TRANSLITERATION OF TAMIL / SANSKRIT USING ROMAN FONTS
a.natarajan
Greetings!
I request your comments on my proposal.
I propose the following scheme for transliteration of Sanskrit / Tamil
characters using the key-board for out-put on video-screen.
If a Sanskrit group is eventually formed, we need an easy method of
transliterating dhEvanAgarI characters using Roman fonts (keys on our
key boards) resorting minimally to special characterics such as:( ' ~ . -) etc.
[The internationally accepted scheme for transliteration on paper appears
satisfactory and needs no change.]
A. Natarajan at AT&T
*****************************************************************************
A transliteration scheme for transliterating Sanskrit or Tamil
characters using Roman fonts, for display on video monitors.
****************************************************************
SANSKRIT
Vowels:
a, A, i, I, u, U, R, Rr, L, E, ai, O, ow, M, :
Consonants:
ka, Ka, ga, Ga, N~Ga
ca, Ca, ja, Ja, N~Ja
ta, Ta, da, Da, Na
tha, Tha, dha, Dha, na
pa, Pa, ba, Ba, ma
ya, ra, la, va, sa, sha, Sa, Ha
***********************************************************************
TAMIL
Vowels:
a, A, i, I, u, U, e, E, ai, o, O, ow, H~K
Consonants:
ka, ga; N~G; ca,sa; N~J; ta,da; Na; tha,dha; na; pa,ba; ma;
ya; ra; la; va; zha; La; Ra; n~a
***************************************************************************
NOTE:
Tamil hard consonant "Ra" and Sanskrit vowel "R" will not create any clash
because Sanskrit vowel "R" never gets another vowel after it.
Tamil hard consonant "La" and Sanskrit vowel "L" will not create any clash
because Sanskrit vowel "L" never gets another vowel after it.
Sincerely
A. Natarajan at AT&T BL.
(908)-957-6836 att!mtfmi!sri
Examples:
*********************************************************************
TAMIL:
dhan~antharum kalvi tharum oru nALum thaLarvaRiyA man~antharum
dheyva vadivuntharum nallan~a ellAntharum an~bar en~bavarkkE
kanantharum pUN~GkuzhalAL abirAmi kadaikkaNgaLE.
*********************************************************************
SANSKRIT:
SaraSvathi namaSthuByaM varadhE kAmarUpiNI |
vidhyAraMBaM karishyAmi SidhDhirBavathu mE SadhA ||
**********************************************************************
Apology:
I do not remember if I posted this earlier! If I did, I apologize
for the repeat-post! (vayaSAga Aga N~JApaka sakthi kuRainthu varukiRathu!;
I am getting old and have difficulty remembering things!)
Respectfully
Dhanalakshmy at IBM ( Sorry! It should have been Natarajan at AT&T!)
Srinivasan Pichumani
Thanks for initiating the discussion. As you point out,
>>The internationally accepted scheme for transliteration on paper
>>appears satisfactory and needs no change
It is actually much better than that -- since it has been around for
almost 150 years and has been used extensively in Sanskrit/Indological
research for all this time -- by both non-Indians and Indians.
Actually it is as exact as a Sanskritist will ever want.
Although I don't know the exact history of this transliteration
standard, it seems to me that it has a fairly strong Germanic
influence - notice the choice of the Roman alphabet to represent
the Sanskrit vowels.
Now, in my view, any transliteration scheme adapted to the
computer keyboard/monitor should very closely follow this
scheme and vary only where absolutely necessary -- remember that
a lot of people schooled in this country and who have learnt
Sanskrit know this transliteration scheme very well.
Also, variations like the South Indian penchant of adding 'h' to
consonants like 't' to make the 'th' sound as in thyagarajan etc
only has regional validity. 'h' is used both in North India and in
the standard transliteration scheme to denote aspiration -- for good
reasons -- when it follows a consonant.
