courtroom proceedings, attending a
committee meeting, buying stamps at the post offce, and dozens of
other different experiences people have in which there is interper
sonal exchange of talk. The kind of talk is likely to differ according
to the different contexts of interaction. However, the structure of
the talk, the basic pattern of 'I speak-you speak-I speak-you
speak', will derive from that fundamental kind of interaction we
acquire frst and use most ofen. This is the structure of conversa
tion. Conversation structure is what we have been assuming as
familiar throughout much of the preceding discussion. It is time to
look more closely at that structure as a crucial aspect of pragmatics.
Conversation analysis
There are many metaphors used to describe conversation struc
ture. For some, conversation is like a dance, with the conversa
tio
nal partners coordinating their movements smoothly. For
others it's like traffc crossing an intersection, involving lots of
alternating movement without any crashes. However, the most
Widely used analytic approach is based, not on dancing (there's no
music) nor on traffc flow (there are no traffc signals), but on an
analogy with the workings of a market economy.
CONVERSATI ON AND PREFERENCE STRUCTURE 71
ln this markct, thcrc is a scarcc commodity callcd thc foor
which can bc dchncd as thc right to spcak. Having control ot this
scarcc commodity at any timc is callcd a turn. ln any situation
whcrc control is not mcd in advancc, anyonc can attcmpt to gct
control . 1his is callcd turn-taking. ccausc it is a torm ot social
action, turn-taking opcratcs in accordancc with a local manage
ment system that is convcntionally known by mcmbcrs ot a social
group. 1hc local managcmcnt systcm is csscntially a sct ot con-
vcntions tor gctting turns, kccping thcm, or giving thcm away.
1his systcm is nccdcd most at thosc points whcrc thcrc is a poss-
iblc changc in who has thc turn. Any possiblc changc-ot-turn
point is callcd a Transition Relevance Place, or TRP. Within any
social group, thcrc will bc tcaturcs ot talk or abscncc ot talk typ-
ically associatcd with a TRP.
1his typc ot analytic mctaphor providcs us with a basic pcr-
spcctivc in which spcakcrs having a convcrsation arc vicwcd as
taking turns at holding thc hoor. 1hcy accomplish changc ot turn
smoothly bccausc thcy arc awarc ot thc loca| managcmcnt systcm
tor taking thosc turns at an appropriatc TRP. 1hc mctaphor can
bc applicd to thosc convcrsations whcrc spcakcrs coopcratc and
sharc thc hoor cqually. lt can also bc uscd to dcscribc thosc con-
vcrsations whcrc spcakcrs sccm to bc in compctition, hghting to
kccp thc hoor and prcvcnting othcrs trom gctting it. 1hcsc pat-
tcrns ot convcrsational intcraction dittcr substantial|y trom onc
social group to anothcr. ln ordcr to illustratc thc systcm at work,
wc will tocus on thc convcntions ot onc social groupmiddlc
class English spcakcrs in publicwhilc rcmaining awarc that
othcr social groups will havc substantially dittcrcnt assumptions
about thc mcaning ot various tcaturcs.
Pauses, overlaps, and backchannels
Most ot thc timc, convcrsation consists ot two, or morc, particip-
ants taking turns, and only onc participant spcaking at any timc.
Smooth transitions trom onc spcakcr to thc ncxt sccm to bc va|-
ucd. 1ransitions with a long silcncc bctwccn turns or with sub-
stantial overlap i. c. both spcakcrs trying to spcak at thc samc
timc arc tclt to bc awkward. Whcn two pcoplc attcmpt to havc a
1
convcrsation and discovcr that thcrc is no ' how' , or smooth
72 S URVEY
rhythm to their transitions, much more is being communicated
than is said. There is a sense of distance, an absence of familiarity
or ease, as in the interaction shown in [ I] between a student and
his friend's father during their frst meeting.
[ r] Mr. Strait: What's your maj or Dave?
Dave: English-well I haven't really decided yet.
( 3 seconds)
Mr. Strait: So-you want to be a teacher?
Dave: No-not really-well not if I can help it.
( 2. 5 seconds)
Mr. Strait: Wha-I Where do you- go ahead
Dave: I mean it's a-oh sorry / I em-
As shown in [ r ] , very short pauses ( marked with a dash) are
simply hesitati ons, but longer pauses become silences. The
silences in [ r] are not attributable to either speaker because each
has completed a turn. If one speaker actually turns over the foor
to another and the other does not speak, then the silence is attrib
uted to the second speaker and becomes signifcant. It's an atrib
utable si l ence. As shown in [ 2] , the non-response of Dave i s
treated, by his girlfriend, as possibly communicating something.
[2] Jan: Dave I'm going to the store.
( 2 seconds)
Jan: Dave?
( 2 seconds)
Jan: Dave-is something wrong?
Dave: What ? What's wrong?
Jan: Never mind.
Silence at a T is not as problematic for the local management
system as overlap. I the expectation is that only one person speaks at
a time, then overlap can be a serious problem. Returing to example
[I], the fnal two lines illustrate overlaps, conventionally marked by a
double slash (/) at the beginning of the overlapping tal. Typically,
the
fst overlap occurs as both speakers attempt to initiate talk. I
accordance with the local management system, one speaker will stop
to allow the other to have the foor. However, for two speakers who
are having difculty geting into a shared conversational rhythm, the
stop-start-overlap-stop patter may be repeated.
CONVERSATI ON AND PREFERENCE S TRUCTURE 73
1hc typc ot ovcrlap shown n j t] s smply part ot a dmcult mst
convcrsaton wth an untamlar pcrson. 1hcrc arc othcr knds ot
ovcrlap and thcy arc ntcrprctcd dttcrcntly. For many ottcn
youngcr spcakcrs, ovcrlappcd talk appcars to mncton likc an
cxprcsson ot soldarty or closcncss n cxprcssng smlar opn-
ons or valucs. As shown n j j] , thc cttcct ot thc ovcrlappng talk
crcatcs a tcclng ot two voccs collaboratng as onc, n harmony.
j j ] Mn. Dd you scc hm n thc vdco!
Wcndy. Ycah-thc part on thc bcach
Mn. Oh my god hc was so scxy
Wcndy. hc was | ust bcng so cool
Mn. And all thc wavcs crashng around hm|
Wcndy. ycah that was rcally wld|
H cxamplc j j ] , ovcrlap communcatcs closcncss. ln cxamplc jq] ,
ovcrlap communcatcs compctton.
jq] |oc. whcn thcy wcrc n
hpowcr las- wat CAN I FINISH?
|crry. that's my pont l sad-
ln cxamplc jq] , thc spcakcrs may appcar to bc havng a dscus-
son, but thcy arc, n tact, compctng tor thc hoor. 1hc pont at
whch ovcrlap occurs s trcatcd as an ntcrrupton and thc hrst
spcakcr actually has to makc a commcnt about proccdurc wth a
loudcr vocc, shown by thc captal lcttcrs n ' CAN I FINISH? ' )
rathcr than about thc topc ot convcrsaton.
y drawng attcnton to an cxpcctaton that hc should bc
allowcd to hnsh, thc hrst spcakcr n j q] s appcalng to somc ot
thc unstatcd ' rulcs' ot convcrsaton structurc. Each potcntal
spcakcr is cxpcctcd to wat untl thc currcnt spcakcr rcachcs a
TRP. 1hc most obvous markcrs ot a TRP arc thc cnd ot a struc-
tural unt a phrasc or clausc and a pausc. otcc that, n jq] , thc
hrst spcakcr has uucrcd 'whcn thcy wcrc n-' at thc pont whcrc
thc sccond spcakcr bcgns to talk. 1hcrc s no pausc and t s not
thc cnd ot a phrasc or clausc. 1hs s a clcar ntcrrupton and
brcaks thc ' rulcs' .
ormally, thosc who wsh to gct thc hoor wll wat tor a poss-
iblc TRP bctorc | umpng n. Ot coursc, thosc holdng thc hoor n a
compctitvc cnvronmcnt wll avod provdng TRPs. 1o do so,
q b1BNEY
thcy must avoid an opcn pausc at thc cnd ot a syntactic unit. As
il|ustratcd in j y] , thc spcakcr hlls cach ot his pauscs ' um' or ' uh' ,
which arc placcd insidc, not at thc cnd ot, syntactic units. |ust
prior to this turn, anothcr spcakcr had attcmptcd to takc thc hoor,
so thc spcakcr in j y] sccms conccrncd to protcct his turn.
j y] l wasn't talking about-um his hrst book that was-uh
rca|ly | ust likc a start and souh isn'tdocsn't count
rca|ly.
Anothcr typc ot hoor-ho|ding dcvicc is to indicatc that thcrc is a
largcr structurc to your turn by bcginning with cxprcssions ot thc
typc shown in j6] .
j 6] a. 1hcrc arc thrcc points l'd |ikc to makc-hrst . . .
h. 1hcrc's morc than onc way to do this-onc cxamplc
wou|d bc . . .
c. Didn't you know about Mclvin!-oh it was last
Octobcr . . .
d. Did you hcar about Cindy's ncw car !-shc got it in . . .
1hc cxprcssions in j6a. ] and j6b. ] arc associatcd with discus-
sions ot tacts or opinions whcrcas thosc in j 6c. ] and j 6d. ] arc prc-
|udcs to storytclling. ln a|| cascs, thcy arc uscd to gct thc rcgular
cxchangc ot turn proccss suspcndcd and a|low onc spcakcr to
havc an cxtcndcd turn. Within an cxtcndcd turn, howcvcr, spcakcrs
sti|l cxpcct thcir convcrsationa| partncrs to indicatc that thcy arc
|istcning. 1hcrc arc many dittcrcnt ways ot doing this, inc|uding
hcad nods, smilcs, and othcr tacia| cxprcssions and gcsturcs, but
thc most common vocal indications arc cal|cd backchannel signals,
or simp|y backchannels. Somc ot thcsc arc prcscnt in Mary's con-
tributions to j;] .
[;] Callcr. it you usc your |ong distancc scrvicc a |ot thcn you'll
Mary. uh-uh
Callcr. bc intcrcstcd in thc discount l'm talking about bccausc
Mary. ycah
Callcr. it Lon|y savc you moncy to switch to a chcapcr scricc
Mary. mmm
Thcsc typcs ot signa|s 'uh-uh', 'ycah',

) providc tccdback to
thc currcnt spcakcr that thc mcssagc is bcing rcccivcd. 1hcy nor-
ma|ly indicatc that thc listcncr is to||owing, and not ob|ccting to,
CONVERSATI ON AND PREFERENCE STRUCTURE ;y
what thc spcakcr is saying. Givcn this normal cxpcctation, thc
abscncc ot backchanncls is typically intcrprctcd as signihcant.
During tclcphonc convcrsations, thc abscncc ot backchanncls may
prompt thc spcakcr to ask it thc listcncr is still thcrc. During tacc-to-
tacc intcraction, thc abscncc ot backchanncls may bc intcrprctcd as
a way ot withholding agrccmcnt, lcading to an intcrcncc ot dis-
agrccmcnt. H convcrsation, silcncc is signihcant and will bc intcr-
prctcd as mcaningtul.
Conversational style
Many ot thc tcaturcs which charactcrizc thc turn-taking systcm ot
convcrsation arc invcstcd with mcaning by thcir uscrs. Evcn within
a broadly dchncd community ot spcakcrs, thcrc is ohcn sumcicnt
variation to causc potcntial misundcrstanding. For cxamplc, somc
individuals cxpcct that participation in a convcrsation will bc vcry
activc, that spcaking ratc will bc rclatlvcly tast, with almost no
pausing bctwccn turns, and with somc ovcrlap or cvcn complction
ot thc othcr's turn. 1his is onc conversational style. lt has bccn
callcd a high involvement stle. lt dittcrs substantially trom anothcr
stylc in which spcakcrs usc a slowcr ratc, cxpcct Iongcr pauscs
bctwccn turns, do not ovcrlap, and avoid intcrruption or complc
tion ot thc othcr's turn. 1his non-intcrrupting, non-imposing stylc
has bccn callcd a high considerateness sle.
Whcn a spcakcr who typically uscs thc hrst stylc gcts into a con-
vcrsation with a spcakcr who normally uscs thc sccond stylc, thc
talk tcnds to bccomc onc-sidcd. 1hc activc participation stylc will
tcnd to ovcrwhclm thc othcr stylc. cithcr spcakcr will ncccssarily
rccognizc that it is thc convcrsational stylcs that arc slightly dihcr-
cnt. lnstcad, thc morc rapid-hrc spcakcr may think thc slowcr-
paccd spcakcr | ust docsn't havc much to say, is shy, and pcrhaps
boring or cvcn stupid. ln rcturn, hc or shc is likcly to bc vicwcd as
noisy, pushy, dominccring, sclhsh, and cvcn tircsomc. Fcaturcs ot
convcrsational stylc will ohcn bc intcrprctcd as pcrsonality traits.
Adjacency pairs
Dcspitc dittcrcnccs in stylc, most spcakcrs sccm to hnd a way to
copc with thc cvcryday busincss ot social intcraction. 1hcy arc
;6 S URVEY
L
ccrtain|y hc|pcd in this proccss by thc tact that thcrc arc many
almost automatic pattcrns in thc structurc ot convcrsation. Somc
c|car cxamplcs arc thc grcctings and goodbycs shown in [ 8] to
[ o] .
[ 8] Anna. Hcl|o.
j] Anna. How arc you!
j o] Anna. Scc ya|
i|l. Hi.
i||. Finc.
i|l . yc.
1hcsc a utoma tic scqucnccs arc cal|cd adjacency pairs. 1hcy
always consist ot a first par and a second par, produccd by dittcr-
cnt spcakcrs. 1hc uttcrancc ot a hrst part immcdiatcly crcatcs an
cxpcctation ot thc uttcrancc ot a sccond part ot thc samc pair.
Fai|urc to producc thc sccond part in rcsponsc wi|| bc trcatcd as a
signihcant abscncc and hcncc mcaningtul. 1hcrc is substantia|
variation in thc torms which arc uscd to h|l thc s|ots in ad| accncy
pairs, as shown in j t t ] , but thcrc must a|ways bc two parts.
[ t ] First Part Sccond Part

A. What's up! . othin' much.
A. How's it goin' ! . |us' hangin' in thcrc.
A. How arc things ! . 1hc usua|.
A. How ya doin' ! . Can't comp|ain.
Thc cxamplcs in j t t ] arc typica||y tound in thc opcning scqucnccs
ot a convcrsation. Othcr typcs ot ad| accncy pairs arc i|lustratcd in
[ tz] , inc|uding a qucstion-answcr scqucncc [ t za. ] , a thanking-
rcsponsc j t zb. ] , and a rcqucst-acccpt [ tzc. ] .
[ z] First Part Sccond Part
a. A. What timc is it! . About cight-thirty.
b. A. 1hanks. . You'rc wc|comc.
c. A. Could you hc|p
mc with this! . Surc.
Mot all hrst parts immcdiatcly rcccivc thcir sccond parts, how-
cvcr. It ottcn happcns that a qucstion-answcr scqucncc wi|| bc
dclaycd whilc anothcr qucstion-answcr scqucncc intcrvcncs. 1hc
sc
qucncc will thcn takc thc torm ot Q-Qz-Az-At , with thc
midd|c pair Qz-Az} bcing cal|cd an inserion sequence. A|though
thcrc appcars to bc a qucstion Qz} in rcsponsc to a qucstion
| Q } , thc assumption is that oncc thc sccond part Az} ot thc
CONVERSATI ON AND PREFERENCE STRUCTURE ;;
|
I'

