According to Chomskyan linguists, grammar (syntax) is autonomous, and linguistic competence and linguistic performance are distinct. Within the paradigm of function-oriented linguistics, however, syntax is not regarded as separable from semantics and pragmatics. Among the various models concerning the interaction of syntax with semantics and pragmatics, it is claimed in Dynamic Syntax (Kempson, Meyer-Viol & Gabbay Reference Kempson, Meyer-Viol and Gabbay2001, Cann, Kempson & Marten Reference Cann, Kempson and Marten2005) that knowing a language is knowing how to use it, and both semantic interpretation and pragmatic inference play a central part in linguistic formalism. Based on Dynamic Syntax (DS), this monograph – a reworked version of the author’s doctoral dissertation and one of the series Routledge Studies in Chinese Linguistics – is a systematic study of the interfaces of Chinese syntax with semantics and pragmatics. The dynamic framework in this book is mainly based on two dynamic notions: syntactic and semantic underspecification, and pragmatic enrichment. Syntactic and semantic underspecification generally means that the semantic content of various syntactic constructions is underspecified and it needs to be characterized or enriched by pragmatic inference. Focusing on this interaction of syntax with semantics and pragmatics, the DS approach is adopted to explore several kernel Chinese grammatical phenomena.
This monograph consists of nine chapters. Chapter 1 is a general description of Chinese syntax. Unlike inflectional languages such as English, Chinese is known for its relative simplicity in word formation and flexibility in surface word order. Both semantics and pragmatics play a key role in such word order flexibility. The syntax–semantics interaction is evidenced by the fact that word order in Chinese often indicates semantic functions. For example, the preverbal NP is systematically treated as the subject and is definite, while the postverbal NP is treated as the object and is indefinite. Similarly, preverbal temporal adverbials are usually punctual, whereas postverbal temporal expressions tend to be durative. The syntax–pragmatics interaction, on the other hand, is manifested by a profound effect of pragmatics on both linguistic performance and grammatical constructions. Such interaction in linguistic performance is best exemplified by the omission of unnecessary adjuncts in communication, while the pragmatic effect on grammatical constructions is exemplified by the implication interpretation of various grammatical constructions, such as the unfortunate or pejorative message carried by the bei construction (further discussed below).
Chapter 2 mainly introduces the DS framework and its technical tools. In general, DS attempts to explain linguistic capacity from the perspective of language use. Such a model can be considered as a parsing process representation, which demonstrates the goal-directed incremental updating of sentence structure relative to the context. This representational construction is realized by a process of ‘tree growth’ in an incremental step-by-step manner. Transition from one parsing state to another indicates the interaction of computational rules with lexical and pragmatic information.
Chapters 3–8 offer a detailed dynamic analysis of the core Chinese syntactic phenomena within the DS framework. While Theta Theory (Chomsky Reference Chomsky1981) claims that the number of syntactic arguments is solely determined by the lexical semantics of the verb, it is argued in Chapter 3 that Chinese verbs are underspecified with regard to the number and the type of arguments they can take. For instance, shui ‘sleep’ in Chinese is conventionally described as an intransitive verb. However, the following sentence is quite acceptable:
Shafa as a locative NP in this sentence indicates the location of sleeping, and is treated as an object argument of the verb, rather than an adjunct. A decontextualized syntactic theory cannot offer an adequate explanation for this phenomenon. Chinese verbs are not inherently transitive or intransitive, rather their argument structures are determined by semantic and pragmatic information.
Chinese has long been regarded as a topic-prominent language, with the topic construction enjoying more freedom than it does in English. According to Chafe (Reference Chafe and Li1976), topic constructions can be classified into English-type and Chinese-type. The English-type refers to a topic that is subcategorized by the verb, while the Chinese-type refers to a topic that just specifies the frame of reference for the following comment. Chapter 4 shows that both types of topic structures are allowed in Chinese and respect the ‘aboutness condition’, which means that there is a relevance-based aboutness relation between the topic and comment. This aboutness is captured by the LINK strategy in the DS, which links the topicalized expression to the unfolding propositional structure with or without a copy of the topic formula. The English-type topic is copied in the parsing because it is an argument of the verb in the comment, while the Chinese-type topic is not. The following examples illustrate:
The English-type topic Zhangsan in (2) projects an unfixed node which eventually merges with the open argument node in the comment clause by acting as the patient of the verb, and the Chinese-type topic jiu-ge miyu ‘nine riddle’ in (3) provides a frame of reference for the comment to yield a complete propositional formula. Both the topic types thus understood provide a relevant point of departure from which the comment is developed.
