Pragmatics in the Analysis of

Discourse and Interaction

MARTIN WEISSER

What Is Pragmatics?

Pragmatics, as a linguistic discipline, broadly deals with the analysis and recognition
of meaning in texts. However, in which way this type of meaning is distinct from that
investigated by its neighboring discipline, semantics, is still a matter of debate among at
least some linguists. One thing is clear in this respect, though, which is that, in pragmatic
meaning, contextual influences play a far larger role than in its semantic counterpart. To
begin our identification and discussion of the issues that may be involved in the generation
of pragmatic meaning (see Verschueren, 1999) in and through context, it is perhaps best
to start by looking at the contrasting ‘attitudes’ expressed by different researchers in the
field toward what exactly belongs to the field of pragmatics.

‘Schools’ of Pragmatics
As already hinted at above, there are in fact two different traditions in pragmatics,
sometimes referred to as the “Anglo-American” and the “European Continental” schools
(Horn & Ward, 2004, p. x). The first subscribes to the “component view” (p. x), a view that
sees pragmatics as a separate level of linguistics—such as those of phonetics/phonology,
syntax, and semantics—and the second adopts the ‘perspective view’ (p. x), which sees
pragmatics as a function of language that influences the other levels and incorporates a
larger situational context that also includes sociolinguistic factors. Following Leech (1983,
pp. 10–11), the former can also be referred to as “PRAGMALINGUISTICS” and the latter
as “SOCIOPRAGMATICS.”
Although both approaches are strongly influenced by concepts originating in the ana-
lysis of dialogue in conversation analysis (CA), proponents of pragmalinguistics still tend
to focus more on issues that relate to the ‘sentence’ or micro-level (see Mey, 1993, p. 182),
often also employing more hypothetical, constructed examples, whereas sociopragmaticists
prefer to concentrate more on real-life examples and larger contexts, that is, the macro-level
(p. 182). A further trait of pragmalinguistics is a strong philosophical background that
frequently expresses itself in the use of analyses that employ the same type of traditional
formal logic we also encounter in formal semantics.

General ‘Theoretical’ Issues in Pragmatics

Amongst the more theoretically oriented topics investigated by both schools, mainly on
the micro-level, are implicature, presupposition, reference, deixis, definiteness, and indefiniteness,
as well as speech acts. As implicature is explained in detail elsewhere in this volume, only
some brief examples of some of the other concepts are provided here, but speech acts will
be discussed in more detail later.

The Encyclopedia of Applied Linguistics, Edited by Carol A. Chapelle.

© 2013 Blackwell Publishing Ltd. Published 2013 by Blackwell Publishing Ltd.
DOI: 10.1002/9781405198431.wbeal0938
2 pragmatics in the analysis of discourse and interaction

Presupposition
Presupposition refers to the fact that, in any conversation or text, certain elements of a
discussion can be assumed to be known or understood without first having been explicitly
introduced. Thus, if I ask you When did you stop smoking?, the underlying assumption is
that, prior to this moment, there was actually a period where you were in the habit of
smoking. Often, presuppositions like the above are linked to certain grammatical construc-
tions that involve:

1. a change of state, as in the above example;

2. so-called factive verbs, such as regret, as in I regret the decision I made yesterday, where
the assumption is that I did indeed make a decision at the said time;
3. implicative verbs, such as manage or succeed, as in They managed to finish and hand in their
essays on time, which presupposes that the agents (they) had to make an effort to achieve
this outcome;
4. some form of iteration, as in I had to go to the office again., which expresses the fact that
I had already been there at least once before (presumably today);
5. and so forth.

A similar thing happens when I tell you that My brother is older than me, when you did
not even know that I had a brother in the first place. However, rather than complaining
about the fact that I had not mentioned my having a brother before, you will simply assume
this as being true, unless, of course, you have any reason to doubt my sincerity.