I
I' i ' |
'1 " , ', 1
' !
|
ll
t
inscrtion scqucncc is providcd, thc sc

ond part At ot thc initial

qucstion Qt will tollow. 1his pattcrn is illustratcd in j t j ] .
j t j ] Agcnt. Do you want thc carly hight! ( Qt
Clicnt. What timc docs it arrivc! ( Qz
Agcnt. inc torty-hvc. ( Az
Clicnt. Ycah-that's grcat. ( At
An inscrtion scqucncc is onc ad| accncy pair within anothcr.
Although thc cxprcssions uscd may bc qucstion-answcr
scqucnccs, othcr torms ot social action arc also accomplishcd
within this pattcrn. As shown in j tq] , thcrc is a pair which con-
sists ot making a rcqucstacccpting thc rcqucst Qt-At , with
an inscrtion scqucncc ot a qucstion-answcr pair Qz-Az which
sccms to tunction as a condition on thc acccptancc At bcing
providcd.
j tq] |can. Could you mail this lcttcr Qt Rcqucst
tor mc!
Frcd. Docs it havc a stamp on it!
|can. Ycah.
Frcd. Okay.
Qz
Az
At Acccptancc
1hc dclay in acccptancc in cxamplc j tq] , crcatcd by thc inscrtion
scqucncc, is onc typc ot indication that not al l hrst parts nc-
ccssarily rcccivc thc kind ot sccond parts thc spcakcr might
anticipatc. Dclay in rcsponsc symbolically marks potcntial
unavaila
[
ility ot thc i mmcdiatc i . c. normally automatic
cxpcctcd answcr. Dclay rcprcscnts distancc bctwccn what is
cxpcctcd and what is providcd. Dclay is always intcrprctcd as
mcaningml. ln ordcr to scc how dclay is locally intcrprctcd, wc
nccd somc analytic tcrms tor what is cxpcctcd within ccrtain
typcs ot ad| accncy pairs.
Preference structure
Ad| accncy pairs arc not simply contcntlcss noiscs in scqucncc.
1hcy rcprcscnt social actions, and not all social actions arc cqual
whcn thcy occur as sccond parts ot somc pairs. asically, a hrst
part that contains a rcqucst or an ottcr is typically madc in thc
cpcctation that thc sccond part will bc an acccptancc. An acccpt-
;8 S URVEY
ancc is structurally morc likcly than a rctusal. 1his structural likc-
lihood is callcd preference. 1hc tcrm is uscd to indicatc a socially
dctcrmincd structural pattcrn and docs not rctcr to any ndivid-
ual's mcntal or cmotional dcsircs. !n this tcchnical usc
ot thc word, prctcrcncc s an obscrvcd pattcrn in talk and not a
pcrsonal wish.
Preference structure dividcs sccond parts into preferred and dis
preferred socal acts. 1hc prctcrrcd is thc structurally cxpcctcd
ncxt act and thc disprctcrrcd is thc structurally uncxpcctcd ncxt
act. 1hc gcncral pattcrns arc prcscntcd in 1ablc 8 . t .
First part
Asscssmcnt
Invitation
Ottcr
Proposal
Kcqucst
Sccond part
Prctcrrcd
agrcc
acccpt
acccpt
agrcc
acccpt
Disprctcrrcd
disagrcc
rctusc
dcclinc
disagrcc
rctusc
TABLE 8 . t The general patterns of preferred and dispreferred
structures (following Levinson 1983)
In consdcring rcqucsts or ottcrs as hrst parts, acccptancc s thc
prctcrrcd and rctusal is thc disprctcrrcd sccond part. !n cxamplcs
j y a.-d. ] , thc rcsponscs in cach sccond part all rcprcscnt prc-
tcrrcds. 1hus, acccptancc or agrccmcnt is thc prctcrrcd sccond
part rcsponsc to rcqucst j t a. ], an ottcr j t y b. ], an asscssmcnt
j yc. ], or a proposal j t y d. ] .
j t y] First Part
a. Can you hclp mc!
b. Want somc cottcc!
c. !sn't that rcally grcat!
d. Maybc wc could go tor a walk.
Sccond Part
Surc.
Ycs, plcasc.
Ycs, it is.
1hat'd bc grcat.
To gct a scnsc ot how cxpcctcd thcsc prctcrrcd sccond parts
arc in thc cxamplcs in j t y] , imaginc cach ot thc hrst parts bcing
mct with siIcncc. Wc might say that in any ad| accncy par, silcncc
in thc sccond part is always an indication ot a disprctcrrcd
rcsponsc.
CONVERSATI ON AND PREFERENCE STRUCTURE ;
lndccd, silcncc ottcn lcads thc hrst spcakcr to rcvisc thc hrst
part in ordcr to gct a sccond part that is not silcncc trom thc othcr
spcakcr. 1his may bc clcarcr via an cxamplc, such as j t 6] , whcrc
|ack's silcncc in rcsponsc to Sandy's commcnt prompts Sandy to
rcstatc hcr asscssmcnt. |ack thcn agrccs a prctcrrcd with Sandy's
asscssmcnt.
j t 6] Sandy.
Sandy.
|ack.
ut l'm surc thcy'll havc good tood thcrc.
t. 6 scconds
Hmml gucss thc tood isn't grcat.
ahpcoplc mostly go tor thc music.
oticc that |ack's silcncc occurs whcrc hc would havc had to pro-
ducc a disagrccmcnt i. c. a disprctcrrcd rcsponsc rcgarding
Sandy's asscssmcnt. on-rcsponsc communicatcs that thc
spcakcr is not in a position to providc thc prctcrrcd rcsponsc.
Howcvcr, silcncc as a rcsponsc is an cxtrcmc casc, almost risk-
ing thc imprcssion ot non-participation in thc convcrsational
structurc. Gcncrally spcaking, whcn participants havc to producc
sccond part rcsponscs that arc disprctcrrcd, thcy indicatc that
thcy arc doing somcthing vcry markcd.
ln cxamplc j t;] , thc hrst spcakcr has madc a statcmcnt that thc
sccond spcakcr appcars to disagrcc with. Agrccmcnt would bc thc
prctcrrcd sccond part, cliciting a rcsponsc such as 'Ycah' or cvcn 'l
think so' . 1hc sccond spcakcr _ulic hnds hcrsclt in thc position
ot producing a disprctcrrcd.
j t;] Cindy. So chiropodists do hands l gucss.
|ulic. Emwcllout thcrcthcy thcy mostly work on
pcoplc's tcct.
|ulic's disprctcrrcd sccond part is markcd
p
ith initial hcsitations,
as it it is dimcult to pcrtorm this action csscntially corrccting thc
othcr . 1hcrc is a dclay 'cm', plus pausc in gctting startcd and
thc actual statcmcnt which indicatcs disagrccmcnt only comcs
attcr a prctacc 'wcll' , an appcal to thc vicws ot othcrs ' out
thcrc' , and a stumbling rcpctition [ 'thcy thcy' } . Lvcn thc statc-
mcnt contains an cxprcssion 'mostly' which makcs thc informa-
tion lcss challcnging to thc claim in thc hrst part. 1hc ovcrall
cttcct is that this spcakcr is prcscnting hcrsclt as having dimculty
and is unwilling to havc to say what is bcing statcd.
8o S URVEY
I

Hcstatons and prctaccs arc also tound in dsprctcrrcd sccond

parts to nvtatons, as shown n j t 8] .
t 8] ccky. Comc ovcr tor somc cottcc latcr.
Wally. OhchI'd lovc tobut you sccII'm
supposcd to gct this hnshcdyou know.
As s ottcn thc casc, thc cxprcsson ot a rctusal a dsprctcrrcd
sccond can bc accomplshcd wthout actually sayng ' no' .
Somcthng that sn't sad ncvcrthclcss gcts communcatcd n j t &] .
Ahcr a prctacc ' Oh' and a hcstaton 'ch' , thc sccond spcakcr
n t &] produccs a knd ot tokcn acccptancc 'I'd lovc to' to show
apprccation ot thc nvtation. 1hcn, thc othcr's undcrstandng s
nvokcd 'you scc' and an account s prcscntcd 'I'm supposcd to
gct ths hnshcd' to cxplan what prcvcnts thc spcakcr trom
acccpting thc nvtaton. 1hcrc s also a m

anng convcycd hcrc

that thc spcakcr's crcumstanccs arc bcyond hs control bccausc
ot an oblgaton 'I' m supposcd to' and, oncc agan, thc nvtcr's
undcrstandng 'you know' s nvokcd.
1hc pattcrns assocatcd wth a disprctcrrcd sccond n Englsh
arc prcscntcd as a scrcs ot optonal clcmcnts n j t] .
j t] How to do a dsprctcrrcd Examplcs
a.
b.
c.
d.
c.
t.
g.
h.
.

k.
dclaylhcstatc
prctacc
cxprcss doubt
tokcn Ycs
apology
mcnton oblgaton
appcal tor undcrstandng
makc t non-pcrsonal
gvc an account
usc mtgators
hcdgc thc ncgatvc
pausc cr cm ah , , ,
wcll, oh
I'm not surc, I don't
know
that's grcat, I'd lovc to
I'm sorry, what a pty
I must do 7I'm
cxpcctcd n Y
you scc, you know
cvcrybody clsc, out
thcrc
too much work, no
tmc lch
rcally, mostly, sort ot,
kinda
I gucss not, not possiblc
CONVERSATI ON AND PREFERENCE STRUCTURE & t
1hc ovcrwhclming chcct ot a dis prctcrrcd is that morc timc and
morc languagc arc uscd than in a prctcrrcd. Morc languagc csscn-
tially rcprcscnts morc distancc bcmccn thc cnd ot thc hrst part
and thc cnd ot thc sccond part. From a pragmatic pcrspcctivc, thc
cxprcssion ot a prctcrrcd in rcsponsc to an ottcr or invitation,
tor cxamplc clcarly rcprcscnts closcncss and quick conncction.
1hc cxprcssion ot a disprctcrrcd, as mappcd out in j t] , would
rcprcscnt distancc and lack ot conncction. From a socia| pcrspcct-
ivc, it is casy to scc why participants in a convcrsation might try to
avoid crcating contcxs tor disprctcrrcds. Onc obvious dcvicc tor
accomplishing this is to usc thosc prc-scqucnccs dcscribcd at thc
cnd ot Chaptcr 7. 1hc bcst way to avoid a disprctcrrcd sccond is
not to gct to thc point whcrc a hrst part ot thc pair is uttcrcd. lt
must tol|ow, thcn, that convcrsations bctwccn thosc who arc
closc tamiliars wi|l tcnd to havc tcwcr claboratc disprctcrrcds
than convcrsations bctwccn thosc who arc still working out thcir
social rclationship. 1hc amount ot talk cmploycd to accomplish a
particular social action in convcrsation is a pragmatic indicator ot
thc rclativc distancc bcmccn thc participants.

8 z S URVEY
1
|
l
I

i
J
I
.
Discourse and culture
The emphasis in the preceding chapter was on the sequential
structure of conversation, particularly on aspects of the turn
taking procedures for control of the foor, with less attention paid
to what speakers had to say once they got the floor. Having gained .
the floor, speakers have to organize the structure and content of
what they want to say. They have to package their messages in
accordance with what they think their listeners do and do not
know, as well as sequence everything in a coherent way. If those
speakers decide to write out their messages, creating written text,
they no longer have listeners providing immediate interactive
feedback. Consequently, they have to rely on more explicit struc
tural mechanisms for the organization of their texts. In this
expanded perspective, speakers and writers are viewed as using
language not only in its interpersonal function (i.e. taking part in
social interaction), but also in its textual function (i.e. creating
well-formed and appropriate text), and also in its ideational func
tion (i.e. representing thought and experience in a coherent way).
Investigating this much broader area of the form and function of
what is said and written is called discourse analysis.
Discourse analysis
Discourse analysis covers an extremely wide range of activities,
from the narrowly focused investigation of how words such as 'oh'
or 'well' are used in casual talk, to the study of the dominant ideo
logy in a culture as represented, for example, in its educational or
political practices. When it is restricted to linguistic issues, dis
COurse analysis focuses on the record (spoken or written) of the
DISCOURSE AND CULTURE 83
proccss by which languagc is uscd in somc contcxt to cxprcss
intcntion.
aturally, thcrc is a grcat dcal ot intcrcst in thc structurc ot dis-
coursc, with particular attcntion bcing paid to what makcs a wcll-
tormcd tcxt. Within this structural pcrspcctivc, thc tocus is on
topics such as thc cxplicit conncctions bctwccn scntcnccs in a tcxt
that crcatc cohcsion, or on clcmcnts ot tcxtual organization that
arc charactcristic ot storytclling, tor cxamplc, as distinct trom
opinion cxprcssing and othcr tcxt typcs.
Howcvcr, within thc study ot discoursc, thc pragmatic
pcrspcctivc is morc spccializcd. lt tcnds to tocus spccihcally on
aspccts ot what is unsaid or unwriucn yct communicatcd within
thc discoursc bcing analyzcd. ln ordcr to do thc pragmatics ot dis-
coursc, wc havc to go bcyond thc primarily social conccrns ot
intcraction and convcrsation analysis, look bchind thc torms and
structurcs prcscnt in thc tcxt, and pay much morc attcntion to
psychological conccpts such as background knowlcdgc, bclicts,
and cxpcctations. ln thc pragmatics ot discoursc, wc incvitably
cxplorc what thc spcakcr or writcr has in mind.
Coherence
Gcncrally, what languagc uscrs havc most in mind is an assump-
tion ot coherence, that what is said or writtcn will makc scnsc in
tcrms ot thcir normal cxpcricncc ot things. 1hat 'normal' cxpcri-
cncc will bc locally intcrprctcd by cach individual and hcncc will
bc ticd to thc tamiliar and thc cxpcctcd. ln thc ncighborhood
whcrc l livc, thc noticc in j t a. ] mcans that somconc is sclling

