Definite descriptions by Paul Elbourne

Abstract

Reviewed by: Definite descriptions by Paul Elbourne Barbara Abbott Definite descriptions. By Paul Elbourne. (Oxford studies in semantics and pragmatics 1.) Oxford: Oxford University Press, 2013. Pp. 251. ISBN 9780199660209. $50. This book presents a development and refinement of the analysis of definite descriptions (and pronouns) that appears in Elbourne’s (2005) book Situations and individuals. That analysis is Fregean, according to which definite descriptions are referential expressions that presuppose the existence and uniqueness of a referent. The main foci of the book are reviewing and adding arguments in favor of the analysis, and addressing problems that have been raised against it. The main competitor is, of course, the Russellian, quantificational analysis, and in addition to arguing for the Fregean approach, E presents problems for this competitor. Ch. 1 (‘Introduction’, 1–16) briefly reviews not two but six analyses of definite descriptions: (i) Russell’s (1905); (ii) Donnellan’s (1966), according to which descriptions are ambiguous between a quantificational and a referential interpretation; (iii) Frege’s (1892); (iv) Graff (Fara)’s (2001) predicate analysis; (v) Heim’s (1982) familiarity view; and (vi) Szabó’s (2000) existential theory. E notes three prominent problems for the Fregean view (true negative existence statements, presupposition failures that result in falsity rather than lack of a truth value, and true sentences about propositional attitudes toward nonexistent entities) that are addressed later in the book, and then briefly argues against analyses (iv)–(vi) above. Ch. 2 (‘Situation semantics’, 17–41) gives a detailed formal presentation of the variety of situation semantics to be presupposed. This is important, because a crucial part of E’s analysis is the inclusion of situation variables in definite descriptions. Situations, following Barwise & Perry 1983, are entities possessing properties at spatiotemporal locations. (One small issue crops up occasionally: whether including a location excludes abstract entities.) Sentence meanings are sets of situations, here called ‘Kripkean propositions’. The actual utterance of a sentence on a particular occasion results in an Austinian proposition: ‘[t]he pair of a sentence meaning and the topic situation that it is meant to describe’ (18). Interestingly, E notes that he does not actually believe in the existence of nonactual situations, holding instead that ‘meanings are internal mental structures’ (19). Ch. 3 (‘The definite article’, 42–51) gives an analysis that, as E notes, is based on one given by Heim and Kratzer (1998). E’s version is below (47). (1). This makes the definite article a function from properties (expressed by the NP with which the combines) to functions from situations to entities—those that uniquely possess the property in the situation in question. Definite descriptions have the structure shown in 2. (2). [[the NP] s] Crucially, though, any particular definite description has the possibility of receiving two analyses, depending on whether the situation variable in 2 (which E refers to as a ‘situation pronoun’) is bound. Ch. 4 (‘Presupposition’, 52–103) is an extended discussion of how the presuppositions associated with definite descriptions appear or fail to appear in the contexts associated with modals, conditionals, disjunctions (assumed here to be equivalent to conditionals), negation, and sometimes just ordinary sentences. In the case of modals, the requirement of a unique referent for a definite description must be satisfied relative to the situation from which the alternative possible worlds are accessible. Similarly for conditionals, except those whose antecedents actually assert the existence and uniqueness of a referent for a definite description occurring in the consequent. In that case, the presupposition does not project. When we come to negation, presuppositions generally project, with the exception of true negative existentials. In this case E proposes implicit [End Page 762] modalization, so that the existence and uniqueness of the entity in question is satisfied relative to imaginary situations, and such an entity is asserted not to exist in the Austinian topic situation (typically one in the actual world). The longest discussion is devoted to ‘presupposition obviation’—cases like Strawson’s (1954:226) example of someone declaring that they had had lunch with the King of France, which we would judge to be false, rather than truth-valueless. Ch. 5 (‘Referential and attributive’, 104–19) shows how E’s analysis can provide...