也論漢語「主語指向型」動補結構的句法構造 (On subject-oriented resultatives in Mandarin)

漢語「主語指向型」動補結構一直被學界看作是動補結構的「例外」。本文著重研究了這類動補結構的句法語義特點，利用多種句法手段證明「主語指向型」動補結構內部存在差異，即有些並不屬於動補結構，有些表面上看似「主語指向」的動補結構其實是基礎動補結構的衍生結構，在底層結構中這類補語仍是指向賓語。這一分析不僅維持了關於動補結構的統一分析理論（如 DOR 理論），而且為現代漢語的動補結構提供了跨語言的分析視角。

Subject-oriented resultatives have long been regarded as exceptions or challenges to current analyses of resultatives. This paper divides them into three types: (a) xie-lei (write-tired); kan-dong (read-understand); and (c) ting-fan (hear-annoyed), a mixed type. It argues that type (a) is a real resultative construction, that type (b) is not a resultative construction, and type (c) could be understood as either (a) or (b). It is shown that Mandarin subject-oriented resultatives do not violate the Direct Object Restriction (DOR) on resultatives, as type (a) is a derived construction from its object-oriented counterpart and type (b) is irrelevant to DOR. On the basis of a unified analysis of Mandarin resultative constructions proposed in the paper, it is argued that subject-oriented resultatives are neither exceptional nor challenging to current analyses of resultatives, especially DOR.