posted March 30, 2001 05:45 PM            

[The Annakodai Seal is much older than the Vallipuram Gold Plate.]

=============================================

Anaikkottai Seal by Dr. K. Indrapala, Department of History, University of Jaffna in Early Settlements in Jaffna an Archeological Survey (Appendix II), Ponnambalam Ragupathy Ph.D. thesis, University of Jaffna, 1987 pp. 199-202

In the first two weeks of December 1980 an archaeological survey team from the University of Jaffna in the northern most
part of Sri Lanka brought to light an important megalithic burial complex at a place called Anaikkoddai, the first of its
kind to be discovered in the Jaffna district. Among the unearthed articles in one of the burials, was a pre-Christian metal
seal with two lines of writing.

While the newly-discovered megalithic burial complex is in itself of great significance to the archaeology of that district and has already created much excitement there, the metal seal, the significance of which is as yet unknown to many, appears to be an extraordinary find with implications for the study of the so-called non-Brahmi or graffiti marks found primarily on pottery in the megalithic sites of South India and Sri
Lanka as well as in sites further north. It may even prove to be of interest to the students of the intriguing Indus script.

The inscription on the seal is deeply indited and well preserved. The second line of the inscription is clearly in Brahmi of
about the third or second century B.C. It consists of three letters and an anusvara ("pu/li" = dot). The first line consists
of three characters or symbols, written in the same way as the ideograms on an Indus seal. What is interesting is that these
are not unfamiliar characters, for they occur both among the numerous graffiti marks on megalithic pottery as well as among the Indus ideograms.

This is the first known instance of these symbols occurring on a seal in the form of an epigraph alongside a Brahmi inscription and hence the special significance of the seal. This poses a series of interesting questions.

One is no doubt tempted to ask whether we ave at last stumbled upon a bi-lingual inscription in the Indus and Brahmi scripts. But we must leave this question aside for the moment. What is of immediate relevance is the question whether this will provide a clue to the proper understanding of the graffiti marks on the megalithic pottery of South India and Sri Lanka.

The so-called graffiti marks have been found on a large number of potsherds, both in the megalithic and premegalithic contexts in South Asia. The earliest material comes from the Indus Valley sites belonging to the Harappan culture and it continues in the post-Harappan chalcolithic cultures in Pakistan, Gujarat, Rajasthan and Maharashtra along the western side of the subcontinent. The later material belonging to the megalithic phase comes from Karnataka, Tamil Nadu and Sri Lanka in the southern part of the South Asian region.

For almost exactly a hundred years, scholars have shown interest in these graffiti marks and attempts were made to collect them and discover their significance. Perhaps the first attempt was made in 1881, and in recent years Mr.B.B.Lal, a former Director-General of Archaeology in India, made a systematic survey of these materials, which resulted in the publication of a 'preliminary note' entitled : 'From the megalithic to the Harappa: Tracing Back the Graffiti, on the Pottery' More graffiti marks have been discovered since and some of them have been published.

Mr.B.B.Lal's investigation revealed that "Out of the total of 61 symbols noted so far, as many as 47 are common to the
megalithic pottery on the one hand and the Harappan and post-Harappan chalcolithic on the other". In terms of statistics, his conclusion was that 89 percent of the megalithic symbols go back to the chalcolithic-Harappan times. And he
concluded "But to stress the point that the symbols do have a phonetic, syllabic or alphabetic value would indeed be presumptuous in the present state of our knowledge".

Writing about ten years later, Prof.T.V.Mahalingam, in his report on
the Tirukkampuliyur excavations, doubted that these graffiti had any association with script and concluded that "we may not
be far wrong if we take the graffiti marks to
represent such totemic symbols by the people of the ancient past".

Among those who investigated the meaning of these graffiti, it was Mr.G.Yazdani who more than 64 years ago, thought that these constituted a script and that these symbols were characters used to express ideas. Although later researchers tended to dismiss his view Mr.Yazdanils seems to be the most acceptable theory.

