Pragmatics

Pragmatics is the study of the context-dependent aspects of MEANING that are systematically abstracted away from in the construction of LOGICAL FORM. In the semiotic trichotomy developed by Charles Morris, Rudolph Carnap, and C. S. Peirce in the 1930s, SYNTAX addresses the formal relations of signs to one another, SEMANTICS the relation of signs to what they denote, and pragmatics the relation of signs to their users and interpreters. Although some have argued for a pragmatics module within the general theory of speaker/hearer competence (or even a pragmatic component in the grammar), Sperber and Wilson (1986) argue that, like scientific reasoning -- the paradigm case of a nonmodular, horizontal system -- pragmatics cannot be a module, given the indeterminacy of the predictions it offers and the global knowledge it invokes (see MODULARITY AND LANGUAGE). In any case, a regimented account of language use facilitates a simpler, more elegant description of language structure. Those areas of context-dependent yet rule-governed aspects of meaning reviewed here include deixis, speech acts, presupposition, reference, and information structure; see also IMPLICATURE.

Pragmatics seeks to "characterize the features of the speech context which help determine which proposition is expressed by a given sentence" (Stalnaker 1972: 383). The meaning of a sentence can be regarded as a function from a context (including time, place, and possible world) into a proposition, where a proposition is a function from a possible world into a truth value. Pragmatic aspects of meaning involve the interaction between an expression's context of utterance and the interpretation of elements within that expression. The pragmatic subdomain of deixis or indexicality seeks to characterize the properties of shifters, indexicals, or token-reflexives (expressions like I, you, here, there, now, then, hereby, tense/aspect markers, etc.) whose meanings are constant but whose referents vary with the speaker, hearer, time and place of utterance, style or register, or purpose of speech act (see Levinson 1983, chap. 2).

If pragmatics is "the study of linguistic acts and the contexts in which they are performed" (Stalnaker 1972: 383), speech-act theory constitutes a central subdomain. It has long been recognized that the propositional content of utterance U can be distinguished from its illocutionary force, the speaker's intention in uttering U. The identification and classification of speech acts was initiated by Wittgenstein, Austin, and Searle. In an explicit performative utterance (e.g., I hereby promise to marry you), the speaker does something -- that is -- performs an act whose character is determined by her intention, rather than merely saying something. Austin (1962) regards performatives as problematic for truth-conditional theories of meaning, because they appear to be devoid of ordinary truth value; an alternate view is that a performative is automatically self-verifying when felicitous, constituting a contingent a priori truth like I am here now. Of particular linguistic significance are indirect speech acts, where the form of a given sentence (e.g., the yes/no question in Can you pass the salt?) belies the actual force (here, a request for action) characteristically conveyed by the use of that sentence. (See Levinson 1983; chap. 4, and Searle and Vanderveken 1985 for more on speech-act theory and its formalization.)

Although a semantic or logical presupposition is a necessary condition on the truth or falsity of statements (Frege 1892, Strawson 1950), a pragmatic presupposition is a restriction on the common ground, the set of propositions constituting the current context. Its failure or nonsatisfaction results not in truth-value gaps or nonbivalence but in the inappropriateness of a given utterance in a given context (Stalnaker 1974; Karttunen 1974). In presupposing Φ, I treat Φ as an uncontroversial element in the context of utterance; in asserting Ψ, I propose adding the propositional content of Ψ to the common ground or, equivalently, discarding ~Ψ from the set of live options, winnowing down the context set (possible worlds consistent with the shared beliefs of S[peaker] and H[earer]) by jettisoning worlds in which Ψ does not hold.

In stating Even Kim left I assert that Kim left while presupposing that others left and that Kim was unlikely to have left. Such presuppositions can be communicated as new information by a speaker who "tells his auditor something . . . by pretending that his auditor already knows it" (Stalnaker 1974: 202). S's disposition to treat a proposition as part of the common ground, thereby getting H to adjust his model of the common ground to encompass it, is codified in Lewis's rule of accommodation for presupposition (1979: 340): "If at time t something is said that requires presupposition P to be acceptable, and if P is not presupposed just before t, then -- ceteris paribus and within certain limits -- presupposition P comes into existence at t"." Accommodation, a special case of Gricean exploitation, is generalized by Lewis to descriptions, modalities, vagueness, and performatives.

How are the presuppositions of a larger expression determined compositionally as a function from those of its subexpressions? Karttunen's (1974) solution to this "projection problem" partitions operators into plugs, holes, and filters, according to their effect on presupposition inheritance, whereas Karttunen and Peters (1979) propose a formalization of inheritance of pragmatic presuppositions qua "conventional implicatures." Gazdar (1979) offers an alternative mechanism in which the potential presuppositions induced by subexpressions are inherited as a default but are canceled if they clash with propositions already entailed or implicated by the utterance or prior discourse context.

