Review Hart of Kavya in South India

[Tieken, H.J.H.]

A propos G. Hart's review of the book Kavya in South India : Old Tamil Cankam Poetry (Gonda Indological Studies X), Groningen 2001.

Though it is generally unwise to write a reaction on a review in this case I feel I have to make an exception because in JAOS 124/1 (2004, appeared 2005), 180-184, Hart has managed to present what looks like a travesty of my book

In Kavya in South India I argue that the current dating of Old Tamil Cankam poetry in the first centuries of our era is untenable. Instead, I try to show that the literary corpus dates only from the eighth or ninth century and that it consists mainly of adaptations in Tamil of certain genres known from Kavya literature and in particular genres written in Prakrit or Apabhramsa.

According Hart the new, late, date of Cankam poetry I propose is based on four "claims". One of these claims is, in Hart's words, that the poems have the same style as Sanskrit prose Kavya and so must be derivative (my italics). Indeed, presented in this way my argument must sound nonsensical. However, Hart does not mention that before that I have tried to show that the scenes of the poems are fictive. More in particular I have argued that at the time of their composition the so-called heroic poems aimed to evoke scenes from a heroic past in which bards wandered from court to court in search of liberal patrons. These poems are to be dated after the period they describe, in the same way as the so-called love poems originate in a milieu far removed, physically as well as mentally, from the little villages depicted in them. This removes the basis on which the current early dating of Cankam poetry relies. This dating is mainly based on the assumption that the poems describe a contemporary society. This part of my argument is tucked away by Hart in the rather vague sentence that I spend "some time discussing the esthetic implications of akam (interior or love) poems, and claim they constitute a condescending and often sarcastic urban and sophisticated take on village live" (p. 180).

After this, the agreement between the Tamil poems and Sanskrit Kavya appears in quite a different light. As the points of agreement are not restricted to the style of the poems but involve whole genres - village poetry (Sattasai) in Akam, minor operatic scenes (lasya) in the Kalittokai and festival songs (carcari) in the Paripatal - the possibility of independent origination becomes unlikely. Given the above conclusion that Cankam poetry is most probably to be dated much later than has been done so far, in the case of borrowing the role of Tamil as the source cannot be taken for granted.

In trying to place the origin of Cankam poetry among the Pandyas of the eighth or ninth century, important information has been drawn from the inscriptions of that dynasty. All that Hart has to say on this point is to suggest some corrections on my translations of some phrases from the inscriptions. He says nothing about the main argument, namely that Tamil in Cankam poetry has exactly the same function as it has in the Pandya inscriptions, and only in the inscriptions of that dynasty.

Twice Hart refers to a "great deal of evidence ... [which] suggest [s] strongly that the poems were composed between the first and the third centuries A.D." (p. 180 and 183), implying that my ideas would be completely out of touch with the actual state of affairs. On p. 183 he presents some of this "evidence". It includes the fact that Iravatham Mahadevan has shown that writing was used in that period by the common people in Tamil Nadu. Apart from the fact that Mahadevan is obviously carried away in his enthousiasm here (as I will show in a review of Mahadevan's book, to appear in ZDMG, some of the stone masons responsible for the inscriptions could not write or read at all), the main question is of course what all this means. Writing was also known at Asoka's court but as far as we know it did not lead to the production of literary works. Secondly, the names of some "Cankam" kings appear in early inscriptions. A "classic" case is of course the identification of the Irumporai kings mentioned in the Pugalur inscriptions (2nd century) with three generations of kings of that dynasty mentioned in the Patirruppattu. The identification keeps cropping up, most recently in Eva Wilden's review of my book in WZKS XLVI (2002), p. 124. I wonder if anyone has recently cared to have a look at the two sets of names, in which Ko Atan Cel Irumporai of the inscriptions corresponds to Celva-k-katunko Vali-y Atan of the Patirruppattu, Perunkatunkon to Perun-ceral Irumporai and Katunkon Ilankatunko to Ilan-ceral Irumporai. I for one fail to see any resemblance, but even if the names had been the same, what does this prove? Do we date Kalidasa in the Sunga period because in his Malavikagnimitra he has the names of the kings of that dynasty right? In fact, the same question applies to Hart's argument that the poems describe trade with the Romans. For so does the seventh-century writer Dandin in his Dasakumaracarita. I was somewhat surprised to see that Hart includes among the evidence the so-called Gajabahu synchronism, which has been relegated to the world of literary legend already more than twenty years ago by Obeyesekere. I was equally amazed by Hart's argument that the "poems name hundreds of poets and kings and string them together in a narrative that is chronologically coherent." For one thing this has nothing to do with the absolute dating of the poetry. Secondly, the "narrative" referred to is not found in the poems but in the later colophons. Furthermore, Hart uses a strange argument here. According to him it would be extremely unlikely that so many names could have been remembered for eight centuries, a reservation he does not make in the case of the transmission of the thousands of poems!  Finally, Hart refers to some of the linguistic peculiarities in the poems, such as the rare use of Sanskrit loanwords in the poems and the absence of the present tense marker -kinr-. In this connection I would like to refer to my article "The Nature of the Language of Cankam Poetry", which has appeared South Indian Horizons (Felicitation Volume for Francois Gros), Pondicherry 2004, 365-387. Not that I claim to have solved the language problem in that article. What I have tried to do, is to deal with the language as a literary language, which might have been fashioned, or created, in accordance with a specific literary function.

Of course I should not expect Hart graciously to connect two findings which I myself indeed presented in separation. Nonetheless, where he casts doubt on my identification of the poems of the Kalittokai as examples of lasyas, he could have mentioned the fact that I am not the only one to make that identification. In fact, as I have tried to show, the same identification appears to have been made by the persons who were responsible for the compilation of the Kalittokai.

When all is said and done there does not seem to be real evidence for dating Cankam poetry in the beginning of our era. In addition to that, the claim that Cankam poetry is the product of a native Tamil culture undiluted by Sanskrit influences will most likely have to be revised.

In a review of my book in his recent study of the Purananuru P. Marudhanayagam, the director of the Pondicherry Institute of Language and Culture, begins by calling me a traitor (utpakaivar) of the Tamil cause. In this way he makes it abundantly clear where he stands. This cannot be said of Hart, who professes to have proceeded carefully and dispassionately.

Herman Tieken