Reference, Deixis, and (In-)Definiteness

As already pointed out above, when we talk about ‘things,’ we always do this in a certain
context, and within this context, it is important to identify the participants in a verbal
exchange, as well as locating this exchange and its participants in space and time. Identifying
who or what exactly we are talking about is generally a question of assigning an explicit
reference to a particular person or entity. Linguistically, this can be achieved by using a
deictic or ‘pointing’ expression, such as a pronoun (I, you, they, etc.), demonstrative (this,
that, etc.), a (definite) noun phrase (the woman on the other side of the road, the Queen of
England, etc.), or a proper noun/name (John, Mary, etc.).
If a person is the target of the reference—and especially if there is a personal pronoun
involved, we speak of person deixis, as in Can I ask you something?, where I and you both
identify speaker roles in the verbal interaction. If the reference takes the form of a temporal
adverbial (construction), such as in We’ll be there tomorrow, this is referred to as temporal
or time deixis, which serves to ‘locate’ the utterance relative to the current time; if it (liter-
ally) locates something in space, relative to the speaker and hearer, this is known as spatial
deixis.
One thing that is particularly interesting about many of the shorter expressions used
in these types of deixis is that they only take on an exact meaning in a particular context,
but are otherwise semantically rather vague, something that led the philosopher Frege to
draw a distinction between the sense of an expression, that is, the basic, decontextualized
meaning, and its actual reference (see Carlson, 2004). This distinction can easily be made
clear by looking at the meaning of the local adverb there in the last example because we
can only really be sure about its exact meaning if we are aware of where the place referred
to is located relative to the speaker and hearer, either because it is physically pointed at
or the location is definitely already known to all parties involved.
Two special types of deixis are social and discourse deixis. The former is basically a type
of person deixis, but with the additional feature of indicating the social ‘status’ of the
 pragmatics in the analysis of discourse and interaction 3

interlocutors. This may take the form of honorifics—such as the use of Sir or Madam—in
English, indications of the degree of familiarity, such as in the use of different second
personal pronouns in French (tu vs. vous) or German (du vs. Sie), or even different speech
styles (or levels) that may involve changes in vocabulary and morphosyntax (as in, for
example, Korean) at the same time. These are also closely linked to issues of politeness,
which is another topic that figures prominently in some types of pragmatics.
Discourse deixis, on the other hand, can be seen as a special type of local or temporal
deixis because it locates utterances not in the ‘physical’ space or event structure the
interlocutors find themselves in, as we saw in our earlier examples, but actually creates
references to prior co-text, as in expressions like As I mentioned earlier . . . , or On page 22. . . .
The issue of definiteness versus indefiniteness basically revolves around the degree of
‘familiarity’ or ‘unique identifiability’ that is expressed through a reference. Thus, for
example, definite articles and pronouns that occur in or as a noun phrase are generally
assumed to produce the most specific references that easily help us to identify a referent
unambiguously, while indefinite articles or general quantifiers, such as some or any, sup-
posedly work less well for uniquely identifying a referent (see Abbott, 2004). However,
there are some problems with these assumptions, as you can easily see if you look at
sentences like I just saw a man standing across the street and looking into our window or the
frequently cited The king of France is bald, where in the first example, clearly a very specific
man, performing a specific action, is referred to using an indefinite article, while in the
second example, a nonexistent personality is referred to via a so-called definite description,
that is, a noun phrase with a definite article.

Speech Acts and Cooperative Behavior

The notion of speech as a kind of action was originally expressed by the ‘ordinary language
philosopher’ Austin in his book How to Do Things with Words (Austin, 1962). There, he
contradicted the traditional assumption, dating back to the days of Aristotle, that sentences
are only used to express propositions, that is, facts that are either true or false:

It was too long the assumption of philosophers that the business of a ‘statement’ can only
be to ‘describe’ some state of affairs, or to ‘state some fact’, which it must do either truly
or falsely. (Austin, 1962, p. 1)

To prove this point, he developed the theory of performative verbs (pp. 14–24), that is,
verbs whose use actually constitutes an action, such as promise, christen, name, declare, and
so forth, which are generally also used in conjunction with a first person pronoun and can
also be reinforced by the adverb hereby, as in, for example, I hereby declare this meeting closed.
He later extended this theory into a more general theory of verbal action that distinguished
between three different types of function of an utterance:

