LINGUIST List 14.458

Sun Feb 16 2003

Review: Lang Description/Historical Ling: Shi(2002)

Editor for this issue: Naomi Ogasawara <>

What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in.

If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Simin Karimi at


  • Karen Chung, The Establishment of Modern Chinese Grammar

    Message 1: The Establishment of Modern Chinese Grammar

    Date: Sat, 15 Feb 2003 19:03:51 +0000
    From: Karen Chung <>
    Subject: The Establishment of Modern Chinese Grammar

    Shi, Yuzhi (2002) The Establishment of Modern Chinese Grammar: The Formation of the Resultative Construction and Its Effects. John Benjamins Publishing Company, 262pp, hardback ISBN 90-272-3062-5 (Eur), 1-58811-203-9 (US), $98.00, Studies in Language Companion Series 59.

    Announced at

    Karen Steffen Chung, National Taiwan University

    This book, a revision of the author's Ph.D. dissertation, attempts to identify and trace the origins of the resultative construction in Chinese, which the author considers 'the most fundamental syntactic change in the history of Chinese' (p. 228). By 'resultative' is meant constructions with a verb describing an action, plus a morpheme indicating a telic outcome of that action, such as _chi1bao3_ [eat + full] 'to eat to the point of being full', or _da3po4_ [strike + break] 'to [hit and] break something'. Shi collected his data from a variety of texts, both classical and vernacular, including the Confucian _Analects_, Tang dynasty vernacular Buddhist texts (_bian4wen2_), Yuan drama, and novels such as _The Dream of the Red Chamber_. Modern examples, some apparently made up, and some from non-Mandarin dialects, are also cited.

    The author believes that the VR (verb + resultative) construction had its origins in the separable resultative structure with an order of VXR, in which X could be an object, adverb, or negative particle. Subsequently the V fused with the R so that there could be no intervening material, thus yielding the modern VR form. The author posits three phases in the emergence of the resultative construction:

    (1) Around 800, VR phrases whose Rs were the underlying predicate of O (the object) started to appear in VRO form; VR phrases whose Rs were the underlying predicate of V (i.e. expressed the temporal internal structure of the verb) reached a low degree of fusion. (2) Around A.D. 1000, VR phrases whose Rs were the underlying predicate of V reached a high degree of fusion. (3) Ca. A.D. 1200, VR phrases whose Rs were the underlying predicate of the subject (i.e. the agent of the verb) became loosely fused. (p. 229)

    The key factors Shi posits in the formation of the resultative construction are (1) the emergence of disyllabification in Chinese, which promoted fusion of the V and R elements; (2) idiomatization of the most frequently used collocations, which first produced individual lexical items but eventually a generalized syntactic pattern; (3) elimination of the _er2_ 'and' connective, which produced both resultative constructions and serial verbs; and (4) establishment of a verb-result relationship in the construction, in which telic states followed and other adverbs preceded the verb.

    Shi believes the VR construction was the motivating cause behind several major changes in Chinese syntax. Because in many cases an object could not for various reasons immediately follow the VR construction, objects needed to be placed elsewhere in the sentence, though VRO was still an option for some verb types.

    If the object was definite, it could be preceded by the particle _ba3_ and placed before the VR construction, to form a 'disposal' construction, e.g. _Wo3 ba3 shu1 kan4wan2le_ 'I finished reading the book.' One specific book is referred to here; an indefinite reading is not possible. Shi notes that some verbs in initial position in serial verb constructions became grammaticalized into 'prepositions' or co-verbs, or in this case the object marker _ba3_.

    If the object was indefinite, it could be accommodated under the new 'verb copying' construction, e.g. _ta1 he1 jiu3 he1zui4le_ 'He drank until drunk/he got drunk' - no particular bottle of alcohol is referred to here. Other possibilities were to topicalize the patient by fronting it, with or without the subject: _Shu1 yi3jing1 kan4wan2le._ '[As for] the book, [I] have already finished reading [it].' Shi believes all of these structures, and even the comparative _bi3_ construction, are traceable to the emergence of the VR construction.

    Some of Shi's points are open to debate. Shi proposes three types of resultatives distributed over a cline (p. 32-33): 'syntactic collocation' (e.g. _kan4wai1_ 'to misinterpret, put a wrong spin on s.t.'), 'verb + bound resultative' (e.g._xue2hui4_ 'to learn s.t. thoroughly'), and 'compound verb' '(e.g._kan4kai1_ 'to view s.t. with detachment' and _shuo1ming2_ 'to explain'). He attempts to distinguish the first two categories by claiming that expressions in the first category are synthetic, concatenated syntactically, rather than being set phrases; in the second he includes idiomatized phrases; and in the third, lexified compounds. He claims that what they all have in common is that they do not allow any intervening material, in contrast to the situation in Middle Chinese.

