Frontline Volume 20 - Issue 13, June 21 - July 04, 2003
India's National Magazine
from the publishers of THE HINDU

Home Contents

Printer Friendly Page Send this Article to a Friend


A magnum opus on Tamil-Brahmi inscriptions


Early Tamil Epigraphy. From the Earliest Times to the Sixth Century A.D. by Iravatham Mahadevan; Crea-A:, Chennai (email:, and the Department of Sanskrit and Indian Studies, Harvard University, Cambridge, USA, 2003; pages 719 + xxxix; Rs.1,500.


Iravatham Mahadevan copying a Brahmi inscription at Tiruvadavur.

IT is rarely that one comes across a study that marks, in the usual manner of description, "a milestone" in the history of a discipline like epigraphy. In the last century, the 1960s saw a new awakening in the field of south Indian epigraphy and palaeography - owing to the efforts of one man, Iravatham Mahadevan, an administrator-turned scholar. He created history by reviving interest in the earliest surviving and "enigmatic" cave inscriptions of Tamil Nadu in the Brahmi script, which had defied all earlier attempts at successful decipherment and reading. His first publication, Corpus of Tamil-Brahmi Inscriptions (1966/68), triggered a series of institutional and individual explorations. The Tamil Nadu State Department of Archaeology, the Department of the Chief Epigraphist, Government of India, and individual scholars vied with one another to make new discoveries of cave and rock inscriptions in Brahmi.

More than the romance of discovery, these explorations proved to the scholarly world how rigorous the discipline of epigraphy had become and how important an interdisciplinary method was for such studies to be meaningful. That epigraphy could no longer be treated as an appendage to archaeological studies, but was a major discipline in itself was firmly established. South India's rich epigraphic sources form nearly 70 per cent of the total number of inscriptions in India, and the "Tamil-Brahmi" inscriptions represent their beginnings in Tamil Nadu in a language (Tamil) other than Prakrit.

The recently published book on Early Tamil Epigraphy (From the earliest Times to the Sixth Century A.D.), the result of more than forty years of dedication and penance, is truly Mahadevan's magnum opus. His earlier study of the Indus script is no less significant. It is the most scientific and sober analysis of an undeciphered script in a language that remains unknown. Further, the Indus script has been the focus of an unresolved controversy, to which not only genuine scholarly interest but also politically motivated hijacking has contributed. However, it is Tamil-Brahmi that has been Mahadevan's lifelong, magnificent obsession.

Coin with the Brahmi legend "Kuttuvan Kotai", a Chera king. 3rd century A.D.

The names of two pioneers of epigraphic studies are indelibly imprinted in our minds: James Princep (1850s), who deciphered the Asokan and post-Asokan Brahmi used for the Prakrit language, and A.C. Burnell (1874), who attempted the earliest work on South Indian palaeography. The contributions of Indian epigraphists like D.C. Sircar, H. Krishna Sastri, T. N. Subrahmanian and K.G. Krishnan have made epigraphy the most important among the sources relevant for the study of the pre-modern periods of Indian history. The deciphering of the Grantha, Vatteluttu, Nagari and Tamil scripts of the south Indian inscriptions dating from the 7th century A.D. and their evolutionary stages, based on their resemblance to the modern forms of the scripts, seemed relatively easier and more successful than that of the early Brahmi inscriptions.

The early Brahmi inscriptions posed a greater challenge on account of their archaic characters and orthographic conventions, which were different from the original Brahmi used for Prakrit. The challenge seemed insuperable even to the most competent among the pioneering epigraphists. The major breakthrough in the decipherment of the cave inscriptions of Tamil Nadu came with K.V. Subrahmanya Aiyer (1924). He was the first to recognise that these are inscribed in Brahmi, but with certain peculiarities and new forms of letters, due to its adaptation for the Tamil language which has sounds (phonetic values) not known to the Prakrit (Indo-Aryan) language and northern Brahmi script. Yet, this lead was not seriously followed and was soon forgotten. Even Subrahmanya Aiyer did not pursue his line of enquiry to its logical conclusion.

Other scholars like V. Venkayya and H. Krishna Sastri were constrained by the assumption that all Brahmi inscriptions were invariably in Prakrit or Pali, as Brahmi was used predominantly for Prakrit in all other regions of India from the Mauryan (Asokan) period. Their readings failed to convey any meaning. By reviving Subrahmanya Aiyer's early decipherment and reading and at the same time more systematically studying these inscriptions in all their aspects, including palaeography, orthography and grammar, and seeking corroboration from the early Sangam classics and the Tolkappiyam, the basic work on Tamil grammar, Mahadevan has virtually re-deciphered these inscriptions and shown them to be inscribed in Tamil. Hence the name "Tamil-Brahmi," one variety of the Brahmi script.

