Review: Cantonese: A Comprehensive Grammar.
书评: 广东话语法
Arthur Holmer

Abstract 摘要
Cantonese is spoken by between 50 and 70 million people worldwide (depending on which source is cited), is the dominant language of the economically very successful territory of Hong Kong, and is enjoying a rising importance in southern Mainland China. Further, Cantonese exerts, through the medium of Canto-pop and other types of popular culture, a substantial cultural influence across the whole Sinosphere, second only to its close relative, Mandarin. For these reasons, Cantonese is undoubtedly one of the largest and most important languages in the world, regardless of which criteria are used to determine this. Nevertheless, it has often in popular consciousness been overshadowed by Mandarin, and erroneously even been demoted to the status of dialect. It is therefore particularly gratifying to see that Matthews and Yip’s Cantonese: A Comprehensive Grammar, originally from 1994, has now reappeared in a second, revised and improved edition. This is good news for all Sinologists interested in this fascinating language, as well as for typologists working on differences between Sinitic languages. The language described in this grammar is Hong Kong Cantonese, which is the culturally most influential variety, and also the variety spoken by the authors themselves. The present edition includes written Cantonese characters alongside the transcription. This is a valuable addition to the book, not only for the reasons outlined in the Introduction (to enhance readability for readers knowledgeable in Chinese, and to allow ambitious readers to learn written Cantonese). For readers with some knowledge of Mandarin but with little or no understanding of the different sound changes which have taken place in the Sinitic languages, the use of Cantonese characters gives the reader a rough clue as to which Cantonese words are cognate with corresponding Mandarin words and which are not (or perhaps, which are perceived by the Cantonese-speaking community as cognate, as there are a few mismatches). Further, the illustration of written Cantonese, as it is seen in authentic popular media and internet culture, is a neat counterexample to the myth that Mandarin and Cantonese are simply two spoken dialects sharing a common written language. In fact, much of the contribution made by this work (or any good description of Cantonese) involves a certain degree of myth-busting. Some truths can never be repeated too often, e.g. that the Cantonese tonal system involves six phonemically distinct tones in Hong Kong (and seven in Guangzhou), not nine, ten or even eleven, as sometimes reported. Similarly, the perceived structural similarity between Mandarin and Cantonese is often exaggerated, since the formal register of Cantonese contains countless verbatim morpheme-for-morpheme loans of entire constructions from Mandarin, as the authors point out. The best antidote for these kinds of myths is a detailed and well-structured description of the actual facts, and the authors generally succeed well with this ambition.