The Roman Empire as known to Han China.

• Author: Pulleyblank, Edwin G.

In the centuries on either side of the beginning of the Common Era two great empires dominated opposite ends of the Eurasian continent, Rome in the west and Han in the east. They knew little of one another and had no direct interaction. Nevertheless it was at this time that some information began to filter through in both directions. Chinese silk, in particular, became known in the west and a long-distance trade through intermediaries developed across Central Asia, the famous Silk Road. In the Roman world stories, some based on fact though often much distorted in transmission. others completely fanciful, began to circulate about the Seres, that is, the Silk People. A little later the name Sinae, based, like Sanskrit Cina and our present China, on Qin, the name of the short-lived dynasty that preceded Hah and united China in 221 B.C.E., also appears in western sources. At the same time the Chinese began to hear about a country in the far west which they called Da Qin, Great Qin, apparently thinking of it as a kind of counter-China at the other end of the world. The surviving records of the accounts of Da Qin that they wrote down have naturally attracted much interest among Western scholars ever since modem scholarship on China first began to develop. A well-known monograph, China and the Roman Orient, published in 1885 by Friedrich Hirth, contains a selection of Chinese texts with their translations and an extended commentary. The Japanese scholar Shiratori Kurakichi wrote several extensive articles in the 1920s and 1930s; these were assembled and translated into English in volume 15 (1956) of the Memoirs of the Research Department of the Toyo Bunko. Other notable contributions were made by the eminent French sinologists Edouard Chavannes and Paul Pelliot and there have been many other studies on various aspects over the years. Professors Leslie and Gardiner now offer their reassessment of this material.

In view of the time that has elapsed since much of the scholarship mentioned above was completed and the complexity of the problems involved, a fresh look at the subject is certainly in order. Regrettably, it has to be said that this book does not succeed in this desirable objective and does little to advance our understanding. The way in which the book is arranged gives the impression that Hirth's antiquated work, which it seeks to replace, has provided the basic model. Thus, part A in ten chapters offers translations of Chinese texts related to Da Qin from Han down to Northern Song, together with a few pages in chapter ten on classical sources that seem to refer to China. Part B, entitled "Historical, Geographical and Linguistic Identifications," contains eleven more chapters on specific topics - "Historical Background," "The Earliest Information and Contacts," "The Routes," "The Cities of Ta-ch'in," "Vassal States of Ta-ch'in," "Jewels and Exotica: the Products of Ta-ch'in," "Some Special Problems," "Legends," "The Important Countries and Regions of West Asia," "Other Identifications," "Legendary Areas and the Far West." Finally there is a conclusion, entitled "Ta-Ch'in in the Roman Empire." There are three short appendices, on Fu-lin, Ta-Ch'in in Chu-fan-chih by Chao Ju-kua (a book of Southern Song date), and Ta-Ch'in in South-east Asia and India; followed by a list of Chinese dynasties, a chronological chart [of the time span covered by the book], a series of maps, lists of identifications of proper names by the authors and by other scholars, glossaries of key terms in the translation, and color names and a list of place names with their Chinese characters. The text proper is followed by bibliographies of primary and secondary sources and a full index.

This is all very impressive. Unfortunately, one's satisfaction diminishes as soon as he looks more closely. First of all, it is regrettable that the two authors did not follow Hirth's example and include the Chinese texts, preferably alongside the translations rather than in a separate section; this should not present insuperable difficulties to the printer, in view of the technological facilities that are available nowadays. By cutting out redundancies and intelligent cross-referencing it would, in fact, have been possible to include the Chinese texts and still reduce the length of the book, while giving it greater unity and coherence. In part B the authors often retranslate, in different wording, passages from the texts they have already translated in part A. An "Introductory Note" at the beginning of the book is evidently designed to anticipate such criticism: "We have attempted to make most of the chapters semi-independent. Some of them can thus be read as separate essays. As a result of this, there are a number of primary passages that we have translated in different chapters as well as in our Part A 'The Texts'. We have, however, refrained from giving copious notes to the Texts, postponing most of our comments until the later chapters." In fact, the "Texts" are copiously annotated and the retranslated passages are also copiously annotated. Not only does this waste much space but it makes it difficult, without a great deal of use of the index (which fortunately seems quite reliable but was presumably done by someone other than the authors themselves) and flipping back and forth for the reader to tell whether, as is by no means always the case, what is said in one place is consistent with what is said in another. Moreover, for some key points the authors do not provide argumentation in the book itself but refer back to a journal article: Leslie and Gardiner 1982. They do this, for example, in the case of a proposed rearrangement of the text of Hanshu 96A which plays a key role in their argumentation.

If Leslie and Gardiner could be shown to have made a significant contribution to the analysis of the various Chinese sources referring to the far west in Hah and post-Hah times, one would have to put up with the inconveniences in their presentation. Their argumentation is also very muddled, however, and their conclusions are mostly either banal or unconvincing.

To begin with, there is one important respect in which the two authors are clearly poorly equipped for the task. They are babes in the woods when it comes to Chinese historical phonology, an essential tool for identifying the many foreign place names transcribed in Chinese characters. In section 2.3, entitled "Some Philological Principles," they complain of the "often wild attempts of Miyazaki, Soma and many others to identify the various place-names by unwarranted distortion of the sounds of Chinese characters" and go on to say, "we need to use acceptable reconstructions of the ancient pronunciations." Yet there is little evidence that they have any conception of what this commendable principle means. They claim to have given Karlgren's reconstructions "for the most important cases . . . occasionally noting also suggestions by Pelliot, Haloun, Boodberg or Pulleyblank." In fact, even the insertion of Karlgren's readings is quite sporadic and often inaccurate, ignoring diacritics and making mistakes in phonetic symbols. To give only one example, for Zesan (which they mistranscribe in Wade-Giles as Ts'e-san), they give Karlgren's (Ancient Chinese) reading as "d'ek-san" instead of "d'Dk-san." The diacritic over the d- is all important in order to account for the modem affricate initial and the difference in the vowel is also not a trivial matter.

Inserting Karlgren's reconstructions is, in any case, of little value for those, like the present authors, who do not understand their basis or their limitations. The inadequacy of Karigren's system for understanding transcriptions of the Han period was, I believe, amply demonstrated in my first study of the problem (Pulleyblank 1962; see also 1983). References by Leslie and Gardiner to my work or that of anyone else except Karlgren are exceedingly rare, which is perhaps just as well, since they have no understanding of the issues involved. They obviously feel defensive about this. They make much of quotations they have found from various scholars who, not being linguists, have tried to cast doubt on the value of linguistic arguments in identifying place names, including Hirth, Maenchen-Helfen, and Daffina. For this purpose they even quote, out of context, a sentence of mine. In a review of China in Central Asia by Hulsewe and Loewe, I did indeed say at one point: "In the present state of knowledge about Chinese of the second century B.C. linguistic judgments can only be somewhat subjective" (Pulleyblank 1981: 284). As is obvious from the context, however...