plants, but thc noticc in j tb. ] docs not mcan that somconc is scll-
ng garagcs.
j t] a. Plant Salc
b. Garagc Salc
Although thcsc noticcs havc an idcntical structurc, thcy arc intcr-
prctcd dittcrcntly. lndccd, thc intcrprctation ot [ tb. ] , that somc-
onc is sclling ouschold itcms trom thcir garagc, is onc that
rcquircs somc tamiliarity with suburban litc.
1his cmphasis on tamiliarity and knowlcdgc as thc basis ot
cohcrcncc i s ncccssary bccausc ot cvidcncc that wc tcnd to makc
8q S URVEY
instant ntcrprctatons ot tamiliar matcral and tcnd not to scc
possiblc altcrnatvcs. For cxamplc, thc qucston prcscntcd n j z] is
casly answcrcd by many pcoplc.
jz] How many animals ot cach typc did Moscs takc on thc
Ark!
l you immcdiatcly thought ot ' two' , thcn you acccsscd somc
common cultural knowlcdgc, pcrhaps cvcn without notcing that
thc namc uscd ' Moscs' was inappropriatc. Wc actually crcatc a
cohcrcnt ntcrprctaton or a tcxt that potcntally docs not havc t.
Wc arc also unlkcly to stop and puzzlc ovcr 'a malc and a
cmalc what! ' as wc rcad about thc accidcnt rcportcd n j j ] .
[ j] A motor vchiclc accidcnt was rcportcd in tront ot Kcnncdy
1hcatrc nvolvng a malc and a cmalc.
Wc automatically ' hll in' dctails tor cxamplc, a malc pcrson
driving onc ot thc motor vchiclcs to crcatc cohcrcncc. Wc also
construct tamlar sccnarios in ordcr to makc scnsc ot what mght
hrst appcar to bc odd cvcnts, as in thc ncwspapcr hcadlinc in jq] .
[q] Man Robs Hotcl with Sandwich
l you crcatcd an intcrprctaton tor jq] that had thc sandwch
pcrhaps in a bag bcing uscd as it it was a gun, thcn you actvatcd
thc knd o background knowlcdgc cxpcctcd by thc wrtcr as
conhrmcd by thc rcst ot thc ncwspapcr articlc . You may, ot
coursc, havc crcatcd a quitc ditcrcnt kind ot intcrprctation tor
cxamplc, thc man was catng thc sandwch whlc robbing thc
hotcl . Whatcvcr t was, t was incvitably bascd on what you had
in mnd and not only on what was in thc 'tcxt' in jq] .
Background knowledge
Our ablty to arrvc automatically at intcrprctations ot thc
unwrttcn and thc unsad must bc bascd on prc-cxstng know-
lcdgc structurcs. 1hcsc structurcs tunction lkc tamlar pattcrns
trom prcvous cxpcrcncc that wc usc to intcrprct ncw cxpcri-
cnccs. 1hc most gcncral tcrm tor a pattcrn ot this typc s schema
plural, SChemata) . A schcma is a prc-cxsting knowlcdgc struc-
turc n mcmory.
lt thcrc is a hxcd, static pattcrn to thc schcma, t is somctmcs
DI S COURS E AND CULTURE &y
called a fame. A frame shared by everyone within a social group
would be something like a prototypical version. For example,
within a frame for an apartment, there will be assumed compon
ents such as kitchen, bathroom, and bedroom. The assumed ele
ments of a frame are generally not stated, as in the advertisement
in [ 5] .
[ 5] Apartment for rent. $500. 763-668 3 .
A normal ( local) interpretation of the small fragment of discourse
in [5 ] will be based on not only an ' apartment' frame as the basis
of inference ( if X is an apartment, then X has a kitchen, a bath
room, and a bedroom) , but also an 'apartment for rent' adverise
ment frame. Only on the basis of such a frame can the advertiser
expect the reader to fll in 'per month' and not 'per year' afer
' $ 5 00' here. If a reader of the discourse in [ 5] expects that it
would be 'per week', for example, then that reader clearly has a
different frame (i. e. based on a different experience of the cost of
apartment rental ! ) . The pragmatic point will nevertheless be the
same: the reader uses a pre-existing knowledge structure to create
an interpretation of what is not stated in the text.
When more dynamic types of schemata are considered, they are
more ofen described as scripts. A script is a pre-existing know
ledge structure involving event sequences. We use scripts to build
interpretations of accounts of what happened. For example, we
have scripts for what normally happens in all kinds of events,
such as going to a doctor's offce, a movie theater, a restaurant, or
a grocery store as in [6] .
[ 6] I stopped to get some groceries but there weren't any bas
kets lef so by the time I arrived at the check-out counter I
must have looked like a j uggler having a bad day.
Part of this speaker's normal script for ' getting groceries' ob
viously involves having a basket and goi ng to the check- out
counter. Everything else that happened in this event sequence is
assumed to be shared background knowledge ( for example, she
went through a door to get inside the store and she walked
around picking up items from shelves) .
The concept of a script is simply a way of recognizing some
expected sequence of actions in an event. Because most of the
8 6 S URVEY
dctai|s ot a script arc assumcd to bc known, thcy arc un|ikc|y to bc
statcd. For mcmbcrs ot thc samc cu|turc, thc assumption ot
sharcd scripts allows much to bc communicatcd that is not said.
Howcvcr, tor mcmbcrs ot dittcrcnt cu|turcs, such an assumption
can |cad to a grcat dca| ot miscommunication.
Cultural schemata
Evcryonc has had thc cxpcricncc ot surprisc whcn somc assumcd
componcnt ot an cvcnt is uncxpcctcd|y missing. I rcmcmbcr my
hrst visit to a Moroccan rcstaurant and thc abscncc ot onc ot my
'rcstaurant script' rcquircmcntsthcrc wcrc no chairs | 1hc |argc
comtortab|c cushions wcrc an cxcc|lcnt rcp|accmcnt. It is a|most
incvitab|c that our background know|cdgc structurcs, our
schcmata tor making scnsc ot thc wor|d, wi|| bc cu|tura||y dctcr-
mincd. Wc dcvc|op our cultural schemata in thc contcxts ot our
basic cxpcricnccs.
For somc obvious dittcrcnccs tor cxamp|c, cushions instcad ot
chairs , wc can rcadi|y modi[ thc dctai|s ot a cu|tura| schcma.
For many othcr subtlc dittcrcnccs, howcvcr, wc ohcn don't rccog-
nizc that thcrc may bc a misintcrprctation bascd on dittcrcnt
schcmata. In onc rcportcd cxamp|c, an Austra|ian tactory supcr-
visor c|car|y assumcd that othr tactory workcrs wou|d know
that Eastcr was closc and hcncc thcy wou|d a|| havc a ho|iday. Hc
askcd anothcr workcr, origina||y trom Victnam, about hcr p|ans,
as in j;] .
[;] You havc hvc days ott. What arc you going to do!
Thc Victnamcsc workcr immcdiatc|y intcrprctcd thc uttcrancc in
tcrms ot bcing |aid ott rathcr than having a ho|iday . Somcthing
good in onc pcrson's schcma can sound |ikc somcthing bad in
anothcr's.
Cross-cultural pragmatics
Thc study ot dittcrcnccs in cxpcctations bascd on cultura|
sch
cmata is part ot a broad arca ot invcstigation gcncra||y known
as cross-cultural pragmatics. 1o |ook at thc ways in which mcaning
is constructcd by spcakcrs trom dittcrcnt cu|turcs wi|| actua||y
DI S COURS E AND CULTURE 8;
rcquirc a complctc rcasscssmcnt ot virtually cvcrything wc havc
considcrcd so tar in this survcy. 1hc conccpts and tcrminology
may providc a basic analytic tramcwork, but thc rcalization ot
thosc conccpts may dittcr substantially trom thc Lnglish languagc
cxamplcs prcscntcd hcrc.
Whcn wc rcvicwcd thc coopcrativc principlc and thc maxims,
wc assumcd somc kind ot gcncral middlc-class Anglo-Amcrican
cultural background. What it wc assumcd a cultural prctcrcncc
tor not saying what you know to bc thc casc in many situations !
Such a prctcrcncc is rcportcd in many culturcs and would clcarly
rcquirc a dittcrcnt approach to thc rclationship bctwccn thc
maxims ot quality and quantity in a morc comprchcnsivc
pragmatics.
Whcn wc considcrcd turn-taking mcchanisms, wc did not
cxplorc thc powcrtul rolc ot silcncc within thc normal convcrsa-
tional practiccs ot many culturcs. or did wc includc a discussion
ot a socially prcscribcd 'right to talk' which, in many culturcs, is
rccognizcd as thc structural basis ot how intcraction procccds.
Whcn wc cxplorcd typcs ot spccch acts, wc did not includc
any obscrvations on tbc substantial dittcrcnccs that can cxist
cross-culturally in intcrprcting conccpts likc 'complimcnting' ,
'thanking', or 'apologizing' . 1hc typical Amcrican English stylc
ot complimcnting crcatcs grcat cmbarrassmcnt tor

somc ativc
Amcrican lndian rcccivcrs it's pcrccivcd as cxccssivc} , and can
clicit a rcaction similar to apologizing trom somc |apancsc
rcccivcrs it's pcrccivcd as impossiblc to acccpt . lndccd, it is
unlikcly that thc division onc cultural group makcs bctwccn any
two social actions such as 'thanking' or ' apologizing' will bc
matchcd prcciscly within anothcr cu
|
turc.
1hc study ot thcsc dittcrcnt cultural ways ot spcaking is somc-
timcs callcd contrastive pragmatics. Whcn thc invcstigation tocuscs
morc spccihcal|y on thc communicativc bchavior ot non-nativc
spcakcrs, attcmpting to communicatc in thcir sccond languagc, it
is dcscribcd as interlanguage pragmatiCS. Such studics incrcasingly
rcvcal that wc all spcak with what might bc callcd a pragmatic
accent., that is, aspccts ot our talk that indicatc what wc assumc is
communicatcd without bcing said.
lt wc havc any hopc at all ot dcvcloping thc capacity tor cross-
cultural communication, wc will havc to dcvotc a lot morc
8 8 S URVEY

attcntion to an undcrstanding ot what charac

crizcs pragmatic
acccnt, not only in othcrs, but in oursclvcs. l hopc that this brict
survcy has providcd a bcginning, and an inccntivc to cxplorc
turthcr.
DI S COURSE AND CULTURE 89
S ECTI ON Z
0Bdl hg5
Chapter 1
Defi nitions and background
T0X 1
GEORGI A GREEN: Pragmatics and Natural Language
Understanding. Lawrcncc Erlbaum 8, pagc j
1hc broadcst ntcrprctaton ot pragmatcs s that t s thc study ot
undcrstandng ntcntonal human acton. 1hus, t nvolvcs thc
ntcrprctaton ot acts assumcd to bc undcrtakcn n ordcr to
accomplsh somc purposc. 1hc ccntral notons n pragmatcs
must thcn ncludc bclct, ntcnton or goal , plan, and act.
Assumng that thc mcans andor thc cnds nvolvc communca-
ton, pragmatcs stll cncompasscs all sorts ot mcans ot commun-
caton, ncludng nonconvcntonal, nonvcrbal, nonsymbolc oncs
as, tor cxamplc, whcn a ltcguard throws a vollcyball n thc drcc-
ton ot a swmmcr strugglng n thc occan. 1hc ltcguard bclcvcs
that thc swmmcr wants assstancc, and that thc swmmcr wll
undcrstand that thc vollcyball thrown n hs drccton s ntcndcd
by thc ltcguard to bc assstancc, and that thc swmmcr wll
know how to takc advantagc ot thc vollcyball's propcrty ot bcng
lghtcr than watcr. 1hat makcs at lcast thrcc bclcts and onc ntcn-
tion on thc part ot thc ltcguard, ncludng two bclcts about thc
swmmcr's bclcts, and onc about thc swmmcr's dcsrcs.
>
From this description, it seems as if every act in life is part of
pragmatics. Do you think that pragmatics is the study of all
actions, or should it be limited to only certain actions? What
kind of limitations would you propose?
READI NGS 91
| The fnal sentence in this brief extract mentions cbeliefs about
. . . beliefs' .. How can we know about a person's beliefs when
we are analyzing their actions and utterances?
| If the swimmer doesn't want assistance (in the example), how
does that afect the analysis?
T0X Z
'Pragmatics. mcaning and contcxt. ' Filc 70 in Language Files:
Materials for an Introduction to Linguistic.s. 6th cdn.
Ohio Statc Univcrsity Prcss 1yy1, pagc 223
1o tully undcrstand thc mcaning ot a scntcncc, wc must also
undcrstand thc contcxt in which it was uttcrcd. Consi dcr thc
word ball. ln a scntcncc such as, He kicked the ball into the net,
wc may visualizc a round, black and whitc socccr ball about ninc
inchcs in diamctcr. ln a scntcncc such as She dribbled the ball
down the court and shot a basket, wc would visualizc a baskct
ball. Givcn yct anothcr scntcncc, She putted the ball in from two
feet away, wc would visualizc anothcr ball, a golt ball. ln thcsc
cxamplcs, thc word ball is undcrstood in dittcrcnt ways dcpcnd-
ing on what typc ot action is associatcd with it. Whatcvcr undcr-
stood mcaning is common to ball in all ot thcsc contcxts will bc
part ot thc word's corc mcaning. lt wc think ot cnough typcs ot
balls, wc can comc up with an invariant corc mcaning ot ball that
wil l allow spcakcrs to rctcr to any bal l in any contcxt.
cvcrthclcss, cvcn though wc can discovcr a word's ' invariant
corc' , wc normally undcrstand morc than this. lt is thc LL!EX
that hlls in thc dctails and allows tull undcrstandingsuch as thc
usual color of a socccr ball, thc sizc ot a baskctball, or thc wcight
ot a golt ball. 1hc study ot thc contribution ot contcxt to mcaning
is ottcn callcd pragmatics.
| What do you think is the 'invariant core' meaning of the word
cbalr, as proposed here? Can you think of any use of the word
'balr that would not have that Ccore' meaning? Can cthe con
text' cause a word not to have its 'core' meaning?
| What does the term ccontext' seem to refer to in this text? If
you have a different concept of 'context', how would you
revise this paragraph to illustrate it more clearly?
92 READI NGS
l In what ways is the view ofpragmatics in this text similar to or
different from the way pragmatics is defned in Text I ?
Chapter 2
Deixis and distance
lex3
CHARLES FI LLMORE: Santa Cruz Lectures on Deixis.
Indiana University Linguistics Club 1975 , pages 40-2
The most obvious place deictic terms in English are the adverbs
'here' and 'there' and the demonstratives 'this' and 'that' , along
with their plural forms; the most obvious time deictic words are
adverbs like 'now' or 'today' . There are important distinctions in
the uses of these and other deictic words which I would like us to
be clear about right away. I will frequently need to point out
whether a word or expression that I am referring to can be used in
one or more of three different ways, and these I will call gestural,
symbolic, and anaphori. By the gestural use of a deictic expres
sion I mean that use by which it can be properly interpreted only
by somebody who is monitoring some physical aspect of the com
munication situation; by the symbolic use of a eictic expression I
mean that use whose interpretation involves merely knowing cer
tain aspects of the speech communication situation, whether this
knowledge comes by current perception or not; and by the
anaphoric use of an expression 1 mean that use which can be cor
rectly interpreted by knowing what other portion of the same dis
course the expression is coreferential with.
I can illustrate the distinction I'm talking about by taking the
word 'there'. It has all three uses. Its gestural use can be seen in a
sentence like, 'I want you to put it there' . You have to know where
the speaker is pointing in order to know what place he is indicat
ing. The symbolic use is exemplifed in the telephoner's utterance,
'Is Johnny there? ' . This time we understand the word 'there' as
meaning ' in the place where you are' . An example of the
ana phoric use of 'there' is a sentence like 'I drove the car to the
parking lot and left it there' . In that case the word refers to a
place which had been identifed earlier in the discourse, namely
the parking lot. Take another example, this time one showing j ust
READI NGS 93
thc distinction bctwccn thc gcstural and thc symbolic usc. lt dur-
ing my lccturc you hcar mc usc a phrasc likc ' this hngcr' , thc
chanccs arc tairly good that you will look up to scc what it is that
l want you to scc, you will cxpcct thc word to bc accompanicd by
a gcsturc or dcmonstration ot somc sort. On thc othcr hand, it
you hcar mc usc thc phrasc 'this campus' , you do not nccd to look
up, bccausc you know my mcaning to bc 'thc campus in which l
am now locatcd' , and you happcn to know whcrc l am. 1hc tor-
mcr is thc gcstural usc, thc lattcr thc symbolic usc.
| Can you transfer this discussion to temporal deixis (as
described in Chapter 2), considering then' (instead of cthere')
in gestural, symbolic, and anaphoric uses?
| Given the three categories described here, which category
seems to ft the typical uses of deictic' expressions such as yes
terday' and Ctomorrow'?
Placc indications takc pan in thc dcictic systcm ot a languagc by
virtuc ot thc tact that tor many locating cxprcssions, thc location ot
onc, or anothcr, or both, ot thc spccch act participants can scrvc as a
spatial rctcrcncc point. Somctimcs all that mcans is that tor an
cxprcssion which in a nondcictic usc rcquircs mcntion ot
a rctcrcncc ob| cct, in its dcictic usc thc rctcrcncc ob| cct, takcn to bc
thc spcakcr's body at thc timc ot thc spccch act, simply gocs unmcn-
tioncd. 1akc, tor cxamplc, thc cxprcssion ' upstairs' . lt l say,
'|ohnny livcs upstairs' , you will undcrstand mc as mcaning upstairs
ot thc placc whcrc l am at thc timc l say thc scntcncc, unlcss thc
immcdiatcly prcccding discoursc has providcd somc othcr rctcr-
cncc point. lt l say 'Harry livcs ncarby', thc samc can bc said. You
will undcrstand that Harry livcs ncar to thc placc whcrc l am whcn
l say thc scntcncc, again, cxccpt tor thc casc whcrc a rctcrcncc point
has bccn idcntihcd in thc immcdiatcly prcccding discoursc.
| Is the speaker's body always the unmentioned reference point,
as Fillmore suggests here? Consider the uses of words like
crront', back', cdown (the street) ', cabove', outside', and any
others that seem to be similar to cupstairs' and Cnearby' in the
' I
examples.
r
1

