Chinese narrative poetry, the late Han through the T'ang dynasties . By Dore J. Levy. pp. xiv, 225. Durham, N.C., Duke University Press, 1988. £26.00.

This book marks a new departure in the study of traditional Chinese poetry. The author has subjected four comparatively long shift-lyric poems to an analysis specifically of their narrative features. For all its simplicity, this imaginative approach produces interesting insights into these familiar pieces. They are "Poem of Affliction" (Peifen shift) by the late Han poetess, Ts'ai Yen, "Song of Everlasting Sorrow" (Chiang hen ko) and "Ballad of the P'i-P'a {Fi-p'a hsing) by Po Chii-i, and " Song of the Lady of Ch'in " (Ch'in fu yiri) by Wei Chuang, both of the midto late T'ang. The fact that the first piece dates from the late Han and the last three from the T'ang means that the title of this book is a misnomer. A more accurate designation would be: A Study of Major Narrative Features in Four Chinese Shih-Poems of the Late Han and T'ang. The title, "Chinese Narrative Poetry", suggests that the author's emphasis lies in the poetry of each of the four pieces as an individual, artistic whole. On the contrary, the author has chosen to focus on various narrative elements in three separate chapters, analysing the four poems piecemeal within these chapters. The result is fragmentary and truncated, cited passages and critical comment awkwardly disconnected. I would have preferred four separate chapters on one narrative poem at a time, each chapter opening with a full translation of it, and proceeding to discuss its common and unique narrative aspects. Another disputed point arises from the author's title, and from her heading of Chapter One, "Narrative elements in traditional Chinese poetics". In this chapter Dr Levy reviews early Chinese statements on literature and poetry with a particular reference to her topic of narrative art. This chapter heading, her book's title, and sub-title suggest to the reader that the discussion that follows in Chapters Two to Four will reveal some historical awareness of her subject. This is not the case. In fact, incomprehensible as it might seem, Dr Levy's analysis of the four pieces is conducted with no apparent knowledge of and scant reference to the enormously important and valuable repertoire of narrative verse among pieces known as yiieh-fu, or Music Bureau pieces. Significantly, the seminal and pioneering work of Jean-Pierre Dieny, 1968, on yiieh-fu, though cited in the Bibliography, thereafter remains lost to the discussion in her book. It is not even clear whether this book has been read by the author. Had it been assimilated, Dr Levy would have been able to find very many more narrative qualities native to the Chinese oral tradition replicated in the later pieces she deals with. As it is, the narrative qualities she does highlight are those of the Western literary tradition, in the main, such as point of view, description (Chapter Two), character types and roles (Chapter Three), temporal and other sequences (Chapter Four). This lack of historical perspective in Dr Levy's book makes spurious her claim that it constitutes "A model for the study of Chinese narrative poetry", the title of her Introduction, and her claim in the Epilogue that" I have attempted to provide a theoretical basis for a literary history of Chinese narrative poetry" (p. 124).