Re: Transliteration schemes

["Dr.K. Kalyanasundaram" <kalyan@igcsun3.epfl.ch>]

Dear Vasu:
Thanks for posting up the transliteration scheme used
in the Tamil Lexicon of Madras Univ. and also in Rajam
Aiyar's book.  My gif had a mistake.
All the consonants shown in tamil script must be marked 
as ik, il etc insteadof being there as ka, la etc. I have
now corrected this mistake and loaded the revised gif.
The summary questions that I listed and also those that
you have listed need to be answered.

Here are some comments as a follow up of yesterday's 
posting(s):

i) What I have indicated as the LC scheme in my gif is 
what is being used in the Inst. of Asian Studies in their
extensive, recent publications including the comprehensive
Encyclopaedia of Tamil Literature.
When I compare the Tamil Lexicon scheme with the LC scheme,
I find that the two schemes are identical but for one single
character ak. In Tamil lexicon is h with a under-dot while
it is k with an under-bar in LC scheme. Am I correct- please
clarify.

ii) Last time when I was in Madras (oops Chennai), I visited
the famous Inst. of Asian Studies and talked to several people
(incl. distinguished tamil scholars such as Prof. John Samuel, 
Prof. Shanmugam Pillai ) on our attempts to build a standard 
for tamil computing. The impression I got was that, there will be
a very strong reluctance to accept any revisions to existing 
widely used transliteration schemes using diacritical markers. 
The same was the message I got with many of the
tamil scholars I met at Roja Muthaiah Tamil Library in Chennai.
It is just like the resistance one faces to replace
existing tamil typewriter layouts with anything else, irrespective
of the merits of the proposed one. 
We are to be aware that transliteration scheme such as that
used in Library of Congress has been around for a long time and
has already built a strong base:
it is used not only in tamil but for all dravidian languages
and hence by a large number of indologists (mostly scholars
of western origin). This takes on new dimension when we consider
growing attempts to go for multi-lingual computing under one 
window (Unicode).
It may be very difficult to push through major reforms in 
transliteration schemes with diacritical markers. The LC scheme
is used not only in the venerable Library of Congress but also
by major american univ. libraries (Berkeley, chicago, UPenn etc).
I am not sure if TNC would be able to push through any reforms 
that will require these major world libraries with huge tamil
book collections to rewrite their catalogues and also carry through
indologists. The point is we may have to be very very 
conservative in proposing changes. I am sure Prof. Schiffman
is aware of these hard realities.

iii) regarding the words "tamil", "Chennai" etc, I want to 
clarify that these words are english words for equivalent 
tamil terms "thamizh" and "cennai". Often there is confusion
that the words used as such are bad transliterations. 

iv) I am sure all of us will agree that, along with written
script, the phonetic form of any language evolves continuously. 
Even at a given time, the pronounciations vary from region to 
region. Given this, wouldn't be too much trying to integrate 
subtle variations in pronounciations into transliteration schemes?
The simpler and logical any transliteration scheme is, more easier
would be its acceptance. 

Kalyan

PS: Since transliteration schemes are used more widely by western
scholars, it would be very useful if we can hear the opinion of
(bring in?) distinguished tamil scholars of this group. I am happy
that Prof. Schiffman is participating in these. I have in
mind people like Prof. George Hart, Prof. Norman Cutler,...
Can you (or people like Kumar Kumarappan) reach them by phone 
or email and request them to send their viewpoints? Kumar wrote
earlier in Tamil.Net that these issues were discussed early this
year at UC Berkeley. Unfortunately we do not have any written 
documentations of the proceedings of these conferences.