CONTENTS

CHAPTER 1
PRAGMATICS AND OTHER AREAS OF LINGUISTIC INQUIRY

1.1 Introduction 8
1.2 Syntax – semantics – pragmatics 9
1.3 The American vs. the Continental approach to pragmatics 11

CHAPTER 2
DEIXIS

2.1 Introduction 14
2.2 Deictic expressions in semantic theory 15
2.3 Types of deixis 18
2.3. 1 Person deixis 19
2.3.2 Time deixis 21
2.3.3 Spatial deixis 23
2.3.4 Discourse deixis 24
2.3.5 Social deixis 24
2.4 The deictic centre 26
2.5 Deictic usages 27
2.6 Non-deictic usages 28
2.7 Conclusions 29

CHAPTER 3
PRESUPPOSITION

3.1 Introduction 30
3.2 Theories of presupposition 31
3.2.1 The Frege-Strawson tradition 31
5
3.2.2 Pragmatic presupposition 33
3.3 Types of semantic presupposition 34
3.4 Presupposition triggers 37
3.5 Projection 39
3.6 Cancellability 42
3.7 Current issues in presupposition theory 45
3.7.1 Local contexts 45
3.7.2 Presupposition and anaphora 49
3.7.3 Accommodation 51
3.7.4 Presupposition failure 54
3.8 Conclusions 56

CHAPTER 4
IMPLICATURE

4.1 Introduction 57
4.2 Natural vs. non-natural meaning 58
4.3 The Cooperative Principle and the maxims of conversation 60
4.4 Generalized vs. particularized conversational implicatures 62
4.5 Properties of conversational implicatures 65
4.6 Scalar implicatures 68
4.7 Conventional implicatures 73
4.8 Relevance 77
4.8.1 Ostensive-inferential communication 78
4.8 2.The concept of relevance 79
4.8 3 Implicatures and explicatures 81
4.9 Conclusions 83

CHAPTER 5
THE THEORY OF SPEECH ACTS

5.1 Introduction 84
5.2 Austin’s theory of speech acts 85
5.2.1 Locutions, illocutions, and perlocutions 85

6
5.2.2 Felicity conditions 88
5.2.3 The performative formula 89
5.2.4 Austin’s revised approach to
the constantive vs. performative distinction 91
5.3 The influence of Grice 93
5.3.1 Strawson’s objection to Austin 93
5.3 2.Searle’s defense of Austin 94
5.4 Indirect speech acts 98
5.4.1 Indirect speech acts and politeness 100
5.4.2 Indirect speech acts and sentence type 101
5. 5 Conclusions 102

CHAPTER 6
CONVERSATION ANALYSIS

6.1 Introduction 103

6.2 Turn-taking organization and the local management system 106
6.3 Turn-taking irregularities 110
6.3.1. Timing in turn-taking theories 111
6.3.2 Silences in conversation: pauses, gaps, and lapses 112
6.3.3 Overlapping talk 113
6.4 Adjacency pairs 115
6.5 Pre-sequences 116
6.6 Repair 119
6.7 Tying techniques 120
6.8 The monitoring of conversation 121
6.9 Accounts for conversation 129
6.10 Conclusions 130
A glossary of pragmatics 131
References 158

7
 CHAPTER 1
PRAGMATICS AND OTHER AREAS OF
LINGUISTIC INQUIRY

1.1 Introduction
If you want to know what a particular human activity is all about, you may
ask questions like: ‘What are the rules of tennis?’ or ‘What is cricket like?
And then you will get to the point where you want to figure out what these
sports enthusiasts are actually doing when playing out there in the field.
Similarly, if you want to know what a particular religion is all about, you are
of course entitled to ask what its beliefs are; but you will be more interested
in, and enlightened by the practices that are said to be characteristic of such a
religion. A similar line of reasoning could be applied to a religion, a system
of beliefs or any human endeavour.
Any pragmatician who is asked what pragmatics is all about will
answer that it is a science that has to do with language and its users. But in
order to have an understanding of what pragmatics really stands for, one
must try and find out how the game is played, what pragmaticians do for a
living and what makes them different the people working in other related
branches of language studies such as syntax or semantics. So the relevant
question that arises is ‘What could be called a typical pragmatic perspective
on matters of language?’
To get an idea of how pragmatics works consider the example in (1)
taken form 21 August, 1992 issue of The Chicago cultural weekly Reader.
This particular issue of the magazine included an advertisement for a
downtown cocktail lounge ‘Sweet Alice’. The ad carried the text given in
(1):
(1) I brought some sushi home and cooked it, and it wasn’t bad at all.
What can be made of the example in (1)? If we take a look at the syntax of
the sentence, the sentence is well-formed. However, there is something
funny about the sentence. It is obvious that the sentence is intended as a joke
since everybody knows that sushi is eaten raw, and that you are not supposed
to cook it. Cooking sushi may strike someone as funny or stupid or
outrageous, depending on one’s point of view. In an informal way, the
8
sentence above makes no sense. And a linguist familiar with other bizarre
sentences might argue that since everybody knows that sushi is defined as
being eaten raw, a sentence such as the above is wrong in the same way as
are sentences of the type ‘Colourless green ideas sleep furiously’ which
made a certain American linguist famous in the early sixties. The linguist
might go on to argue that the sentence is semantically wrong; it does not
make sense because the semantics of one its parts (the sushi) clashes with the
semantics of another part (the cooking).
But, then, why should such a silly sentence be used in advertisement
for a cocktail bar? This is where pragmatics sheds new light. Pragmatics tells
us it is alright to use language in surprising unconventional ways, as long as
we know, as language users, what we’re doing. As language users we can let
ourselves be semantically shocked if there is a reason for it, or if it serves a
purpose.
In this particular case, the shocking semantics is a reflection of a
euphoric effect. It can be interpreted as invoking the silly state of mind that
becomes our privilege after a couple of drinks. This is precisely the reason
why the ad felicitously achieves its illocutionary force (i.e. it is effective as
an invitation to join the crowd at Sweet Alice’s) and has appropriate
perlocutionary effects on the audience (transmitting a feeling of euphoria).
Thus, pragmatics is where action is. But what is the action that goes
with the above ad? Obviously the ad in (P) is an attempt to ‘sell’ something:
a cocktail bar, a particular ambiance, a particular clientele, a promise of good
times. The ad invites us in, so to speak. But it doesn’t do that by saying
‘Come into my parlour/or cocktail bar’, such an invitation would be too
blatant to be effective or to arise our interest. It talks to us in a voice that
appeals to us as individual language users with a particular history and
understanding of the world and living within a particular context. The
parlour or the cocktail bar is sketched as a desirable place and the invitation
is by innuendo only: a pragmatic act of inviting.

1.2 Syntax – semantics – pragmatics

We owe the modern usage of the term pragmatics to the philosopher Charles
Morris who outlined the framework of a science of signs – i.e. semiotics –
with a view to unifying logical positivism with behavioural empiricism and
pragmatism. One of the central tenets of theory of signs is the view that
symbols have three types of relations:
1. to objects,
2. to persons, and
9
 3. to other symbols.
He later called these relations semantics, pragmatics, and syntactics
(Posner 1987: 25). Within his theory of semiotics, Morris is concerned with
explaining the tri-relation between syntactics, semantics, and pragmatics in a
dyadic way, which is very different from the semiotics of C. S. Peirce. He
distinguished three inter-related sciences or branches linguistic inquiry
(Morris 1946):
- Syntactics (i.e. syntax) – the study of the formal relation of signs to one
another. In other words, syntax is the study of the relationships between
linguistic forms, how they are arranged in well-formed sequences and which
sequences are well-formed. This type of study generally takes place without
considering any world of reference or any user of the forms.
- Semantics – the study of “the relations of signs to the objects to which the
signs are applicable (their designata)”. Thus, semantics is the study of the
relationships between linguistic forms and entities in the world, i.e. how
words literally connect to things. Semantic analysis also attempts to establish
the relationship between verbal description and states of affairs in the world
as accurate (true) or not (i.e. false), regardless of who produces that
description. In this regard, we can speak of semantics as being truth-
conditional.
- Pragmatics – the study of “the relation of signs to interpreters”. Pragmatics
is the study of the relationships between linguistic forms and the users and
interpreters of those forms. According to Morris’ theory of signs, the user is
seen in a wider perspective, since the human being, as interpreter of signs,
does not only have a mental identity, but also a biological and social identity
which affects his interpretations of signs.
It is a sufficiently accurate characterization of pragmatics to say that
it deals with the biotic aspects of semiosis, that is, with all the
psychological, biological, and sociological phenomena which occur
in the functioning of signs. (Morris 1938: 108)
Pragmatics is the study of speaker meaning. Pragmatics is concerned
with the studying of meaning as communicated by a speaker and interpreted
by a listener (or reader). It has more to do with the analysis of what people
mean by their utterances rather than what the words or phrases that occur in
those utterances might mean by themselves.

10
 Pragmatics is the study of contextual meaning. In other words,
pragmatics 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 intend to say in
accordance with such aspects of the speech event as who they are talking to,
where, when and under what circumstances.
Pragmatics is the study of how more gets communicated than is said.
This branch of linguistic investigation also explores the way addressees draw
inferences about what is said in order to arrive at an interpretation of the
speaker’s intended meaning.
Pragmatics is the study of the expression of relative distance. The
choice between what is said and what remains unsaid may be tied to the
notion of distance, not necessarily distance in space, but also social distance.
Closeness whether it is physical, social or conceptual implies shared
experiences and a more or less similar world view. Taking into account how
close or distant the addressee is, speakers determine how much needs to be
said (Mey 1993).

1.3 The American vs. the Continental approach to pragmatics

Since Morris’s introduction of the trichotomy syntax, semantics, and
pragmatics, the last term has come to be used in two very distinct ways.
Firstly, according to the American approach, pragmatics is the study of
those relations between language and context that are grammaticalized, or
encoded in the structure of language. In other words this approach deals with
the expression of pragmatic relations with the help of strictly linguistic
means, such as the rules of grammar operating on phonological,
morphological and syntactic elements. This is a strictly linguistically
oriented definition. Such a scope for pragmatics would include the study of
deixis and probably the study of presuppositon and speech acts. It would
exclude the study of principles of language usage that could not be shown to
have an impact on the grammar of languages. Thus the extremely important
pragmatic concept of conversational implicature and the principle of
politeness would lie outside the scope of this approach to pragmatics.
However, such a scope for pragmatics has the advantage of drawing a clear-
cut distinction between pragmatics and neighbouring fields like
sociolinguistics and psycholinguistics.
Secondly, within the Continental approach to pragmatics a theory
of language as a user’s interest should rest on a theory of the user. Since the
user is a member of a particular human society, such a theory should
comprise everything that characterizes the user as a person whose use of
11
language depends on the rules and norms that are valid at any time, in any
place in the community in which he/she is living. This is a return to the view
of pragmatics prposed by Morris that “pragmatics is about everything human
in the communication process, psychological, biological and sociological”.
This definition, which is still generally used on the Continent, incorporates
as much societal context as possible. Within this theoretical framework
sociolinguistics or conversation analysis, for instance, become part of
pragmatics, being, as some linguists argued, applied pragmatics.
In this book we shall assume, for working purposes, that pragmatics
equates with a theory of language understanding that takes context into
account, in order to complement the contribution that semantics makes to
meaning. In other words, pragmatics is the study of all those aspects of
meaning not captured in a semantic theory. However, one objection to such a
definition could be that the scope of pragmatics would vary according to the
kind of semantic theory adopted.
At this point one should attempt a conceptual clarification and try to
delimit the scope of pragmatics by a boundary drawing exercise. Thus the
upper bound of pragmatics is provided by the borders of semantics and the
lower bound by sociolinguistics (and perhaps psycholinguistics, too)
Given the difficulties of drawing a neat dividing line between
semantics and pragmatics, the best strategy seems to restrict semantics to
truth-conditional content. Pragmatics is meaning minus truth conditions. As
far as the lower bound is concerned, i.e. the border between pragmatics and
sociolinguistics, here things are even more problematic. Drawing a boundary
between sociolinguistic and pragmatic phenomena is an extremely difficult
enterprise. Let us consider an instance of sociolinguistic phenomena and ask
how it falls with respect to 2 of our definitions of pragmatics, the most
restrictive definition (i.e. the American approach) and the broadest one (i.e.
the Continental approach).
Consider honorifics most simply exemplified by the polite singular
pronoun of address in some European languages (i.e. the T/V distinction). If
we take the view that pragmatics is concerned only with grammatically
encoded aspects of the context, then we might propose a tidy division of
labour between pragmatic and sociolinguistic accounts of honorifics:

a. Pragmatics would be concerned with the meaning of honorifics (e.g.

with the specification that V encodes that the addressee is socially
distant or superior)
b. Sociolinguistics would be concerned with the recipes for the usage of
such items (e.g. the specification that among some segment of the
12
speech community, V is used to aunts, uncles, teachers, etc.). If we
take pragmatics to be the study of the contribution of context to
language understanding, consider what happens when an aunt that
normally gives her nephew T, switches on one occasion to V. In
order to predict and account for the intended ironic or angry meaning
a pragmatic theory must have available the detailed recipe for usage
that tells us that V is not the normal usage, and thus not to be taken at
face value. So on this broader scope for pragmatics, pragmatic
accounts of language understanding will at least need access to
sociolinguistic information. It is from this point of view that we can
consider sociolinguistics to be applied pragmatics.

13
 CHAPTER 2
DEIXIS

2.1 Introduction
The term deixis comes from the Greek word for ‘pointing’. Its equivalent
philosophical term indexicality comes from the corresponding Latin word.
Deixis is the phenomenon whereby some linguistic expressions are
systematically dependent on the context for their interpretation. Consider the
utterance in (1):
(1) Put this book over there.
Establishing the identity of the referent, i.e. which book is being referred to,
and which place it is to be put, depends on features of the context outside the
utterance itself. In this particular case the features of the context are typically
gestures pointing to the referent of this and there.
A similar point could be argued about the personal pronoun in the
example in (2):

(2)

A: Who’s there?
B: It’s me.

The utterance ‘It’s me’ is always true but is totally uninformative when it
comes to establishing the speaker’s identity. The referent of the pronoun I
changes with the person uttering it. Everybody can say I and whoever says it
points to another object than everybody else. Just the same point could be
made about first- and second-person pronouns we and you, about
demonstratives and specific time and place adverbs like now, here or there,
to mention just a few. These lexical items are referred to as deictic
terms/indexical expressions or simply indexicals. They are a particular kind
of referential expressions where the reference is not just semantic but
includes a reference to a particular context in which the semantics is put at
work. In other words, one should always refer to the context if in order to
establish the proper reference of deictic words.

14
 In this chapter the terms ‘deixis’ and ‘indexicality’ will be used co-
extensively, since they simply come from different traditions (Bühler 1934
and Peirce 1955) and have become associated with linguistic and
philosophical approaches respectively. However, a technical distinction will
be made: indexicality will be used to label the broader phenomena of
contextual dependency, and deixis the narrower linguistically-relevant
aspects of indexicality.

2.2 Deictic expressions in semantic theory

Deictic expressions are one of the abiding puzzles in the philosophy of

language. Expressions like today have a constant meaning, but on the other
hand their referent constantly changes (since the reference of today will
always be different tomorrow). To a certain extent, they resemble proper
names due to their little descriptive content and resistance to paraphrase. On
the other hand, deictic expressions differ from proper names in their
constantly changing reference (Kaplan 1989:562).
In philosophical approaches to semantics, handling indexical
expressions is a two-stage process involving a mapping from contexts into
propositional contents, followed by a mapping from, say, worlds to truth-
values. In Montague’s (1974) early theory the content of deictic expressions
was captured by mapping contexts (reduced to a set of indices for speakers,
addressees, indicated objects, times and places) into intensions. In Kaplan’s
(1989) theory, all expressions have this characteristic mapping (their
character) from contexts into intensions (their proposition-relevant content),
but only indexicals have variable character, which can be thought of as their
meaning. Thus the meaning of I is its character, which is a function or rule
that variably assigns an individual concept, namely the speaker, in each
context (Kaplan 1978). Non-indexical expressions have constant character,
but may (rigid designators) or may not (other referring expressions) have
constant content, as illustrated in (3) below:

(3)

constant character variable character

constant content proper names indexicals

variable content definite descriptions ‘deferred ostension’

15
 Situation semantics (Barwise and Perry 1983) provides us with
another version of the two-stage theory. Within the framework of situation
semantics utterances are interpreted with respect to three situations (or states
of affairs): the utterance situation (corresponding to Montague’s indices), the
‘resource situation’ (which handles other contextually determined reference
such as anaphora) and the ‘described situation’ (corresponding to the
propositional content). Indexicals get their variables fixed in the utterance
and/or resource situations. The value of the variables (e.g. the referent of I or
that) is then transferred to the described situation (e.g. “I gave him that
book” has the described content, say, ‘Bill gave him that book’). Meaning is
relational, and the meaning of an indexical is characterized as the relation
between utterance/resource situations and described situations.
The central tenet of two-stage theories is that indexicals do not
contribute directly to the proposition expressed, the content of what is said,
or the situation described. Instead, they is take us to an individual, a referent,
which is then slotted into the proposition expressed or the situation
described, or as Nunberg (1998:159) puts it: “The meanings of indexicals are
composite functions that take us from an element of the context to an
element of a contextually restricted domain, and then drop away”.
The literature lists several empirical properties of indexicals. Firstly,
Wettstein (1984), among others, argued that pure indexicals such as I, now,
here have their semantico-pragmatic content exhausted by a specification of
the relevant index (speaker, time and place of speaking respectively).
However, if one considers closely related indexicals such as we, today,
nearby, it becomes apparent that these deictic expressions convey additional
semantic conditions (“at least one person in addition to the speaker”, “the
diurnal span which contains the coding time”, “a place distinct from here but
close to here”, respectively). Thus, deictic expressions may exhibit both
descriptive properties and contextual variables. In addition, nearly all deictic
expressions heavily depend on pragmatic resolution (Levinson 2000:177ff).
Consider the example in (4):

(4) Come here

The place deictic item may refer, say, to this sofa or to this city according to
context.
Secondly, Cresswell (1973:111ff) points out that the relevant
contextual features cannot be determined in advance. Consider the example
in (5):

(5) This is the largest walnut tree on the planet.

16
In various contexts, the speaker can uniquely establish the referent of this by
pointing to a tree some distance away, or while standing underneath it, or by
touching a picture in a book, or, if the addressee were blind, by running the
addressee’s hand over the bark. The mode of demonstration, if any, just does
not seem to be determined in advance.
Thirdly, there is the feature which Quine called ‘deferred ostension’
and which has become familiar especially through the work of Nunberg
(1978, 1998). Suppose the participants in the situation of utterance are
listening to a program on a radio station and one of them says:

(6) CNN has just bought this.

In delivering the utterance in (6) in that particular context, the speaker

doesn’t refer to the current program but the radio station.

Similarly, in a context in which the speaker points to a Coca-Cola bottle and

says:

(7) That used to be a different shape.

In (7) the speaker doesn’t refer to the current bottle, but the type of container,
and asserts that tokens used to be of a different shape. Thus, in both (6) and
(7) the object the speaker is pointing to is not the object referred to.
Another empirical feature would be that third-person, non-deictic
expressions can have indexical uses, as when the speaker says, pointing to a
man in a black tuxedo, “He is President Linton’s nephew”.
Indexicals have projection properties which follow from the fact that
demonstratives and many other deictic expressions have no substantial
descriptive content, so that once the contextual parameters have been fixed
they are ‘directly referential’ (Kaplan 1990). A true demonstrative remains
transparent in an intensional context. Consider the utterance in (8):

(8) John said he broke that

In (8) the referent of that can only be identified as the object the speaker is
pointing at, at speaking time, and not the object John pointed at.
Deictic expressions cannot be assigned attributive meanings.
Compare the utterances in (9):

(9a) The man who can lift this sword is our king.
17
(9b) That man who can lift this sword is our king.

The utterance in (9a) has both referential and attributive reading (i.e.
“Whoever can lift this sword is our king”). By contrast, the utterance in (9b)
has only a referential reading.

Finally, deictic expressions do not generally fall under the scope of negation
or modal operators. The utterance in (10) cannot be understood as ‘I am not
indicating X and X is a planet’ (Enç 1981).

(10) That is not a planet.

These treatments of indexicality considered so far demonstrate that there is a

clear class of indexical expressions, which have an inbuilt variable whose
value is instantiated in the context.

2.3 Types of deixis

In the European linguistic and philosophic traditions one usually mentions

the categories of person, place and time in this connection. The explanation
for this tripartition is that all indexing/pointing is done by living human
beings and therefore all indexical expressions have to be related to:

- the person who has uttered them

- pointing in a particular place
- and at a particular time.

Before embarking on a detailed analysis of the major categories of person,

time, place, social and discourse deixis, let us consider the lesser-known
categories of empathic and reminder or recognitional deixis.
Emphatic deixis concerns the use of this in context where the first
option would be that to signal empathy. Similarly, it involves the shift from
this to that to signal emotional distance (Lyons 1977). Consider the example
in (11):

(11) Tell that bastard to shut up!

Pragmaticians (Mey 1993; Horn and Ward 2004) also mention the
case of reminder or recognitional deixis. Consider de examples in (12):

(12a) I met this girl the other day.

18
(12b) Do you remember that holiday we spent in the rain in Devon?

In the utterance in (12a) this girl is used to refer to ‘a certain young lady’
whose identity needs no further introduction because either her identity is of
no interest to the story, or because her identity has already been established
in some other way.

2.3. 1 Person deixis

The grammatical category of person directly reflects the different roles that
individuals play in the speech event: speaker, addressee and audience. When
these roles shift in the course of conversational turn-taking the deictic centre
shifts with them (hence Jespersen’s 1922 term shifters for deictic
expressions): A’s I becomes B’s you, A’s here becomes B’s there and so
forth.
Person deixis concerns the encoding of the role of participants in the
speech event in which the utterance is delivered. The category first person is
the grammaticalization of the speaker’s reference to himself, the second
person is the encoding of the speaker’s reference to one or more addresses1,
the third person is the grammaticalization of the speaker’s reference to
persons and entities that are neither speakers nor addresses of the utterance in
question2. The traditional paradigm of first, second and third persons can be
captured by the two semantic features of speaker inclusion [S] and addressee
inclusion [A]: first person [+S], second person [+A, -S], and third person [-
S,-A], which is therefore a residual, non-deictic category. Most languages
directly encode these participant-roles in pronouns and/or verb agreement, as
well as vocative. The majority explicitly mark third person [-S, -A].
Although the traditional notions 1st, 2nd and 3rd persons hold up
remarkably well, there are many kinds of homophony, or different patterns
of syncretism, across person paradigms (Cysouw 2001). Much of this
complexity is due to how the notion of ‘plurality’ is conceptualized within
the paradigm: first-person plural clearly does not entail more than one person
in [+S] role. ‘We’ notions are especially troubling, since many languages
distinguish such groups as: [+S+A] vs. [+S+A+O] (where O is Other, i.e. one

1
There are also exceptions to the alleged universality of 1st and 2nd person marking. For
instance, in some S. E. Asian languages like Thai there are titles (on the pattern of ‘servant’
for 1st person, ‘master’ for 2nd person) used in place of pronouns and there is no verb
agreement (Cooke 1968).
2
Some languages have no third person pronouns, although they often indirectly mark third-
person by zero agreement markers. Yélî Dnye, a Papuan, i.e., non-Austronesian language, is
a case in point.
19
or more persons that are neither speakers nor addresses), vs. [+S-A], vs. +S-
A+O. In some pronominal systems ‘plural’ can be neatly analyzed as
augmenting a minimal deictic specification with ‘plus one or more additional
individuals’ (AUG). Thus the distinction between I and We might be
analyzed as (+S,-Aug), (+S,+AUG). In some contexts, the English personal
pronoun we may be ambiguous between an inclusive (i.e. including the
addressee) or an exclusive (i.e. excluding the addressee) interpretation.
Most languages evince a number of intriguing details that show that
the roles of speaker and addressee roles can be motivated by such
grammatical detail (Levinson 1983; Goffman 1981). Consider the utterance
in (13):

(13) Bill is to get ready now.

The utterance in (13) is appropriately said to a person who will then run
along and tell Bill “Get ready now!”3. In such a scenario, the speaker is
analytically dividing the notion of an ‘addressee’ into two distinct sub-roles:
a person actually spoken to by the speaker, and an illocutionary target of the
utterance, who, as with any imperative, is expected to perform the action. In
a similar way, some languages have specific ways of indicating that the
speaker is merely the mouthpiece for someone else, thus distinguishing the
actual speaking role from the illocutionary source of the message. Compare
the utterances in (14):

(14a) You are to hand in your paper now.

(14b) Hand in your paper now.

The imperative form in (14a) indicates that the speaker is not the originator
of the message, but rather the speaker conveys the message on someone
else’s behalf.
Another important phenomenon related to person is the whole field of
honorifics, which typically make reference to speaking and recipient roles,
but which will be dealt with under social deixis. Last but not least, in some
contexts third-person pronouns are distal forms in terms of person deixis;
using a third-person pronoun where a second-person one could be possible is
a way of communicating distance, not necessarily spatial, but emotional and
social distance.

3
Some languages have a category of 3rd person imperative covering such a scenario.

20
2.3.2 Time deixis

Within the deictic centre the central time, i.e. the temporal ‘ground zero’, is
the moment at which the utterance is issued or the ‘coding time’ in
Fillmore’s (1997) terminology. Time deixis concerns the encoding of
temporal points or spans relative to the time at which an utterance was
spoken, that is relative to coding time. Time deixis is grammaticalized in the
system of tenses deictic adverbs of time, such as now, yesterday, this week,
last year, etc.
Thus, now means some span of time including the moment the
utterance is delivered, today refers to that diurnal span in which the speaking
event takes place. Similarly we count backwards from coding time in
calendrical units in such expressions as yesterday, three weeks ago, or last
years or forwards in tomorrow, or next Friday. These deictic expressions of
time depend for their interpretation on knowing the coding time. In written
or recorded uses of language, a distinction is made between coding time and
receiving time4.
If we don’t know the coding time of a note on an office door such as
the one given in (15) we won’t know if we have a short or a long wait ahead:

(15) Back in an hour.

.
In English, units of time measurement may either be fixed by reference to the
calendar, or not. Consider the utterance in (16):

(16) He’ll do it this week.

The deictic expression this week may refer to a span of seven days from
utterance time or to the calendar unit beginning on Sunday (or Monday)
including utterance time. Similarly, this year means the calendar year
including the time of utterance, but this May tends to mean the next monthly
unit so named (or alternatively, the May of this year, even if past), while this

4
In some languages there are often conventions about whether one writes ‘I am writing this
today so you will receive it tomorrow’ or something more like ‘I wrote this yesterday so that
you receive it today’.

21
morning refers to the first half of the diurnal unit including coding time, even
if that is in the afternoon (Fillmore 1975).
The most pervasive aspect of temporal deixis is the system of tenses.
The grammatical categories called tenses usually encode a mixture of deictic
time distinctions and aspectual distinctions. Linguists tend to set up a series
of pure tense grammatical distinctions that roughly correspond to the extra-
linguistic time distinctions, and then catalogue the discrepancies (cf. Comrie
1985:18ff). For example, one might gloss the English present tense as
specifying that the state or event holds or is occurring during a temporal span
including the coding-time; the past tense as specifying that the relevant span
held before coding-time; the future as specifying that the relevant span
succeeds coding-time; the preterite (as in He had gone) as specifying that the
event happened at a time before an event described in the past tense. It is
clear that there is a deictic temporal element in most of the grammatical
distinctions linguists call tenses, although the system of tenses captures only
partially the English usage (The soccer match is tomorrow; John will be
sleeping now; I wondered whether you were free now, etc.). Tenses are
traditionally categorized as ‘absolute’ (deictic) versus ‘relative’ (anterior or
posterior to a textually specified time), so that the simple English past (He
went) is absolute and the preterite (He had gone) is relative (anterior to some
other, admittedly deictically specified, point). Absolute tenses may mark two
(e.g. past vs. non-past) or they may mark up to nine distinct spans of time
counted out from coding-time (Comrie 1985).
The interpretation of tenses often involves Gricean implicatures5. The
example in (17a) implicates that that he no longer does so, although this is
clearly defeasible as shown in (17b):

(17a) Believe it or not, Mike used to teach semantics.

(17b) Believe it or not, Mike used to teach semantics and in fact he still has
to do so.

Many deictic expressions in the temporal domain are borrowed from

the spatial domain. In English, the temporal prepositions and connectives
like in (e.g. in the morning), on (e.g. on Tuesday), at (e.g. at 5.00 p.m.),
before and after, are all derived from spatial descriptions. The
demonstratives in English follow the same pattern (cf. this week), and in
some languages (Anderson and Keenan 1985:298) ‘here’ and ‘there’ are the
sources for ‘now’ and ‘then’.

5
See Levinson (2000: 95) for a relevant framework of analysis, and Comrie (1985) for the
role of implicature in the grammaticalization of tense.
22
2.3.3 Spatial deixis

Place deixis concerns the encoding of spatial locations relative to the

location of the participants in the speech event. Such deictic adverbs like
here and there are the most direct and most universal examples of spatial
deixis (Diessel 1999:38). Most languages grammaticalize at least a
distinction between proximal (i.e. close to speaker) and distal (i.e.non-
proximal to speaker, sometimes close to addressee). In English, this
distinction is encoded in demonstratives (this, that), the deictic adverbs of
place here (proximal) and there (distal) and in the demonstrative pronouns
this/these and that/those and in some verbs such as come and go. Verbs of
‘coming’ and ‘going’ encode motion to or away from the deictic center6.
Consider the examples in (18):

(18a) Come to bed. – come signals movement towards the deictic

centre/central place, i.e. the speaker’s location
(18b) Go to bed. – go signals movement away from the deictic centre/central
place

However, not all cases when the ‘towards the deictic center’ feature
is lexicalized in verbs of coming are crystal-clear. Firstly, speakers may
project themselves into other locations prior to their actually being in those
locations, as shown in (19):

(19) I’ll come later – come signals movement towards the addressee’s
location

Secondly, there are cases when what they encode turns out to be quite
differentiated (Wilkins and Hill 1995, Wilkins, Hill and Levinson 1995). If
someone comes towards the speaker but stops short before he arrives at the
tree which is near the speaker, the speaker can say ‘He came to the tree’ in
English, but not Italian, where he would say the equivalent of ‘He went to
the tree’.
There is another aspect that is relevant to place deixis. Many analysts
have drawn attention to the ambiguity of the utterance in (20):

(20) The dog is behind the television.

6
However, not all languages lexicalize the ‘towards the deictic center’ feature in their verbs.
23
The dog could be at that side of the television opposite from the screen, or it
could be on the other side of the TV from the speaker, whichever side the
speaker is on. The former interpretation is called the ‘intrinsic’ frame of
reference or perspective in the literature, and the latter mostly ‘deictic’.

2.3.4 Discourse deixis

Discourse deixis concerns the use of expressions within some utterance to

refer to earlier or forthcoming segments of the discourse containing that
utterance. Consider the examples in (21):

(21a) I bet you haven’t heard this story – this story refers to a forthcoming
portion of discourse
(21b) That was the funniest story I’ve ever heard – where that refers to a
preceding portion of discourse.
(21c) “‘You are wrong’. That’s exactly what she said” – that refers to a
preceding portion of discourse.
(21d) “It sounded like this: whoosh” – this refers to a forthcoming portion of
discourse

Since discourse unfolds in time, time-deictic and place-deictic

expressions can be used to refer to portions of discourse, as in: in the
last/next chapter/paragraph, in this chapter, two paragraphs below. Clearly,
in the examples in (21) the reference to parts of a discourse can be
established only on the basis of knowing the current coding point or current
reading/recording. Such reference is quintessentially deictic in character.
An important area of discourse deixis concerns discourse markers,
like anyway, but, however, actually, in conclusion (see Schiffrin 1987).
These relate a current contribution to the prior utterance or portion of text.
For instance, the discourse marker anyway indicates that the utterance that
contains it does not address preceding discourse, but it addresses one or more
steps back.

2.3.5 Social deixis

Social deixis concerns the marking, in linguistic expressions, of the social

relationship holding between speaker and addressees or speaker and some
other referent. This marking is reflected in direct or oblique reference to the
social status or role of participants in the speech event. It is encoded
throughout the morphological system in honorifics and in choices regarding
24
the use of pronouns, summon forms, vocatives and titles of address. It
essential to distinguish a number of axes on which such relations are defined
(Levinson 1983, Brown and Levinson 1987). Table 2.1 summarizes the
relations that typically encoded in social deixis.

Table 2.1. Relational social deixis

Axis Honorific types Other linguistic

encodings

1. Speaker to referent Referent honorifics titles

2. Speaker to addressee Addressee honorifics address forms

3. Speaker to non-addressed Bystander honorifics taboo vocabularies

participant
4. Speaker to setting Formality levels register; diglossia

Table 2.1 shows that the distinction between (1) and (2) is crucial. In (1)
‘honour’ (or a related attitude such as respect, deference) can only be
expressed by referring to the entity to be ‘honoured’, i.e. the referent. In (2),
on the other hand, the same deferential attitude may be expressed while
talking about unrelated matters. Thus, the relation in (2) encodes respect to
the addressee without referring to him. In this scheme, respectful pronouns
like Vous, Sie, dumneavoastră used to when addressing singular addressees
are referent honorifics, which happen to refer to the addressee. In S. E. Asian
languages the elaborate honorifics systems are built up from a mixture of (1)
and (2). Examples include humiliative forms replacing the first person
pronoun (on the principle that lowering the self raises the other) honorific
forms for referring to the addressee or third parties (both referent honorifics),
and suppletive forms for such verbs as eating or going, which give respect to
the addressee regardless of who is the subject of the verb (see Brown and
Levinson 1987, Errington 1988, Shibatani 1999)7.
The relation between speaker and bystander, given in (3), is encoded
in‘bystander honorifics’, used to convey respect to a non-addressed party
who is, nevertheless, present in the interaction. In some Australian languages
there are taboo vocabularies used in the presence of real or potential in-laws,
or those who fell in a marriagable section for ego but were too close to marry
7
See Errington (1988), Agha (1993) for a detailed account of the ten-level of Javanese
etiquette.

25
(Dixon 1980:58-65, Haviland 1979). The Papuan language Yélî Dnye, a
Papuan language, evinces a similar taboo vocabulary related to in-laws,
especially parents and siblings of the spouse. This involves a replacement
vocabulary for body-parts and items like clothing and baskets associated
with the taboo person, and special indirect ways of referring to such people
in their presence (Levinson 1983).
The relation the relation between speaker and setting, given in 4,
involves respect – or some other special attitude – conveyed to the setting or
event. The German system of address includes the following options Du vs.
Sie and First Name vs. Herr/Frau + Last Name which hold across formal or
informal contexts. By contrast, British English speakers generally switch
from First Name to Title + Last Name according to the formality of the
situation (see Brown and Gilman 1960, Lambert and Tucker 1976). Although
most languages are used differently in formal settings (e.g. eat is replaced by
dine, home becomes residence, etc), in some the distinction formal vs.
informal is firmly grammaticalized in the existence of high and low diglossic
variants, with distinct morphology for formal and literary uses. The relations
mentioned in (1), (2) and (3) concerns relative rank and respect.
The second type of socially deictic information is is not relational but
absolute and concerns the use of (a) the forms reserved for certain speakers,
i.e. authorized speakers (e. g. the British royal we, the Thai the morpheme
khrob as a polite particle that can only be used by male speakers, or the
Japanese Emperor’s special first-person pronoun) and (b) forms reserved for
authorized recipients (e.g. restrictions on most titles of address such as Your
Honour, Mr President).
Levinson (1979) argues that the social deictic contents of honorifics
should be considered to be conventional implicatures overlaid on the
referential content, since the deictic content is not deniable and does not fall
under the scope of logical operators.

2.4 The deictic centre

Deixis is generally, but not invariably organized in an egocentric way, i.e.

the speaker and the referent of I are identical (Levinson 1983). If we think of
deictic expressions as anchored to specific points in the communicative
event, then the unmarked anchorage points that make up the deictic centre
are assumed to be as follows:

- the central person – the speaker

- the central time – the time at which the speaker produces the utterance, i.e.
the coding time
26
- the central place – the speaker’s location at utterance time
- the discourse centre – the point which the speaker is currently at in the
production of his utterance
- the social centre – the speaker’s social status and rank, to which the status
and rank of the addressee(s) or referents is relative

However, deictic words may be used in ways that shift this deictic center to
other participants. This state of affair is referred to as deictic projection
(Lyons 1977) or shifts in point of view (Fillmore 1971). Contrast the
utterances in (22):

(22a) I am going to London.

(22b) I am coming to London.

In (22b) the deictic centre is organized around the addressee, since the verb
come indicates movement towards the deictic centre and the speaker is
clearly outside London. Moreover in some languages distal terms can be
used to distinguish between ‘near to the addressee’ and ‘away from both the
speaker and the addressee’.

2.5 Deictic usages

Peirce pointed out that indexicality involves “the dynamical coexistence” of

an indexical sign with its object of reference. It is normally associated with
linguistic expressions that are semantically insufficient to achieve reference
without contextual support. These expressions have, therefore, as basic a
deictic usage. Filmore (1971) distinguishes between symbolic and gestural
usages. Some deictic expressions such as this city resists a gesture (i.e. they
have symbolic usage), while this finger requires one (i.e. it has a gestural
usage). Filmore (1971) defines deictic terms used in a gestural deictic way as
deictic expressions whose reference depends on an audio-visual-tactical and,
in general a physical, monitoring of the speech event. Consider the examples
in (23) which illustrate a gestural deictic usage:

(23a) This one’s genuine, but this one is a fake.

(23b) He’s not the Duke, he is. He’s the butler.

In the examples above, some kind of selecting gesture or pointed gaze is

required in order for the addressee to correctly identify the referent of the
deictic expressions given in italics.

27
 It should be pointed out that the property of indexicality is not
exhausted restricted to the use of inherently indexical expressions. Just about
any referring expression can be used deictically, as illustrated by the
examples in (24):

(24a) The funny noise is our old dishwashing machine – said pointing chin to
kitchen
(24b) What a great picture! – said looking at a picture

By contrast, symbolic usages of deictic expressions generally require

for their interpretation only knowledge of the basic spatio-temporal
parameters of the speech event or, in the case of person deixis, knowledge of
participant-roles. Thus, it is sufficient to know the location of the participants
in order to interpret (25a), or to know the set of potential addressees in the
situation in order to interpret (25b), or to know the non-addressed participant
in the speech event in (25c), or to know when the interaction is taking place
in order to know which calendar year is being referred to in (23d).