1. locution(ary act)
2. illocution(ary act)
3. perlocution(ary effect). (pp. 98–101)

Function (1) essentially comprises the (physical or syntactic) form of the utterance,
(2) the basic intention of the speaker in producing it, and (3) the effect it produces in the
hearer. Let’s try to clarify this using an example: if I say something like Can you reach the
top shelf? to you, then the locutionary act would simply consist in me uttering these words
with the appropriate rising intonation, the illocutionary part would most likely be me
indirectly asking you for help in getting something off this top shelf (that I myself cannot
4 pragmatics in the analysis of discourse and interaction

reach), and the perlocutionary effect would hopefully be for you to recognize my indirect
question and ask me in turn what I’d like you to fetch for me from there. However, accord-
ing to Austin, these functions can only be successfully performed if a number of felicity
conditions (“Conditions for Happy Performatives”) (Austin, 1962, pp. 12–13) are fulfilled,
for instance that the hearer understand and accept the intentions of the speaker or that
certain conventions or conventional procedures are followed, such as, for example, in a
marriage procedure.
John Searle later developed Austin’s ideas further and even claimed that speech acts
play the most central role in communication:

The unit of linguistic communication is not, as has generally been supposed, the symbol,
the word or sentence, but rather the production or issuance of the symbol or word
or sentence in the performance of the speech act . . . More precisely, the production or
issuance of a sentence token under certain conditions is a speech act, and speech acts . . . are
the basic and minimal unit of linguistic communication. (Searle, 1969, p. 16)

To differentiate between the ‘locutionary elements’ of a sentence, he draws a distinction

between “propositional” and “illocutionary force indicator[s]” (p. 30). The force of an
utterance, according to him, is indicated through “illocutionary force indicating devices”
(or IFIDs, for short) which include “word order, stress, intonation contour, punctuation,
the mood of the verb, and the so-called performative verbs” (p. 30).
He identifies five main classes of speech acts, assertives (e.g., stating, claiming, reporting),
directives (e.g., ordering, requesting, demanding), commissives (e.g., promising, offering),
expressives (e.g., thanking, apologizing, congratulating), and declarations (e.g., resigning,
dismissing, sentencing, christening).
In an attempt to explain how meaning can be expressed and understood in such
an ‘indirect way,’ another ordinary language philosopher, H. P. Grice, formulated the
cooperative principle (often abbreviated CP). In this, he basically claims that participants
in a verbal interaction behave largely cooperatively, both in the construction and inter-
pretation of what they say. The CP embodies a set of categories, generally these days referred
to as maxims, although Grice himself in fact used this term to refer to the subparts of each
category. A brief summary of these categories is given below:

1. Quantity: ‘say just as much, but no more, than required.’

2. Quality: ‘only say what you think is true and you can prove.’
3. Relation: ‘only say what is relevant.’
4. Manner: ‘be clear, unambiguous, and brief.’

The CP has often been criticized a series of strict rules that are generally untenable, but
what most of its critics seem to have overlooked is that they are just guidelines for best
case behavior, as well as expressing certain expectations most interlocutors would bring
into a verbal exchange, anyway. However, there are certainly accidental or deliberate
cases—referred to as flouting, violating, infringing, opting out of, or suspending one or more
of the categories (see Thomas, 1995, for some examples)—where the CP is not adhered to
in almost all conversations.
As there seems to be a certain degree of overlap between the different categories of
the CP, frequent attempts were made to reduce and ‘streamline’ the categories involved
in explaining the mechanisms involved in the cooperative construction of meaning, most
notably by Sperber and Wilson, Horn, and Levinson. The simplest of these is Sperber and
Wilson’s, who make do with only a single principle, that of relevance, claiming that “the
expectations of relevance raised by an utterance are precise and predictable enough to
guide the hearer toward the speaker’s meaning.” (Wilson & Sperber, 2006, p. 608)
 pragmatics in the analysis of discourse and interaction 5

