When the archaeological expeditions led by Marc Aurel Stein, Paul Pelliot and Albert von Le Coq to eastern Turkistan in the early twentieth century returned to Europe, they brought with them a wealth of manuscripts in dozens of languages, several of which were previously unknown to scholars. Amongst these newly rediscovered languages were two that came to be known as Tocharian A and Tocharian B. Scholars now view this designation as a misnomer, since the languages in question were connected with the inhabitants of Turfan, Kucha and other cities in eastern Turkistan, not with the territory of Tocharistan, in northern Afghanistan.
As the author briefly explains in the Introduction to the book under review (an adaptation of his Master's thesis), Tocharian A and B differ enough from each other that they should be considered separate languages, rather than dialects of the same language. Of the two, Tocharian B was still a spoken language when the extant manuscripts were written, but Tocharian A was used only in Buddhist liturgical contexts. The Tocharian B texts are of particular interest, since they exhibit numerous variant forms. These were initially explained by Werner Winter as representatives of three regional dialects of Tocharian B: western, central and eastern.Footnote 7
Winter's theory was challenged by Peter Stumpf, in his 1976 Habilitationsschrift, published posthumously (by Winter) in 1990.Footnote 8 Stumpf claimed that, rather than regional dialects, the variants represent distinct stages in the development of the language: “stages I and II, further divided into IA and IB, corresponding to the western dialect, and IC, corresponding to the central dialect; stage II corresponds to the eastern dialect” (p. 19).
As Peyrot notes, despite important research that has been carried out on the Tocharian B corpus, “many questions of both the synchronic and diachronic analysis of Tocharian have not been definitely settled” (p. 21). In light of this, the author sets out to examine phonological, morphological, syntactic, lexical and orthographic variations in the Tocharian B corpus within the framework of Stumpf's thesis.
Although Peyrot uses Stumpf's model as the framework for how the Tocharian B variants relate to each other, he does not merely repeat Stumpf's work. Rather, he makes use of a much larger corpus of texts than Stumpf did (Stumpf's work was based primarily on texts in the Turfan Collection in Berlin, whereas Peyrot has also consulted manuscripts found in Paris, London, St Petersburg, Japan and China). As a result, the number of examples of variation, especially in morphology, is expanded considerably from Stumpf's earlier study. With the expanded corpus, Peyrot aims “to see to what extent this variation can be explained with Stumpf's model”. (p. 24)
Before embarking on his systematic analysis of variants in the corpus, the author outlines his method in Chapter 2, including 1) his use of the terms “archaic”, “classical” and “late” to describe the stages in his adaptation of Stumpf's scheme; 2) difficulties in classifying texts which include forms from more than one stage; and 3) the special case of metrical texts, where “archaic” forms may reflect this literary genre, rather than stages of linguistic development.
Chapter 3, the bulk of the book, is taken up with a listing of all known Tocharian B variants, subdivided into Phonology (37 examples), Morphology (53 examples), Syntax (2 examples), Lexicon (19 examples) and Writing and Spelling (5 examples). Peyrot deals with the material in an exhaustive and competent manner. Each subsection in the chapter is methodically broken down into similar features (e.g. under Morphology, variants affecting nouns, pronouns, numerals and verbs are each clustered together).
Examples are meticulously referenced to the manuscript numbers listed in the appendices at the back of the book, including line numbers. Although the average reader is unlikely to read through this chapter from start to finish, the data assembled forms a valuable reference for those who wish to investigate any of the particular variants in more depth. The chapter concludes with an explanation of the author's criteria for classifying the texts listed in the appendices as archaic, classical or late. Finally, the variations discussed in the chapter are handily summarised in a chart on pp. 183–185.
Chapter 4 is devoted to Peyrot's interpretation of the data he has assembled. Here, the author concludes that the evidence he has compiled not only upholds Stumpf's theory of different linguistic stages in Tocharian B, but also strengthens it. At the same time, Peyrot notes instances where he differs from Stumpf's original interpretation of certain variations. This is followed by an outline of the way in which the different stages of Tocharian B relate to each other, again handily summed up in a diagram on p. 191.
In the light of his analysis of the corpus, Peyrot concludes that Winter's theory of three dialects of Tocharian B cannot be maintained, although he does concede the “possibility that the late language was in fact a dialect in the sociolinguistic sense” (pp. 192–193). He also supports Stumpf's conclusion that the original home of Tocharian B was Kucha, where all stages of the language are attested. The author examines Stumpf's assertion that the existence of impure Tocharian B texts, containing elements of more than one stage, indicates that they occurred contemporaneously and were therefore probably different registers of the same language. Again, the assembled data seems to support this theory, although Peyrot disagrees with Stumpf's suggestion that the archaic language constituted a higher and more prestigious register than the classical language.
The results of palaeographical analysis and carbon dating of Tocharian manuscripts are used by Peyrot to assign rough dates to the three stages of Tocharian B (archaic, classical and late), again summarised in helpful diagrams on pp. 205–206. This is followed by an examination of Stumpf's theory that the classical language was an artificial construct meant to bridge the archaic and the late or colloquial language, a position that the author dismisses as too radical. The chapter concludes with a brief discussion of the relationship between Tocharian A and Tocharian B.
Chapter 5 summarises the findings of the book, comprising a reiteration of the contents of the previous four chapters, followed by a summary of the major findings, linked to the original aims set out in Chapter 1. This is followed by several maps of the areas where the Tocharian manuscripts were found and a series of appendices, comprising the lists of manuscripts mentioned above (subdivided into Berlin, Paris, London and Other) and a final list of text types, where all texts belonging to each of the different stages are grouped together for ease of comparison. A Bibliography and a list of all Tocharian B words mentioned in the text complete the volume.
The book is masterful in its command of the material, but is written in such a way that its primary audience will undoubtedly be limited to specialists in Tocharian or other ancient Central Asian languages. Although I recognise that the work is primarily a linguistic study, the historian in me would have preferred more explanation of the historical, social, religious and political context in which Tocharian B was spoken. Thus, when I read an exchange along the lines of “«Give me that [hearer sphere] porridge as alms!» They said to him, «This [speaker sphere] porridge is not to be eaten by you.»” (p. 122), I am more interested in knowing what was going on in the social relationships between the interlocutors than I am in the specific morphological point that this text exemplifies. However, in order to stick to his aim, the author understandably could not delve into such matters.
The book's strong points include its exhaustive approach to the Tocharian B corpus, the extensive appendices and the index of Tocharian words cited. However, given the subject matter, it cannot be described as a fascinating read. Perhaps this is unavoidable with linguistic studies such as this. The author's use of diagrams and charts is helpful, but more could have been incorporated in order to help the reader, especially when new ideas are introduced, such as Stumpf's theory of three stages (p. 19). In addition, especially for any non-specialists who might consult the work, a table of all transcription symbols used, with their IPA equivalents, would have been a useful addition (particularly for characters not commonly encountered, such as ś). On a minor point, it also seems odd that place names in the maps were not capitalised.
Finally, the book suffers somewhat from not having been well edited by a native English speaker, so that unfortunate spelling and grammatical errors are scattered through the text: “The designations ‘A’ and ‘B’ sticked to these varieties” (p. 15), “1.2 Incomplete Language Discription” (p. 24), “Syntactic variation is very difficult to asses” (p. 158), “a layer of classical texts with few or no influence of the late language, (p. 208)” etc. However, although these are irritating, they do not impact the overall value of the book for all who are interested in the complex linguistic history of Central Asia.