(25a) This city is really beautiful.

(25b) You can all come the party tonight.
(25c) He is my father – said of man entering the room
(25d) They can’t afford a holiday this year.

2.6 Non-deictic usages

In non-deictic usages, the deictic terms are interpreted relative to the text and
not relative to the situation of utterance. Central to non-deictic usage is the
phenomenon of anaphora. An anaphoric usage is when some term picks out
as referent the same entity or class of objects that some prior term in the
discourse picked out. Consider the sentence in (26):

(26) Susan came in and she turned on the music.

Thus, in (26) she is interpreted as referring to whoever it is that Susan refers

to.
However, there are cases when anaphora is so closely linked to deixis
that it is not always separable. Consider the sentence in (27):

(27) I’ve been living in Vienna for 5 years and I love it there.

In (27), here is used both anaphorically and deictically.

28
It is used anaphorically because it refers back to whatever place Vienna. At
the same time it contrasts with here on the deictic dimension of space,
locating the utterance outside Vienna. Similarly, it is possible for gestural
usage to combine with an anaphoric usage, as in (28):

(28) I hurt a finger: this one.

In the utterance in (28), this one refers picks out the same referent as a finger
does, which accounts for its anaphoric non-deictic usage. However, the
utterance should be simultaneously must be accompanied by a presentation
of the relevant finger, if it is to be felicitously used.
Another boundary problem is posed by contexts where indexical
expressions are not so clearly demarcated. Consider the example in (29):

(29) Let’s go to a nearby restaurant.

In the example in (29) nearby is clearly used deictically, since it locates the
utterance close to the speaker’s location. However, in (30) nearby is used
non-deictically since it may be interpreted relative to some preceding portion
of the discourse text or it may presume some point of measurement in the
same way in which the adjective tall is relative to some implicit standard.

(30) Churchill took De Gaulle to a nearby restaurant.

2.7 Conclusions

This chapter has addressed a number of topics that establish deixis as a

central subject in the theory of language. At first sight deixis seems to be a
simple phenomenon, reminiscent of the direct ‘here and now’ relevance of
animal communication systems. However, the intersection of this context-
dependence with the property of abstract symbolic representation in
language leads to deep complexities. Deixis turns out to be very puzzling
both philosophically and psychologically, since deixis introduces context-
dependency into almost every utterance.
Research has focused on understanding how deixis works and on how
it can be incorporated into a formal theory of meaning. However, the
complexity and pervasiveness of deictic phenomena makes it difficult for
most of the theoretical models proposed to account for the richness of the
underlying contextual systems.

29
 CHAPTER 3
PRESUPPOSITION

3.1 Introduction
When engaging in conversation speakers take a lot for granted. In other
words, they presuppose information. Presupposing information applies not
only to spoken discourse, but also to written discourse. As I wrote this, I
presupposed that readers would understand English. But I also presupposed,
as I wrote the previous sentence, repeated in (1), that there was a time when I
wrote it, for otherwise the clause as I wrote this would not have identified a
time interval.
(1) As we wrote this, I presupposed that readers would understand English.
Moreover, I presupposed that readers would be able to identify the reference
of the demonstrative this, i.e., the chapter itself. And I presupposed that there
would be at least two readers, for otherwise the bare plural readers would
have been inappropriate.
The presupposition that an interlocutor would understand English
corresponds to an assumption I made in using English words, but it has
nothing to do with the meanings of any of those words. On the other hand,
the existence of a time when I wrote the chapter is a requirement associated
with the use of a specific word, as. It is a requirement built into the meaning
of the temporal preposition as that in a phrase as X, the X has denote some
time. Thus, the lexical item as is said to be a presupposition trigger.
Similarly, the demonstrative this is a presupposition trigger requiring
something to refer to, the bare plural is a presupposition trigger requiring
existence of multiple individuals, and the modal auxiliary would is a
presupposition trigger requiring a salient future or hypothetical circumstance.
Following Stalnaker (1972; 1974), one can say that the
presupposition that the interlocutor speaks English or the presupposition that
the interlocutor is interested in what the speaker (or writer) has to say is
speaker presupposition or pragmatic presupposition. The presuppositions
associated with specific triggers are said to be conventional or semantic. This
terminological distinction is of theoretical import.
To call presuppositional expressions “conventional” or “semantic” is
not necessarily to imply that the presuppositions they trigger don’t depend on
the context in any way. As we shall in the following sections, the
30
interpretation of conventional presupposition triggers depends on the context
in significant ways. Some theorists regard it as an open question whether
there are any purely conventional presuppositions.
What makes presuppositions of special interest is their ubiquity in
both spoken and written discourse and their behavior that differs
significantly from other aspects of meaning.

3.2 Theories of presupposition

3.2.1 The Frege-Strawson tradition
Many of the issues that are central to the phenomenon of presupposition
were first raised by Frege in his seminal study “On Sense and Reference”
published in 1892. Frege examines that the implication of existence
associated with proper names and argues that it arises with both affirmative
sentences and their negative counterparts.
Frege argues that ‘if anything is asserted there is always an obvious
presupposition that the simple or compound proper names used have a
reference’. If one asserts (2) there is a presupposition, i.e. a piece of
information which taken for granted, that the name Kepler designates
something.

(2) Kepler died in misery.

According to Frege, the presupposition ‘Kepler designates something’ is not
part of the meaning of ‘Kepler died in misery’; if it were then ‘Kepler died in
misery’ would have the logical form ‘Kepler died in misery & Kepler
designates something’ and thus the sentence ‘Kepler didn’t die in misery’
would be equivalent to ‘Kepler didn’t die & the name Kepler has no
reference’. That he argued would be absurd. Therefore ‘Kepler designates
something’ is the presupposition of both ‘Kepler died in misery’ and its
negative counterpart ‘Kepler didn’t die in misery’. In other words, the
implication of existence associated with proper names in affirmative and
negated sentences is an implication of their use.
Frege outlines a theory of presupposition pointing out the following
details of relevance:

(i) Referring phrases and temporal clauses carry presuppositions to

the effect that they do have a referent;

31
 (ii) A sentence and its negative counterpart share the same set of
presuppositions;
(iii) In order for a sentence or assertion to be true or false its
presupposition must be true or satisfied.

The first extended analysis of this type of implication appears in

Strawson (1950) who distinguishes between expressions (simple or complex,
hence including sentences) and uses of expressions. He points out that what
can be said of expressions cannot necessarily be said of uses of expressions,
and vice versa. According to Strawson (1950), sentences are not true or false,
only statements are. In particular, the statement The King of France is wise
may have been true in say 1670 and false a hundred years later, but
nowadays the statement cannot be said to be either true or false; due to the
non-existence of a present King of France the question of its truth or falsity
does not even arise. Strawson claimed that there is a special kind of
relationship between (3) and (4), namely that (4) is a precondition for (3)
being judgeable as either true or false.

(3) The King of France is wise

(4) There is a present King of France

He called this relation presupposition and argued that it is a special species

of inference, distinct from logical implication or entailment, a species which
derives from conventions about the use of referring expressions or definite
noun phrases. More formally, he held that a statement A presupposes a
statement B if and only if B is a precondition of the truth or falsity of A.
Following Frege, Strawson argued positive and negated sentences, when
uttered, will preserve their presuppositions.
Basically a semantic presupposition can be defined as follows:
A sentence A semantically presupposes another sentence B iff
(a) in all situations where A is true, B is true
(b) in all situations where A is false, B is true
(Burton-Roberts 1989)
Frege’s and Strawson’s claim that presuppositions are preserved in
negative sentences and statements provides us with an operational test for
identifying presuppositions generally referred to in the literature as

32
constancy under negation (Levinson 1983). Given a sentence, in order to
identify the presupposition(s), we simply negate it (i.e. negate the verb in a
simple sentence or the verb of the main clause in a complex sentence) and
see what inferences are shared by both the positive and the negative
sentence. Consider the sentence in (5).
(5) John managed to stop in time
The sentence in (5) has the following inferences:
(6) John stopped in time
(7) John tried to stop in time
From the sentence in (8), which is the negation of (5), one cannot infer (6)
since the main point of (8) is to deny (6).
(8) John didn’t manage to stop in time
On the basis of the constancy under negation test and the assumption of its
sufficiency (7) is the presupposition of both (5) and (8), whereas (6) is the
entailment of (5). Thus, negation alters the entailments of a sentence or
statement, but it leaves its presuppositions untouched.
A semantic entailment can be defined as follows:
A semantically entails B (A║—B) if and only if every situation that makes A
true, makes B true i.e. in all worlds in which A is true, B is true.
3.2.2 Pragmatic presupposition

The most significant philosophical counterpoint to the Frege-Strawson

approach to presupposition is due to Stalnaker (1972, 1973, 1974, 1998).
Stalnaker argues that a pragmatic notion of presupposition is needed, so that
the proper object of philosophical study is not what words or sentences
presuppose, but what people presuppose when they are speaking. According
to Stalnaker, the notion of pragmatic presupposition is not inconsistent with
the type semantic presuppositions within the Strawson-Frege tradition. He
points out that whenever something is semantically presupposed, we should
expect it to be pragmatically presupposed, as well, although not necessarily
the other way round. Thus if a definite description semantically presupposes
the existence of a salient referent, then it follows that speakers using definite
descriptions will pragmatically presuppose the existence of such referents.

33
 One consequence of Stalnaker’s view is that presupposition failure
need not produce a semantic catastrophe. There are, however, two weaker
types of failure that can occur: (i) a speaker uttering some sentence S can fail
to assume that some proposition P is in the common ground, even though
most utterances of S would be accompanied by the presupposition that P; and
(ii) a speaker can presuppose something that is not in the common ground.
The former idea was used by Stalnaker to account for some tricky examples
of Karttunen (1971b), involving a subclass of factive verbs that Karttunen
referred to as “semifactives”. The naturally occurring examples in (9a) and
(9b), which involve the (semi-)factive verb know, illustrate the point. The
sentence in (9a) involves a first person, present tense use of know, and there
is clearly no presupposition that Mullah Omar is alive. On the other hand,
(9b) involves a past tense, third person use of know, and in this case it does
seem to be presupposed (at least in the fictional context of the story) that
Luke was alive.

(9a) I don’t know that Mullah Omar is alive. I don't know if he’s dead either.
(General Dan McNeill, Reuters, 19 May 2008)

(9b) Vader didn’t know that Luke was alive, so he had no intentions of
converting Luke to the Sith. (Web example)

Examples like (9) led Karttunen to propose that know only triggers a
presupposition in some person and tense forms; hence the term
“semifactive”.
In the context of his pragmatic approach to presupposition, Stalnaker
argued that these examples are not problematic. In the pragmatic account, the
verb know does not necessarily have to presuppose that its complement is
true. When an addressee hears the sentence in (9a), he will realize that if it
were in the common ground that Mullah Omar was alive, then the speaker
would know this, and so the speaker’s claim would be false. Therefore the
hearer can reason that the speaker is not presupposing the complement of
know to be true. On the other hand, when a hearer hears the sentence in (9b),
it is consistent to assume that Luke was alive. Since speakers using know
typically presuppose the truth of the complement, one can assume that this is
the case here.

3.3 Types of semantic presupposition

1. The existential presupposition is associated with possessive
constructions, definite noun phrases and simple and compound proper
34
names. By using any of these expressions the speaker is assumed to be
committed to the existence of the entities named:
(10a) your car → you have a car
(10b) the King of Spain → the King of Spain exists/there is a present King of
Spain
(10c) the dog → there is a dog
(10d) the girl next door → there is a girl living next door
2. A number of verbs such as know, realize, regret as well as phrases
including be and adjectives such as aware, odd, glad are used with the
implication that the piece of information following them is treated as a fact
and is referred to as factive presupposition:
(11a) She realized he was ill. → He was ill.
(11b) We regret telling him. → We told him.
(11c) I’m glad that it’s over. → It’s over.
3. There are also other forms that give rise to lexical presuppositions. In
lexical presuppositions, the use of one form with its asserted meaning is
conventionally interpreted with the presupposition that another non-asserted
meaning is understood. Each time we say that someone managed to do
something the asserted meaning is that ‘the person succeeded in some way’.
When we say that someone didn’t manage the asserted meaning is that ‘the
person didn’t succeed’. In both cases, however, there is a presupposition (i.e.
a piece of information which is non-asserted, but nevertheless taken for
granted) that ‘the person tried to do something’. Thus, managed is
conventionally interpreted as asserting ‘succeeded’ and presupposing ‘tried’.
Other examples of lexical items that trigger lexical presuppositions include
stop, start, forget, again:
(12a) He stopped smoking. → He used to smoke.
(12b) They started complaining. → They weren’t complaining before.
(12c) You are late again. → You were late before.
4. So far, we have considered presuppositions associated with the use of
lexical items and phrases. However, we can also identify the class of

35
structural presuppositions which are triggered by sentence structures. In
this case, certain sentence structures are interpreted conventionally and
regularly presupposing that part of the structure is already assumed to be the
case. For example, a wh- interrogative construction is conventionally
interpreted with the presupposition that the information after the wh- form is
assumed to be true.
(13a) When did he leave? → He left.
(13b) Where did you buy that book? →You bought the book.
(13c) Who is the professor of linguistics at MIT? → Someone is the
professor of linguistics at MIT.
Yes/no interrogatives and alternative interrogatives presuppose the
disjunction of their possible answers:
(14a) Is there a professor of linguistics at MIT? → Either there is a professor
of linguistics at MIT or there isn’t.
(14b) Is Newcastle in England or is it in Australia? →Newcastle is in
England or Newcastle is in Australia.

So far we have considered contexts in which the presuppositions is a

piece of information which is assumed to be the case. There are, however,
instances of non-factive presuppositions associated with a number of verbs
such as dream, imagine, pretend, etc. These verbs are used with the
presupposition that the piece of information that follows is not the case:
(15a) I dreamt that I was rich. → I was not rich.
(15b) We imagined we were in Hawaii. → We were not in Hawaii.
(15c) He pretends to be ill. → He is not ill.

Second type if- clauses are associated with counter-factual

presuppositions, meaning that what is presupposed is not only not true, but
is the opposite of what is true, i.e. it is contrary to facts.

(16) If you were my friend, you would have helped.

→ You are not my friend.

36
3.4 Presupposition triggers
Associated with these types of presuppositions are various lexical items or
aspects of surface structure. These presupposition-generating linguistic items
are referred to as presupposition triggers. The following list includes
various types of presupposition triggers:
Definite descriptions (Strawson 1950)
(17a) John saw the man with two heads. → There exists a man with two
heads.
(17b) The Prime Minister of Italy stood up and waved his hand. → Italy has
a (unique) prime minister.
Proper names names (van der Sandt, 1992)
(18) The author is Julius Seidensticker. → Julius Seidensticker exists.
Factive verbs (Kiparsky and Kiparsky, 1970)
(19a) Berlusconi knows that he is signing the end of Berlusconism.
→ Berlusconi is signing the end of Berlusconism.
(19b) Martha regrets going to John’s party. → Martha went to John’s party.
(19c) John was aware that Martha was at home. → Martha was at home.
(19d) John realized that he was in debt. →John was in debt.
(19e) It was odd how proud he was. → He was proud.
Further factive verbs include: know, be sorry/indifferent/sad/glad +
that- clause.
Implicative verbs (Levinson 1983)
(20a) John managed to open the door. →John tried to open the door.
(20b) John forgot to post the letter. → John ought to have posted the letter/
intended to post the letter.
Aspectual verbs (Simons, 2001; Abusch, 2002; Lorenz, 1992)

37
(21a) China has stopped stockpiling metals. → China used to stockpile
metals.
(21b) Joan began to beat her husband. → Joan hadn’t been beating her
husband.
(21c) Bush continued to rule the world. → Bush had been ruling the world.
Further aspectual verbs: start, finish, carry on, cease, leave, enter, arrive,
come.
Iteratives (Levinson 1983)
(22a) The flying saucer came again. → The flying saucer came before.
(22b) Clinton returned to power. → Clinton held power before.
Further iteratives: another time, come back, repeat.
Manner adverbs (Abbott, 2000)
(23) Jamie ducked quickly behind the wall. → Jamie ducked behind the wall.
Quantifiers (Roberts, 1995; Gawron, 1995; Abusch and Rooth, 2000;
Cooper, 1983)
(24) I have written to every headmaster in Rochdale. → There are
headmasters in Rochdale.
Sortally restricted predicates of various categories (e.g., bachelor)
(Thomason, 1972)
(25) Julius is bachelor.→ Julius is an adult male.
Temporal clauses (Beaver and Condoravdi, 2003; Heinämäki, 1974)
(26a) The dude released this video before he went on a killing spree. → The
dude went on a killing spree.
(26b) While Chomsky was revolutionizing linguistics, the rest of social
science was asleep. → Chomsky was revolutionizing linguistics.
(26c) Since Churchill died we’ve lacked a leader. → Churchill died.
(26d) As John was getting up, he slipped. → John was getting up.

38
Cleft sentences (Delin, 1995; Prince, 1986)
(27) It was Jesus who set me free. → Somebody set me free.
Non-restrictive relative clauses (Levinson 1983)
(28) Hillary, who climbed Everest in 1953, was the greatest explorer of our
day. → Hillary climbed Everest in 1953.
Intonation (e.g., focus, contrast) (Jackendoff, 1972; Geurts and van der
Sandt, 2004; Roberts, 1998)
(29) HE set me free.→ Somebody set me free.

3.5 Projection
The most thoroughly researched presuppositional phenomenon is projection
(Langendoen and Savin, 1971). There is a basic expectation that the
presupposition of a simple sentence will continue to be true when that simple
sentence becomes part of a complex sentence. This expectation is related to a
more general view proposed by Frege that the meanings of sentences are
compositional, i.e. the meaning of the whole expression is a function of the
meaning of the parts. However, the set of the presuppositions of a complex
sentence is not the simple sum of the presuppositions of the parts; some
presuppositions of the component sentences do not survive to become
presuppositions of the complex sentence. Moreover, a theory that will predict
correctly which presuppositions of component sentences will be inherited by
the complex sentence is difficult to formulate. This compositional problem is
known as projection and it is a distinctive feature of presuppositions.
Two issues should be considered in relation to presuppositional
projection. First, presuppositions survive in linguistic contexts where
entailments cannot, i.e. the presuppositions of component sentences are
inherited by the whole complex sentence where the entailments of those
components would not be. Second, presuppositions disappear in other
contexts where we might expect them to survive and where entailments
would.
Consider the sentence in (30). The sentence in (30) has all the
presuppositions in (31). The presuppositions in (31) all follow from
utterances of the sentence in (30). The entailments of the sentence in (30) are
given in (32). A speaker who sincerely uttered (30) is expected to be
committed to the truth of (31) and (32).

(30) It’s the boy that stole the apples.

39
(31a)There is a (salient and identifiable) boy.
(31b) There were (salient and identifiable) apples.
(31c) Somebody stole the apples.
(32a) The boy did something illegal.
(32b) The boy took possession of the apples.
Now consider the sentences in (33):
(33a) It isn’t the boy that stole the apples. (negative sentence)
(33b) If it’s the boy that stole the apples, he will be punished. (antecedent of
a conditional)
(33c) Is it the boy that stole the apples? (interrogative sentence)
(33d) Maybe/It is possible that it’s the boy that stole the apples. (possibility
modal)
(33e) Presumably/probably it’s the boy that stole the apples. (evidential
modal, probability adverb)
(33f) The teacher thinks it’s the boy that stole the apples. (belief operator)
In all the examples in (33), sentence (30) is embedded under various
operators. What is notable is that whereas the statements in (32) do not
follow from any of these embeddings, the presuppositions do follow. Thus,
the presuppositions are said to be projected. Some contexts make the
inference more robust than others. For instance, it is hard to imagine
sincerely uttering (33a) without believing some apples to be salient. On the
other hand, it is easier to imagine a situation in which (33f) could be uttered
when in fact the apples were not stolen, but hidden. But in the absence of
special factors, to which I will turn shortly, someone who sincerely uttered
any of the sentences in (33) might be expected to believe all of the
presuppositions in (31a)–(31b).
In what follows I will turn to examining the survival properties of
presuppositions. Projection from negation is used as a standard diagnostic
test for presupposition hence the term “constancy under negation test”. Thus
(34) presupposes (35) and entails (36)

(34) The chief constable arrested three men.

40
(35) There is a chief constable.
(36) The chief constable arrested two men.

From the negation of (34), as in (37), one cannot entail (36); the
presupposition in (35), however, does survive negation.

(37) The chief constable did not arrest three men.

Karttunen (1973) proposes a taxonomy of embedding constructions

that divided them into three classes: plugs, holes and filters. Plugs comprise a
class of predicates and operators which Karttunen claimed block the
projection of presuppositions, while holes are a class of predicates and
operators which allow presuppositions to project freely.
His list of complement-taking verbs or sentential operators that allow
presuppositions to ascend to become presuppositions of the complex
sentence but would block entailments includes: factive verbs, modal
operators, negation, disjunctive and conditional contexts. For instance,
consider the example in (38) and (39) that illustrate modal contexts (i.e.
when embedded under modal operators like possible, probable, there’s a
chance that, etc.) in which presuppositions survive, but entailments do not.
(38) continues to presuppose (35). However, (38) does not entail (36)
because it cannot logically be inferred from the mere possibility of a state of
affairs that any part of that state of affairs is actually the case.

(38) It’s possible that the chief constable arrested three men.

The same behaviour occurs under deontic modalities such as ought, should
and the like. Consider the example in (39):

(39) The chief constable ought to have arrested three men.

(39) presupposes “There is a chief constable”, but does not entail “The chief
constable arrested two men”.
Another set of contexts in which presuppositions survive while
entailments do not, are complex sentences formed by the connectives and,
or, if......then their equivalents. Consider for example in (40):

(40) The two thieves were caught again last night.

The sentence in (40) entails among other things:

41
(41) A thief was caught last night

The iterative again triggers the lexical presupposition in (42):

(42) The two thieves had been caught before.

If (40) is embedded in the antecedent of a conditional as in (43), (41) is not

an entailment of (43), but the presupposition (42) survives:

(43) If the two thieves were caught again last night PC Brown will get an
honourable mention.

Similarly if (40) is embedded in a disjunction, its presuppositions, but not its

entailments, survive:

(44) Either the two thieves were caught again last night, or PC Brown will be
losing his job.

3.6 Cancellability

What makes the “projection phenomenon” problematic is the fact that,

although presuppositions typically project quite often they do not. Most of
the empirical and theoretical work on presupposition since the 1970s has
focused on the task of describing and explaining the contexts when
presuppositions project and the contexts when they fail to be inherited by
complex sentences because they are cancelled by virtue of intra-sentential
context (they are defeasible).
When a presupposition does not project, it is sometimes said to be
“canceled”. The most straightforward way cases of cancellation occur when
the presupposition is overtly denied in co-ordinate sentences as in (45) and
(46).

(45) John doesn’t regret doing a useless PhD in linguistics, because in fact he
never did one!
(46) John didn’t manage to pass his exam, in fact he didn’t even try.

This cannot be done with entailments because of direct contradiction.

Consider the example in (47) which is pragmatically infelicitous (marked by
the sign “#”).

42
(47) #John doesn’t regret doing a useless PhD because in fact he does regret
doing a useless PhD.

Presuppositional inferences can typically be cancelled by overt denial

only when the presupposition trigger is embedded under some other
operator. When the presupposition is not embedded, such cancellation (by
the same speaker) is usually pragmatically infelicitous, just as is cancellation
of entailed content which is not embedded. Thus the denial of a
presupposition in (48) and the denial of an entailment in (49) both lead to
pragmatically infelicitous utterances.

(48) #John regrets doing a PhD because in fact he never did one
#Joan has stopped beating her husband and in fact she never did beat him.

(49) #It’s the boy that stole the apples, but he didn’t do anything illegal.

The fact that presuppositions associated with unembedded triggers are not
cancellable is one of the features that distinguish most presuppositions from
conversational implicatures (Grice, 1989). For instance, the example in (50a)
might ordinarily lead to the scalar implicature in (50b). But while this
implicature is cancelable, as shown in (50c), the presupposition of existence
that there is a boy is not cancelable, as shown by the oddity of (50d).

(50a) The boy stole most of the apples.

(50b) The boy did not steal all of the apples.
(50c) The boy stole most of the apples—in fact, he stole them all.
(50d) #The boy stole most of the apples, but there was no boy.

The typical behavior of entailments, presuppositions, and conversational

implicatures is summarized in (51)

(51)

Entailments Presuppositions Implicatures

Project from embeddings no yes no
Cancelable when embedded — yes —
Cancelable when unembedded no no yes

43
 In addition to the cancellation of presuppositions by overt denial,
there is the possibility of what Horn (1986) has called suspension. The use of
a following if-clause can suspend the speaker’s commitment to
presuppositions as shown in (52) and (53):

(52) John didn’t cheat again, if indeed he ever did.

(53) Harry doesn’t regret being a CIA agent, if he actually ever was one.

The presuppositions of constituent parts of complex sentences are

blocked under certain verbs of propositional attitude like want, believe,
imagine, dream and verbs saying like say, tell, mumble, retort, etc. Consider
the examples in (54) and (55):

(54) Loony old Harry believes he’s the King of France.

(55) Nixon announced his regret that he did not know what his subordinates
were up to.

These examples do not seem to give rise to the expectable presuppositions

given in (56) and (57) respectively:

(56) There is a present King of France.

(57) Nixon did not know what his subordinates were up to.

In view of this behaviour such verbs of propositional attitude and verbs of

saying are referred to as plugs, because in contrast to holes, they prevent the
presuppositions of lower sentences from ascending to become
presuppositions of the whole. However, it is far from clear that this is
generally true. Consider for example:

(58) The mechanic didn’t tell me that my car would never run properly again
(59) My car used to run properly.

Here (58) continue to presuppose (59) despite the presence of plugs. Thus
the existence of plugs is questionable.
One of the most troublesome aspects of the projection problem is the
behaviour of presuppositions in complex sentences formed by using the
connectives and, or, if….then, and related expressions such as but,
alternatively, suppose that, etc. Karttunen (1973) calls these connectives
filters because they let some presuppositions through but not others.

44
 It has been shown that presuppositions tend to survive in disjunctions
and conditionals where entailments do not. Thus one might argue that these
constructions are holes that let presuppositions through. However,
counterexamples can be supplied to show that this behaviour does not always
apply. Consider the example in (60):

(60) If John does computational linguistics, he will regret doing it

(61) John will do computational linguistics.

In (60) the main clause, by itself, would presuppose (61). However, the
whole conditional sentence does not presuppose (61) because presupposed
content mentioned in the subordinate if-clause and is therefore made
hypothetical. This turns out to be general. Similarly, consider the disjunction
in (62).

(62) Either John will not in the end do computational linguistics, or he will
regret doing it.

In (62) the second clause alone presupposes (61), but the whole does not.
The presupposition seems to be cancelled because the alternative expressed
in the first clause is the negation of the presupposition of the second clause.

3.7 Current issues in presupposition theory

3.7.1 Local contexts

By the early 1970s, the focus of research had been on presuppositions

triggered by definite descriptions. Since the early 1970s, the empirical
domain of presupposition theory has expanded to cover a wide range of
triggers including including factives (Kiparsky and Kiparsky, 1970),
implicatives (Karttunen, 1971a), focus particles (Horn, 1969), verbs of
judging (Fillmore, 1971) and sortal constraints (Thomason, 1972). By the
early 1970s, empirical linguistic research such as Morgan (1969),
Langendoen and Savin (1971), Karttunen (1973) had been expanded in the
direction of arbitrary embeddings of presupposition triggers in contexts other
than unembedded triggers and negation. Karttunen’s (1973) fine-grained
analysis showed that there was more variation in projection behavior than
had been previously described and that none of the Frege-Strawson derived
systems could account for every case.
Karttunen (1973) proposed taxonomy of embedding constructions
that included three classes: plugs, holes and filters. Plugs comprise a class of
45
predicates and operators which, according to Karttunen, block the projection
of presuppositions, while holes are a class of predicates and operators which
allow presuppositions to project freely. Karttunen argues that in (63) the
presupposition there is a King of France is not projected because told that is
a plug. However, in (64) perhaps is a hole and thus the presupposition of
existence (i.e. there is a King of France) can be predicted.

(63) Mary told Jim that the King of France was bald.
(64) Perhaps the King of France is bald.

Karttunen’s filters include the binary logical connectives “if then”, “and”,
and “or”. These constructions are analyzed as allowing only some
presuppositions to project. Thus the example in (65) shows that the
presupposition that there was a knave does not project because it is
embedded in a conditional or, to use Karttunen’s metaphor, filtered out.
However, the presupposition that there are (or, at least, there were) some
tarts triggered by the definite description the tarts does project from the
conditional.

(65) If there is a knave, then the knave stole the tarts.

Karttunen concluded that the consequent of a conditional acts as a hole to

some presuppositions, but filters out all those presuppositions which are
entailed by the antecedent, or, more generally, by a combination of the
antecedent and contextually supplied background information. The example
in (66), provided by Karttunen, shows the crucial role played by the context.

(66) Either Geraldine is not a mormon or she has given up wearing her holy
underwear.
(Karttunen, 1973)

In (66) the second half contains two presupposition triggers: the definite
description “her holy underwear” and the aspectual verb “give up”, which
trigger the presuppositions that Geraldine had and she used to wear holy
underwear, respectively. The filtering condition that Karttunen proposes for
disjunctions removes from the right disjunct any presuppositions that are
entailed by a combination of the context and the negation of the left disjunct.
If the sentence in (66) is uttered within a context that supports the
proposition that all mormons have holy underwear which they wear
regularly, it follows from this proposition and the negation of the left
disjunct (i.e., the proposition that Geraldine is a mormon) that Geraldine has
46
holy underwear and has worn it regularly. But these are exactly the
presuppositions triggered in the right disjunct, so they are filtered out.
Consequently, (66) has no presuppositions.
Karttunen’s study (1973) did not clarify why certain presuppositions
should be filtered out if they were entailed by additional context. However,
Karttunen (1974) proposes an alternative treatment of presuppositional
projection based on the idea of local contexts of evaluation. According to
Karttunen (1974), the parts of a sentence are not necessarily evaluated with
respect to the same context as that in which the sentence as a whole is
evaluated. He argues that a local context may be more informative than
the global context. For instance, when evaluating a conjunction, the second
conjunct is evaluated in a local context which contains not only the
information in the global context, but also whatever information was
supplied by the first conjunct. Karttunen (1974) defined local contexts of
evaluation for a range of constructions. Moreover, he suggested the
following requirement: presuppositions always need to be entailed or
“satisfied” (as he puts it) in the local context in which the trigger is
evaluated. As a result of this requirement, the overall presuppositions of a
sentence will be whatever propositions must be in a context of an utterance
in order to guarantee that the presuppositions associated with presupposition
triggers are satisfied in their local contexts of interpretation.
For each connective or operator he considered, Karttunen spelled out
how local satisfaction should be calculated. Recent developments such as
Schlenker (2008) and Beaver (2008) provide a general way of calculating
what the local context should be. The following reformulation draws on
Karttunen’s model and incorporates Schlenker’s insights along the lines
proposed by Beaver (2008).
As an illustration, let us say that some clause in a complex sentence
is redundant relative to some context of utterance if it can be replaced that
clause by a tautology without affecting the amount of factual information
conveyed by the sentence in that context. For example, in (67), the first
conjunct is redundant in any context of utterance. Here, the same factual
information would be conveyed by “John is John and John owns a horse”,
where the first conjunct is replaced by the tautology “John is John”.

(67) John owns an animal and John owns a horse.

A clause is left-redundant if it is possible to tell by looking at the material in

the sentence to the left of the clause that the clause is redundant. So “John
owns an animal” is not left-redundant in (67) (unless the context of utterance
already entails that John owns an animal), because there is no material before
47
that clause, implying that it is impossible to tell by looking at material to the
left of the clause that the clause is redundant. By contrast, “John owns an
animal” is left-redundant in (68) and (69):

(68) John owns a horse and John owns an animal.

(69) If John owns a horse then John owns an animal.

This leads us to the following definition of presupposition via Karttunen’s

(1974) notion of presupposition satisfaction in global contexts.

A presupposition P is satisfied at point X in S iff P would be left-redundant if

added at point X. A sentence presupposes whatever propositions must hold in
global contexts of utterance such that each locally triggered presupposition is
satisfied where its trigger occurs.

(based on Karttunen, 1974 and Schlenker, 2008)

As an example, let us consider the presuppositions predicted for (70),

repeated for convenience below:

(70) Either Geraldine is not a mormon or she has given up wearing her holy
underwear.
(Karttunen, 1973)

Before considering the satisfaction of the presuppositions for (70) within a

global context, it should be pointed out that for sentences of the form
“A or B”, the negation of A is satisfied within the right disjunct. Thus,
“Geraldine is a mormon” is satisfied in the right disjunct of (70)8. The clause
“she has given up wearing her holy underwear” triggers the presupposition
that “Geraldine has had holy underwear that she wore” due to the definite
description her holy underwear and the aspectual verb give up. This
presupposition will be satisfied provided the global context of utterance,
combined with the negation of the left disjunct, entails that she has had holy
underwear that she wore. This will be the case if and only if the context
supports the conditional “if Geraldine is a mormon, then she has had holy
underwear that she wore.” Hence, this conditional is the presupposition
Karttunen (1974) predicts for (70).

8
More generally, anything entailed by a combination of propositions in the context and the
negation of the left disjunct will be satisfied in the right disjunct.
48
3.7.2 Presupposition and anaphora

A number of authors have pointed out that presuppositions behave, to a

certain extent, like anaphors (Kripke 2009 and Soames 1989). With respect
to this similarity, van der Sandt’s (1989; 1992) analysis is crucial. He pointed
out that for every case where a pronoun is assigned an anaphoric reading, but
is not interpretable as a bound variable, a similar configuration is possible
with presuppositions. Thus in each set of examples from (71) to (74) the (a)
and (b) cases exemplify a phenomenon, while the (c) and (d) cases show that
a slight divergence from the original configurations given in the (a) and (b)
variants produces infelicity. Moreover, the (a) and (c) examples include the
anaphoric pronoun it, and the (b) and (d) examples include the factive verb
know, triggering a presupposition that its propositional complement is true
(i.e., John left). The examples in (71) – (74) show anaphora to be felicitous
in configurations in which presuppositions are felicitous. Similarly, they
show anaphora to be infelicitous in configurations in which presuppositions
are infelicitous. This parallelism demonstrates that the relations between
anaphoric pronouns and their antecedents on the one hand, and
presupposition triggers and their antecedents on the other, are sensitive to
very similar configurational constraints.

Inter-sentential (discourse) anaphora

(71a) There was a storm. It was fierce.

(71b) John left. Lucy knows that John left.
(71c) #It was fierce. There was a storm9.
(71d) #Lucy knows that John left. John left.

Anaphora and conditional contexts

(72a) If a farmer owns a donkey then he beats it10.

(72b) If John left then Lucy knows that John left.
(72c) #If a farmer doesn’t own a donkey, then he beats it.
(27/72d) #If John didn’t leave left then Lucy knows that John left.

Modal subordination

(73a) A bear might come to the door. It would eat you.

9
The symbol # indicates the sentence is pragmatically infelicitous in the given context.
10
The example is due to Kamp (1981).
49
(73b) John might have left. Lucy would know that John has left.
(73c) #A bear might come to the door. It’s brown.
(73d) #John might have left. Lucy knows that John has left.

Anaphora and the disjunction operator or

(74a) Either there’s no bathroom in this house, or else it’s in a funny place11.
(74b) Either John didn’t leave, or else Lucy knows that he left.
(74c) #Either there is a bathroom, or else it’s in a funny place.
(74d) Either John left, or else Lucy knows that he left.

In order to account for these parallels, van der Sandt proposed a unified
treatment of presupposition and anaphora, applying Discourse
Representation Theory so as to deal with both phenomena. Presupposed
information is information that is presented as given, and in van der Sandt’s
theory this means that presuppositions need to have discourse referents to
bind to. However, whereas pronouns used anaphorically can rarely be
interpreted in the absence of an antecedent, the same does not hold for all
presupposition-inducing expressions. For instance, a speaker may
felicitously assert that he met “John’s sister” even if he knows that his
audience wasn’t aware that Fred has a sister. In such cases, presuppositions
are generally accommodated, which means that the addressee accepts the
information as given, and revises his representation of the context
accordingly.
Van der Sandt’s theory (1989; 1992) incorporates the notion of
accommodation as follows. Presuppositions, according van der Sandt,
introduce information that is linked to discourse referents that are already
available in the hearer’s representation of the discourse, and in this respect
they behave like pronouns. However, if a suitable discourse antecedent is not
available, a new one will be accommodated, and the presupposition is linked
to that. Being told that “she” is wonderful is not particularly helpful or
informative if it isn’t clear who the pronoun is refers to. By contrast, if the
speaker refers to “John’s sister” there is more to go on, and accommodation
becomes feasible. Van der Sandt hypothesizes that pronouns make up a
special class of presuppositional expressions. While all presupposition
triggers prefer to be linked to antecedents, pronouns almost always must be
linked to antecedents, because cannot be construed by way of
accommodation. Consider the example in (75).

11
The example is due to (Partee 1984).

50
(75) If Bill is gay, then his son is gay, too.

This sentence contains the definite NP his son and the focus particle too.
Assuming that the antecedent of the pronoun is “Bill”, the definite NP
triggers the presupposition that Fred has a son, while the focus particle
triggers the presupposition that someone other than Bill’s son is gay. Note
that in this example the presupposition triggered by the definite NP is
“inherited” by the sentence as a whole, while the one triggered by too is not.
The example in (75), when uttered, would license the inference that
(according to the speaker) Bill has a son, but not that someone else besides
Bill’s son is gay.

3.7.3 Accommodation

The concept of accommodation was first addressed by Karttunen (1974) and

Stalnaker (1974), although it was named as such by Lewis (1979). Karttunen
introduces the concept as follows:

Ordinary conversation does not always proceed in the ideal orderly fashion
described earlier. People do make leaps and shortcuts by using sentences
whose presuppositions are not satisfied in the conversational context. This is
the rule rather than the exception […] I think we can maintain that a sentence
is always taken to be an increment to a context that satisfies its
presuppositions. If the current conversational context does not suffice, the
listener is entitled and expected to extend it as required.

