-
3. Imposters as Antecedents
- The MIT Press
- Chapter
- Additional Information
3 Imposters as Antecedents 3.1 Some Basic Antecedence Facts As indicated in chapter 1, English singular imposters of all sorts invariably determine 3rd person verbal agreement (we discuss verbal agreement for plural and coordinated imposters in chapter 9), and most (but, for example, not yours truly) have the superficial morphology of standard 3rd person DPs. It is then natural to assume, as the Notional View would have it, that imposters are syntactically unexceptional 3rd person DPs tout court, which just have unexpected non-3rd person meanings or uses. This would mean that the imposter and nonimposter uses of this reporter would have identical syntactic structures. Consider how this conclusion would interact with the existence, illustrated in (1), of the syntactic relation between DPs that, following tradition, we call antecedence. We assume general understanding of antecedence here, but provide a richer and detailed account in chapter 4. Antecedence is the relation holding between the pairs of italicized phrases—obligatorily in (1a,c), as one possibility in (1b,d). (1) a. Mercedes embarrassed herself. b. Mercedes claimed she was uncomfortable. c. Mercedes and Ronald/The two lovers embarrassed themselves. d. Mercedes and Ronald/The two lovers claimed they were uncomfortable. The antecedence relation is of course commonly represented by coindexing, a matter discussed in chapter 4. There it is observed that in imposter cases in particular, coindexing fails to correspond to antecedence in key ways. A fundamental basis for this claim is that while antecedence is asymmetrical, coindexing is evidently a symmetrical relation. Nonetheless, as a purely notational device, it is convenient to represent antecedence relations with coindexing in the common way and so, while coindexing plays no theoretical role in this 16 Chapter 3 book, we still use the relevant notation without further comment whenever it appears useful. The Notional View has an obvious entailment. If imposters are syntactically unexceptional 3rd person DPs and if it is granted that pronominal antecedence is a syntactic relation, the antecedence behavior of imposters should invariably be just that of standard 3rd person DPs.And the syntactic nature of antecedence seems to be widely assumed in modern linguistics. A not infrequently invoked alternative that also has traditional roots is that pronominal antecedence is at least in part semantically driven. For the present discussion, we assume without argument the syntactic character of antecedence; but we return to the question in chapter 8. Well-supported principles require pronominal forms to agree with their antecedents in person, number, gender, familiarity, and perhaps other features (hereafter, ϕ-features). Various formulations are shown in (2) (see also Heim 2008, 50, (45), and the quotation from Curme in (1) of chapter 4). (2) a. Sag, Wasow, and Bender 2003, 208 The Pronominal Agreement Condition (PAC) “Coindexed NPs agree.” b. Lasnik and Uriagereka 1988, 47 “An anaphor must agree in syntactic features with its antecedent.” c. Carnie 2007, 11 “An anaphor must agree in person, gender and number with its antecedent.” d. Payne and Huddleston 2002, 486 “Personal pronouns agree with their antecedent in person and number; in the 3rd person singular, they also agree in gender.” A matching condition like (2a) would combine with the Notional View to determine that any pronominal form an imposter antecedes would necessarily match it in being 3rd person and in having identical number and gender as well. The following sections test this entailment and show that it fails. 3.2 3rd Person Pronominals There are innumerable instances where pronominal forms with imposter antecedents behave just as specified in (2) for a 3rd person DP. We break these data down into singular, plural, and coordinated cases, for reasons that will become clear. It is not accidental that we have chosen reflexive forms in particular to illustrate our claims about antecedence. It is almost universally accepted that the relation between a reflexive form and its antecedent is syntactically governed and cannot be taken to be purely semantic/pragmatic. In terminology consid- [18.208.203.36] Project MUSE (2024-03-19 11:51 GMT) Imposters as Antecedents 17 ered in greater detail in chapter 8, it is never claimed that the relation between a reflexive and its antecedent is one of accidental coreference. But this is not infrequently claimed for nonreflexive pronouns; see chapter 8. The examples in (3) provide relevant data for singular imposters as antecedents for 3rd person reflexives (which obligatorily require an antecedent). (3) a. Yours truly has volunteered herself for the position. b. Daddy has helped himself to dessert. c. Would the colonel like...