BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Probe-Goal Approach to Parametric Differences in Nominal Phrases
  - Yi-An Lin\, Department of Linguistics
DTSTART:20080225T171500Z
DTEND:20080225T180000Z
UID:TALK10454@talks.cam.ac.uk
CONTACT:Louise Radok
DESCRIPTION:In this talk\, I intend to postulate a machinery of encoding r
 eferential properties (i.e. definiteness) and grammatical number (i.e. plu
 rality) to explain parametric differences in nominal phrases. Compared wit
 h articled languages such as English\, definiteness in article-less langua
 ges (i.e. Chinese) can be exhibited with bare nominals due to the lack of 
 articles\; In addition\, the expression of grammatical number in most clas
 sifier languages (i.e. Chinese\, Japanese\, and Korean) does not rely on n
 umber morphology\, which is different from non-classifier languages (i.e. 
 English\, French and German) (see Chierchia 1998). However\, according to 
 Pereltsvaig’s (2007) Universal-DP Hypothesis\, the syntactic structure o
 f nominal phrase is universal regardless of the presence of lexical items 
 which realize the heads of the functional projections. This is contrast to
  Cheng and Sybesma’s (1999) proposal that Chinese and English have diffe
 rent encoding mechanisms of definiteness and Li’s (1999) analysis of plu
 ral marking in Chinese and English. What Cheng and Sybesma believe is that
  definiteness is encoded by the function head\, classifier (Cl)\, in Chine
 se\, whereas it is encoded by the D head in English. As for Li’s analysi
 s of plural morpheme -men in Mandarin and -s in English\, she argues that 
 they are realized in different functional heads according to diverse head 
 movements. She proposes that -men is realized in D given Noun-men is defin
 ite in meaning\, whereas -s is realized in the function head\, number (Num
 ). However\, her analysis is rejected based on the fact that English and M
 andarin have an unvaried Adjective-Noun order. The cross-linguistic machin
 ery of encoding referential properties and grammatical number I propose fo
 r the nominal phrase is the syntactic structure schematized as DP>NumP>S(p
 ecificity)P>light noun projection (nP)>NP. The head of DP is the locus of 
 the [definite] feature\, the head of NumP\, which hosts numerals in its sp
 ecifier position\, is the locus of the [number] feature\, the head of SP\,
  where demonstratives are merged in its specifier position\, is the locus 
 of the [specific] feature\, and the head of nP\, which is lexically realiz
 ed as the classifier in classifier languages\, is the locus of the [refere
 ntial] feature. In terms of feature interpretability (Chomsky 1995)\, the 
 mentioned feature carried by each functional projection is interpretable. 
 However\, to solve the problem of the phenomenon “definiteness spreading
 ” as in Swedish\, I further assume that the head of each functional proj
 ection bears not only the interpretable feature but also several uninterpr
 etable features related to the other functional projections. For instance\
 , the D head is composed of an interpretable [definite] feature and the un
 interpretable [number]\, [specific]\, [referential] features. According to
  Chomsky’s (2001) Probe-Goal theory\, the interpretable feature under ea
 ch functional head interacts with the uninterpretable feature under other 
 functional heads through the operation Agree. For example\, the D head wit
 h the interpretable [definite] feature and the unvalued uninterpretable [r
 eferential] feature serves as the Probe\, while the n head with the interp
 retable [referential] feature and the unvalued uninterpretable [definite] 
 feature serves as the Goal. The interpretable [definite] feature on D matc
 hes and deletes the unvalued uninterpretable [definite] feature on n by Ag
 ree\, whereas the interpretable [referential] feature on n matches and del
 etes the unvalued uninterpretable [referential] feature on D by Agree. Suc
 h an analysis can easily explain the “definiteness spreading” phenomen
 on given that the [definite] features (whether interpretable or uninterpre
 table) can be spelt out on different functional heads simultaneously. Addi
 tionally\, given the difference of feature strength\, the word order varia
 tion among different languages and within the same language can be explain
 ed by DP-internal head and/or phrasal movement(s). As for my account of pl
 ural marking in English and Chinese\, I propose that the realization is a 
 PF operation just like the Affix Hopping of English agreement marker –s 
 in the clausal domain. \nIn summary\, the current model provides a unified
  account and requires only the different feature strength which leads to D
 P-internal head and/or phrasal movement(s) to explain parametric differenc
 es in nominal phrases cross-linguistically. In addition\, Pereltsvaig’s 
 (2007) proposal that the projection of DP is a property of Universal Gramm
 ar can be maintained within the current Probe-Goal approach\, which allows
  natural languages not to lexically realize each functional head.\n
LOCATION:GR-04\, English Faculty Building
END:VEVENT
END:VCALENDAR