(Karttunen 1974: 191)

The concept of accommodation is among the more contentious topics in

presupposition theory. Various definitions of the concept of accommodation
have been advanced, some of which being stricter than others. Consider the
example in (76) due to Heim (1982).

(76) John read a book about Schubert and wrote to the author.

In order to determine the intended meaning of the definite description the

author, the hearer has to infer (i) that there is an author and (ii) that the
author referred to wrote the book that John read. Whereas on a broad
definition of accommodation both these inferences are accommodated, on a
strict construal only (i) is, and (ii) is a bridging inference. Thus, one can
51
argue that there is something like an “accommodation module”, which as
such has nothing to do with world knowledge; whereas if the notion is
construed more broadly, accommodation is of a piece with bridging. To
facilitate the following discussion, I will adopt a strict notion of
accommodation and take the view that what is accommodated is the
presupposition as triggered.
When approaching the concept of presupposition the first question
that arises is: Where are presuppositions accommodated? Though it may
seem odd at first, this question is crucial an expression may occur in several
contexts at the same time. To illustrate, consider the following:

(77a) (c0) Maybe (c1) Betty is trying to give up smoking.

(77b) (c0) Maybe (c1) Wilma thinks that (c2) her husband is having an affair.

Here c0 refers to the global context in which a given sentence is uttered, and
c1and c2 are local, contexts. In (77a), the modal operator maybe creates an
local context of possible states of affairs in which Betty is trying to give up
smoking; the same applies to (77b), mutatis mutandis. The presupposition
triggered in (77a), that Betty used to smoke, can be accommodated globally,
i.e., in c0, or locally, in c1. In the former case, the utterance is construed as
meaning that Betty used to smoke and may be trying to kick the habit; in the
latter, it conveys that, possibly, Betty used to smoke and is trying to give up
smoking. Similarly, in (77b), the presupposition that Wilma is married may
be accommodated globally, or locally in the most deeply embedded context.
But here there is a third option, as well: if the presupposition is
accommodated in c1, the sentence is interpreted as “Maybe Wilma is married
and she thinks that her husband is having an affair”, and we speak of
intermediate accommodation.

Heim (1983) proposes the following empirical generalization:

PGA: Global accommodation is preferred to non-global accommodation.

The examples in (77b) illustrate cases when the PGA (preference for global
accommodation) clearly holds: non-global interpretations may be possible,
but they require special contexts. One such context may be that the
presupposition contains a variable which is bound by a quantifier. Consider
the example (78).

(78) Most Americans mow their lawns on Saturday.

52
In (78), with the possessive presupposition trigger their lawn, there is a
global context (outside of the scope of the quantifier most), a local context
corresponding to the scope of the quantifier (occupied by the VP mow their
lawns on Saturday), and also an intermediate context in the restrictor of the
quantifier (occupied by Americans). The most natural interpretation of this
sentence is that most Americans who own a lawn mow it on Saturday. So in
this case intermediate accommodation seems to be the preferred option.
There are other cases where intermediate accommodation is virtually
impossible, as shown in (79).

(79) (c0) If (c1) Bill is coming to the reception, (c2) he may (c3) bring his
wife.

It is quite unlikely that this may be construed, with intermediate

accommodation in c1, which is the antecedent of the conditional, as “If Bill is
married and is coming to the reception, he may bring his wife.”
Virtually all theories of presupposition projection accept that the
PGA holds. This preference can be accounted for in terms what Beaver
(2001) calls the “Atlas Principle”, after Atlas (1976):

AP: One accommodation alternative is preferred to another if the former

yields a stronger meaning than the latter (i.e., if the first meaning unilaterally
entails the second).

In general, the predictions made by the Atlas Principle will tend to

comply with the PGA. However, this is does not have to be the case in all
contexts. A major worry about the Atlas Principle is that it is ad hoc. Despite
its resemblance to Grice’s second Quantity Maxim, it is debatable whether
hearers generally prefer stronger interpretations to weaker ones.
One last issue we would like to mention is that accommodation isn’t
always equally easy. Consider the examples in (80a) and (80b).

(80a) Fred is looking for the person.

(80b) Jane is pregnant, too.

In (80a) the presupposition triggered by the definite description the person

clearly requires a salient discourse referent that is clearly identifiable to. The
sentence would be infelicitous when uttered out of the blue. The same
applies to the presupposition triggered by too in (80b), i.e. that some salient
person different from Jane is pregnant. For both (80a) and (80b),
accommodation is possible if the context provides a hook, in the form of a
53
salient individual that could conceivably be a person, or pregnant,
respectively. In a context where the interlocutors have been discussing their
tech stocks, for instance, without mentioning any particular people, (80b)
would be odd.

3.7.4 Presupposition failure

According to Frege (1892), if an expression A suffers from presupposition

failure, then any sentence containing A will lack a truth value. Russell (1905)
rejected this view arguing that such a sentence will always be true or false;
Strawson (1950) reaffirmed Frege’s position, more or less. However, in
subsequent work, Strawson partly recanted his initial view and questioned
whether presupposition failure invariably entails lack of truth value.
Strawson (1964) argues that presupposition failure may but need not
cause a sentence to be infelicitous. Consider the examples in (81a) and (81b)
due to Strawson.

(81a) Jones spent the morning at the local swimming pool.

(81b)The exhibition was visited yesterday by the King of France.

If there is no swimming pool locally, it is “natural enough”, according to

Strawson, to say that (81a) is false, and since the king of France doesn’t
exist, the same applies to (81b). And if these sentences are false, their
negations must be true. So, if these interpretations are accepted, these are
cases in which presupposition failure does not prevent us from judging a
sentence as true or false. But Strawson hasn’t changed his mind about
Russell’s example:

Confronted with the classical example, “The king of France is bald”, we may
well feel it natural to say, straight off, that the question whether the statement
is true or false doesn’t arise because there is no king of France.

(Strawson 1964: 90)

However, Strawson points out that speakers who subscribe to this judgment
may want to reconsider their verdict if the context is set up the right way. For
instance, if Russell’s sentence is used to answer the question, “What
examples, if any, are there of famous contemporary figures who are bald?”,
one may be more inclined to say that the answer is simply false.
These facts are accounted for by Strawson in terms of topicality. The
most likely purpose of a sentence like (81a) is to describe what Jones has
54
been doing in the morning, rather than, say, refer to the local swimming
pool. That is, in the absence of further information about the context in
which this sentence is uttered, its topic will be Jones’s exploits. Similarly, a
sentence like (81b) will normally be used to convey information about the
exhibition. If so, although the sentence purports to refer to the king of
France, it is not about him; the king of France is neither the topic of
discourse, nor part of the topic. On this view, presupposition failure results in
infelicity only if it affects the topic of a sentence; otherwise the sentence will
be judged true or false, as appropriate.
Strawson’s analysis takes into account the context-dependence of
speakers’ intuitions. Strawson argues that Russell’s sentence in (82) is by
default construed as being about the king of France, and consequently there
is a strong tendency to judge the sentence infelicitous because the
presupposition that there is a present king of France is not met.

(82)The king of France is bald.

If, however, the discourse is about royal baldness in general, the grammatical
subject of (82) is used to say something about that topic, and Strawson’s
account predicts that the sentence is more likely to be judged false, which
seems correct. Word order may play a crucial role in speakers’ intuitions
about presupposition failure. As Strawson (1964:91) points out, if we
compare (81b) with (83), where the defective description is in subject
position, we would be “a shade more squeamish” to say that the sentence is
simply false. This is precisely what one should expect if speakers’ intuitions
were topic-dependent.

(83) The king of France visited the exhibition yesterday.

Assuming that Strawson’s observations are correct, the following questions

arise from his line of reasoning: (a) are non-topical definites non-
presuppositional? or (b) do they have presuppositions whose failure does not
affect speakers’ truth-value judgments? Some authors argue for the former
(e.g., Reinhart 1982, Horn 1989); this is a view embraced by Strawson, as
well. Von Fintel (2004) argues for the latter: topical or non-topical, “the king
of France” always triggers the presupposition that there is a king of France;
it’s just speakers’ truth-value judgments that fluctuate.
Von Fintel’s position is in line with what in theories of projection has
been a working hypothesis for many years. According to these theories,
presuppositions are never really cancelled: if a presupposition seems to
disappear, it is because it projects to a non-global context.
55
 So far I have addressed presuppositions triggered by definite
descriptions. Some other triggers show similar variability. Presuppositions
associated with strong quantifiers like all and most behave very much like
those associated with definite descriptions (de Jong and Verkuyl 1985,
Lappin and Reinhart 1988, Geurts 2007). Consider the example in (84).
When uttered out of the blue, (84) will sound odd (i.e. infelicitous), but when
uttered in response to Strawson’s question, “What examples, if any, are there
of famous contemporary figures who are bald?”, it seems more likely to be
judged false:

(84)All South-American monarchs are bald.

On the other hand, consider the following:

(85a) #BENEDICT XVI is the incumbent pope, too.

(85b) #Carnap managed to be born on May 18, 1891.

Given that there is only one incumbent pope and that it can’t have taken
Carnap any effort to be born on May 18, 1891, both (85a) and (85b) suffer
from presupposition failure.

3.8 Conclusions

The aim of this chapter was to outline an overview of the major strands of
work on presupposition. A question that still remains to be addressed is
whether presupposition triggers form a natural class of linguistic expressions.
This is a possibility that has only just begun to be explored in detail, with
several theorists proposing to dichotomize presupposition triggers in some
way or other (e.g., Zeevat 1992, Geurts 1999b, Abusch 2002).
There are several philosophically and linguistically interesting
dimensions along which the set of presupposition triggers can be classified,
such as referentiality, anaphoricity, ease of accommodation, ease of
cancellation, and maintenance of truth under presupposition failure. Such
dimension of classification might a more complex taxonomy of different
types of triggers.

56
 CHAPTER 4
IMPLICATURE

4.1 Introduction

In everyday naturally-occurring conversations speakers often convey more

information that goes beyond what they strictly speaking say. Exaggeration
and irony are cases in point, for instance. H.P. Grice (1975, [1989])
introduced the technical notion of a conversational implicature to refer to the
act of meaning one thing by saying something else. Consider the following
examples in (1) and (2) due to Grice (1975 [1989]: 39-40):

(1)
A: Can I get petrol somewhere around here?
B: There’s a garage around the corner.
[A can get petrol at the garage around the corner.]

(2)
A: Is Karl a good philosopher?
B: He’s got a beautiful handwriting.
[Karl is not a good philosopher.]

In each of these examples B’s utterance conveys a proposition – the

proposition given in italics within the square brackets – that differs from
what B has explicitly said. For instance, while B in (1) merely says or asserts
that there’s a garage around the corner, the main point of B’s utterance is to
convey the proposition that A can get petrol at the garage around the corner.
In introducing the notion, Grice drew a line between what is said,
which he understood as being closely related to the conventional meaning of
the words uttered, and what is conversationally implicated, which can be
inferred from the fact that an utterance has been made in context. Since
Grice’s seminal work, conversational implicatures have become one of the
major research areas in pragmatics. This chapter focuses on the notion of a
conversational implicature. It addresses the key issues that lie at the heart of
the recent debate, and explicates tests that allow us to reliably distinguish
between semantic entailments and conventional implicatures on the one hand
and conversational implicatures on the other hand.

57
4.2 Natural vs. non-natural meaning

The concept of implicature was first proposed by Grice in the William James
Lectures delivered at Harvard in 1967 and still only partially published
(Grice, 1975, 1978). Before we outline his key ideas, it would be useful to
briefly review his other major theory, namely his theory of non-natural
meaning as it sheds new light on his theory of implicature.
In his seminal paper ‘Meaning’ (1957), Grice proposed a distinction
between natural(N) and non-natural(NN) meaning. The distinction reflects what
he later describes as “a reasonably clear intuitive distinction between cases
where the word ‘mean’ has what we might think of as a natural sense, a
sense in which what something means is closely related to the idea of what it
is a natural sign for (as in ‘Black clouds mean rain’), and those where it has
what I call a non-natural sense, as in such contexts as ‘His remark meant so-
and-so’.” (1989: 291). Cases of each kind are given in (3) and (4) below:

(3) These spots mean measles.

(4) That remark means he has measles.
Grice proposed several recognition tests whereby the two types of meaning
might be distinguished. Cases of meaningN are factive. A speaker who says
“those black clouds mean rain” commits himself to the claim that it will rain
(or has rained): in such cases, x meansN p or x meantN that p entail p. By
contrast, cases of meaningNN are non-factive. A speaker who says ‘his
remark meant it is raining’ does not commit himself to the claim that it is
raining now, or, in fact, has been raining at all. This observation was
reflected in one of a series of tests, in which Grice contrasted ways that
utterances containing uses of the word ‘mean’ (in both senses) might be
satisfactorily paraphrased. While (5a) is a plausible paraphrase of utterance
(3), (5b) is not a plausible paraphrase of utterance (4). It may be true, but it
does not convey the same sense of ‘means’ as that in the original utterance.
(5a) The fact that he has those spots means he has measles.
(5b) ??The fact that he made that remark means he has measles.
Another recognition test that Grice proposed, this time for involves
the paraphrasing of utterances such as (3) and (4) using direct quotation. (6a)
below is not a satisfactory paraphrase of (3), but (6b) is a satisfactory
paraphrase of (4). Thus, only non-natural meaning allows paraphrasing using
direct quotation.
58
(6a) ??Those spots mean “he has measles”.
(6b) That remark means “he has measles”.
In another test, Grice proposed that no conclusion about what is (was) meant
by
(something) could be drawn from an utterance that describes a case of
meaningN.
Such a conclusion, however, can indeed be drawn from a case of meaning NN
(see (7ab)):
(7a) ??What was meant by those spots was that he has measles.
(7b) What was meant by that remark was that he has measles.
Of the two types of meaning, Grice paid special attention to
meaningNN; in particular, how the kind of meaning exemplified in (6b) and
(7b) might be characterized in terms of intentions and the recognition of
intentions. The examples in (8a-e) below are adapted from Grice (1989:
218). In all these sentences the speech event triggered a response in an
audience. In both senses of the word, then, Mary’s pale complexion and her
utterance might be said to mean she is unwell:
(8a) Mary is asleep. Her mother notices that she is pale and concludes she is
unwell.
(8b) Feeling unwell, Mary intends her mother to see how pale she is.
However, she doesn’t want this intention to be noticed (it might decrease her
chances of getting a day off school), so she pretends to be asleep.
(8c), as (8b), except that here Mary’s mother realizes Mary is pretending to
be asleep, and hence infers her intention not to have her intention noticed.
(8d) Feeling unwell, Mary deliberately and openly lets her mother see how
pale she is, so she will notice and help.
(8e) Mary says to her mother “I don’t feel well”.
Grice argued that certain intentions must be present before any
instance of meaningNN is dealt with. Firstly, the response itself must be
intended. Consequenlty, (8a) is ruled out as a case of meaningNN; secondly,
the addressee must recognize the intention to produce that response – this
59
rules out (8b), as well; thirdly, the speaker must intend that the audience
should recognize the intention to produce that response—this rules out (8c).
Finally, the crucial condition, the one that rules out (8d), and makes (8e) a
case of meaningNN, is that only in this example does Mary have a true
(Gricean) communicative intention: the intention that the recognition of her
intention to produce the desired response will play a crucial role in producing
the response itself.
The natural meaning is conventional and entailed. The non-natural
meaning is non-conventional and variable, depending on the context. Grice
coined the term implicature to cover any non-conventional meaning that
implied, i.e. conveyed indirectly through hints and understood implicitly
without being explicitly stated.
According to Grice, what a speaker means by an utterance can be
divided into what the speaker “says” (the logical cognitive content, the
minimum necessary to specify the truth conditions of a sentence) and what
the speaker thereby “implicates”. Grice coined the term implicature cover
any non-conventional meaning that is implied, i.e. conveyed indirectly
through hints and understood implicitly without being explicitly stated.
Consider the following example in (9):
(9)
A: How did Manchester United play last night?
B: They won.
On some occasions, B might intend A to understand that Manchester United
played brilliantly, if they played, for instance, in a European competition.
Conversely, in other contexts, B might implicate that they played rather
poorly, if they played a non-league side in the FA Cup. In both cases,
however, it is clear that whatever B implied, suggested or meant is distinct
from what is said, which was simply that the team referred to by they scored
more goals than the team they were playing against. And yet this
conventional meaning is often less important than the other meaning that A
understands from B’s utterance and B nowhere states.

4.3 The Cooperative Principle and the maxims of conversation

Grice argued that speakers intend to be cooperative when they engage in

conversation. One way of being cooperative is for a speaker to give as much
information as is expected. Thus an addressee who knew that Manchester
United were playing a top team in a European competition might be
60
expecting the speaker to say that they had done quite well considering the
circumstances. Since they won would be more than was expected, the
addressee might implicate that Manchester United had played brilliantly.
Conversely, an addressee who knew that Manchester United were playing a
non-league side might be expecting the speaker to say that they had scored
dozens of goals or that they had wiped the opposition out. Hearing only they
won, less than might be expected, the addressee would draw the inference
that they played rather poorly. Because they won in the first context is more
than the addressee was expecting and in the second less, in each case it gives
rise to a non-conventional meaning which was called by Grice implicature.
He developed an influential theory to explain and predict
conversational implicatures, and describe how they arise and are understood.
According to Grice (1975), there is a set of assumptions which guide
conversational exchanges and which can be formulated as guidelines for
efficient and effective use of language in conversation to further cooperative
ends. Grice identified as guidelines of this sort the cooperative principle
and its four maxims of conversation. The cooperative principle states:
“Make your conversational contribution such as is required at the
stage at which it occurs, by the accepted purpose of the talk exchange
in which you are engaged.”
(Grice 1989)
This general principle manifests itself, according to Grice, in a set of Maxims
of Conversation that govern rational interchange and bridge the gap between
what is said and what is meant, i.e. between natural and non-natural
meaning.

Quantity
(i) Make your contribution as informative as is required (for the current
purposes of the exchange)
(ii) Do not make your contribution more informative than is required

Quality
Make your contribution one that is true.
(i) Do not say what you believe to be false.
(ii) Do not say that for which you lack adequate evidence
Relation: Be relevant
61
Manner: Be perspicuous
(i) Avoid obscurity of expression
(ii) Avoid ambiguity
(iii) Be brief (Avoid unnecessary prolixity)
(iv) Be orderly
Consider example (10) again, repeated here for convenience and the
inferential chain proposed by Grice in order to calculate conversational
implicatures.

(10)
A: Can I get petrol somewhere around here?
B: There’s a garage around the corner.
[A can get petrol at the garage around the corner.]

According to Grice (1975 [1989] :39-40), B in (10) would violate the Maxim
of Relation (Be relevant!) if she were to mean only what she says: B would,
in such a case, be uncooperative. Thus, assuming that B adheres to the CP,
we must interpret her utterance in (10) as carrying a conversational
implicature: B means to convey more than what she says. More specifically,
B must, in this particular case, be interpreted as conversationally implicating
that A can get petrol around the corner, for otherwise she would convey
irrelevant information. According to Grice, the implicature generated by the
conversational exchange in (10) can be calculated as follows:

Calculation of Relation Implicature in (10):

B just said that there is a garage around the corner in response to my

question whether I can get petrol somewhere around here. That information
is irrelevant for my purposes, unless the garage is open and I can get petrol
there. B is cooperative and wouldn’t respond with irrelevant information to
my question. Moreover, B has done nothing to prevent me from thinking that
I can get petrol at the garage around the corner.

(Grice 1975 [1989] :39-40).

4.4 Generalized vs. particularized conversational implicatures

The CP and the maxims are neither sociological generalizations nor

prescriptions for proper conversational etiquette. Rather they are underlying

62
rational or logical principles that, by their observance or their apparent
violation, generate conversational implicatures. When these principles are
not adhered to on a superficial level, hearers still assume that they are
adhered to at some deeper level. The inferences that arise in order to preserve
the assumption of cooperation and to bridge what is said to what is meant are
called by Grice conversational implicature.
There are two ways in which conversational implicatures can be
created: speakers may either abide by the maxims, in which case they
generate what Levinson (1983) calls standard implicature, or they may
flout one or several maxims, giving thus rise to floutings or exploitations.
Grice further distinguished between generalized and particularized
conversational implicatures. Generalized conversational implicatures
arise irrespective of the context in which they occur. In other words, they do
not depend on particular features of the context and if any of those features
changes this does not trigger a change in the inferred meaning. The
utterances in (11) illustrate generalized conversational implicatures.
(11a) I walked into a house.
Conversational implicature: The house is not mine.

(11b) Bill has got some of Chomsky’s papers.

Conversational implicature: Bill hasn’t got all of Chomsky’s papers.

(11c) Fred thinks there is a meeting tonight.

Conversational implicature: Fred doesn't know for sure that there is a
meeting tonight.

(11d) Mary has 2 children.

Conversational implicature: Mary has no more than 2 children.

The utterances in (11) always give rise to the same implicature, no matter
what the context.
Particularized conversational implicatures, on the other hand, are
derived not from the utterance alone, but from the utterance in context.
Consider the excerpt in (12):
(12)
A: What on earth has happened to the roast beef?
B: The dog is looking very happy?
63
The state of looking very happy in reference to the dog would ordinarily not
convey anything about the roast beef. So the implicature in this case depends
on the context as well as the utterance itself. In this particular context the the
utterance The dog is looking very happy may generate the implicature
Perhaps the dog has eaten the roast beef.
Generalized conversational implicatures are inferred irrespective of
the context of utterance and result from the speaker’s abiding by maxims of
Quantity and Manner. Particularized conversational implicatures are inferred
in relation to a particular context and result from the existence of the Maxim
of Relation. In both cases, however, these inferences arise from the
assumption that the speaker is observing the maxims of conversation and the
CP. Thus both generalized and particularized conversational implicatures can
be regarded as instances of standard implicature.
Conversational implicatures can also be derived on the basis of the
speaker’s intentionally or unconsciously flouting or exploiting a maxim (i.e.
on the speaker’s not abiding by the maxim) as in the following examples
where the speaker flouts the maxim of quantity:
(13a) If he does it he does it.
Conversational implicature: It’s no concern of ours.

(13b) War is War.

Conversational implicature: Terrible things may happen.

Floutings or exploitations of the maxim of quality give rise to irony and

metaphor.
Grice proposes a definition of implicature which can be stated as follows:
(14)
S saying that p conventionally implicates q iff:
(i) S is presumed to be observing the maxims, or at least (in the case of
floutings) the co-operative principle
(ii) in order to maintain this assumption it must be supposed that S thinks
that q
(iii) S thinks that both S and the addressee H mutually know that H can
worked out that to preserve the assumption in (i), q is in fact required

64
 Moreover, Grice argues that, for the addressee H to be able to
calculate the implicature q, H must know, or believe that he knows the facts
in (15):
(15)
(i) the conventional content of the sentence (P) uttered
(ii) the co-operative principle and its maxims
(iii) the context of P (e.g. its relevance)
(iv) certain bits of background information (e.g. P is blatantly false)
(v) that (i) – (v) are mutual knowledge shared by speaker and addressee
From all this a general pattern of working out an implicature can be
adduced:
(16)
(i) S has said that p
(ii) there’s no reason to think S is not observing the maxims, or at least the
co-operative principle
(iii) in order for S to say that p and be indeed observing the maxims or the
co-operative principle, S must think that q

4.5 Properties of conversational implicatures

Grice isolates the following five characteristic properties of conversational
implicatures:
1. They are cancellable or defeasible
The notion of defeasibility is crucial in pragmatics since it is exhibited by
various kinds of inferences. An inference is defeasible if it is possible to
cancel it by adding some additional premises the original ones.
“a putative conversational implicature that p is explicitly cancelable
if, to the form of words the utterance of which putatively implicates
that p, it is admissible to add but not p, or I do not mean to imply that
p, and it is contextually cancelable if one can find situations in which

65
 the utterance of the form of words would simply not carry the
implicature”
(Grice 1989:44)
Consider the example (17) and its straightforward implicature (18) which
results from the Maxim of Quantity:
(17) Mary has three cats.
(18) Mary has only three cats and no more.

Notice that (17) entails (19)

(19) Mary has two cats.

Implicatures can be cancelled by mention in an if clause:
(20) Mary has three cats, if not more.

(20) no longer carries the implicature in (18). Entailments, being non-

defeasible, cannot be suspended in a similar way:

(21) # Mary has three cats, if not two

Moreover, implicatures are overtly and directly deniable without any sense
of contradiction:
(22) Mary has three cats, in fact five
(23) Mary has three cats and maybe more

2. They are non-detachable

The second important property of implicatures is that (with the exception of
those arising from maxim of Manner as they are specifically linked to the
form of the utterance) they are non-detachable. By this Grice means that the
implicature is attached to the semantic content of what is said, not to the
linguistic form, and therefore the implicatures can be detached from an
utterance simply by changing the words of the utterance for synonyms.
“it will not be possible to find another way of saying the same thing,
which simply lacks the implicature in question, except where some
special feature of the substituted version is itself relevant to the

66
 determination of an implicature (in virtue of one of the maxims of
manner)”
(Grice 1989:39)
As an illustration of this property, consider the ironic interpretation (25) of
(24):
(24) John’s a genius.
(25) John’s an idiot.

The same implicit meaning can be conveyed by any of the sentences in (26)
in a context in which it is mutually known that (24) is false:

(26a) John’s a mental prodigy.

(26b) John’s an exceptionally clever guy.
(26c) John’s an enormous intellect

3. They are calculable

As Grice (1989:31) argues, “the presence of a conversational implicature
must be capable of being worked out; for even if it can in fact be intuitively
grasped, unless the intuition is replaceable by an argument, the implicature
(if present at all) will not count as a conversational implicature”.
For any implicature it is possible to construct an argument of the type
of in (16) showing how from the literal meaning or the sense of the
utterance, on the one hand, and the co-operative principle and the maxims,
on the other, it follows that an addressee will make the inference in question
to preserve the assumption of co-operation.

4. They are non-conventional

They are not part of the conventional meaning of linguistic expressions. A
linguistic expression conveying a single meaning can give rise to different
conversational implicatures in different contexts. Since one has to know the
literal meaning or sense of a sentence before one can calculate its implicature
in a context, the implicature cannot be part of that meaning. Moreover, it can
be shown that an utterance can be true while its implicature can be false, and
vice versa, as in:
(27) Herb hit Sally.
which by Quantity would implicate
67
(28) Herb didn’t kill Sally by hitting her.
(since if Herb killed Sally, the speaker would, in saying just (27) be
withholding information in a non-cooperative way); but a speaker might say
(27), attempting to mislead, in a situation in which (27) is true, but (28) is
false.
5. They are context-dependent
A linguistic expression conveying a single meaning can give rise can give
rise to different implicatures on different occasions when various aspects of
the context are changed. Moreover, on any one occasion the set of associated
implicatures may not be exactly determinable. An example such as the one
given in (29)
(29) John’s a machine.
Could convey that John is cold, efficient, or never stops working, or has little
in the way of grey matter. Thus implicatures can evince certain
indeterminacy in at least some cases, incompatible with the stable senses
usually assumed in semantic theories.

4.6 Scalar implicatures

The earliest published indication of the phenomenon which is nowadays
widely known as scalar implicature appeared in 1952, in a footnote to Peter
Strawson’s Introduction to Logical Theory. In discussing the relationship
between the statement ‘there is not a book in his room which is not by an
English author’ and the assumption ‘there are books in his room’, Strawson
draws attention to the need to distinguish between strictly logically relations
and the rules of ‘linguistic conduct’. He proposes as one such rule the
following one: ‘one does not make the (logically) lesser when one could
truthfully (and with equal or greater linguistic economy) make the greater,
claim.’ It would be misleading, although not strictly false, to make the less
informative claim about English authors if in a position to make the much
more informative claim that there are no books at all. Strawson
acknowledges that ‘the operation of this “pragmatic rule” was first pointed
out to me, in a different connection, by Mr H. P. Grice.’
Grice (1975) identifies a meaning-creating clash between quantity and
quality, with respect to a direct question:
(30)
68
A is planning with B an itinerary for a holiday in France. Both know that A
wants to see his friend C, if to do so would not involve too great a
prologation of his journey.
A: Where does C live?
B: Somewhere in the South of France
[B does not know in which town C lives]12
Levinson (1983) defines quantity implicatures in a way that assumes
relevance (clause (31b) and then singles out the speaker’s lack of knowledge
as the driving force behind the implicature (31d):
(31)
a. S has said p
b. There is an expression q, more informative than p (and thus q entails p),
which might be desirable as a contribution to the current purposes of the
exchange (and here there is perhaps an implicit reference to the maxim of
Relevance)
c. q is of roughly equal brevity to p; so S did not say p rather than q simply
in order to be brief (i.e. to conform to the maxim of Manner)
d. Since if S knew that q holds but nevertheless uttered p he would be in
breach of the injunction to make his contribution as informative as is
required, S must mean me, the addressee, to infer that S knows that q is not
the case (K~ q), or at least that he does not know that q is the case (~ Kq).
(Levinson 1983)
Horn (1989:232ff) gives Horn’s (1972) original definition of scalar
implicatures, which makes use of scales like the above, along with critical
discussion of its details, and critical discussion of subsequent attempts by
others to define this notion. Levinson (1983:134) defines a conversational
implicature that is triggered by a violation of Quantity1 based on the use of

12
Gloss: There is no reason to suppose B is opting out; his answer is, as he well knows, less
informative than is required to meet A’s needs. This infringement of the first maxim of
Quantity can be explained only by the supposition that B is aware that to be more
informative would be to say something that infringed the maxim of Quality, ‘Don’t say what
you lack evidence for’, so B implicates that he does not know in which town C lives.

69
an informationally weak term on an implicational scale. Consider the
following standard examples of scalar implicature:

(32)
A: Who ate the cookies?
B: I ate some of the cookies.
[B didn’t eat all of the cookies.]

(33)
A: Who is the best in class?
B: John is sometimes the best in class.
[John isn’t always the best in class.]

While Grice was aware that implicatures such as those in (32) and
(33) are generalized conversational implicatures, it is important to emphasize
that they are generalized conversational implicatures of a rather special type.
To see their special status it should be pointed out that the lexical items some
and sometimes are members of what Horn (1972) calls an implicational
scale – that is, a set of lexical items that form a linear ordering according to
their informational (or even, as in the case of ‘some’, logical) strength.
Consider the following examples of implicational scales or Horn scales:

(34)

<all, most, some>

<always, often, sometimes>
<certain, probable, possible>

Utterances of sentences that contain an item at a lower point on the scales in

(34) usually conversationally implicate the negation of the propositions
expressed by analogous sentences containing an item from further up the
scale. For instance, an utterance of the sentence in (32) expresses the
proposition that B ate some of the cookies, which is logically and
informationally weaker than the propositions that B ate many, most, or all of
the cookies. Thus, when B says or asserts that she ate some of the cookies,
she usually conversationally implicates that she did not eat many, most, or all
of them. A similar line of reasoning applies to (33). Thus, scalar implicatures
are cases of utterances in which an informationally weaker content is
asserted to conversationally implicate that an informationally stronger
content is false.

70
 Horn scales, defined first in Horn (1972), are more or less
conventionalized scales of lexical items organized by informativity (in some
sense). Levinson (1983) gives the examples in (35), which should be handled
with care (some theorists argue, for instance, that few should not be included
in the first one, since it has different polarity/monotonicity than the others):

(35) Scales from Levinson (1983:134)

<all, most, many, some, few>

<and, or>
<n, . . . , 5, 4, 3, 2, 1>
<excellent, good>
<hot, warm>
<always, often, sometimes>
<succeed, Ving, try to V, want to V>
<necessarily p, p, possibly p>
<certain that p, probable that p, possible that p>
<must, should, may>
<cold, cool>
<love, like>
<none, not all>

A few other standard lexical scales are given in (36)

(36)

<first, second, third, fourth, fifth>

<definite, indefinite>
<lover, friend>
<need, want>
<old, middle-aged, young>
<general, colonel, major, captain, …>

Some theorists distinguish between the more or less lexicalized scales that
follow from some general definition of entailment and the more
particularized – “pragmatic” – scales that can crop up with particular
speakers, or given particular discourse situations.
Even for apparently standard lexical scales like those in (35) and
(36), one needs to be aware that the order of the scale is context dependent:

71
(37)
a. <cold,warm, hot> coffee/champagne
b. <gain, lose >20 pounds while dieting/weight-lifting
c. <gain, lose> 20 dollars

In addition, a given word can have different scalar peers in different

contexts:

(38)
a. morality: <bad, evil>; food: <bad, rotten>; abilities: <bad, useless>
b. morality: <good, divine>; food: <good, delicious>; abilities: <good,
skilled>

Despite the conventional aspect of scalar implicatures – they are,

after all, partly triggered by the presence of lexical items that are
semantically associated with implicational scales – it is important to
emphasize that scalar implicatures are conversational and not conventional
implicatures. This can be demonstrated by their behavior in relation to
diagnostic tests that show that scalar implicatures are cancellable, non-
detachable, and calculable. Consider firstly constructions cancelling the
implicatures in (32) and (33):

(32’) I ate some of the cookies. In fact, I ate all of them.

(33’) John is sometimes the best in class. In fact, he always is.

The implicatures in (32) and (33) are also non-detachable:

(32*)
A: Who ate the cookies?
B: I ate at least one of the cookies.
[B didn’t eat all of the cookies.]

(33*)
A: Who is the best in class?
B: At times, John is the best in class.
[John isn’t always the best in class.]

Finally, scalar implicatures are calculable: in the standard Gricean

framework, scalar implicatures are triggered by potential violations of the
Maxim of Quantity1 according to which speakers must make their

72
contribution to the conversation as informative as is required13. The
implicature in (14/32), for instance, can be calculated as follows:

Calculation of Scalar Implicature in (32):

In response to my question of who ate the cookies, B just said that she ate
some of them. If B had eaten all of the cookies, then her contribution to our
conversation would not be as informative as is required. But B is cooperative
and wouldn’t keep information from me that I asked her for. Moreover, B
has done nothing to prevent me from thinking that she didn’t eat all of the
cookies. So that must be what B meant to convey by her utterance.

As further reading, I recommend Hirschberg (1985), still the most

comprehensive treatment of this topic, Levinson (2000), Geurts (2009) and
Sauerland (2010, 2012) for an overview of recent work on scalar
implicatures, including alternative views to and a critical discussion of the
Gricean approach briefly presented in this section.

4.7 Conventional implicatures

Grice (1975) defined conventional implicatures (CIs) specifically to
distinguish them from conversational implicatures, his domain of inquiry.
Conversational implicatures exist in virtue of the maxims and the
cooperative principle, whereas CIs are idiosyncratic properties of the
grammar. Put another way, conversational implicatures are not inherently
linguistic, whereas CIs are inherently linguistic.
CIs cannot be teased apart from the lexical items that produce them,
nor can we understand where they arise by appeal to the nature of the context
and the maxims. The result is that they are not contextually variable. In other
words, conventional implicatures are independent of the Cooperative
Principle and its maxims. A conventional implicature is simply attached by
convention to a particular lexical item or phrase.
According to Grice, calculability is the crucial property that
distinguishes between conversational and conventional implicatures. A
conventional implicature is, as Grice says, not “calculable.” The basic tenet
of conventional implicature (CI-) thesis is that there are certain locutions

13
Scalar implicatures also allow for what Horn has called metalinguistic negation: ‘A didn’t
eat some of the cookies, she ate all of them!” For interesting discussion see (Horn 1984;
1989, pp. 362-375).

73
which do not affect the truth or falsity of what is said but they generate
implicatures, by virtue of their conventional meaning.
Grice is usually credited with the discovery of conventional
implicature. However, it was actually Frege’s idea and Grice merely labeled
it. In “On Sense and Reference”, Frege pointed out:

Subsidiary clauses beginning with ‘although’ … express complete

thoughts. This conjunction actually has no sense and does not change
the sense of the clause but only illuminates it in a peculiar fashion.
[footnote: Similarly in the case of ‘but’ and ‘yet’.] We could indeed
replace the concessive clause without harm to the truth of the whole
by another of the same truth value; but the light in which the clause is
placed by the conjunction might then easily appear unsuitable, as if a
song with a sad subject were to be sung in a lively fashion.
(Frege 1918/1994, p. 155)
Much later, in “The Thought: A Logical Inquiry”, Frege puts his idea this
way:
With the sentence ‘Alfred has still not come’ one really says ‘Alfred has not
come’ and, at the same time, hints that his arrival is expected, but it is only
hinted. It cannot be said that, since Alfred’s arrival is not expected, the sense
of the sentence is therefore false. … The word ‘but’ differs from ‘and’ in that
with it one intimates that what follows is in contrast with what would be
expected from what preceded it. Such suggestions in speech make no
difference to the thought.
(Frege 1918/1994: 522; my italics)
Thus, still and but have no bearing on the truth or falsity of what is said.
Grice makes a similar point about therefore:
If I say (smugly), He is an Englishman; he is, therefore, brave, I have
certainly committed myself, by virtue of the meaning of my words, to its
being the case that his being brave is a consequence of (follows from) his
being an Englishman. But while I have said that he is an Englishman, and
said that he is brave, I do not want to say that I have said (in the favored
sense) that it follows from his being an Englishman that he is brave, though I
have certainly indicated, and so implicated, that this is so. I do not want to
say that my utterance of this sentence would be, strictly speaking, false
should the consequence in question fail to hold.
74
(Grice 1989: 25)
Grice supplies another example which he argues more plausibly and
which involves but:
(39) She is poor but she is honest.
The statement in (39) implies poverty and honesty are not compatible but in
spite of this the subject is still honest. Grice argues that the putative contrast
between being poor and being honest is “implied as distinct from being
stated” (Grice 196: 127).
But has the same truth-conditional content as the conjunction and.
From a logical point of view the value of but in conjoining two phrases or
sentences is the same as that of and (i.e. the conjoined sentences with but
have the same truth conditions as those with and: one is true in exactly the
same cases as the other). However, but has an additional conventional
implicature, i.e. the implicature of a sense of contrast. So, “She is poor but
she is honest” will always necessarily imply “Surprisingly she is honest in
spite of being poor”.
In the examples considered the propositions are said to be
implicatures because their truth value does not affect the truth value of the
entire utterance, so that the falsity of such a proposition is compatible with
the truth of the entire utterance. So, according to the CI-thesis, the truth of
(3/38/39) would not be affected if there were no contrast, real or presumed,
between being poor and being honest.
On Grice’s conception a conventional implicature is a proposition
which is conveyed due to the presence of a certain term with a certain
meaning but whose falsity is compatible with the truth of the utterance. This
view gained currency among linguists, thanks to Karttunen and Peters
(1979), who proposed it as a replacement for the notion of semantic
presupposition. Karttunen and Peters (1979: 11) argued that “a large set of
cases that have been called presupposition are really instances of
conventional implicature. The most obvious examples are those associated
with particles like too, either, also, even, only, and so on. This class also
includes the presuppositions of certain factive verbs, such as forget, realize,
take into account, and so on, and those that accompany implicative verbs like
manage and fail.” These terms that are said to generate conventional
implicatures are called in the literature “alleged conventional implicature
devices” (ACIDs) (Bach 1999). The following list given in (40) includes,
according to Bach (1999), representatives members:

75
(40)
adverbs: already, also, barely, either, only, scarcely, still, too, yet
connectives: but, nevertheless, so, therefore, yet
implicative verbs: bother, condescend, continue, deign, fail, manage, stop
subordinating conjunctions: although, despite (the fact that), even though

Karttunen and Peters offer one piece of evidence in support of the CI-
thesis, namely the occurrence of even in (41):

(41) Even Bill likes Mary.

