Ellipsis and Syntactic Representation
Yüklə 121,54 Kb. Pdf görüntüsü
|
Christopher Kennedy 19 (61) a.
How talented £ do the Sox start [ DP an
outfield]? b.
How successful £ did Smith produce [ DP a ¥¦£ play]? What is surprising is that ellipsis in the comparative clause seems to elimi- nate LBC violations (Pinkham 1982; Kennedy and Merchant 2000). This is illus- trated by the examples in (62), which are just like the sentences in (59) except that a constituent containing the gap has been elided. (62)
a. The Cubs start a more talented infield than the Sox (do). b. Jones produced as successful a film as Smith (did). Given the main conclusion of the previous section — that elided constituents have syntactic representation and are subject to configurational constraints — the well- formedness of the examples in (62) is extremely puzzling. On this view, these sentences have the structures shown in (63), which should violate the LBC just as much as those in (60), assuming that the LBC is stated in terms of configurational relations (e.g., in terms of the ECP, as proposed in Corver 1990). (63) a.
£ the Sox (do) [ VP
DP an ¥¦£ infield]]] b. Jones produced as successful a film as [wh £ Smith (did) [ VP
DP a ¥¦£ film]]] The solution to this puzzle developed and substantiated in Kennedy and Mer- chant 2000 (K&M) builds on an analysis of LBC effects in which extraction of at- tributive modifiers is ruled out not by a configurational constraint governing empty categories or operator/variable dependencies, but rather by a constraint on the mor- phophonological instantiation of syntactic feature combinations. If this version of the LBC is correct, then these constructions actually instantiate the prediction of the syntactic theory of ellipsis stated in (29b): since ellipsis does not require pro- nunciation of the omitted structure, elided constituents should be insensitive to syn- tactic constraints that derive from morphophonological properties of lexical items. K&M’s arguments for a PF characterization of the LBC are summarized in the next section.
4 4 It should be noted that K&M focus specifically on the case of extraction of left-branch attributive modifiers, not e.g. extractions such as (i), which violate the head-movement constraint, assuming that how is the head of a Degree Phrase (DegP) with the AP headed by strong as its complement (see Abney 1987; Corver 1990; Grimshaw 1991; Kennedy 1999). (i)
How
is Sammy
strong?
As originally observed by Grosu (1974) (see also Corver 1990), the various phenomena that Ross subsumed under the Left Branch Constraint are more properly explained in terms of distinct con- straints.
20 Ellipsis and Syntactic Representation 4.2 A PF Analysis of the Left Branch Constraint We begin with the syntax of attributive modifiers. Following Svenonius 1992, K&M assume that the base position of an attributive modifier (categorically a DegP) is as a right adjunct to NP. Inverted DegPs, such as those in (64)-(65), indicate that some attributive modifiers must move to the left of the determiner (Bolinger 1972; Bresnan 1973; Woisetschl¨ager 1981; Baker 1989; Corver 1990; Hendrik 1990). (64) a.
b. I ate [too big a piece]. c. If I ever see [that disgusting a movie] again, I’ll ask for my money back. d. Bob didn’t write [as detailed a proposal] as Sheila did. e. He took [so big a piece] that he couldn’t finish it. (65) a.
[A how interesting a play] did Brio write? b.
I ate [a too big piece]. c.
If I ever see [a that disgusting movie] again, I’ll ask for my money back.
d.
Bob didn’t write [an as detailed proposal] as Sheila did. e.
He took [a so big piece] that he couldn’t finish it. K&M propose that the position an inverted DegP — as well as an inter- mediate landing site for the wh-operator in comparatives — is the specifier of a functional phrase within the nominal projection but above DP: “FP” (cf. Bennis, Corver, and den Dikken 1998; see Corver 1990 for arguments that the landing site of inversion is not SpecDP). The basic structure is illustrated in (66). (66)
FP DegP
how interesting £ F’ F DP D a NP DegP ¥¦£ NP N play Christopher Kennedy 21 K&M provide three pieces of independent evidence for this syntactic anal- ysis. First, the head of FP can sometimes be morphologically realized as of, as in the following examples (cf. constructions like a bear of a guy discussed in Bennis et al. 1998). (67)
How long of a novel did Brio write? a. I ate [too big of a piece]. b. If I ever see [that disgusting of a movie] again, I’ll ask for my money back. c.
d. He took [so big of a piece] that he couldn’t finish it. Second, an attributive modifier can be “caught” by pseudogapping. This is illustrated by (68a)-(70a), which are ambiguous between the (b) and the (c) inter- pretations. (68)
a. I have written a successful play, but you have a novel. b. I have written a successful play, but you have written a novel. c. I have written a successful play, but you have written a successful novel. (69)
a. The Cubs need a left-handed hitter more than they do a pitcher. b. The Cubs need a left-handed hitter more than they need a pitcher. c. The Cubs need a left-handed hitter more than they need a left-handed pitcher. (70)
a. I buy expensive shoes because I don’t suits. b. I buy expensive shoes because I don’t buy suits. c. I buy expensive shoes because I don’t buy expensive suits. Given the assumption that pseudogapping involves movement of the “remnant” phrase out of VP, followed by VP-deletion (Kuno 1981; Jayaseelan 1990; Lasnik 1995; Johnson 1997; see Sag 1976; Levin 1986; Miller 1992 for qualifications), the possibility of the (c) interpretations follows directly from the structure in (66): these readings correspond to structures in which the attributive modifier raises to SpecFP, and DP moves out of (the deleted) VP. This is illustrated for (69a) in (71), where the deleted VP is enclosed in a box.
Yüklə 121,54 Kb. Dostları ilə paylaş:
|