Semantics and pragmatics

Author: Richard Breheny

© Dr Richard Breheny

Table of contents

• I Methodology
• II Context and content
• III Content and inference
• Bibliography
• Related links

Broadly, pragmatics concerns itself with phenomena relating to language use and interpersonal communication while semantics is concerned with the meaning and interpretation of language. While there is clearly some overlap in the concerns of the two disciplines, they largely co-exist in a symbiotic relationship. Here we will give some indication of how.

I Methodology

As with most branches of linguistics, semantics looks at natural language as a fairly abstract phenomenon. Abstract entities such as phrases and sentences are thought to have invariant linguistic properties including semantic properties. The semantic properties bear on the interpretation of sentences and phrases when used in utterances. In particular, semantics is concerned with accounting for the information bearing potential of language and certain (largely inferential) relations among sentences or utterances. A common rule of thumb applied by semanticists is to suppose that if a given utterance of a declarative sentence S is perceived as an assertion that p then the meaning of the sentence determines what p is. A common problem which a blind application of this rule of thumb runs into has to do with what has been identified as (particularised or generalised) conversational implicature. Conversational implicature has been identified as being that part of what an utterance means which is not determined by what the sentence uttered means. To illustrate, it is clear that B's response below implies that he cannot go swimming because he has a cold, but it is also clear that this is an implication which attaches to the utterance in the context and does not necessarily follow from the meaning of the sentence uttered:

A: Do want to go swimming?
B: I have a cold.

While it is relatively easy to distinguish in this example between what information derives from the meaning of the sentence uttered and what follows from this sentence's being uttered in the context, untutored folk intuitions about what is being asserted, denied, proposed etc by an utterance of a declarative sentence often do not distinguish between these components. Some well known examples include the following where the implication indicated is normally taken to be part of the explicit content but is arguably not part of the conventional or linguistic meaning of the sentence uttered:

I have broken a finger –>the speaker broke her own finger
Mary has two children ->? Mary has no more than two children
John hasn't given up smoking -> John has smoked

In a series of landmark papers (collected in Grice 1989), Grice introduced a distinction between speaker meaning and conventional (or linguistic) meaning. The difference is what is conversationally implicated. He also introduced a theory of how implicatures are generated given conventional meaning and context. This theory is based on the assumption that conversation is a rational, co-operative activity and (in line with Grice's view on meaning in general) it places the speaker's intentions at the centre of the account. Grice used his theory to argue that the linguistic meaning of natural language was much closer to that of the formal languages proposed by Russell and Frege than had previously been supposed and that it was much more stable than his erstwhile colleagues at Oxford (Austin, Strawson) had claimed. To illustrate, the sentence, "Mary has two children" does not carry the implication that Mary has no more than two children in all contexts of use (cf "If Mary has two children, she can claim a tax refund"). Without a theory of implicature, one would have to say that sometimes the sentence has its 'logical' meaning and sometimes it means something more. Grice's theory of implicature considerably lightens the burden on a semantic theory and enables one to claim that everyday discourse is far more 'logical' or less fuzzy than it would otherwise seem. (see Grice 1989 esp chs 1-3, Levinson 1983, Gamut 1991 ch 6).

A number of refinements to Grice's theory have been proposed (see Levinson 2000), as have alternative theories (see Sperber & Wilson 1986). However, the essence of the Gricean methodology remains and the moral that Grice teaches the semanticist is ignored at the semanticist's peril: the data for semantics is far more problematic and far less straightforward than it at first appears.

