Colombo Telegraph

A Historian In Focus: The Dark Side Of S Pathmanathan

By Darshanie Ratnawalli

Darshanie Ratnawalli

Cognitive problems and knowledge deficiencies of S. Pathmanathan, Professor Emeritus of History? Yes. First, a caveat. Although there is a school of thought that Sri Lanka shows a lack of discernment in the making of her professors emeritus (“X was made a professor emeritus after just one publication” remarked a senior academic grimly), they clearly do not mean S. Pathmanathan. In his chosen area (the middle or the medieval period of SL history), Professor Pathmanathan has enough publications (most of them downloadable here) and his peers mention him respectfully enough. “I stress that Pathmanathan, in his Kingdom Of Jaffna, does not indulge in such outrageous statements. In fact, note the paraphrase of his carefully circumscribed statements in fn.59 above”- (Michael Roberts: 2004).

There is an Other Side though. I first learnt of it from K.S. Sivakumaran in History of Lankan Thamilians revisited. It contains a translation of statements from a Tamil newspaper article by Pathmanathan on Brahmi lithic inscriptions of Sri Lanka. Although the translator’s language does not inspire confidence, I will assume that it’s a faithful translation because the statements are bald, simple, without nuance and the least likely to have suffered in translation unless the translator made them up from scratch (unlikely).

“In Lanka the Brahmi inscriptions are written in Prakrit language…Paranavitana tried to convince that these inscriptions were written in Sinhala language…In Lankan Brahmi inscriptions Thamil Brahmi letters are found in many places. Arya Abeysingha and Saddamangala Karunaratne have explained this feature showing examples. But Paranavithana hides these findings. He has completely ignored the Thamil Brahmi letters. Three letters were differently written in Thamil Brahmi and Ashoka Brahmi. These two kinds were in existence in Lanka until the demise of Brahmi script.
The formation of letters of two different languages –Thamil and Prakrit- were found in the inscriptions from the beginning and its end. This shows that the inscriptions were written in both languages.”

The ignorance in this is so mindboggling that at first I did not know how to tackle it. Then I knew. Contrast. Place an educated statement (R. Champakalakshmi in “A magnum opus on Tamil-Brahmi inscriptions”) next to it. 

“…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.”

The formation of letters of two different languages were certainly not found in the stone Brahmi records (For these are the subject of Pathmanathan’s statements) of Sri Lanka at anytime. They were not bilingual records. There are in Old Sinhalese[i] (300 BC-300/400 AD) and later in Proto Sinhalese. The term “Old Sinhalese” is used because the language found in them is the oldest example of the Sinhalese language from which the evolution of the present language can be traced. It’s a name that can be used irrespective of language family. If the language known now as “Sinhala” is renamed “Swahili” tomorrow, the language in the early Brahmi lithic records would be old Swahili. Old Sinhalese is written in these records in a script, which Mahadevan calls Sinhala- Brahmi (2003:117) and Paranavitana called Ceylon Brahmi. Saddhamangala Karunaratne (1960/1984:39) expressed it best; 

“It’s generally accepted that the language of the Brahmi inscriptions of Ceylon is an Aryan form of speech and is the precursor of Modern Sinhalese…To distinguish it from Sinhalese proper, we propose to name it Sinhalese Prakrit, using the term, Prakrit in its widest sense. If objections are raised against the term Prakrit, then it may also be argued that the script is not Brahmi but Old Sinhalese. In this dissertation, Prakrit and Brahmi are used to distinguish the broad division into two strata in point of time of the same language and script, which has continued for 22 centuries from the earliest recorded times. The end of the 7th c. A.D. is the point at which the division is made.”    

There were points of similarity between Sinhala Brahmi and Tamil Brahmi or to paraphrase Pathmanathan, between the letter formations of old Sinhala and old Tamil. There were also differences that made them distinctive and made it possible to single out certain features as signature Tamil Brahmi and certain other features as signature Sinhala Brahmi. The alveolar nasal ṉ was such a signature feature that instantly identified Tamil Brahmi, differentiating it from both Asokan Brahmi and Sinhala Brahmi. The following paragraph (Osmund Bopearachchi: 2001) would illustrate;   

“This coin is of utmost importance in that it presents us a personal name in a clear Tamil nominative form with an aksara ṉa, representing an alveolar nasal, which is not found in Ceylonese Brahmirock inscriptions, but which is well-known from South Indian inscriptions in Tamil Brahmi, and now also from two of our coins... I. Mahadevan, the foremost authority on Tamil Brahmi, in a recent article, accepting our initial reading added: “The authors have correctly identified the Tamil alveolar nasal n here and point out that “in Tamil texts, this character terminates proper names”.”[ii]

No less was Sinhala Brahmi distinctive. You could pick out Sinhala Brahmi and Tamil Brahmi separately from a lineup.

“…P. Jeyakumar has asserted that “undeniably cultural elements from Sri Lanka must have spread to Tamilakam” and presented evidence from potsherd graffiti showing influences of Sri Lankan Brahmi and Sinhalese-Prakrit…In his monumental work on Tamil epigraphy, Iravatham Mahadevan has drawn attention to several instances of Sri Lankan influence in the Brahmi inscriptions of Tamil Nadu. Mahadevan is of the view that, among the inscribed potsherds found in Tamil Nadu, ‘A small but significant group of pottery inscriptions is in the Sinhala-Prakrit language written in the Early Sinhala-Brahmi script (ca. 2nd century BC to 1st century AD). These inscriptions were discovered at Arikamedu, Alangulam, Kodumanal and Poompuhar (Kaveripattinam)… Mahadevan has also published a couple of interesting articles on this subject (‘An Old Sinhalese Inscription from Arikamedu’,…‘Old Sinhalese Inscriptions from Indian Ports- New Evidence for India-Sri Lanka Contacts’…) A remarkable graffito in Prakrit from the ancient port of Kaveripattinam has been read as a Sinhalese-Brahmi inscription by S. Iracavelu.”- (Indrapala, 2005:337)

What shape did the signature of Sinhala Brahmi take? 

“Linguistic features like the genitive suffixes sa and ha; shortening of long vowels; de-aspiration of aspirates and unique change of ja to jha found in the seven inscribed potsherds collected at Arikamedu, Alangkulam and Kodumanal in Tamil Nadu are known to Sri Lankan Prakrit only (Mahadevan 1995:55-65).”- (Rajan: 2009). The extensive preference for the palatial “ś” (sha) over the dental “s” should also be added to this list.

Which brings us to the similarities between the two scripts and the interesting accusation brought by Pathmanathan that Saddhamangala Karunaratne revealed something and then Paranavtana hid it (or Paranavitana hid something and then Karunaratne revealed it? it’s not clear from the translation). But we will assume it’s the latter because what’s the point of hiding what someone has already revealed? Even the latter accusation is a technical impossibility because Karunaratne’s (revealing) publication, which Pathmanathan obviously has in mind (Karunaratne: 1984-EZ vol VII), came before Paranavitana’s (concealing) publication (IC 1:1970). Karunaratne: 1984 was simply his PhD thesis of 1960 published as it was, without any updates whatsoever. Anyway all that is esoteric Watson. Here is what Karurunaratne (1984/60:16) “revealed” about the ‘ma yanna’.

“The commonest form of “ma” found in the earliest inscriptions of Ceylon, is …(drawing of the tubular ma)….Since then several inscriptions have been found in South India with the identical “ma” and the graffiti from Arikamadu, also in South India, have added to the list.”

This is how Paranavitana (IC: 1970: xix-xx) “hid” it;

“Two types of the letter ma occur in the inscriptions of Section I. The type of ma which is common in Asokan inscriptions is formed of a circle on which is placed a semicircle opening upwards. Examples of this type of ma  from Inscriptions numbered 835, 236, 1023, 269 and 623 are given, respectively, in Columns…But the type of ma which is normal in our inscriptions is a u-shaped tubular form in the middle of which is a cross-bar extending from one vertical to the other. Morphologically, this tubular ma (also referred to as cup-shaped), is easily seen to be the earlier of the two forms of ma….This type of ma, which is common in Ceylon Brahmi, occurs also in the few records in that script found in the Tamil-nad. But it is not unknown to North Indian Brahmi, being found in Inscriptions no. 187 and 361 at Sanchi. The intermediate form, differing but little from the tubular type occurs in No. 201 from the same site.”

The same “hide-reveal” pattern can be found in the respective descriptions of Karunaratne (84:16) and Paranavitana (70: xviii) for the letter “i yanna[iii]. That is, signs of hiding would only be visible to those (to paraphrase P. G Wodehouse) dropped on their heads as babies. Such early accidents would also prevent folk from comprehending that when features are shared between two entities they stop being signature to either.