They argue that “the truth of what (41) says depends solely on whether Bill
likes Mary” (Karttunen and Peters 1979:12). Suppose (41) is embedded in
(42):
(42) John just noticed that even Bill likes Mary.
The crux of the argument is that (42) “does not mean that he has just noticed
that other people like Mary or just noticed that Bill is the least likely person
to do so” (1979, p. 13). Karttunen and Peters assume that noticing a complex
fact requires noticing its constituent facts. Their reasoning is that since (42)
does not entail that John just noticed that other people like Mary or that Bill
is the least likely person to do so, (42) says merely that John just noticed that
Bill likes Mary, i.e., that ‘even’ does not contribute to what John is being
said to have noticed.
Bach (1999) refutes Karttunen and Peters’ line of argument by
supplying the example (42).

(43) John just noticed that Bill has three cars.

Bach argues that (43) could be true even if John has long known of two of
Bill’s cars and just became aware of a third one.
In Bach’s view, the same line of argument applies to implicative
verbs, such as manage. While functioning syntactically as a main verb,
manage seems to modify the verb in its infinitival complement, entailing that
the action in question requires effort or involves difficulty. Proponents of thr
CI-thesis view this implicitly conveyed meaning a convetional implicature.
Bach argues that this proposition is not a mere conventional implicature, for
it is part of the content of sentences in which manage occurs. The content of
(44), for example, includes both the finishing and the entailed difficulty.

76
(44) Bill managed to finish his homework.
In Bach’s view the same point made above about noticing applies to the
example in (44): one could notice that Bill managed to finish his homework
even if one already knew that finishing it would be difficult.

4.8 Relevance
Over the past twenty years, relevance theory has become a key area of study
within semantics and pragmatics. Relevance theory is an approach to
implicature developed by Sperber and Wilson (1986, 1995) as part of a
broader attempt to shift pragmatics into a cognitive framework. In relation to
implicature, as conceived by Grice (1975), relevance theory can be viewed
as a reductionist theoretical approach for two reasons. Firstly, it reduces all
pragmatic principles that have been proposed to underlie conversational
implicature to a single ‘Principle of Relevance’. Secondly, it reduces all the
different species of meaning in the Gricean/neo-Gricean framework (such as
what is said, conventional implicature, scalar implicature, generalised
conversational implicature, particularised conversational implicature and so
on) to two broad categories: explicature and implicature.
Relevance theory is based on the assumption that human beings are
endowed with a biologically rooted ability to maximize the relevance of in-
coming stimuli (linguistic utterances or nonverbal behavior). Relevance is
not only a characteristic property of external stimuli (e.g. utterances), but
also of internal representations and thoughts, all of which may become inputs
for cognitive processing. Assessing relevance is a typical mental activity of
human beings, always geared to obtaining the highest reward from the
stimuli which they process.
The following sentences summarize Sperber and Wilson’s relevance
theory: (a) in a given context, the decoded meaning of the sentence is
compatible with a number of different interpretations; (b) these
interpretations are graded in terms of accessibility; (c) hearers rely on a
powerful criterion when selecting the most appropriate interpretation; and (d)
this criterion makes it possible to select one interpretation among the range
of possible interpretations, to the extent that when a first interpretation is
considered a candidate to match the intended interpretation, the hearer will
stop at this point.
In what follows I will examine the basic tenets of Sperber and
Wilson’s relevance theory. The central focus of the discussion is on such key
concepts as ostensive-inferential communication, the dichotomy implicature-
explicature, and the notion of relevance.

77
4.8.1 Ostensive-inferential communication

Sperber and Wilson take the Gricean inferential approach to communication

as the starting point, but they disagree with Grice on some aspects. Grice
underlined the crucial role intentions 14 play in human communication. His
emphasis on the expression and recognition of intentions laid the foundations
of the inferential model of communication. However, Sperber and Wilson do
not embrace the complex schema of human reasoning which Grice proposed
for the calculation of conversational implicatures.
Sperber and Wilson argue that communication can exist without the
need for a code. All that the communicator has to do in order to
communicate a thought is to get the addressee to recognize his/her intention
to convey it. The proponents of relevance theory view ostension and
inference as two sides of the same coin. These two concepts refer to the
production and interpretation of certain stimuli respectively. Unlike coding
and decoding, ostension and inference are non-conventional. The addressee’s
attention is drawn to a given fact in order to infer the content of that which
the speaker tries to communicate. In Sperber and Wilson’s own words:

The communicator produces a stimulus which makes it mutually

manifest to communicator and audience that the communicator
intends, by means of this stimulus, to make manifest or more
manifest to the audience a set of assumptions. (1986:63)15

Sperber and Wilson distinguish two levels of intention: informative

(an intention to inform the addressee of a given fact) and communicative
(the intention to inform the addressee of that informative intention). The
former entails the identification of the latter, which is typically activated by
verbal ostensive communication in which it is clear to both speaker and
addressee (mutually manifest in Sperber and Wilson’s terminology) that the
speaker has the intention to communicate the intention to inform the
interlocutor of something. For ostensive communication to be efficient, the
addressee has to realize that the stimulus produced by the communicator (i.e.
speaker) is intentional, i.e. is directed to the addressee and it is a conscious
modification of the environment to draw addressee’s attention to a group of
facts.

14
Intentions can be roughly defined as mental representations of a desired state of affairs.
15
Sperber and Wilson (1986:39) define the term manifest as follows: “A fact is manifest to
an individual at a given time if and only if he is capable at that time of representing it
mentally and accepting its representation as true or probably true.”
78
 Relevance theory explains the addressee’s inference of the speaker’s
intended meaning by resorting to a claim that is central to Grice’s theory of
implicature: ostensively communicated utterances generate expectations
which activate the addressee’s search for the speaker’s intended meaning.
Unlike Grice, who explained these expectations in terms of the assumptions
hearers make that speakers are following the cooperative principle and its
maxims, Sperber and Wilson account for these expectations in cognitive
terms and propose a Cognitive Principle of Relevance, without relying the
Co-operative Principle.
In Sperber and Wilson’s view, Gricean maxims are required to bridge
the gap between what is said and what is meant. Sperber and Wilson have
shown that people are normally loose when they speak and only on very
specific occasions do they intend their utterances to be regarded as literally
true. They propose a single explanatory framework based on general
expectations of relevance that will account for all loose uses of language
(metaphor, hyperbole, irony, vagueness, etc.).

4.8 2.The concept of relevance

According to Sperber and Wilson, for a piece of information to be relevant, it

must produce some effect on the addressee’s cognitive environment. These
effects are called contextual effects and they are said to be of three types:
contextual implications, strengthening of existing assumptions and
contradiction of existing assumptions.
Conceptual implications are inferences which follow from the
combination of the propositional content of an utterance and its contextual
assumptions. Consider the example in (45):

(45)
A: Are you coming to the rock concert?
B: I’ve got a meeting at half past six.

If this all the information that we have, we cannot know with any certainty
what B implies by delivering the utterance in (45). We are not certain
whether she will go to the rock concert or not. If (45) is processed in a
context containing the assumptions in (46):

(46)
a. The rock concert starts at 7 o’clock and finishes at 8 o’clock.
b. B’s meeting will last at least one hour.

79
From the utterance in (45) considered within context in (46), the contextual
implication in (47) will follow:

(47)
B is not going to the rock concert.

Thus, (47) does not follow from the propositional content of (45) alone or
from the assumptions in (46), but from the inferential combinations of both.
The example in (48) illustrates the strengthening of existing
assumptions:

(48)
A: I have the impression that Paul’s new girlfriend is a foreigner.
B: I guess she is, she speaks with a French accent.

In (48), A indicates that he is not totally sure of the truth of his utterance.
Assuming that A’s context contains the following premise:

(49) Someone who speaks with a French accent is a foreigner.

then B’s utterance supplies information that can serve as further evidence
that supports the truth of A’s assumption.
The contradiction of existing assumptions can be illustrated by the
exchange in (50):

(50)
A: I think Bill and Jane have split up.
B: Nope, they are just coming down the street kissing each other.

In (50), B supplies information that proves A’s assumption is wrong. The

clash between assumptions will be solved in favour of the strongest one,
since “information available from perception is usually assigned much
greater strength than information based on inference” (Gutt 1991: 29).
Sperber and Wilson (1986) point out that an utterance is not relevant
unless it yields some contextual effects. However, this not enough since
relevance is not an absolute notion, but a relative one. The contextual effects
of an utterance must be related to the effort necessary to achieve those
effects. They argue that “other things being equal, an assumption with
greater contextual effects is more relevant; and, other things being equal, an
assumption requiring a smaller processing effort is more relevant” (Sperber
and Wilson 1986: 125).
80
 Relevance is not an intrinsic feature of utterances. It is property
derived from the relationship between a given utterance and the addressee’s
assumptions in a particular situation. What may be relevant for somebody at
a given moment may not be relevant for somebody else or for the same
person in a different situation.
The intention to communicate is base on the fact that the speaker
intends to modify the hearer’s cognitive environment in some way. This is
called the presumption of optimal relevance and it has been defined by
Sperber and Wilson in the following terms:

Presumption of optimal relevance:

(a) The set of assumptions {I} which the communicator intends to make
manifest to the addressee is relevant enough to make it worth the addressee’s
while to process the ostensive stimulus.
(b) The ostensive stimulus is the most relevant one the communicator could
have used to communicate {I}. (Sperber and Wilson 1986: 158)

From this, the principle of relevance is derived and formulated as follows:

Principle of relevance:

Every act of ostensive communication communicates the presumption of its

own optimal relevance (Sperber and Wilson 1986: 158)

Thus, the principle of relevance expresses the assumption will make his
utterance as relevant as possible in the circumstances in which it is produced.
This does not necessarily mean that a satisfactory degree of relevance is
always achieved. Some utterances do not yield any contextual effect and
consequently they are not relevant.

4.8.3 Implicatures and explicatures

One of the key differences between Grice’s model and Sperber and Wilson’s
lies in the distinction between explicatures and implicatures. The meaning
explicitly communicated by means of an utterance is an explicature and the
content that is derivable from the proposition expressed by the utterance in a
context called an implicature. The term explicature is used by Sperber and
Wilson to cover aspects of meaning which Grice included in the term
implicature (e.g. the so-called generalized conversational implicatures,
most of which are now pictured as explicit information, see Carston, 2002).
81
Sperber and Wilson’s (1986/95, 182) definitions are as follows:

(I) An assumption communicated by an utterance U is explicit [hence an

‘explicature’] if and only if it is a development of a logical form encoded by
U. [Note: in cases of ambiguity, a surface form encodes more than one
logical form, hence the use of the indefinite here, ‘a logical form encoded by
U’.]

(II) An assumption communicated by U which is not explicit is implicit

[hence an ‘implicature’].

To illustrate this distinction, consider the example in (51):

(51)

A: How is Mary feeling after her first year at university?

B: She didn’t get enough units and can’t continue.

On the basis of the definitions above, it seems relatively clear that (52a) is an
explicature of B’s utterance and (52b) is an implicature.

(52) a. Mary did not pass enough university course units to qualify for
admission to second year study and, as a result, Mary cannot continue with
university study.
b. Mary is not feeling very happy

Sperber and Wilson propose two types of explicitly communicated

information: the basic-level explicature, and the higher-order explicature.
The latter also includes the speaker’s attitude (to regret that…, to be happy
that…, etc.) or a higher-order speech-act schema (to be asking that…, to be
ordering that…, etc.). Both explicatures and implicatures allow for degrees
(i.e., strong and weak explicatures/implicatures), depending on the
addressee’s responsibility for their derivation and the amount of mental
processing required.
To conclude this section on relevance theory, it should be pointed out
that it is based upon the following two tenets: (i) the ostensive stimulus is
relevant enough for it to be worth the addressee’s effort to process it; and (ii)
the ostensive stimulus is the most relevant one compatible with the
communicator’s abilities and preferences (Sperber and Wilson, 1995: 267
and 270). As Wilson and Sperber (2002a: 257-258) correctly point out,
communicators “cannot be expected to go against their own interests and
82
preferences in producing an utterance. There may be relevant information
that they are unable or unwilling to provide, and ostensive stimuli that would
convey their intentions more economically, but that they are unwilling to
produce, or unable to think of at the time”. All this is covered by the
definition of optimal relevance, which states that the ostensive stimulus is the
most relevant one “that the communicator is WILLING AND ABLE to
produce” (ibid., 258).

4.9 Conclusions

Grice’s concept of conventional implicatures is the most controversial part of

his theory of conversation. The category of conventional implicatures blurs
the distinction between what is said, usually conceived as determined by the
semantic conventions of language, and what is implicated, usually thought of
as a matter of inference as to a speaker’s intentions in saying what he or she
does. Conventional sentence meaning contributes crucially to what is said,
which is considered essentially different from implicatures; but now we have
the result that some elements of conventional meaning do not contribute to
what is said but to implicatures (albeit conventional) (Bach 1999). Thus, for
some expressions, it places the study of the conventional meaning within the
scope of pragmatics (the study of intended meaning), rather than semantics
(the study of conventional meaning).

83
 CHAPTER 5

THE THEORY OF SPEECH ACTS

5.1 Introduction
Making a statement may be the paradigmatic use of language. However,
there are all sorts of other things we can do with words such as from
aspirating a consonant, constructing a relative clause, insulting a guest,
starting a war, making a declaration of love, making a promise, giving
thanks, etc. Pre-theoretically, these are acts done in the process of speaking.
The theory of speech acts, however, is especially concerned with those acts
that are not completely covered under one or more of the major divisions of
grammar – phonetics, phonology, morphology, syntax, semantics – or under
some general theory of actions (Sadock 1974).
Within a theory of speech acts, any speech act is actually the
simultaneous performance of several acts, distinguished by different aspects
of the speaker’s intention: there is the act of saying something, what one
does in saying it, such as requesting or promising, and how one is trying to
affect one's audience. The theory of speech acts is partly taxonomic and
partly explanatory. It offers a systematic classification of types of speech acts
and it accounts for the ways in which they can succeed or fail. A major task
for the theory of speech acts is to account for how speakers can succeed in
what they do despite the various ways in which linguistic meaning
underdetermines use (Bach and Harnish 1979).
Real-life acts of speech usually involve interpersonal relations of
some kind. By saying certain words to an audience, a speaker affects I a
certain way the interpersonal relation with that audience. Thus it would seem
that ethnographic studies of such relationships and the study of discourse
should be central to speech act theory, but in fact, they are not. Such studies
have been carried out rather independently of the concerns of those
philosophers and linguists who focused their attention on speech acts. This is
perhaps not a good thing, as Croft (1994) has argued, but since it is the case,
anthropological and discourse-based approaches to speech acts will not be
covered in this chapter.

84
5.2 Austin’s theory of speech acts

The modern study of speech acts begins with Austin’s (1962) widely
acclaimed study How to Do Things with Words, the published version of his
William James Lectures delivered at Harvard in 1955. In his William James
Lectures, Austin attacked the doctrine of logical positivism which flourished
in the 30s. According to logical positivism, unless a sentence can be tested
for its truth or falsity, it is strictly speaking meaningless. Thus according to
this doctrine, most ethical, aesthetic and literary discourses, not to mention
most everyday utterances are viewed as meaningless.
Austin argues that certain sorts of sentences, e.g., I christen this ship
the Joseph Stalin; I now pronounce you man and wife; and the like, seem
designed to do something, here to christen and wed, respectively, rather than
merely to say something which can be assed as true or false. Thus, these
sentences are not used to describe states of affaires, but are rather actively
used to do things. Austin refers to these special sentences and the utterances
realized by them as performatives, in contrast to what he calls constatives,
the descriptive sentences that until Austin were the principal concern of
philosophers of language. Performatives, unlike constatives which can be
assessed in terms of truth or falsity, cannot be true or false.
While the distinction between performatives and constatives is often
invoked in the literature, by the end of the book Austin revises this
distinction. Austin argued that this distinction is not ultimately defensible.
The point of Austin’s lectures is, in fact, that every ordinary utterance has
both a descriptive and an effective aspect: that saying something is also
doing something.

5.2.1 Locutions, illocutions, and perlocutions

The initial distinction between constatives and performatives is replaced by

Austin with a three-way contrast among the kinds of acts that are performed
when language is put to use. Austin isolates three basic senses in which in
saying something one is doing something, and hence three kinds of acts that
are simultaneously performed.
Locutionary acts, according to Austin, are acts involved in the
production of speech, such as uttering certain sounds or using particular
words and using them in conformity with the grammatical rules of a
particular language and with certain senses and certain references as
determined by these rules. Thus, a locutionary act has a locutionary
meaning.

85
 Illocutionary acts, Austin’s central innovation, are acts done in
speaking (hence illocutionary), including and especially that sort of act that
is the apparent purpose for using a performative sentence: christening,
marrying, and so forth. Illocutionary acts have illocutionary forces indicated
by the performative verb they include or by their explicit performative
paraphrase.
The third of Austin’s categories of acts is the perlocutionary act,
which can be viewed as a by-product of speaking, whether intended or not.
The perlocutionary act is the bringing about of effects on the audience. The
perlocutionary act has perlocutionary effects. According to Austin,
perlocutionary acts consist in the production of effects upon the thoughts,
feelings, or actions of the addressee(s), speaker, or other parties, such as
causing people to refer to a certain ship as the Joseph Stalin, producing the
belief that Sam and Mary should be considered man and wife, convincing an
addressee of the truth of a statement, etc.
To take an example, consider the utterance in (1) uttered by a
husband to his wife at a party:
(1) It’s getting late
The locutionary meaning of the utterance in (1) is a simple reference to the
lateness of the hour; the illocutionary force of the act, which constitutes a
suggestion or a proposal, is equivalent to saying ‘Let’s go home now’; the
perlocutionary effect may be that of persuading the addressee to perform the
suggested action.
Though it is crucial under Austin’s system that we be able to
distinguish fairly sharply among the three types of acts, in practice the
distinction is often beset with problems and it is difficult to draw the
requisite lines. Austin argues that the locutionary act and the illocutionary act
are detachable and therefore the study of meaning may proceed
independently but is necessarily supplemented by a theory of illocutionary
acts. More troublesome is the distinction between the illocutionary act and
the perlocutionary act.
Austin suggests an operational test that allows us to distinguish
between the illocutionary and the perlocutionary acts. Under this test, if the
hypothetical illocutionary force can be paraphrased as an explicit
performative, then the act performed is an illocutionary act; if this is not
possible , then the act performed is a perlocutionary act.
Austin (1962:101) illustrates the distinction between these kinds of
acts with the example in (2).

86
(2) Shoot her!

He trisects the utterance as follows:

Locutionary Act

He said to me “Shoot her!” meaning by shoot “shoot” and referring by her to

“her.”

Illocutionary Act

He urged (or advised, ordered, etc.) me to shoot her.

Perlocutionary Act

He persuaded me to shoot her.

Austin suggests the following operational test that helps us

distinguish between the illocutionary and the perlocutionary acts: if the
hypothetical illocutionary force can be paraphrased as an explicit
performative, then the act performed is an illocutionary act; if this is not
possible, then the act performed is a perlocutionary act.
Austin’s proposal for discriminating between an illocution and a
perlocution is that the former is conventional, in the sense that at least it can
be made explicit by the performative formula; but the latter cannot (Austin
1962:103). This, however, is more a characterization of possible
illocutionary acts than a practicable test for the illocution of a particular
sentence or an utterance of it. As Saddock (1977) points out, while the test
can give direct evidence as to what is not an illocutionary act, it fails to tell
us what the illocution is. If, for example, someone says delivers the utterance
in (3):

(3)The bull is about to charge!

and thereby warns the addressee of impending danger, the utterance in (3)
can be interpreted as a speech act of warning, i.e. an illocutionary act of
warning because the explicit performative paraphrase could be “I warn you
that the bull is about to charge”. Another reasonable interpretation would be
that in this case, the warning of the addressee, i.e., the production of a feeling
of alarm, is a perlocutionary by-product of asserting that the bull is about to
charge. Many authors, such as Searle (1969, 1975a) and Allan (1998), seem
87
to accept the idea that potential expression by means of a performative
sentence is a sufficient criterion for the recognition of illocutions, while
others, e.g., Sadock (1977), do not.

5.2.2 Felicity conditions

An important aspect of Austin’s theory of speech acts concerns the situations

when performatives can go wrong. It appears that constatives are just those
utterances that are false when they fail, whereas failed performatives are not
aptly described as false, but rather as improper, unsuccessful, or, in general,
infelicitous. For instance, if someone says:

(4) I christen this ship Queen Elisabeth

he may not succeed in so christening the vessel, if, for instance, it is already
named otherwise or his not the appointed namer or there are no witnesses or
bottles of champagne.
Performatives have to meet certain conditions if they are to succeed
or be happy. Austin refers to these conditions as felicity conditions and
distinguishes the following three categories:

A. (i) There must be a conventional procedure having a conventional

effect; (ii) the circumstances and persons must be appropriate, as
specified in the procedure.
B. The procedure must be executed (i) correctly and (ii) completely.
C. The person must have the requisite thoughts, feelings and intentions
as specified in the procedure, and if consequent conduct is specified
then the relevant parties must so do.

Violations of these conditions are all of equal status. Austin distinguished

three broad categories of infelicities: misinvocations, misexecutions and
abuses. Violations of A and B give rise to misfires i.e. the intended actions
fail to come off and consequently the act is void, as illustrated by the
following examples:

A. Misinvocations, which disallow a purported act. For example, a random

individual saying the words of the marriage ceremony is disallowed from
performing it. Similarly, no purported speech act of banishment can succeed
in our society because such an act is not allowed within it.
88
e.g., the clergyman baptizing the wrong baby or the right baby with the
wrong name (violation of A);

B. Misexecutions, in which the act is vitiated by errors or omissions,

including examples in which an appropriate authority pronounces a couple
man and wife, but uses the wrong names or fails to complete the ceremony
by signing the legal documents. Here, as in the case of misinvocations, the
purported act does not take place.

e.g, the absence of an uptake in the case of bets i.e. the procedure is not
carried out completely (violation of B)

C. Abuses, where the act succeeds, but the participants do not have the
ordinary and expected thoughts and feelings associated with the happy
performance of such an act. Insincere promises, mendacious findings of fact,
unfelt congratulations, apologies, etc. come under this rubric.

At this point, Austin’s theory of speech acts can be summed up along the
following lines:

- some sentences, performatives, are special: uttering them DOES

things and does not merely say things i.e. report states of affairs;
- these performative sentences achieve their corresponding actions
because there are specific CONVENTIONS linking the words to
institutional procedures;
- unlike constatives, which can be assessed in terms of truth or falsity,
performatives can only be assessed as felicitous or infelicitous
according to whether their felicity conditions are met or not.

5.2.3 The performative formula

Austin investigated the possibility of defining performative utterances in
terms of a grammatical formula for performatives. The formula he initially
proposed has a first person singular subject and an active verb in the simple
present tense that makes explicit the illocutionary act that the speaker intends
to accomplish in uttering the sentence, as shown in (5):
(5) “I verb-present-active X …”

89
Consider, as an illustration, the examples in (6):
(6a) I bet you 5 pounds it will rain tomorrow.
(6b) I am betting you 5 pounds it will rain tomorrow.
(6c) I betted you 5 pounds it would rain the following day.
(6e) He bets you 5 pounds it will rain tomorrow.
Only the example in (6a) is a performative, the others are constatives.
However, there are plenty of other uses of first person indicative sentences in
the simple present that can be said in demonstration, as a report of a
concurrent action. Consider the example in (7):
(7) I now beat the eggs till fluffy
Moreover, there are sentences which fit the formula in (5) but which can be
descriptive of activities under a variety of circumstances. Consider the
examples in (8):
(8a) I bet him every morning that it will rain – constative
(8b) On page 29 I protest against the verdict – constative
Consequently, Austin proposes another criterion in order to isolate
performatives alone: namely the self-referential adverb hereby:
(9) “I hereby verb-present-active X …”
Austin calls the forms that follow the formula in (9) explicit performatives,
opposing them with primary performatives (rather than with implicit or
inexplicit performatives.)
A contentious issue, however, seems to be cases when performative verbs
can be used non-performatively, as in the example (10), below:
(10)
A: How do you get me to throw all these parties?
B: I promise to come.
Although “I promise to come” fit the performative formula, nevertheless the
utterance is used non-performatively, i.e. descriptively.

90
Apparently the formula is not a necessary criterion, since there are many
forms that differ from this canon and nevertheless seem intuitively to be
explicit performatives. The examples in (11) are performatives that deviate
from the performative formulae:
(11a) You are fired - passive voice
(11b) The court finds you guilty - the subject is not first person
(11c) You did it! – no performative verb
(11d) Guilty! – no verb
Austin therefore concluded that the performative formula was neither a
necessary nor a sufficient condition in order for a sentence to be called
performative.
There still are numerous clear cases that fit the performative formulae, but
the fact that explicit performatives seem to shade off into constatives and
other non-performative sentence types greatly weakens the utility of the
formulae as as a litmus for illocutionary force. The example in (12) is still
another case when the illocutionary act cannot be accomplished in terms of
an explicit performative formula.
(12) *I fire you.
5.2.4 Austin’s revised approach to the constantive vs. performative
distinction
In view of all these problems, Austin eventually rejects what he proposes at
the beginning. First, there is a shift from the view that performatives are a
special class of sentences with peculiar syntactic and pragmatic properties to
the view that there is a general class of performative utterances. This general
class of performative uttereances include explicit performatives (the old
familiar class) and implicit performatives or primary performatives, the
latter including lots of other kinds of utterances, if not all. Thus, there is a
shift from the dichotomy performative vs. constative to a general theory of
speech acts in which constatives are just particular members of a special sub-
case of performatives.
Explicit performatives such as I hereby warn you are interpreted as a
relatively specialized way of being unambiguous or specific about the act
one is performing in speaking. Instead of explicit performative, one can

91
employ implicit performatives that make recourse to less explicit devices
that mark the illocutionary force, as shown in (13) below:

(13)
- mood as in:
e.g. Shut it! – instead of ‘I order you to shut it’

- adverbs
e.g. I’ll be call you without fail – instead of ‘I promise to call you’
- particles
e.g. Therefore X – instead of ‘I conclude that X’
- intonation to distinguish between a warning a question or a protest in
It’s going to charge
An important feature of implicit performatives, according to Austin, is that in
principle any implicit performative can be put into the form of an explicit
performative.
Initially Austin claimed that the only kinds of utterances that are not
doing actions as well as, or instead of reporting facts or events are
statements/constatives. Later in his William James Lectures, Austin rejects
the dichotomy between perfomatives and constatives and argues that there is
no incompatibility between utterances being truth-bearers and
simultaneously performing actions. An utterance like the one given in (14)
seems simultaneously to perform the action of warning and to issue a
prediction, which can later be assessed as true or false.

(14) I warn the dog will bite

Moreover, he argued that presupposition failure, in the domain of constatives
plays the same role as felicity conditions do in the case of perforamatives. A
constative such as the one given in (15) can be put the form of an explicit
performative as in (16).
(15) I am alone
(16) I state that I am alone
Austin concludes that the distinction between statements, as truth-
bearers, and performatives, as action-performers, can no longer be
92
maintained. Statements or constative are merely a special case of
performatives. All utterances, in addition to conveying a certain meaning,
perform specific actions, through having specific illocutionary forces.

5.3 The influence of Grice

Grice’s seminal articles (1957, 1967), while not dealing directly with the
problems that occupied Austin, nevertheless have had a profound impact on
speech act theory. Grice (1957) argued that ordinary communication takes
place not directly by means of convention, but in as a result of a speaker’s
having certain intentions and getting his or her audience to recognize those
intentions. Moreover, it is the speaker’s intention to secure this recognition.
Grice suggests that this holds both for speech and for other sorts of
intentional communicative acts. In his view, the utterance is not
communicative in itself, but only provides clues to the intentions of the
speaker. Grice proposed some maxims of cooperative behavior and
accounted for speakers exploit them in order to secure recognition of the
speaker’s intentions in uttering certain words under particular circumstances
(cf. Chapter 4 of this book). Grice distinguished between what is said in
making an utterance, that which determines the truth value of the
contribution, and what is meant, i.e. the total of what is communicated.
Things that are communicated beyond what is said, in as strictly technical
sense, are called, by Grice, implicatures. The implicatures that are based on
the assumption that the speaker is being cooperative Grice called
conversational implicatures.

5.3.1 Strawson’s objection to Austin

Strawson (1971) criticized the Austinian view arguing that speech acts such
as christening and marrying are wrongly identified as typical of the way
language works. Strawson pointed out that such illocutionary acts ordinarily
take place in highly formal, ritualistic, or ceremonial situations such as ship
launchings and weddings. Strawson conceded that in such ceremonial
situations the use of language does involve convention. However, he claimed
that what one says on such occasions is part of a formalized proceeding
rather than an example of ordinary communicative behavior.
He argued that for more ordinary speech acts, such as those
accomplished by uttering declarative sentences of various sorts, the act
succeeds by Gricean means rather than by abiding by conventions of
language use. In other words, the act is felicitous if it arouses in the
addressee the awareness that it was the speaker’s intention to achieve a

93
certain communicative goal and to get the addressee to reach this conclusion
on the basis of his or her having produced a particular utterance.

5.3 2.Searle’s defense of Austin

Austin’s work has given rise to many theories regarding speech acts, the
most significant of which being Searle’s taxonomy (1969). Searle 1969, a
work that is second only to Austin’s in its influence on speech act theory,
proposes a neo-Austinian analysis in which convention, contrary to Grice’s
and Strawson’s claims, once again is instrumental and difficult not to be
taken into consideration. Although Searle does acknowledge the role of
Gricean intentions in communication, nevertheless he argues that such an
account is incomplete because 1) it fails to distinguish communication that
proceeds by using meanings of the kind that only natural languages make
available, and 2) it fails to distinguish between acts that succeed solely by
means of getting the addressee to recognize the speaker’s intention to
achieve a certain (perlocutionary) effect and those for which and those for
which that recognition is “…in virtue of (by means of) H[earer]’s knowledge
of (certain of) the rules governing (the elements of) [the uttered sentence] T.”
(Searle 1969: 49-50). These Searle labels illocutionary effects.
Of the various locutionary acts that Austin mentions, Searle singles
out the propositional act as especially crucial. In Searle’s view, this act
consists of two components, a referential act, in which a speaker picks out
or identifies a particular object through the use of a definite noun phrase, and
a predication, which Searle did not see as a separate locutionary act (or any
other kind of speech act) but only as a component of the total speech act, i.e.
the combination of illocutionary force with propositional content.
Searle (1969) pointed out that quite often the form of an utterance
displays bipartite structure, one part of which determines the propositional
act, and the other part the illocutionary act. Searle symbolized as p the parts
of an utterance that together are used by a speaker to signal the propositional
act. Formal features of the utterance which determine the literal illocutionary
force he called the illocutionary force indicating device (IFID), which he
symbolized as F. Thus, he proposed the formula in (17) for the form of a
complete utterance used to accomplish a complete speech act, including the
propositional portion of the locution and the IFID.
(17) F(p)
Among Searle’s arguments for the validity of the formula in (17) was the
claim that negation can be either internal or external to the IFID, at least at
94
the abstract level of grammatical analysis that Chomsky (1965) called deep
structure. Thus, if p is (underlyingly) I will come and F is I promise, there are
two negations, namely I promise not to come and I do not promise to come.
Searle argued that the latter must be construed as an illocutionary act of
refusing to promise something, not as an illocutionary act of asserting,
stating, or describing oneself as not making a certain promise.
A central tenet of Searle’s theory of speech acts is the view that
“speaking a language is performing acts according to rules” (Searle 1969:36-
7). By “rule” he means a conventional association between a certain kind of
act and its socially determined consequences. He views these as constitutive
rules, in the same sense that the rules of chess are constitutive of the game
itself. To perform an illocutionary act, according to Searle, is to follow
certain conventional rules that are constitutive of that kind of act. In order to
discover the rules, Searle, following Austin, proposed to examine the
conditions that must obtain for an illocutionary act to be felicitously
performed.
His analysis was carried out in detail for promises, a kind of
illocution that Searle described as “fairly formal and well articulated,”
(Searle 1969:54) and from which “many of the lessons learned … are of
general application.” (Searle 1969:54) For the illocutionary act of promising,
the rules that he postulated are (Searle 1969:63):
1. Pr (the IFID for promising) is to be uttered only in the context of a
sentence (or larger stretch of discourse) T the utterance of which predicates
some future act A of S.
2. Pr is to be uttered only if the hearer H would prefer S’s doing A to his not
doing A, and S believes hearer H would prefer S’s doing A to his not doing
A.
3. Pr is to be uttered only if it is not obvious to both S and H that S will do A
in the normal course of events.
4. Pr is to be uttered only if S intends to do A.
5. The utterance of Pr counts as the undertaking of an obligation to do A.
Searle called Rule 1 the propositional content rule. Rules 2 and 3 are
preparatory rules. Rule 4 is a sincerity rule, and rule 5 is the essential
rule. Searle found a similar set of rules to be operative in the case of other
kinds of illocutions, such as assertion, thanking, and warning, as shown in
the following Table 5.1:

95
Table 5.1 The constitutive rules for assertion, thanking and warning (based
on Searle 1969)
Assert Thank (for) Warn

Propositional Any proposition p Past act A done by Future event or

content H state, etc., E

Preparatory 1. S has evidence A benefits S and S 1. H has reason to

(reasons, etc.) for believes A benefits believe E will
the truth of p S. occur and is not in
H’s interest.
2. It is not obvious
to both S and H 2. It is not obvious
that H knows to both S and H
(does not need to that E will occur.
be reminded of,
etc.) p

Sincerity S believes p S feels grateful or S believes E is not

appreciative for A. in H’s best
interest.

Essential Counts as an Counts as an Counts as an

undertaking that p expression of undertaking to the
represents an gratitude or effect that E is not
actual state of appreciation in H’s best
affairs. interest.

Violations of Searle’s preparatory conditions produce infelicities of Austin’s

type A, misinvocations. Violations of the sincerity conditions correspond
more or less directly to Austin’s class C of infelicities, the abuses that do not
nullify or vitiate the illocutionary act but nevertheless make it flawed.
Neither the propositional content condition, nor the essential condition can
be related very clearly to Austin’s taxonomy of infelicities.
Searle’s taxonomy of speech acts includes five types of illocutionary
acts, characterized as follows:

Representatives/assertives – represent a state of affairs; they have a word-

to-world direction of fit, i.e. the intention is to make the words fit the world;
the psychological state expressed is a belief, in which any proposition can
occur; characteristic performative verbs that help us identify the illocutionary

96
force, and therefore referred to as illocutionary force indicating devices,
include: claim, assert, report, confirm, confess, etc.; e.g. – statements; they
carry the value true or false: they should match the world in order to be true;
their illocutionary point is to commit the S to something being the case, to
the truth of the expressed proposition.
Directives – direct the addressee towards doing something, i.e. they get the
H to do some volitional act; they a world-to-word direction of fit, i.e. the
world is adapted to the uttered words (they make the world fit words via the
hearer); the psychological state expressed: a wish, a desire; examples of
directives include orders, wishes; performative verbs: ask, order, command,
implore, beg, challenge, etc.
Commisives are those kinds of speech acts that speakers use to commit
themselves to some future action; they have a world-to-word direction of fit,
i.e. they make the world fit words via the speaker; the psychological state
expressed: an intention (they express what the speaker intends); examples of
commisives include promises, threats, refusals, pledges, etc.; performative
verbs: promise, swear, vow, etc.
Expressives are speech acts that state what the speaker feels. They express a
wide range of psychological states and can be likes, dislikes, statements of
pleasure, pain, joy, sorrow. They have no direction of fit; examples of
expressives include congratulations, condolences, etc.; performative verbs:
thank, congratulate, apologize, condole, etc.
Declarations are those kinds of speech acts that change the world via their
utterance. They correspond to Austin’s original class of performatives (i.e.
explicit performatives). They express no psychological state; they heavily
rely on extralinguistic conventions: the speaker has to have a special
institutional role, in a specific context in order to perform a declaration
felicitously; they have two directions of fit: word-to-world (make words fit
the world) and world-to-word (make the world fit words i.e. words change
the world); examples of declarations include excommunication, christening,
etc.; performative verbs: declare, baptize, name, appoint, elect, pronounce,
etc.
Two further features of Searle’s (1969) theory deserve a special
mention. First, he embraces Austin’s idea that a sufficient test for
illocutionary acts is that they can be performed by uttering an explicit
performative. He argues that more than one illocutionary act can be
accomplished by the utterance of a single, simple sentence. To illustrate this
point he gives as an example the case of a wife who says at a party, “It’s
97
really quite late”. In doing so the wife simultaneously performs the
illocutionary act of stating a fact and the illocutionary act of making a
suggestion equivalent to “I suggest that we go home.” Second, he noticed
that an illocutionary act is typically performed with a certain perlocutionary
effect in mind, an effect that follows from the essential condition: “Thus
requesting is, as a matter of its essential condition, an attempt to get the
hearer to do something …” (Searle 1969:71).