    In my view, however, there are only two categories: separable resultative *constructions* and inseparable verb *compounds*. Among the examples cited by Shi in all three categories, there is only one true compound, and that is _shuo1ming2_, 'to explain', which cannot be separated under any circumstances. All the others, such as _zhua1jin3_, 'to hang on tightly, emphasize', can take not only the affirmative and negative potential forms, with an inserted _de2_ or _bu4_, but also sometimes intensifiers such as _de hen3_ which modify the resultative satellite, e.g. _chi1 de hen3 bao3_ 'to eat to the point of being very full', _zuo4 de hen3 lei4_ 'to become very tired working at something.' In response to this, Shi claims that the potential form is a different construction entirely and does not constitute an exception to his rule (p. 33ff). Yet the criterion of whether or not intervening material of any kind is allowed - which is Shi's own acknowledged criterion - is the one thing that can neatly distinguish constructions from compounds. Shi does discuss the potential form in chapters 4-6, but fails to address and solve this fundamental issue.

    Shi claims that the R of a VR construction must be monosyllabic (p. 39), which would exclude trisyllabic phrases like _xi3gan1jing4 _ 'to wash clean' (Shi gives only _xi3jing4_ as an example on p. 33) and _jiang3qing1chu3_ 'to say s.t. clearly', or even four-syllable ones like _jiao1dai4qing1chu3 _ 'to give explicit instructions'. He does however include three-syllable constructions in his 'first category' of syntactic resultatives, like _yan2jiu4qiong2_ 'to research exhaustively' on p. 32. On p. 85 he quotes a four-syllable example from 1200 A.D.: _yue1su4 fen1ming2_ 'to strictly discipline'. He claims that such expressions have a low degree of fusion, and that they do not allow objects; yet objects, fronted in a _ba3_ construction, are certainly possible for some of these expressions, e.g. _Qing3 ni3 ba3 suo3you3de wan3 xi3gan1jing4_. 'Please wash all of the dishes.' This issue is not fully clarified in the book.

    A relatively minor point: Shi notes (p. 13) that 'elements in the R position often lost their independent tonal values'. This is true of Beijing Mandarin, but not of all other varieties of standard spoken Chinese; Taiwan Mandarin usually preserves the full tonal values of all morphemes, including resultatives, excepting only a very small closed group of function particles.

    One of Shi's many interesting observations: unlike English resultatives, such as 'to lick the platter clean', a change of state indicated by a Chinese resultative does not have to immediately follow the V action; e.g. _chi1pang4_ 'to eat until fat' takes place over an extended period of time (p. 30). Shi also notes that the transitivity of a VR construction is determined by neither element alone, but the two together (p. 35), and that its interpretation is mainly determined by what makes sense in the particular context in which it occurs rather than by any strict syntactic criteria. Shi offers many other such inspired insights which must be left unmentioned here due to space restraints.

    It seems that this book went to press without much editorial or typographical quality control at all. Sometimes a point is repeated on the same page in almost the same words (e.g. p. 3). There are inaccurate or not very polished 'Chinese English' translations (e.g., 'He beat a fly to death.' [p. 1] and 'He patted one fly dead.' [p. 166] instead of the 'He swatted a fly [and killed it]/He killed a fly [by swatting it].'). There are many misspellings, both of English words and Pinyin Romanization, and tones are not indicated. There are incorrect characters, including some that apparently had been automatically converted from simplified to traditional form by computer, and the wrong option chosen; the word _lei4_ 'tired' is consistently written with an incorrect character. Some English words are split by a hyphen even though they do not occur at the end of a line. Margins and spacing are not always even. As it stands, this book requires a good deal of patience and indulgence on the part of the reader. Moreover, the book is overpriced, even considering its relatively limited market and the average prices of scholarly books these days.

    Still, the merits and original findings of this book, which are many and noteworthy, should not be overshadowed by its shortcomings. It makes a significant contribution to the study of resultatives in Chinese and their historical development -- I personally gained many valuable and useful insights and ideas from this book -- and it should be available in any institutional research library with a linguistics section, especially if it were carefully edited and reissued.


    Karen Steffen Chung teaches English and linguistics in the Department of Foreign Languages and Literatures of National Taiwan University in Taipei. Her chief current research interests are phonetics, prosody and Chinese morphology; see her websites at