(Raghavan, in your post on the net, you had wondered why h is used
-- the above-mentioned point is probably one of the reasons)
Therefore, other than the vowels
>> a, A, i, I, u, U, R, Rr, L, E, ai, O, ow, M, :
I have a problem with all of your consonants; and would advise you
to just stick to the standard transliteration (which is striking
for its simplicity/exactitude)
k kh g gh ng
c ch j jh jn
T Th D Dh N
t th d dh n
p ph b bh m
y r l v
sh Sh s h
Except for the capitalization of all the retroflex/cerebral sounds,
there is no change in this adaptation from the standard. Also,
instead of using the " ' " mark over "s" for the palatal sibilant,
I have used "sh" (as in shAnti -- peace!) & instead of using the dot
underneath s for the retroflex, I have used "Sh".
Raghava, your suggestion of just capital E, O and aa,ii makes
sense. However, once again, I am loath to do this because of
the change from the standard scheme.
Regards,
-Srini.
ps: Please look at a standard workbook for Sanskrit classes at your
University like the one Prof.Madhav Deshpande -- an excellent
Vedic/Sanskrit scholar, linguist, and grammarian -- uses for his
Sanskrit class at the U of M. The introductory section
on the alphabet and classifications such as palatal/cerebral etc
in such books are very precise.
Or else check out any book dealing with Sanskrit poetry, drama,
criticism etc and you will find this transliteration scheme.
Corvin Russell
Basically I completely agree with Srini. We could invent a new
transliteration scheme every day, but since the ideal convention in
this case is the one with which the most people are familiar, we
definitely are better off sticking to as close a version of the
standard transliteration as we can. Actually, there is a network
convention in use in Academic distribution lists like INDOLOGY@Liverpool
and Buddha-L. It is even closer to the standard transliteration
than Srini's version, but most netters will find the use of periods
cumbersome. Since we already have a de facto convention here of
representing retroflexes and long vowels by capitals, let's keep it that way.
The additional Tamil consonants can be dealt with in like
fashion. As far as the Tamil lateral flap (I think that's what it
is phonetically), standard transliteration is an l with a bar under it.
Most Tamilians seem to prefer "zh", and we should probably keep it this
way. However, it is NOT a helpful transliteration for speakers of
American English. Tamilians seem to think foreigners will have
a dreadful time with this letter, and ask us to pronounce
"vAzhai pazham" for their amusement. However, the predominant
pronunciation (not the retroflexed "L" pronunciation) is very, very close
to the American R (and is not that hard for us to pronounce!), and
I think a better transliteration would use an R for this purpose. But
typographical limitations preclude this.
A standard alternate for vocalic ocalic "R" is "ri", which, though
innacurate, is almost never ambiguous. Long "Rr"will almost never
appear on the net. Since we need something to represent Hindi's retroflexed
r, the "D" with a dot under it in Nagari, why not use "R" and save "ri" and
"rI" for the vocalic "r"s.
Also, Urdu sounds have so far been neglected. However, they
can most easily be represented as follows, I think:
q KH f z
Corvin
Kathiravan Krishnamurthi
>The additional Tamil consonants can be dealt with in like
>fashion. As far as the Tamil lateral flap (I think that's what it
>is phonetically), standard transliteration is an l with a bar under it.
>Most Tamilians seem to prefer "zh", and we should probably keep it this
>way. However, it is NOT a helpful transliteration for speakers of
>American English. Tamilians seem to think foreigners will have
>a dreadful time with this letter, and ask us to pronounce
>"vAzhai pazham" for their amusement. However, the predominant
>pronunciation (not the retroflexed "L" pronunciation) is very, very close
>to the American R (and is not that hard for us to pronounce!), and
>I think a better transliteration would use an R for this purpose. But
>typographical limitations preclude this.
thamizhar's, a lot of them do not pronunce zha, La, la correctly.
there are lots of them who cannot flex and say vaazhaip pazham.
they would say vaala palam. Another problem is knowing
whether to use la or La. In Canada, I noticed people
using in their marriage invitation, mangkaLam instead
of mangkalam. mangkaLam is the end of songs noramally in
bhajan's. mangkalam is auspicious.
The spelling used by some canadian thamizhs is dreadful
to say the least.
>A standard alternate for vocalic ocalic "R" is "ri", which, though
>innacurate, is almost never ambiguous. Long "Rr"will almost never
>appear on the net. Since we need something to represent Hindi's retroflexed
>r, the "D" with a dot under it in Nagari, why not use "R" and save "ri" and
>"rI" for the vocalic "r"s.
>Corvin
anban
Kathiravan