Official Transliteration?

[otherstuff1957]

I read somewhere that someone claimed to use the Thai Governments "official transliteration" in their place names. Does anyone know where I can find a copy of this tranlisteration list? And if so, is it any good.

I have the consonants down pretty well, but the Thais don't even seem to remember that their 32 vowels even exist !

TIA

[Richard W]

I read somewhere that someone claimed to use the Thai Governments "official transliteration" in their place names. Does anyone know where I can find a copy of this tranlisteration list? And if so, is it any good.

I have the consonants down pretty well, but the Thais don't even seem to remember that their 32 vowels even exist !

TIA

The web page starts at หลักเกณฑ์การถอดอักษรไทยเป็นอักษรโรมันแบบถ่ายเสียง. The guts of the systems are given at Transliteration Table, which is problably all you really need to know. The key points are:

1. <ก> = <k>, and all that that entails.

2. No <c> or <chh>, but just <ch>.

3. The system's phonetic.

4. Tones aren't marked

4. Vowel length isn't marked - nor are final glottal stops, so goodbye final -h.

5. Falling diphthongs end in -i or -o.

6. <uea> is one of the rising diphhongs (at long last! => Don Mueang)

The lack of length marking should guarantee that other systems continue to be used. Have just one symbol (<o>) for all 4 of โอะ, –(โอะ ลดรูป), โอ, เอาะ and ออ is not good either. Note that <ao> means เอา or อาว, so the transcription 'Aoi' remains outside the scheme.

This transliteration is nowhere near a transliteration in the sense of allowing one to go back and forth between Thai and the Roman alphabet. Far too much information is lost.

[meadish_sweetball]

This transliteration is nowhere near a transliteration in the sense of allowing one to go back and forth between Thai and the Roman alphabet. Far too much information is lost.

Agreed. Thai could do with a romanisation system such as that for Vietnamese, or Pinyin for Mandarin Chinese. AUAs or Mary Haas's systems would be good starting points, as they are very close to being fully phonetic.

Pipe dreams...

[konangrit]

This transliteration is nowhere near a transliteration in the sense of allowing one to go back and forth between Thai and the Roman alphabet. Far too much information is lost.

Agreed. Thai could do with a romanisation system such as that for Vietnamese, or Pinyin for Mandarin Chinese. AUAs or Mary Haas's systems would be good starting points, as they are very close to being fully phonetic.

Pipe dreams...

Good point, but as you said it isn't going to happen. Not so important anyway, may aswell spend the time and effort it takes learning a transliteration system to learn the Thai alphabet instead.

[Richard W]

Thai could do with a romanisation system such as that for Vietnamese, or Pinyin for Mandarin Chinese. AUAs or Mary Haas's systems would be good starting points, as they are very close to being fully phonetic.

Good point, but as you said it isn't going to happen. Not so important anyway, may aswell spend the time and effort it takes learning a transliteration system to learn the Thai alphabet instead.

One needs a transliteration scheme for writing Thai names in English text, especially if some of the readership does not know the Thai alphabet.

I'm currently looking for a good way to transcribe Thai words in a linguistics forum. The context is Indian civilisation, so a phonetic transcription is no use. We're restricted to the ASCII (i.e. 7-bit) character set. I've seen one published accurate system, but it looks hideous. I've never encountered it in use. I fear I'm going to have to invent my own.

[chanchao]

7 bit ASCII: Look at the soc.culture.thai newsgroup system. It's hideous of course, but very accurate! Newgroups are 7-bit ASCII, so it was developed out of necessity: Can't write Thai in newsgroups and can't even write accents & umlauts and stuff..

PM me if you can't find it on google.

It takes almost as much time to learn to use reliably as learning Thai would.

[chanchao]

Got your PM, found the link. I post it here as well in case someone's interestied. Keep in mind that this system is the way it is because this was a forum where Thai language could NOT be written and neither could any advanced characters beyong basic a-z and A-Z. Internet used to be like that in the 80's and early 90's.

http://www.faqs.org/faqs/thai/language/

Cheers,

Chanchao

[otherstuff1957]

aswell spend the time and effort it takes learning a transliteration system to learn the Thai alphabet instead.

Good point. I am learning the Thai alphabet and I find that it's much easier to understand Thai when you can read words as well as hear them! I'm going to help some students learn how to spell their names and nicknames in English and didn't want to get too creative!

[chanchao]

Tell all the ones named 'Pun' that they will go to he11 if they keep writing their names as 'Ple'.

(Pun comes from 'Apple' you see, pronounced 'Appun' in Thai because Thai words just don't end in L, ever. But when they then write their names as Ple then NONE of the odd billion or so English speakers in the whole wide world is going to pronounce it right, thus defeating the purpose of transliterating their name in the first place.)

[Richard W]

Tell all the ones named 'Pun' that they will go to he11 if they keep writing their names as 'Ple'.

What happens to the rest of the vowel? I thought 'apple' was normally transliterated แอ็ปเปิล (another instance of Thai spelling not being able to show vowel length!).

Incidentally, there is an ISO standard for transliterating (as opposed to transcribing) Thai, ISO 11940:1998. From what I can glean from the net, it appears that แอ็ปเปิล should be transliterated as xæ˘ppeil (or perhaps x˘æppeil), though for those without rich enough fonts, that should be , by the principles of Transliteration of Indic Scripts - How to Use ISO 15919, something like x^:aeppeil. Has anyone seen this transliteration scheme in use? The most striking features will be macrons on consonants and dots under 'p', 'k', 'y', 'a', 'i', 'o' and 'u'.

