-
7. Pronominal Agreement
- The MIT Press
- Chapter
- Additional Information
7 Pronominal Agreement 7.1 Background In chapter 3, we presented the basic data of pronominal agreement with imposters . In chapter 4, we developed an account of antecedence motivated largely by the goal of characterizing pronominal agreement, and we gave a first version of a pronominal agreement principle based on antecedence. In chapters 5 and 6, we provided substantive if partial views of the internal structure of imposters and camouflage constructions. Here, we introduce a novel conception of pronominal agreement, one designed to provide the basis of an account of the agreement determined in a range of imposter and camouflage constructions. But, as we will argue in chapter 13, it also provides the basis for other instances of pronominal agreement not linked to these specific construction types. Moreover , we will show that the same ideas are central to our account of the ϕ-feature values of coordinate nodes in chapter 9. 7.2 Secondary Sources: Introduction Recall statement (1a) from Sag, Wasow, and Bender 2003, 208, cited in (2) of chapter 3, and our last version of a related statement, (1b), given in (28) of chapter 4. (1) a. Coindexed NPs agree. b. The Pronominal Agreement Condition (second version) If P is a nonexpletive pronominal whose immediate antecedent is A, then P agrees with A in those ϕ-features for which P is not inherently valued. These conditions are inadequate as an account of natural language pronominal agreement for at least two reasons directly relevant to current interests. First, in chapter 4 we rejected appeal to coindexation for describing pronominal 90 Chapter 7 agreement in favor of a primitive antecedence relation. That renders the present treatment incompatible with (1a). Second, and far more important, pronominal agreement involving imposter and camouflage DPs evidently reveals characteristics not treated by either (1a) or (1b).1 As we have documented, both special DP types can in a variety of cases determine two alternative sets of ϕ-feature values for pronominals they antecede. One of these value sets matches those of the shells of imposters and camouflage structures. This variety of pronominal agreement is entirely congruent with the verbal agreement determined by imposters and camouflage DPs and is the type of agreement generally expected for regular DPs, the type assumed in accounts like (1a,b). The other type of pronominal agreement that imposters and camouflage expressions permit lies outside the framework of standard agreement and involves pronominal ϕ-feature values that clash with those of the shells. For concreteness, we focus on specific imposter and camouflage cases drawn from earlier discussions. (2) a. This middle-class citizen lives within my means. (based on (4f ) of chapter 1) b. Your Majesty should praise yourself. (see (29a) of chapter 6) In an informal survey of nine people, five found (2a) acceptable and one found it “not perfect, but not bad.” Three people found it unacceptable. As noted in chapter 3, there is a certain amount of interspeaker variability where singular imposters antecede non–3rd person pronominals (the examples in chapter 3 involve reflexives). See (11) below for more discussion of this issue. The remainder of this chapter treats the variety of English where (2a) is acceptable. There is little doubt that the 1st person imposter this middle-class citizen antecedes the 1st person pronominal my, because the construction X lives within Y’s means has the property that Y must be anteceded by the subject of lives and agree with it. (3) a. I live within my/*your/*that guy’s means. b. You should live within your/*my/*their means. Hence, a proper analysis of the pronominal agreement in (2a) must somehow account for the fact that the 3rd person DP this middle-class citizen can determine a 1st person ϕ-feature value on the pronoun it antecedes. In the SHCC case (2b), the full camouflage DP is once more itself 3rd person (as evidenced by the subject-verb agreement in examples like Your Maj- [3.230.128.106] Project MUSE (2024-03-19 07:01 GMT) Pronominal Agreement 91 esty is justifiably proud of yourself ). But it determines a 2nd person reflexive pronominal. Note, in particular, that it is the subject itself that antecedes the reflexive, not its 2nd person possessor. In this case, then, as in the imposter case, a proper account of pronominal agreement must be more complicated than an elementary view like (1a,b). We have built our conceptions of imposters and camouflage structures in chapters 5 and 6 expressly to permit a more flexible account of...