DEMONSTRATIVES

Demonstratives are one type of indexical. Like other indexicals, demonstratives can be used to refer to different objects on different occasions. Some examples of demonstratives are that, this, you, he, she, there, then, this dog and that yellow house.

Indexicals and Demonstratives

Philosophers of language commonly distinguish between the meaning of a linguistic expression and its referent. For example, the definite descriptions the president of the United States in 2003 and the husband of Laura Bush in 2003 refer to the same individual (namely, George W. Bush), but differ in meaning. Indexicals (also known as context-sensitive expressions ) lead many philosophers to distinguish between two different sorts of meaning. Consider the paradigm indexical I and suppose that Al and Bob both utter the sentence I live in Chicago. Their utterances of I have the same meaning, in one sense of meaning. Let us call the type of meaning that their utterances share linguistic meaning. But there are reasons to think that their utterances also differ in some other type of meaning. Al's utterance of I refers to Al, whereas Bob's utterance refers to Bob. Al and Bob also say different things: Al says that Al lives in Chicago, whereas Bob says that Bob does. Finally, one of their utterances may be true while the other is false. Let us call the type of meaning that their utterances do not share content. All utterances of I have the same linguistic meaning, but utterances that are produced by different speakers have different contents. Similarly for the sentence I live in Chicago.

A speaker's utterance of I refers to that speaker no matter what object he might want to refer to, even if he intends to refer to Napoleon Bonaparte as he utters I (because he mistakenly thinks that he is Napoleon) and even if he points at someone else as he produces his utterance. By contrast, the referent of a speaker's utterance of he depends on the speaker's intentions or pointing gestures. If Al intends to refer to George W. Bush, and points at Bush as he utters He is a Republican, then his utterance of he refers to Bush. If Al instead intends to refer to Bill Clinton, and points at Clinton, then his utterance of he refers to Clinton. Indexicals, such as he, whose reference and content depend on the actions or intentions of the speaker are commonly called demonstratives. Those that do not, such as I, are often called pure indexicals. David Kaplan (1989a, 1989b), whose work on indexicals is highly influential, claims that he, she, this, and that are demonstratives, whereas I, today, tomorrow, yesterday, now, and here are pure indexicals.

The pronouns he and she are often used as demonstratives, but they also have nondemonstrative uses. For instance, the pronoun he is used in roughly the same way as a bound variable of formal logic in the sentence Every man thinks that he is handsome (on the interpretation "for every man x, x thinks that x is handsome"). The pronoun she is used as an anaphor in Fiona sat down. She picked up a pencil. It is controversial how these uses of he and she are related to their demonstrative uses.

Reference-Fixing for Demonstratives

We previously observed that the reference of a demonstrative utterance depends on the speaker's intentions or pointing gestures. Kaplan's early work (1989b) tends to emphasize the importance of pointing gestures in determining reference. In this early work Kaplan says that an utterance of a demonstrative is typically accompanied by a demonstration, which is a public presentation of an object that is typically, though not always, a pointing gesture. The demonstration determines a demonstrated object (a demonstratum ) in a context, and the demonstratum is the referent of the demonstrative, in the context. Kaplan's later work (1989b) tends to emphasize the role of speakers' intentions. According to it demonstrations are directed towards objects by directing intentions and it is directing intentions that determine the referents of demonstrative utterances.

One difficulty for the view that pointing gestures determine reference is that not all utterances of demonstratives are accompanied by pointing gestures. Kaplan's early theory allows there to be demonstrations that are not pointing gestures, but unfortunately leaves unclear what demonstrations (in general) are. A problem for the view that directing intentions determine reference is that it is not clear what directing intentions are. Speakers typically have many different intentions when they utter demonstratives. When Gail utters he, she may simultaneously (1) intend to refer to Hal, (2) intend to refer to the man she sees, and (3) intend to refers to the man to whom others are referring with he. But these intentions may conflict, and it is unclear which of them (if any) is a directing intention.

Kaplan's Semantics for Pure Indexicals

Kaplan (1989b) presents a semantics for indexicals that attempts to describe their various meanings. We shall first consider how his theory works with pure indexicals, and then consider how to extend it to simple demonstratives. (The following text concentrates on Kaplan's informal remarks about the semantics of indexicals. His formal logical system uses the apparatus of possible-worlds semantics.)

Kaplan's theory begins with the idea that a linguistic expression has a content with respect to, or in, a context of utterance. For instance, the word I has a content in every context, depending on who the agent of the context is. For every context C, there is an agent of C, a location of C, a time (or day) of C, and a possible world of C. The content of the word I in any context C is the agent of C, the content of here at C is the location of C, and the content of now and today at C is the time (day) of C. The linguistic meaning, or character, of an expression is a function whose value at any context C is its content in C. For instance, the character of I is a function on contexts whose value at any context C is the agent of C.

The content of a declarative sentence in a context is a proposition, which is an entity that can be an object of attitudes such as belief, doubt, and assertion. When a speaker assertively utters a sentence, she asserts the proposition that is the content of her sentence in her context, and if she is sincere, then she believes that proposition. On Kaplan's (informal) semantics, propositions have constituent structures that resemble the constituent structures of sentences: Just as sentences have words as parts or constituents, so propositions have individuals, properties, and relations as parts or constituents. If the content of sentence S in context C is proposition P, then the constituents of P are (roughly) the contents, in C, of the words in S. For example, suppose that C* is a context in which Inga is the agent. Then the content of I laugh with respect to C* is a proposition whose constituents are Inga and the property of laughing. If Inga laughs in the possible world of context C*, then I laugh is true in context C*.