[chanchao]

Yes but nicknames are rarely longer than one syllable.. Hence 'Pun'/เปิล as a shortening of Apple. Shorten 'Apple' the same way and you're left with 'ple'.

[weary]

Well, what do I know?

[Richard W]

Actually "pun" doesn't entirely work for that Thai nickname either, does it? Pun is pronounced [p'an] in Received Pronunciation English, as in pun = 'play on words'.

But the Thai nickname's vowel quality is arguably more like a French "eu" or the vowel sound in Black American English "love" or "fun".

[p'an] could just as easily be interpreted as <pan>. <pun> still has a lower mid vowel in my pronunciation. The best English match is a conservative pronunciation of <pun>, but not so conservative as in <put>!

I suppose one might get away with a spelling 'Ple, but otherwise they should stick to Poen. (I took Ple as being เปล, whence my question about the missing vowel.)

[bina]

never knew what "pun" actually meant:

we have a red headed freckled israeli running the apple orchards named paul and all the thai workers call him "pun"...i thought it was just because they couldnt say his name (the L at the end), but now i know they actually punned his name (he's red, and works with apples, so 'pun' is a great nickname.... actually, i named a big male dark red billy goat 'pun' because he was red and had big balls like 'pun' from the orchards; now the joke''s on me

[Richard W]

I've been thinking about ISO 11940:1998. My local library (in England) wondered what I was talking about, because it isn't in their indexes. I'm not sure I want to spend 2,400฿ on it. I doublechecked and discovered it wasn't approved until 2003.

The ISO Working Group in this area appears to be looking for transcription schemes that transliterate back to the source language (fair enough) but also that can be computerised. With that in mind, I tried a very simple automatic implementation (which I can help by sprinkling the Thai with phinthus). The alphabet comes out in this scheme as:

kị̀ k̄hị̀ ḳ̄hwd khwāy k'hn raḳhạng ngū

cān c̄hìng chā̂ng sò c̣hex h̄ỵing

chḍā pṭạk ṭ̄hān mṇṯho p̄hū̂t'hèā ṇer

dĕk tèā t̄hung thh̄ār ṭhng h̄nū

bımị̂ plā p̄hụ̂ng f̄ā phān fạn s̄åp̣heā mā̂

yạks̄̒' reụ̄x ling h̄wæn

ṣ̄ālā vs̄'ī s̄eụ̄x h̄īb cuḷā xā̀ng nkḥūk

(The original is legible in the Tahoma and Titus Cyberbit Basic fonts. The principal loss you may see above is the loss of the macrons over the high consonants. The text above can be displayed by cutting and pasting into an XP Notepad window and setting the font to Tahoma or TITUS Cyerbit Basic.

Applying ISO 15919 principles, in ASCII this comes out as:

k,.i ~kh,.i ~.khwd khwaay k'hn ra.kh.ang nguu

caan ~ch,ing ch^aang s,o .chex ~h.ying

ch.daa p.t.ak ~.thaan m.n_tho ~ph^uut'h,eaa .ner

d^:ek t,eaa ~thung th~haar .thng ~hnuu

b_im^.i plaa ~ph^.ung ~faa phaan f.an ~s~a.pheaa m^aa

y.ak~s'\ re.uux ling ~hwaen

~.saalaa v~s'ii ~se.uux ~hiib cu.laa x,aang nk.huuk

Transliteration of ศาลา corrected 19/6/04.

(For a macron over vowels, I've doubled the vowel, but for a macron over a consonant I've used a tilde. Doubling the consonants is ambiguous.)

Implicit vowels are supposed to be represented by <.o> ('o' with a dot underneath), and I'm not sure whether ด้วย should come out as <d^.owy> or <d^wy> (accent on the 'w' - rather as in Welsh).

Has anyone seen this transcription, and have I understood it? It seems it's intended users are librarians.

[chanchao]

> It seems it's intended users are librarians.

Hm.. It kinds of makes me think they might as well just use Thai. All Thai characters are in the unicode Tahoma set as well. Any system that can display those characters can also display Thai.

Cheers,

Chanchao

[Richard W]

Displayability is not the issue. (Actually, I've cheated a bit, but not in this example, Lakkhangyao is a barred i, but Tahoma doesn't support this character. I'm therefore using a plain i, which is unambiguous.) I suspect the chief copying tool might actually be pen and paper! Another unfriendly feature is the use of undotted i (as in Turkish) for mai muan. In theory, I believe you need to be able to distinguish i grave and undotted i grave! (Or is ใหม่ meant to be mangled to <hım.ò> ?)

The idea is that the letters be recognisable without needing to learn a vast array of alphabets. I can recognise a Russian name in transliteration (if I know the transliteration) far more easily than I recognise it in Cyrillic. I would severely struggle with Devanagari and have little real hope with Chinese.

[sabaijai]

Tell all the ones named 'Pun' that they will go to he11 if they keep writing their names as 'Ple'.   

(Pun comes from 'Apple' you see, pronounced 'Appun' in Thai because Thai words just don't end in L, ever.  But when they then write their names as Ple then NONE of the odd billion or so English speakers in the whole wide world is going to pronounce it right, thus defeating the purpose of transliterating their name in the first place.)

I'd transliterate the nickname as Poen myself, following RTGT. At least some folks know how to pronouce /oe/.

Edited June 28, 2004 by sabaijai