Ongoing issues: There is still considerable scope for debate about Grice's proposal and the Gricean program. For example, many remain to be convinced whether a program of formal, logically oriented semantics plus Gricean pragmatics can really come close to capturing what goes on in natural language. In particular, many are unconvinced that the open-endedness of meanings of expressions could be accounted for in this way. Consider a colour predicate such as 'red'. We could say about this expression that it's meaning shifts around according to context, or the purposes of the conversation. A red book is usually a book whose cover is red, a red grapefruit is yellow on the surface. In an appropriate context, we can describe a car produced with a special red plastic as a red car to discriminate it from one which has been produced with a special blue plastic, regardless of the colour of the cars (see Chomsky 2000, Lakoff 1987 for this kind of view). A defender of the Gricean program could however point to the fact that there is a commonality to all these construals of 'red' in that redness is involved in each predication. There is scope within the Gricean program to allow that words may have context-dependent meanings (see the next section). Clear cut examples include indexicals ('I', 'he', 'come', 'local' etc) others include scalar terms ('tall', 'fast' etc). So some kind of story involving hidden parameters may be possible for this predicate. Alternatively, it has been proposed that pragmatic principles are involved in 'local' adjustments to meaning so that the interpretation of a word in a context is open to negotiation, even assuming that that word has a fixed linguistic meaning. (See Davidson 1986, Recanati 1995, Sperber & Wilson 1998, Carston 2002, Lascarides & Copestake 1998, Blutner 1998).

II Context and content

Let us reconsider the rule of thumb for semantics: if a given utterance of a declarative sentence S is perceived as an assertion that p then the meaning of the sentence determines what p is. A moment's reflection on this will reveal yet another problem for semantics. This problem has to do with what we might call the context-dependence of natural language meaning (and what Barwise & Perry have called the 'efficiency of natural language'). Taking "I have a cold" to illustrate, we see that, in one sense, what we know about this sentence does not tell us all we need to know to determine what p is when it is asserted. We have to go beyond linguistic knowledge to consider facts about the utterance situation in order to be able to say what p is (we need to know who utters the sentence). On the other hand, we could say that our linguistic knowledge does determine p but relative to a context. In the case at hand, the meaning of 'I' would specify that the speaker is the referent of the term. Somewhat more problematic is the case of third person pronouns, as is illustrated by the well-known pair below:

                     The town councillors refused the protestors permission to demonstrate.
                                 They feared a riot.
                     The town councillors refused the protestors permission to demonstrate.
                                They were anarchists.

Although we may have firm intuitions about who the pronouns in the second sentence refer to in each case, it is not difficult to construct a plausible story whereby they refer to the other group, regardless of the context of utterance. Unlike "I", most context dependent expressions exhibit this kind of indeterminacy with regards their interpretation. While it is straightforward to specify a semantic rule for "they" which still links its interpretation to a feature of the context (something like the object of the speaker's referring intention in uttering a given token - see Kaplan 1989), this still leaves a problem of co-ordination between the speaker and her audience. One important area of research then concerns what principles constrain the possible referents of context-dependent expressions. There are a variety of views about these principles. It has been proposed that they are principles of discourse considered as a text (see Hobbs 1978, Asher & Lascarides 1993, 1998); that they are principles which affect the conduct of discourse by the participants (see Sperber & Wilson 1986, Carston 2002, Parikh 2001, Stalnaker 1978) and that they are cognitive principles which govern discourse processing (Walker et al 1998, Garrod & Sanford 1994). These views are not necessarily mutually exclusive and some current work involves co-ordinating levels of explanation (see Asher 1999, Recanati 2002)

III Content and inference

When the principles of discourse have resolved the indeterminacies inherited from the linguistic meaning of a sentence, there is still an ongoing issue about whether the resulting proposition is always the content of the utterance in question. A classic illustration of this issue involves temporal sequence in conjunction:

Mary didn't get married and fall pregnant. She fell pregnant and got married.

It is generally agreed that the temporal sequence implied in each of the above two sentences is not determined by the meaning of the sentences themselves. They are more like conversational implicatures. However, these implications are part of the discursive or logical content of the discourse (since otherwise it would strike us as contradictory). Given cases like these, many have argued that the logical content of discourse is some combination of linguistically and pragmatically determined information (Carston 2002, Recanati 1989, Perry 1998, Stalnaker 1978, Asher 1999), although there is resistance to this view (Stanley 2000, Cappelen & Lepore 1997, Dekker 2002).

