Eliminating T as an independent syntactic head
direct link: http://ling.auf.net/lingbuzz/004532
In this paper, I propose to eliminate T as an independent syntactic head from the computational system of human language. Adopting feature inheritance (Chomsky 2013, 2015) and assuming that tense (and other features, if any) is inherited from C by T, T does not have any contents in lexicon. It is doubtful that T can exist as a legitimate syntactic head, though tense feature definitely exists in human language. Based on Hosono’s (2018) analysis on verb movement, I propose that in languages such as French and English, a verbal head moves and merges to the root, and it inherits tense (and any other functional features) after C merges; in V2 languages, a verbal head directly moves to C, and feature inheritance does not occur. In both cases, I argue, the object which has traditionally been called TP is unlabeled. I also introduce the framework of workspace proposed by Chomsky et al. (2017) and Chomsky (2017), who change the definition of the merging operation from Merge to MERGE and claim that only external and internal MERGE are legitimate, rejecting other merging operations, e.g. countercyclic movement such as verb movement. I argue that without T, the problem on the countercyclic property of verb movement is solved. This is the first draft. Any comments are very welcome.
|Format:||[ pdf ]|
(please use that when you cite this article)
|keywords:||t(ense), feature inheritance, verb movement, labeling, workspace, merge, countercyclic property., morphology, syntax, phonology|