|
<the, a>: (In)definiteness and implicature Laurence Horn and Barbara Abbott In this paper we review some recent accounts of the contrast between the definite and indefinite articles of English, which (i) attempt to analyze the two articles as identical semantically, differing only in their pragmatics and (ii) analyze this (putatively pragmatic) difference as one of familiarity vs. novelty (à la Heim 1982). We argue that such accounts are not successful, and replace them with a revision of Russell’s analysis, according to which the difference is semantic and crucially involves uniqueness. Szabó (2000, 2003) and Ludlow & Segal (2004) assign to both indefinite and definite descriptions the logical form of the Russellian (1905) pure existential analysis of indefinites. While the Szabó and Ludlow & Segal treatments differ in the details, essentially Bertrand found a solution and Bertrand found the solution are both assigned the logical form ($x)(x is a solution & Bertrand found x); the difference in import between the indefinite and definite descriptions is taken to be “pragmatic”, via conversational implicature. But without a conventional distinction between indefinites and definites, there is nothing from which to derive the posited pragmatic distinction. A more plausible position is that the definite article contributes a conventional implicature, as Ludlow & Segal suggest in a footnote, although we argue that they err in associating this implicature with familiarity or givenness. On our account, there is indeed a conventional (encoded) distinction between indefinite and definite, in that the latter is marked for uniqueness or maximality (from which familiarity is pragmatically derivable in relevant contexts), while the former is unmarked for this feature. The relationship between the and a is thus a privative dyad of the form <S, W> along the lines of e.g. <thumb, finger>, <square, rectangle>, or <himself, him>, in which S (the stronger scalar value) is marked for a feature with respect to which W (the weaker) is unmarked, and is hence more informative, whence the implicature from the use of W (a solution) that the speaker was not in a position to have used the corresponding S value (the solution). Building on Abbott (2003, 2008) and Horn (2007), we follow Hawkins (1991) in invoking a <the, a> scale and in taking non-uniqueness to constitute an upper-bounding Q-based conversational implicature (cf. Horn 1972, 1989) and provide additional empirical evidence for the scalar nature of this relation, via such attested data as (1) Yet time and again, North
Korea is cited as not only "a" but "the" major to US security. (Note also that the contrasts above clearly involve uniqueness, not familiarity.) But wedepart from Hawkins in positing uniqueness or maximality as conventionally implicated rather than asserted or entailed in definites. For Hawkins, “The F is G” entails that there is only one (salient) F, assumed by the speaker to be identifiable by the hearer. This is indeed part of the conventional contribution of the (contra Szabó and Ludlow & Segal)—but is it part of the truth-conditional meaning of the sentence, or just an appropriateness condition on its assertion? The main argument for a conventional implicature approach is that there is no clear evidence that a statement containing a definite description is ever taken to be false on the grounds that uniqueness/maximality is violated, in the way that it may be when the existential premise fails. In addition—as exemplified in contexts like (2) Whaddaya mean, she’s poor
BUT honest? —objections to statements on the grounds that a contained definite description fails to apply uniquely (as opposed to objections because the existence clause is not satisfied) typically involve the metalinguistic devices characteristically employed to reject conventional implicatures rather than ordinary negation functioning as a falsity operator. Our account also touches on the apparent—but, we argue, not actual—problems for a conventional account of uniqueness posed by “weak definites” that do not implicate that the hearer can uniquely a suitable referent (Carlson et al. 2006), as in (3a) and (4a). (3) a. You need to see the
doctor. [“fake definite” reading available] References Abbott, Barbara (2003) A reply to Szabó’s “Descriptions and uniqueness.” Philosophical Studies 113: 223-31._____ (2008). Issues in the semantics and pragmatics of definite descriptions in English. In J. Gundel & N. Hedberg (eds .), Interdisciplinary Perspectives on Reference Processing, 61-72. Oxford: Oxford U. Press.Carlson, Greg, Rachel Sussman, Natalie Klein and Michael Tanenhaus (2006). Weak definite noun phrases. NELS 36.Grice, H. P. (1989). Studies in the Way of Words. Cambridge: Harvard U. Press. Hawkins, John (1991). On (in)definite articles: implicatures and (un)grammaticality prediction. Journal of Linguistics 27: 405-42. Heim, Irene (1982). The semantics of definite and indefinite noun phrases. University of Massachusetts PhD dissertation. Horn, Laurence (1972). On the Semantic Properties of Logical Operators in English. UCLA dissertation, reprinted by IULC, 1976. _____ (1989). A Natural History of Negation. Chicago: U. of Chicago Press. _____ (2007). Toward a Fregean pragmatics: Voraussetzung, Nebengedanke, Andeutung. In I. Kecskes & L. Horn (eds.), Explorations in Pragmatics, 39-69. Berlin: Mouton de Gruyter. Ludlow, Peter and Gabriel Segal (2004). On a unitary semantical analysis for definite and indefinite descriptions. In M. Reimer & A. Bezuidenhout (eds.), Descriptions and Beyond, 420-36. Oxford: Clarendon Press. Russell, Bertrand (1905). On denoting. Mind 14: 479-93.Szabó, Zoltán (2000). Descriptions and uniqueness. Philosophical Studies 101: 29-57._____ (2003). Definite descriptions without uniqueness: A reply to Abbott. Philosophical Studies 114: 279-91. .
|