Considerations of Genre

Let me first say a few words about the concept of genre as it is to be understood in this paper. Few people today, if any, would subscribe to a prescriptive theory of genres and urge writers and scholars to approximate some "Platonic" notion of pure forms.^[4] Rather, those who still think that generic distinctions are indispensable to both writers and literary scholars generally adopt a descriptive theory of genres. That is, they usually derive the sense of literary kinds from existing works of literature and tend to allow the mixing of established generic norms.^[5] As E. D. Hirsch, Jr., has aptly observed, the "essential elements of all

[4] See Allan Rodway's discussion of "two inclusive classes of kind-criticism" in his article "Generic Criticism: The Approach through Type, Mode and Kind" in Contemporary Criticism , ed. Malcolm Bradbury and David Palmer, Stratford-Upon-Avon Studies, no. 12 (London: Edward Arnold, 1970), p. 87. René Wellek and Austin Warren have observed that classical theory of genre is usually "prescriptive" while modern theory is by contrast "descriptive." See their Theory of Literature , 3d ed. (New York: Harcourt, Brace & World, 1956), pp. 233–35.

[5] Allan Rodway, "Generic Criticism," p. 87.

genres are historical and culture-bound,"^[6] and the evolution of genres within any given literary tradition is always a complex and dynamic process, intimately related to changing historical and cultural contexts. The creativity of individual writers also plays a significant part in the dynamic evolution of genres. But individual contributions are themselves closely related to the complex historical and cultural conditions in which the writers lived. In this discussion, therefore, genres are to be regarded not "as a hierarchy of fixed forms, each individual work belonging to one and only one such category," but "more flexibly as clusters of stylistic and thematic traits that a number of works hold more or less in common and that change irregularly over the course of time."^[7]

It is true that genres were usually conceived by scholars in Sung times as a sort of "hierarchy of forms" and were used more often than not for prescriptive or evaluative, rather than purely descriptive, purposes. Nevertheless, the traits and conventions associated with genres are by no means always rigidly fixed in their formulations. The sense of a genre's identity changes over the course of time in accordance with the evolution of a whole literary culture; certain traits remain always essential to a particular genre's identity from the time of its emergence, while others lose their vitality in the genre's history and are replaced by or integrated with traits from other genres. Thus, the formation of the self-identity of the song lyric can best be approached from an evolutionary perspective. Although the song lyric never stopped evolving during the more than four centuries that elapsed from the early T'ang to the end of the Northern Sung, two periods were of particular relevance to the formation of its generic identity: the period around the year 850, when the song lyric began to evolve in the hands of literati poets into an "independent literary genre,"^[8] and the late Northern Sung, when the song lyric reached a mature and sophisticated stage of development that allowed a self-conscious generic identity to appear. These two periods will serve as the foci of my attention, even though I will be discussing the evolution of the early song lyric as a whole.

[6] E. D. Hirsch, Jr., Validity in Interpretation (New Haven: Yale University Press, 1967), p. 107.

[7] One can find a brief but lucid and sensible discussion of the concept of genre in Lawrence Buell, New England Literary Culture (Cambridge: Cambridge University Press, 1986), pp. 16–17.

[8] Kang-i Sun Chang has discussed this important development in early tz'u in her book The Evolution of Chinese Tz'u Poetry: From the Late T'ang to the Northern Sung (Princeton: Princeton University Press, 1980), pp. 29–32.