Discourse

Discourse is the ground of our experience of language and of linguistic meaning. It is in discourse that we learn language as children, and in discourse that we most adequately convey our thought. The individual utterances in a discourse are notoriously vague and full of potential AMBIGUITY. Yet in the context of the discourse in which they occur, vagueness and ambiguity are rarely a problem. That is, the overall discourse profoundly influences the interpretation of individual linguistic constituents within it, as witnessed by our discomfort with the ethics of taking what someone says out of context.

Discourse can be characterized in three principal ways. We may think of discourse as a type of event, in which human agents engage in a verbal exchange; in the limit case, the monologue, there is only one agent, but even then there is an intended audience, if only reflexive or imaginary. We may also think of discourse as the linguistic content of that exchange, an ordered string of words with their associated syntactic and prosodic structures. Or we may characterize discourse as the more complex structure of information that is presupposed and/or conveyed by the interlocutors during the course of the discourse event in view of the explicit linguistic content of the exchange. The information structure of a discourse may be characterized as an ordered set containing several distinguished kinds of information, for example: a set of discourse participants (the interlocutors); the linguistic content of the discourse, with each utterance indexed by the speaker; the information presupposed or proffered by speakers during the discourse via the linguistic content of their utterances; an association of the proffered information with various topics or questions under discussion, the topics and subtopics hierarchically organized by virtue of their relations to the (often only inferred) goals and intentions of the various speakers (the intentional structure of the discourse); a set of entities discussed (the domain of discourse); a changing set of the entities and topics which the interlocutors focus on during their discussion, organized as a function of time (the attentional structure of discourse); and other kinds of information and structures on information, as well (Lewis 1979; Grosz and Sidner 1986; Roberts 1996). The information structure of a discourse is far richer than its linguistic content alone. Only the full range of contextual information for a given discourse, including a grasp of the interlocutors' inferred intentions, the intended rhetorical relations among their utterances, and other nonlinguistically given information that they share, can resolve all the potential ambiguities in the linguistic strings uttered, clarify the often inexplicit connections between utterances, and lead us to grasp what the speaker(s) intend to convey.

These three ways of characterizing discourse -- as an event revolving around verbal exchange, as the linguistic content of that exchange, and as the structure on the information involved -- are not mutually exclusive; there is no verbal exchange without linguistic content, and the latter can be taken as one aspect of the abstract information structure of the exchange. However, most of the work on discourse in artificial intelligence, linguistics, philosophy of language, PSYCHOLINGUISTICS, sociology, and anthropology can be classified according to which of the three aspects it focuses on. For example, sociologists, sociolinguists, and ethnographers interested in conversational analysis (Sacks, Schegloff, and Jefferson 1978) focus on the discourse event itself and its social character, including the way that interlocutors organize their participation in such events in an orderly, apparently conventional manner, varying somewhat from culture to culture. Speakers take turns, the opportunity for taking a turn being cued by a number of conventional means (including set phrases, intonation, and pauses), and negotiate the beginning and end of a discourse as well as the shift from topic to topic within it. In sociologically informed anthropological linguistics (Duranti 1997), discourse events are taken to play a crucial role in the creation, reproduction, and legitimation of a community's social alliances and cleavages. Those working in the tradition of discourse analysis (see van Dijk 1985; Carlson 1983) focus instead on the linguistic content of the verbal exchange, the text, some arguing that it is generated by the syntactic rules of a text grammar. But probably the majority of theorists who work on discourse today would agree that discourse is not a linguistic structure generated by a grammar, but instead is structured by nonlinguistic, logical, and intentional factors -- aspects of what we have called the information structure of discourse.

A number of prima facie unrelated pragmatic phenomena depend on the information structure of discourse, suggesting that this aspect of discourse can provide the basis for a unified theory of the role of pragmatic factors in linguistic interpretation. Dynamic theories of semantic interpretation take the meaning of an utterance to be a function from contexts (the context of utterance) to contexts (the context of utterance updated with the information proffered in the utterance). One can view the three basic types of speech acts -- assertions, questions, and orders -- as functions that update the information structure of a discourse in different ways. Assertions update the information proffered in the discourse (see DYNAMIC SEMANTICS; Stalnaker 1979); a question sets up a new (sub)topic for discussion, hence affecting the intentional structure of the discourse (Ginzburg 1996; Roberts 1996); an order, if accepted, commits the person ordered to behave in accordance with the order, and this commitment is part of the information indirectly conveyed by the discourse. Many secondary subtypes of these three basic types of speech acts have been proposed in the literature, including, for example, predictions, confirmations, concessives, and promises; rhetorical questions as well as probes for information; requests, permissions, advisories, and commands; and so forth (see Searle 1969). Work in artificial intelligence on plan inference (see Cohen, Morgan, and Pollack 1990) has argued that the secondary speech act type of an utterance can be derived from its basic speech act type and proffered content, its inferred role in the intentional structure of the discourse, and the inferred domain goals of the interlocutor at the time of utterance (i.e., her general goals at that time, not necessarily just those expressed in discourse).