At the next level, we find Horn’s dual distinction between the Q(uantity) and the
R(elation) principle (Horn, 1984, p. 13), where the former incorporates the ‘sub-maxims’
“MAKE YOUR CONTRIBUTION SUFFICIENT” and “SAY AS MUCH AS YOU CAN”
(p. 13) and the latter “MAKE YOUR CONTRIBUTION NECESSARY” and “SAY NO MORE
THAN YOU MUST” (p. 13).
In contrast to Horn, Levinson posits the three principles of Q(uantity), I(nformativeness),
and M(anner), where the latter may be better classified as markedness, though. One further
important distinction to both Grice and Horn is that each of Levinson’s principles involves
cooperative efforts on both the part of the speaker and the hearer.
Technically speaking, though, one could argue that both Horn’s and Levinson’s prin-
ciples do not really represent full reductions of the Gricean system, in particular also
because they both still do not seem to fully reject the role of Quality (Horn, 1984; Levinson,
2000), but which simply does not form part of their systems, as they are primarily con-
cerned with implicatures and inferencing.

Pragmatics Applied

Having discussed all the relevant theoretical issues in pragmatics, we can now turn to the
practical aspects of how and where these theories have been and may be applied in the
analysis of discourse and interaction. Issues like reference, deixis, and (in)definiteness,
which are more grammatical in nature, can relatively easily be incorporated into the
regular investigation and teaching of grammar, usually in the context of analyzing written
language. Here, we would mainly be dealing with issues such as cohesion or coherence.
Research into the former can improve our understanding of how texts ‘hang together’
through the use of conjunctions, ellipses, pro-forms, and so forth, and which strategies
allow authors to link up the different parts of a text to improve its flow, as well as to make
it more ‘compact’ by eliminating redundancies. Investigating coherence, although it may
formally involve some of the same features, is more concerned with identifying the logic
behind the sequencing of utterances, whether reference is assigned unambiguously, and
deixis successfully employed to reflect spatial and temporal order or meaning.
Those issues like presupposition, implicature and speech acts that embody even more
context-dependent cooperative or cultural/conventional mechanisms, as well as maybe a
certain degree of indirectness, and tend to involve more lexicogrammatical or semantic
aspects, are much more difficult to handle because they follow less clearly defined or
researched patterns. Due to this fact, research into them has often been focusing on speech
acts in relation to exchange structures in spoken interaction, as in the Birmingham tradi-
tion of discourse analysis (see Coulthard, 1992), which initially concentrated on classroom
interaction, or on identifying certain social norms and sequencing of speech events, two
of the major foci in conversational analysis (see Seedhouse, 2005). More recent approaches
to discourse analysis, as well as sociopragmatics, also tend to focus more strongly on the
sequencing and social aspects of linguistic utterances, investigating issues such as adjacency
pairs or preferred responses, which has also had an influence on communicative approaches
to language teaching. Different other strategies of verbal interaction, such as the use of
discourse markers (Jucker & Ziv, 1998; Fischer, 2006), backchannels, or the use of (meaningful)
filled pauses, have more recently also become more of a focus of attention. Issues of polite-
ness and its relationship to (in)directness have been investigated in studies on cross-cultural
and interlanguage pragmatics (see Rose & Kasper, 2001, or Bardovi-Harlig, 2002, for some
general background).
As pointed out earlier, though, a fair degree of the ‘research’ in the Anglo-American
tradition is still carried out using constructed examples, so the adoption of a more data/
corpus-oriented approach in general would certainly be preferable. Research in corpus
6 pragmatics in the analysis of discourse and interaction

linguistics and attempts in computational linguistics to improve dialogue systems (Leech &
Weisser, 2003) have already led to the development of extended speech act taxonomies or
other aspects (Allen & Core, 1997) of identifying the different levels of meaning involved
in constructing meaning in dialogues. Identifying these different levels then also makes it
possible to annotate large bodies of data (semi-)automatically, for which both statistical
(Stolcke et al., 2000) and symbolic (Weisser, 2004) means have been developed.
Although it is of course still possible to conduct non-computer-based pragmatic analyses
in the traditions referred to above, a more widespread dissemination and adoption of
pragmatically annotated corpora for research would probably make it considerably easier
to investigate all the diverse topics in spoken and written interaction.

