Copyright © 1999 Massachusetts Institute of Technology
The content of most linguistic expressions in one way or other depends on the context of utterance; in fact, in logical semantics (literal) meaning is analyzed as a function assigning contents (or intensions) to contexts. The most prominent way for the context to determine content is by way of INDEXICALS, that is, by expressions whose sole function is to contribute a component of the situation in which an utterance is made to the content expressed by that utterance. Indexicals can be either lexical items, such as the English personal pronoun I, which always refers to the speaker, or grammatical forms, such as the first person verbal suffix in Latin, which has the same function. Indexicals are special cases of deictic expressions whose reference depends on context -- a case in point being the possessive pronoun my, which describes something as belonging to the speaker as determined by the utterance situation (Zimmermann 1995).
All languages seem to contain deictic expressions and to make ample use of them. Traditionally, deictics are classified according to the aspect or feature of the utterance context that determines their reference. Three major kinds of deixis are usually distinguished: (1) person deixis, where the context provides one or more participants of the conversation (speaker, addressee), or a group to which they belong; (2) spatial deixis, where the context provides a location or a direction, especially as a reference point for spatial orientation on which other deictics depend (an "origo," in Bühler's 1934 sense); and (3) temporal deixis, with the context contributing a specific time, which may be the time of utterance, the time of a reported event, or the like. Among the clear cases of deixis in English are (1) the first and second person pronouns, I, you, we; (2) the local adverbs here and there; and (3) the temporal adverbs now and yesterday. Other examples of deixis, including demonstratives such as this whose referents depend on an accompanying gesture plus the speaker's referential intentions (Kaplan 1989a, 1989b), do not clearly fall under 1-3. DISCOURSE anaphors, such as aforementioned, or third person pronouns, such as he, her, receive their interpretation by reference to their linguistic context and are thus sometimes also considered as deictic. As has become clear by work in DYNAMIC SEMANTICS, however, such anaphoric elements quite regularly undergo variable binding processes, as in Every man who owns a donkey beats it, resulting in a quantified, rather than context-dependent reading. The same has been observed for the context dependence of certain relational nouns (Partee 1989) such as enemy whose arguments are usually given by the utterance situation, as in The enemy is approaching, but may also be quantified over, as in Every participant faced an enemy.
Languages differ considerably in the number and kinds of deictic locutions they have. Some have the place of utterance as their only spatial parameter, where others have complex systems classifying space according to various criteria (Frei 1944), including distance from the speaker's position as measured in varying degrees (up to seven in Malagasy, according to Anderson and Keenan 1985), visibility (also in Malagasy, and in many other languages), and perspective (as in English left and right). Person deixis, too, is subject to variation: many languages (e.g., French) distinguish more than one second person, depending on the social relationship between speaker and hearer, a phenomenon known as "social deixis" (Levinson 1979); another common distinction (to be found, for example, in Tagalog) is between an inclusive and an exclusive first person plural, depending on whether the addressee does or does not belong to the group designated. The variation in temporal deixis is harder to estimate, partly because it is not always clear whether tenses are truly deictic, but also because languages tend to be more or less localistic, extending their system of spatial deixis to time by analogy and frozen metaphor (cf. TENSE AND ASPECT).
The role of the context in providing the perspective from which the utterance is interpreted becomes particularly vivid in shifted contexts, also known as "relativized deixis" (Fillmore 1975), or "Deixis am Phantasma" (Bühler 1934), where at least some of the deictic parameters are not provided by the utterance situation. Among these shifts are various forms of pretense like play-acting, impersonation, analogous deixis (Klein 1978) -- the speaker of We took this road may refer to what is represented by the map in front of her -- or even first-person inscriptions on gravestones (Kratzer 1978).
Speakers often take the hearer's perspective in describing the spatial location of objects (Schober 1993), as in Please press the left button / the button to your left. This context shift can be made out of politeness or, especially when the hearer does not know the speaker's location, for communicative efficiency. In the latter case, however, deictic orientation may also be replaced by an intrinsic perspective provided by an object with a canonical front (as in behind the house, denoting the backyard).