*The writer can be found at

[i] This is one of the most abnormal aspects of Sri Lankan Academia. A significant number of academics lack the most rudimentary knowledge about linguistics and consequently go public with entrancing displays of ignorance. In 1981, A. Veluppillai remarked; “The present writer has to confess that he could not understand why Paranavitana rushed to the conclusion that the language of the inscription was Old Sinhalese. Prakrit was the language of inscriptions throughout South Asia except in Tamilnadu, for the first few centuries before and after Christ. The language of Sri Lanka inscriptions up to the fourth century A.D. had been termed Sinhalese Prakrit. The Sinhalese Prakrit, even in the earliest records did not reflect the Prakrit of any particular North Indian region. Linguistic features of Prakrit of Western and Eastern India and Deccan had been noticed in these records. Paranavitana’s comment on the language of the Gold Plate should be viewed from this point of view.” A sincere confession from A. Veluppillai should have read; “The present writer has to confess that he could not understand why Paranavitana called the language of this inscription Old Sinhalese because the present writer is not sufficiently well read and perceptive. This prevents him from understanding that a) Paranavitana is not the only Indic linguist in the world and other linguists call this language Old Sinhala too b) Old Sinhala and Sinhalese Prakrit are synonyms and complementary terminology signifying two aspects of the same entity (‘Old Sinhala’ indicating that it’s the linguistic precursor of modern Sinhalese and Sinhalese Prakrit signifying the familial affinities of the language).  My readers who do not want things too complex can get a basic handle on Sri Lankan inscriptional linguistics by reading the third chapter(which I have uploaded here) of Richard Salomon’s “Indian Epigraphy, A Guide to the Study of Inscriptions in Sanskrit, Prakrit, and the Other Indo-Aryan Languages”. Read (Sinhalese Prakrit) under (Inscriptions in other MIA dialects) and 3.4.7 (Sinhalese) under 3.4 (The New Indo-Aryan Languages). If you want to go deeper, read James W. Gair in “Studies in South Asian Linguistics. Sinhala and Other South Asian Languages”. The PDF I have uploaded has three good articles; “Sinhala, An Indo Aryan Isolate”, “How Dravidianized was Sinhala Phonology? Some Conclusions and Cautions” and “Some Aspects of the Jaffna Tamil Verbal System”   

[ii] There are examples from Tamil Nadu where the Old Sinhalese ownership-suffix “sa” is affixed to Tamil names ending in an alveolar nasal ṉa. These excerpts from “Visaki and Kuviran- Historical Implications of Names in the Tamil-Brahmi Inscriptions” by Y. Subbarayalu will illustrate;  “In no. 170 of the above table, whose latter half only is available, the genitive case marker sa is added to the name “..taṉ”, which with “aṉ” ending must definitely be a Tamil name like Ataṉ or a Tamilized Prakrit name like Visakaṉ…Another similar instance is found in a cave inscription (Mahadevan 2003:no. 24, p.351), which gives the word Utayaṉa-sa, “of Utayaṉ”… the occurrence of the genitive suffix “śa” in many as five names and of the genitive suffix “ha” in two of the pottery inscriptions is another significant piece of correspondence between the two areas. The two genitive suffixes are peculiar to Sri Lankan Prakrit (Paranavitana 1970, p.xl). The use of the palatal sibilant “ś” in the place of the dental sibilant “s” normally found in other Prakrits is a special feature of Sri Lankan Prakrit. Though the Prakrit influence from Sri Lanka is clearly perceptible, Sri Lanka is not the only source. In fact, the impact of North Indian Prakrits is found more influential than that of Sri Lankan one…”

[iii] Karunaratne (1984/60:16);

“The “i” form ..(graphic of the Brahmi letter, two dots on either side of a vertical line) found in the earliest Ceylonese inscriptions is quite different from the standard Mauryan form which was …(graphic 3 dots).The identical form has been found in the early Brahmi inscriptions of South India and in the graffiti of Arikamedu, also in South India.”

Paranavitana(IC-1970 :xviii); “The symbol for the initial i, which is not uncommon in our documents, is a vertical stroke with two dots on either side of it, and not three dots arranged to form a triangle as in Asokan Brahmi. This type of i also occurs in the few Brahmi inscriptions discovered in caves in the Pandya country, and has been read as i as well as ī. The symbol formed by a vertical stroke with two dots on either side of it occurs in a Mathurā inscription of close of the first century A.C., with the value of ī .The modern Sinhalese, as well as the Tamil letter for ī, is obviously evolved from this symbol, and it is possible that it was originally meant to represent the ī sound, but was read as i in Old Sinhalese which had no long vowels.”

Back to Home page