Page 247 - Encyclopedia of Philosophy of Language

Necessarily p, because logical truth applies only to all proper points of reference, and he also allows for structures with improper points of reference).
In an early paper, 'Dthat' (1970, repr. in Yourgrau 1990), Kaplan recognized a demonstrative use of defi- nite descriptions following Donnellan's distinction between referential and attributive use, and intro- duced a special demonstrative dthat to capture this use. This operator turns a description (or an arbitrary singular term) into a directly referring expression. The singular term then functions as the demonstration of the demonstrative dthat, so that its demonstratum is the denotation of that singular term in that context. This makes a = dthat [a] logically true for any singular term a, but not Necessarily (a = dthat [a]), for given a
context, descriptive terms may under different cir- cumstances refer to different objects, while the referent of dthat [a] remains fixed. We can think of dthat as an intensional operator, one that rigidifies any singular term, or alternatively as an incomplete demonstrative, whose associated demonstration is not itself part of the content. Kaplan allows for intensional operators, which operate on content, but not for terms that oper- ate on character (such terms, which could easily be introduced into his logic, he calls 'monsters'); for otherwise, a sentence like (4)
In some contexts it is true that I am not tired now (4)
would be true in a context c0 if some speaker is not tiredinsomecontextc,.Butthen/in(4)hasnothing to do with me, the speaker in c0, which violates its directly referential character. According to Kaplan, the only way to control the character of a term is to use quotation marks.
3.2.2 Applications: Problemsand Prospects
Kaplan's theory has interesting epistemological conse- quences. First, we can now distinguish logical truth (or truth in all contexts of utterance, which is a matter of character) from necessary truth, which is truth in all circumstances and thus a matter of content. 'I am here now' is analytically true as I could not have uttered it falsely, but not necessarily true: at this moment, I might just as well have been somewhere else.
Further, Kaplan identifies content with the object of thought (what Frege called the thought expressed), and character with cognitive significance. Therefore, identity claims such as dthat[a] = dthat [a] and dthat [a] = dthat[b] may express the same thought, whereas their cognitive significance differs. People may believe the former without therefore being committed to accept the latter as well. Character may be seen as a manner of presenting a content, but unlike a Fregean sense, the same manner of presentation associated with /will usually present different contents to differ- ent persons, namely themselves. This implies that cog- nitive states are context-sensitive: we may hold
different prepositional attitudes to the same content if it is presented under different characters. Again, however, this distinction cannot formally be made in Kaplan's logic (see also the papers by Salmon and Soames in Almog, et al. 1989; see Recanati (1993) for an application of the theory of direct reference to the problem of de re thoughts; for Fregean criticism of Perry and Kaplan, see Gareth Evans in Yourgrau
1990).
Another area where Kaplan's ideas can be fruitfully
applied is that of figurative language (see Leezenberg (1995) for an extensive treatment along Kaplanian lines). Josef Stern has claimed that metaphors are context-dependent in a way similar to indexicals. He postulates a 'metaphorical operator' Mthat analogous to Kaplan's dthat, which converts any literal expression p into a 'metaphorical expression' Mthat [p] of nonstable character, sensitive to a 'meta- phorically relevant feature of the context' (Stern 1985:695). Thus, Stern treats metaphor as the desta- bilization of character. However, for him, neither Mthat nor dthat is an intensional operator: they do not work on content, but on character and are thus, in Kaplan's vocabulary, 'monsters.'
The same applies to analyses of context-dependent expressions other than those referring to individuals (be those individualspersons or places). Kaplan does not treat demonstratives involving properties or plural referents, like these books, this big, and thus; but other expressions are even more problematic. Bartsch (1987) calls attention to the context-dependence of property expressions like good. She calls these thematically weakly determined: they require a knowledge of the 'thematic dimension' of the context, i.e., the kind of property an utterance is about: in saying John is good, do we speak of his moral qualities, his tennis skill or something else? This analysis also allows for monsters: for one can supply a thematic dimension for John is good by adding a predicate-limiting adverbial like with respect to tennis playing or morally. Such adverbials, however, may shift the contextually given dimension, and thus are monstrous operators in Kaplan's sense. So apparently, we should distinguish between those context-dependent expressions that are directly ref- erential and those that are not. Kaplan's analysis still holds for indexicals like /, here, and this, but it does not apply to the latter ones, so perhaps we had better not call them 'indexicals' after all.
See also: Montague Grammar; Names and Descrip- tions; Reference: Philosophical Issues.
Bibliography
Almog J, Perry J, Wettstein H (eds.) 1989 Themes from Kaplan. Oxford University Press, Oxford
Bartsch R 1987 Context-dependent interpretations of lexical items. In: Groenendijk J, De Jongh D, Stokhof M (eds.)
Indexicals
225