The Anaikkoddai seal seems to confirm Mr.Yazdani's view. So far scholars appear to have been misled by the notion that these symbols occur only on pottery. But their occurrence in a line on a seal, like letters in any short inscription and similar to those on the Indus Valley seals, together with another line in Brahmi - very much like the bi-lingual legends on a coin, indicates that these symbols were in fact used as characters in a script not only on pottery but also on other materials. That the symbols on the megalithic pottery stood for words or names is also confirmed by the occurrence of names in Brahmi scripts on potsherds of a slightly later period excavated in Arikamedu and elsewhere in Tamil Nadu as well as in Kantarodai, Anuradhapura and other sites in Sri Lanka. After the spread of Brahmi, naturally this phonetic script displaced the earlier symbols.

If as we are inclined to believe, the so-called graffiti marks on the megalithic pottery are ideograms or characters with
meaning, they have to be evidently treated as survival of the Indus writing system. For, as we have seen, the vast majority of these graffiti could be traced back to the Indus ideograms and this is not a mere coincidence. Spatially as well as chronologically a relationship could be established between the two sets of characters. Spatially they extend from the
Indus Valley right down the western part of India to the south and beyond to Sri Lanka. Chronologically they begin in the Third Millennium B.C. in the Harappan chalcolithic culture, continue into the post-Harappan phase, then into the megalithic phase and overlap into the period of the Brahmi script. The Anaikkoddai seal belongs to this final phase, after which the easy phonetic Brahmi script supplanted the more difficult
character writing.

On this premise, the first line of the Anaikkoddai seal inscription consists of Indus-derived characters and each of them
must have a value. Being a legend on a seal, they no doubt stand for a name. And the Brahmi writing in the second line obviously stands for the same name, as the case of the Greek and Brahmi legends on some of the coins of the Indo-Greek rulers, Pantaleon and Agathocles. So, for the first time we have a chance of deciphering one complete legend in the characters of this Indus-derived script with the aid of a Brahmi
transliteration.

The inscription in Brahmi consists of three letters and an anusvara. They are crowded within a small space and the first letter, though at first sight it seems to present some difficulty in reading, shows on closer examinations all the features of the vowel-consonant 'Ko'. The middle stroke of this letter is not horizontal but diagonal and the arm of the right (as it appears in negative on the seal) is not very prominent. The second letter is clearly 've' and the third letter is a clear 'ta'. There is a dot or anusvara above the letter 'ta'. Two readings seem to be possible, depending on the point at which we
read the anusvara. If we read it before 'ta' the inscription would read as 'Koventa', but if we read the anusvara after 'ta', the reading 'Kovetan', is possible. Either way, the word would be Dravidian and both readings would have the same meaning.

'Koventa' consists of two words 'Ko' and "venta". Ko in Tamil and Malayalam means 'King' and is related to words in other lesser known Dravidian languages, such as 'Koc' in Parji and 'Kosu' in Gadba. "Venta"is no doubt a variant of or
related to the Tamil and Malayalam 'ventan', 'Ventu', also meaning
'king'. It is also related to the Parji word 'vedid', meaning 'good'. "Koventa" would then appear to be a tautological compound and it is interesting to note here that such a compound. 'Koventan' as well as its variant form 'Koventu' does actually occur in the earliest literature of the Tamils.

In the reading 'Kovetan', 'Ko' is of course 'King' and 'Vetan' would also mean 'King'. "Ventan" (which could also be read as "Vettan", as the double consonants do not sometimes occur in the early Tamil-Brahmi inscriptions) is a variant of
"Ventan" and its root could clearly be seen in such ancient Tamil "vetalikar" (vetu (or veta) + alikar - king's clowns or dancers). "Vetan" has to be split as vetu (or veta) + "an" the 'an' being the masculine singular ending. If as is obvious the name on the seal is Old Tamil or Proto-Malayalam then the second reading namely "Koventan~" is preferable to the first
which has no masculine singular ending. But the first reading is possible if the name is in some other Dravidian language,including Proto-Tamil While "Koventa" could be meaningfully split into only two parts ('Ko' and 'venta'), "Kovetan" could be split into three parts going by some of the forms of the
masculine nouns in the early Tamil-Brahmi inscriptions of Tamil Nadu. The three parts are "Ko", "veta", "an", comparable to
"asiriyku -an", "Pana-an" and "Katala-an-" in the Mankulam Inscription No.1. This accords well with the occurrence of
three characters in the first line on the seal as the equivalent of the name in Brahmi. We are therefore, inclined to adopt
the reading "Kovetan".