yq READI NGS
T0X 4
QUENTI N S MI TH: '1hc mu|tip|c uscs ot indcxica|s' in
Synthese ;8, 8, pagcs 8 z-j
'l am in last placc' is ottcn uscd to indicatc that thc spcakcr is in
last placc. ut this scntcncc is also uscd on a numbcr ot occasions
to indicatc that somcbody c|sc is in |ast placc. l am watching a
racc and thc pcrson upon whom l havc bct, o. o, drops to thc
|ast p|acc. 'l am in last placc| ' l cxclaim in anguish to my compan-
ion. My companion knows pcrtcct|y wcll what 1 mcanthat the
person upon whom I have bet is in last placc. lndccd, shc rcplics in
kind, disagrccing with my statcmcnt. 'o you arcn't| Look| ' shc
cxclaims, pointing at o. o, 'You arc passing o. j | '
| Can you think of any other contexts where 'I' is not to be liter
ally interpreted as 'the person who is speaking'?
| Do examples such as these mean that we need a new defnition
of the meaning of the word 'I' in English? If yes, what would
have to be in that defnition? If no, how would you explain
this type of 'extra' usage?
T0Xb
GEOFFREY NUNBERG: 'lndcxicality and dcixis' in
Linguistics and Philosophy 6, tj, pagc q
. . . you might point at a picturc ot |ohn Ashbcrry to idcnti| his
most rcccnt book, using thc dcmonstrativc that, with no rcstric-
tion on thc things you cou|d say about it.
q 1hat is in all thc bookstorcs on thc top shclt, tcmporarily
out ot stock .
ut whilc |ohn Ashbcrry might casily say ot himsclt ' l am in al| thc
bookstorcs,' it would bc odd tor him to say 'l am on thc top shclt' or
'l am tcmporarily out ot stock,' unlcss it could bc supposcd that thc
tact that an author's book was on thc top shc|t or was tcmporarily
out ot stock carricd somc notcworthy implications tor him.
|
Following on from these examples, could you point to an
empty space on the bookshelf and and ask the owner of the
bookstore, 'Is that out of stock?'? If yes, do we have to re
formulate the defnition of deixis (i. e. 'pointing via language')
when there's nothing being pointed to?
READI NGS y
| Why do you think the idea of 'some noteworthy implictions'
is mentioned in this text? Does identifing the reference of
deictic expressions depend on information about a person's
thoughts and feelings? If yes, can you think of other examples
(involving' other deictic forms) ?
| How does the example with 'I' in this text ft in with your
analysis of 'I' in Text 4?
Chapter 3
Reference and i nference
10X 6
KEITH DONNELLAN: 'Rctcrcncc and dchnitc dcscriptions' in
Philosophical Review 75, 1966, pagcs 28 5-6
l will call thc two uscs ot dchnitc dcscriptions l havc in mind thc
attributivc usc and thc rctcrcntial usc. P spcakcr who uscs a
dchnitc dcscription attributivcly in an asscrtion statcs somcthing
about whocvcr or whatcvcr is thc so-and-so. A spcakcr who uscs
a dchnitc dcscription rctcrcntially in an asscrtion, on thc othcr
hand, uscs thc dcscription to cnablc his audicncc to pick out
whom or what hc is talking about and statcs somcthing about
that pcrson or thing. Hthc hrst casc thc dchnitc dcscription might
bc said to occur csscntially, tor thc spcakcr wishcs to asscrt somc-
thing about whatcvcr or whocvcr hts that dcscription, but in thc
rctcrcntial usc thc dchnitc dcscription is mcrcly onc tool tor doing
a ccrtain | obcalling attcntion to a pcrson or thingand in gcn-
cral any othcr dcvicc tor doing thc samc | ob, anothcr dcscription
or a namc, would do as wcll. ln thc attributivc usc, thc attributc ot
bcing thc so-and-so is all important, whilc it is not in thc rctcrcn-
tial usc.
1o illustratc this distinction, in thc casc ot a singlc scntcncc,
considcr thc scntcncc, ' Smith's murdcrcr is insanc. ' Supposc hrst
that wc comc upon poor Smith toully murdcrcd. rom thc brutal
manncr ot thc killing and thc tact that Smith was thc most lovablc
pcrson in thc world, wc might cxclaim, ' Smith's murdcrcr
,
s
insanc

' l will assumc, to makc it a simplcr casc, that in a quitc

ordinary scnsc wc do not know who murdcrcd Smith though this
96 READI NGS
is not in the end essential to the case) . This, I shall say, is an
attributive use of the defnite description.
The contrast with such a use of the sentence is one of those situ
ations in which we expect and intend our audience to realize
whom we have in mind when we speak of Smith's murderer and,
most importantly, to know that it is this person about whom we
are going to say something.
I Before Donnellan's proposal, many philosophers argued that
if a description does not ft anything, then it fails to refer.
What is Donnellan's perspective on this?
I Using Donnelan's ditinction (plus any additional ditinctions you
think are needed), how would you account for the use of a denite
description that does not accurately ft the person or thing?
I Can the attributive versus referential distinction be related to
Fillmore's distinction (Text 3) between gestural, symbolic,
and anaphoric uses of deictic expressions?
Tex 7
M. A. K. HALLI DAY and RUQAIYA HASAN: Cohesion in
English. Longman I 976, page 3 I
There are certain items in every language which have the property
of reference, in the specifc sense in which we' are using the term
here; that is to say, instead of being interpreted semantically in
their own right, they make reference to something else for their
interpretation. In English these items are personals, demonstra
tives and
c
omparatives.
We start with an example of each:
a. Three blind mice, three blind mice.
See how they run! See how they run!
b. Doctor Foster went to Gloucester in a shower of rain.
He stepped in a puddle right up to his middle and never went
there again.
c. There were two wrens upon a tree.
Another came, and there were three.
In
( a) , they refers to three blind mice; in ( b) there refers to
Glo
ucester; in ( c) another refers to wrens.
READI NGS ' 97
1hcsc itcms arc dircctivcs indicating that intormation is to bc
rctricvcd hom clscwhcrc. So much thcy havc in common with all
cohcsivc clcmcnts. What charactcrizcs this particular typc ot
cohcsion, that which wc arc calIing REFERENCE, is thc spccihc
naturc ot thc intormation that is signaIlcd tor rctricval. ln thc casc
ot rctcrcncc thc intormation to bc rctricvcd is thc rctcrcntial
mcaning, thc idcntity ot thc particular thing or class ot things that
is bcing rctcrrcd to, and thc cohcsion lics in thc continuity ot rct-
crcncc, whcrcby thc samc thing cntcrs into thc discoursc a sccond
timc.
| In this analysis, the assumption is that certain words refer to
other words. Do you think that this is a helpful or misleading
assumption? .
| Do you agree with the fnal statement that Cthe same thing
enters into the discourse a second time'? How about example
(c), where the analysis proposes that the word Canother' refers
to Cwrens'?
| If the word cthere' in (b) is an example of cohesion by refer
ence, is the word cthere' in the second line of (c) the same?
How do you decide?
| Is Donnellan's distinction in Text 6 relevant to what these
authors are saying?
Chapter 4
Presupposition and entailment
TexS
ROBERT L. STALNAKER: 'Pragmatic prcsupposition' in
Milton Munitz and Pctcr Ungcr cds. . Semantics and
Philosophy. cw York Univcrsity Prcss t ;q,
pagcs t - 200
Although it is normally inappropriatc bccausc unncccssary tor
mc to asscrt somcthing that cach ot us assumcs thc othcr alrcady
bclicvcs, my asscrtons will ot coursc always havc conscqucnccs
which arc part ot thc common background. For cxamplc, in a
contcxt whcrc wc both know that my ncighbor is an adult malc,
8 READI NGS
I say 'My ncghbor s a bachclor,' whch, lct us supposc, cntals
that hc s adult and malc. I mght | ust as wcll havc sad 'my ncgh-
bor s unmarrcd. ' 1hc samc ntormaton would havc bccn con-
vcycd although thc nuanccs mght not havc bccn cxactly thc
samc . 1hat s, the increment of information, or ot contcnt, con-
vcycd by thc hrst statcmcnt s thc samc as that convcycd by thc
sccond. It thc asscrtcd proposton wcrc acccptcd, and addcd
to thc common background, thc rcsultng stuaton would bc
thc samc as t thc sccond asscrton wcrc acccptcd and addcd to thc
background.
1hs noton ot common background bclct s thc hrst approx-
maton to thc noton ot pragmatc prcsupposton that I want to
usc. A proposton P s a pragmatc prcsupposton ot a spcakcr n
a gvcn contcxt | ust n casc thc spcakcr assumcs or bclcvcs that P,
assumcs or bclcvcs that hs addrcsscc assumcs or bclcvcs that P,
and assumcs or bclcvcs that hs addrcsscc rccognzcs that hc s
makng thcsc assumptons, or has thcsc bclcts.
> Do you agree that the two utterances quoted in the frst para
graph would add exactly the same information to the com
mon background?
| According to the defnition presented in the second para
graph, would it be correct, or not, to say that a pragmatic pre
supposition is any belief of the speaker? (1 t may be helpful to
look again at Chapter q,pages 25-3 0.)
>
Can you think of circumstances where it is not inappropriate
for someone eto assert something that each of us assumes the
other already believes' ?
Tex 9
GERALD GAZDAR: Pragmatics. Implicature, Presupposition,
and Logical Form. Acadcmc Prcss 1979, pagc to6
6y John got to safety before the boiler blew up.
66 John got to the safety handle before the boiler blew up.
It wc assumc n 66 that |ohn's gcttng to thc satcty handlc prc-
vcntcd thc bolcr blowng up, thcn 66 docs not, but 6y docs,
prcsupposc that thc bolcr blcw up. It wc trcat before as bcng
'ambguous' , thcn wc arc agan lch wth no prncplc tor dccdng
READI NGS yy
whcthcr or not thc prcsupposition attachcs to a particular scn-
tcncc. otc also that, it all prcsupposing constructions arc
ambiguous, thcn thc notion ot ' intclicity' or ' unacccptability' is
inapplicablc, sincc wc will always havc an altcrnativc rcading
with rcspcct to which thc scntcncc wilI bc acccptablc.
| How do you account for the fact that cbefore' creates a pre
supposition in example (65), but not in (66) ? Can you think of
other examples where the use of cbefore' does, or does not,
lead to a presupposition?
| Does 'afer' work the same way? Should we defne 'before'
and 'afer', not only as opposites, but also as creating different
presuppositions?
Chapter S
Cooperation and implicature
Tex 10
PAUL GRI CE: 'Logic and convcrsation' in P. Colc and |. L.
Morgan cds. . Syntax and Semantics Volume j : Speech Acts.
Acadcmic Prcss 1975, pagc 48
l would Iikc to bc ablc to think ot thc standard typc ot convcrsa-
tional practicc not mcrcly as somcthing that alI or most do I N
FACT tollow but as somcthing that it i s REAS ONAB LE tor us to
tollow, that wc S HOULD NOT abandon. For a timc, l was
attractcd by thc idca that obscrvancc ot thc CP jco-opcrativc
principlc] and thc maxims, in a talk cxchangc, could bc thought
ot as a quasi-contractual mattcr, with parallcls outsidc thc rcalm