Chapter 5 is devoted to analyzing the Chinese passive construction. The morpheme bei is claimed to have grammaticalized from a verb to a passive voice particle, which indicates that the preceding argument is the passive recipient of the action. For example:
The displaced argument Zhangsan in (4) is placed in a position of prominence and the passivity of this patient in relation to the action of ‘seeing’ is emphasized along with the implicature that he might be suspected of committing some illegal act. Wu argues that the pre-bei constituent, Zhangsan, as a left-dislocated argument, projects an unfixed node with a locational requirement. In other words, its position is underspecified. As the parsing proceeds, this unfixed node is further linked onto a type-t (proposition) requiring structure, which indicates that Zhangsan is the dislocated internal argument of the predicate kanjian ‘see’. Thus, the Chinese passive construction is given an adequate explanation under DS.
The Chinese copular morpheme shi is examined in Chapter 6. Wu notes that shi can occur in predicative, emphatic or elliptical construction, which yield different interpretations, as shown in the following examples, respectively:
This variation in interpretation illustrates that the interpretation of shi is highly context-dependent. Thus, it is claimed that shi possesses the characteristics of an anaphoric expression which takes its value from the context. It is semantically underspecified and acts as a semantic placeholder which needs to be pragmatically enriched by contextual information.
A further construction with shi, the cleft construction, is explored in Chapter 7. Similarly to the emphatic construction with shi discussed in Chapter 6, the Chinese cleft construction also gives rise to a focus interpretation, as exemplified by the following sentence:
By being dislocated to the right periphery of the sentence, Lisi is placed in a position of prominence just like Mulan in (6) above. But unlike Mulan in the emphatic construction, the referent of the post-copular focused constituent Lisi is identified by the pre-copular description – Wangwu jian guo de ‘who Wangwu has met’ in this case. This identification is explicated in DS by treating Lisi as projecting an unfixed node as the update for the ‘subject’ node of the sentence.
In addition to the structural underspecification addressed in the preceding chapters, Chapter 8 is devoted to the analysis of semantic underspecification, focusing on the different usages of the third-person pronoun ta, both referential and expletive, as shown in the following examples:
The difference in interpretation results from the fact that in (9), the referential ta is the copy of its antecedent gou ‘dog’, while in (10), the expletive ta takes its value from the postverbal indefinite expression shi-wan mian ‘ten bowls of noodles’. From a dynamic processing perspective, both types of ta can be treated as placeholders projecting an interim value that is provided contextually. Thus, a unified account of various meanings of ta is viewed as deriving from the interaction of structural underspecification and pragmatic enrichment.
Chapter 9 offers a conclusion and enumerates the monograph’s contributions to Chinese linguistics and implications for future linguistic theorizing.
Wu should be credited with applying the DS framework to the study of Chinese grammar. The author has shown that the in-depth exploration of the pragmatic factors which DS allows leads to insightful analyses of aspects of Chinese grammar. This book is also characterized by its systematic account of interactions of Chinese syntax with semantics and pragmatics. In spite of the broad scope of the inquiry, its discussion and argumentation are easy to follow due to the author’s lucid and logical writing. Each chapter begins with a clear introduction of the syntactic phenomenon to be discussed and a critical review of the previous relevant literature, which prepares the reader to achieve a better understanding of the topic in question to be presented. This helps provide a reference and starting point for future research in the relevant field.
What needs to be noted, however, is the author’s subjectivity in the choice and analysis of the data. First, nearly all the data used in the study are based on the author’s introspection. Though these examples seem natural to Chinese native speakers, we suggest that the study would be more convincing if the data were selected from well-respected corpora such as the BCC corpus (further introduced below). Second, some analyses are not sufficiently persuasive. For example, in Chapter 5, when reviewing the verb hypothesis of bei, Wu claims that unlike verbs such as ai ‘receive’, shou ‘get’ and zaoshou ‘undergo’, with similar meanings, bei ‘cannot be used in V-not-V form’ (100), and that thus the following utterance is unacceptable:
However, this utterance sounds perfectly natural to us as well as to other Chinese speakers we have consulted. In addition, by searching the authoritative online BCC corpus – a Chinese corpus (15 billion words) set up by Beijing Language and Culture University which encompasses written and spoken, and diachronic and synchronic Chinese data (Xun et al. Reference Xun, Rao, Xiao and Zang2016) – we found four cases of the bei-mei-bei construction. When bei-bu-bei was searched, 24 examples were found. Thus, it is clear that bei-‘not’ -bei (bei-mei-bei/bei-bu-bei) as a construction can be used in certain contexts, though we agree with Wu that bei is not a verb and has become a functional particle (a passive marker) via the process of grammaticalization. Although as a grammatical marker, ‘bei has no lexical content at all’ (101), we would argue that bei still possesses some meaning potential of ‘being affected’ as a kind of meaning residue after grammaticalization. This observation is partly evidenced by the ‘unfortunate or pejorative message’ conveyed by the bei construction (8).
On the whole, this book makes an excellent contribution to the field by filling a significant theoretical gap, and shedding new light on the interaction of Chinese syntax with semantics and pragmatics. It is strongly recommended for graduate students as well as researchers with an interest in the study of Chinese syntax, semantics, pragmatics and their interfaces.