SEE ALSO: Analysis of Dialogue; Analysis of Discourse and Interaction: Overview; Corpus
Linguistics: Overview; Inference and Implicature; Pragmatics: Overview; Speech Acts;
Speech Acts Research

References

Abbott, B. (2004). Definiteness and indefiniteness. In L. Horn & G. Ward (Eds.), The handbook of
pragmatics. Oxford, England: Blackwell.
Allen, J., & Core, M. (1997). Draft of DAMSL: Dialog act markup in several layers. Retrieved
August 30, 2011 from ftp://ftp.cs.rochester.edu/pub/packages/dialog-annotation/manual.
ps.gz
Austin, J. L. (1962). How to do things with words. Oxford, England: Oxford University Press.
Bardovi-Harlig, K. (2002). Pragmatics and second language acquisition. In R. Kaplan (Ed.), The
Oxford handbook of applied linguistics. Oxford, England: Oxford University Press.
Carlson, G. (2004). Reference. In L. Horn & G. Ward (Eds.), The handbook of pragmatics. Oxford,
England: Blackwell.
Coulthard, M. (Ed.). (1992). Advances in spoken discourse analysis. London, England: Routledge.
Fischer, K. (Ed.). (2006). Approaches to discourse particles. Amsterdam, Netherlands: Elsevier.
Horn, L. (1984). Toward a new taxonomy for pragmatic inference: Q-based and R-based
implicature. In D. Schiffrin (Ed.), Meaning, form and use in context (GURT ’84) (pp. 11–42).
Washington, DC: Georgetown University Press.
Horn, L., & Ward, G. (Eds.). (2004). The handbook of pragmatics. Oxford, England: Blackwell.
Jucker, A., & Ziv, Y. (Eds.). (1998). Discourse markers: Description and theory. Amsterdam,
Netherlands: John Benjamins.
Leech, G. (1983). Principles of pragmatics. London, England: Longman.
Leech, G., & Weisser, M. (2003). Pragmatics and dialogue. In R. Mitkov (Ed.), The Oxford hand-
book of computational linguistics. Oxford, England: Oxford University Press.
Levinson, S. (2000). Presumptive meanings. Cambridge, MA: MIT Press.
Mey, J. (1993). Pragmatics: An introduction. Oxford, England: Blackwell.
Rose, K., & Kasper, G. (Eds.). (2001). Pragmatics in language teaching. Cambridge, England:
Cambridge University Press.
Searle, J. (1969). Speech acts: An essay in the philosophy of language. Cambridge, England: Cambridge
University Press.
Seedhouse, P. (2005). Conversation analysis as research methodology. In K. Richards &
P. Seedhouse (Eds.), Applying conversation analysis. London, England: Palgrave Macmillan.
Stolcke, A., Coccarro, N., Bates, R., Taylor, P., Van Ess-Dykema, C., Ries, K., . . . & Meteer, M.
(2000). Dialogue act modeling for automatic tagging and recognition of conversational
speech. Computational Linguistics, 26(3).
Thomas, J. (1995). Meaning in interaction: An introduction to pragmatics. London, England: Longman.
Verschueren, J. (1999). Understanding pragmatics. London, England: Edward Arnold.
 pragmatics in the analysis of discourse and interaction 7

Weisser, M. (2004). Tagging dialogues in SPAACy. In J. Véronis (Ed.), Le traitement automatique

des corpus oraux (Special issue). Traitement Automatique des Langues, 45(2), 131–57.
Wilson, D., & Sperber, D. (2006). Relevance theory. In L. Horn & G. Ward (Eds.), The handbook
of pragmatics (pp. 607–32). Oxford, England: Blackwell.

Suggested Readings

Grice, H. P. (1989). Studies in the way of words. Cambridge, MA: Harvard University Press.
Levinson, S. (2004). Deixis. In L. Horn & G. Ward (Eds.), The handbook of pragmatics. Oxford,
England: Blackwell.
McCarthy, M. (1991). Discourse analysis for language teachers. Cambridge, England: Cambridge
University Press.