Bibliography

Asher, N. (1999). Discourse structure and the logic of conversation. In K. Turner (ed) The Semantics/Pragmatics Interface from Different Points of View. Dordrecht: Elsevier
Asher N, & A. Lascarides, (1993). Temporal Interpretation, Discourse Relations and Commonsense Entailment. Linguistics and Philosophy 16, 5:437-493.
Asher, N. & A. Lascarides, (1998). Bridging. Journal of Semantics. 15:83-113.
Cappelen, H. and E. Lepore (1997). On an Alleged Connection between Indirect Quotation and Semantic Theory. Mind and Language 12:278-296.
Chomsky, N. (2000). New Horizons in the Study of Language and Mind. Cambridge: CUP
Davidson, D. (1986). A nice derangement of epitaphs. In R. Grandy & R. Warner (eds) Philosophical Grounds of Rationality. Oxford: Clarendon. 157-174
Dekker, P. (2002). Grounding Dynamic Semantics. To appear in M. Reimer & A Bezuidenhout (eds) Descriptions and Beyond. Oxford: OUP.
Blutner, R. (1998). Lexical Pragmatics. Journal of Semantics 15:115-162.
Carston, R. (2002). Thoughts and Utterances. Oxford: Blackwells
Gamut, L.T.F. (1991). Logic, Language & Meaning Chicago: U. of Chicago Press
Garrod, S. & A. Sanford (1994). Resolving sentences in a discourse context: How discourse representation affects language understanding. in M.Gernsbacher (ed) Handbook of Psycholinguistics. San Diego: Academic Press. 675-697
Grice, H.P. (1989). Studies in the Way of Words. Cambridge, Ma. Harvard University Press
Hobbs, J. (1979). Coherence and Coreference. Cognitive Science 3:67-90.
Kaplan, D. (1989). Demonstratives & Afterthoughts. In J. Almog et al. (eds.) Themes From Kaplan. New York: OUP. 481-614
Lakoff, G. (1987). Women, Fire and Dangerous Things. Chicago: U. of Chicago Press
Lascarides, A. & A. Copestake (1998). Pragmatics and Word Meaning. Journal of Linguistics 34: 387-414
Levinson, S. (1983). Pragmatics. Cambridge: CUP
Levinson, S. (2000). Presumptive Meanings. Cambridge Ma.: MIT Press
Perry, J. (1998). Indexicals, contexts and unarticulated constituents. In A. Aliseda et al (eds) Computing Natural Language. Stanford: CSLI Press.
Recanati, F. (1989). The pragmatics of what is said. Mind and Language, 4, 295-329.
Recanati, F. (1995). The Alleged Priority of Literal Interpretation. Cognitive Science 19: 207-232.
Recanati, F. (2002). Does linguistic communication rest on inferences?. Mind & Language 17, 1&2: 105-126.
Sperber D. & D. Wilson.. (1986). Relevance: communication and cognition. Blackwell: Oxford. (second edition, 1995)
Sperber D. & D. Wilson (1998). The mapping between the mental and the public lexicon. In P. Carruthers & J. Boucher (eds.) Language and thought. Cambridge: CUP. 184-200
Stalnaker, R. (1978). Assertion. In P. Cole (ed.) Syntax and Semantics vol. 9: Pragmatics. New York: Academic Press
Stanley, J. (2000). Context and logical form. Linguistics and Philosophy, 23: 391-434.
Walker, M., A. Joshi & E. Prince (1998). Centering Theory in Discourse. Oxford: OUP

Related links

Papers in semantics and also web resources links relevant to semantics/pragmatics:
http://semanticsarchive.net/
Papers on pragmatics and semantics:
http://cogprints.ecs.soton.ac.uk/view/subjects/ling.html
A good source of basic information on keywords:
http://cognet.mit.edu/MITECS/Front/introduction.html

Referencing this article

Below are the possible formats for citing Good Practice Guide articles. If you are writing for a journal, please check the author instructions for full details before submitting your article.

• MLA style:
  Canning, John. "Disability and Residence Abroad". Southampton, 2004. Subject Centre for Languages, Linguistics and Area Studies Guide to Good Practice. 7 October 2008. http://www.llas.ac.uk/resources/gpg/2241.
• Author (Date) style:
  Canning, J. (2004). "Disability and residence abroad." Subject Centre for Languages, Linguistics and Area Studies Good Practice Guide. Retrieved 7 October 2008, from http://www.llas.ac.uk/resources/gpg/2241.