5.4 Indirect speech acts

Searle (1969) distinguished between effects which are achieved by getting
the hearer to recognize that the rules governing the use of an illocutionary
force indicating device are in effect, which he called illocutionary effects,
and those effects that are achieved indirectly as byproducts of the total
speech act, for which he used the term perlocutionary effects. Saddock
(1970, 1972) points out that the effect might be very similar and proposes
that the same words be used to described it, whether it is an illocutionary or
perlocutionary effect. For instance, a speaker might warn a hearer by uttering
an explicit warning that the dog is about to bite, in which case we have an
illocutionary effect of warning. Alternatively, a speaker might warn the
addressee (in the sense of making him feel alarmed) by making a statement
to the effect that the dog is about to bite. In this second case the speaker
produces in the addressee an illocutionary effect of understanding that the
speaker is stating that the dog is about to bite, which in turn, under the right
circumstances, causes him or her to be warned. In this case the effect of
warning is a perlocutionary effect.
Sadock (1970, 1972) argued that in certain cases, there was some
conventional indication in the form of the utterance of what might be taken
as an indirect, perlocutionary effect. The central sort of example is the
utterance at a dinner table of an apparent question like “Could you pass the
salt?”. The utterance appears to be a question, but when produced at a dinner
table, a commonly achieved effect is to arouse in the addressee a feeling of
obligation to pass the salt. In the literature utterances of the form “Could you
pass the salt?” are literal or direct speech acts when interpreted as questions,
and indirect speech acts when interpreted as requests.
An indirect speech act is an illocutionary act that is performed
subordinately to another (literal) illocutionary act. It is indirect in the sense
that its success is tied to the success of the first act. Thus the addressee
should correctly identify the first act (i.e. the direct/literal act) if he wants to
interpret the indirect speech act felicitously. The speaker’s primary
illocutionary intent is that of the indirect speech act. His secondary
98
illocutionary intent is that of the act literally associated with the particular
sentence form.
Indirect speech acts are also based on the mechanism of
conversational implicature. There will be an inferential chain linking the
literal and the indirect speech acts. The first step within this inferential chain
is to notice that the literal illocutionary force is not relevant enough under the
circumstances. Let us consider the utterance in (18):

(18) Can you pass the salt?

In uttering it the speaker is performing two speech acts: the literal act and the
indirect speech act. Interrogative sentences are used to formulate questions,
which in their turn are requests for information. In our example the
information is about the ability of the H to pass the salt and the relevant
answers are ‘yes, I can’ or ‘no, I can’t’. This is the direct/literal act such an
utterance is doing. However, this is not the speaker’s primary illocutionary
intent. The intended meaning/interpretation is that of an indirect speech act,
i.e. a request for action or a directive the explicit performative being ‘I
request of you to pass the salt’. In order to arrive at the interpretation
intended by the speaker we need to draw a series of inferences. This
inferential chain includes the following steps:
1) S could not be merely asking whether I (the addressee) have the ability to
pass the salt because it is mutually believed that the answer is ‘yes’; so the
question will be irrelevant, it will flout the maxim of relevance and it will
also sin against the maxim of quality by breaching the sincerity condition
associated with yes/no questions, namely S does not know and he sincerely
wants to know whether something is or is not the case.
2) Since S is believed to be cooperative (he is believed to be abiding by the
CP) than there is another illocutionary act that he is performing and that links
the circumstances to asking whether I (the addressee) have the ability to pass
the salt, such that in asking this question S is performing that act.
(a) A preparatory condition for any directive is H’s (the addressee’s) ability
to perform the act predicated, in this case passing the salt.
(b) Therefore S has asked me a question, the affirmative answer to which
would entail that the preparatory condition for requesting me the salt is
satisfied

99
(c) Since S has alluded to the satisfaction of a preparatory condition for a
request, then it is likely that he wants me to bring about the obedience
conditions for that request.
Thus applying this line of reasoning we reach the conclusion that:
3) S is asking me (H) whether I have the ability to pass the salt and is thereby
requesting me (H) to pass him the salt.
Consider the following statement which can be used as an apology.
Only the derivation of the indirect speech act is discussed:
I should never have done that.
1) S could not be merely stating that S should never have done that. Because
it is mutually believed by S and H that the act affected H in some negative
way, that S should not do such things. So S is not conveying information and
would be violating the maxim of quantity.
2) Since S is believed to be cooperative, there is another illocutionary act that
S is performing and that links the circumstances to stating that one should
never have done that act, such that in stating that one should never have done
that act, S could also be performing a second act.
3) S is stating that S should never have done that and thereby apologizing for
having done it. Basis for this inference: it is mutually believed that people
often regret doing things they believe they should not have done. Therefore
S’s primary illocutionary intent under these circumstances is to apologize for
doing A.

5.4.1 Indirect speech acts and politeness

Most theories of indirect speech acts hardly focus on the reasons why
speakers use indirect rather than direct speech acts, nor do they seek an
explanation for which particular indirect speech acts will be used under
which conditions. It is obvious that speakers’ choice of a particular indirect
speech act to the detriment of the corresponding direct one is, to a large
extent, related to aspects of politeness. Brown and Levinson (1987) include
extensive investigations of how models of politeness can yield answers to
these interesting questions. They argue that a fundamental principle of
politeness (deriving from a need to preserve addressee’s “face”) is: Don’t
impose.

100
 Requests are, by definition, impositions. Thus, they are likely to
trigger a clash with the principle of politeness. One way of resolving this is
the use of indirect forms. The direct imposition can be mitigated by avoiding
a direct demand and instead asking whether the addressee is willing to or
capable of carrying out the act. This gives the addressee the technical option
of not carrying out the implied request without losing face. Hence Would you
pass the salt? or Can you pass the salt? are more polite than Pass the salt.
These studies of politeness have given rise to a considerable interest in cross-
cultural comparisons of indirection strategies, and intercultural
communication in naturally occurring conversation.

5.4.2 Indirect speech acts and sentence type

In most languages, perhaps even all, sentences can be classified on the basis
of formal features into a small number of sentence types, with each type
associated with a certain illocutionary act potential (IAP). Thus in English,
sentences can be classified as declarative, with IAP including acts of stating,
asserting, claiming, testifying, and so on; interrogative, with IAP including
asking, inquiring, querying, and so on; and imperative, with IAP including
requesting, demanding, commanding, directing, and so on. Consider the
examples in (19):
(19a) You wear a seat belt – declarative; IAP: asserting
(19b) Do you wear a seat belt? interrogative; IAP: asking/ inquiring
(19c) Wear a seat belt! – imperative; IAP: commanding
To count as a type within such a system, the formal features defining the
types must be mutually exclusive. A sentence cannot be simultaneously of
the declarative and interrogative type, or of the interrogative and imperative
type. Furthermore, every sentence should be of one or of another type
according to the formal, structural features that it displays.
Whenever there is a direct relationship between a structure and a
communicative function, the speech act is a direct speech act. Whenever
there is an indirect relationship between a structure and a function, the
speech act is an indirect one. Thus, a declarative used to make a statement or
an assertion is a direct speech act, whereas when the declarative is used to
make to make a request the act is an indirect speech act. Consider the
utterance in (20):

(20) It’s cold outside

101
The example in (20) is interpreted as a statement/assertion if it is paraphrased
as ‘I hereby tell you about the weather’, in which case it is functioning as a
direct speech act. When it is used to make a request/command, in which case
it can be paraphrased as ‘I hereby request of you that you close the window’,
it is functioning as an indirect speech act.
Different structures can be used to accomplish the same basic
communicative function as the examples in (21) show. If the speaker wants
the addressee not to stand in front of the TV, he could use any of the
utterances given in (21):
(21a) Move out the way!
(21b) Do you have to stand in front of the TV?
(21c) You’re standing in front of the TV.
(21d) You’d make a better door than a window.
Of these, only (21a) is the direct speech act, the (21b – 21c) others are
indirect speech acts.

5. 5 Conclusions

Speech act theory has been developed mainly by Austin (1962) and Searle
(1969, 1979). The theory of speech acts is based on the assumption that the
minimal unit of human communication is not a sentence or other expression,
but rather the performance of certain kinds of communicative acts, such as
requests and promises. For instance, the communication of a request by a
speaker (S) to a hearer (H) is an attempt by S to get H to do something. This
communication is successful or felicitous if H does perform the requested
act.
For philosophy of language in particular, the theory of speech acts
underscores the importance of the distinction between language use and
linguistic meaning (i.e. the distinction between pragmatics and semantics).
This distinction sharpens the formulation of questions about the nature of
linguistic competence, by separating issues related to capacities exercised in
linguistic interaction from those specific to knowledge of language itself.

102
 CHAPTER 6

CONVERSATION ANALYSIS

6.1 Introduction

Conversation analysis (CA), an approach to discourse derived from

ethnomethodology16. At this point terminology needs clarification. A
distinction should be drawn between conversation and conversational
analysis, on the on hand, and conversation analysis and interactional
sociolinguistics, on the other hand, since, although used interchangeably by
some linguists, these approaches to conversational discourse differ
significantly in terms of methodology.
The term conversation analysis as used in this study refers to a line
of linguistic inquiry, also referred to as ethnomethodological conversation
analysis, whose earliest contributions are associated with authors such as
Sacks, Schegloff, Jefferson, Pomerantz and others. Such works are distinct
from studies also sometimes referred to by the same name or by the term
conversational analysis, and identified with such authors as Gumperz,
Tannen, Schiffrin and others, whose declared intent is to describe
conversational uses of language. Such works fall within the scope of
interactional sociolinguistics or pragmatics17. It should be pointed out that
the works on gendered conversational styles reviewed in the previous chapter
were indeed carried out within the framework of interactional
sociolinguistics and not conversation analysis, the approach that we propose
in this chapter.

16
Ethnomethodology is a sociological perspective pioneered by Harold Garfinkel which
concerns itself, among other things, with talk viewed as a means of sustaining reality being
at the same time part of that reality. Ehnomethodologists focus on various bits and pieces of
everyday life in an attempt to show how those who deal with such bits and pieces go about
doing so. As Leiter (1980:5) states, ‘the aim of ethnometodology […] is to study the process
of sense making (idealizing and formulizing) that members of society […] use to construct
the social world and its factual properties (its sense of being ready-made and independent of
perception).’
17
For a discussion of the boundaries of pragmatics and sociolinguistics, see Levinson
(1983).
103
 Interactional sociolinguistics as articulated by Gumperz 18 and
Goffman 19 has been extensively applied within linguistics by Brown and
Levinson (1987), Schiffrin (1987), and Tannen (1989, 1994). Despite the
different sets of interest advanced by the two main proponents of
interactional sociolinguistics, Gumperz and Goffman, some stemming from
concerns about language and culture20 (Dil 1971; Gumperz 1982; Gumperz
1985), others from concerns about self and society (Goffman 1963; 1967;
1974) the underlying issues providing unity to interactional sociolinguistics
are the interaction between self and other, and self and context.
Both conversation analysis and interactional sociolinguistics are
necessarily analyses of language in use. As such, they cannot be restricted to
the description of linguistic forms independent of the purposes or functions
which these forms are designed to serve in human affairs. In contrast to
analysts whose work fits into one of the previously established subfields of
linguistics (and we refer here to theoretical linguistics), conversation analysts
and particularly sociolinguists of interactional orientation tend to focus on
the fact that language is designed for communication and varies according to
the contexts of use, being situated in particular circumstances of social life,
reflecting and adding meaning and structure in those circumstances
(Schiffrin 1987: 3; Schiffrin 1994a: 97).
In fact, it so is difficult to draw a clear dividing line between these
two approaches to discourse that some linguists refer to conversation
analysis as interactional sociolinguistics (Mercer 2000). Both conversation
analysis and interactional sociolinguistics draw upon naturally occurring
interactions (in institutional settings or among friends) for data and both
approaches to discourse pay a great deal of attention to transcriptions of
features of talk likely to serve as contextualization cues. Within this body of
research, social life is viewed as being constituted at the micro-level of social
interaction. Thus the major focus of concern is on the interpretive and
inferential processes whereby interactants acting in real time are able to
adopt the most appropriate strategies to achieve their desired social
meanings, including their identities, footings and alignments with others.

18
Gumperz’s work focuses on how interpretations of context are instrumental in the
communication of information and in understanding the speaker’s intention and/or discourse
strategy.
19
Goffman’s work focuses on how the organization of social life (in institutions,
interactions, etc.) provides contexts for making sense of both the conduct of self and the
communication with another.
20
Dil (1971) is a collection of Gumperz’s essays through 1971. The research reported in it is
grounded in an assumption basic to social and cultural anthropology: the meaning, structure
and use of language are socially and culturally relative.
104
These analytic approaches, together with the ethnography of speaking21,
have certain crucial elements in common, identified by Schiffrin as follows:

CA is like interactional sociolinguistics in its concern with the

problem of social order, and how language both creates and is created
by social context. It is also similar to the ethnography of speaking in
its concern with human knowledge and its belief that no detail of
conversation (or) interaction can be neglected a priori as unimportant.
All three approaches also focus on detailed analysis of particular
sequences of utterances that have actually occurred. (Schiffrin
1994a:233)

Conversation analysis, however, is different from these two

approaches in various significant ways to the extent to which it provides its
own assumptions, its own methodology and terminology. Conversation
analysis refrains from premature generalization, viewing the empirical
conduct of speakers as the central resource out of which analysis must
develop (Heritage 1984: 243). Moreover, what is said provides not only the
data on which analysis is based, but also the evidence for hypotheses and
conclusions. It is participants’ linguistic behaviour that provides evidence for
the presence of units, existence of patterns, and formulations of rules.
Similarly, CA avoids analyses based on single texts. Instead, there
are examined as many instances as possible of a particular phenomenon
across several texts in order to bring to light the systematic properties of the
sequential organization of talk. Thus the main tenet of CA is the search for
recurrent patterns, distributions and forms of organization in large corpora of
talk (Sacks et al. 1974; Schegloff 1980). These large corpora of talk are
transcribed in an attempt to reproduce what is said rendering both linguistic
details such as pronunciation and extra-linguistic details such as in-breaths
and in so doing researchers aim at avoiding any presuppositions about what
might be important for either participants or analysts themselves. Unlike
variation or interactional sociolinguistics, CA transcripts of talk pay little
attention to social relationships holding among participants or to what has
been called social context (e.g. social identities of participants, setting,
personal attributes, etc.).
In what follows we will briefly describe the main areas of interest
within conversation analysis suggesting some ways in which our
understanding of gendered language and gendered selves can be enhanced by

21
For a discussion of the points of contact and the differences in terms of methodology and
analysis of data see A. Duranti (1988).
105
the findings of conversation analytic works and the research practices
underlying those findings.

6.2 Turn-taking organization and the local management system

The main point that researchers working within CA framework wanted to

make was that although we cannot specify a set of formal rules for
generating the set of ‘all and only correct’ conversational utterances,
ordinary conversation is by no means un-ruled. From the very beginning, the
focus of attention for conversation analysts became the organization and
structuring of conversation and not so much its ‘correctness’.
The main point of conversational structure is to keep the flow of
conversation going, by avoiding clashes (when two or more people are
speaking at the same time), or instances when participants feel trapped in
sterile verbal exercise, or by helping along a conversation that has halted or
has trouble in maintaining its proper speed. As far as conversational structure
is concerned, Sacks (1992:32-42) mentions the following features:

- in a single conversation people talk one at a time,

- speaker change recurs, i.e. conversation is characterized by
turn-taking,
- while speaker change recurs, one-party-at-a-time is preserved.

The mechanism responsible has been shown to be capable of operating in

quite different circumstances:

- the number of parties engaged in conversation may vary

(from 2 to 10 or even more),
- turn size also varies from minimal utterances to many minutes
of continuous talk,
- if there are more than 2 parties then provision is made for all
parties to speak without there being any specified order or
queue,
- length of conversation is not specified in advance
- talk can be continuous or discontinuous
- transitions from one turn to the next turn with no gap and no
overlap are not uncommon
- occurrences of more than one speaker talking at a time,
although possible, are brief

106
 - the same system is equally functional both in face-to-face
interaction and in the absence of visual monitoring, as on the
telephone,
- the same system holds across various types of conversations
(small talk, conflict talk, business talk, etc.); it also holds
across things like gender, occupation, social class, political
persuasion, etc. (e.g. the fact that one party talks at a time and
speaker change recurs is not a feature of , say male
conversation or female conversation, or of middle-class
conversation)
(Sacks, Schegloff and Jefferson 1974)

Sacks, Schegloff and Jefferson (1974) suggest that the mechanism

that governs turn-taking22 and accounts for the features mentioned above
(one party talks at a time and speaker change recurs) is a set of rules with
ordered options which operates on a turn-by-turn basis and is therefore
termed the local management system. This set of rules provides for the
allocation of a next turn to one party and coordinate transfer so as to
minimize gap and overlap. The rules are locally managed in that they apply
to all possible points when speaker change may occur. Such locations where
these rules may apply are called transition-relevance places (TRP). These
places reflect the existence of various unit types through which a speaker
may construct a turn. These turn constructional units (TCU) are
determined by features of linguistic surface structure, hence their syntactic
nature (they can be phrases, clauses or sentences). Initially, a speaker is
assigned just one of these turn constructional units whose length is largely
within the speaker’s control due to the flexibility of natural language syntax.
The end of such a unit constitutes a point at which speakers may change, that
is a transition-relevance place (TRP).
One organizational and possibly even a grammatical fact about turns
is that they can comprise more than one TCU. One relevant aspect of the
possibility of multi-turns is that some TCUs can be designed for their
position in the turn. For instance, a first unit in its turn can be built and
delivered in a way which projects additional ones to follow. Similarly, non-
first TCUs can be conceived and delivered as “subsequents” (Schegloff
1996).

22
The main bibliographical resources in this area are Sacks, Schegloff and Jefferson (1974)
on turn-taking and Schegloff (1882, 1996) on turn-organization, but also Sacks (1992),
Jefferson (1973, 1984), Lerner (1991, 1996a), and Schegloff (1999, 2000, 2002).

107
 The number of TCUs that make up a turn can vary according to the
position of the turn. Second position turns in a sequence may be more
expansible than first position turns. For example, many turns following
questions appear to provide for multi-unit answers. Similarly, some practices
such as story preface work not to get an additional TCU in the turn, but to
neutralize the transition-relevance place of the possible completion of TCU
until some projected feature is delivered, e. g. until something that can be
analyzed as “funny” or “strange”, or the like has been told. This is the key
feature of the production of many discourse units in conversation such as
narratives or other types of extended units of talk (Schegloff 1996).
As pointed out by Sacks, Schegloff and Jefferson (1974:701-3), the
turn-taking organization is composed of two types of resources, plus a set of
rules in which they are integrated. One resource is the turn-constructional
resource, composed of a set of types of units of talk – the TCUs mentioned
above. The second resource is the turn-allocational resource comprising
turn-allocational practices ordered as follows: the self-selecting practices are
invocable only when the other-selecting ones have not been employed. In
addition the following set of rules governs the transition of speakers so as to
minimize gap and overlap and they apply at a TRP:

RULE 1 – applies initially at the first TRP of any turn:

a. If current speaker selects next speaker in current turn, then current

speaker must stop speaking and next speaker must speak next,
transition occurring at the first TRP after next-speaker selection.
b. If current speaker does not select next speaker, then any(other) party
may self-select, first speaker gains rights to the next turn.
c. If current speaker has not selected next speaker, and no other party
self-selects under option (b), then current speaker may (but need not)
continue (i.e. claim rights to a further turn-constructional unit)

RULE 2 – applies at all subsequent transition-relevance places.

When rule 1(c) has been applied by current speaker, then at the next TRP
rules 1(a)-(c) apply recursively until speaker change is effected.

Turn-allocational resources can be roughly grouped under ‘current

selects next’ and ‘self-selection’. Next-speaker selecting techniques may
include such obvious cases as an addressed question that selects its addressee
to speak next, or when a party starts to speak when not selected, i.e. self-

108
selects himself to speak. Among the most commonly used turn-allocational
techniques we can mention the following:

a. An important general technique, and perhaps the central one, whereby

current speaker selects next involves the addition of an address term, or
some other device for achieving addressing (such as the direction of gaze), to
the first part of an adjacency pair. However, addressing a party will not
necessarily in itself, select him as a speaker. Thus A addressing a question to
B, selects him as next speaker; but when B speaks next and addresses an
answer to A (which constitutes the second part of the same adjacency pair),
A is not necessarily selected as next speaker in the next turn (Sacks,
Schegloff and Jefferson 1974).

b. The use of what Sacks (1992) refers to as an appendor clause/question is a

technique for reselecting the last speaker as next speaker, as in the example
below:

(1)

A: …..and it turned out very good

→B: from the old man’s point of view?

c. A turn’s talk, whether or not initially constructed as a first pair-part, can be

turned into current selects next by the affiliation to it of a tag question such
as You know?, Don’t you agree?, etc.

d. The central technique for self-selecting is ‘starting first’. Rule 1b

explicitly stipulates that the speaker who starts first gets the right to the next
turn. A self-selector aiming for an earliest start should begin his earliest start
with a unit-type’s beginning. A next-turn’s beginning, however, is subject to
sources of overlap and may impair its part in the turn’s utterance
construction (Schegloff 2000, 2002). Consequently, appositional beginnings
such as well, but, and, so, etc., represents a class of constructions which are
commonly used and which satisfy the constraints of beginning. Given their
function, they have been termed turn-entry devices or pre-starts, just as tag
questions are turn-exit devices or post-completers (Sacks, Schegloff and
Jefferson 1974).
These practices of turn-taking not only organize distribution of
opportunities to talk among speakers taking part in the speech exchange, but
they also constrain the size of turns, by making the possible completion of a
turn transition-relevant. As pointed out by Schegloff et al. (2002: 6) “this
109
interactive dimension – in which possible completion can (but need not
always) occasion or trigger the start of a next turn by another – has
consequences for speakers’ constructions of turns, and thereby for the form
which turns (and their building blocks, “turn-constructional units”) take”.
So far we have addressed the way in which the local management
system controls turn-taking in ordinary conversation, the role it assumes in
constructing simple or multi-unit turns. This has been done in order to
highlight the areas of potential intersection with interactional sociolinguistics
and more specifically with language and gender research. It has been pointed
out that “if we are to understand language in its context of deployment, we
need in the first instance to understand how and for what it is deployed by its
participants, and how its deployments are understood by them and reflected
in their own responsive conduct” (Schegloff et al. 2002).
Of the ways turn-taking and turn organization should matter to the
gender-related conversational styles described by interactional sociolinguists
we will mention only one. The unmarked value of the transition space is one
beat of silence; in other words, at a possible transition-relevance place a next
speaker ordinarily allows one beat of silence to pass before starting a next
turn (Jefferson 1984). Departures from this value are potentially marked.
Longer silences at the transition space can be taken as incipient rejection of,
or disalignment from what preceded it (Pomerantz 1984; Sacks 1987;
Schegloff 1988a, 1995). Conversely, shorter silences or their absences (i.e.
the so called instances of latching) can be interpreted as signalling agreement
or alignment with what went before and with the addressee. In consistently
using such departures from the unmarked value participants can be shown as
being oriented towards a competitive conversational style or a cooperative
conversational one based on micro strategies that are accelerators of social
relationships, styles which are said to be favoured by one gender group or the
other.

6.3 Turn-taking irregularities

In their seminal paper proposing a model for turn-taking organization, Sacks

et al. (1974) devoted considerable attention to the phenomenon of speaker
change. The analysis of large corpora of talk enabled them to identify the
following three cases in organizing a speaker change: there may be a silence
in-between, before the speaker change is effected; there may be overlap; or
the speaker change may be effected with no silence and no overlap. The
empirical data demonstrated that the most common case in conversation is
one-party-at-a-time, and that speaker changes typically occur without any
silence in-between and without any overlapping speech (no gaps, no
110
overlaps). The tendencies are accounted for in terms of a force acting to
minimize gap and overlap in conversation. However, they recognize that, on
some occasions, there are slight departures from the one-party-at-a-time rule.
In their own words ‘‘Transitions from (one turn to a next) with no gap and
no overlap are common. Together with transitions characterized by slight
gap or slight overlap, they make up the vast majority of transitions.’’ (Sacks
et al.1974:700).

6.3.1. Timing in turn-taking theories

Two competing theories have been proposed to account for how turn-taking
is achieved. More specifically, they explain how next speakers know when to
start talking. The theory proposed by Sacks and co-workers, known as the
projection theory. Its main tenet is that a next speaker anticipates or projects
when the current speaker will finish on the basis of contextual information,
and then starts talking at the projected TRP (Sacks et al., 1974). Within the
projection theory the notion of no-gap–no-overlap is a key concept.
The second turn-taking theory is known as the reaction or signal
theory, the proponents of which include, among others, Duncan (1972),
Kendon (1967), Yngve (1970). According to the reaction theory, a next
speaker starts talking as a direct reaction to a signal that the current speaker
is finished, or is about to finish.
Apparently, some of the followers of Sacks and colleagues have
interpreted no-gap–no-overlap in a narrow sense as literally zero gap and
zero overlap. This understanding of no-gap–no-overlap has then been taken
as support for a stronger claim: that turn-taking must rely entirely on the
ability to project upcoming turn-endings.
Typically, those who interpret no-gap–no-overlap as literally zero
gap and zero overlap argue that prosodic or other acoustic turn-taking signals
immediately before the silence cannot be of any relevance, simply because
there is no time to react to such signals (e.g. Bockgard, 2007; de Ruiter et al.,
2006; Levinson, 1983). Instead, it is argued that turn-taking is so precise and
next speakers manage to start with no gap and no overlap because they rely
on anticipating the end of the current turn on the basis of its syntactic
features and contextual cues.
Crucially, for projection to result in zero gap and zero overlap, next
speakers have to project not only what the current speaker will say, but also
the exact point in time when the current finishes. Between-speaker interval
distributions provide empirical evidence that can support or challenge the
claims of precision timing in turn-taking.

111
 In their seminal article on turn-taking and in subsequent work, Sacks,
Schegloff and Jefferson (1974) acknowledge the existence such turn-taking
irregularities as silence and overlapping talk. The next two sections focus
these deviations from no-gap-no-overlap feature as they are conceptualized
and integrated in the main turn-taking system advanced by the projection
theory.

6.3.2 Silences in conversation: pauses, gaps, and lapses

Sacks et al. (1974) distinguished between three kinds of silences in

conversations: pauses, gaps, and lapses23. Their classification was based on
what preceded and followed the silence in the conversation, and on the
perceived length of the silence. Pauses refer to silences within turns; gaps
referred to shorter silences between turns or at possible transition-relevance
places (TRPs), before a subsequent application of Rules 1(b) or 1(c); lapses
referred to longer silences between turns, as a result of the non-application of
Rules 1(a), (b), (c). However, the classification they propose is more
complex since the context in which the silence occurs is also taken into
account. Thus, in their account, a silence followed by more speech by the
same speaker is classified as a pause even when it occurs at a TRP.
Consequently, a silence followed by a speaker change is classified as a gap
or a lapse even when it does not occur at a TRP. Hence, gaps and lapses
could in practice only occur when there was a speaker change. A selected
next speaker’s silence is classified as a significant/attributable silence after
the application of Rule 1(a) Sacks et al. (1974)24.

23
In the relevant literature any combination involving the following three factors has been
used in the description and classification of silences in conversation: (ii) turn/speaker, and
(iii) silences/pauses/intervals/transitions have been used for concepts similar to gaps and
duration of gaps at some point in time (e.g. Bull, 1996; Roberts, Francis and Morgan, 2006;
ten Bosch, Oostdijk and Boves, 2005; ten Bosch, Oostdijk and de Ruiter, 2004b). Regarding
the terminology for silences the literature is rich in terms such as alternation silences (Brady,
1968), switching pauses (Jaffe and Feldstein, 1970), (positive) switch time or switch pauses
(Sellen, 1995), transition pauses (Walker and Trimboli, 1982), (positive) floor transfer
offsets (de Ruiter et al., 2006), or just silent or unfilled pauses (e.g. Campione and Veronis,
2002; Duncan, 1972; Maclay and Osgood, 1959; McInnes and Attwater, 2004; Weilhammer
and Rabold, 2003).
24
Various factors have been shown to influence the distribution of pause, gap and overlap.
Empirical research has shown that increased stress (induced in an interview situation
designed to elicit information of an intimate and embarrassing nature) is associated with
markedly shorter gaps (Jaffe and Feldstein, 1970). Similarly, competitive conversations, for
example conversations involving arguments, have significantly shorter gaps than
cooperative conversations, such as friendly chats (Trimboli and Walker, 1984). Some
studies reported that gap durations tend to increase with cognitive load (Cappella 1979). It
112
 Subsequent work by the same authors has shown transitions with
slight gap to occur with greater frequency. For example, Jefferson (1984: 8),
argued that transitions with slight gap ‘‘of all the transition-place points, this
is the most frequently used. A recipient/next speaker does not start up in
‘terminal overlap’, nor ‘latched’ to the very point of possible completion, but
permits just a bit of space between the end of a prior utterance and the start
of his own’’. As a result of its high frequency, this position is viewed as the
unmarked one and described by Jefferson as (1984:8-9) a position when “one
doesn’t get a sense of a next utterance being ‘pushed up against’ or into the
prior, nor of its being ‘delayed’. It simply occurs next’’. Similarly, Schegloff
(2000) who uses the term normal value of the transition space for the same
case, quantifies ‘‘just a bit of space’’ as roughly one syllable, corresponding
to a silent interval of about 150–250 ms.

6.3.3 Overlapping talk

The rules governing turn constructional units described above are

instrumental in accounting for the instances when overlapping talk25 occurs.
There are a number of systematic bases for the occurrence of overlap, of
which the following need to be emphasized:

(a) By allocating a turn to the self-selector who starts first, Rule 1b can
encourage an earliest possible start for each self-selector, providing thus for
overlap by competing self-selectors for a next turn, if each calculates his start
to be earliest possible start at a possible TRP and generating simultaneous
starts.

(b) Overlap can also derive from the projectability of possible completion or
transition-relevance places. Variation in the articulation of the projected last
part of a last component in a turn’s talk is likely to produce overlap between
a current turn and a next turn.

has been shown that more complex tasks and lack of familiarity with tasks result in longer
gaps (Bull and Aylett, 1998). Several studies have furthermore observed longer gaps in
dialogues where the participants have eye contact than in dialogues without eye contact
(Beattie and Barnard, 1979; Bull and Aylett, 1998; Jaffe and Feldstein, 1970; ten Bosch,
Oostdijk, and de Ruiter, 2004).
25
It becomes apparent that there are two ways of conceptualizing gaps and overlaps in the
previous literature. Gaps and overlaps may be treated as entirely different phenomena. Or,
alternatively, they are conceptualized as two sides of a single continuous metric (with
negative values for overlaps, and positive values for gaps) that measures the relationship
between one person ending a stretch of speech and another starting one (de Ruiter et al.,
2006; Norwine and Murphy, 1938; Sellen, 1995).
113
 Overlapping talk has been divided into two classes: competitive – i.e.
problematic, in need of overlap resolution system26 - and non-competitive.
The following configurations of simultaneous talk are treated as non-
problematic:

a. terminal overlaps – in which a next speaker starts a next turn by

virtue of a current speaker’s incipient finishing, but overlaps a bit of its end;
in this case, the very circumstances under which the overlap occurs project
its almost instant self-liquidation.
b. continuers, such as uh huh, which give the opportunity to take a
full turn while displaying an understanding that current speaker is producing
an extended turn;
c. various types of conditional accesses to turn, where the
intervention by the recipient of a not-yet-complete turn is allowed within the
turn’s space as long as it facilitates the completion of the turn-in-progress;
instances of such phenomena include word search, anticipatory completion
and collaboratively built sentences.
d. various choral phenomena such as laughter, collective greetings,
leave-takings in multi-person settings or congratulations. These are either
mandated or allowed to be produced simultaneously and not serially.

Most overlaps are over very quickly. Others, however, persist at great
length. The latter are managed and resolved by participants by means of an
overlap resolution device, which is composed of three elements: a set of
resources (i.e. departures from the normal course of production) for overlap-
oriented turn production, a set of places where these resources come into
play, and a so-called interactional ‘logic’ according to which those resources
in those places are “moves of a describable sort in a competitive sequential
topography” (Schegloff 2000a).
The disruptive nature of competitive overlapping talk is easily
recognizable as it is reflected in deflections from the normal course of
production such as louder volume, higher pitch, faster or slower pace,
repairs of certain items within the same turn (when, for instance, a speaker
repeatedly cuts off a word or a phrase in progress and the repeats it only to
cut it off at the same point and redo the entire operation).
It has been argued by those concerned with language and gender
research that men and women view overlapping talk differently and use it for
different purposes: men tend to use overlapping talk to grab the floor and to

26
For an extensive account, see Schegloff (2000a, 2002).
114
dominate the conversational space while women use it as means of showing
involvement in the conversation and in their conversationalist partner (cf.
Maltz and Borker 1982; Eckert 1990; Coates 1993; Fishman 1997).
However, this is where the argument stopped. Although they supplied
excerpts from naturally-occurring conversations to support their statement, in
the analysis of these excerpts participants to conversation were not actually
shown to orient themselves to one interpretation or another.
Conversation analytic resources and practices can prove fruitful to
language and gender research on overlapping talk since it supplies a
theoretical framework that broadens our understanding of gender-related
overlapping talk by showing participants in talk-in-interaction as selecting
and deploying conversational strategies to accomplish one recognizable
action or another (i.e. doing competitive or cooperative overlapping talk).

6.4 Adjacency pairs

The phenomenon of adjacency pairs is deeply inter-related with the turn

taking system as techniques for selecting a next speaker. Adjacency pairs can
be defined as sequences of two utterances that are:

- adjacent,
- produced by different speakers,
- ordered as a first part and a second part,
- typed, i.e. they belong in certain types, so that a particular first part requires
a particular second or range of second parts

(Sacks and Schegloff 1973)

Examples of adjacency pairs include: assertion-assent/dissent, summons-

answer, question-answer, apology-acceptance/refusal, compliment-
acceptance/rejection; threat-response, challenge-response, accusation-
denial/confession, telling-assessment, boasting-appreciation/derision. In
addition to these features, Sacks and Schegloff (1973) propose the following
rule that governs the use of adjacency pairs:

Having produced a first part of some pair, current speaker must stop
speaking, and next speaker must produce at that point a second part of the
same pair.

A first part of an adjacency pair sets up a conditional relevance which is (or

can be) satisfied by a second part of the same an adjacency pair.
115
 Not all potential second parts to a first part of an adjacency pair are of
equal structural complexity. There is a ranking operating over the
alternatives, so that, for each pair, there is at least one preferred and one
dispreferred category of response. The notion of preference is not a
psychological one, but a structural and it corresponds closely to the linguistic
concept of markedness. Preferred second parts occur as structurally simpler
turns and are therefore unmarked. By contrast, dispreferred second parts are
marked by various kinds of structural complexity. Consequently, they
require much effort on the part of both the speaker, who produces, and the
addressee, who processes them.
If we take the example of requests, the general ranking from
structurally simpler to structurally more complex utterances corresponds to
the ranking acceptance – rejection. One has to work harder, use more
linguistic resources to say no to a request than to say yes. Rejections are
usually accompanied by lots of background material intended to avoid giving
the impression that one just declined to perform the requested action, but
rather that the refusal is due to circumstances beyond one’s control,
circumstances that generally have to be specified. This specification takes
time and requires a greater effort and more linguistic resources which may
surface as complex syntactic structure, a significant delay, some preface
marking their dispreferred status (e.g. the particle well), an account of why
the dispreferred second occurred, pauses, self-interruptions, false starts,
repetitions, etc.

6.5 Pre-sequences

Conversation analysts claim that adjacency pairs are the fundamental unit of
conversational organization. Although conversation proceeds in a pair-wise
fashion, there are also ways that a pair can be expanded before its initiation
though the use of pre-sequences, after its completion by means of post-
elaborations, or even during its creation, through the use of insertion
sequences.
Pre-sequences are usually, and in some cases always, felt to be
precursors to some other utterance or a sequence of utterances. Technically,
they can be viewed as purely formal tools of conversation management, as
techniques for selecting next speaker. However, they are more than that.
They occupy a position which is midway between formal and content-related
aspects of conversation, since they may or may not forward a sequence to its
prefigured action according to the category of response they elicit.

116
Types of pre-sequences

Pre-sequences take the form of adjacency pairs. A characteristic feature of

pre-sequences is that, unless the right response is got to the first part of the
pre-sequence there isn’t going to be the case of a sequence. For instance, if
If one does a pre-invitation, then, unless one gets the right return, one doesn’t
do an invitation. Thus, a pre-sequence is a sequence which includes a turn
recognizable as potential initiation of another specific type of turn.
By prefiguring an upcoming action, pre-sequences invite
collaboration in that action (as in pre-closings) or collaboration in avoiding
explicit action (as in pre-self-identifications). Pre-invitations are treated as
transparent by recipients so that their responses are attuned to the fact that an
invitation is forthcoming in the next turn. In the exchange in (2) nothing is
intended to mean ‘nothing that would make the invitation irrelevant’.

(2)

A: What are you doing? (Position 1)

B: Nothing (Position 2)
A: Wanna a drink? (Position 3)
B: sure (Position 4)

Similarly, in (3) just shows that the question is interpreted as a pre-invitation

and it says ‘give me the invitation and I’ll accept it’

(3)
A: Are you studying?
B: No, I’m just reading.

A pre-request usually precedes a request of some kind. Its function is to

make sure that the request about to be made is within the limits of the
possible, from the point of view of the person who is requested to perform
the action. The function of pre-announcements is to gain ratified access to an
extended turn at talk. As an illustration, consider the example in (4):

(4)

A: Guess what?
B: What?
A: Professor Smith came in this morning and put another book on his order

117
The analysis of empirical data has shown that pre-sequences evince the
following structure:

T1 (Position1): pre-sequence first part, generally checking whether some

condition obtains for the action performed in T3
T2 (Position2): pre-sequence second part, generally indicating that the
condition obtains, often with a question or request to proceed to T3 (i.e. go-
ahead)
T3 (Position3): the prefigured action, conditional on the ‘go ahead’ in T2
T4 (Position 4): response to the action in T3

(Schegloff 1988)

T3 is conditioned by T2. As an illustration of a pre-sequence that forwards

the sequence to its core action, as a result of the go-ahead in position 2,
consider the example in (2).

In the absence of an encouragement in T2, the sequence can be expected to

abort on the following lines:

T1 – as above
T2 – answer indicates that the precondition on action does not obtain often
so formulated as to discourage the foreseeable action
T3 – withholding the prefigured action, usually with a report of what would
have been done in T3, by way of explanation of T1

(Schegloff 1988)

The example in (5) illustrates a pre-arrangement which does not forward the
sequence to the prefigured actions. The turn in position 2 shows that
arrangements for future contact cannot be made their function is to check
whether arrangements for future contact can be made.