Subsequent work identifies empirical and conceptual problems for these models. Heim (1983) identifies an operator's projection properties with its context-change potential. Presuppositions are invariant pragmatic inferences: A sentence Σ presupposes φ if every context admitting entails φ. If a context c (a conjunction of propositions) is true and c admits Σ, then Σ is true with respect to c if the context incremented by Σ is true. But if Σ is uttered in a context c not admitting it, the addressee will adjust c to c', a context close to c but consistent with Σ. Heim's projection theory thus incorporates Stalnaker-Lewis accommodation, which appeals in turn to the Gricean model of a cooperative conversational strategy dynamically exploited to generate pragmatic inferences (see DYNAMIC SEMANTICS).

Soames (1989) provides a conspectus of formal approaches to presupposition, and see also van der Sandt (1992) for an anaphoric account of PRESUPPOSITION, projection, and accommodation formulated within discourse representation theory. On van der Sandt's theory, the very presupposition that presuppositions are determined compositionally is challenged, leading to a reassessment of the entire projection-problem enterprise.

Although speech acts and presuppositions operate primarily on the propositional level, reference operates on the phrasal level. Reference is the use of a linguistic expression (typically a noun phrase) to induce a hearer to access or create some entity in his mental model of the discourse. A discourse entity represents the referent of a linguistic expression -- that is, the actual individual (or event, property, relation, situation, etc.) that the speaker has in mind and is saying something about.

Within philosophy, the traditional view has been that reference is a direct semantic relationship between linguistic expressions and the real world objects they denote (see SENSE AND REFERENCE and REFERENCE, THEORIES OF). Researchers in computer science and linguistics, however, have taken a different approach, viewing this relation as mediated through the (assumed) mutual beliefs of speakers and hearers, and therefore as quintessentially pragmatic. Under this view, the form of a referring expression depends on the assumed information status of the referent, which in turn depends on the assumptions that a speaker makes regarding the hearer's knowledge store as well as what the hearer is attending to in a given discourse context.

Given that every natural language provides its speakers with various ways of referring to discourse entities, there are two related issues in the pragmatic study of reference: (1) What are the referential options available to a speaker of a given language? (2) What are the factors that guide a speaker on a given occasion to use one of these forms over another? The speaker's choice among referring expressions (e.g., zero forms, pronominals, indefinites, demonstratives, definite descriptions, proper names) is constrained by the information status of discourse entities. Unidimensional accounts (e.g,. Gundel, Hedberg, and Zacharski 1993) provide a single, exhaustively ordered dimension ("assumed familiarity," "accessibility," "givenness") along which the various types of referring expressions are arranged. More recently, Prince (1992) offers a two-dimensional account in which entities are classified as, on the one hand, either discourse-old or discourse-new (based on whether or not they have been evoked in the prior discourse) and, on the other hand, either hearer-old or hearer-new (based on whether they are assumed to be present within the hearer's knowledge-store).

Related to information status is the notion of definiteness, which has been defined both as a formal marking of NPs and as an information status. Research into the meaning of the English definite article has generally been approached from one of two perspectives (Birner and Ward 1994); its felicitous use has been argued to require that the referent of the NP be either familiar within the discourse or uniquely identifiable to the hearer. In the absence of prior linguistic evocation, the referent must be accommodated (Lewis 1979) into the discourse model by the hearer.

Research into the discourse functions of syntax is based on the observation that every language provides its speakers with various ways to structure the same proposition. That is, a given proposition may be variously realized by a number of different sentence-types, or constructions, each of which is associated with a particular function in discourse. Consider the sentences in (1).

(1): a. John did most of the work on that project.; b. Most of the work on that project was done by John.; c. Most of the work on that project John did.; d. It's John who did most of the work on that project.

The same proposition expressed by the canonical word-order sentence in (1a) can also be expressed by the (truth-conditionally equivalent) passive sentence in (1b), by the topicalization in (1c), and by the cleft sentence in (1d), among others, each of which reflects the speaker's view on how it is to be integrated by the hearer into the current discourse. For example, the topicalization (1c) allows the speaker to situate familiar, or discourse-old (Prince 1992) information in preposed position, thus marking the preposed constituent as related -- or "linked" -- to the prior discourse, whereas use of the cleft in (1d) reflects the speaker's belief that her hearer has in mind the fact that somebody did most of the work in question. Finally, with the passive in (1b), in which the canonical order of arguments is reversed, the speaker may present information that is relatively familiar within the discourse before information that is relatively unfamiliar within the discourse.

