In lieu of an abstract, here is a brief excerpt of the content:

  • The Syntax of D-Linking
  • Tomio Hirose

1 Introduction

The two wh-questions in (1) differ in terms of the now familiar notion "D-linking" (Pesetsky 1987, 2000).

(1)

  1. a. What is your plan?

  2. b. Which is your plan?

Unlike the answer to (1a), the answer to (1b) is "supposed to be drawn from a set of individuals previously introduced into the discourse, or . . . part of the 'common ground' shared by speaker and hearer" (Pesetsky 2000:16). If what and which are syntactic atoms with no internal syntax, their observed semantic difference, and therefore the D-linking semantics, too, must be lexical in origin. However, there is no a priori reason why the origin of D-linking should, as a UG property, be lexical. In this squib, I argue that UG permits the possibility that D-linking is instead a syntactic phenomenon. More specifically, I argue that the D-linking semantics can arise through the local binding of pro by the (null) interrogative operator. I base my argument on two genetically and geographically unrelated languages, Japanese and Plains Cree. The emerging picture is that as far as D-linking is concerned, the semantic difference between (1a) and (1b) can be reduced to the type of the constituent that the interrogative operator binds.

2 Japanese dono

The Japanese D-linked wh-word to be examined is dono 'which', a wh-word that is used only attributively, as in dono hon 'which book'.1 Its D-linking nature is illustrated by the fact that (2) can be followed by (3a), but not (3b).2 [End Page 499]

(2) Hon-ga    hosii-n-da     tte?

book-NOM want-NML-COP Q

'You want a book, eh?'

(3)

  1. a. Donna hon-ga     hosii no?

    what   book-NOM want Q

    'What (kind of) book do you want?'

  2. b. #Dono hon-ga     hosii no?

    which book-NOM want Q

    'Which book do you want?'

That is, (3a) can be uttered felicitously without a preestablished set of books in the minds of the speaker and the hearer, but (3b) cannot.

Morphologically, the initial portion do- of dono can be identified as the wh-morpheme, as the following paradigms illustrate:

(4)

  1. a. do-re    'which one'

  2. b. do-no   'which'

  3. c. do-ko   'where'

  4. d. do-o     'how'

(5)

  1. a. ko-re    'this one'

  2. b. ko-no   'this'

  3. c. ko-ko   'here'

  4. d. ko-o     'this manner'

If do- is the wh-morpheme shared with other wh-words, the D-linking semantics of dono can prima facie be attributed to the remaining portion -no. However, the issue is not so straightforward. The second portion -no of dono may be an instance of the genitive marker -no of the language if Fukui's (1995) assumption is correct. Fukui (1995: 105, fn. 10) regards -no in the demonstrative pronouns kono 'this', ano 'that', and sono 'it' as the genitive marker because these pronouns take different forms when appearing elsewhere than in prenominal position, namely, kore 'this', are 'that', and sore 'it'. Given this assumption, -no can no longer be identified as the locus of the D-linking semantics associated with dono. Now, if neither do- nor -no, the only two morphological constituents of dono, is responsible for its D-linking nature, what is?

3 Local Binding of pro

A seemingly far-fetched but still logically possible way to answer the above question is to attribute the D-linking semantics of dono to something invisible/inaudible, namely, a zero constituent. Postulating a zero constituent within dono itself is not implausible because the other Japanese wh-words listed in (4) have the form "do- plus something." If dono indeed encompasses a zero constituent, what kind of zero element is it? I propose that the zero constituent in question is pro. For concreteness' sake, I assign dono hon 'which book' the following structure: [End Page 500]

(6) [DP[D'[NP[DP do- prox] -no [NP hon]] ] Op[Q]x]3

The wh-morpheme do- and pro form a DP, a prenominal modifier of the NP hon (see Noguchi 1997:781). I assume that the genitive marker -no is inserted onto the constituent that immediately precedes an NP (Fukui 1995:143). The NP...

pdf

Share