(5)

A: What are you doing today? (Position 1)

B: I’m supervising in the morning (Position 2)
A: Would you like to come by after that? (Position 3)
B: I can’t. I am afraid no (Position 4)

118
6.6 Repair

The practices of repair27 refer to efforts in talk-in-interaction to deal with

trouble or problems in speaking, hearing and/or understanding. A distinction
has been drawn between self-repair (i.e. repair done by the speaker of
problem/repairable item) and other-repair (repair done by another
party/correction offered to the speaker). Self-repairs can be self-initiated i.e.
done by the current speaker without prompting or other-initiated i.e. repair
done by the current speaker with prompting. These practices of repair are
organized by reference to where it is initiated and this is best characterized
by relationship to the trouble-source i.e. the talk which is being repaired. The
following are the main positions from which repair is initiated: the turn
which contains the trouble source (same turn repair); the next turn; the third
turn; the fourth turn relative to the turn in which the problem items occur.
Repairs are often used as strategic devices: correcting oneself, in
particular, can be a way of gaining time to think, or a means of preventing
someone else from jumping into the conversation at an upcoming TRP. As
such they can frequently occur in competitive conversational styles where
participants are oriented towards dominating the floor.
Of particular importance in applying a CA treatment of this
phenomenon to interactional sociolinguistics are the manipulative instances
of repair deployed to dominate the floor (and thereby to produce singly-
developed floors as opposed to collaborative floors28), as well as the non-
deployment of repair in contexts that make it relevant; in these latter contexts
the non-deployment of repair may be interpreted as an indication of
agreement and intersubjectivity. Engaging in such practices of repair will
display participants’ orientation to raising problems in understanding and
thereby provide a data-internal warrant for the use of terms such as
‘communal/agentic’ speaker as relevant characterizations of the parties.
Similarly, lack of deployment of practices of repair in contexts where
they are relevant will make room for interactional strategies that are
instrumental in creating and achieving intersubjectivity or “reciprocity of
perspectives” (see Cicourel 1974, Schutz 1962, Schutz and Luckmann 1977)
by invoking shared knowledge and shared experience. Among such
conversational strategies, of special interest would be minimal responses
functioning as markers of agreement or the use of general extenders.

27
Relevant bibliographical resources include among others Schegloff, Jefferson and Sacks
(1977); Schegloff (1979, 1987a, 1992, 1997a, 1997b, 2000b) and Jefferson (1974, 1987)
28
For an extensive analysis of these two types of floor, see Edelski (1993).

119
Consistent use of these strategies (or their non-deployment) would reinforce
participants’ orientation towards communal or agentic conversational styles
respectively.

6.7 Tying techniques

Tying techniques play an important role in ordering the various parts of a

conversation. Sacks (1995) identified a large range of tying techniques and a
very large group of specifically tying items. The presence of such items in
certain positions in an utterance signals that the utterance is tied to some
prior utterance. The elaboration of tying is a means of showing that one
understood any given utterance that went before. Tying utterances together is
a local operation, i.e. it is done in a next utterance and therefore done by
only one person.
Sacks (1995) identified the following ways of tying utterances
together:

1. If some given utterance is ended in a noun for tying one’s utterance

to the previous one, one can turn that noun into an –ing participle and make
the latter syntactically consistent with the preceding utterance.
2. Another rule concerns the use of prepositional phrases. If an
utterance ends in a preposition, then if you begin the next utterance with a
noun phrase that functions as the complement of that preposition, you tie
your utterance to the previous one.
3. Another widespread tying technique involves the use of
conjunctions. If an utterance ends up with what might be a sentence, if you
begin the next with a conjunction and make it syntactically consistent, then
you have tied the second utterance to the first one.
4. Items such as now, well, then when they occur appositionally in front
of an utterance, operate as tying terms.
5. There is a variety of tying terms, and therefore tying rules which,
while they can operate within a sentence or within a single utterance
composed of several sentences, they can also operate across utterances:

- one group of such prototypical terms are pronouns

- another prototypical group involves pro-verbs (do, be, say)
- various combinations of: pronouns and verbs; pronouns and pro-verbs;
nouns and pro-verbs

Some tying techniques, especially those mentioned in (1) and (2) can
be used as a means of building sentences collaboratively. In producing a
120
collaboratively built sentence, a speaker A produces a syntactically and
semantically complete turn made up of a single sentence which is then turned
by another party B into a seemingly incomplete turn which then B completes
by adding a dependent clause or prepositional phrase which are syntactically
and semantically consistent with the initial turn. The first utterance is a
complete sentence on its own, but the second utterance is not a sentence. It
only becomes a sentence together with the first one. The example in (W/6),
due to Sacks (1995), illustrates the use of collaboratively built sentences in
teenagers’ discourse:

(6)

A: We were in an automobile discussion

B: discussing the psychological motives
C: for drag racing on the streets

(Sacks 1995)

As Sacks (1995) put it: “If you want to find a way of showing somebody that
what you want is to be with them, then the best way to do that is to find some
way of dividing a task that is not easily dividable and which clearly can be
done by either person alone”. Thus, collaboratively built sentences can be
used as strategic devices to show that the people who produce them are close
to each other and they know what is on each other’s minds.

6.8 The monitoring of conversation

Speakers can monitor the conversation by employing various strategies that

allow them to avoid silence (since the main point of conversational structure
is to keep the flow of conversation going) or face-threatening topics (e.g.
God, sex, death, political orientations). One way of doing that is for a
speaker to start a stretch talk by inserting a joke. An important feature of
jokes is that they come in rounds: when a joke is used that is the first joke
and that will provide the occasion for each person present to have a
particular chance to talk, i.e. a chance to tell a joke. Telling a joke is
instrumental in avoiding face-threatening acts due to their unaffiliated
character which results from the fact that the speaker does not disclose his
position by using it. Moreover when there are more than 2 persons present
and their right to talk may be obscured due to large status differences, the
occurrence of a first joke gives the others the opportunity to talk.

121
 Another strategy that can be used in the monitoring of conversation,
especially when silence seems to be present is asking questions. Apart from
functioning as requests for information or clarification, questions can be used
as strategic devices for keeping the flow of conversation going.
As a rule, a person who asks a question in ordinary conversation has
the right to talk again after the addressee talks. This rule provides a
convenient way of generating blocks of talk. On the other hand, one can
produce an answer to a question and append to it another question and thus
taking over, i.e. gaining the floor and therefore being in control of the
conversation.
Sacks (1995) points out that there can be a sense in which, while one
is asking questions they could not be said to be in control. Consider the
examples in (7) and (8):

(7)
A: Is your husband a police officer?
B: No. (the questioner is in control of the conversation)

Sacks (1995)

(8)

A: Is your husband a police officer?

B: No, he is such-and-such (the questioner is not in control of the
conversation)

Sacks (1995)

In the example in (8) B’s answer is based on a kind of co-operation which is

not present when the questioner is in control of the conversation and which
allows the speaker to claim rights to a relatively extended turn at talk.
A related aspect, so far as the monitoring of conversation is
concerned, has to do with the phenomenon known as appendor
clauses/questions. Using an appendor clause is a technique for reselecting
the last speaker as next speaker (Sacks 1995).

(9)

A: They make miserable coffee

→B: across the street? - appendor clauses/questions.

122
(10)

A: And this is something fearful, you know

→B: to you? - appendor clauses/questions.

Appendor clauses are routinely interpreted by the recipient as producing a

question that the person who started that sentence (i.e. the recipient) should
answer and routinely the recipient does answer. In employing this strategy a
frequently used device is phrases with or without a question intonation
attached, as appendor clauses, to the end of a syntactically and semantically
complete sentence. Consider the examples in (9) and (10) above and the
example in (11).

(11)

A: …..and it turned out very good

→B: from the old man’s point of view? - appendor clauses/questions

Thus, a first person starts an utterance that can be a complete sentence and
the second person appends a prepositional phrase to it, which is syntactically
coherent with the previous sentence and may have question intonation.
Another way of speakers’ going about monitoring when they want or
ought to be talking or when embarrassing silence seems to be present is the
use of floor seekers. A characteristic feature of floor seekers is that they are
utterances that have as their feature relative to selection, that they are
attempts to have their speaker selected as next speaker in the next turn. They
normally do not select a next speaker (anyone can answer them). Their
function, however, is to have the person who produces them reselected by
whoever self-selects as next speaker (Sacks 1995). The example in (12) is an
illustration of floor seekers.

(12)

→A: You know what happened to me last night? – floor seeker

B: What?
A starts telling the story

They are the first part of an adjacency pair which regularly takes the format
of a pre-sequence, namely a pre-announcement. Like various other first part
members they do not necessarily question intonation. They can be delivered
simply flat out as “I heard a good joke”. Moreover, if they are not responded
123
they are repeated. By producing a floor seeker one can get the floor, i.e. the
right to deliver the next turn. Even if what they get is ‘Big deal’ that is a
good enough take-off point.
Floor seekers are used as strategic devices to get an occasion to tell a
story or a joke. One can take almost anything said and use it in a controlled
fashion, turning it thus into a way of getting the floor. Consider the example
in (13):

(13)

A: I was at the police station this morning

B: Big deal
A: Big deal, yeah. Somebody stole all my radio equipment outta my car.

(Sacks 1995)

This strategy enables us to distinguish between: (i) having the floor in the
sense of being a speaker while others are hearers, something which floor
seekers achieve; and (ii) having the floor in the sense of being a speaker
while others are doing whatever they please.
Within an extended turn one way of indicating that one is listening
includes head nods, smiles, facial expressions, gestures and most commonly
vocal indications referred Minimal responses, also referred to as back-
channel communication (Yngve 1970), assent terms (Woods 1988) or
accompaniment signals29(e.g. uh-uh, yeah, mm, etc.). These types of signals
provide feedback to the current speaker that his message is being received.
Schegloff (1982) identifies two usages of minimal responses. The
most common usage is to exhibit on the part of its producer an understanding
that an extended unit of talk is underway by another and that this unit is not
yet complete. When so used utterances such as uh huh are termed
continuers30. In producing a minimal response functioning as a continuer,
speakers display their understanding that an extended turn is underway and

29
Throughout the literature two main characterizations have been offered to deal with
minimal responses. According to one, they are evidence of attention and understanding on
the listener’s part. Thus according to Kendon (1967:44), in producing a minimal response
the addressee ‘…appears to do no more than signal that he is attending and following what
is being said’. A second characterization of such behaviour is that it ‘…keeps the
conversation going smoothly’ (Dittman and Llewellyn 1967:342) or ‘… appears to provide
the auditor with a means for participating actively in the conversation thus facilitating the
general coordination of action by both participants’ (Duncan and Fiske 1977: 202-203).
30
Other researchers have used the term facilitative for a continuer (Reid 1995).
124
show their intention to pass the opportunity to take a turn at talk and in so
doing they facilitate the continuation of the turn that is underway. As an
illustration of the use of minimal responses as continuers consider the
excerpt in (14) which shows Maria to be attending to the ongoing talk at
lines 2, 8, 13, and 18. This involves, on the one hand, Maria’s refraining
from initiating a turn in order to show that she does not object to her
partner’s having the floor and producing an extended unit of talk.

(14)
1 Alina: şi acum io ştiam c-aveam nişte oase în frigider de vită
and now I knew I had some beef bones in the fridge
2 Maria: mhm
mhm
3 Maria: şi zic lasă din oasele alea fac ciorbă şi din carne fac spaghete
and I thought I’d make some soup with those bones and spagetti with the
meat
4 şi cînd dimineaţa am constatat că de fapt aveam aveam numai carne
but in the morning I noticed that in fact I had only meat
5 şi zic a:: şi ciorbă din carne de vită şi spaghete tot–
and I thought, well, beef soup and spagetti as well
6 toate sînt prea cu carne de vită
too much beef in everything
7 lasă o să fac p[ui
well, I’ll cook some [poultry
8 Maria: [mhm=
mhm
9 Alina: =cu spaghete=
with spaghetti
10 Alina: =şi le-am făcut
so I cooked them
11 le-am fiert pe ele
I‘ve boiled them
12 apoi am făcut un sos cu ceapă, usturoi, bulion şi un cub de Knorr de pui
then I made a sauce with onion, garlic, tomato paste and a Knorr poultry
cube
13 Maria: mhm
mhm
14 Alina: ca să aibă cît de cît gust de pui
to taste a little bit like poultry
15 Maria: [((laughing))
16 Alina: [((laughing))
17 Alina: aşa şi apoi am pus carne de pui tăiată aşa fîşii lun[gi
so, and next I put some beef carved in long strips
18 Maria: [mhm
125
 mhm

(Hornoiu 2016)

Not objecting to her partner’s contribution and encouraging her to carry on

with her stories enables the conversation to flow smoothly, when willingness
to take part in a conversation is equivalent to willingness to be part of a
relationship. On the other hand, fitting to the details of the locally preceding
talk is achieved through the use of minimal responses functioning as
continuers. In this example, Maria satisfies her partner’s desire to be liked,
admired and listened to.
As the example in (14) shows, the inserter signals that she is
constantly attending to what is being said, demonstrating her participation,
her interest in the interaction and in the speaker since none of these minimal
responses is delayed. Minimal responses functioning as continuers are
skilfully inserted between the breaths of a speaker, causing sometimes slight
overlaps and there is nothing in tone or structure to suggest that the person
who produces them is attempting to take over the talk. Thus they are signals
of solidarity and support.
Minimal responses can also be used as markers of agreement when
they signal the passing up of an opportunity to initiate repair (Schegloff
1982). If any talk can be viewed as a trouble-source, then ‘after any talk’ can
be a place for repair to be initiated on it. Vocalizations like ‘uh huh’, ‘mm
hmm’, ‘yeah’, ‘right’ as well as head-gestures such as nods, in passing an
opportunity to initiate a full turn at talk, can be seen to be passing an
opportunity to initiate repair on the immediately preceding talk. Passing the
opportunity to raise problems of understanding may be taken as indicating
the absence of such problems and hence the absence of disagreement. Thus
minimal responses that occur in places where other-initiated repairs are
potentially relevant, i.e. after completion of any unit of talk (Schegloff
1982), are taken as indications of agreement. Consider the excerpt in (15)
which illustrates the use of minimal responses as indicators of agreement.

(15)

1 Raluca: dacǎ ar fi totul prea frumos aşa ştii cînd e prea bine ştii cǎ a greşit undeva
if everything was too good, then, you know, when it’s too good, you know
that he’s done something wrong
2 Mona: mhm
mhm
3 Raluca: ştii şi te şi plictiseşti la un moment dat
you know, and you get bored at some point
126
4 Mona: mhm
mhm
5 Raluca: de atîta bine omu’ se plictiseşte
if everything is too good, one gets bored

(Hornoiu 2016)

The minimal responses in (15) occur at transition relevant places, i.e.

after the completion of a unit of talk when speaker change may occur and
other-initiated repairs are potentially relevant. However, in all cases those
who utter them refrain from initiating repair, thus implying that they agree
with their interlocutor’s opinions (Schegloff 1982). During face-to-face
interaction this absence may be interpreted as a way of withholding
agreement, leading to an inference of disagreement.
Speakers can also keep the flow of conversation going through the
use of phatic questions. Questions are interactionally powerful utterances, as
they open a two-part sequence. In conversation questions primarily serve two
functions. They often function as information-seeking devices. A speaker
can assume the role of ‘expert’, while the others ask the ‘expert’ questions.
On the other hand, asking question may be a way of inviting others to
participate; of checking that what has been said is acceptable to those
present, of ensuring that conversation continues (Coates 1993; 1995) as the
following excerpts show.
The excerpt in (16) illustrates the use of phatic questions as a means
of keeping the flow of conversation going.

(16)

1 Iulia: ai fost la croitoreasă?

have you gone to the dressmaker’s?
2 Maria: eh, am fost pe naiba
to hell I’ve gone
3 Iulia: de ce? [păi n-ai zis ca te duci astăzi?
why? didn’t you say you’d go today?
4 Maria: [păi ( )
well ( )
5 Maria: nu ţi-am zis c-am sunat-o mă
I haven’t told you that I called her
6 Iulia: eh, şi? ce a zis?
and? what did she say?
7 Maria: am fost bolNA:vă
127
 I’ve been ILL
8 Iulia: da’ ce a avut?
what was wrong ?
9 Maria: eh, a avut pe dracu’ ( ) a avut de o lună de zile–
eh, she didn’t have shit ( ) she’s been ill for a month
10 Iulia: n-a avut clienţi
she had no clients
11 Maria: a::re prea multi clienţi
she’s got too many clients
12 Iulia: are clienţi?
she’s got clients?
13 Maria: ia clienţi= prea mulţi şi p’ ormă nu le face faţă=
she takes on too many clients and then she can’t manage them
14 Iulia: =şi
and
15 Iulia: =e:h, păi da’ e neserioasă lasă că te duc–
e::h, well, then she’s not reliable I’ll take you
16 să vedem cum îmi ies mie pantaloni şi te duc la:: la tipa asta ştii
let’s see how my trousers turn out, and I’ll take you to this lady, you
know
17 Maria: păi chiar m-am gândit la ea ştii
well, actually. I thought about her you know
18 Iulia: da şi ţi le-a dat până la urmă sau nu?
well, and did she give them to you or not after all?
19 Maria: nu mâine tre’ să mă duc la ea să văd dacă–
no, I’ve got to go to her tomorrow to see if
20 deci s-o sun înainte
so, I’ll call her first
21 cre’ că m-au costat telefoanele
I think I’ve run up huge phone bills
22 am sunat-o de şa:pte mii de ori
I’ve been calling her seven thousand times

(Hornoiu 2016)

In line 1 Iulia uses her first question (“ai fost la croitoreasă”) to establish a
topic (i.e. the dressmaker) on which to talk. Since Maria seems to be
unwilling to discuss it, Iulia makes a second attempt to launch the
conversation in line 3 where she makes use of another question to invite
Maria to participate. Maria’s answer in 5 seems to be more encouraging,
offering a piece of information that can be turned into a good take-off point.
From this point on, Iulia asks a series of questions in lines 6, 8 and 18. Our
assumption is that Iulia is not as much interested in the information these

128
questions elicit as she is in ensuring that conversation goes on. In other
words, these questions are rather phatic in intent: what exactly is talked
about seems to be less important than the fact that talk itself occurs.
Apparently Iulia has achieved her aim of keeping the conversation
flowing: Maria seems to be willing to contribute to the topic as her turns at
speaking become longer. Both conversational exchanges demonstrate how
questions can be used to keep the flow of conversation going when
willingness to engage in interaction is equivalent to being part of a
relationship.

6.9 Accounts for conversation

In the absence of some obvious reason for the conversation to take place, by
virtue of two persons being co-present and nothing else, one can initiate an
exchange by making use of a type of question whose function is to indicate
that although it does not seem to be the case, there is indeed a reason. Sacks
(1995) is the first to draw attention to this strategy which he calls accounts
for conversations The examples in (17) illustrate this conversational
strategy:

(17a) Don’t I know you from somewhere else?

(17b) Didn’t I see you at such-and-such place?
(17c) Didn’t you go to such-and-such school?

This type of questions provides for the fact that the speaker and the
addressee seem to know each other, even though this is not the case. If,
however, it turns out that they do, then the conversation in question can take
place as a ‘further conversation’. They make up a class of questions which
provide an account for a conversation to take place and develop. Other
examples include the utterances indicated by ‘→’ in (18):

(18)

→A: When does the train arrive?

B: 7.15
→A: Are you going to X, too?

These questions are standardized in that they provide for the relevance of the
turn-taking rules, such that one knows what an answer to such a question
looks like, so that one who has asked the question can know when the
utterance that stands for an answer will have been finished providing thus
129
that the other can talk again. Unlike discursive talk, where it may not be clear
when the speaker has finished, this type of standardized questions is
convenient when someone makes an effort to get acquainted to the other
person, as they do not need to wait too long after the other has stopped. Thus,
accounts for conversation are an efficient strategy that enables the
conversation to unfold smoothly with no or very little silence (waiting too
long may imply withdrawal from the interaction and may be perceived as
withdrawing from the relationship) or overlapping talk (interrupting the
interlocutor is generally viewed as being rude).

6.10 Conclusions
Because this book does not focus on conversation analysis, certain key areas
of CA research such as preference organization and repair have not been
addressed; even topics for topics that we have considered, such as turn-
taking, sequential organization, overlapping talk, it provided a cursory
examination. It is, therefore, advisable for readers to familiarize themselves
with more in-depth introductions to CA.
Good comprehensive accounts can be found in Hutchby and Wooffitt
(1998) and ten Have (1999). Short, sophisticated chapter-length
introductions can be found in Drew (1994) and Heritage and Atkinson
(1984). Heritage (1984a) offers an excellent introduction to CA in his superb
account of ethnomethodology and its origins. A short but useful discussion
of the relationship between CA, Goffman and Garfinkel can be found in
Heritage (2001).
On the relationship between CA and ethnomethodology should first
consult Garfinkel (1967), Heritage (1984a) and Clayman and Maynard
(1995) offer useful. In order to explore how the origins of CA were
influenced by Goffman’s studies of the moral and social order of everyday
life, an extended and focused treatment can be found in Schegloff (1988a).
Manning (1992) offers a very good oveview of Goffman’s work.

130
 A GLOSSARY OF PRAGMATICS

A
ADDRESSEE see under speech event participants
ADJACENCY PAIRS see under conversational analysis
AGREEMENT MAXIM One of the maxims of politeness proposed by Leech.
The maxim is fairly straightforward. It instructs the speaker to:
Maximise agreement with hearer.
Minimise disagreement with hearer.
The effect of this maxim is illustrated in the following:
A: Do you agree with me?
B: Yes. (slightly less polite); Absolutely. (more polite)
A: Do you agree with me?
B: No (less polite); Up to a point, but … (more polite)
ANAPHORA, ANAPHOR An anaphor is an expression that must be interpreted
via another expression called the ‘antecedent’, which typically occurs earlier
in the discourse. The term ‘anaphora’ refers to this phenomenon. In the
examples below, the anaphor and its antecedent are given in bold type:
1. John came in and he lit the fire.
2. Pete was driving a blue car. I’m pretty sure it wasn’t insured.
3. Barack Obama arrived in London this morning. The President will
address the Cabinet tomorrow.
This type of anaphora is called ‘coreferential anaphora’, because anaphor
and antecedent share the same referent.
In ‘non-coreferential anaphora’, as in John shot a fox; Mike shot one, too,
the default interpretation is that anaphor and antecedent have different
referents. In some cases, the antecedent occurs later in the discourse; this is
131
sometimes called ‘cataphora’. A distinction should be made between
anaphoric and exophoric expressions. The latter refer directly rather than
through antecedents:
(Woman pointing to a man) He was the one who snatched my bag.
APPROBATION and MODESTY MAXIMS These belong in the set of maxims
of politeness proposed by Leech. The former being oriented towards the
hearer, while the latter towards the speaker. Leech’s formulations of these
maxims (slightly modified) are as follows:
Approbation Maxim: Maximise praise of hearer.
Minimise dispraise of hearer.
Modesty Maxim: Minimise praise of self.
Maximise dispraise of self.
‘Dispraise’ includes criticism, blame, belittlement, etc. These two maxims
are self-explanatory: exaggerate anything that puts the hearer in a relatively
good light, and understate anything that puts the hearer in a relatively bad
light. These maxims instruct participants in the speech event that self-
directed boasting is impolite and self-belittlement is polite:
A: You were brilliant!
B: Yes, wasn’t I? (less polite)/ I was lucky. (more polite)
A: What a fool I’ve been!
B: Indeed. (less polite)/ These things happen. (more polite)

B
BYSTANDER see under speech event participant

C
CATAPHORA see under anaphora
CLAUSAL IMPLICATURES see under generalised vs. particularized
conversational implicatures

132
COMMON GROUND This refers to aspects of knowledge that participants in
the speech event assume to be shared and therefore do not need to be spelled
out. It includes what can be perceived in the immediate context, together
with knowledge of the language, general world knowledge, shared attitudes,
shared world views, and so on.
COMPETENCE VS PERFORMANCE This distinction is proposed by Chomsky.
Competence is conceived as the neutral representation of the system as it
exists in the mind of an ideal language user. Performance basically refers to
the processes involved in the production of utterances, rather than the
produced utterances themselves.
COMPOSITIONALITY (PRINCIPLE OF) According to the principle of
compositionality, the meaning of a complex expression is a compositional
function of the meanings of its parts. In other words, we work out the
meaning of an expression containing more than one meaningful element by
combining the meanings of its constituents. So, to get the meaning of, say,
The cat ate the fish, we add together the meanings of the individual items:
‘The cat ate the fish’ = ‘the’ + ‘cat’ + ‘ate’ + ‘the’ + ‘fish’
However, not all expressions of a language conform to this principle. Those
that do are described as ‘compositional’; those that do not are described as
‘non-compositional’ or ‘semantically opaque’. Semantic opacity (which is a
matter of degree) is a prototypical characteristic of idioms.
CONSTRAINTS ON RELEVANCE This notion is used by proponents of
Relevance Theory to account for certain aspects of non-propositional
meaning, especially the meanings of elements such as but or what’s more.
For instance, compare the pair but vs. and. Clearly, the sentences Liz is
blonde and beautiful and Liz is blonde but beautiful do not convey the same
meaning. Yet they have the same propositional content: they are true and
false in the same circumstances. The difference between but and and relates
to the relevance of what follows. The word but indicates that the information
that follows it contradicts some belief or assumption on the part of the
hearer. Similarly, what’s more is interpreted as indicating that what follows,
in the speaker’s opinion, reinforces some prior belief or opinion.
CONTEXT An essential factor in the interpretation of utterances and
expressions. The most important aspects of context are: (1) preceding and
following utterances and/or expressions (‘co-text’), (2) the immediate

133
physical situation, (3) the wider situation, including social and power
relations, and (4) knowledge presumed shared between speaker and hearer.
CONVENTIONAL IMPLICATURES These are components of the meanings of
utterances which are not propositional in nature, but which have a stable
association with particular linguistic expressions and which therefore cannot
be cancelled without causing anomalous sentences or utterances. For
instance, John hasn’t registered yet and John hasn’t registered are
propositionally identical, but the presence of yet in the former implicates that
John is still expected to arrive. Contradicting this implicit meaning leads to
oddness: ?John hasn’t registered yet and I know for a fact he does not intend
to.
CONVERSATIONAL ANALYSIS This is an area of study, nowadays usually
considered a branch of pragmatics, concerned with the structure of naturally
occurring conversation. The approach is strictly empirical. The empirical
data consist of recordings and transcripts of naturally-occurring
conversations. The empirical data are analyzed without theoretical
preconceptions, whether semantic, philosophical, or deriving from other
branches of pragmatics. The aim is to extract regularities of regarding the
overall organization and aspects of turn-taking. The basic unit of description
in conversational analysis is the ‘turn’ (sometimes called the ‘turn
constructional unit’). This is an uninterrupted contribution of one speaker to
a conversation, generally followed and preceded by a change of speaker
unless it represents the beginning or end of the conversation. Turns are said
to be ‘latched’ if there is no detectable gap between the end of one turn and
the beginning of the next. Sometimes turns, or parts of turns may be
delivered in ‘overlap’. When overlapping talk occurs a resolution system is
applied to deal with it, since the main point of conversational structure is to
minimize gap and overlap. A slight pause may signal a ‘transition-relevance
place’, a place where speaker change may occur. A speaker may start to say
something, then change their mind about what to say; this phenomenon is
known as a ‘repair’. Conversations are structured in a number of ways. For
instance, certain utterances serve to initiate a conversation (e.g. Hi!), while
others serve to terminate them (e.g. See you later!). Some turns form natural
pairs, called ‘adjacency pairs’. Examples of these are question and answer,
greeting and response greeting, invitation and acceptance or refusal, and
apology and acceptance or rejection.
CONVERSATIONAL IMPLICATURES One of two basic types of implicature
(the other type being conventional implicatures). Conversational
implicatures have four main characteristic properties:
134
1. They are not entailments, that is, they do not follow logically from what
is said. For instance, we can infer from ‘Pete has a cousin’ that ‘At least one
of Pete’s parents is not an only child’, but since this is an entailment it is not
a conversational implicature. In the example below:
A: Can I speak to Jane?
B: Jane’s in the shower.
the inference from B’s answer, that Jane is not able to take a telephone call,
is not an entailment, but an implicature
2. They are ‘cancellable’ (or ‘defeasible’). They are relatively weak
inferences and can be denied by the speaker without contradiction. For
instance, B’s reply in the following would normally be taken to mean ‘I
don’t intend to tell you’:
A: How old is she?
B: That’s none of your business.
If B added ‘But I’ll tell you, anyway’ this would cancel the inference, but no
self-contradiction arises. This is a characteristic feature of conversational
implicatures. In contrast, an attempt to cancel an entailment leads to a
contradiction:
?Pete has a cousin, but both his parents are onlychildren.
3. Implicatures are context dependent or ‘context sensitive’, in that the same
proposition expressed in a different context can give rise to different
implicatures. Compare the two exchanges below:
A: Can I speak to Jane?
B: Jane’s in the shower.
vs.
A: I think I’ll take a shower.
B: Jane’s in the shower.
In the latter B’s utterance implicates ‘You can’t take a shower just yet’, not
‘Jane can’t accept a phone call’.

135
4. They are ‘non-detachable’. In a particular context the same proposition
expressed in different words will give rise to the same implicature. In other
words, the implicature is not tied to a particular form of words (cf.
conventional implicatures). For instance, if B in 2 above had said ‘That
doesn’t concern you’, the implicature would be the same.
5. They are ‘calculable’. They can be worked out on the basis of using
general principles (e.g. the co-operative principle and the Gricean
conversational maxims) rather than requiring specific knowledge, such as a
private arrangement between A and B that if one says X it will mean Y. (See
also under generalised vs particularised conversational implicatures.)
CO-OPERATIVE PRINCIPLE This principle was proposed by the philosopher
Grice, in his theory of implicature, to account for how conversational
implicatures arise. In Grice’s view, a conversation is a co-operative activity
in which participants tacitly agree to abide by certain norms. His formulation
of the general principle runs as follows: “Make your conversational
contribution such as is required, at the stage at which it occurs, by the
accepted purpose or direction of the talk exchange in which you are
engaged.” Grice spelled out the norms in greater detail in the form of a set of
maxims of conversation.
COREFERENTIAL ANAPHORA see under anaphora
COST-BENEFIT SCALE see under Tact and Generosity Maxims
CO-TEXT see under context

D
DECLARATIVE This is usually considered to be grammatically the most
basic sentence form (John is lazy, as opposed to Is John lazy?, Don’t be lazy,
John! or How lazy John is!). Prototypically, the declarative form encodes a
statement, that is, a proposition, together with a commitment to its truth.
The basic meaning of a declarative is related to the meanings of explicit
performative verbs such as state, assert, declare, aver, announce, and so
on, but is more general than any of these. A declarative sentence can acquire
additional illocutionary force through implicature, as in There’s ice on the
road uttered as a warning.
DEIXIS, DEICTIC EXPRESSIONS Deictic expressions form a subtype of
definite referring expressions. They can be viewed as expressions which
‘point to’ their referents. Usage of the term ‘deixis’ is variable, but most
136
typically it designates referring expressions which indicate the location of
referents along certain dimensions, using the speaker (and time and place of
speaking) as a reference point or ‘deictic centre’. The deictic centre is
generally organized in an egocentric way, i.e. the referents of these
expressions are established relative to the speaker, the time when the speaker
delivers his utterance (i.e. the coding time) and the speaker’s location at
coding time. An example is the use of this and that. In Can you pass me that
newspaper, the newspaper in question is typically relatively distant from the
speaker; however, once the speaker receives the newspaper, any further
reference to it will require a different deictic element: I’m going to have to
stop buying this newspaper. A change of this sort, made necessary by a
change in the relation between the referent and the speaker, is diagnostic for
a deictic element (items like this and that are sometimes called ‘shifters’).
Certain verbs of motion encode direction relative to one of the participants in
a speech situation, and may thus be called deictic. For instance, come
denotes motion towards either speaker or hearer (Come and see me
sometime, I’ll come and see you tomorrow), while go denotes motion
towards a third person (You/I should go and see him). The verbs bring and
take have a similar relation (I’ll bring it to you, I’ll take it to him). There are
three main sub-types of deixis: spatial, temporal, and person deixis, and
two minor sub-types: social and discourse deixis. Occasionally, the deictic
centre is not the speaker (see under projected deixis).
DISCOURSE DEIXIS It concerns reference to discourse items which occur
either before or after the current time of speaking. When functioning as
discourse deixis, that typically refers to a previously occurring item, and this
to something which is still to come: That was the best story I’ve heard for a
long time, Wait till you hear this … Expressions such as therefore, however,
on the other hand, anyway which relate portions of earlier discourse to
portions of later discourse, are also included under discourse deixis. They
express how the utterance is related to surrounding discourse.
DISCOURSE MARKERS A category of expressions which includes such items
as well, oh, then, so, but. They are grammatically optional, in that omitting
them does not result in ungrammaticality, and they mark boundaries between
units of discourse. Their typical functions include:
1. They carry expressive meaning (they may also have propositional
meaning).
2. They contribute to, or emphasize coherence relations in discourse.

137
3. They act as constraints on relevance. (These are not mutually exclusive.)

E
EAVESDROPPER see under speech event participant

EXCLUSIVE FIRST PERSON PLURALS see under person deixis

EXPLICATURE see under relevance theory

F
FELICITY CONDITIONS These are conditions that must be satisfied for a
speech act to be successfully performed. They can be grouped under three
headings: preparatory conditions, sincerity conditions, and essential
conditions.
1. Preparatory conditions define an appropriate setting for the act, including
the speaker’s intentions and qualifications. For instance, someone uttering
the words I pronounce you man and wife has not sealed the union of a man
and a woman unless he or she is properly qualified, and does so in the course
of an official marriage ceremony; the issuer of a command must have
authority over the addressee, and the act must be both possible and not
already carried out. If the preparatory conditions are not satisfied, the speech
act has not been validly performed. In other words, it is said to have
‘misfired’.

2. Sincerity conditions require the speaker to be sincere: someone who

promises to do something must genuinely intend to do it; someone
congratulating someone must feel pleasure at that person’s good luck or
success; someone making a statement must believe it to be true, and so on. If
the sincerity conditions for a particular speech act are not met the speech act
has nevertheless been performed, but the speaker is guilty of an ‘abuse’.

3. Essential conditions define the essential nature of the speech act. For
instance, if someone makes a promise, they must intend their utterance to
count as putting them under an obligation to carry out what is promised; in
the case of I name this ship … the speaker must intend the utterance to count
as conferring a name on the ship; in making a statement, a speaker must
intend it to be taken as true, and so on. If the essential conditions for a
particular speech act are not met, then merely producing the right form of
utterance does not result in the speech act being performed. For instance,
producing The King of France is bald in a logic class would not normally
138
count as a statement committing the speaker to its truth. Notice that this is
different from sincerity: someone telling a lie intends their statement to be
taken as the truth.

FLOUTING THE (CONVERSATIONAL) MAXIMS Some conversational

implicatures arise when a speaker follows the maxims of conversation
(these are called standard implicatures), but others can arise when a speaker
deliberately goes against one or more of the maxims, provided that (1) it is
clear to the hearer that the ‘flouting’ is deliberate and (2) the speaker can
nonetheless be assumed to be obeying the Co-operative Principle and is
therefore breaking the rules for good communicative reasons. Consider the
following:

A: Where did you go last night?

B: Out.

In some circumstances B’s reply could be taken as a signal of non-co-

operation, equivalent to Mind your own business. But a situation can easily
be imagined where B gives no sign of opting out of the conversation.
Suppose Grandma, who has firm ideas about how teenage girls like B should
and should not spend their evenings, is within earshot. B’s reply could then
be interpreted as ‘I’d rather not say while Grandma is listening.’ In many
cases, deliberate flouting of one or more maxims is a signal that an utterance
is not to be interpreted literally. For instance, if interpreted literally, Boys
will be boys gives is totally uninformative, That man is a snake and The
chicken salad in the corner wants his coffee now are obviously untrue, and
Oh, brilliant! is not a comment taken in a literal sense when someone’s
‘repair’ of the toaster has resulted in the lights fusing. But all make sense
with appropriate non-literal readings.

FOREGROUNDING here are various linguistic devices for increasing the

salience of part of an utterance. One obvious device is to pronounce it with
emphatic stress:

PETE did the washing up yesterday.

Pete did THE WASHING UP yesterday.
Pete did the washing up YESTERDAY.

Notice that these different forms not only highlight different items, but also
introduce different presuppositions. Foregrounding can also be achieved
grammatically:
139
It was Pete who did the washing up yesterday.
It was yesterday that Pete did the washing up.
What Pete did yesterday was the washing up.
It was the washing up that Pete did yesterday.

Structures like those illustrated above are called ‘focusing devices’, and the
foregrounded part of the utterance is called the ‘focus’.

G
GENERALIZED VS PARTICULARIZED CONVERSATIONAL IMPLICATURES
These implicatures arise on the basis of the speaker’s following the maxims
of conversation. A distinction can be drawn between two types of
conversational implicature. An implicature counts as ‘generalized’ if it is a
default reading, that is to say it arises unless it is explicitly cancelled and is
to that extent independent of context. For instance, Some of the parents came
to the meeting will normally imply that not all of them did. But in Some of
the parents, if not all of them, came to the meeting the implicature ‘not all’ is
cancelled. The fact that this is not anomalous shows that we are not dealing
with an entailment. A ‘particularized’ implicature is one that depends on
specific contexts and is not a default message component. For instance, Jane
is in the shower does not convey a default message component ‘She cannot
come to the telephone’. This requires a particular context, such as the one
given below:

A: Can I speak to Jane?

B: She’s in the shower.

Generalised conversational implicatures can be further divided into I-

implicatures, M-implicatures, and Q-implicatures. These labels relate to
Grice’s maxims of conversation. I-implicatures are concerned with
informativeness and result from the speaker’s following the Maxim of
Quality; M-implicatures relate to Grice’s Maxim of Manner; Q-implicatures
are worked out on the basis of the Maxim of Quantity.

1. I-implicatures depend on the notion that we do not need to spell out what
the hearer would expect to be normally the case. For instance, in the case of
This car costs £15,000, we do not need to be told that the price includes the
wheels.

140
2. M-implicatures are based on the principle that if a speaker avoids a
standard way of saying something, then they do not wish to convey the
standard meaning. For instance, if an offering at breakfast is described as
‘partially charred pieces of bread’, rather than ‘toast’ we are entitled to
assume that it somehow falls short of standard expectations for toast.

3. Q-implicatures depend on the principle that a speaker will make the

strongest possible statement that is consistent with the facts. For instance,
Pete has three children normally implies ‘Pete has no more than three
children’. However, in special circumstances, this implicature may be
suppressed, as in

A: You have to have three children to qualify for this allowance.

B: Pete has three children.

B’s reply is perfectly justified if Pete has five children.

Q-implicatures like these, which depend on a scale of values of some sort,

are known as ‘scalar implicatures’. An example of item giving rise to scalar
implicature would be some which implicates “not all”. Some people believe
in God implicates Not all people believe in God. The so-called ‘clausal
implicatures’ also fall under this heading. For instance, ‘If P then Q’
implicates that the stronger statement ‘P, therefore Q’ cannot validly be
made. If Pete left early, he won’t have got the message implicates that the
stronger statement Pete left early, so he won’t have got the message cannot
be made.

GENEROSITY MAXIM see under tact and generosity maxims

GESTURAL DEIXIS This refers to the use of a deictic expression in a

situation where, prototypically, speaker and hearer are together and the
hearer can see what the speaker is doing. Gestural deixis, as the name
implies, typically involves a gesture on the part of the speaker. Examples are:
It was this big (speaker indicates a size with his hands); I want you, you, and
you to come with me (speaker points to three people); This is totally
unacceptable (speaker points to an offensive poster). An example involving
temporal deixis which does not strictly demand co-presence of speaker and
hearer (it could be done by telephone) but does require moment-by-moment
monitoring of the speech event by the hearer is:
Press the button … (pause) … NOW! See also symbolic deixis.
141
GIVEN VS NEW INFORMATION These notions are concerned with what is
called the ‘information structure’ of utterances. In virtually all utterances,
some items are assumed by the speaker to be already present in the
consciousness of the hearer, mostly as a result of previous discourse, and
these constitute a platform for the presentation of new information. As the
discourse proceeds, the new information of one utterance can become the
given information for subsequent utterances, and so on. The distinction
between given and new information can be marked linguistically in various
ways. The indefinite article typically marks new information, and the definite
article, given information: A man and a woman enteredthe room. The man
was smoking a pipe. A pronoun used anaphorically indicates given
information: A man entered the room. He looked around for a vacant seat.
The stress pattern of an utterance can indicate new and given information (in
the following example capitals indicate stress):

Pete washed the dishes. (in answer to Who washed the dishes?)
Pete washed the dishes. (in answer to What did Pete do?)

Givenness is a matter of degree. Sometimes the degree of givenness is so

great that the given item(s) can be omitted altogether (ellipsis):

A: What did you get for your birthday?

B: A computer. (The full form would be I got a computer for my birthday.)

GRAMMATICAL PERFORMATIVITY This refers to the signaling of

illocutionary force by grammatical means. See under declarative,
interrogative, imperative, exclamations.

H
HEDGE An expression which weakens a speaker’s commitment to some
aspect of an assertion:
She was wearing a sort of turban.
I’ve more or less finished the job.
As far as I can see, the plan will never succeed.
She’s quite shy, in a way.

142
I
IMPLICATURES These are related to the meanings of utterances which,
although intended, are not strictly part of ‘what is said’ in the act of
utterance, nor do they follow logically from what is said. There are two basic
sorts of implicature: (a) those which have a stable association with particular
linguistic expressions (conventional implicatures), such as the element of
surprise associated with yet in Haven’t you finished yet? (speaker does not
actually say he or she is surprised), and (b) those which must be inferred in
relation to a context and the maxims of conversation (conversational
implicatures), such as the implied negative in B’s reply in:

A: Can I speak to Jane?

B: She’s in the shower. Can you call back?

INCLUSIVE FIRST PERSON FORMS see under person deixis

INDEXICALITY For present purposes, indexicality can be taken as equivalent

to deixis.

INDIRECT SPEECH ACT This is an utterance that has the typical form of one
kind of speech act, but which functions, either typically or in specific
contexts, as a different type of speech act. Many instances of indirect speech
acts are highly conventionalized. The following are typical:

1. You will do as I say has the form of an assertive (i.e. makes a statement),
but commonly functions as a directive (i.e. tries to get someone to do
something).
2. Would you mind if I opened the window? superficially is a question
inquiring about the hearer’s attitude to a hypothetical event, but is a frequent
way of requesting permission.
3. Could you lend me a hundred pounds? literally is a question regarding the
hearer’s ability to do something, but is conventionally used as a (relatively
polite) directive or request.
4. What did I tell you? is literally a question, but conventionally functions as
an equivalent to I told you so!

INFORMATION STRUCTURE see under given vs new information, topic vs

comment, foregrounding

143
INTENSIFIERS These are words or expressions which strengthen or weaken
the degree of a property indicated by a relative adjective (or adverb).
Examples are: very, extremely, slightly, quite, rather, fairly, a little, a bit, on
the X side. Very and extremely strengthen the property (relative to the degree
indicated by the bare adjective) and slightly, a little, a bit weaken it. The
interpretation of several of these items depends on how they are pronounced.
Compare the following (stress indicated by upper case): This is QUITE
SUPERB; Well, it was QUITE good; I thought it was quite GOOD.

INTERROGATIVE The prototypical function of a sentence in the interrogative

form is to ask questions. A prototypical question expresses (a) a lack of
knowledge on the part of the speaker (exam questions are atypical in this
respect, in that the questioner already knows the answer), (b) a desire for the
lack to be made good, (c) a desire for a response from the addressee that will
fulfil (b), and (d) a belief that the addressee can supply such a response. In
English, there are two basic types of question: ‘Yes-No questions’ and ‘Wh-
questions’. Yes-No questions effectively present a proposition and ask
whether it is true or not. Thus Is John here? presents the proposition ‘Pete is
here’ and expects the answer Yes if it is true and No if it is false. In contrast,
Wh-questions present a proposition with a term missing, and request an
answer which fills in the gap to form a true proposition. Hence Where is
John? presents the skeleton proposition ‘John is –’; an answer in the kitchen
means that the proposition ‘John is in the kitchen’ is true. (For questions
functioning as requests for action, see indirect speech act.)

IRONY A species of figurative language, in which the intended meaning of

an expression is usually some kind of opposite of the literal meaning, as, for
instance, when someone says You’ve been a great help! to a person whose
actions or words have just precipitated a disaster. The literal meaning of an
ironic expression typically echoes the words or assumed opinions of
someone else, and is intended to mock or ridicule. Irony arises from the
speaker’s flouting the maxim of quality.

L
LANGUE VS PAROLE This is a distinction first drawn by Saussure, one of the
founding fathers of modern linguistics, which had a profound influence on
the development of the subject. It is basically a distinction between a
language as an abstract system, which is the true object of the study of
linguistics, and the use made of that system, in the sense of what speakers of

144
the language actually say on particular occasions, which, for one reason or
another, may not conform precisely to the underlying system. See also a
related distinction between competence and performance.

LATCHED TURNS see under conversational analysis

LATENCY A latent element is one which must be recovered from context if

an expression is to be understood properly. A good example is the direct
object of the verb watch in, for example, Pete was watching. This is
acceptable only if what Pete was watching is known or can be recovered
from context. It is not true of all verbs that if their direct object is not
mentioned it becomes latent. Although logically speaking Pete must be
reading something, the direct object in Pete was reading is not latent. Notice
that in Pete was watching and so was Liz both must be watching the same
thing, but in Pete was reading and so was Liz there is no implication that
they were reading the same thing. (Latency is also known as ‘zero anaphora’
or ‘definite deletion’.)

M
MAXIMS OF CONVERSATION These spell out in greater detail the
consequences of the Co-operative Principle. They were proposed Grice:
Maxim of Quantity, Maxim of Quality, Maxim of Relation, and Maxim of
Manner. They are rules of conversational conduct that people do their best to
follow, and that they expect their conversational partners to follow. They
have a rational basis, and are not matters of pure convention.

1. Maxim of Quantity: this deals with the amount of information presented. It

comprises two submaxims:
(a) Make your contribution as informative as is required for the current
purposes of the talk exchange in which you are engaged.
(b) Do not make your contribution more informative than is required.

2. Maxim of Quality: this has to do with truth-telling:

(a) Do not say what you believe to be false.
(b) Do not say that for which you lack adequate evidence.

There are many occasions when this maxim seems not to be followed (see,
for instance, under politeness), but it is arguable that it represents a valid
default position, that is to say we do not depart from it without good reason.

145
3. Maxim of Relation: Be relevant.
The truth of a statement is no guarantee that it is an appropriate contribution
to a conversation: it must also connect suitably with the rest of the
conversation. (According to some scholars, a suitable version of this maxim
renders the others unnecessary: see under Relevance Theory.) These three
maxims can be combined into one: make the strongest statement that can
relevantly be made that is justifiable by your evidence.

4. Maxim of Manner: This comprises four submaxims:

(a) Avoid obscurity.
(b) Avoid ambiguity (this means ‘ambiguity in context’).
(c) Avoid unnecessary prolixity (i.e. excessive wordiness).
(d) Be orderly (this means that if time relations are not explicitly expressed,
events should be related in the order in which they occur).

For a discussion of how the maxims help to account for conversational

implicatures, see under standard implicatures and flouting the maxims.

MINIMAL RESPONSES see under conversational analysis

MODESTY MAXIM see under Approbation and Modesty Maxims

N
NATURAL VS CONVENTIONAL SIGNS Conventional signs are those which
are established for communicative use in some community and which have
to be specially learned (and often taught). Linguistic signs are obvious
examples; so are traffic signs and the like. There are two interpretations of
‘natural’ in respect to signs. According to one interpretation, natural signs
are based on causal connections in the natural world. In this sense we say
that smoke is a sign of fire and dark clouds are a sign of rain. According to
another interpretation, natural signs are signs produced by communicating
beings that do not have to be learned but are instinctive, like animal cries and
human signs such as smiling, weeping, and gasping.

O
OVERHEARER see under speech event participant

OVERLAPPING TURNS see under conversational analysis

146
P
PARTICULARISED CONVERSATIONAL IMPLICATURES see generalized vs
particularised conversational implicatures

PERFORMANCE see under competence vs performance

PERFORMATIVE VERB A performative verb is one which designates a

specific speech act and which, if used appropriately, counts as the
performance of the speech act. For instance, saying I promise to be careful
counts as a promise to engage in a particular course of action. Other
examples of performative verbs are: ask, beg, command, congratulate, deny,
deplore, declare, implore, warn, etc. In general, only certain forms of the
verb count as performing the speech act, mainly first person simple present
active and third person present passive. For instance, I congratulate you on
your promotion and Passengers are warned not to lean out of the windows
count as congratulation and warning respectively. But although I promised to
do it and He will promise to do it refer to the performance of acts of
promising, they do not count as promises, rather they are descriptions of
states of affairs. A performative verb in a performative use can typically be
accompanied by hereby: I hereby promise to pay the sum of one hundred
pounds (but ?I hereby promised to do it).

PERSON DEIXIS Person deixis designate the basic roles in a speech event,
namely the speaker (‘first person’), the person(s) spoken to (‘second
person’), and the person or persons who are neither speaker nor addressee
(‘third person’). Person deixis include pronouns (I, you, him; mine, yours,
hers; myself, yourself, herself), possessive adjectives (my, your, her), and
verb inflections (Latin amo, amas, amat, ‘I love, you love, he or she loves’).
Personal pronouns can have singular and plural forms. A plural form may
apply even if only one referent is designated, provided that the referent can
be taken to represent a group. For instance, the first person plural we is
normally produced by a single speaker who represents a group. Some
languages have a different first person plural form according to whether the
represented group includes both the speaker and the addressee (‘inclusive’
form) or the speaker and others, but not the addressee (‘exclusive’ form).
First person plural pronouns refer directly to a plurality of speakers only in
the case of choral speaking.

147
POLITENESS Insofar as linguistic behaviour is concerned, politeness is a
matter of minimizing the negative effects of what one says on the feelings of
others and maximizing the positive effects (known as ‘negative politeness’
and ‘positive politeness’ respectively). Politeness can also be either speaker-
oriented or hearer-oriented. Speaker oriented politeness involves not saying
things about oneself that would place one in a favourable position relative to
the hearer; boasting, for instance, is for this reason inherently impolite.
Utterances which directly involve the hearer fall into the domain of hearer-
oriented politeness. Leech proposes a general ‘Politeness Principle’:
Minimize the expression of impolite beliefs. This principle both constrains
and is constrained by the Cooperative Principle. Clearly, there are
occasions when it is more important to convey relevant true information
even if it has negative effects on the hearer. Like the cooperative principle,
the politeness principle is expanded by means of a set of maxims (see the
entries for Tact and Generosity Maxims, Approbation and Modesty
Maxims, Agreement Maxim, Sympathy Maxim). There are, in addition,
three minor principles: Banter Principle: this allows us to be polite while
being superficially rude, as when one says to a good friend Look what the
cat’s brought in! The underlying message is ‘We are such good friends we
don’t need to be polite.’ Irony Principle: this allows us to be impolite while
being superficially polite, as in You should be very proud of yourself, said to
someone who has made a mess of something. Pollyanna Principle: this
enjoins us to avoid drawing attention to things ‘which are not mentioned in
polite company’. It is this principle which underlies the use and development
of euphemisms. Politeness also enters into ways of addressing people. Many
languages have a choice of pronouns for designating the addressee according
to the relationship between speaker and addressee and, to some extent, the
situation (these are sometimes called ‘T/V pronouns’). Examples are French
tu/vous, Italian tu/Lei, German du/Sie, and Turkish sen/siz. The exact
conventions for using these forms differ from language to language, but we
may take French as an example. We may first distinguish asymmetrical
usage from symmetrical usage. Asymmetrical usage is relatively rare in
modern French but it can still be observed in, for instance, a school setting,
where pupils will address a teacher as vous and the teacher will address a
pupil as tu. The distinction marks a difference of social status. In the more
common symmetrical use, vous (sometimes called the ‘polite form’) marks
either psychological distance (respectful or otherwise) or a formal situation
(or both), while tu (the ‘familiar form’) indicates intimacy/familiarity or an
informal situation (or both). (The rules are quite subtle – the foregoing is a
first approximation.) English does not use T/V pronouns, but, as in many

148
languages, politeness enters into the choice of forms of address, such as Pete,
Smith, Mr Smith, Professor Smith, Sir Peter, and so on.

PRAGMATICS Most linguists draw a distinction within the study of meaning

between semantics and pragmatics, but there are several ways of drawing the
distinction. The main ones are as follows:
1. Semantics deals with truth conditional aspects of meaning; pragmatics
deals with non-truth conditional aspects.
2. Semantics deals with context-independent aspects of meaning; pragmatics
deals with aspects where context must be taken into account. Context is
understood here in a broad sense that includes previous utterances (discourse
context), participants in the speech event, their interrelations, knowledge,
and goals, and the social and physical setting of the speech event.
3. Semantics deals with conventional aspects of meaning, that is, where there
is an established connection between form and meaning. Pragmatics deals
with aspects of meaning that are not ‘looked up’ but which are ‘worked out’
on particular occasions of use.
4. Semantics is concerned with the description of meanings; pragmatics deals
with the uses made of those meanings. This is sometimes expressed by
saying that semantics takes a formal approach and pragmatics a functional
approach. These different definitions have consequences in terms of what is
included in pragmatics. In general, linguists agree that the following belong
to pragmatics: politeness phenomena, reference and deixis, implicatures, and
speech acts.

PRESUPPOSITION A presupposition is a proposition whose truth is taken for

granted by the producer of an utterance and which must be known and taken
account of for the utterance to make sense to an interpreter. Take the case of
Pete has stopped smoking. Someone using this sentence takes it for granted
that Pete was previously a smoker, although this is not explicitly stated. And
the presumption that Pete had been a smoker is necessary for the sentence to
make sense to a hearer, even if that fact was not previously known.
Presupposition is not the same as entailment. Take a genuine case of
entailment such as that between ‘Pete killed the beetle’ and ‘The beetle
died’. First, if we negate the entailing sentence, the entailment fails: ‘Pete did
not kill the beetle’ entails neither ‘The beetle died’ nor ‘The beetle did not
die’. However, ‘Pete has not stopped smoking’ carries the same presumption
as the affirmative version, as does ‘Has Pete stopped smoking?’. Second, an
entailment cannot be denied without contradiction: ‘Pete killed the beetle,
but it did not die’ is a contradiction. A presupposition, on the other hand, can
be denied (although it needs a special intonation): ‘Pete HASN’T stopped
149
smoking because he never DID smoke’. Presuppositions are ubiquitous. The
following are some examples:

Utterance: The flying saucer landed right here.

Presupposition: A flying saucer landed. Flying saucers exist.
Utterance: Liz regrets / does not regret selling the house.
Presupposition: Liz sold the house.
Utterance: Liz plays / does not play the piano brilliantly.
Presupposition: Liz plays the piano.

There has been some dispute as to whether presuppositions are a semantic or

a pragmatic phenomenon. If they are inherent properties of certain linguistic
expressions then they are semantic in nature; if, on the other hand, they are a
property of utterance(s)-in-context then they are pragmatic. Currently, the
weight of scholarly opinion is in favour of a pragmatic analysis.

PROJECTED DEIXIS This is when deictic expressions are used in their usual
way, but the deictic centre is not the speaker but some other participant in the
speech event, most commonly the addressee. For instance, the verb come has
deictic properties in that its basic use is to denote movement towards the
speaker, as in Come here! However, in Shall I come and see you? the
movement in question is towards the addressee.

PROPOSITION The simplest type of proposition consists of an argument (an

entity about which something is ‘said’) and a predicate (what is ‘said’ about
the argument). In the proposition ‘John is tall’, ‘John’ is the argument and
‘(is) tall’ is the predicate. Some predicates need more than one argument to
form a complete proposition: ‘like’, for instance, requires two (‘John likes
Liz’) and ‘give’ requires three (‘John gave Liz a present’). Predicates may be
described as one-place, two-place or three-place according to the number of
arguments they take. A proposition has a truth value, that is, it is either true
or false. It is not a linguistic expression, nor is it tied to any particular
linguistic expression. The same proposition may be expressed by different
linguistic means, and a given sentence may be used to express different
propositions on different occasions. So, for instance, John is here, My
brother is here, Liz’s boyfriend is here can very well express the same
proposition, provided the same individual and place are designated. The
proposition is what is asserted in a statement, what is questioned in a
question (Is John here?) and what is denied in a negation (John is not here).

150
R
RATIFIED PARTICIPANT see under speech event participant

RELEVANCE THEORY This is a theory of implicature proposed by Sperber

and Wilson which incorporates some aspects of Grice’s approach (see under
Co-operative Principle, conventional implicature, conversational
implicature, maxims of conversation), but takes it in a new direction. As in
Grice’s theory, the notion of (conversational) implicature is central.
Relevance theorists, however, regard Grice’s maxims as arbitrary in number
and deficient in explanatory power. They argue that one of Grice’s maxims,
the Maxim of Relation (‘Be relevant’), when appropriately developed, can do
the work of all the others, and they accordingly raise it to the status of an
overall principle to supplant Grice’s Co-operative Principle. This principle,
called the ‘principle of relevance’. The principle of relevance has two parts,
the ‘cognitive principle of relevance’ and the ‘communicative principle of
relevance’. According to the cognitive principle of relevance, the human
cognitive system interprets an utterance in such a way as to maximize its
relevance. This means achieving the greatest number of ‘contextual effects’
(changes in the information stored in the cognitive system) while minimizing
the cost in terms of processing effort. According to the communicative
principle of relevance, a bona fide communicator, simply by producing an
utterance, implicates his or her belief that it is optimally relevant. The
speaker is the more active participant in two-way communication; the hearer
is more passive. The speaker’s task is to produce an utterance which will
enable the hearer to construe the intended message by following the standard
procedure. To do this, the speaker must take account of the hearer’s
knowledge and of how accessible its different parts are. The standard
procedure for the hearer is to test possible interpretations in order of
processing effort required, beginning with the most accessible, until one is
found whose contextual effects justify the processing effort expended. If the
speaker has done his or her job properly, the first such interpretation will be
the correct one. According to proponents of the relevance theory, two phases
of interpreting an utterance can be distinguished. The first is the extraction of
the ‘explicature’. The explicature is basically what is explicitly encoded in
the linguistic form of the utterance, together with certain elaborations that are
needed to make it logically complete and unambiguous. Consider, for
instance, B’s reply in the following:

A: When did you arrive?

151
B: Yesterday.

In order to function as explicature this needs at least two sorts of elaboration.

First, it must be expanded to I arrived yesterday. Second, yesterday
designates a particular day, and this must be specified in absolute terms (not
just as ‘the day before the day including the time of utterance’). The second
phase is the combination of explicature with context to produce
implicatures.

REPAIR see under conversational analysis

SCALAR IMPLICATURES see under generalised vs particularized

conversational implicatures

SENTENCE MEANING VS UTTERANCE MEANING Sentence meaning is the

meaning a sentence has by virtue of the words it contains and their
grammatical arrangement, and which is not dependent on context. Utterance
meaning is the meaning a sentence carries when it is used in a particular
context, with referents assigned to all referring expressions, and taking into
account any conversational implicatures, inferences that arise due to the
Gricean maxims of converation.

SOCIAL DEIXIS Social deixis concerns the use of expressions whose function
is to indicate the position of the referent on the scales of social status and
intimacy relative to the speaker. A prototypical example is the use of the so-
called T/V pronouns. For more details, see under politeness.

SPATIAL DEIXIs Spatial deixis concerns the use of expression to indicate

location in space relative to the speaker. The most basic spatial deictic
expressions are the adverbs here and there. These can be glossed ‘place near
to the speaker’ and ‘place not near to the speaker’. Modern English has only
two terms, but older forms of English and some dialects have a third term
yonder, yon, which indicates a greater distance than there and can be glossed
‘far from the speaker’. Notice that the boundary between here and there is
heavily context-bound: here in this room, here in Manchester, here in
Britain, here on earth, and so on. The demonstratives this and that are
usually considered to be spatial deictic items in their basic use, although they
152
often have a more abstract meaning. For instance, this theory and that theory
do not locate the theory in literal space, but do indicate a more abstract
closeness and distance from the speaker. This and here are instances of
proximal deixis (showing closeness to the speaker), and that and there distal
deictic items (non-proximal to speaker, sometimes proximal to hearer). Some
scholars emphasize correspondences between proximal deictics and first
person deictics, and between distal and second and third person deictics.
Spatial deictics can indicate psychological or emotional, rather than strictly
spatial distance: This is great news, That theory is rubbish, I can’t stand that
man.

SPEECH ACTS These are acts which crucially involve the production of
language. The theory of speech acts was originally proposed by Austin who
distinguished three basic types: locutionary acts, illocutionary acts and
perlocutionary acts.
1. Locutionary act: the production of an utterance, with a particular sense and
reference.
2. Illocutionary act: an act performed by a speaker in saying something (with
an appropriate intention and in an appropriate context), rather than by virtue
of having produced a particular effect by saying something. For instance, if
someone says I order you to leave now they have performed the act of
ordering, simply by virtue of having uttered the words, whether or not the
addressee acts in the desired way.
3. Perlocutionary act: a speech act which depends on the production of a
specific effect. For instance, for the verbal act of persuasion to have
occurred, in Pete persuaded Liz to marry him, it is not enough for Pete to
have uttered certain words – what is essential is that a previously reluctant
addressee is caused to act in an appropriate way. Every illocutionary act has
a particular ‘illocutionary force’. This may be explicitly signalled by the use
of a performative verb such as beg, promise, command, suggest,
congratulate, or thank, or a particular grammatical form, as in Go away!,
Have you seen Pete?, or it may be implicit, in which case it must be inferred,
largely on the basis of contextual evidence. For instance, an utterance such as
You will never see me again may function, in different circumstances, as a
threat, a promise, a simple statement of fact, or a prediction. For a particular
illocutionary act to be successfully performed, it is typically the case that
certain contextual conditions need to be satisfied. These conditions are called
felicity conditions.

There are various ways of classifying illocutionary acts. The following is

proposed by Searle:
153
Assertives: these commit the speaker to the truth of what is said: assert, aver,
boast, claim, report.
Directives: these are aimed at getting someone to act in a certain way: beg,
implore, request, warn (to), recommend
(to), ask (to).
Commissives: these have the effect of committing the speaker to some action
in the future: promise, undertake,
offer, threaten.
Expressives: these express the speaker’s feelings or attitude: thank,
congratulate, forgive.
Declarations: these are said to produce a change of some sort in the world:
resign, sack, appoint, name, christen, sentence (in court), bid (at auction).

SPEECH EVENT PARTICIPANTS The following may be recognized:

Speaker: the person producing an utterance.
Addressee: the person to whom an utterance is directed in a speech event.
Ratified participant: a member of a conversational group not directly
addressed, who is expected to attend to what is said.
Overhearer: this may be a ‘bystander’ (who may understand and join in
conversation without causing offence) or an ‘eavesdropper’ (who is not
supposed to hear).

STANDARD (CONVERSATIONAL) IMPLICATURES These are conversational

implicatures which can be inferred from an utterance, provided we assume
that the speaker is following the maxims of conversation. Consider this
scenario:

A: Can I speak to Jane?

B: She’s in the shower.

A will infer from B’s answer that this is an inconvenient time to speak to
Jane, although B does not explicitly say so. A’s inference is due to B’s
assumption A is obeying the Maxim of Relation, and that the answer is
therefore relevant. The most obvious relevance is that calling Jane to the
phone would cause inconvenience. Or take the following example:

A: Did Pete post the letter and pay the newspaper bill?
B: He posted the letter.

B’s answer implicates either that Pete did not pay the newspaper bill, or that
B does not know whether he did or not (more context would be needed to
154
choose between these). B appears not to be following the Maxim of
Quantity, in that the utterance does not provide the required amount of
information. However, assuming that B is nonetheless obeying the Co-
operative Principle, we can infer that there is good reason for the poverty of
information. One possibility is adherence to the Maxim of Quality: B is
giving as much information as he or she has evidence for. A might therefore
infer that B does not know whether or not Pete paid the paper bill.

STRENGTH OF IMPLICATURES Implicatures can vary in strength according

to the degree of the speaker’s commitment to them, how easily the speaker
could deny intending them, and how easily the hearer can avoid drawing
them. For instance, B’s answer in (1) carries a strong implicature that he or
she has exactly four children:

1. A: How many children do you have?

B: Three.
However, the implicature of exactitude is much weaker in (2):

2. A: You need to have four children to qualify for this allowance.

B: I have three children.

And in (3), the implicature of exactness (‘four minutes and no less than four
minutes’) is probably absent altogether:
3. A: You have to be able to do the 1000 metres in three minutes to enter.
B: I can do it in three minutes.

SYMBOLIC DEIXIS This refers to the use of a deictic expression where close
monitoring of the situation by the hearer is not required because the relations
between the speaker and the things referred to are relatively stable and do not
change over the course of a conversation or discourse: I’ve lived here all my
life, Nobody cares these days, Those bastards are just out to get you.

SYMPATHY MAXIM One of the Maxims of Politeness proposed by Leech. Its

specifications are as follows:

Maximise sympathy (expression of positive feelings) towards hearer.

Minimise antipathy (expression of negative feelings) towards hearer.

On the basis of this principle, congratulations, condolences, and

commiserations are inherently polite. If negative feelings must be expressed,

155
they should be played down. Consider the following utterances delivered
after an accident that someone experienced due to carelessness:

Serves you right! (not polite)

You really need to be more careful (more polite)
These things happen (even more polite)

T
Tact and Generosity Maxims These are members of the set of Maxims of
Politeness proposed by Leech. The former is oriented towards the hearer,
while the latter is oriented towards the speaker. The following is a slight
modification of Leech’s formulation:

Tact Maxim: Minimise cost to the hearer. Maximise benefit to the hearer.
Generosity Maxim: Minimise benefit to self. Maximise cost to self.

Both of these maxims apply particularly to speech acts which are directives
or commissives. Both of them relate to the idea of a ‘cost-benefit scale’.
Actions (requested or offered) can be ranked according to the cost or benefit
(physical, psychological, financial, or whatever) to the person carrying them
out. For instance, digging the garden probably represents a greater (physical)
cost than mowing the lawn, which in turn is greater than picking some
flowers. The cost-benefit scale operates in conjunction with a ‘scale of
indirectness’, which applies to the way a command, request, offer, (and so
on) is formulated. In the case of directives, for instance, the most direct form
is the imperative: Wash the dishes. Progressively more indirect are: I want
you to wash the dishes; Can you wash the dishes?; Could you wash the
dishes?; I wonder if you would mind washing the dishes. The general
principle for both commissives and directives is that, for politeness, anything
which involves cost to the hearer or benefit to the speaker should be
‘softened’ by being expressed indirectly, and the greater the cost the more
the indirectness required. Conversely, anything that involves benefit to the
hearer or cost to the speaker should be expressed directly. Thus, Could you
wash the dishes? and I’ll do the dishes are more polite than Wash the dishes
and Maybe I should wash the dishes respectively.

TEMPORAL DEIXis Temporal deictic expressions indicate the timing of an

event relative to the coding time. The only ‘pure’ English temporal deixis
(those which give no other information) are now, which designates a time

156
period overlapping with the time of speaking, and then, which basically
means ‘not now’, and can point either into the future or the past: I was much
younger then; You’ll be somewhat older by then. Many temporal deictic
expressions give additional information, such as tomorrow (‘the day after the
day which includes the time of speaking’) and last year (‘the (calendar) year
previous to the one which includes the time of speaking’). The system of
verb tenses is also deictic (I washed the dishes, I am washing the dishes, I
will wash the dishes) since it relates the utterance to the coding time. It is
useful, when speaking about tenses, to distinguish three points in time: the
time at which the event occurred (ET), the time at which the utterance was
produced (UT), and the reference time (RT). In the so-called primary tenses,
past, present and future, UT and RT are the same. There are also secondary
or compound tenses in which UT and RT are different. In the case of the past
preterite, RT is in the past relative to UT: Liz had already left when I arrived
(Liz’s leaving preceded my arrival, which preceded the time of speaking). In
the ‘future perfect tense’, RT is in the future relative to UT: By the time I
arrive, Liz will have left. It is also possible to have a ‘future-in-the-past
tense’: Liz was about to leave when I arrived. Some languages have different
past tense forms according to how far back in time the denoted event
occurred.

TURN (CONVERSATIONAL) see under conversational analysis

UPTAKE An uptake is generally said to be an essential feature of a felicitous

speech act. It indicates the hearer’s acceptance of the validity of the speech
act. An example could be You’re on in the case of bets. Consider the case of
a promise such as I’ll do it tomorrow. Suppose the speaker sincerely intends
to put himself under an obligation to carry out the act. But suppose also the
hearer, for one reason or another, refuses to accept that an obligation has
been entered into. In such cases the debatable point is whether in such cases
the promise is still a fully-fledged one.

157
 REFERENCES

Abbott, Barbara. 2000. Presuppositions as nonassertions. Journal of

Pragmatics, 32(10): 1419–1437.
Abbott, Barbara. 2006. Where have some of the presuppositions gone? In
Betty Birner and Gregory Ward (eds.), Drawing the boundaries of
meaning: Neo-Gricean studies in pragmatics and semantics in
honor of Laurence R. Horn, Philadelphia: John Benjamins.
Abbott, Barbara. 2008. Issues in the semantics and pragmatics of definite
descriptions in English. In J.K. Gundel and N. Hedberg (eds.)
Reference: Interdisciplinary Perspectives, Oxford: Oxford University
Press.
Abusch, Dorit. 2002. Lexical alternatives as a source of pragmatic
presuppositions. In Jackson, Brendan (ed.), Proceedings of SALT
XII, Ithaca, NY: CLC Publications, 1–20.
Abusch, Dorit and Rooth, Mats. 2000. Empty-domain effects for
presuppositional and non-presuppositional determiners. In Hans
Kamp and Barbara Partee (eds), Context Dependency in the Analysis
of Linguistic Meaning.
Agha, Asif. 1994. Honorification. Annual Review of Anthropology 23:277-
302.
Allan, Keith. 1998. Meaning and speech acts. Online publication,
http://www.arts.monash.edu.au/ling/speech_acts_allan.shtml.
Anderson, Stephan and Keenan, Ed. 1985. Deixis. In T. Shopen (ed.),
Language Typology and Syntactic Description, Vol 3: Grammatical
Categories and the Lexicon, Cambridge: Cambridge University
Press, pp. 259-308.
Atkinson J.M. and John Heritage (eds.). 1984. Structures of Social Action:
Studies in Conversation Analysis, Cambridge: Cambridge University
Press, 299–345.
Atlas, Jay. 1976. On the Semantics of Presupposition and Negation: an Essay
in Philosophical Logic and the Foundations of Linguistics, Ph.D.
thesis, Princeton University.
Atlas, Jay.1977. Negation, ambiguity and presupposition. Linguistics and
Philosophy, 1: 321–336.
Atlas, Jay. 1979. How linguistics matters to philosophy: Presupposition,
truth, and meaning. In Oh, C. K. and Dineen, D. A. (eds.),
Presupposition, New York: Academic Press, vol. 11 of Syntax and
Semantics, 265–281.
158
Atlas, Jay and Levinson, Stephen. 1981. It-clefts, informativeness and
logical form: Radical pragmatics. In Cole, Peter (ed.), Radical
Pragmatics, Academic Press, 1–61.
Austin, John L. 1962. How to do things with words, London: Oxford
University Press.
Austin, John. 1975. How to Do Things with Words, (2nd edn.), Claredon
Press.
Bach, Kent. 1999. The Myth of Conventional Implicature. Linguistics and
Philosophy, 22: 327-366.
Bach, Kent, and Robert M. Harnish. 1979. Linguistic communication and
speech acts. Cambridge, MA: MIT Press.
Barwise, John and Perry, John. 1983. Situations and Attitudes. Cambridge,
MA: MIT Press.
Beattie, G.W., and Barnard, P.J. 1979. The temporal structure of natural
telephone conversations (directory enquiry calls). Linguistics, 17,
213–229.
Beaver, David. 1992. The kinematics of presupposition. In Dekker, P. and
Stockhof, M. (eds.), Proceedings of the Eighth Amsterdam
Colloquium, University of Amsterdam: ILLC, 17–36.
Beaver, David. 1997. Presupposition. In van Benthem, J. and ter Meulen, A.
(eds.), The Handbook of Logic and Language, Amsterdam: Elsevier,
939–1008.
Beaver, David. 2001. Presupposition and Assertion. Dynamic Semantics,
Stanford, CA: CSLI Publications. [Available online]
Beaver, David. 2008. As brief as possible (but no briefer). Theoretical
Linguistics, 34: 213–228.
Beaver, David and Condoravdi, Cleo. 2003. A uniform analysis of “before“
and “after“”, in Proceedings of Semantics and Linguistic Theory
XIII, Cornell, Ithaca: CLC Publications, 37–54.
Beaver, David and Krahmer, Emiel. 2001. A partial account of
presupposition projection. Journal of Logic, Language and
Information, 10: 147–182.
Beaver, David and Zeevat, Henk. 2007. Accommodation. In Ramchand, G.
and Reiss, C. (eds.), The Oxford Handbook of Linguistic Interfaces,
Oxford University Press, 533–538.
Blamey, Steven. 1989. Partiality. In Gabbay, D. and Guenther, F. (eds.),
The Handbook of Philosophical Logic, Dordrecht: Reidel, vol. IV,
553–616.
Blutner, Reinhard. 2000. Some aspects of optimality in natural language
interpretation. Journal of Semantics, 17(3): 189–216.

159
Bockgard, Gustav 2007. Syntax och prosodi vid turbytesplatser: Till
beskrivningen av svenskans turtagning. In E.Engdahl and A.-
M.Londen (Eds.), Interaktion och kontext: Nio studier av svenska
samtal, Lund: Studentlitteratur, pp. 139–183.
Bochvar, Dmitrii Anatoljevich. 1939. Ob odnom trehznachom iscislenii i
ego primeneii k analizu paradoksov klassicskogo rassirennogo
funkcional 'nogo iscislenija. Matematiciskij Sbornik, 4: 87–112.
English Translation: History and Philosophy of Logic 2, 1981, “On a
Three-valued Calculus and Its Applications to the Analysis of the
Paradoxes of the Classical Extended Functional Calculus”.
Böer, Steven E. and Lycan, William G. 1976. The myth of semantic
presupposition. In Zwicky, A. (ed.), Papers in non-phonology,
Columbus, Ohio: Department of Linguistics, Ohio State University,
vol. 21 of Working Papers in Linguistics, 1–90.
Brady, P.T. 1968. A statistical analysis of on–off patterns in 16
conversations. The Bell System Technical Journal, 47, 73–91.
Bridge, Derek, 1991, Computing Presuppositions in an Incremental Natural
Language Processing System, Ph.D. thesis, University of Cambridge
Computer Laboratory. Available as Technical Report No. 237.
Brown, Roger W. and Gilman, Albert. 1960. The pronouns of power and
solidarity. In T. Sebeok (ed.), Style in Language. Cambridge, MA:
MIT Press, pp. 253-276.
Brown, Penelope and Levinson, Stephen. 1987. Politeness: Some Universals
in Language Usage. Cambridge: Cambridge University Press.
Bull, M. 1996. An analysis of between-speaker intervals. In Proceedings of
the Edinburgh Linguistics Department Conference’96, Edinburgh,
(pp.18–27).
Bull, M., and Aylett, M. 1998. An analysis of the timing of turn-taking in a
corpus of goal-oriented dialogue. In Proceedings of the fifth
international conference on spoken language processing (ICSLP‘98),
Sydney, Australia, (Vol.4,pp.1175–1178).
Bühler, Karl 1982 [1934]. The deictic field of language and deictic words.
R.J. Jarvella & W. Klein (eds.), Speech, place and action. New York:
Wiley, pp. 9-30.
Campione, E., and Veronis, J. 2002. A large-scale multilingual study of
silent pause duration. In Proceedings of the first international
conference on speech prosody (Speech prosody 2002), Aix-en-
Provence, France, pp.199–202.
Cappella, J.N. 1979. Talk–silence sequences in informal conversations In
Human Communication Research, 6, 130–145.

