Thursday 10 March – Wei-Tien Dylan Tsai

Speaker: Wei-Tien Dylan Tsai

Title: Operator Binding and the Syntax of Analyticity

Date: Thursday 10 March

Time: 15.15-16.30

Venue: Eyckhof 2-003

Abstract: 

In this talk, we entertain the possibility that peripheral features play a crucial role in the formation of the upper layer of a sentence, which can be checked by either external Merge or internal Merge (i.e., Move) according to the parameter-settings of individual languages. Along this line, topic prominence is regarded as the result of peripheral feature checking, and the null topic hypothesis à la Huang (1984) is reinvented as a null operator merger to fulfill interface economy in the left periphery. In this regard, Chinese provides substantial evidence from obligatory topicalization in outer affectives, evaluatives, and refutory wh-constructions, which applies only when the licensing from a D(efiniteness)-operator is blocked.

The idea also extends naturally to the issues concerning pro-drop and bare nominals in general. Namely, we may well compare Chinese obligatory topicalization to those residual cases of verb-second (V2) in English, all being manifestation of the strong uniformity. Topic prominence can also be reinvented in this new light: The null topic operator can be regarded as the quantifier part of a definite argument, just like a Q(uestion)-operator or a polarity operator (call it an $-operator) that functions as the quantifier part of a wh-phrase:

1. D-operator binding for bare nominals in-situ:

            [D_x-Top]         . . .       N(x) . . .

1. Q-operator binding for wh- in-situ:

            [Q_x-Int]           . . .       wh(x) . . .

III.      $-operator binding for indefinite wh:

[$x-Mood] . . . wh(x) . . .

Consequently, a Chinese topic is either an XP in the Spec-head relation with Top, or a discontinuous DP consisting of a peripheral D-operator and an in-situ nominal. In the case of pro-drop, a P(redication)-operator is raised instead to trigger predication on a null discourse topic in Huang’s (1994) sense. P-operator is essentially the syntactic counterpart of a l-operator in semantics (cf. Chierchia 1984, 1985): 

1. P-operator binding for pro-drop:

            [_PredP Topic [P_x-Pred [_TP  . . . e_x . . . ] 

Furthermore, we present evidence showing that the scope-taking property of mei ‘every’ is realized not through Quantifier Raising (QR), but by merging a distributive operator (i.e., the implicit counterpart of dou) directly to the left periphery:

1. “-operator binding for mei-NPs:

[“x-Dis] . . . mei-NP(x) . . .

We this complete the spectrum of the operator binding construals in question. In a robust analytic language such as Chinese, where a phrase can be constructed in a sentential magnitude (cf. Tsai 1994), this is not only possible but also necessary.