A Critical Re-edition and Annotated Translation plus an Electronic Dictionary and Grammar of the Earliest Tamil Literary Corpus

Eva Wilden, EFEO Pondicherry - Jean-Luc Chevillard, CNRS Paris

Objective:

What are the remaining authentic sources for Classical Tamil Literature of the early first millennium and how far do they allow us to reconstruct a more faithful picture of the language and literary texts of that time? Since the texts of the so-called Cankam corpus are even today the main witnesses (apart from a very limited number of brief inscriptions) of the history and culture of South India roughly up to the 5th century A.D., they are the most important source texts for any question pertaining to that period, and for a great many after that. As this simple fact has long been recognized, one would expect these texts to be widely studied, well-edited and made accessible to non-specialists of the language in the form translations. This is, alas, not the case. Even if we take into consideration the fact that the corpus was nearly forgotten and had to be resurrected from oblivion at about the end of the 19th century  so that the actual time of philological ground work has not been much longer than a hundred years (and Tamilists have always been few in number)  the scholarly output of that period of work has to be termed unsatisfactory:

  • 1. The art of text edition has gone from preliminary to worse. Instead of collecting the remaining witnesses, the precious manuscripts which in the South Indian climate are disintegrating so fast, and providing editions which give at least all the extant evidence, we are today faced with cheap reprints and (mostly) single-manuscript editions all deviating from each other without giving any account for that fact. In other words, we hardly know which are precisely the texts we are dealing with. There is not a single edition of an Old Tamil text which deserves to be termed critical, and today there are hardly a handful of scholars in the world who would have enough knowledge and experience to do a better job. It is hardly surprising that the translations, which are based on such texts cannot not be called translations in any philological sense of the word. Most of them are an unfortunate cross between a literary translation and a reproduction of the, in nearly all cases, modern commentary. They are completely unreliable and, as a rule, do not lay open their working procedures. A tradition of philological annotation, as known from other disciplines, is nearly totally missing.
  • 2. The language of the Cankam is as yet imperfectly understood. Some work at least has been done on morphology, though not yet passing beyond the impressionistic state; there are no statistics which would open the path to the investigation of historical developments within the corpus and the language. Our knowledge of the syntax is very inadequate. This is one of the severe obstacles not only for understanding the texts as they are, but also for editing them, since the idea of what is possible and what is not can at best be called hazy. Semantics is one further weak point. We have a number of indexes  the existing digital indices are based on one edition that cannot be called standard and give a quite distortive picture of the language  and these are an invaluable tool for research into morphology and syntax, but most of them refrain from adding a meaning to the word/form in question (just listing the occurrences). Thus our idea of meanings and shades of meaning does not go far beyond the TAMIL LEXICON which was basically a first tentative. Veritable semantic studies are nearly totally absent.
  • 3. The main interest in Cankam literature has been historical in a broad sense of the word: the history of the Tamil country, of Tamil religion, Tamil culture and folklore, and of Tamil literary concepts (Cankam love poems being among the finest specimens of their kind preserved in world literature). And when we take into account the fact that these texts are still more or less the only (at least by far the most extensive) source available for the Tamilnadu of the early Christian centuries, this hardly comes as a surprise. But of how much actual value is the exploration of sources so imperfectly understood? To make matters more complicated, there is an enormous mass of commentary and subsidiary literature (dating, however, hardly further back than to the middle ages) which has been made use of quite liberally. In other words, our knowledge is, for the time being, a nearly indissoluble mix of sources from roughly 18 centuries  or even 2000 years, if we include the modern commentaries, which in their turn have made use of and often mirror the medieval material.
  • Relevance to the Pondicherry institutions:

    The joined forces of the IFP and EFEO have played a very active role in the exploration of Classical Tamil. In accordance with the general orientation of the institutions, the research into South Indian religions with an emphasis on Saivism, the approach has not been guided so much by chronological considerations  which could have meant, for example, studying the oldest texts first  but by content. Thus one important French translation has been that of the ParipATal (F. Gros 1968), one of the markedly later Cankam anthologies, which can be seen as a link between Cankam and Bhakti (devotional) literature, a field for which there are several important Pondicherry publications, among others the edition of the TEvAram, both in print and recently on CD. But interests were wider. The dire need for a reliable lexicon was perceived early on, and a foundation has been laid by the collation of a word index of the whole Cankam literature, including most pre-medieval literary works apart from the devotional ones, on no less than 300 000 files, which give not only all the single word occurrences, but also the meanings as adduced by the commentary literature. This work has only partly been published, namely in the form of a three-volume word-index which does not give meanings. Both publications concerning Cankam, the text edition and the word-index, date back to the late sixties.

    Method:

    In order to provide a stable philological basis for the study of Old Tamil documents we first of all need access to the sources, which means, in a South-Indian context, palm-leaf and paper manuscripts. Pondicherry is most conveniently situated for collecting manuscripts at last (a thing that very soon might not be possible anymore, because the writing material is deteriorating fast). Most of them are supposed to be found in Tamil Nadu itself, some in Kerala and a few stray ones have gone as far as Calcutta. Since 2002 the photographer team of the EFEO has managed to digitise more than a hundred Cankam manuscripts. By now the collection has entered its final phase, though with the fresh searches going on in South India at this time it is not to be excluded that further material as yet might turn up. Those manuscripts will be generally available in the library of the EFEO, predominantly in order to provide the material basis for better editions, but hopefully also in order to stimulate a re-thinking and subsequent improvement in the editorial practice of the whole field of Tamil studies (an idea well in accordance with the Sanskrit policy followed by the present head of the EFEO center in Pondicherry, Dominic Goodall).

    The logical next step are re-editions themselves, that is, critical editions of the whole Cankam corpus: editions based on all the material still available  which includes not only manuscripts, but also the extant editions (which may partly be based on manuscripts no longer available) and the often extensive quotations in the ancillary poetological literature (the editions of which, to be sure, are also in dire need of improvement...)  and giving a faithful account of the testimonies. Since the whole corpus comprises only about 30 000 verses, such a work is voluminous, but by no means interminable. In fact we hope to achieve it within the next eight to ten years. The first edition of this kind is about to be published, to more are expected to come out in the following year. They will be accompanied by (hopefully) more adequate, at any rate properly annotated English translations which try to give an account of the state of affairs and the present lacunae in scholarly knowledge.

    Hand in hand with that work a new word index and a grammar are under preparation. Since the whole is planned and conceived as a digital database, it will be easily possible to improve the single entries along with the developing editions. With these three steps, collecting the material, providing critical editions and generating a Cankam lexicon and grammar, we would have, for the first time, a reliable basis not only for linguistic (most urgently syntactic and semantic) studies in Old Tamil language, but also the necessary foundation for all the studies in history, anthropology, literature etc., that nowadays have to remain so frustratingly vague and impressionistic (or, worse, completely apodictic).