Journal of the American Oriental Society 130.3 (2010) 361 The Production of Unpleasurable Rasas in the Sanskrit Dramas of Ārya Kṣemīśvara Adheesh Sathaye University of British Columbia Writing at the Kannauj court sometime around 915 c.e., the dramatist Ārya Kṣemīśvara would have found himself at a remarkable moment in the history of Sanskrit literature. It was a time of great innovation in literary theory, not least of which was a “paradigm shift” from alaṃkāra (iguration) to rasa (emotional lavor) as the fundamental unit of poetic analysis. This was part of a larger uniication of poetics (alaṃkāra-śāstra) with dramaturgy (nāṭya- śāstra)—long-independent intellectual pursuits that would be deinitively brought under one umbrella a little more than a century after Kṣemīśvara’s time. At the very heart of these changes lay Ānandavardhana’s (c. 850) proposal of a new type of signiication—sugges- tion (dhvani)—which, unlike denotation (abhidhā) or indication (lakṣaṇā), was not an isol- able feature of the poetic text, but a phenomenon involving the sensibilities of the literary connoisseur. 1 Ānandavardhana’s Dhvanyāloka proved to be a landmark text, forcing practically every theorist after him to confront what we, following Roland Barthes, might call the “writerly” nature of rasa. In S/Z (1970) Barthes approached the question of textual interpretation by describing the “writerly” (scriptible) text as one in which the reader must act as a kind of “writer” in order to produce meaning. He distinguished this from the traditional idea of the static, “readerly” (lisible) text, in which the reader may only be a passive receiver of a meaning preigured by the original (and authoritative) author. In the French intellectual and political context of the 1960s and 70s, the writerly text represented for Barthes and other newly “poststructural” theorists the means through which literature might achieve an eman- cipatory goal, “to make the reader no longer a consumer, but a producer of the text” (Barthes 1974: 4). 2 In the middle of the ninth century Ānandavardhana produced an equally liberative moment for Sanskrit poetics, opening up the ield to a new, overarching goal: the experience of rasa (McCrea 2008). Rasa, of course, was nothing new; before the advent of dhvani, however, most critics regarded it to be an intrinsic feature of a poem or play to be appreciated during performance. Thus, for Lollaṭa (early ninth century), this aesthetic lavor ( rasa) was an aug- mented or enhanced stable emotion (sthāyibhāva) of a character and shared by the actor play- ing the part; for Śaṅkuka (late ninth century), rasa was also located in the character but only imitated by the actor (Warder 1972–2004, vol. 1: 36–37; Kane 1971: 370–71). Intrinsic to these models was the proposition that rasa, like any other igure, belonged to the “readerly” text and not to the spectator, who only consumed it. Ānandavardhana’s approach, in contrast, 1. For discussions on the uniication of Sanskrit poetics and dramaturgy, see Tubb 1998: 58; Gerow 1977: 256; McCrea 2008: 44–45. For analyses of the novelty of Ānandavardhana’s dhvani concept, see McCrea 2008: 162–63; Pollock 2001: 200; Gerow 1977: 252–53. 2. Barthes’s concept of readerly vs. writerly texts grows from his study of the “Death of the Author” (1977 [1968]), in which he had irst attempted to loosen the authoritative grip of the author (now simply seen as a “scrip- tor”) on textual meaning; it would culminate in his Pleasure of the Text (1975 [1973]), in which he proposed that writerly texts generate a diferent, more ecstatic kind of enjoyment (jouissance) than the pleasure (plaisir) of read- erly texts.