TamilNet Transcription: mails and responses

[TamilNet, Friday, 01 June 2007, 18:47 GMT]

From: Tom Kurudeepan
Date: Wed, 16 May 2007

Vanakkam to all those working with Tamilnet with Transliteration, Transcription and Automation, of Thamil. First of all let me sincerely thank you all for the 'Thamil Thondu' you are all doing. Secondly, it is my humble suggestion that, we are living in a time very much of an anglosized world. It is my strong suggestion that in the 'Grantha' alphabets, if you could include some form of equivalent characters that will represent, F, B, G, also. Eventhough Thamil has most of the sounds being an ancient language, it may be important to grow with time, people and languages in communication.


TamilNet response:
Our present transcription attempt is not to modify the concept of Tamil alphabet. This needs wide consensus.

From: Sinnathurai Srivas
Date: Wed, 16 May 2007

Quoting TamilNet:
"In words such as தங்கம், தங்கச்சி, வங்கம், வங்களாவடி (thangkam, thangkachchi, vangkam, vangka'laavadi) etc., the k is largely retained in Sri Lankan Tamil, especially in the Jaffna dialect. The accent makes the difference.

"It is not uncommon to see people pronouncing gangai as kangkai. Also, look at words such as கங்குமட்டை (kangkumaddai).

"We do agree that k is often pronounced as h, coming under rule number 3 (exceptions) of Muthukumar and as pointed out by Srivas. A word such as நகர் (nakar), can be pronounced as nagar as well as nahar. The former is favoured by those who are familiar with Sanskrit pronunciation and the latter largely prevails in Sri Lankan Tamil. In the example of மகள் (maka'l), it is pronounced as maga'l in the northern districts of Tamil Nadu where Kannada overlaps with Tamil and pronounced as maha'l in Sri Lanka. Which one has to be taken if we aim for standard automation?"

The reason for g changing to k in Jaffna is mainly due to the gradual influence of the misunderstanding among some that Tamil alphabet are phonemic and not letters representing group of phonemes.

Quoting TamilNet
"It is interesting to note that the letters for s and h are frequently found in Tamil Brahmi, ie. the alphabet of the earliest written Tamil. But, when the occassion came to write the word மகள் in the early Brahmi inscriptions, it was written as maka'l and not as maha'l, despite the fact that the script for h was very much known to Tamil Brahmi."

This paragraph need removing ASAP. Again the misunderstanding that Tamil ezuththukaL also represent phonemes and not agroup of phonemes leads to ths above conclusion. Most (or all) Indic languages use phonemic based alphabet while Tamil use alphabet to indicate a group of phonemes. Before Brami, Tamil had defined writing system using ezuththukkaL. When the Brami script was introduced Tamil used only the needed. In Tholharpiyam there is some reference on how to identify vadasoRkaL as the Tamil system would have ambiguity in which phoneme to use in an unknown situation.

As the systems (Phonemic against phonimic based groups) was already in place, no need for Brami to change an existing system.

(This applies to ee and oo vowels as these does not exist in other Indic languages)

Quoting TamilNet
"Tamil Brahmi was not unaware of the letters used for a variety of sounds in Ashokan Brahmi. Yet, for some reason, the Tamils thought of restricting the alphabet to 30 and evolved a parallel Grantha Alphabet to write Sanskrit."

Agin phonemic vs phonimic based Tamil is the reason for the above. Please read http://www.araichchi.net/chiirmai/papers/isis/Tamil-ISIS.htm

Please remove inacurate references to Brami ASAP.

From: Sinnathurai Srivas
Date: Mon, 21 May 2007

Tamil characters were of different shape in the early days. The structure guiding the written form was defined from very early days. However the shapes of characters used for writing was never defined. Tamil was and is probably the only language that defines the structure of written characters as part of Grammar. We need to analyse the differences between character sets representing a group of phonemes and character sets representing phonemes in a one to one relationship. The shape of Tamil characters was not defined in Tholharpiyam while the structure of Tamil character was defined beginning with rule number one "there are thirty ezuthukaL and three relational symbols". It is very evident that when pre-Bhrami shapes became one form of accepted character shapes, to maintain the rule of Tamil writing structure, extra rules were introduced in Tamil. Some of these extra rules were also mentioned in Tholharpiyam.

Even today an alternative script called Grantham, attempting to write Tamil refuses to include the traditional long vowels "ee" and "oo" in its alphabet soup. Yet, these long vowels are important part of Tamil. The philosophy behind Grantham is that Tamil writing should follow the Northern philosophy of one phoneme one-letter and sub philosophies such as long "ee" and long "oo" should not be used. The Fundamental philosophy of closely related group of phonemes represented by same character, which supports scalable use of phonemes in Tamil is also broken by the philosophy followed by Grantham. For example the "Ph" in the word "gnaaphaham" (gnaapakam) cannot be expressed by Grantha philosophy.

