Transliterating Lao Script Pali

Explore the ancient language of the Tipitaka and Theravāda commentaries
Post Reply
Richard W
Posts: 16
Joined: Thu May 20, 2021 12:17 am

Transliterating Lao Script Pali

Post by Richard W »

I've been contributing Pali items and tables to the English Wiktionary for a few years now, and I'm finally getting round to setting up proper transliteration. My main interest was in recording the spellings actually used in SE Asia; we record words in the script in which they are found. Our transliteration standard is IAST, which is mostly very straightforward. However, there are difficulties with the sort of Lao script Pali mostly found online, for it only uses the letters used for the Lao language. Thus the voiced aspirates are missing, the retroflexes are missing, and ch and ñ are missing.

Is there any standard for transliterating Pali written in this reduced alphabet? I have adopted the convention that where the text is ambiguous, I convert the letter before me. Thus, as dhamma and damma ('in need of training') are written ທັມມະ, they are both transliterated as damma. When the extended alphabet is used, damma is written that way (or ທມ຺ມ), but dhamma is written ຘັມມະ (or ຘມ຺ມ). I had been transliterating texts according to the standard IAST spellings of the words, but I think that is technically wrong and will have to change.
User avatar
Dhammanando
Posts: 6492
Joined: Tue Dec 30, 2008 10:44 pm
Location: Mae Wang Huai Rin, Li District, Lamphun

Re: Transliterating Lao Script Pali

Post by Dhammanando »

Have you tried using Vinodh Rajan's Aksharamukha converter?

http://aksharamukha.appspot.com/converter

It's also available as an Android program.

https://play.google.com/store/apps/deta ... sharamukha

If you type in "dhamma", selecting IAST for the input text and Lao for the output, then the word will be given in its popular form: ທັມມະ

But if you select Lao Pali for the output, then it will appear in its scholarly form: ຘມ຺ມ

Likewise with all the aspirates:

kha gha cha jha ṭha ḍha tha dha pha bha
khā ghā chā jhā ṭhā ḍhā thā dhā phā bhā
khi ghi chi jhi ṭhi ḍhi thi dhi phi bhi
khī ghī chī jhī ṭhī ḍhī thī dhī phī bhī
khu ghu chu jhu ṭhu ḍhu thu dhu phu bhu
khū ghū chū jhū ṭhū ḍhū thū dhū phū bhū
khe ghe che jhe ṭhe ḍhe the dhe phe bhe
kho gho cho jho ṭho ḍho tho dho pho bho

Popular
ຂະ ຄະ ