再论汉语是否有限定和非限定区分

Revisiting the Finite vs. Non-finite Distinction in Mandarin Chinese

按国外相关文献,语言限定性和非限定性区分有不同标准和不同类型。近期,有人提出新证据,表明汉语有句法层面上的限定和非限定区分。但分析表明,这些新证据和早期这方面的证据一样不能成立。汉语缺少按形态和句法来区分限定和非限定的证据,但从功能意义上的限定性范畴来看,汉语存在语义限定和非限定区分。汉语的语义限定性问题和完句性问题可以归并,两者都可以从语句所述情景的时空定位或者事件类型的个体化来加以探讨。汉语事实表明,作为人类语言共性的限定性范畴不是语法限定性,而是语义限定性。

The opposition of finite vs. non-finite as a clausal property is made chiefly on a language-specific basis. Cross-linguistically speaking, different criteria or dimensions are involved in establishing such a distinction, which could be distributional, morphological, syntactic, semantic or interpretational. Different languages may adopt one or two or more of them without much consistency. It is suggested in Lasser （1997） and Gretsch and Perdue （2007）, among others, that two types of finiteness be distinguished： M（orpho-syntactic） - finiteness and S（emantic） -finiteness, with the latter being an interpretational component as well as the invisible function of the former. Importantly, M-finiteness implies S-finiteness, but S-finiteness can hold in the absence of M-finiteness. With respect to Mandarin Chinese, it has been claimed that the crucial criterion for the finite vs. non-finite opposition in the language is the existence of Tense as part of IP. Earlier evidence in this regard from J. Huang （ 1982, 1984, 1987, 1989） and Li （ 1985, 1990）, among many others, was invalidated by Xu （ 1986a, 1986b）, Y. Huang （ 1994, 2000）, and Hu, et al. （2001）. Recently, new analyses are produced by T. Lin （2011, 2012, 2015）, reverting to the theme that the Chinese language is characterized by the tense-based finite vs. non-finite distinction. （i） Granted that modal verbs in Chinese take clausal complements, T. Lin argues that epistemic modals s-select finite clauses, but that root modals s-select non-finite clauses. （ii） Based on Paul （2002）, T. Lin concludes thatobject fronting in Chinese can only occur in finite but not non-finite clauses. （iii） Different scope readings of object quantifiers in Chinese arise owing to the finite vs. non-finite distinction. It turns out, however, that none of these new analyses are valid. The conclusion is that M-finiteness in general and the grammatical-tense-based finiteness vs. non-finiteness distinction in particular are of no relevance to Chinese. The distinction between finiteness vs. non-finiteness can only be made on a semantic or interpretational basis in the language. That only S-finiteness is found in Chinese tallies with the claim that Chinese is a genuine tenseless language. Given that S-finiteness is mainly defined in terms of temporal anchoring, it is our suggestion that the finiteness issue of Mandarin Chinese be treated as a furthering of the tradition in the well-documented sentence-（in） completeness study of the language.