160
Carston, Robyn. 2002. Relevance Theory and the saying/implicating
distinction. In L. Horn and G. Ward (eds.) 2002. Handbook of
Pragmatics. Blackwell.
Chierchia, Gennaro. 1995. Dynamics of Meaning, Chicago: University of
Chicago Press.
Chomsky, Noam. 1965. Aspects of the theory of syntax, Cambridge MA:
MIT Press.
Clark, Herbert H. and Marshall, C. 1981. Definite reference and mutual
knowledge. In Joshi, A. and Webber, B. L. and Sag, I. (eds.),
Elements of Discourse Understanding, Cambridge: Cambridge
University Press.
Coates, Jennifer. 1993. Women, Men and Language, London and New
York: Longman.
Coates, Jennifer. 1995. Language, gender and career. In S Mills (ed.),
Language and gender: interdisciplinary perspectives, London and
New York: Longman, pp. 13-30.
Cole, Peter and Jerry L. Morgan. (eds.) 1975. Syntax and semantics, volume
3: Speech acts. New York: Academic Press.
Comrie, Bernard 1985. Tense. Cambridge: Cambridge University Press.
Comrie, Bernard 1989. Aspect. Cambridge: Cambridge University Press.
Cooke, Joseph. 1968. Pronominal reference in Thai, Burmese and
Vietnamese. Berkeley, CA: University of California Press.
Cooper, Robin. 1983, Quantification and Syntactic Theory, Dordrecht:
Reidel. de Jong, Franciska and Verkuyl, Henk, 1985, “Generalized
quantifiers: the properness of their strength”, in ter Meulen, A. and
van Benthem, J. (eds.), Generalized Quantifiers: Theory and
Applications, Dordrecht: Foris, vol. 4 of GRASS, 21–43.
Cresswell, Max. 1973. Logics and languages. London: Methuen.
Cysouw, Michael. 2001. The paradigmatic structure of person marking. PhD
dissertation, KU Nijmegen.
Davis, S. 1991. Pragmatics. A Reader, Oxford: Oxford University Press.
Delin, Judy, 1995. Presupposition and shared knowledge in it-clefts.
Language and Cognitive Processes, 10(2): 97–120.
van Eijck, Jan, 1993. The dynamics of description. Journal of
Semantics, 10: 239–267.
de Ruiter, J. P., Mitterer, H., and Enfield, N. J. 2006. Predicting the end of a
speaker’s turn: A cognitive corner stone of conversation. Language,
82, 515–535.
Diessel, Holger. 1999. Demonstratives: Form, function and
grammaticalization. Amsterdam: Benjamins.

161
Dil, A. S. 1978 (ed.), Language, culture and history: essays by Mary R.
Haas, Stanford, Calif.: Stanford University Press.
Dittman, A. and Llewellyn, G. 1967. The phonemic clause as a unit of
speech decoding. Journal of Personality and Social Psychology, 6,
341-349.
Dixon, Robert. 1980. The languages of Australia. Cambridge: Cambridge
University Press.
Drew, Paul. 1994. Conversation analysis. The Encyclopaedia of Language
and Linguistics, Two volumes. Pergamon Press and Aberdeen
University Press, 2: 749–54.
Duncan, Starkey, Jr. 1972. Some signals and rules for taking speaking turns
in conversations. Journal of Personality and Social Psychology, 23,
283–292.
Duncan, Starkey Jr. and Fiske, Donald. 1977. Face-to-face interaction:
Research, methods and theory, Hillsdale N.Y.: Lawrence Erlbaum
Associates.
Duranti, Alessandro. 1988. Ethnography of speaking: toward a linguistics of
the praxis. In F. Newmeyer (ed.) Linguistics: the Cambridge Survey.
Vol. IV Language: The Socio-cultural context, New York: Cambridge
University Press, pp.210-228.
Eckert, Penelope. 1990. Cooperative competition in adolescent “girl talk”.
Discourse Processes, Vol. 13, No. 1, pp. 91-122
Edelski, C. 1993. ‘Who’s got the floor?’ in D. Tannen (ed.) Gender and
Conversational Interaction, New York Oxford: Oxford University
Press, pp. 189-230
van Eijck, Jan, 1994, Presupposition failure: a comedy of errors. Formal
Aspects of Computing, 6(6): 766–787.
van Eijck, Jan, 1995, “Presuppositions and dynamic logic”, in Papers from
the Second CSLI Workshop on Logic, Language and Computation,
Makoto Kanazawa, Christopher Piñon, and Henriette de Swart (eds.),
Stanford: CSLI Publications.
Enç, Murvet. 1981. Tense without scope: An analysis of nouns as indexicals.
PhD dissertation, University of Wisconsin-Madison.
Errington, Joseph. 1988. Structure and style in Javanese: a semiotic view of
linguistic etiquette. Philadelphia: University of Pennsylvania Press.
Fauconnier, Gilles, 1985, Mental Spaces: Aspects of Meaning Construction
in Natural Language, Cambridge, Massachusetts: MIT Press.
Fillmore, Charles. 1971. Verbs of judging: An exercise in semantic
description. In Fillmore, C. and Langendoen, D. T. (eds.), Studies in
Linguistic Semantics, New York: Holt, Reinhardt and Winston, 272–
289.
162
Fillmore, Charles. 1971/1975 Santa Cruz Lectures on Deixis. Indiana
University Linguistics Club, Bloomington.
Fillmore, Charles. 1997. Lectures on Deixis. Stanford, CA: CSLI
Publications.
Fishman, Pamela. 1997. Interaction: The Work Women Do. In N. Coupland
and A. Jaworski (eds.), Sociolinguistics. A Reader and Coursebook,
London: Macmillan, pp. 416-429.
Fraassen, Bas C. van, 1968, Presupposition, implication, and self- reference.
The Journal of Philosophy, 65(5): 136–152.
Frege, Gottlob. 1892. Uber Sinn und Bedeutung. Zeitschrift fur
Philosophie und philosophische Kritik, C: 25–50. English
Translation: “On Sense and Meaning”, in McGuinness, B. (ed),
Frege: collected works, Oxford: Basil Blackwell, 157–177.
Frege, Gottlob. 1918/1994. The Thought: A Logical Inquiry. In R. Harnish
(ed.), Basic Topics in the Philosophy of Language, Prentice-Hall,
Englewood Cliffs, N.J., pp. 517-535.
Gamut, L. T. F. 1991. Logic, Language and Meaning, vol. 1, University of
Chicago Press.
Garfinkel, Harold. 1967. Studies in Ethnomethodology. Englewood Cliffs,
NJ: Prentice Hall.
Gawron, Jean Mark. 1995. Quantification, quantificational domains and
dynamic logic. In Shalom Lappin (ed.), Handbook of Contemporary
Semantic Theory, Oxford: Blackwell, 247–267.
Gazdar, Gerald. 1979a. Pragmatics: Implicature, Presupposition and
Logical Form, New York: Academic Press.
Gazdar, Gerald. 1979b. A solution to the projection problem. In Oh, C.-K.
and Dineen, D. (eds.), Syntax and Semantics 11: Presupposition,
New York: Academic Press, 57-89.
George, Benjamin R. 2008. Presupposition Repairs: a Static, Trivalent
Approach to Predicting Projection, Master's thesis, UCLA.
Geurts, Bart. 1996. Local satisfaction guaranteed. Linguistics and
Philosophy, 19: 259–294.
Geurts, Bart. 1998. Presuppositions and anaphors in attitude contexts.
Linguistics and Philosophy, 21: 545–601.
Geurts, Bart. 1999a. Presuppositions and Pronouns, vol. 3 of CRiSPI,
Amsterdam: Elsevier. [Available online]
Geurts, Bart. 1999b. Specifics. In Geurts, B. and Krifka, M. and van der
Sandt, R. (eds.), Focus and presupposition in multi-speaker
discourse, Utrecht: ESSLLI 99, 99–129.
Geurts, Bart, 2000. Buoyancy and strength. Journal of Semantics, 17(4):
315–333.
163
Geurts, Bart. 2007. Existential import. In Comorovski, I. and von
Heusinger, K. (eds.), Existence: syntax and semantics, Dordrecht:
Springer, 253–271.
Geurts, Bart. 2009. Scalar implicatures and local pragmatics. Mind and
Language 24(1):51–79.
Geurts, Bart and van der Sandt, Rob. 1999. Domain restriction. In Bosch, P.
and van der Sandt, R. (eds.), Focus: Linguistic, Cognitive, and
Computational Perspectives, Cambridge: Cambridge University
Press, 268–292.
Geurts, Bart and van der Sandt, Rob. 2004. Interpreting focus. Theoretical
Linguistics, 30(1): 1–44.
Goffman, Erving 1981. Forms of talk. Philadelphia: University of
Pennsylvania Press.
Green, Georgia. 1996. Pragmatics and Natural Language Understanding,
(2nd edn), Lawrence Erlbaum Associates, Inc. Publishers.
Grice, H. Paul. 1957. Meaning. Philosophical Review 66:377-88.
Grice, H. Paul. 1967. Logic and Conversation. In Cole and Morgan, eds.,
41-58.
Grice, H. Paul. 1975. Logic and conversation. In Peter Cole and Jerry
Morgan, eds., Syntax and Semantics, volume 3: Speech Acts, 43–58.
New York: Academic Press.
Grice, H. Paul. 1989. Studies in the Way of Words, Cambridge,
Massachusetts: Harvard University Press.
Groenendijk, Jeroen and Stokhof, Martin, 1991, Dynamic Predicate Logic.
Linguistics and Philosophy, 14: 39–100.
Groenendijk, Theo M.V. Janssen and Martin Stokhof (eds.) 1984. Formal
Methods in the Study of Language, Vol.I, Mathemathische Centrum,
Amsterdam, reprinted in Jeroen Groenendijk, Theo M.V. Janssen and
Martin Stokhof (eds.), (1984), Truth, Interpretation and Information,
Foris, Dordrecht, pp.1-41.
Gunji, Takao. 1981. Towards a Computational Theory of Pragmatics -
Discourse, Presupposition and Implicature, Ph.D. thesis, Ohio State
University.
Gutt, E.-A. 1991. Translation and Relevance: Cognition and Context.
Oxford: Blackwell. (Second edition 2000. Manchester: St Jerome
Publishing.)
Hamblin, Charles L. 1970. Fallacies, London: Methuen & Co.
Haviland, John. 1979. How to talk to your brother-in-law in Guugu
Yimithirr. In T. Shopen (ed.), Languages and their speakers.
Cambridge, MA: Winthrop, pp. 161-239.

164
Hawkins, John A. 1978. Definiteness and Indefiniteness: A Study in
Reference and Grammaticality Prediction, Croom Helm.
Heim, Irene. 1982. On the Semantics of Definite and Indefinite Noun
Phrases, Ph.D. thesis, University of Massachusetts at Amherst.
Heim, Irene. 1983. On the projection problem for presuppositions. In
Barlow, M. and Flickinger, D. and Westcoat, M. (eds.), Second
Annual West Coast Conference on Formal Linguistics, Stanford
University, 114–126.
Heim, Irene. 1992. Presupposition Projection and the Semantics of Attitude
Verbs. Journal of Semantics, 9: 183–221.
Heinämäki, Orvokki. 1974. Semantics of English Temporal Connectives,
Ph.D. thesis, University of Indiana at Bloomington.
Heritage, John. 1984a. Garfinkel and Ethnomethodology. Cambridge: Polity
Press.
Heritage, John. 1984b. A change of state token and aspects of its sequential
placement. In J. Maxwell Atkinson and J. Heritage (eds) Structures of
Social Action: Studies in Conversation Analysis. Cambridge:
Cambridge University Press, 299–345.
Heritage, John. 2001. Goffman, Garfinkel and conversation analysis. In M.
Wetherell, S.Taylor and S.Yates (eds) Discourse Theory and
Practice. London: Sage, in association with the Open University, 47–
56.
Hirschberg, Julia. 1985. A Theory of Scalar Implicature. Ph.D. thesis,
University of Pennsylvania.
Horn, Laurence R. 1969. A presuppositional analysis of only and even. In
Papers from the Fifth Regional Meeting of the Chicago Linguistics
Society, University of Chicago, 98–107.
Horn, Laurence R. 1972. On the Semantic Properties of Logical Operators
in English. UCLA dissertation. Distributed by Indiana University
Linguistics Club, 1976.
Horn, Laurence R. 1985. Metalinguistic Negation and Pragmatic Ambiguity,
Language, 61: 121–174.
Horn, Laurence R. 1986. Presupposition, theme and variations. CLS 22:2,
168-192.
Horn, Laurence R. 1989. A Natural History of Negation. Chicago: University
of Chicago Press.
Horn, Laurence R. 1989. A Natural History of Negation, University of
Chicago Press. Reissued with new introduction (2001), David Hume
Series, CSLI Publications, Stanford.
Horn, Laurence R. and Ward, Gregory (ed.). 2004. The Handbook of
Pragmatics, Malden, MA: Blackwell Publishing.
165
Hornoiu, Diana 2016. Trends in Spoken Romanian. Corpus Design, Data
Analysis and Findings, Constanta: Ovidius University Press.
Horton, Diane. 1987. Incorporating agents’ beliefs in a model of
presupposition. Tech. Rep. CSRI-201, Computer Systems Research
Institute, University of Toronto.
Horton, Diane and Hirst, Graeme. 1988. Presuppositions as beliefs. In
Proceedings of the International Conference on Computational
Linguistics, COLING, 255–260.
Hutchby, Ian and Wooffitt, Robin. 1998. Conversation Analysis: Principles,
Practices and Applications. Oxford: Polity Press.
Jackendoff, Ray. 1972. Semantic Interpretation in Generative Grammar,
MIT Press.
Jaffe, J., and Feldstein, S. 1970. Rhythms of dialogue. New York, NY, USA:
Academic Press.
Jespersen, Otto 1968 [1922]. Language: its nature, development and origin.
London: Allen and Unwin.
Kamp, Hans. 1981. A Theory of Truth and Semantic Representation. Jeroen
Kaplan, David 1978. On the logic of demonstratives. Journal of
philosophical Logic 8:81-98.
Kaplan, David. 1989. Demonstratives. In J. Almog, J. Perry and H. Wettstein
(eds.), Themes from Kaplan, Oxford: Oxford University Press, pp.
481-564.
Kaplan, David. 1990. Words. The Aristotelian Society, Supplementary
Volume, LXIV.
Karttunen, Lauri. 1971a. Implicative verbs. Language, 47(2): 340–358.
Karttunen, Lauri. 1971b. Some Observations on Factivity. Papers in
Linguistics, 5: 55–69.
Karttunen, Lauri. 1973. Presuppositions of Compound Sentences.
Linguistic Inquiry, 4: 167–193.
Karttunen, Lauri. 1974. Presuppositions and Linguistic Context.
Theoretical Linguistics, 1: 181–194.
Karttunen, Lauri and Peters, Stanley. 1977. Requiem for presupposition. In
BLS3, Proceedings of the Third Annual Meeting of the Berkeley
Linguistic Society, Berkeley: California: 266–278.
Karttunen, Lauri and Peters, Stanley. 1979. Conventional Implicatures. In
Montague Grammar. Oh, C.-K. and Dineen, D. (eds.), Syntax and
Semantics 11: Presupposition, New York: Academic Press, 1–56.
Kasher, Asa (ed.) 1998. Pragmatics: Critical Concepts, Vol III: Indexicals
and reference. London: Routledge.
Kay, Paul, 1992. The inheritance of presuppositions. Linguistics and
Philosophy, 15: 333–381.
166
Kempson, Ruth. 1975. Presupposition and the Delimitation of Semantics,
Cambridge: Cambridge University Press.
Kendon, A. 1967. Some function of gaze direction in social interaction.
Acta Psychologica 26, pp 22-63
Kiparsky, Paul and Kiparsky, Carol. 1970. Fact. In Bierwisch, M. and
Heidolph, K. (eds.), Progress in Linguistics, The Hague: Mouton,
143–173.
Kleene, Stephen. 1952. Introduction to Metamathematics, Amsterdam:
North-Holland.
Krahmer, Emiel. 1994. Partiality and dynamics. In Dekker, P. and Stokhof,
M. (eds.), Proceedings of the Ninth Amsterdam Colloquium: Volume
2, University of Amsterdam: ILLC, 391–410.
Krahmer, Emiel. 1998. Presupposition and Anaphora, Stanford: CSLI
Publications.
Kripke, Saul. 2009. Presupposition and anaphora: Remarks on the
formulation of the projection problem. Linguistic Inquiry, 40: 367–
386.
Lambert, William and Tucker, Gordon. 1976. Tu, vous, usted. Rawley, MA:
Newbury Press.
Langendoen, D. Terence and Savin, Harris. 1971. The projection problem
for presuppositions. In Fillmore, C. and Langendoen, D. T. (eds.),
Studies in Linguistic Semantics, New York: Holt, Reinhardt and
Winston, 373–388.
Lappin, Shalom and Reinhart, Tanya. 1988. Presuppositions of quantifiers:
a processing account. Linguistics, 26: 1021–1037.
Leech, Geoffrey. 1983. Principles of Pragmatics, Longman.
Leiter, K. 1980. A Primer on Ethnomethodology, London: Oxford
University Press.
Lerner, Gene. 1991. On the syntax of sentences-in-progress. Language in
Society, 20, pp 441-458
Lerner, Gene. 1994. Responsive List Construction. Language and Social
Psychology 13, pp. 20-33.
Levinson, Stephen. 1979. Pragmatics and social deixis. Proceedings of the
Fifth Annual Meeting of the Berkeley Linguistic Society, pp. 206-23.
Levinson, Stephen C. 1983. Pragmatics. Cambridge: Cambridge University
Press.
Levinson, Stephen C. 2000. Presumptive Meanings: The Theory of
Generalized Conversational Implicature. Cambridge, MA: MIT
Press.
Levinson, Stephen C. 2003. Space in language and cognition: explorations
in cognitive diversity. Cambridge: Cambridge University Press.
167
Levinson, Stephen C. 2004. Deixis. Laurence R. Horn, and Gregory Ward,
(ed.). The Handbook of Pragmatics, Malden, MA: Blackwell
Publishing, pp. 97-121.
Lewis, David. 1979. Scorekeeping in a language game. Journal of
Philosophical Logic, 8: 339–359. Also in Bäuerle, R. and Egli, U.
and von Stechow, A. (eds.). 1979. Semantics from Different Points of
View, Springer, Berlin.
Lorenz, Sven. 1992. On the role of reasoning about change in the projection
of presuppositions. Report 234, IWBS, IBM Germany.
Lyons, John 1977. Semantics. 2 Vols. Cambridge: Cambridge University
Press.
Maclay, H., and Osgood, C.E. 1959. Hesitation phenomena in spontaneous
English speech. Word, 15, 19–44.
Maltz, Daniel. N. and Ruth A. Borker. 1982. A cultural approach to male-
female miscommunication. In J. Gumperz (ed.) Language and Social
Identity. Cambridge: Cambridge University Press, pp. 197-215.
Manning, P. 1992. Erving Goffman and Modern Sociology. Cambridge:
Polity Press.
Marcu, Daniel, 1994. A Formalisation and an Algorithm for Computing
Pragmatic Inferences and Detecting Infelicities, Ph.D. thesis,
University of Toronto, Computer Systems Research Institute.
Available as Technical Report CSRI-309, Computer Research
Institute, University of Toronto.
Mey, Jacob. 1993. Pragmatics: An Introduction, Blackwell, 1993.
McCawley, James, 1981, Everything That Linguists Have Always Wanted to
Know about Logic (but Were Ashamed to Ask), University of
Chicago Press.
McInnes, F., and Attwater, D. 2004. Turn-taking and grounding in spoken
telephone number transfers. Speech Communication, 43, 205–223.
Mercer, Robert, 1987, A Default Logic Approach to the Derivation of
Natural Language Presuppositions, Ph.D. thesis, University of British
Columbia.
Mercer, Robert, 1992. Default Logic: Towards a common logical semantics
for presupposition and entailment. Journal of Semantics, 9: 223–
250.
Mercer, Neil. 2000. Words and Minds: How We Use Language to Think
Together, London: Routledge.
Montague, Richard. 1974 [1968]. Pragmatics. R.H Thomason (ed.), Formal
Philosophy. New Haven, CT: Yale University Press, pp. 95-118.

168
Morgan, Jerry. 1969. On the Treatment of Presupposition in
Transformational Grammar. In Papers from the Fifth Regional
Meeting, Chicago Linguistic Society, University of Chicago, 167–77.
Morris, Charles W. 1938. Foundations of the Theory of Signs, Chicago: The
University of Chicago Press.
Morris, Charles W. 1946. Signs, Language and Behavior. New York:
Prentice-Hall. Reprinted, New York: George Braziller, 1955.
Reprinted in Charles Morris, Writings on the General Theory of Signs
(The Hague: Mouton, 1971), pp. 73–397.
Nerlich, Graham. 1967. Presupposition and Classical Logical Relations.
Analysis, 27(3): 104–106.
Norwine, A.C., and Murphy, O.J. 1938. Characteristic time intervals in
telephonic conversation. The Bell System Technical Journal, 17, 281–
291.
Nunberg, Geoffrey. 1978. The pragmatics of reference. Mimeo, Indiana
University Linguistics Club.
Nunberg, Geoffrey. 1998. Indexicality and deixis. Ada Kasher, Asa (ed.)
Pragmatics: Critical Concepts, Vol III: Indexicals and reference.
London: Routledge, pp. 145-184.
Partee, Barbare H. 1973. Nominal and Temporal Anaphora. Linguistics and
Philosophy 1, 243-86.
Peirce, Charles 1955 [1940]. Philosophical writings of Peirce. In J. Buchler
(ed.), New York: Dover Publications.
Peters, Stanley. 1979. A truth-conditional formulation of Karttunen's
account of presupposition. Synthese, 40(2): 301–316.
Posner, Roland. 1987. Charles Morris and the Behavioral Foundations of
Semiotics. In M. Krampen et al (eds.) Classics of Semiotics, New
York: Plemun Press.
Prince, Ellen. 1986. On the syntactic marking of presupposed open
propositions. In Farley, A. M. and Farley, P. T. and McCullough,
K.-E. (eds.), Papers from the Parasession on Pragmatics and
Grammatical Theory, 22nd Regional Meeting of the Chicago
Linguistic Society, University of Chicago, 208-222.
Reid, J. 1995. A study in gender differences in minimal responses. In
Journal of Pragmatics, vol.24, issue 5, pp. 489-512.
Reinhart, Tanya. 1982. Pragmatics and Linguistics: An Analysis of
Sentence Topics, Philosophica, 27: 53–94.
Roberts, Craige. 1995. Domain Selection in Dynamic Semantics. In Bach, E.
Jelinek, E. and Kratzer, A. and Partee, B. (eds.), Quantification in
Natural Languages, Dordrecht: Kluwer, 661–700.

169
Roberts, Craige. 1998. Focus, the flow of information, and universal
grammar. In Culicover, P. and McNally, L. (eds.), The Limits of
Syntax, New York: Academic Press, 109-160.
Roberts, F., Francis, A. L., and Morgan, M. 2006. The interaction of inter-
turn silence with prosodic cues in listener perceptions of ‘‘trouble’’ in
conversation. Speech Communication, 48, 1079–1093.
van Rooij, Robert. 2007. Strengthening conditional presuppositions.
Journal of Semantics, 24: 289–304.
Rothschild, Daniel. 2008. Presupposition projection and logical
equivalence. Philosophical Perspectives, 22(1): 473–497.
Russell, Bertrand. 1905. On Denoting. Mind, 14: 479–493.
Sacks, Harvey. 1978. Some technical considerations of a dirty joke. J.N.
Schekein (ed) Studies in the organization of conversational interaction, New
York: Academic, pp. 249-270.
Sacks, Harvey. 1995. Lectures on Conversation, Vol. 1&2, Oxford: Basil
Blackwell.
Sadock, Jerrold M. 1969. Hypersentences. Papers in Linguistics 1: 283-371.
Sadock, Jerrold M. 1970. Whimperatives. Studies presented to R.B. Lees by
his students, ed. by J. M. Sadock and A. L. Vanek, 223-239.
Edmonton: Linguistic Research.
Sadock, Jerrold M. 1972. Speech act idioms. Papers from the Eighth
Regional Meeting of the Chicago Linguistic Society: 329-39.
Sadock, Jerrold M. 1974. Toward a Linguistic Theory of Speech Acts. New
York: Academic Press.
Sadock, Jerrold M. 1977. Aspects of Linguistic Pragmatics. In Rogers, et al.,
eds. 67-78.
Sauerland, Uli. 2010. Embedded implicatures and experimental constraints:
A reply to Geurts & Pouscoulous and Chemla. Semantics and
Pragmatics 3(2):1–13.
Sauerland, Uli. 2012. The Computation of Scalar Implicatures: Pragmatic,
Lexical or Grammatical?. Language and Linguistics Compass 6(1):
36-49.
Sæbø, Kjell Johan. 1996. Anaphoric presuppositions and zero anaphora.
Linguistics and Philosophy, 19: 187–209.
van der Sandt, Rob. 1982. Kontekst en Presuppositie: Een Studie van het
Projektieprobleem en de Presuppositionele Eigenschappen van de
Logische Konnektieven, Ph.D. thesis, Nijmegen Institute of
Semantics.
van der Sandt, Rob. 1988. Context and Presupposition, London: Croom
Helm.

170
van der Sandt, Rob. 1989. Presupposition and discourse structure. In
Bartsch, R. van Benthem, J. and van Emde Boas, P. (eds.), Semantics
and Contextual Expression, Dordrecht: Foris, 287–94.
van der Sandt, Rob. 1992. Presupposition projection as anaphora
resolution. Journal of Semantics, 9: 333–377.
Schegloff, Emanuel A. 1988. Presequences and Indirection: Applying
Speech Act Theory to Ordinary Conversation. Journal of Pragmatics,
12, 55-62.
Schegloff, Emanuel A., Jefferson, Gail and Sacks, Harvey. 1977. The
preference for self-correction in the organization of repair in
conversation. Languge, vol.53 number 2, pp. 361-382.
Schegloff, Emanuel A. 1979. The relevance of repair to a syntax-for-
conversation. In T. Givon (ed.) Syntax and Semantics (vol. 12)
Discourse and Syntax, New York: Academic Press, pp. 261-286.
Schegloff, Emanuel A. 1982. Discourse as an interactional achievement:
some uses of “uh huh” and other things that come between sentences.
In D. Tannen (ed.) Analyzing Discourse: Text and Talk, Washington
D.C.: Georgetown University Press, pp. 71-93.
Schegloff, Emanuel A. 1987a. ‘Some sources of misunderstanding in talk-
in-interaction’ in Linguistics, 25, pp.201-218.
Schegloff, Emanuel A. 1987b. ‘Between Micro and Macro: Contexts and
Other Connections’ in JC Alexander and al. (eds.) The Micro-Macro
Link, Berkeley and Los Angeles: University of California Press, Ltd.,
pp 207-34.
Schegloff, Emanuel A. 1988a. On an actual virtual servo-mechanism for
guessing bad news: A single case conjecture. Social Problems,
35(4), pp.442-457.
Schegloff, Emanuel A. 1988b. ‘Goffman and the analysis of conversation’
in P Drew and A Wootton (eds.) Erving Goffman: Exploring the
interaction order, Cambridge: Polity Press, pp. 89-135
Schegloff, Emanuel A. 1991. Reflections on Talk and Social Structure. In D.
Boden and D. Zimmerman (eds) Talk and Social Structure. Studies in
Ethnomethodology and Conversation Analysis, Polity Press,
1991, pp. 44-70.
Schegloff, Emanuel A. 1992. Repair after Next Turn: The Last Structurally
Provided Defense of Intersubjectivity in Conversation. American
Journal of Sociology 97, pp. 1295-1345.
Schegloff, Emanuel. A. 1996a. Turn organization: On intersection of
grammar and interaction. In E Ochs, E. Schegloff and S. Thompson,
S. (eds.), Interaction and Grammar, Cambridge and New York:
Cambridge University Press.
171
Schegloff, Emanuel A. 1996b. Confirming allusions: Toward an empirical
account of action. American Journal of Sociology 102, pp.161-216.
Schegloff, Emanuel A. 1996c. Issues of Relevance for Discourse Analysis:
Contingency in Action, Interaction and Co-participant Context. In
E.H. Hovy and D.R. Scott (eds.) Computational and Conversational
Discourse: Burning issues – An Interdisciplinary Account, Berlin
Heidelberg New York: Springer, pp. 4-33.
Schegloff, Emanuel A. 1997a. Third Turn Repair. In G.R. Guy, C. Feagin,
D. Schiffrin and J Baugh (eds.) Towards a social science of
language: Papers in honor of William Labov, Volume 2, Social
interaction and discourse structures, Amsterdam: John Benjamins,
pp.31-40.
Schegloff, Emanuel A. 1997b. Practices and actions: Boundary cases of
other-initiated repair. Discourse Process, 23, pp. 499-545.
Schegloff, Emanuel A. 1999. Discourse, pragmatics, conversation, analysis.
Discourse Studies 1 (4), pp. 405-35.
Schegloff, Emanuel A. 2000a. Overlapping talk and the organization of
turn-taking for conversation. Language in Society, 29, pp 1-63.
Schegloff, Emanuel A. 2000b. When ‘others’ initiate repair. Applied
Linguistics 21, pp. 205-243
Schegloff, Emanuel A. 2002. Accounts of Conduct in Interaction:
interruption, overlap, and turn-taking. In J.H. Turner (ed.) Handbook
of Sociological Theory, New York Boston Dordecht London
Moscow: Kluwer Academic/Plenum Publishers, pp. 287-321.
Schegloff, Emanuel A. et al. 2002. ‘Conversation Analysis and Applied
Linguistics’ in Annual Review of Applied Linguistics 22, Cambridge:
Cambridge University Press, pp. 3-31
Schegloff, Emanuel A. 2004. Putting the interaction back into dialogue.
Behavioral and Brain Sciences, 27 (2), pp. 207-208.
Schegloff, Emanuel A. 2005. ‘On Complainability’, in Social Problems,
52(4), pp. 449- 476.
Schiffrin, Deborah. 1987. Discourse markers. Cambridge: Cambridge
University Press.
Schiffrin, Deborah. 1987. Approaches to Discourse. Oxford and Cambridge:
Blackwell.
Schiffrin, Deborah. 1987. Discourse markers, Oxford: Basil Blackwell.
Schiffrin, Deborah. 1994a. Approaches to Discourse. Oxford and
Cambridge: Blackwell.
Schiffrin, Deborah. 1994b. ‘Making a List’ in Discourse Processes 17, pp.
377-406

172
Schlenker, Philippe, 2007. Anti-dynamics: Presupposition projection
without dynamic semantics. Journal of Logic, Language and
Information, 16(3): 325–256.
Schlenker, Philippe. 2008. Be articulate: A pragmatic theory of
presupposition. Theoretical Linguistics, 34: 157–212.
Schlenker, Philippe. 2009. Local contexts. Semantics and Pragmatics,
2(3): 1–78.
Searle, John R. 1969. Speech Acts: An Essay in the Philosophy of Language.
Cambridge: Cambridge University Press.
Searle, John R. 1975a. Indirect Speech Acts. In Cole and Morgan, eds. 59-
82.
Searle, John R. 1975b. A taxonomy of speech acts. Minnesota Studies in the
Philosophy of Science, Vol 9: Language, mind and knowledge, ed. by
K. Gunderson. 344-69. Minneapolis: University of Minnesota Press.
Reprinted in Expression and Meaning. John R. Searle, 1-29.
Cambridge: Cambridge University Press.
Searle, John. 1984. Expression and Meaning. Studies in the Theory of Speech
Acts, Cambridge University Press.
Sellars, W. 1954. Presupposing. Philosophical Review, 63: 197–215.
Sellen, A.J. 1995. Remote conversations: The effects of mediating talk with
technology. Human–Computer Interaction, 10, 401–444.
Shibatani, Masayoshi. 1999. Honorifics. K. Brown and J. Miller (eds.),
Concise Encyclopedia of Grammatical Categories. Amsterdam:
Elsevier, pp. 192-201.
Simons, Mandy. 2001. On the conversational basis of some
presuppositions. In Hastings, R. and Jackson, B. and Zvolensky, Z.
(eds.), Proceedings of Semantics and Linguistics Theory 11, Ithaca,
NY: CLC Publications, 431–448.
Simons, Mandy, 2003, Presupposition and accommodation: Understanding
the Stalnakerian picture. Philosophical Studies, 112(3): 251–278.
Simons, Mandy, 2004. Presupposition and relevance. In Zoltán Szabó
(ed.) Semantics vs. Pragmatics, Oxford University Press, 329–355.
Simons, Mandy. 2006. Foundational issues in presupposition. Philosophy
Compass, 1(4): 357–372.
Simons, Mandy. 2007. Observations on embedding verbs, evidentiality, and
presupposition. Lingua, 117(6): 1034–1056.
Singh, Raj. 2007. Formal alternatives as a solution to the proviso problem.
In Gibson, M. and Friedman, T. (eds.), Proceedings of the 17th
Conference on Semantics and Linguistic Theory, Ithaca, NY: CLC
Publications, Cornell University, 264–228.

173
Soames, Scott. 1979. A projection problem for speaker presuppositions.
Linguistic Inquiry, 10: 623–666.
Soames, Scott. 1982. How presuppositions are inherited: a solution to the
projection problem. Linguistic Inquiry, 13: 483–545.
Soames, Scott. 1989. Presupposition. In Gabbay, D. and Guenther, F.
(eds.), Handbook of Philosophical Logic, Dordrecht: Reidel, vol. IV,
553–616.
Sperber, Dan and Wilson, Deidre. 1986a. Relevance: Communication and
Cognition. Oxford: Blackwell. (Second edition 1995.)
Sperber, Dand and Wilson, Deidre. 1986b. On defining relevance. In R.
Grandy and R. Warner (eds.) Philosophical Grounds of Rationality:
Intentions, Categories, Ends. Oxford: OUP. 143-158.
Stalnaker, Robert. 1972. Pragmatics. In Davidson, D. and Harman, G.
(eds.), Semantics of Natural Language, Reidel: Dordrecht, 389–408.
Stalnaker, Robert. 1973. Presuppositions. The Journal of Philosophical
Logic, 2: 447–457.
Stalnaker, Robert. 1974. Pragmatic presuppositions. In Munitz, M. and
Unger, P. (eds.), Semantics and Philosophy, New York University
Press, 197–214.
Stalnaker, Robert. 1998. On the representation of context. Journal of
Logic, Language and Information, 7: 3–19.
Strawson, Peter F. 1950. On referring. Mind, 59: 320–44.
Strawson, Peter F. 1952. Introduction to Logical Theory, London: Methuen.
Strawson, Peter F. 1954. A Reply to Mr. Sellars. Philosophical Review,
63: 216–231.
Strawson, Peter F. 1964. Identifying reference and truth-values. Theoria,
30: 96–118. Reprinted in: Logico-linguistic Papers. Methuen,
London, 1974, 75–95.
Strawson, Peter. F. 1971. Intention and convention in speech acts. In J.R.
Searle (ed.) The philosophy of language, London: Oxford
University Press, 23-38.
Tannen, Deborah. 1989. Talking voices: repetition, dialogue, and imagery in
conversational discourse, Cambrigde: Cambridge University Press.
Tannen, Deborah. 1994. Gender and Discourse, Oxford: Oxford University
Press.
ten Bosch, L., Oostdijk, N., and Boves, L. 2005. On temporal aspects of turn-
taking in conversational dialogues. Speech Communication, 47, 80–
86.
ten Bosch, L., Oostdijk, N., and de Ruiter, J.P. 2004a. Durationala spects of
turn-taking in spontaneous face-to-face and telephone dialogues. In

174
 Text, speech and dialogue (pp. 563–570). Berlin, Heidelberg:
Springer.
ten Bosch, L., Oostdijk, N., and de Ruiter, J.P. 2004b. Turn-taking in social
talk dialogues: Temporal, formal, and functional aspects. In
Proceedings of the ninth conference on speech and computer
(SPECOM 2004), Saint-Petersburg, Russia (pp. 454–461).
Ten Have, P. 1999. Doing Conversation Analysis: A Practical Guide.
London and Thousand Oaks, CA: Sage.
Thomason, Richmond. 1972. A semantic theory of sortal incorrectness.
Journal of Philosophical Logic, 1: 209–258.
Trimboli, C., and Walker, M.B. 1984. Switching pauses in cooperative and
competitive conversations. Journal of Experimental Social
Psychology, 20, 297–311.
Veltman, Frank 1996. Defaults in update semantics. Journal of
Philosophical Logic, 25: 221–261.
von Fintel, Kai, 1995, Restrictions on Quantifier Domains, Ph.D. thesis,
University of Massachussetts at Amherst.
von Fintel, Kai. 2004. Would you believe it? The king of France is back!
Presuppositions and truth-value intuitions. In Reimer, M. and
Bezuidenhout, A. (eds.), Descriptions and Beyond, Oxford University
Press, 269–296.
Walker, M.B., and Trimboli, C. 1982. Smooth transitions in conversational
interactions. The Journal of Social Psychology,117, 305–306.
Weilhammer, K., and Rabold, S. 2003. Durational aspects in turn-taking. In
Proceedings of the 15th international congress of phonetic sciences
(ICPhS 2003), Barcelona, Spain , pp.2145–2148.
Wettstein, Howard. 1984. How to bridge the gap between meaning and
reference. Synthese 58:63-84.
Wilkins, David and Hill, Deborah. 1995. When ‘GO’ means ‘COME’:
Questioning the basicness of basic motion verbs. Cognitive
Linguistics 6:209-259.
Wilkins, David, Hill, Deborah and Levinson, Stephen. 1995. Bedeutet
KOMMEN und GEHEN in verschiedenen Sprachen immer dasselbe?
Max-Planck-Gesellschaft Jahrbuch, pp. 307-312.
Wilson, Deidre. 1975. Presupposition and Non-truth-conditional Semantics,
London: Academic Press.
Wilson, Deidre and Sperber, Dan. 2002. Truthfulness and relevance. Mind
111. 583-632.
Yeom, J.I., 1998, A Presuppositional Analysis of Specific Indefinites:
Common Grounds as Structured Information States, Garland
Publishing.
175
Yngve, Victor. H. 1970. On getting a word in edgewise. In Papers from the
Sixth Regional Meeting, Chicago Linguistic Society, Chicago:
Chicago Linguistic Society, pp. 567-577
Zeevat, Henk. 1999. Explaining presupposition triggers. In Proceedings of
the Twelfth Amsterdam Colloquium, ILLC, Amsterdam, 19–24.
Zeevat, Henk. 1992. Presupposition and accommodation in update
semantics. Journal of Semantics, 9: 379–412.

176