Laurent Sagart Review: James A. Matisoff Handbook of Proto-Tibeto-Burman. System and philosophy of Sino-Tibeto-Burman Reconstruction. The book is meant to be a companion to the Sino-Tibetan Etymological Dictionary and Thesaurus volumes where the full cognate sets will eventually appear, but the first fascicle of the first volume, said p. After a general introductory chapter, the Handbook describes the PTB syllable canon.
|Published (Last):||1 May 2012|
|PDF File Size:||1.23 Mb|
|ePub File Size:||7.21 Mb|
|Price:||Free* [*Free Regsitration Required]|
Laurent Sagart Review: James A. Matisoff Handbook of Proto-Tibeto-Burman. System and philosophy of Sino-Tibeto-Burman Reconstruction. The book is meant to be a companion to the Sino-Tibetan Etymological Dictionary and Thesaurus volumes where the full cognate sets will eventually appear, but the first fascicle of the first volume, said p.
After a general introductory chapter, the Handbook describes the PTB syllable canon. The rest of the book is organized into chapters, each dealing with a piece of the PTB syllable: initial consonants, prefixes, rhymes, vowel length, final nasals, final stops, final liquids, final -s, suffixes. There is an imposing critical apparatus, including a copious amount of front matter, two appendices on OC by Z.
Handel and R. Cook , six indexes, and a reference list. The style is readable, somewhat chatty. Matisoff presents a genetic tree for ST p. Sino-Tibetan at "perhaps years BP", p. The binary, Chinese-vs. Benedict appears to have regarded TB as a subgroup partly because he had himself inherited this view and partly on the basis of cognate counts Benedict This view, though not implausible in itself, needs to be buttressed by evidence of unique TB innovations. Such evidence however, is still missing.
Recently van Driem has argued again without presenting evidence of uniquely shared innovations that Chinese and Tibetan belong in the same primary branch of the family. In a footnote fn. Claims made in the book can only be verified by checking the cognate sets, but finding the cognate sets is not straightforward. One must check all the index references for a particular reconstruction: the reference with the cognate set is not singled out. These problems discourage verification. After discovering the first elements of evidence of a genetic relationship between Chinese and Austronesian, where TB elements were not prominent, I claimed in a conference paper Sagart that Chinese stood closer genetically to Austronesian than to Tibeto-Burman.
That was an error. I now fully recognize that Sino-Tibetan is a valid grouping: I consider that ST as a whole, not just Chinese, forms a genetic unit with Austronesian. See Sagart b. The vast majority of ethnic Tujias speak Mandarin Chinese. My use of that term should not imply that I am presently convinced that it is a valid grouping. As for data, M. Yet Written Tibetan, the oldest TB literary language, with its abundant literature and convenient lexicography, is under-represented.
Because many TB languages are not well documented, the sound correspondences for all of them cannot be known, as Matisoff acknowledges elsewhere Unfortunately, when he gives cognate sets, he does not say which forms obey known sound correspondences, and which are included on the basis of educated guesses. Because the boundary between sound correspondences and guess-work is not marked in the cognate set, the reasoning which led M. Matisoff forgets that for Popper falsifiability goes on a par with explicitness.
Only explicit theories can be tested, and hence falsified. I cannot speak for Jones and Starostin but I can assure Matisoff that my Sino-Tibetan-Austronesian theory Sagart is very explicit on sound correspondences and makes predictions that can be tested linguisticallyin a number of ways. TB sound system of PTB Overall, the reconstructed system is inherited from Benedict; changes are in the direction of replacing phonological or morphological problems with untestable variation hypotheses: for instance syllable- initial alternation between velar stops and zero in languages like Lushai and WB, treated by Benedict as conditioned by prefixes, is regarded by M.
In the next section, a suggestion wll be made that the velar vs. Matisoff offers little discussion of the evolution of manner contrasts: on this central issue of TB comparison he refers the reader to the minimal, year-old chart of initial correspondences for WT, WB, Kachin, Garo and Lushei in STC Benedict thought two manner types for stops voiced vs.
His chart of correspondences makes testable predictions on the manner of articulation of Tibetan and Burmese and, to some extent, Lushei stops, but basically excludes no stop reflexes at a given point of articulation in Garo and Jingpo. Matisoff maintains the same range of assumptions, but treats root alternation, not mentioned by name, simply as lexical variation at the PTB level.
In Sagart , I have argued that the transitivity-related voicing alternation, at least, is better explained as induced by a prefix intransitive m-, see below than as lexical variation in PTB, and even in PST. Every effort should be made to replace them with explicit prefixation hypotheses. Matisoff earlier produced such a hypothesis for Burmese-Lolo. It would have been interesting to know how similar proposals fare in other TB languages or branches. Matisoff gives a proto-system of 23 initial consonants.
Discussion is speedy and not quite to the point. The reader is invited to admire e. The reader who wishes to find what the reflex of a particular initial is in, say, Tibetan, must to turn to the index of reconstructed roots, look up each root beginning with a particular initial, and check the text for occurrences of that root in Tibetan.