Square seal (silver) from Karur, with symbols like the Srivatsa and legend "Kuravan". Ist century B.C.

The Tamil-Brahmi inscriptions are mostly short, donative inscriptions. They are found in inaccessible rock-caverns with stone beds for ascetics, mainly of the Jaina faith and occasionally Buddhist. The inscriptions number 89 in all, so far discovered and read, apart from the 21 Early Vatteluttu inscriptions studied by Mahadevan in order to show the transformation of the Tamil-Brahmi into the Vatteluttu and also the inscriptional usage of Prakrit and Sanskrit words and the emergence of the Tamil script. The distribution of these inscriptions reveals a clear pattern: they occur on trade routes connecting the west (Kerala) coast with the east (Tamil) coast and the upper parts of south India with Tamil Nadu. The distribution also coincides with the distribution of coin finds (indigenous punch-marked and dynastic and foreign, that is, Roman) and pottery with Brahmi inscriptions in urban/craft centres, while potsherds with inscriptions occur even in rural areas.

Mahadevan persuasively relates the significance of this pattern (Maps I, I-A and II) with the intensive trade activities of the period (the 2nd century B.C. to the 3rd century A.D.). He points out, for the first time, that the relatively large number of potsherds with Brahmi inscriptions even in rural areas, signet rings, seals and other objects inscribed with Brahmi characters, indicate a transition from orality to literacy in this part of the country, where Tamil was both the spoken and "official" language. Prakrit was never given the hegemonic status that it had attained in all other parts of India, where Prakrit/Pali was the language of the elite and administration.

This certainly is a significant finding as the Tamil literary works (the Sangam classics) represent the earliest and only large corpus known in a Dravidian language, a language that was spoken in the Tamil region, which then included the territory that is now Kerala. What is of even greater importance is the fact that the Brahmi script was brought to the Tamil region by the Jainas and Buddhists in the post-Asokan period. It may be added that the Jainas and Buddhists also fostered the Tamil language and authored some of the most remarkable literary works, above all the two epics - Silappatikaram and Manimekalai. Even Tolkappiyam and many of the 18 didactic works, including the Tirukkural, are often assigned to Jaina authorship.

Early Tamil Epigraphy, which is organised in three parts and thematic sections (chapters) with charts and tables, inter-linked by cross references, is highly readable, delightfully so, because it addresses the lay and specialist reader with equal ease. For it takes up serious issues such as palaeography (the evolution of script), orthography (the system of spelling), grammar and linguistic analysis of the inscriptions (in Part Two) with the competence of a specialist in each field, without deviating from the simplicity of expression that only a master of the subject can adopt.

In Part One, the author takes us on a fascinating journey through the hazardous fieldwork of pioneers, the copying, deciphering and reading of inscriptions. The inscriptions are found in inaccessible hills (rocky outcrops) and out-of-the-way sites, to which the author made two major field trips, equally difficult, but immensely interesting and rewarding. Every inscription was rechecked, re-deciphered and read both with the help of estampages supplied by the Government Departments of Epigraphy and fresh copying and fresh photographs, following a new method of tracing each letter on the rough and often undressed rock surface. In the process of making his fieldwork productive, Mahadevan collected around it a number of younger and enthusiastic epigraphists, who are now actively engaged in pursuing research in this field. The author generously acknowledges their contribution in his book.

Parts Two and Three, the key sections of the book, make this work unique - for the following reasons.

First, Early Tamil Epigraphy is the most comprehensive source for the study of the Tamil-Brahmi and Early Vatteluttu inscriptions, including inscriptions on pottery, seals, rings and other objects. Second, the occurrence of the largest number of inscriptions on pottery in the Tamil region not only in well-known urban sites but also in rural areas indicates that Tamil society was in the process of transition from orality to literacy. Third, this is the first work to take up the study of the orthography in addition to the palaeography of the inscriptions. This has made it possible to recognise that these inscriptions are inscribed in the Tamil language (Old Tamil). These are the earliest known lithic records in Dravidian, as rare lexical items and grammatical morphemes not found even in the earliest layer of Old Tamil occur in these records. On the other hand, no Brahmi inscriptions in Telugu or Kannada have been found so far, since Prakrit is the language of the early "Southern-Brahmi" inscriptions in Andhra Pradesh and Karnataka.

