Parametrizing the Timing of Transfer in Japanese and English and its Consequences
Ryoichiro Kobayashi
July 2017
 

The aim of this short paper is to propose an analysis of labeling of some {XP, YP} structures in Japanese. In English, Chomsky (2013) proposes that the label of {DP, vP} = XP in (1a) is determined as vP, since the DP internally merges with TP, rendering the lower copy invisible from LA. Another {XP, YP} problem arises when the DP merges TP in (1b). The label of YP is determined as <φ, φ> via feature-sharing (Agree) between the DP and T(P). In languages with φ-agreement, the strategy in (1b) is available. Then, what would the label of YP in φ-featureless languages like Japanese (Fukui 1986) be like? This is the central question the current study deals with. (1) a. {xp DP, {vP v ,{VP … }}}: XP = vP (Labeling via Internal Merge) b. {xp DP, {TP T, {vP DP, vP}}}: YP = <φ, φ> (Labeling via feature-sharing) I propose that the timing and domain of Transfer is much more flexible in Japanese than in English due to the lack of its unvalued formal features. Combining the Labeling via Transfer (Takita et al. 2016) and the insight of Fukui and Kasai (2004), I will show how the current proposal explains the labeling problems in structure building, multiple nominative, and scrambling constructions, all of which involve {XP, YP} structures in Japanese. Let us first review the previous studies in section 2, and then move to the proposal and its consequences in section 3
Format: [ pdf ]
Reference: lingbuzz/003581
(please use that when you cite this article)
Published in: To appear in Proceedings of WAFL 13
keywords: labeling algorithm, transfer, multiple nominatives, scrambling, japanese, syntax
Downloaded:427 times

 

[ edit this article | back to article list ]