The sentence I am speaking is false in some contexts, according to Kaplan, because there are contexts in which the agent is not speaking. However, every agent of every context exists in the possible world of that context. Therefore, I exist is true in every context. Thus, Kaplan (1989b) claims that I exist is a logical truth. But the content of I exist in a context is (usually) not a necessary truth. For example, the content of I exist with respect to context C* above is the proposition that Inga exists. This is not a necessary proposition. Therefore on Kaplan's theory, the sentence I exist is a logical truth that fails to express a necessary truth in many contexts.

Extending Kaplan's Theory to Simple Demonstratives

One way to extend Kaplan's (1989a, 1989b) theory to simple demonstratives involves adding more items to contexts. (A second way, which will not be discussed here, involves Kaplan's dthat-terms.) For instance, one can suppose that every context has a sequence of demonstrata, and that the content of that₁ with respect to context C is the first demonstratum of C, the content of that₂ is the second demonstratum of C, and so on. One can then say that the character of that₁ is a function whose value at every context C is the first demonstratum of C, the character of that₂ is a function whose value at every context C is the second demonstratum of C, and so on. Similarly, one can suppose that every context has a sequence of addressees and that the content of you ₁ is the first addressee, the content of you ₂ is the second addressee, and so on.

There are two difficulties with this extension. First, We saw earlier that the referent of a demonstrative in a context is determined, somehow, by the pointing gestures and intentions of the speaker. The preceding theory assumes that every context has a sequence of demonstrata. But as Kaplan (1989b) points out, the agents of many contexts are not pointing at any objects and do not have any intentions that are relevant to determining a referent for a demonstrative. So it is highly artificial to suppose that every context contains a sequence of demonstrata. Second, the English word that is a single linguistic expression with a single linguistic meaning. But the previous theory instead provides an infinite number of distinct subscripted demonstratives (that ₁, that ₂, … ), each with its own character (Braun 1994).

Belief and Demonstratives

Propositions that have individuals as constituents are known as singular propositions. Kaplan's theory says that singular propositions can be asserted and believed. This claim is problematic. Suppose that John is wearing a shirt with a large stain on its back. Suppose that he sees the back of his shirt in a mirror, but does not realize that he is viewing his own shirt. Then he may sincerely say I am wearing a clean shirt and he is not wearing a clean shirt, as he points at the person reflected in the mirror. On Kaplan's theory, John asserts and believes the contradictory singular proposition that John is wearing a clean shirt and John is not wearing a clean shirt. But surely he does not believe a contradictory proposition.

Gottlob Frege gives similar reasons for thinking that the content of a proper name is not its referent, but is instead a sense that determines a referent. Kaplan (1989b) and John Perry (2000) respond to Fregean criticisms of Kaplan's theory and criticize Frege's theory of demonstratives.

Complex Demonstratives

Complex demonstratives are expressions of the form that CN, where CN is a common noun phrase. Examples are that car, that man who is wearing a baseball hat, and that yellow house. Kaplan's theory does not mention complex demonstratives, and it is not entirely clear how they should be integrated into a theory of demonstratives. The major issue concerns the property expressed by the common noun phrase inside a complex demonstrative. Is this property a part of the content of the entire complex demonstrative? For instance, does the content of an utterance of that yellow house include the property of being yellow? Or, alternatively, is the content of the complex demonstrative simply the object to which the utterance refers?

On the one hand, it seems that a speaker's intentions and demonstrations are relevant to determining the referent of that yellow house. In this respect, that yellow house resembles the simple demonstrative that. We concluded earlier that the content of a simple demonstrative is just its referent. Therefore, We have some reason to think that the content of a complex demonstrative is also its referent and to think that the property of being yellow is not a part of the content of that yellow house. David Braun (1994) and Nathan Salmon (2002) argue for this view of complex demonstratives. On the other hand, the complex demonstrative that yellow house has a syntactic form much like the syntactic forms of the definite description the yellow house and the quantifier phrase some yellow house. Most philosophers think that the contents of the yellow house and some yellow house include the property of being yellow. That is some reason to think that the content of that yellow house also includes the property of being yellow. Jeffrey C. King (2001) argues for this latter view.

See also Anaphora; Frege, Gottlob; Indexicals; Logical Form; Meaning; Modality and Quantification; Proper Names and Descriptions; Propositional Attitudes; Propositions; Sense.

Bibliography

Braun, David. "Demonstratives and Their Linguistic Meanings." Noûs 30 (1996): 145–173.

Braun, David. "Indexicals." In Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. Stanford, CA: Metaphysics Research Lab, Center for the Study of Language and Information, Stanford University, 2001.

Braun, David. "Structured Characters and Complex Demonstratives." Philosophical Studies 74 (1994): 193–219.

Kaplan, David. "Afterthoughts." In Themes from Kaplan, edited by Joseph Almog, John Perry, and Howard Wettstein, 565–614. New York: Oxford University Press, 1989a.

Kaplan, David. "Demonstratives." In Themes from Kaplan, edited by Joseph Almog, John Perry, and Howard Wettstein, 481–563. New York: Oxford University Press, 1989b.

King, Jeffrey C. Complex Demonstratives: A Quantificational Account. Cambridge: MIT Press, 2001.

Perry, John. The Problem of the Essential Indexical and Other Essays. Expanded ed. Stanford, CA: CSLI Publications, 2000.

Salmon, Nathan. "Demonstrating and Necessity." Philosophical Review 111 (3) (2002): 497–537.

David Braun (2005)