ot discoursc. lt you pass by whcn l am struggling with my
strandcd cag l no doubt havc somc dcgrcc ot cxpcctation that you
will ottcr hclp, but oncc you | oin mc in tinkcring undcr thc hood,
my cxpcctations bccomc strongcr and takc morc spccihc torms
in thc abscncc ot indications that you arc mcrcly an incompctcnt
mcddlcr , and talk cxchangcs sccmcd to mc to cxhibit, charactcr-
isticalIy, ccrtain tcaturcs that | ointly distinguish coopcrativc
transactions.
1 . 1hc participants havc somc common immcdiatc aim, likc
gctting a car mcndcd, thcir ultimatc aims may, ot coursc, bc
100 READI NGS
independent and even in conflict-each may want to get the
car mended in order to drive off, leaving the other stranded.
In characteristic talk exchanges, there is a common aim even
if, as in an over-the-wall chat, it is a second order one,
namely that each party should, for the time being, identif
himself with the transitory conversational interests of the
other.
2. The contri butions of the participants should be dovetailed,
mutually dependent. .
3 . There is some sort of understanding (which may be explicit
but which is often tacit) that, other things being equal, the
transaction should continue in appropriate style unless both
parties are agreeable that it should terminate. You do not
j ust shove off or start doing something else.
But while some such quasi-contractual basis as this may apply to
some cases, there are too many types of exchange, like quarreling
and letter writing, that it fails to ft comfortably.
I Can you spell out why 'quarreling and letter writing' do not ft
comfortably with the conditions presented here?
I What would you call the three 'features' listed here if you were
to make them into maxims for cooperative transactions?
C Grice emphasizes the word 'reasonable' as he describes his
consideration of the cooperative principle and his maxims as a
kind of contract. Would the cooperative principle, the
maxims, and the three features listed here be treated as
'reasonable' in all societies and cultures?
Tex 11
J . L. MORGAN: ' Two types of convention in indirect speech
acts' in P. Cole ( ed. ) : Syntax and Semantics Volume 9:
Pragmatics. Academic Press 1978, pages 277-8
Just above I presented cases involving particular expressions and
the conventionalization of their use for certain implicatures, as in
the case of If you've seen one, you've seen them all, or the original
example, Can you pass the salt? I said in the later case that it had
become a convention of usage to use this expression, with its lit
eral meaning, to convey an implicature of request. The question
READI NGS 101
now arises, can there be this kind of conventionalization of rules
of conversation? I think there can. For example, it is more or less
conventional to challenge the wisdom of a suggested course of
action by questioning the mental health of the suggestor, by AN
appropriate linguistic means, as in:
( 3 7) Are you crazy?
( 3 8 ) Have you last your mind?
( 39) Are you out of your gourd?
and so on. Most Americans have two or three stock expressions
usable as answers to obvious questions, as in:
( 40) Is the Pope Catholic?
( 41 ) Do bagels wear bikinis?
But for some speakers the convention does not specif a particu
lar expression, and new ones are manufactured as they are
needed. It seems that here a schema for implicature has been con
ventionalized: Answer an obvious yes/no question by replying
with another question whose answer is very obvious and the same
as the answer you intend to convey.
In a similar way, most speakers have a small number of expres
sions usable as replies to assertions, with the implicature that the
assertion is transparently false-(42) , for example:
( 42) Yes, and I'm Marie the Queen of Romania.
But again, for some speakers the convention sp-ecifes only a
general strategy, rather than a particular expression: To convey
that an assertion is transparently false, reply with another asser
tion even more transparently false.
I Do you know any other 'stock expressions' for these tpes of
occasions (request, challenge, answer to obvious questions,
reply to a false assertion) ? How would you explain (to some
one learning English as a foreign language, for example) how
to work out the communicated meaning from the literal
meaning?
I The author uses the term 'convention' in talking about the
kinds ol implicatures involved here. Do you think that the
examples presented here can be analyzed in terms of conven
tional implicatures (as discussed in Chapter 5, pages 45-6) ?
102 READI NGS
r What do you think about the idea that an implicature may
begin by being based on inference, but can become so conven
tionalized that no one has to make the inference any more? Is
that the same process as we use in interreting idioms?
Chapter 6
Speech acts and events
Tex 12
J OHN S EARLE: Speech Acts. Cambridge University Press
1969, pages 5 8-9
One crucial distinction between promises on the one hand and
threats on the other is that a promise is a pledge to do something
for you, not to you; but a threat is a pledge to do something to
you, not for you. A promise is defective if the thing promised is
something the promisee does not want done; and it is further
defective if the promisor does not believe the promisee wants it
done, since a non-defective promise must be intended as a
promise and not as a threat or warning. Furthermore, a promise,
unlike an invitation, normally requires some sort of occasion or
situation that calls for the promise. A crucial feature of such occa
sions or situations seems to be that the promisee wishes (needs,
desires, etc. ) that something be done, and the promisor is aware of
this wish ( need, desire, etc. ) . I think both halves of this double
condition are necessary in order to avoid fairly obvious counter
examples.
t This paragraph lists several required features for a speech act
to count as a promise. Do you agree that all these features are
necessary? Are other crucial features not included here?
One can, however, think of apparent counter-examples to this
condition as stated. Suppose I say to a lazy student, 'If you don't
hand in your paper on time I promise you I will give you a failing
grade in the course' . Is this utterane a promise? I am inclined to
think not; we would more naturally describe it as a warning or
possibly even a threat. But why, then, is it possible to use the
locution 'I promise' in such a case? I think we use it here because
'I promise' and 'I hereby promise' are among the strongest
READI NGS 103
illocutionary torcc indicating dcviccs tor commitment providcd
by thc English languagc. For that rcason wc ottcn usc thcsc
cxprcssions in thc pcrtormancc ot spccch acts which arc not
strictly spcaking promiscs, but in which wc wish to cmphasizc thc
dcgrcc ot our commitmcnt. 1o illustratc this, considcr anothcr
apparcnt countcr-cxamplc to thc analysis along dittcrcnt lincs.
Somctimcs onc hcars pcoplc say ' l promisc' whcn making an
cmphatic asscrtion. Supposc, tor cxamplc, l accusc you ot having
stolcn thc moncy. 1 say, 'You stolc that moncy, didn't you!' . You
rcply, 'o, l didn't, 1 promisc you l didn't' . Di d you makc a
promisc in this casc! l hnd it vcry unnatura| to dcscribc your
uttcrancc as a promisc. 1his uttcrancc would bc morc aptly
dcscribcd as an cmphatic dcnial, and wc can cxplain thc occur-
rcncc ot thc illocutionary torcc indicating dcvicc ' l promisc' as
dcrivativc trom gcnuinc promiscs and scrving hcrc as an cxprcs-
sion adding cmphasis to your dcnial.
| Do you agree that having used the words 'I promise', you
could later claim that 'strictly speaking' you did not make a
promise because you meant something else?
| What seem to be the conditions for an utterance containing
the IFID 'I promise' to serve as an emphatic denial?
| Is the recognition of speech act conditions related at all to the
cooperative principle as discussed in Text IO? (It may be help
ful to refer to the discussion of felicity conditions in Chapter 6,
pages 50-I.)
Tex 13
GEOFFREY LEECH: Principles of Pragmatics.
Longman 1983 , pagcs 1 77-8
ln rctcrring to human convcrsational bchavior, as to othcr arcas
ot cxpcricncc, our languagc providcs us with catcgorical distinc-
tions. ut it is to commit a tundamcntal and obvious crror to
assumc that thc distinctions madc by our vocabulary ncccssarily
cxist in rcality. Languagc providcs us with vcrbs likc order,
request, beg, plead, | ust as it providcs us with nouns |ikc puddle,
pond, lake, sea, ocean. ut wc should no morc assumc that thcrc
arc in pragmatic rcality distinct catcgorics such as ordcrs and
1 04 READI NGS
rcqucsts than that thcrc arc in gcographical rcality distinct cat-
cgorics such as puddlcs, ponds and lakcs. Somchow, this assump-
tion slips unnoticcd into Scarc's introduction to his taxonomy.
What arc thc critcria by which wc can tcll that ot thrcc actual
uttcranccs onc is a rcport, onc a prcdiction and onc a promisc!
ln ordcr to dcvclop highcr ordcr gcncra, wc must hrst know
how thc spccics promise, prediction, report, ctc. dittcr trom
onc anothcr.
Scarlc, |. 1 979. : Expression and Meaning. Cambridgc.
Cambridgc Univcrsity Prcss, pagc 2. )
ut it would bc strikingly inappropriatc it onc wcrc to bcgin a
trcatisc on cxpanscs ot watcr on thc world's surtacc in this way.
What arc thc critcria by which wc can tcll that ot thrcc actual
cxpanscs ot watcr, onc is a puddlc onc a pond, and onc a
akc! ln ordcr to dcvclop highcr ordcr gcncra, wc must hrst
know how thc spccics puddle, pond, and lake dittcr trom onc
anothcr.
ln dctcncc ot Scarlc it could bc argucd, hrst, that thc comparison is
untair. it onc had choscn monkcys and girattcs say} instcad ot
ponds and puddlcs, thc cxamplc would havc bccn lcss ridiculous.
ut my rcply is a} that onc has no right in advance to assumc that
such catcgorics cxist in rcality although onc might discovcr thcm
by obscrvation} , and b} that in actuality, whcn onc does obscrvc
thcm, illocutions arc in many rcspccts morc likc puddlcs and
ponds than likc monkcys and girattcs. thcy arc, that is to say, dis-
tinguishcd by continuous rathcr than by discrctc charactcristics.
> What exactly is the argument being presented here against the
idea that we can identif a speech act as a prediction or not?
>
What would distinguish the defnition of a puddle, in Leech's
view, from the kind of defnition of a promise presented in
Text I2?
>
Do you think that Leech's argument is based on an important
issue, or just a minor point? How do you think Searle would
respond to this criticism from Leech?
READI NGS 105
Chapter 7
Politeness and interaction
Tex 14
ROB I N LAKOFF: Talking Power. The Politics of Language.
Basic Books 1990, pages 3 4, 3 6, 3 8
Indirectness can function as a form of politeness. Politeness is a
system of interpersonal relations designed to facilitate interaction
by minimizing the potential for confict and confrontation inher
ent in all human interchange. We like to think of conversation as
confict-free, with speakers normally bei ng able to satisf one
another's needs and interests. But, in fact, we enter every conver
sation-indeed, every kind of discourse-with some personal
desideratum in mind: perhaps as obvious as a favor or as subtle as
the desire to be likeable. For some of these needs, participants can
accede to each other, and both gain their desires; but with others,
one must lose, however minimally, for the other to win. One per
son must tell another something that the other doesn't want to
hear; one person must refuse another's request; one person must
end a conversation before the other is quite willing to go. In such
cases, there is the danger of insult and, consequently, the break
down of communication. If societies did not devise ways to
smooth over moments of confict and confrontation, social rela
tionships would be difcult to establish and continue, and essen
tial cohesion would erode. Politeness strategies are the means to
preserve at least the semblance of harmony and cohesion . . . .
l In what ways is this defnition of politeness more or less
specifc than the general social meaning of politeness you are
familiar with?
Distancing cultures weave remoteness into their language. The
attribution of responsibility represents an intrusion of the per
sonal: it suggests that individuals with different interests are
involved in the discussion. So grammatical devices that minimize
a speaker's personal involvement are favored-for instance, pass
ive verb forms and impersonal forms l i ke one. Words that
threaten to convey or evoke dangerous emotion are replaced with
safer ones, which suggest that no emotion is involved. This formal
language is the language of diplomacy, bureaucracy, and the pro-
1 06 READI NGS
fessions. Diplomats speak of an incident when they mean that
their countries are in a virtual state of war; bureaucrats talk of
revenue enhancement when they renege on a promise of no new
taxes; doctors discourse on iatrogenesis when they mean they did
something that made the patient sick. These words provide a
buffer between pure denotative meaning and its emotional wal
lop: the hearer, in all probability, knows perfectly well what the
speaker intends; but the latter has chosen deliberately Latinate
words from a sector of the vocabulary not rich in emotional con
notations, so as to lessen the danger of collision.
[ Can you think of other examples of distance politeness in lan
guage use?
[ Can you think of situations or special circumstances where
the type of distance politeness, as defned here, is ignored?
It is essential to realize that camaraderie can be conventional . . . .
But . . . someone unaccustomed to conventional camaraderie will
take it as genuine, arising out of long acquaintance and the develop
ment of mutual liking and trust. Modern camaraderie probably
began in California as an outgrowth of the human potential move
ment of the 1960s and I970s. For a while it was a bane to visiting
Easterners, who were confounded by the Californian's appearance
of good fellowship and deep caring; the immediate frst-naming,
touching, looking deep into the eyes, and asking truly caring ques
tions: ' Are you really happy with your life?' To the properly
brought-up, Easterner, such behavior was permissible only afer
years of earning it, and maybe not then. Easterners fell into one of
several schools of thought about the character of Californians:
either that they had the simplicity of children and should be patron
ized; or that they were rough frontier sorts, probably raised by
wolves ( and you know how wolves are) ; or that they were truly
wonderful people who could get to know you as well afer two
seconds as would take most of us a lifetime. All of these attitudes
assumed, of course, that the camaraderie was real rather than
conventional.
[
What examples of language use would you predict (or have
you experienced) as representative of 'conventional cama
raderie' in contrast to 'distance politeness' ?
READI NGS 107
Tex 15
GABRI ELE KAS PER: 'Politcncss' in R. E. Ashcr cd. .
The Encyclopedia of Language and Linguistics. Volumc 6.
Pcrgamon 1994, pagc 3 209
Somc typcs ot linguistic action arc carricd out morc trcqucntly in
somc culturcs than in othcrs. Hcarcr-bcnchcial acts such as com-
plimcnting and thanking occur morc rcgularly in somc Wcstcrn
contcxts c.g., thc USA) than in somc Asian culturcs c.g. , main-
land China , rcflccting both thc strong positivc politcncss oricnta-
tion and rcluctancc to imposc on othcrs in mainstrcam Amcrican
culturc, on thc onc hand, and thc assumption, in China, that par-
ticipants act according to thcir social positions and associatcd
rolcs and obligations, on thc othcr. Also, hcarcr-costly acts such as
rcmsals arc pcrccivcd as bcing morc socially ottcnsivc by |apancsc
and Chincsc intcrlocutors and thus tcnd to bc avoidcd, whcrcas it
sccms morc consistcnt with Amcrican intcrlocutors' right to sclt-
dctcrmination not to comply with anothcr pcrson's wishcs.
| Can you think of other 'hearer-benefcial acts' and other
'hearer-costly acts'? For example, what is an invitation or a
complaint? Is it possible that the concepts of 'cost' and
'beneft' may be culturally deterined?
| There is a suggestion in this text that people in the USA are
more concerned with their rights as individuals than with
their social roles and obligations. What kind of evidence from
language behavior would you look for in order to decide
whether this suggestion is true or not?
| Can you characterize the normal behavior of your own social
group as having more 'hearer-benefcial' acts? What about
'hearer-costly' acts? Are there other social groups with whom
you share the same language, but whose politeness strategies
appear to be different?
| Where does Lakoff's 'conventional camaraderie' (Text I4) ft
into the distinction that Kasper is making here?
108 READI NGS
Tex 16
PENELOPE BROWN and S TEPHEN LEVI NS ON: Politeness.
Cambridgc Univcrsity Prcss 8;, pagc z8
ln languagc thc constraints arc morc on torm than on contcnt or
at lcast torm providcs a morc tcasiblc arca ot study . 1hc ways in
which mcssagcs arc hcdgcd, hintcd, madc dctcrcntial, and cmbcd-
dcd in discoursc structurcs thcn bccomc crucial arcas ot study. ut
such arcas arc also thc conccrn ot pragmatics, thc study ot thc
systcmatic rclation ot a languagc to contcxt. 1hc spccial intcrcst
ot sociolinguistics in our vicw is in thc dittcrcntial usc ot such
pragmatic rcsourccs by dittcrcnt catcgorics ot spcakcrs in dittcr-
cnt situations. lt is in this way that wc dcrivc our slogan
'Sociolinguistics should bc applicd pragmatics. '
| Do you agree with the assumption that pragmatics comes frst
and then is 'applied' to the social use of language, or should it
be the other way round?
| Notice that the concepts of 'hedge' and 'hint' are used here.
Recall the use of 'hedges' on implicatures in Chapter j, pages
3 8-9 (which themselves may be termed 'hints'); would such
phenomena in the use of language be better analyzed as
aspects of politeness? Is pragmatics really just the study of lin
guistic politeness?
| Does the 'slogan' at the end of this text provide a better (or
worse) perspective on pragmatics than those offered in Texts
I and earlier?
Chapter S
Conversation and preference structure
Tex 17
HARVEY SACKS : Lectures on Conversation. Volumc t .
lackwcll z, pagcs 3-4
l'll start ott by giving somc quotations.
t ) A: Hcllo.
b: Hcllo.
z A: 1his i s Mr Smith may I hclp you.
b: Ycs, this is Mr rown.
READI NGS o
( 3 ) A: 1his is Mr Smith may l hclp you.
b. l can't hcar you.
A: 1his is Mr Smith.
b. Smith.
1hcsc arc somc hrst cxchangcs in tclcphonc convcrsations col-
lcctcd at an cmcrgcncy psychiatric hospital. 1hcy arc occurring
bctwccn pcrsons who havcn`t talkcd to cach othcr bctorc. Onc ot
thcm, A, is a statt mcmbcr ot this psychiatric hospital. can bc
cithcr somcbody calling about thcmsclvcs, that is to say in troublc
in onc way or anothcr, or somcbody calling about somcbody clsc.
l havc a largc collcction ot thcsc convcrsations, and l got startcd
looking at thcsc hrst cxchangcs as tollows. A scrics ot pcrsons
who callcd this placc wo