Fourth, present day Kerala with its Tamil-Brahmi and Early Vatteluttu records was part of a larger Dravidian-speaking south in the early historical period. It became a separate region and culture zone from the early medieval period (A.D. 600-1300). This fact is corroborated by the Sangam classics as well as by later Malayalam literature and inscriptions. Fifth, the Tamil language with its alphabet of 26 main letters attained fixity by the 6th century A.D. and resisted any new characters for the non-Tamil words introduced into the language. The origin of the Vatteluttu (cursive script of the 5th-6th centuries A.D.) can now be traced to the Tamil-Brahmi. Sixth, although the Southern-Brahmi and the Tamil-Brahmi are derived from the Asokan Brahmi, they evolved independently of each other, despite the close cultural and commercial contacts between upper and lower south India in the early period. There is a significant influence of Jain Ardhamagadhi - and not of Asokan Prakrit - in the language of Tamil-Brahmi inscriptions. Seventh, there is clear evidence of mutual influence between the Tamil-Brahmi and the Simhala-Brahmi, although the latter is used for Simhala-Prakrit, a Middle-Indo-Aryan language, and the former for Tamil, a Dravidian language. Simhala-Brahmi and Tamil-Brahmi show certain orthographic similarities and peculiarities. It is interesting that recent Sri Lankan archaeological and epigraphical studies have also recognised this interaction and influence. Simhala-Brahmi, we are told, is "unique among the Prakrit based variants of Brahmi, for a substratum of Tamil influence seems to have been present and due to the processes of assimilation and epenthesis, which were more thorough going in this language than in Indian Prakrits, the two scripts, one for a Middle-Indo-Aryan (Simhala-Brahmi) and the other for a Dravidian language (Tamil-Brahmi), were able to avoid ligatures, a prominent feature in all other regional scripts."

Ring (silver) from Karur with legend "Velli Campan".

Eighth, Brahmi cannot be derived from the graffiti (symbols), as the latter occurs in the inscriptions side by side with the Brahmi characters in rock inscriptions and pottery (from Kodumanal). Also important is Mahadevan's observation that the resemblance of the cave symbols with the Indus script may show that they are likely to share similar significance, but not necessarily the same phonetic value. Ninth, of great importance is the recognition that the Tolkappiyam, admittedly the earliest work on Tamil grammar, cannot be dated earlier than the 2nd century A.D., as its rules regarding the phonetic needs of Tamil and the signs (medial vowel notations etc.) used for specific sounds not known to the Indo-Aryan appear in the later stages that is, in Late-Tamil-Brahmi. Tenth, the revised chronology presented by the author provides a century-wise dating of the inscriptions and broadly classifies them into two: Early-Brahmi - 2nd century B.C. to 1st century A.D., and Late-`Brahmi - 2nd century A.D. to 4th century A.D., followed by the Early Vatteluttu - 5th to 6th centuries A.D. Eleventh and most important, Early Tamil Epigraphy disproves the claim by Tamil enthusiasts that there existed an earlier independent script for Tamil, which was forgotten, and that Brahmi came into use later.

To show how the author has arrived at these conclusions, one has necessarily to dwell upon the technical aspects of the study in some detail. The Brahmi script was adapted and modified to suit the Tamil phonetic system. Palaeographic changes were made to suit the Tamil language, with the omission of letters for sounds not present in the Tamil language and by additions to represent sounds in Tamil that are not available in Brahmi. All but four of the 26 letters are derived from Brahmi and have the same phonemic values. Even these four - i.e., l,l, r, n - are adapted from the letters with the nearest phonetic values in (Asokan-) Brahmi, i.e., d, l, r, n. Letters were also modified with a special diacritic mark, viz., the pulli (dot). These are reflected in the development of the Tamil-Brahmi in three stages (TB I, II and III): Stage I when the inherent a (short-medial vowel) was absent in the consonants and the strokes (vowel notations) were used for both the short and long medial a, and hence the need for the reading of consonants with reference to context and position; Stage II when the stroke for medial a marked only the long a; and Stage III when the use of diacritics like the pulli was introduced for basic consonants and for avoiding ligatures for consonant clusters (as in Simhala-Brahmi). The pulli was used also for distinguishing the short e and o from the long vowels, for the shortened - i and -u (kurriyalikaram and kurriyalukaram) and for the unique sound in Tamil called aytam, all of which are unknown to the Indo-Aryan ( Prakrit and Sanskrit).

