Restoring "History" to Historical Study.

• Author: Tian, Xiaofei

Geographical writings and local history have continued to be engaging research topics, and the relationship between humans and our environment is more important than ever. In the past several years there have been a number of noteworthy studies of pre-Tang geography writings in Chinese, German, and other languages. (1) In this context, D. Jonathan Felt's book, Structures of the Earth: Metageographies of Early Medieval China, is a commendable attempt to fill a considerable gap in English-language scholarship by studying the geographical writings of early medieval China, a period situated between the Han and Tang dynasties and roughly lasting from the late second through early seventh century.

The book is written in a lucid style. The argument made by the book is straightforward: there is, the author contends, an imperial metageography, first developed by "Qin-Han officials and literati" (p. 4), that centers on the Sinitic civilization and on the imperial court; however, born out of the fragmentation of imperial order following the collapse of the Han empire, early medieval Chinese geographical writings focused on local regions and challenged that imperial metageography by developing four new metageographies. "Tang literati described these [pre-Tang geographical] texts negatively because they perceived them as detrimental to the imperial order" (p. 15), and so the present volume is a corrective both to that distorted picture painted by those Tang literati and to the imperial metageographical framework that, in the eyes of the author, still largely remains "the unconscious spatial framework employed by historians and Sinologists alike" (p. 258).

The first chapter defines geographical writings as a "new literary genre" and examines the kind of cultural work it does. It classifies these writings into six kinds--local regions, natural spaces, foreign lands, the world, capital cities, and other "minor subgenres"--and relates the story of how "Tang officials evaluated and reinterpreted early medieval geographical writing to serve their own imperial ideologies" (p. 85). I will have more to say about this chapter; here I will simply make a passing observation on the terminology. In this book, the word "literary" is used liberally throughout, but in many places it simply means "textual" in the sense of writing. Indeed, many of the geographical writings fit neither the early medieval nor the modern concept of "literature." It is worthwhile keeping in mind that the term dili [phrase omitted] very imperfectly corresponding to "geography," was a bibliographical category subsumed under "history" (shi [phrase omitted] ) in dynastic histories. The statement that "the single defining feature of early medieval geographical writing" was "its spatially organized textual structure," or that "all geographies organized their text as though moving through space" (p. 22), is rather dubious, because nearly all of the textual remains are fragmentary and reconstituted by Qing and modern scholars, and we simply do not know how those texts were originally structured.

Each of the four remaining chapters in this book explores one of the four new metageographies. Chapter two discusses what the author calls the metageography of ecumenical regionalism, by which the author means that many local geographies of the third to fifth century emphasize the unique contribution of local cultures to the Sinitic ecumene (p. 115). The third chapter, "North and South," describes how the metageography of northern and southern dynasties, or in the author's words, the Tabgatch empire and the Jiankang empire, (2) constituted a conflict with "the singular and absolute claim to imperial centrality" (p. 6) by posing two competing imperial centers, in contrast with the "two equal and complementary regional halves of a unified empire" as portrayed, the author claims, by Tang literati (p. 164).

The last two chapters focus on Li Daoyuan's [phrase omitted] (d. 527) commentary on the River Classic (Shuijing zhu [phrase omitted] ). Chapter four contends that Li Daoyuan resorts to the metageography of hydrocultural landscape that "prioritizes river systems as the foundational organizing framework for all other natural and human geographical patterns," which Felt sees as a reaction against the Han imperial metageography advocating the anthropocentric ideal of state control over nature (p. 165). Chapter five posits that Li Daoyuan embedded another organizing metageography, "an Indo-Sinitic bipolar world model," by explicitly equating Mount Kunlun and Mount Anavatapta (pp. 210-11). Felt argues that this world model was a disruption of Sinocentrism and allowed Sinitic literati to think about civilizations within a comparative framework "for the first time" (p. 7).

The book contains many keen insights and accurate observations, such as about the competition between the northern and southern dynasties or about the Buddhist disruption of Sinocentrism (even though whether we can indeed credit Li Daoyuan with a bipolar world model is another question). Yet, it has a tendency to overstate its case, and to construct arguments and marshal evidence on the basis of a preconceived idee fixe. Eager to demonstrate the early medieval geographical writings as a product of the fragmentation of imperial order and a way of resisting imperial order, the author simplifies the complicated ways in which imperial order works, and in the process both overestimates its power and underestimates it.

Nowhere did anyone--certainly not Ban Gu [phrase omitted] (32-92) in his Han shu "Dili zhi" [phrase omitted] [phrase omitted] , which is regarded by Felt as the central manifestation of Han imperial metageography--state that "the Han empire had successfully subdued regionalism" (p. 66), nor was that a reality, either. This is indeed a myth, not, however, propagated by the Han literate elite, but one ironically upheld by the author himself. To argue that a unified empire necessarily and successfully suppresses any expression of regionalism is an exaggeration, and to support it, the author often willfully overlooks any evidence that would point to another direction.

For instance, the Han author Yang Xiong [phrase omitted] (53 BCE-18 CE) produced "Fu on Shu Capital" (Shu du fu [phrase omitted] ) and The Annals of Shu Kings (Shu wang benji [phrase omitted] which are no different from many of the early medieval works of "ecumenical regionalism" discussed in chapter three, but neither work is mentioned in the author's discussion of the so-called Han imperial metageography. The author writes off the Eastern Han work Yue jue shu [phrase omitted] (translated as "Lost histories of Yue"), which is dismissed for exemplifying Han universalism and suppressing regionalism (p. 26), even though Yue jue shu very much evokes the fourth-century Huayang guozhi [phrase omitted] , a regional history of Sichuan, which is praised as a prime example of regionalism (pp. 108-14).

One question we might ask is, besides doing cultural work called for by their own time, what other possible reasons could account for the proliferation of geographical texts in early medieval China. Geographies were by no means the only kind of works that appeared in large numbers: from this period we have more writings than ever before about many subjects, such as the collections of anomaly accounts or of humorous anecdotes, biographies of clans, families, notable personages, and recluses, as well as personal letters. The transformative change in technology--the invention and widespread use of paper--was crucial in the dissemination and preservation of knowledge; not only were more writings produced because of the more...