Such constructions serve an information-packaging function in that they allow speakers to structure their discourse in a maximally accessible way, thereby facilitating the incorporation of new information into the hearer's knowledge-store. Like referring expressions, propositions contain information that can be either discourse-new/old and hearer-new/old.

Vallduví (1992) proposes a hierarchical articulation of information within his theory of informatics. Sentences are divided into the focus, which represents that portion of information that is hearer-new, and the ground, which specifies how that information is situated within the hearer's knowledge-store. The ground is further divided into the link, which denotes an address in the hearer's knowledge-store under which he is instructed to enter the information, and the tail, which provides further directions on how the information must be entered under a given address (see also Rooth 1992). Lambrecht (1994) identifies three categories of information structure: presupposition and assertion (the structure of propositional information into given and new); identifiability and activation (the information status of discourse referents); and topic and focus (the relative predictability of relations among propositions). (See also the Functional Sentence Perspective frameworks of Firbas 1966 and Kuno 1972 and the overview in Birner and Ward, 1998.)

Additional links

-- Laurence Horn and Gregory Ward

References

Austin, J. L. (1962). How To Do Things With Words. Oxford: Clarendon Press.

Birner, B. J., and G. Ward. (1994). Uniqueness, familiarity, and the definite article in English. Berkeley Linguistics Society 20:93-102.

Birner, B. J., and G. Ward. (1998). Information Status and Noncanonical Word Order in English. Amsterdam: Benjamins.

Firbas, J. (1966). Non-thematic subjects in contemporary English. Travaux Linguistiques de Prague 2:239-56.

Frege, G. (1952/1892). On sense and reference. In P. Geach and M. Black, Eds., Translations from the Philosophical Writings of Gottlob Frege. Oxford: Blackwell, pp. 56-78.

Gazdar, G. (1979). Pragmatics. New York: Academic Press.

Gundel, J., N. Hedberg, and R. Zacharski. (1993). Givenness, implicature, and the form of referring expressions in discourse. Language 69:274-307.

Heim, I. (1983). On the projection problem for presuppositions. WCCFL 2 pp. 114-25.

Karttunen, L. (1974). Presupposition and linguistic context. Theoretical Linguistics 1:181-193.

Karttunen, L., and S. Peters. (1979). Conventional implicature. In C.-K. Oh and D. A. Dinneen, Eds., Syntax and Semantics 11: Presupposition. New York: Academic Press, pp. 1-56.

Kuno, S. (1972). Functional sentence perspective: A case study from Japanese and English. Linguistic Inquiry 3:269-320.

Lambrecht, K. (1994). Information Structure and Sentence Form. Cambridge: Cambridge University Press.

Levinson, S. (1983). Pragmatics. Cambridge: Cambridge University Press.

Lewis, D. (1979). Scorekeeping in a language game. Journal of Philosophical Logic 8:339-359.

Prince, E. F. (1981). Toward a taxonomy of given/new information. In P. Cole, Ed., Radical Pragmatics. New York: Academic Press, pp. 223-254.

Prince, E. F. (1992). The ZPG letter: Subjects, definiteness, and information-status. In S. Thompson and W. Mann, Eds., Discourse Description: Diverse Analyses of a Fundraising Text. Amsterdam: John Benjamins, pp. 295-325.

Rooth, M. (1992). A theory of focus interpretation. Natural Language Semantics 1:75-116.

Searle, J., and D. Vanderveken. (1985). Foundations of Illocutionary Logic. Cambridge: Cambridge University Press.

Soames, S. (1989). Presupposition. In D. Gabbay and F. Guenthner, Eds., Handbook of Philosophical Logic, 4. Dordrecht: Reidel, pp. 553-616.

Sperber, D., and D. Wilson. (1986). Relevance. Cambridge, MA: Harvard University Press.

Stalnaker, R. (1972). Pragmatics. In D. Davidson and G. Harman, Eds., Semantics of Natural Language. Dordrecht: Reidel, pp. 380-397.

Stalnaker, R. (1974). Pragmatic presuppositions. In M. Munitz and P. Unger, Eds., Semantics and Philosophy. New York: New York University Press, pp. 197-214.

Strawson, P. F. (1950). On referring. Mind 59:320-344.

Vallduví, E. (1992). The Informational Component. New York: Garland.

van der Sandt, R. A. (1992). Presupposition projection as anaphora resolution. Journal of Semantics 9:333-378.

Pragmatics

See also

Additional links

References

Further Readings