ld not givc thcir namcs. 1hc hospital's

conccrn was, can anything bc donc about it! Onc qucstion l
wantcd to addrcss was, whcrc in thc coursc ot thc convcrsation
could you tcll that somcbody would not givc thcir namc! So l
bcgan to look at thc matcrials. lt was in tact on thc basis ot that
qucstion that l bcgan to try to dcal in dctail with convcrsations.
l tound somcthing that struck mc as tairly intcrcsting quitc
carly. And that was that it thc statt mcmbcr uscd '1his is Mr Smith
may l hclp you' as thcir opcning linc, thcn ovcrwhclmingly, any
answcr othcr than 'Ycs, this is Mr rown' tor cxamplc, 'l can't
hcar you,' 'l don't know, ' 'How do you spcll your namc! ' mcant
that you would havc scrious troublc gctting thc callcr's namc, it
you got thc namc at all. . . .
Looking at thc hrst cxchangc comparcd to thc sccond, wc can
bc struck by two things. First ot all, thcrc sccms to bc a ht bctwccn
what thc hrst pcrson who spcaks uscs as thcir grccting, and what
thc pcrson who is givcn that grccting rcturns. So that it A says,
'Hcllo, ' thcn tcnds to say 'Hcllo. ' lt A says ' 1his is Mr Smith
may l hclp you, ' tcnds to say 'Ycs, this is Mr rown. ' Wc can say
thcrc's a proccdural rulc thcrc, that a pcrson who spcaks hrst in a
tclcphonc convcrsation can choosc thcir torm ot addrcss, and in
choosing thcir torm ot addrcss thcy can thcrcby choosc thc torm
ot addrcss thc othcr uscs.
| Do you think that the 'procedural rule' presented here applies
to all 'frst exchanges' in telephone conversations?
| Can you describe this 'procedural rule' in terms of preference
1 10 READI NGS
.
structure (as outlined in Chapter 8, pages 78-82) by including
example (3) in your analysis?
| What advantages and disadvantages do you think there are in
using telephone data as the basis for analyzing how conversa
tion works?
Tex 18
H. SACKS , E. S CHEGLOFF, and G. J EFFERS ON: 'A simplcst
systcmatics tor thc organization ot turn-taking in convcrsa-
tion' in Language y0, 1974, pagcs 700-1
1o mcrit scrious considcration, it sccms to us, a modcl should bc
capablc ot accommodating i. c. , cithcr bc compatiblc with, or
al|ow thc dcrivation of} thc tollowing grossly apparcnt facts. In
any convcrsation, wc obscrvc thc tollowing.
( 1 ) Spcakcr-changc rccurs, or at lcast occurs.
( 2) Ovcrwhclmingly, onc party talks at a timc.
( 3 ) Occurrcnccs ot morc than onc spcakcr at a timc arc com-
mon, but brict.
( 4) 1ransitions trom onc turn to a ncxt} with no gap and no
ovcrlap arc common. 1ogcthcr with transitions charac-
tcrizcd by slight gap or slight ovcrlap, thcy makc up thc
vast ma| ority ot transitions.
( y } 1urn ordcr is not hxcd, but varics.
( 6) 1urn sizc is not hxcd, but varics.
( 7) Lcngth ot convcrsation is not spccihcd in advancc.
( 8 ) What partics say is not spccihcd in advancc.
( 9) Rclativc distribution ot turns is not spccihcd in advancc.
( 1 0) umbcr ot partics can vary.
( 11 ) 1alk can bc continuous or discontinous.
( 1 2) 1urn-allocation tcchniqucs arc obviously uscd. A currcnt
spcakcr may sclcct a ncxt spcakcr as whcn hc addrcsscs a
qucstion to anothcr party} , or partics may sclt-sclcct in
starting to talk.

( 1 3 ) Various 'turn-constructional units' arc cmploycd, c. g. ,
turns can bc pro| cctcdly 'onc word long' , or thcy can bc
scntcntial in lcngth.
( 14) Rcpair mcchanisms cxist for dcaling with turn-taking
crrors and violations, c.g., it two partics hnd thcmsclvcs
READI NGS 1 1 1
talking at thc samc timc, onc ot thcm will stop prcma-
turcly, thus rcpairing thc troublc.
| Can you divide these fourteen statements into two groups
one that applies to all conversations and one that applies to
only some conversations in some contexts? What kinds of
situations or people appear to create exceptions?
| Should these statements be restricted to any conversation that
is middle-class American and basically friendly? Can you
think of different factors such as social class, culture, ethnic
it, relationship, age-or any others that will have an effect on
how turn-taking proceeds?
Tex 19
JACK B I LMES : Discourse and Behavior Plcnum Prcss 1 986,
pagc 1 66
.
Considcr thc tollowing cxchangc.
A jaddrcssing ] . Whcrc arc you going!
jno rcsponsc]
A 1hc hcll with you.
1his cxchangc makcs scnsc. It is ordcrly, not random. Wc may
charactcrizc 's non rcsponsc with an inhnitc varicty ot ncg-
ativcs. lt is not a qucstion, not a promisc, not a lccturc, and so
torth. Howcvcr, givcn that qucstions call tor answcrs, it is relev
antly not an answcr.
| Why do you think the word (relevantly' is emphasized in this
text? Does this mean that every C(non) response' counts as
relevantly not something in conversation?
| Consider what speaker A says i n reaction to the
C(non)response'. What kind of speech act is this? Does this
utterance tell us anything about the relationship between the
two speakers (i. e. strangers, acquaintances, or intimates) ?
1 1 2 READI NGS
Chapter 9
Discourse and culture
Tex Z
J OHN GUMPERZ and JENNY COOK- GUMPERZ:
'lntroducton. languagc and thc communcaton ot socal
dcntty' n |. Gumpcrz cd. . Language and Social Identity.
Cambrdgc Unvcrsity Prcss 1982, pagc 1 2
Although thc pragmatic condtons ot communcatvc tasks arc
thcorctcally takcn to bc univcrsal, thc rcalizatons ot thcsc tasks
as socal practccs arc culturally varablc. 1hs varaton can bc
analyzcd trom scvcral dttcrcnt pcrspcctvcs, all ot whch ot
coursc co-occur n thc actual practccs.
t ) Dittcrcnt cultural assumptons about thc situaton and
about appropratc bchavor and ntcntons wthn t.
( 2) Dittcrcnt ways ot structurng ntormaton or an argumcnt
n a convcrsaton.
( 3 ) Dttcrcnt ways ot spcakng. thc usc ot a dittcrcnt sct ot
unconscous lngustc convcntons such as tonc ot vocc
to cmphaszc, to sgnal locaI conncctons and to ndcatc
thc sgnhcancc ot what s bcng sad n tcrms ot ovcrall
mcanng and atttudcs.
y 'dttcrcnt cultural assumptons' wc rctcr to thc tact that, cvcn
though pcoplc n stuatons such as wc study agrcc on thc ovcrall
purposc ot thc ntcracton, thcrc arc ohcn radcal dittcrcnccs as to
what cxpcctations and rights arc nvolvcd at any onc timc.
> There is a suggestion here that 'pragmatic conditions' can be
treated as 'universal' (i. e. applicable everywhere). Can you
suggest some examples of pragmatic universals? How about
'Be polite'? Any others?
> Can you think of any examples that would support the idea
that 'appropriate behavior' differs in different cultures (prag
matically speaking) ?
> Do you agree with these authors that there are different ways
of 'structuring an argument'? How is an argument structured
in English? How could it be structured any other way?
READI NGS 1 1 3
Tex ZJ
]NNYTHOMAS. 'Cross-cultural pragmatic failure' in
Applied Linguistic 4/2, 1983, page 105
'Free goods' are those which, in a given situation, anyone can use
without seeking permission, for example, salt in a restaurant
(providing, of course, that you are having a meal in that restau
ant and have not simply wandered in from the street with a bag of
fsh and chips) . Generally speaking, what an individual regards as
'free goods' varies according to relationships and situation. In
one's own family or home, most things ( food, drink, books,
baths) are free goods. H a stranger's house they are not. Cross
culturally, too, perceptions of what constitutes 'free' or 'nearly
free' goods differ. In Britain, matches are 'nearly free', so one
would not use 8 particularly elaborate politeness strategy to
request one, even of a total stanger. Hthe Soviet Union cigarettes
are also virtually 'free' and a request for them demands an equally
minimal degree of politeness, such as Daite sigaretu [give (me) a
cigarette]. A Russian requesting a cigarette in this country and
using a similar strategy would either have wrongly encoded the
amount of politeness sle intended (covert grammatical or pragma
linguistic failure) or seriously misjudged the size of imposition
( sociopragmatic failure) .
| The author is writing tin this countr'} about Britain. Do you
think her observation on salt in a restaurant is based on a uni
versal component of a 'restaurant script'? In a family context,
do you agree that 'most things . . . are treated as free goods'?
What about other cultures you are familiar with?
| The examples in this text are physicl objects. There are also
cultural diferences in what kind of information is considered
'free goods'. What constraints are there, in cultures you are
familiar with, on asking people about certain topics (for
example, their political views, religion, marital status,
income, cost of their possessions, bathroom behavior sexual
practices) ?
| What do you think the distinction is between the two kinds of
'failure' (pragmalinguistic and sociopragmatic) described
here?
1 1q RADINGS
Tex 22
DEB ORAH TANNEN: You Just Don't Understand.
Wm. Morrow 1 990, page 40
A woman was telling me why a long-term relationship had ended.
She recounted a recurrent and pivotal conversation. She and the
man she lived with had agreed that they would both be free, but
they would not do anything to hurt each other. When the man
began to sleep with other women, she protested, and he was
incensed at her protest. Their conversation went like this:
SHE: How can you do this when you know it's hurting me?
HE: How can yo try to limit my freedom?
SHE: But it makes me feel awful.
HE: You are trying to manipulate me.
On one level, this is simply an example of a clash of wills: What he
wanted conficted with what she wanted. But in a fundamental
way, it refects the difference in focus I have been describing. In
arguing for his point of view, the key issue for this man was his
independence, his freedom of action. The key issue for the woman
was their interdependence-how what he did made her feel. He
interpreted her insistence on their interdependence as ' manipula
tion': She was using her feelings to control his behavior.
I Do you agree with the analysis presented here? Are there
other implicatures possible from what is said in the dia./og?
I We are used to thinking that the term 'cross-culturar will
apply to people from different countries. Is it appropriate to
think of the interactions between males and females within
one country (sharing a lot of one culture) as a site for the study
of cross-cultural pragmatics? What kinds of differences might
be worthy of investigation?
READI NGS I I 5
S ECTI ON
ofof0DOo5
1hc rcfcrcnccs which follow can bc classihcd into introductory
lcvcl markcd .00) , morc advanccd and conscqucntly morc tcch-
nical markcd 0) , and spccializcd, vcry dcmanding markcd
aaa ) .
Chapter 1
Definitions and backround

S TEVEN DAVI S cd. } . Pragmatics. A Reader

Oxford Univcrsity Prcss 1991
1his is a collcction of thirty-hvc papcrs, originally publshcd n
| ournals dcaling main|y with philosophical issucs in thc rcccnt
history of pragmatics .
LL
GEORGIA GREEN: Pragmatics and Natural Language
Understanding. Lawrcncc Erlbaum 1989
1his is an introduction which focuscs on linguistc pragmatics as
'thc study of undcrstanding intcntional human action' , with a
strong cmphasis on grammatcal issucs .
00
GEOFFREY LEECH: Principles of Pragmatics.
Longman 1 983
1his introductory tcxt prcscnts a rhctorical modcl of pragmatics,
attcmpting to dcscribc ' principlcs and maxims of good COU~
REFERENCES 1 1 7
municativc bchaviour' . Pragmatics is dchncd as 'thc study ot how
uttcranccs havc mcanings in situations' , with an cmphasis on thc
analysis ot politcncss .
L
STEPHEN C. LEVI NS ON: Pragmatics.
Cambridgc Univcrsity Prcss 1 983
1his widcly uscd introductory tcxt ottcrs scvcral dittcrcnt
dchnitions ot pragmatics and prcscnts ' an ovcrvicw ot somc ot thc
ccntral tasks that pragmaticists wrcstlc with' . 1hc cmphasis is on
Iinguistic and philosophical issucs .
LL
JACOB MEY: Pragmatics: An Introduction.
lackwcl| 1993
1his is a comprchcnsivc introduction to pragmatics as 'thc study
ot thc conditions ot human languagc usc as thcsc arc dctcrmincd
by thc contcxt ot socicty'. 1hcrc is a strong cmphasis on thc ways
in which socicty's institutions govcrn thc usc ot languagc .
LL
JAN NUYTS and JEF VERS CHUEREN cds. .
A Comprehensive Bibliography of Pragmatics. Volumcs 1-4.
|ohn cn| amins 1987
1his rcmarkablc rcsourcc providcs a widc rangc ot rctcrcnccs.
1hc vcry usctul diagrams in thc Sub| cct lndcx pagcs 5 1-69) act
as a guidc to thc widc arcas ot study covcrcd by pragmatics.
Chapter 2
Oeixis and distance
L
S TEPHEN ANDERS ON and EDWARD KEENAN: 'Dcixis' in
1imothy Shopcn cd. . Language Typology and Syntactic
Description. Volume j : Grammatical Categories and the
Lexicon. Cambridgc Univcrsity Prcss 198 5
1his papcr prcscnts a rcvicw ot thc rangc ot dcictic cxprcssions
uscd in a widc varicty ot languagcs.
1 1 8 REFERENCES
0
ROB ERT JARVELLA and WOLFGANG KLEI N ( eds. ) : Speech,
Place and Action: Studies in Deixis and Related Topics.
John Wiley & Sons 1 982
This is a collection of ffeen papers on different aspects of deixis
by both linguists and psychologists, incorporating studies on
deixis and the blind and in the sign language of the deaf .
0
J OHN LYONS : Natural Language and Universal Grammar
Cambridge University Press 1991
Chapters 8 and 9 in this collection of essays provide a lot of
insights into the nature of deixis .
00
ROGER WALES : 'Deixis' in P. Fletcher and M. Garman (eds. ) :
Language Acquisition ( 2nd edn. ) Cambridge University Press
1 98 6
This is a review paper covering studies of the frst appearance
and development of deictic forms in the early language of young
children .
0
J URGEN WEI S S ENBORN and WOLFGANG KLEI N ( eds. ) :
Here and There: Cross-linguistic Studies on Deixis and
Demonstration. John Benj amins 1 982
This is a collection of fourteen papers on different types of deixis
in a wide range of languages.
Chapter 3
Reference and i nference

HERBERT C LARK and DEANNA WI LKES - GI B B S : ' Referring
as a collaborative process' in Cognition 22, 1 98 6
This important paper presents evidence for the ways in which
speakers in conversation collaborate to create referring expressions.
REFERENCES 1 1 9

GI LES FAUCONNIER: Mental Spaces.

