Distributional criteria for verbal valency in Chinese By Yilu Zhao (review)

228 LANGUAGE, VOLUME 74, NUMBER 1 (1998) transfer of properties from Ll to L2 and the creation in IL of structures not found in Ll or L2. In Ch. 2 Y explores the logical problem of L2 acquisition and assumes the subset principle as a principle of learnability; especially, a superset relation of some property or construction in Ll will hinder learners in acquiring a property or construction in L2. Preemption can operate when Ll or IL forms a superset in some respect: a more restricted target structure replaces an overgenerated structure on the basis of no direct negative evidence. This can be accomplished with the uniqueness principle which, informally stated, says that with no positive evidence in the target language to the contrary 'a given meaning in a sentence can have only one form' (60). When learners become aware of a target structure , they will realize that their IL structure is superfluous . Ch. 3 explores in typological and generative terms some basic problems oflearning English for Chinese speakers. Chinese is a topic-prominent language while English is a subject-prominent language. Y assumes that topics occupy an ?-bar position in Chinese as well as English but that English topic structures form a subset of the Chinese topic structures , and learners must therefore narrow their grammars . For example, in Ch. 4 Y explores such CIL sentences as New cars must keep inside. While this sentence might appear to be missing passive morphology , Y argues that it is a topicalized structure with a null subject transferred from Chinese, New cars, proor¿ must keep inside t,. In Ch. 5, Y argues that English ergative sentences such as World War III will happen are inherently difficult for L2 learners since they require a theme as subject yet do notrequire any marking on the verb, unlike the passive which is marked. Overgenerahzing passive morphology to permit themes as subjects, as in The World War III will be happened, is a common IL strategy, not only for Chinese learners. Ch. 6 deals with the issue of íoug/i-movement and raising structures. CIL sentences such as / am difficult to learn superficially resemble the English ioHgA-movement sentence This book is easy to read. Y argues that such sentences do not result from misapplying iOHgA-movement to subjects rather than objects ofcomplement clauses but that they result from the overgeneralization of Chinese subject raising. Raising structures in CIL form a superset with regard to English and are therefore extremely difficult to retreat from, requiring the learner to master other syntactic structures, i.e. Iforget easily or / have difficulty learning, in order to preempt the ill-formed structures and to learn from context the meaning of the English iowgA-movement sentences. The learner will realize from context that in sentences such as My cousin isfun to visit, my cousin can only be interpreted as the object of to visit. In Ch. 7, Y argues that sentences such as In a coeducational school, there are many students have this problem seem superficially to involve missing relative clause markers but in fact originate in the overgeneration of existential sentences with There are. There are is used in CIL because Chinese does not allow indefinites in preverbal subject position; these sentences instead reflect the transfer of Chinese pivotal structures, a kind of serial verb construction. In a pivotal structure, within a GB framework, an NP after the existential verb controls a subject empty category (PRO) in the S-complement. Under such an analysis (proposed for Chinese by Dingxu Shi, Pivotal structure in Chinese. Los Angeles: University of Southern California MS, 1988) the IL sentence would be In a coeducational school, there are many students PRO have this problem. This volume presents a very careful and clear analysis and explanation of certain IL problems and will be very helpful to scholars of IL and to teachers of EngUsh to Chinese. There are appendices containing questionnaires used by Y for grammaticality judgment tasks. The book is generally well presented, except for some typographic errors, especially on page 194, which make it a difficult page to read. [Joanne Sher Grumet, New York University.] Distributional criteria for verbal valency in Chinese. By Yilu Zhao...