There is a blanket statement on p. Here Matisoff has missed a sound correspondence Table 1. WB w- corresponding to WT g-. Note: The forms in bold type are not cited by Matisoff.
Unexpected alternation between labial stops and w-, widespread in TB languages e. Later 23, fn. This is presumably because he cannot state the conditions under which the glide arises. This is not felicitous. Matisoff treats it as just variation, but another prossibility, adopted in Peiros and Starostin , is that this alternation reflects earlier uvular initials. This part is much more detailed and explicit than with initial consonants, a most welcome advance from the STC. Here reflexes of the posited proto-rimes in some important languages usually Tibetan, Burmese, Jingpo, Lahu, Lushai, and a representative of the Bodo-Garo group, sometimes also Mikir, Nung are tabulated and examples are given.
The majority of the rime correspondences presented in the book appear credible at first sight, but the uncertainty on initial consonants and on tones below weighs on the comparisons. Tibetan, again, is under-studied thus four out of thirteen diphthongal TB rimes in Table 14 p. The origin of Vietnamese and Chinese tones has been successfully explained by Haudricourt a,b as resulting from the loss of final laryngeal segments around years ago.
It is highly likely that tonal TB languages acquired their tones in similar ways. Benedict saw a basic two-tone agreement between Chinese and some tonal TB languages. It increasingly seems that the contrast observed by Benedict is real, though phonetically it was probably a contrast between sonorant endings followed by a glottal stop and and sonorants not followed by a glottal stop.
He is aware of the problem p. Vowel length Especially on the basis of Lushai, Benedict reconstructed a PTB vowel length contrast, with marked long vowels. He reconstructed few words with long vowels, however. Matisoff reconstructs many more. This will be of particular interests to students of Chinese, because it has been claimed by Zhengzhang and Starostin that the TB length contrast correlates with the Chinese distinction between type A and B syllables type B syllables are those in which a medial yod appeared in Middle Chinese.
At first sight, the TB length distinction as reconstructed by Matisoff does not correlate particularly well with the Chinese distinction: thus, in the pair of TB rimes -uk and -u:k pp.
I have not conducted a full investigation of the correlation between the two features, however. An intransitive nasal prefix m- was reconstructed for TB by Wolfenden , and this is maintained by later writers, including Matisoff.
Sagart , , , identified the corresponding prefix in Chinese: N-, preserved as prenasalization in early loans to Hmong-Mien. This prefix was first reconstructed for Old Chinese by Sagart but the reconstruction appears to hold good for the whole of ST. Thus Matisoff p. Other prefixes. Matisoff gives more detail than the Conspectus, but his discussion is hardly complete.
Jacques p. It probably does not exist. A new "mysterious" and "semantically elusive" suffix introduced by Matisoff is -k p.
This is better understood as a root- final consonant, lost in conditions that have not been identified. When roots where Matisoff supposes -k have Chinese cognates, Chinese normally has -k. This must be because the form with -k is basic. However, while Lushai tl- can only reflect an earlier gl-, WB ky- can reflect gr- as well as gl-. Gyarong, which distinguishes gr- and gl- Jacques has the same pair as WB: cf.
In effect Matisoff is using empty suffixes as wild cards. Despite the methodological problems described earlier, TB reconstructions at times converge with OC reconstructions, more that M.
There are also problems, discussed below. Undetected loans, anachronisms. Matisoff has no safeguards against loans. Loans from Chinese are not recognized: Matisoff p. This is hard to believe. The oldest indigenous writing system in all of East Asia, from c. Tibetan, the oldest ST literary language outside of Chinese, was not reduced to writing until c. How can PST speakers have known writing brushes?
The isolated WT form, with tell-tale final -r, is without a doubt a late borrowing from a variety of NW Chinese where -t had become -r in the late first millennium CE Coblin 55 for the sound change. Despite his anti-megalocomparativist stance, Matisoff is himself a half-believer in Austro-Tai.
Handbook of Proto-Tibeto-Burman
The majority of the rime correspondences presented in the book appear credible at first sight, but the uncertainty on initial consonants and on tones below weighs on the comparisons. Cooksix indexes, and a reference list. Matisoff gives more detail than the Conspectus, but his discussion is hardly complete. Skip to main content. I consider that ST as a whole, not just Chinese, forms a genetic unit with Austronesian.
University of California Press
Related Books About the Book This page volume is a clear and readable presentation of the current state of research on the history of the Tibeto-Burman TB language family, a typologically diverse group of over languages spoken in Southern China, the Himalayas, NE India, and peninsular Southeast Asia. The TB languages are the only proven relatives of Chinese, with which they form the great Sino-Tibetan family. The exposition is systematic, treating the reconstruction of all the elements of the TB proto-syllable in turn, including initial consonants Ch. III , prefixes Ch.