LITERARY HISTORY. INDIAN HISTORY, WORLD HISTORY 113 informing questions no longer presuppose European primacy and ask. How did the West develop and why didn 't the India develop that way, but rather will be generated from Indian materials themselves and ask. How did India develop, and what might this tell us about the West (and other places)? Given the long histories they enable us to follow—of polyglossia, technologies of the word, practices of reading, speaking, writing—these materials might help transform a lifeless knowledge-practice known as "history of literature" into a more vital, socially embedded "history of literary cultures."1 The great challenge is to figure out good questions to ask, questions adequate to this complex, deep, and I believe unique body of material, and the right theoretical framework within which to ask them. What I want to try here is try to identify a few of these problems—which we are far from sovling but cannot solve until we identify—and suggest why they are important; hence a speculative essay, with far more questions than answers. The problems chosen are those of the outside of the text, of communicative contexts and practices, without which the inside of the text must remain unintelligible for any historicist understanding of literary discourse, which always has primacy in critical scholarship. One of the most important but also one of the hardest questions thrown up in the first instance by our South Asian materials seems to me to concern the conditions of possibility for the very commencement of a literature. Do new literatures enter the world, and if so, what can we say about how and why? It is a question that seems rarely if ever to get asked in any of the many literary histories published over the last twenty years—not only those of Indian languages and literatures but of most others.2 Perhaps this is due in part to the ambiguity of its terms: What do we mean by "literature," and who gets to define it; what do we mean by "commencement," and who gets to decide it? Yet the complexities of the question should not dissuade us from asking it; they should in fact form the substance of the inquiry. The alternative is to empty literary cultures of their history, to render them timeless or facts of nature. Whatever else it may turn out to be, literature is an intentional phenomenon, produced by human agents in changing but determinate conditions, with changing but determinate models of the literary, technologies, languages, and textual communities related to its production. These make up much of what a history of literary cultures should be charged with recovering. I will have less here to say about "conditions of possibility," paramount in importance though these are, because we can get to conditions only once we figure out what phenomena are conditioned, "the very commencement of a literature." What I want to probe critically is the proposition, banal at first sight, that literatures begin—to explore, that is, how at certain times and places certain kinds of language come to be deployed in certain new ways, as never before in their histories, for making certain kinds of texts which for reasons I will try to specify we may call "literary." Although I am speaking here of the commencement of everything from a given genre (from the "nataka" to the "novel") to traditions as such ("Telugu" literature or "Persian" literature), I concentrate here on the latter, which require the most careful handling. Like a poetic form or a genre, a literary tradition itself may not be always-already existent, but begin; that in a culturally-specific sense