(香港中文大學-北京語言大學漢語語言學與應用語言學聯合研究中心、香港中文大學 中國語言及文學系協辦)      

地點:香港中文大學 中國文化研究所 文物館東翼 二樓 活動室


On the Universality of the Adjunct-complement Distinction in Nominal Structure


The adjunct-complement distinction has long been recognized in traditional Chinese linguistic study for the analysis of clausal and verb-phrase structure, but research on nominal structure has recognized only a general dingyu or ‘determinative’ category in the pre-nominal position. While most generative works on Chinese syntax have assumed the distinction for the noun phrase as well as other phrases, some scholars have claimed that for a number of languages with head-final noun phrases (such as Chinese, Japanese, Korean), the adjunct-complement distinction does not exist. Among the scholars who recognize the distinction, opinions also differ as to whether some prenominal elements (e.g., so-called ‘gapless relative clauses’) are to be analyzed as complement clauses or as sentential subjects. Finally, in recent years, within the generative paradigm there have been proposals to derive relative clauses through complementation on the one hand, and analyze so-called noun-complement clauses as appositives, hence adjuncts, on the other. Thus for these scholars, the complement-adjunct distinction exists, but in a completely opposite way.

In this presentation, I shall reaffirm the existence of an adjunct-complement dichotomy for Chinese noun phrase structure, adducing extensive evidence from: word order, subcategorization, stacking, suo-relatives, coordination, long-distance extraction, ellipsis, and nominalization. I shall also re-assess alternative proposals to analyze ‘gapless relatives’. Relevant facts from non-Mandarin dialects, and corresponding phenomena in Japanese and Korean will be examined in the interest of a more adequate cross-linguistic account.

In addition, I shall address the recent generative accounts that derive true relative clauses through a complementation approach and treat all noun complement clauses as appositives. We shall highlight the complement analysis of gapless RCs and argue that they cannot be analyzed as appositives in any way that does justice to their semantics—since gapless RCs denote events (not propositions) while their head nouns are transitive predicates, of type 〈event, 〈e,t〉〉. We also show that even for the normal propositional noun complements, the appositive analysis faces difficulties in light of the evidence we have seen in Chinese. We thus conclude that a proper adjunct-complement distinction that puts complements (including gapless relatives) structurally below adjuncts is not only appropriate for languages with head-final noun phrases but for other languages as well.