In his important work on meaning in conversation, H. Paul GRICE argued that much of the MEANING conveyed in discourse is not directly proffered via the linguistic content of the utterances in the discourse, but instead follows from what is said in view of the intentions of the interlocutors and a set of guidelines, the conversational maxims, which characterize the rational behavior of agents in a communicative situation (see IMPLICATURE and RELEVANCE). Although the crucial role of interlocutors' intentions is sometimes overlooked in older work on implicature, contemporary work (e.g., McCafferty 1987; Welker 1994) pays considerable attention to the role of that facet of the information structure of discourse that we have called its intentional structure in explaining how implicatures are generated. Similarly, work on ANAPHORA in discourse, especially from a computational point of view (e.g., Grosz and Sidner 1986), emphasizes the role both of intentional structure and of attentional structure in constraining the set of possible antecedents for a given pronoun or other anaphoric element across discourse; see also the related work on centering (Walker, Joshi, and Prince 1998). And the intentional structure of discourse may be seen to reflect strategies of inquiry which correspond to classical rhetorical structures (Mann and Thompson 1987), connecting work on the role of such structures in interpretation to the general notion of information structure.

Finally, the information structure of discourse is reflected in a number of ways in the structure of linguistic constituents -- sentences and sentence fragments. These reflections involve the phenomena variously referred to as topic and comment, theme and rheme, link, and focus, among others. Some have argued that (some subset of) these notions play a role as primitive notions in syntactic structure (e.g., Sgall, Hajicova, and Panevová 1986; Vallduví 1992), while others have argued that they are, instead, only functional characterizations of the accidental role of syntactic constituents in particular discourse contexts. However, it is clear that particular sentential constructions (e.g., topicalization in English) and prosodic features (e.g., prosodic prominence) may be specially associated with these functions via associated conventional presuppositions about the information structure of the discourse in which they occur. Such linguistic structures, along with anaphora and ellipsis, are designed to increase discourse coherence and to help interlocutors keep track of the common ground and other features of the information structure of discourse.

Additional links

-- Craige Roberts

References

Carlson, L. (1983). Dialogue Games: An Approach to Discourse Analysis. Dordrecht: Reidel.

Cohen, P., J. Morgan, and M. Pollack. (1990). Intentions in Communication. Cambridge, MA: Bradford Books/ MIT Press.

Duranti, A. (1997). Linguistic Anthropology. New York: Cambridge University Press.

Ginzburg, J. (1996). The semantics of interrogatives. In S. Lappin, Ed., The Handbook of Contemporary Semantic Theory, pp. 385-422. Oxford: Blackwell.

Grosz, B., and C. Sidner. (1986). Attention, intentions, and the structure of discourse. Computational Linguistics 12:175-204.

Lewis, D. (1979). Score-keeping in a language game. In R. Bauerle, U. Egli, and A. von Stechow, Eds., Semantics from a Different Point of View. Berlin: Springer.

Mann, W. C., and S. A. Thompson. (1987). Rhetorical Structure Theory: A Theory of Text Organization. Information Sciences Institute, University of Southern California.

McCafferty, A. (1987). Reasoning about Implicature. Ph.D. diss. University of Pittsburgh.

Roberts, C. (1996). Information structure: Towards an integrated theory of formal pragmatics. In Y.-H. Yoon and A. Kathol, Eds., OSU Working Papers in Linguistics, vol. 49: Papers in Semantics. Ohio State University Department of Linguistics.

Sacks, H., E. A. Schegloff, and G. Jefferson. (1978). A simplest systematics for the organization of turn-taking in conversation. In J. Schenkein, Ed., Studies in the Organization of Conversational Interaction. New York: Academic Press, pp. 7-55.

Searle, J. R. (1969). Speech Acts. Cambridge: Cambridge University Press.

Sgall, P., E. Hajicova, and J. Panevova. (1986). The Meaning of Sentence in its Semantic and Pragmatic Aspect. Dordrecht: Reidel.

Stalnaker, R. (1979). Assertion. In P. Cole, Ed., Syntax and Semantics. New York: Academic Press.

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

van Dijk, T. (1985). Handbook of Discourse Analysis, vol. 3: Discourse and Dialogue. London: Academic Press.

Walker, M., A. Joshi, and E. Prince, Eds. (1998). Centering Theory in Discourse. Oxford: Oxford University Press.

Welker, K. (1994). Plans in the Common Ground: Toward a Generative Account of Implicature. Ph.D. diss. Ohio State University.

Discourse

See also

Additional links

References

Further Readings