Cambridgc Univcrsity Prcss 1994
1his i s a vcry original approach to thc ways i n which wc conncct
words to rctcrcnts, cmphasizing thc assumption ot sharcd know-
lcdgc and thc rolc ot pragmatic conncctions .

TALMY GI VON: Mind, Code and Context: Essays in
Pragmatics. Lawrcncc Erlbaum 1989
1his co|lcction ot cssays covcrs many topics i n pragmatics,
including rctcrcncc Chaptcrs y and 6) , trom a pcrspcctivc that
cmphasizcs mnction what languagc is uscd tor .

J OHN LYONS : Semantics. Volumc 1 .

Cambridgc Univcrsity Prcss 1977
Chaptcr 7, on rctcrcncc, scnsc, and dcnotation, prcscnts a com-
prchcnsivc background to thc basic issucs in thc traditional
scmantic trcatmcnt ot how words arc uscd to rctcr .

GEOFFREY NUNBERG: The Pragmatics of Reference.

lndiana Univcrsity Linguistics Club 1977
1his disscrtation uscs thc idca that words can bc shown to havc
cndlcss possiblc rctcrcnts to arguc tor a pragmatic analysis in
which word-mcanings cannot bc scparatcd trom 'knowlcdgc ot
othcr kinds ot convcntions and social practiccs' .
Chapter 4
Presupposition and entailment

NOEL B URTON-ROBERTS : The Limits to Debate. A Revised

Theory of Semantic Presupposition. Cambridgc Univcrsity
Prcss 1 989
1his book rcprcscnts onc ot thc tcw rcccnt attcmpts to rcconsidcr
thc basic conccpts involvcd in prcsupposition.
1 20 REFERENCES

CHOON- KYU OH and DAVI D DI NEEN cds. . Syntax and

Semantics Volume II Presupposition. Acadcmc Prcss 1979
1hs collccton ot sxtccn papcrs, plus an cxtcnsvc bblography,
|lustratcs thc typcs ot controvcrscs surroundng thc naturc ot
prcsupposton. Many arc prcscntcd n vcry tcchncal languagc .

NEI L S MI TH and DEI RDRE WI LS ON: Modern Linguistics.
Pcngun 1 979
Chaptcrs 7 and 8 ot ths tcxt provdc a dctalcd rcvcw ot prcsup-
poston, cntalmcnt, and thc rolc ot ordcrcd cntalmcnts .

RO B VAN D ER SANDT: Context and Presupposition.

Croom Hclm 1 98 8
1hs book rcconsdcrs thc connccton bctwccn prcsupposton,
contcxt, and thc pro| ccton problcm.
Chapter 5
Cooperation and implicature
M
DIANE B LAKEMORE: Understanding Utterances. An
Introduction to Pragmatics. lackwcll 1992
1hs s an ntroducton to pragmatcs n whch Rclcvancc s takcn
to bc thc ccntral conccpt .

LAURENCE HORN: '1oward a ncw taxonomy tor pragmatc

ntcrcncc. Q-bascd and R-bascd mplcaturc' n Dcborah
Schttrn cd. . Meaning, Form and Use in Context: Linguistic
Applications. Gcorgctown Unvcrsty Prcss 1984
1hs papcr proposcs an altcrnatvc approach to analyzng how
mplcaturcs arsc, usng two nstcad ot tour maxms.
REFERENCES 1 2I
0
PAUL GRI CE: Studies in the Way of Words.
Harvard University'Press 1989
This volume includes the collected papers of the philosopher
whose ideas are considered by many to be the foundation of con
temporary pragmatics .

Proceedings of the Berkeley Lingistic Societ 1 6, 1990
There is a collection of sixteen papers, presented as a parasession
within these published proceedings, on the legacy of Grice, cover
ing a wide range of issues in the analysis of meaning .
0
DAN S PERBER and DEIRDRE WILS ON: Relevance.
Blackwell 1986
Presented as a study of human communication, this book takes
the single mam of Relevance as the key. Arguments and illustra
tions are presented to support the contention that 'communicated
information comes with a guarantee of relevance' .
Chapter 6
Speech ac and event
0
]. AUSTI N: How to Do Things with Words. ( 2nd edn. )
Clarendon Press 1975
The original work which introduced the concept of language use
as a form of acion .
00
KENT BACH and ROBERT HARNI SH: Linguistic
'Communication and Speech Acts. MIT Press 1979
Two linguists present a detailed framework for the analysis of
speech acts.
1 22 REFERENCES
L
J OHN S EARLE: Speech Acts. An Essay in the Philosophy of
Language. Cambridgc Univcrsity Prcss 1969
1hc bcst known work on thc topic, with dctailcd discussion of
both conditions and applications of thc conccpt of a spccch act .

J OHN S EARLE: Expression and Meaning. Studies in the

Theory of Speech Acts. Cambridgc Univcrsity Prcss 1979
A collcction of scvcn papcrs, inc|uding onc on indircct spccch acts
and anothcr on a taxonomy of il|ocutionary acts. 1hcsc frc-
qucntly citcd papcrs rcprcscnt a dcvclopmcnt of thc idcas prc-
scntcd carlicr in Scarlc ( 1969) .
L
J EF VERS CHUEREN: What People Say They Do With Words.
Ab|cx 198 5
1his book prcscnts a critical rcvicw of problcms in spccch act
thcory and a proposal for 3 diffcrcnt approach bascd on thc study
of linguistic action.
Chapter 7
Politeness and interaction
L
S HOS HANA B LUM- KULKA and GABRI ELE KAS PER: .
Journal of Pragmatics 14/2 Spccial lssuc on politcncss ,
1 990
1his collcction of six papcrs includcs a rcvicw papcr by Kaspcr on
currcnt rcscarch issucs as wcll as thrcc rcports on thc dcvclop-
mcnt of politcncss bchavior in childrcn .
L
PENELOPE BROWN and STEPHEN LEVI NS ON: Politeness.
Some Universals of Language Usage. Cambridgc Univcrsity
Prcss 1987
1his i s thc most comprchcnsivc book on linguistic po|itcncss,
REFERENCES 1 23
offering lots of detailed discussion and illustrations from different
languages .
0
PAUL DREW and JOHN HERITAGE (eds. ) : Talk at Work:
Interaction in Institutional Settings. Cambridge University
Press 1992
This is a collection of ffeen papers on the general topic of inter
action in work contexts ( for example, news interviews, court pro
ceedings, doctor's ofce) .
00
M. DUFON, G. KAS PER, S . TAKAHAS HI , and N. YOS HI NAGA:
'Bibliography on Linguistic Politeness' in Journal of
Pragmatics 21 , 1994, pages 5 27-78
This is an extremely useful listing of published work concerned
with language and politeness .
0
ERVI NG GOFFMAN: Forms of Talk. University of
Pennsylvania Press 198 1
This is a collection of fve important papers by one of the most
infuential writers on language and social interaction.
Chapter S
Conversation and preference structure
0
MAXWELL ATKI NS ON and JOHN HERITAGE (eds. ) :
Structures of Social Action: Studies in. Conversation Analysis.
Cambridge University Press 1984
This is a collection of sixteen papers by some of the best known
writers on conversation analysis .
0
JACK BI LMES : 'The concept of preference in conversation
analysis' in Language in Society 1 7, 1 988
1 24 REFERENCES
Jhis papcr prcscnts a rcvicw of thc uscs of thc tcrm ' prcfcrcncc'
and argucs for a morc prccisc application of thc analytic conccpt .
L
ROBERT CRAI G and KAREN TRACY cds. } . Conversational
Coherence: Form, Structure and Strategy. Sagc 1 983
Jhis is a col|cction of fourtccn papcrs on convcrsation as intcrpcr-
sonal communication, vicwcd from a rangc of pcrspcctivcs .
L
HARVEY SACKS : Lectures on Conversation. Volumcs 1-2.
lackwcll 1992
Jhcsc two vo|umcs prcscnt thc original |ccturc matcrial in which
thc foundations of convcrsation analysis wcrc cstablishcd .
L
DEBORAH TANNEN: Conversational Style: Analyzing Talk
Among Friends. Ab|cx 1984
Jhis book prcscnts cxcnsivc illustration of dihcrcnt aspccts of con-
vcrsational sqlc as 'thc basic tools with which pcoplc communicatc' .
L
TEUN VAN DI J K: Handbook of Discourse Analysis.
Volume j : Discourse and Dialogue. Acadcmic Prcss 1 985
Jhis volumc contains sixtccn papcrs i|lustrating a rangc of diffcr-
cnt pcrspcctivcs on aspccts of intcractivc talk.
Chapter 9
Discourse and culture
L
S . B LUM- KULKA, J . HODS E, and G. KAS PER cds. } . Cross
cultural Pragmatics: Requests and Apologies. Ablcx 1 989
Jhis is a collcction of tcn papcrs dcscribing studics undcrtakcn
within thc framcwork of thc Cross-cultural Spccch Act
Rcalization Pro| cct. .
1 25

GI LLIAN BROWN and GEORGE YULE: Discourse Analysis.
Cambridgc Univcrsity Prcss 1983
1his is a standard tcxtbook with a linguistic tocus on thc study ot
discoursc .

J OHN GUMPERZ: Discourse Strategies.
Cambridgc Univcrsity Prcss 1982
1his i s a collcction ot tcn papcrs by onc thc most influcntial
writcrs on social intcraction and cross-cultural communication .

GABRI ELE KAS PER and S HOS HANA B LUM- KULKA cds. .
Interlanguage Pragmatics. Oxtord Univcrsity Prcss 1993
1his is a collcction ot clcvcn papcrs on various pragmatic aspccts
ot sccond languagc lcarning .
0
DEBORAH S CHI FFRI N: Approaches to Discourse.
lackwcll 1994
1his i s a guidc to scvcral dittcrcnt tramcworks tor doing discoursc
analysis .