It is the recognition of the absence of the inherent vowel a (short) in the early phases, e.g. ma, ka, na with strokes or medial vowel notations, which are actually to be read as ma, ka, n (the inverted J symbol for the nominal suffix `an' characteristic of Tamil), and the addition of the pulli as a diacritic, that provided the key to the whole re-decipherment. Herein lies the basic contribution of Mahadevan to the study of the script and alphabet. That these findings are corroborated by the phonetic rules of the Tolkappiyam is significant.

Carefully drawn up charts and a graphemic inventory of the Tamil-Brahmi script illustrate these palaeographic and orthographic changes from the Early Tamil-Brahmi to the Late Tamil-Brahmi and the evolution of the script and its transformation into the cursive Vatteluttu. The Tamil script is basically syllabic and examples of this are provided from Tamil-Brahmi such as segmentation in consonant followed by vowel, vowel followed by vowel, and so on. Complex issues such as the linguistic, grammatic and phonetic differences and the way they were resolved in early Tamil epigraphy are handled with expertise acquired in various disciplines such as linguistics, grammar and lexicography of both Indo-Aryan and Dravidian families of languages.

Potsherd with Brahmi letters from Quseir al-Qadim on the Red Sea coast. Reads "Catan".

WHILE Mahadevan's major finding is that the language of the inscriptions is Old Tamil, his analysis brings out other significant features such as the nature and number of Indo-Aryan loan words - mainly Prakrit loan words - derived from standard epigraphic Prakrit. They are all nouns - names, religious and cultural terms. Some are derived from Jain Ardhamagadhi and interestingly also from Simhala-Prakrit. Sanskrit loan words appear only in the Vatteluttu inscriptions, and increase in the early medieval inscriptions, that is, from the 7th century A.D. Hence the absence of voicing of consonants in Tamil acquires a special significance in the light of the author's discussion of the way in which Prakrit loan words were written with voiceless consonants in Tamil-Brahmi, and later the method by which the problem of the voicing of consonants was solved when the Grantha script was evolved and adopted for the voicing of consonants, aspirates, sibilants and other phonetic needs of Sanskrit in the increasing Sanskrit loan words in the early medieval (A.D. 600-1300) inscriptions of the Pallava, Pandya and Chola periods.

Hence the conclusion that the Tamil alphabet and script attained fixity by the 6th century A.D., resisting the introduction of new letters for non-Tamil sounds, and that the classical age of Tamil began under the Cholas. The graphic presentation with charts and tables on the script and language, their evolution and relative position, influence and interaction among the varieties of Brahmi, such as the Northern-Brahmi, Southern-Brahmi, the Bhattiprolu script, Simhala-Brahmi and Tamil-Brahmi, as also the later Vatteluttu and Grantha, make these sections easy to follow and interesting even to the lay reader. The relative position of Tamil, Telugu, Kannada and Malayalam is also graphically presented in Table 5.5. The Bhattiprolu script, "an isolated epigraphic curio," is legitimately characterised as the Rosetta Stone in the decipherment of Tamil-Brahmi.

All this is of considerable value for the historian. The author consciously draws from and follows closely the historical contexts as well as continuity and change in the subcontinent and Sri Lanka from the Mauryan period to early medieval times, the 6th century A.D. marking the point at which the Tamil letters attained fixity.

The grammar of the inscriptions forms an important section covering all aspects such as the phonemic inventory, dependent sounds (Carpeluttu), vowels, consonants and their distribution, consonant vowels (Uyirmei eluttu) and so on. Sections on morphophonemics, morphology and syntax deal respectively (a) with the process of joining morphemes in a word or words in a sentence, (b) with the forms of words, the syllabic structure of stems, parts of speech, and so on, and (c) the various ways in which the inscriptions make up the sentences with or without verbs as found in the inscriptions.

Mahadevan offers a complete reading and interpretation of all the known inscriptions in Early and Late-Tamil-Brahmi and Early Vatteluttu with illustrations in the form of tracings, estampages and some computer-enhanced prints of direct photographs, carefully listed with fine reproductions, thus preserving these early inscriptions for posterity. There is an exhaustive commentary on the inscriptions, with citations from early Tamil literature and lexicographic works (Nighantus), which aims at situating the Early Tamil inscriptions in the mainstream of Indian epigraphy and which will undoubtedly be a major guide to the study of the Tamil-Brahmi and Vatteluttu. An inscriptional glossary, index to names of places and persons, etymology, grammatical morphemes and so on, together with a useful bibliography make the book a tour de force in scholarship.