A rather coherent area of regular context shift is known as "free indirect speech" in narrative analysis (Banfield 1982; Ehrlich 1990). In a passage such as Mary looked out of the window. Her husband was coming soon, the second sentence is understood to report Mary's thoughts from her own point of view: it is Mary, not the narrator, who believes her husband to be on his way and, whereas the verb come normally expresses movement toward the speaker as determined by the context, in this case it is Mary's position toward which her husband is reported to move (Rossdeutscher 1997). Similarly, the adverb soon is understood to describe an event as happening shortly after the scene described, rather than the utterance, has taken place. This simultaneous replacement of some (but not all) contextual parameters can be seen as a shift of the logophoric center (Kuno 1987), which comprises a large part of the more subjective parameters, including those that determine the interpretation of evaluative adjectives (e.g., boring) and free reflexives. Whereas free indirect speech is a rather well understood phenomenon with predictable features (including a restricted choice of tenses), other shifts of the logophoric center are less easily accounted for. Among these are the optional perspectives in overt attitude reports (Mitchell 1987), as in The CIA agent knows that John thinks that a KGB agent lives across the street, where the underlined phrase can be evaluated from the speaker's, the CIA agent's, or John's point of view.
Anderson, S. R., and E. L. Keenan. (1985). Deixis. In T. Shopen, Ed., Language Typology and Syntactic Description. Vol. 2, Grammatical Categories and the Lexicon. Cambridge: Cambridge University Press, pp. 259-308.
Banfield, A. (1982). Unspeakable Sentences. London: Routledge.
Bühler, K. (1934). Sprachtheorie. Jena, Germany: Fischer.
Ehrlich, S. (1990). Point of View. London: Routledge.
Fillmore, C. (1975). Santa Cruz Lectures on Deixis. Bloomington: Indiana University Lingustics Club.
Frei, H. (1944). Systèmes de déictiques. Acta Linguistica 4:111-129.
Kaplan, D. (1989a). Demonstratives: An essay on the semantics, logic, metaphysics and epistemology of demonstratives and other indexicals. In J. Almog, J. Perry, and H. Wettstein, Eds., Themes from Kaplan. Oxford: Oxford University Press, pp. 481-563.
Kaplan, D. (1989b). Afterthoughts. In J. Almog, J. Perry, and H. Wettstein, Eds., Themes from Kaplan. Oxford: Oxford University Press, pp. 565-614.
Klein, W. (1978). Wo ist hier? Präliminarien zu einer Untersuchung der lokalen Deixis. Linguistische Berichte 58:18-40.
Kratzer, A. (1978). Semantik der Rede: Kontexttheorie -- Modalwörter -- Konditionalsätze. Königstein, Germany: Scriptor.
Kuno, S. (1987). Functional Syntax. Chicago: University of Chicago Press.
Levinson, S. C. (1979). Pragmatics and social deixis: Reclaiming the notion of conventional implicature. In Proceedings of the Fifth Annual Meeting of the Berkeley Linguistics Society, Berkeley, CA: pp. 206-223.
Mitchell, J. (1987). The Formal Semantics of Point of View. Ph.D. diss., University of Massachusetts at Amherst.
Partee, B. (1989). Binding implicit variables in quantified contexts. In C. Wiltshire, R. Graczyk, and B. Music, Eds., CLS 25. Part One: The General Session, Chicago, pp. 342-365.
Rossdeutscher, A. (1997). Perspektive und propositionale Einstellung in der Semantik von kommen. In C. Umbach, M. Grabski, and R. Hörnig, Eds., Perspektive in Sprache und Raum. Wiesbaden: Deutscher Universitätsverlag, pp. 261-288.
Schober, M. F. (1993). Spatial perspective-taking in conversation. Cognition 47:1-24.
Zimmermann, T. E. (1995). Tertiumne datur? Possessivpronomina und die Zweiteilung des Lexikons. Zeitschrift für Sprachwissenschaft 14:54-71. Translated as Tertiumne datur? Possessive pronouns and the bipartition of the lexicon. In H. Kamp and B. Partee, Eds., Context Dependence in the Analysis of Linguistic Meaning. Vol. 1, Papers. Stuttgart, Germany: Institut für maschinelle Sprachverarbeitung, Stuttgart University, 1997, pp. 409-425.
Doron, E. (1990). Point of View. CSLI Report 90-143, Stanford University.
Ehrich, V. (1992). Hier und Jetzt: Studien zur lokalen und temporalen Deixis im Deutschen. Tübingen, Germany: Niemeyer.
Forbes, G. (1988). Indexicals. In D. Gabbay and F. Guenthner, Eds., Handbook of Philosophical Logic, vol. 4. Dordrecht: Kluwer, pp. 463-490.
Jarvella, R. J., and W. Klein, Eds. (1982). Speech, Place, and Action: Studies of Deixis and Related Topics. Chichester: Wiley.
Zimmermann, T. E. (1991). Kontextabhängigkeit. In A. v. Stechow and D. Wunderlich, Eds., Semantik (Semantics). Berlin/New York: Springer, pp. 156-229.
Copyright © 1999 Massachusetts Institute of Technology