ANNA WI ERZBI CKA: Cross-cultural Pragmatics. The
Semantics of Human Interaction. Mouton dc Gruytcr 199I
1his is a book about how cultural valucs and norms shapc dittcr-
cnt modcs ot intcraction.
1 26 REFERENCES
S ECTI ON q
UI o55BQ
Pagc rctcrcnccs to Scction 1, Surcy, arc givcn at thc cnd ot cach cn.
adjacency pair A scqucncc ot two uncranccs by dittcrcnt spcakcrs
in convcrsation. 1hc sccond is a rcsponsc to thc hrst, c.g. qucs-
tion-answcr. j77j
anaphor 1hc word, typicaI|y a pronoun, uscd to maintain refer
ence to somconc or somcthing alrcady mcntioncd, c.g. `P old
man was Iimping towards us. He slowIy camc into vicw. ' j2j
antecedent 1hc initial cxprcssion uscd to idcntiq somconc or
somcthing tor which an anaphor is uscd Iatcr, c.g. 'An old man
was Iimping towards us. Hc slowly camc into vicw. ' j2j
atributable silence 1hc abscncc ot taIk whcn a spcakcr is givcn
thc right to spcak in convcrsation. j7j
atributive use Using an cxprcssion to idcntiq somconc or somc-
thing without bcing committcd to thc cxistcncc ot an actual pcr-
son or thing, c.g. 'thc hrst pcrson to walk on Mars' . jIj
backchannels/backchannel signals VocaI indications ot attcn-
tion, c. g. ' uh-huh' , ' hmm', whcn somconc clsc is talking. j7j
background entailment Any logical conscqucncc ot an uttcr-
ancc. jj
bald on record Uttcranccs, c. g. ordcrs, dircctIy addrcsscd to
anothcr whcrc thc illocuionary force is madc cxplicit. jj
cataphora 1hc usc ot a word typicaIly a pronoun to introducc
somconc or somcthing that is morc tully idcntihcd Iatcr, c.g. 'He
sIowIy camc into vicw. Pold man was limping towards us. ' j2j
coherence 1hc tamiliar nd cxpcctcd rcIationships in cxpcricncc
GLOS SARY 1 27
which we use to connect the meanings of utterances, even when
those connections are not explicitly made. [84]
commissive A speech act in which the speaker commits him or
herself to some future action, e.g. a promise. See Table 6. I. [54]
. constancy under negation Quality of the presupposition of a
statement remaining true when the statement is negated. [26]
content conditions In order to count as a particular type of
speech act, an utterance must contain certain features, e.g. a
promise must be about a future event." [50]
contex The physical environment in which a word is used: cf.
co-tex. [21]
contrastive pragmatics The study of culturally different ways of
using language. [88]
conventional implicature A additional unstated meaning asso
ciated with the use of a specifc word, e.g. 'A but B' implies a
contrast between A and B, so 'contrast' is a conventional
implicature of 'but'. [45]
conversational implicature A additional unstated meaning that
has to be assumed in order to maintain the cooperative principle,
e.g. if someone says 'The President is a mouse', something that
is literally false, the hearer must assume the speaker means to
convey more than is being said. [40]
converational stle Particular way of participating in conversa
tion. [76]
cooperative pri nciple A basic assumption in conversation that
each participant will attempt to contibute appropriately, at the
required time, to the current exchange of talk. [37]
co-tex The linguistic enviroinent in which a word is used: cf.
contex. [21]
counterfactual presupposition The assumption that certain
information is the opposite of true. [29]
cross-cultural pragmatics The study of different expectations
among diferent communities regarding how meaning is con
structed. [87]
cultural schemata Pre-existing knowledge structures based on
experience in a particular culture. [87]
declaration A speech act that brings about a change by being
uttered, e.g. a judge pronouncing a sentence. See Table 6. I. [53]
I 28 GLOS SARY
deference strateg Feature of interactive talk emphasizing nega-
te pline, the non-personal, and feedom fom impositon. [66]
deictic center The speaker's locationtime. [9]
deictic expression See deixis. [9]
deictic projecti on Speakers acting as if they are somewhere
else. [13]
deixis 'Pointing' via language, using a deictic expression, e. g.
'this', 'here' . [9]
directive A speech act used to get someone else to do something,
e.g. an order. See Table 6. 1. [54]
direc speech act Speech act where a direct relationship exists
between the structure and communicative fnction of an utter
ance, e.g. using an interrogative form ( ' Can you . . . ?' ) to ask a
question ( ' Can you swim?' ) : cf. indirec speech act. [55]
discourse analysis The study of language use with reference to
the social and psychological factors that infuence communica
tion. [83]
dispreferred The structurally unexpected next utterance as a
response, e.g. an invitation is normally followed by an accept
ance, so a refsal is dispreferred. [79]
distal Away from the speaker, e. g. 'that', 'there' : cf. proximal. [9] -
ellipsis The absence of a word or words from a structural slot.
[23]
entai l ment Something that logically follows from what is
asserted. [25]
essential condition In performing a speech ac, a requirement
that the uttera
n
ce commits the speaker to the act performed. [51]
exclusive 'we' Addressee excluded: cf. inclusive 'we'. [11]
existenti al presupposition An assumption that someone or
something, identifed by use of a noun phrase, does exist. [27]
explicit performative A speech ac containing a performative verb:
cf. implicit perormative. [52]
expressive A speech act in which the speaker expresses feelings
or attitudes, e.g. an apology. See Table 6. 1 . [53]
face A person's public self-image. [60]
face savi ng act Utterance or action which avoids a potential
threat to a person's public self-image. [61]
GLOS SARY 1 29
face threateni ng act Utterance or action which threatens a
person's public self-image. jIj
fce wants A person's expectations that their public self-image
will be respected. jIj
factive presupposition Te assumption that information stated
afer certain words, e.g. 'know', 'regret', is true: cf. non-fcive
presupposition. [27j
felicit conditions The appropriate conditions for a spech ac to
be recognized as intended. jU]
H par The frst utterance in an adjacency pair, e.g 'How are
you?' See also second par. [77j
foor The current right to speak in a conversaton. j72j
foreground entai l ment The main logical consequence of an
utterance. [j
frame A pre-existing knowledge structure with a fxed static
pattern. jj
general conditions Preconditons on performing a spech ac. [U]
generalized conversational implicature P additional unstated
meaning that does not depend on special or local knowledge:
cf. conversational implicture. j4I]
hedges Cautious notes expressed about how an utterance is to
be taken, e.g. 'as far as I know' used when giving some informa
tion. j]
high considerateness stle A non-interrupting, non-imposing
way of taking part in conversation. [7j
high i nvolvement stle Pactive, fast-paced, overlapping way of
taking part in conversation. j7j
honorifc Exression which marks that the addressee is of higher
status. jIUj
ideational function The use of language as a means of giving
structure to thought and experience. jj
Illocutionary Force I ndicati ng Device (I FI D) Indication in the
speaker's uterance of the communicatve force of that utterance.
j49j
illocuionar ac or frc Te comunicatve force of an utterance.
j4j
1 ]C GLOS SARY
implicature A short version of conversational implicature. [35]
implicit performative A speech act without a perormative verb: cf.
explicit performative. [52]
inclusive 'we' Speaker and addressee included: cf. exclusive 'we'.
[11]
indexicals Like deictic expressions, forms used for 'pointing' via
language. See deixis. [9]
indirect speech act Speech act where an indirect relationship
exists beteen the structure and communicative function of an
utterance, e.g. the use of an interrogative (' Can you . . . ? ' ) not to
ask a question, but to make a request ( ' Can you help me with
this?' ) : cf. direct speech act. [55]
i nference The listener's use of additional knowledge to make
sense of wha is not explicit in an utterance. [17]
inserion sequence A to part sequence that comes between
the frst and second parts of another sequence in conversation.
[77]
interlanguage pragmatics The study of how non-native speakers
communicate in a second language. [88]
i nterpersonal function The use of language for maintaining
social roles and taking part in social interaction. [83]
lexical presupposiion The assumption that, in using one word,
the speaker can act as if another meaning ( word) will be under
stood. [28]
local management system A metaphor for describing the con
ventions for organizing the right to speak in conversation. [72]
locutionary act The basic act of uttering a meaningful linguistic
form. [48]
manner One of the maxims, in which the speaker is to be clear
brief, and orderly. See Table 5 . I . [39]
maxim One of the four sub-principles of the cooperative principle.
See manner, quantit, quality, and relation. See also Table 5 . 1 . [37]
mitigating device Expression used to sofen an imposition, e.g.
'please' . [63]
negative face The need to be independent, not imposed on by
others: cf. positive fce. [61]
GLOS SARY 1 3 1
negative pol iteness Awareness of another's right not to be
imposed on: cf. posiive politeness. j2]
negative politeness strateg An attempt to demonstrate aware
ness of another's right not to be imposed on: cf. positive po
l iteness strategy. j4]
non-factive presupposition The assumption that certain in
formaton, as presented, is not tue: cf. fcive presupposion. j29]
of record Utterances not directly addressed to another. j]
on record Utterances directly addressed to another. j]
overlap More than one speaker talking at the same time in
conversation. jT2]
paricularized conversational impl icature An additional un
stated meaning that depends on special or local knowledge: cf.
converational implicature. j42]
. perormative hypothesis A proposal that, underlying every utter
ance, there is a clause with a verb that identifes the speh ac.
jI]
performative verb A verb that explicitly names the speech act,
e.g. the verb 'promise' in the utterance 'I promise to be there'.
j49]
perlocutionar actefec The efect of an utterance used to per
form a speech ac. j4,49]
person debds Forms used to point to people, e.g. 'me', 'you'. j9]
politeness Showing awareness of another person's public self
image fce wnts. jU]
positive f'c The need to be connected, to belong to a group: cf.
negative fce. j2]
positive politeness Showing solidarity with another: cf. negative
politenes. j2]
positive politeness strategy An appeal to solidarity with
another: cf. negative politeness srteg. j4]
potential presupposition An . assumption typically associated
with use of a linguistic form, e.g. the use of the verb 'regret' in
'He regrets doing that' carries an assumption that he actually
'did that'. j27]
pragmatic accent Aspects of talk that indicate what is assumed
to be communicated without being said. j]
1 ] 2 GLOS SARY
pragmatic connection A conventional association between a
person's name and a kind of obj ect, e.g. 'Shakespeare' used to
identif a book. [20]
pragmatics The study of speaker meaning as distinct from word
or sentence meaning. [4]
pre-announcement ' Utterance before an announcement to check
if an announcement can be made. [68)
preference/preference strucure A patern in which one type of
utterance will be more typically found in response to another in
a conversational sequence, e.g. an acceptance will more tpi
cally follow an invitation than a refsal. [79]
preferred The structurally expected next utterance used in a
response. [79]
pre-i nvitation Utterance before an invitation to check if an
invitation can be made. [68]
preparatory conditions Specifc requirements prior to an
utterance in order for it to count as a particular speech act.
[50]
pre-request Utterance before a request to check if a request can
be made. [67]
presupposition Something the speaker assumes to be the case.
[25]
primar performative A utterance which performs a speech ac
but which does not contain a perormative verb. [52]
projection problem The problem of the presupposition of a simple
structure not surviving when part of a more complex structure.
[30]
proximal . Near speaker, e. g. 'this', 'here': cf. dista. [9]
psychological distance Speaker's marking of how close or dis
tant something is perceived to be. [13]
quality One of the maxims, in which the speaker has to be truth
fl. See Table 5 . 1 . [38]
quantity One of the maxi ms, in which the speaker has to
be neither more or less informative than is necessary. See Table
5 . 1 . [38]
range of reference All the possible referents identifable by use
of a word. [21]
GLOS SARY 1 3 3
reference P act by which a speaker uses a word, or words, to
enable a listener to identify someone or something. jI7]
referential use Using an expression to identify someone or
something when the person or thing is assumed to be known:
cf. atributive use. jI]
referring expression A linguistic form which enables a listener
or reader to identify something. [I7]
relation One of the maxims, in which the speaker has to be rel
evant. See Table j. t .
representative A speech ac in which the speaker states what is
believed or known, e.g. an assertion. See Table 6. t . []
scalar i mplicature Padditional meaning of the negative of any
value higher on a scale than the one uttered, e.g. in saying 'some
children', I create an implicature that what l say does not apply
to 'all children' . j4I]
schema (plural schemata) A pre-existing knowledge structure
in memory typically involving the normal expected patterns
of things, e.g. an apartment schema has a kitchen, a bedroom,
etc. j]
script A pre-existing knowledge stucture for interpreting event
sequences, e.g. a visit to the dentist has a script of specifc events
in sequence (which might start wit giving one's name to the
receptionist and fnish with making a further appointent) . j]
second par The second or response utterance in an adj acency
pair, e.g. 'Fine, thanks' . See frst par. j77]
semantics The study of how words literally connect to things, or
more generally, the investigation of meaning as encoded in lan
guage. j4]
sincerit conditions Requirements on the genuine intentions of a
speaker in order for an utterance to count as a particular speech
ac. jI]
social deixis Forms used to indicate relative social status. [10]
solidarit strateg Pemphasis on the closeness of speaker and
addressee. jj
spatial deixis Forms used to point to location, e.g. 'here', 'there' :
cf. temporal deixis. j9]
speech act P action performed by the use of an utterance to
communicate. j47j
t j q GLOS SARY
speech event A sct ot circumstanccs in which pcoplc intcract in
somc convcntional way to arrivc at somc outcomc. j47, 7J
structural presupposition 1hc assumption that part ot a struc-
turc contains intormation bcing trcatcd as alrcady known. j2J
syntax 1hc study ot thc structurcs connccting linguistic torms.
j4J
tautolog An apparcntly mcaninglcss cxprcssion in which onc
word is dchncd as itsclt, c.g. ' busincss is busincss' . jJ
temporal deixis orms uscd to point to location in timc, c. g.
'now', 'thcn' . ct. spatial deixs. j9J
textual function 1hc usc ot languagc in thc crcation ot wcll-
tormcd tcxt. jJ
Transition Relevance Place (RP) A possiblc changc ot spcakcr
point in an intcraction. j72J
turn 1hc opportmity to spcak at somc point dming a convcrsanon.
j72J
turn-taking 1hc changc ot spcakcr during convcrsation. j 72J
TN distincion ^ distinction bctwccn torms uscd tor a tamiliar
{ ' tu' } and a non-tamiliar ( 'vous' ) addrcsscc, in rcnch and othcr
languagcs. jIUJ
zero anaphora 1hc abscncc ot an cxprcssion in a structural slot
whcrc onc is assumcd, as a way ot maintaining rctcrcncc, c.g.
'Mary mowcd thc lawn and thcn _ watcrcd it. ' j2J
GLOS SARY 1 ]
Acknowl edgements
The author and publisher are grateful to the following for permis
sion to reproduce extracts from copyright material :
Academic Press, Inc. and the authors for extracts from Gerald
Gazdar: Pragmatics. Implicature, Presupposition, and Logical
Form ( 1 979) ; J. L. Morgan: 'Two "types of convention in indirect
speech acts' in Peter Cole ( ed. ) : Syntax and Semantics Volume 9:
Pragmatics ( 1978) .
Basic Books, a division of HarperCollins Publishers, Inc. for an
extract from Robin Tolmach Lakoff: Talking Power: The Politics
of Language, copyright 1990 by Robin Tolmach Lakoff.
Blackwell Publishers for an extract from Harvey Sacks: Lectures
on Conversation ( 1 992) .
Cambridge University Press and the authors for extracts from
Penelope Brown and Stephen Levinson: Politeness ( 1987) ; John
Gumperz and Jenny Cook-Gumperz: ' Introduction: language and
the communication of social identity' in J. Gumperz ( ed. ) :
Language and Social Identity ( 1 982) ; John Searle: Speech Acts
( 1969) .
Elsevier Science Ltd. , The Boulevard, Langford Lane, Kidlington
OX5 1GB, UK for an extract from Gabriele Kasper: 'Politeness' in
Ron Asher (ed. ) : The Encyclopedia of Language and Linguistics
Volume 6 ( 1994) , copyright 1994.
Lawrence Erlbaum Associates, Inc. and the author for an extract
from Georgia Green: Pragmatics and Natural Language
Understanding ( 1 989) .
ACKNOWLEDGEMENTS 1 3 7
Char|cs Fi||morc for an cxtract from Char|cs Fi||morc. Santa
Cruz Lectures on Deixis Indiana Univcrsity Linguistics C|ub
t;y .
Kathlccn Gricc for an cxtract from Pau| Gricc. 'Logic and convcr-
sation' in P. Co|c and |. Morgan cds. . Syntax and Semantics
Volume j : Speech Acts t;y .
Kluwcr Acadcmic Pub|ishcrs for cxtracts from Gcoffrcy
unbcrg. 'Indcxicality and dcixis' in Linguistics and Philosophy
t 6 t j , Qucntin Smith '1hc multiplc uscs of indcxicals' in
Synthese ;8 t8 .
1hc Linguistics Socicty of Amcrica for cxtracts from Harvcy
Sacks et al. : 'A simplcst systcmatics for thc organization of turn-
taking in convcrsation' in Language yo t;{ .
Longman Group Ltd. for cxtracts from M. A. K. Ha||iday and
Rugaiya Hasan. Cohesion in English t;6 , Gcoffrcy Lccch.
Principles of Pragmatics ( I 8j .

Wm. Morrow for an cxtract from Dcborah 1anncn. You Just
Dont Understand to .
cw York Univcrsity Prcss lor an cxtract from Robcrt C.
Sta|nakcr. 'Pragmatic prcsupposition' in Mi|ton Munitz and
Pctcr Ungcr cds. . Semantics and Philosophy t;{ .
1hc Ohio Statc Univcrsity Prcss for an cxtract from. ' Pragmatics.
mcaning and contcxt' Fi|c ;o in Language Files: Materials for an
Introduction to Linguistics 6th cdn. , thc Ohio Statc Univcrsity
Dcpartmcnt of Linguistics, cditcd by Stcfanic |anncdy, Robcrt
Polctto, and 1raccy L. Wc|don, copyright t{ by thc Ohio
Statc Univcrsity Prcss. A|l rights rcscrvcd.
Oxford Univcrsity Prcss for an cxtract from |cnny 1homas. 'Cross
cultural pragmatic fai|urc' in Applied Linguistics {lz t8j .
Plcnum Publishing Corporation and thc author for an cxtract
from |ack ilmcs. Discourse and Behaviour t8 6 .
Dcspitc cvcry cffort to tracc and contact copyright ho|dcrs bcforc
publication, this has not a|ways bccn possiblc. If notificd, thc
publishcr wi|| bc plcascd to rccti any crrors or omissions at thc
carlicst opportunity.
j 8 ACKNOWLEDGEMENTS