By way of historical background to his study, Mahadevan provides a survey of the polity, society and religion in Part One. It may be conceded that since Early Tamil Epigraphy is a work on epigraphy, processes of social, economic, political and religious changes are not major concerns for the author. Yet his overview is too cursory and somewhat inadequate, as it is based mainly on the Tamil-Brahmi and Vatteluttu inscriptions. There is little doubt that the Sangam Chera-Pandya rulers appear for the first time in Tamil-Brahmi inscriptions and that the identity of the Satiyaputas of the Asokan edicts is now established beyond doubt as the Atiyamans of Tagadur (Jambai inscription). Nonetheless, the author's understanding of the nature of the major Tamil polities (Chera-Chola-Pandya) as well-organised kingdoms with a centralised administration, government functionaries like the atikan (adhikari - official) and kanaka (accountant) and territorial units like the natu and ur points to his conventional approach.

Rock-cavern inscription in Jambai. Mentions "Satiyaputo Atiyaman", a Sangam chief, who got this "palli" (cave monastery) made.

There is no attempt to look at the new perspective on early societies that suggests that state institutions were less evolved and administration hardly centralised. The natu was a generic term for any settled region, for example, Chola-natu or Pandya - natu, and a peasant micro-region. It became a revenue unit only later, during the period of the Pallavas and Cholas. Similarly a certain all-pervasive political control is implied in the references to the Kalabhras as the invading and subversive force in Tamil society after the 3rd century A.D., for which it is hard to find epigraphic and archaeological evidence. The so-called Kalabhra interregnum (a dark period in conventional history) in fact marked a period of great flux with no clear political configurations. The derivation of the term Kaviti from the Prakrit Gahapati and its interpretation as a title conferred on merchants and officials, as also the interpretation of Kon as a title conferred on Kaviti, need closer scrutiny. Despite the fact that the author has carefully refrained from any discussion on social structure and relations, the inference that the suffix Ilanko refers to a Vaisya is strange and needs to be substantiated, for even in the inscriptions the term Ilanko refers to a prince.

The predominant references to Jaina ascetics in these inscriptions and the close interactions between Karnataka and Tamil Jainas are duly emphasised. While most of these caverns with stone beds in the interior sites were executed for the Jainas by rulers, merchants and craftsmen, the significant presence of Buddhism in the coastal sites cannot be ignored. The Andhra and Tamil coasts were linked through trade and traders of the Buddhist persuasion and also with Sri Lanka, which had close contact with Amaravati and its art traditions.

The decline of the Jainas (and Buddhists) is rightly attributed to a religious conflict and to the revival of the Brahmanical religions, Saivism and Vaisnavism, revitalised by the Bhakti movement. The theory of "revivalism" however, poses serious problems in the understanding of the religious changes, especially the emergence of organised and institutionalised forces in Brahmanical/Puranic religion and the decline of the "heterodox" Sramanic faiths of Buddhism and Jainism. In the course of the conflict, the Jainas were persecuted, which Mahadevan believes was "uncharacteristic of [the] Indian polity."

Yet there is impressive evidence of patronage, persecution and marginalisation occurring in periods of major socio-religious and economic change. These processes have to be situated in the larger context of the decline of trade and the beginnings of a land-grant system in early medieval India, with predominant agrarian institutions like the Brahmadeya and the temple emerging and Puranic religion providing the major world-view and ideology of the ruling families. Thus the temple appears as an institution in its incipient form even in the Pulankuricci Vatteluttu inscription (circa A.D. 500), although it assumes a multi-faceted institutional role only in the early medieval period, that is, the 7th to the 13th centuries A.D.

Approaches to history may differ. Interpretation and analysis of historical processes may vary and justifiably so. However, the discipline of history will greatly be in debt to Mahadevan for his first authentic study of Tamil-Brahmi. Early Tamil Epigraphy will prove to be a major source of enduring value not only for Tamil-Brahmi and Early Vatteluttu inscriptions, but also for Indian epigraphy as a whole.

Dr R. Champakalakshmi is former Professor of History, Jawaharlal Nehru University, New Delhi; she has specialised in socio-economic history and religion and society in South India.

Printer friendly page  
Send this article to Friends by E-Mail

Subscribe | Contact Us | Archives | Contents
(Letters to the Editor should carry the full postal address)
[ Home | The Hindu | Business Line | Sportstar
Copyright © 2003, Frontline.

Republication or redissemination of the contents of this screen are expressly prohibited
without the written consent of Frontline