Tholharpiyam states that those who wish to adapt pre-Bhrami characters can indicate long "ee" and long "oo" by explicitly marking "e" and "o" Ω necessary. When one uses the philosophy of group of phonemes there is less ambiguity with native words. However, when foreign words are transliterated using group of phoneme philosophy the ambiguity becomes large. Tholharpiyam stipulates that the Tamil writing structure should be maintained in this state of ambiguity. When we face the problem of not all phonemes in use by Tamil are not represented by any writing system and especially phonemes also change within same words, depending on combining situations in sentences, the guidance in Tholharpiyam, in the advice about "vada soRkaL" becomes a valid advice. In present times this is also true for "ulaha soRkaL". Those wish to identify the correct phoneme in Tamil need to use the Tamil diacritics. Tamil diacritics are only now beginning to gain acceptance and will become reality. However transliterating the vast array of phonemes used by Tamil and the characteristics of context sensitive use of phonemes in Tamil into other languages and writing systems will be a major technical issue for a long time to come. Hence, a carefully structured transliteration scheme should be used for attempting a near accurate transliteration of Tamil into other writing systems. (Eg, athanai, aththanai, ∑஀ன√, ∑஀௟஀ன√, ∑ந௟஀ரம௟) One must remember that there are vast arrays of phonemes in use in Tamil and not any other writing system, including IPA can handle this situation at present. Added to this is the context sensitive pronunciation of Tamil words that will remain one of the important issues when attempting to transliterate Tamil to other writing systems.

TamilNet response:
  • 'The reason for g changing to k in Jaffna is mainly due to the gradual influence of the misunderstanding among some that Tamil alphabet are phonemic and not letters representing group of phonemes.' Does this comment mean that the Jaffna Tamil dialect is 'cultivated' and not native?

  • We will be thankful for letting us know whether there is any reference in Tholkaappiyam where a ka becomes ga and where it becomes ha.

  • The existence of pre-Brahmi phonetic writing in Tamil or in any other Indian language is not a foregone conclusion. So far, there is no attestable evidence.

  • There is no phonetic writing system in this world where the alphabets are not group of phonemes. Even from individual to individual, phonemes may differ. The TamilNet attempt is a simple phonetic transcription for media purposes. It is not an ambitious phonemic transcription.

  • We thank Srivas for agreeing with us on 'ambiguity'. It is exactly for the same reason, we wish not to transcribe certain phonemes and stick to the alphabet, because of the elusiveness of standardisation. We wish Srivas well in his attempts of phonemic transcription which will be an asset in specific applications such as computer assisted speech recognition.

From: Uthayan Suntharam
Date: Sat, 19 May 2007

Dear Sirs,

I came across your site lately and found out that you are in the process of developing an alphabet for foreign language transcription especially Thamilj to English.

I have been researching this very issue for some years. I have in fact adapted the roman script for modern use (including all manner of electronic communication, texting, etc.) of Thamilj, without any additional resources being necessary other than a normal English keyboard. I attach some documents which explain my approach.

You will note that I use 'Thamilj' and not 'Tamil'. This is in line with the 'Ariya Thamilj Roman Alphabet' (ATRA) that I have devised. The alphabet is based on all phonemes in current usage in Thamilj. My work also is very much indebted to Rev. Father David's (of St Patrick's College fame) work of comparative linguistics where he compares the most important Dravidian usage (including Thelungu and Kan'adam, as well as some of the more obscure languages of the 28 or so in the Dravidian group) as well as usage in Sinhala which Father David has clearly demonstrated to be half Dravidian and half Indo-European. This ATRA alphabet is also influenced by other languages which adopted ancient Thamilj scripts, such as Thai, which together with Sinhala has recorded some of the Thamilj phonemes within its script.

The ATRA as a written and printed alphabet, like the current Thamilj script, does not show, for example, the difference between 'e' in 'en' (my) and the 'e' in 'enn' (number, count). However, the distinction is clearly made in a second tier - of dictionary definitions. As you read the attached sheets, you will realise that the approach is to simplify (not complicate) adaptation to the Roman script for all modern purposes (including Education - teaching Thamilj to foreigners), and at the same time to promote more accepted pronunciation.

I have used ATRA for texting and for teaching, to which the alphabet adapts very well.

I hope your project (thank you for starting it) will bring out a workable, simple alphabet that can be used by all. You may publish my alphabet system to your other readers for their feedback, if you think fit. You may contact me at my email address.

  1. Ariya Thamilj-Roman Alphabet
  2. Transcription examples1

Yours sincerely,

Ariya Suntharam

TamilNet response:
We would be thankful to know whether the system mentioned by Uthayan Suntharam can be automated.

From: Shanmugam Arulkumaran
Date: Sat, 26 May 2007

Dear Tamilnet,

I am a Jaffna Tamil living and working overseas as a web developer.

First I want to warmly congratulate you on your forward looking idea for Thamizh transcription. We should not find ourselves left behind in this age of Internet Revolution. I sincerely hope that your idea evolves and grows and gains mainstream adoption.

I noticed that you have begun using the transcription for city names and so on. You have also added a black-coloured box on the top-right which allows readers to convert back and forth.

I was wondering whether you have considered using AJAX for this purpose? Currently the user must be redirected to a new page to see the result. By using AJAX, the user does not have to leave the page, he/she does not even need to click "convert>>" to see the results. As users are typing, the text can be converted and displayed instantly.

I respect and admire the work you have done so far, not just the transcription, and I would gladly help you in any web development work needed.

Warm Regards
Shanmugam Arulkumaran

TamilNet response:
Thanks. We invite proposals from those who wish to develop software.



Latest 15 Reports
Find this article at: