Tamil-Brahmi

Tamil-Brahmi, also known as Tamili or Damili, was a variant of the Brahmi script in southern India. It was used to write inscriptions in Old Tamil. The Tamil-Brahmi script has been paleographically and stratigraphically dated between the third century BCE and the first century CE, and it constitutes the earliest known writing system evidenced in many parts of Tamil Nadu, Kerala, Karnataka, Andhra Pradesh and Sri Lanka. Tamil Brahmi inscriptions have been found on cave entrances, stone beds, potsherds, jar burials, coins, seals, and rings.

Tamil Brahmi resembles but differs in several minor ways from the Brahmi inscriptions found elsewhere on the Indian subcontinent such as the Edicts of Ashoka found in Andhra Pradesh. It adds diacritics to several letters for sounds not found in Prakrit, producing ṉ ṟ ṛ ḷ. Secondly, in many of the inscriptions the inherent vowel has been discarded: A consonant written without diacritics represents the consonant alone, whereas the Ashokan diacritic for long ā is used for both ā and short a in Tamil-Brahmi. This is unique to Tamil-Brahmi and Bhattiprolu among the early Indian scripts. Tamil-Brahmi does not, however, share the odd forms of letters such as gh in Bhattiprolu. This appears to be an adaptation to Dravidian phonotactics, where words commonly end in consonants, as opposed to Prakrit, where this never occurs. According to Mahadevan, in the earliest stages of the script the inherent vowel was either abandoned, as above, or the bare consonant was ambiguous as to whether it implied a short a or not. Later stages of Tamil Brahmi returned to the inherent vowel that was the norm in ancient India.

According to Kamil Zvelebil, Tamil-Brahmi script was the parent script that ultimately evolved into the later Vatteluttu and Tamil scripts.

An early mention of a script for writing the Tamil language is found in the Jaina work Samavayanga Sutta and Pannavana Sutta where a script called Damili is mentioned as the seventeenth of eighteen Lipi (scripts) in use in India. Similarly, the tenth chapter of the Lalitavistara, named Lipisala samdarshana parivarta, lists Dravida-lipi and Dakshinya-lipi as two of sixty four scripts that Siddhartha (later the Gautam Buddha) learnt as a child from his gurus in Vedic schools, a list that is found in both Indian Buddhist texts and its ancient Chinese translations. These relationship of early Tamil scripts to these lipi mentioned in Jaina and Buddhist literature relationship is unclear. The pre-1974 work of Mahadevan had established 76 rock inscriptions in Tamil Brahmi from about 21 sites in Tamil Nadu, which states Kamil Zvelebil "establish obvious correlations" with what has been found in early Tamil bardic poems. Nagaswamy treats Tamil-Brahmi script to be synonymous with the Damili script in his publications.

Artifacts such as inscribed potsherds, coins or others are found in Tamil Nadu archaeological sites have graffiti and inscriptions. The potsherds recovered from Kodumanal, for example, have markings that on the basis of stratigraphical analysis appear to be from the 4th century BCE. According to K. Rajan, the "large number of graffiti marks and subsequent Tamil Brahmi script" unearthed in Tamil Nadu and Kerala suggest that this region had a "linguistic cohesiveness well before 5th–4th century BCE". According to Falk these supposed inscriptions are not Brahmi letters, but misinterpreted non-linguistic Megalithic graffiti symbols, which were used in South India during the pre-literate era.

The origins and chronology of Tamil Brahmi are unclear. Several hypotheses have been proposed, with the views of epigraphist Iravatham Mahadevan being generally accepted. According to Mahadevan, the Brahmi script from North India arrived via the southern inscriptions of Ashoka, and evolved into the Tamil Brahmi. This theory presupposes that the Brahmi script itself was either originated within the imperial courts of Mauryan kingdom or evolved from a more ancient foreign script and it was dispersed to South India and Sri Lanka after the 3rd century BCE. The alternate theory proposed by Nagaswamy is that there was an indigenous common source (proto-Vatteluttu) script from which both northern and southern Brahmi script emerged, which he respectively terms as Brahmi and Damili scripts. Richard Salomon favors the Mahadevan theory.

According to Kamil Zvelebil's chronology proposal of 1973, the earliest Tamil Brahmi inscriptions such as the Netunceliyan rock inscriptions at the Mangulam site were derived from Ashokan Brahmi that was introduced to the Tamil region around 250 BCE. It was adapted for the Tamil language by 220 BCE and led to the standardization of the Tamil language and literary norms of Maturai between 200 and 50 BCE. These developments transformed the oral bardic Tamil literary culture to the written Sangam literature in the centuries that followed. The use of Tamil Brahmi continued through the 6th century CE, states Zvelebil.

Since the 1990s, pre-Ashokan dates have been proposed based on excavations and discoveries of graphite covered ancient remains in Sri Lanka. These include those found in Anuradhapura in Sri Lanka, some of which have been dated to the 4th century BCE. The findings of Coningham et al based on the carbon dating of excavated potsherds led to the proposal that the Sri Lankan Brahmi developed before Ashokan era, at least by the 5th to 4th century BCE, from where it came to Tamil region evolving into the Tamil Brahmi, and thereafter spread across South Asia due to trade networks. Sri Lankan nationalists have used this and other fragments of Black-and-Red Ware and Red Ware with Brahmi characters to state that Brahmi was invented on the island and from there it migrated north into the Indian subcontinent. This theory has been criticized by Harry Falk – a scholar of Brahmi and other ancient Indian scripts. First, states Falk, the Coningham team has admitted later that they did not use the carbon dating correction necessary for the Southern hemisphere and used the calibration curves for north Pakistan. Second, the Sri Lankan teams also erred when they deployed a "mathematical trick" whereby they conflated the contested date of lower strata that lacked inscribed shreds with the upper strata where the shreds with Brahmi script were found. According to Falk, a critical study of the feature differences between Ceylonese (Sri Lankan) Brahmi, Tamil Brahmi and Ashokan Brahmi suggest that "all the differences can only be explained once the Ashokan script is taken as primary and the two others as derivations". It is not scholarship that is behind the claims that Ceylonese Brahmi is more ancient and gave rise to Tamil Brahmi and Ashokan Brahmi, rather it is "regional chauvinism", states Falk.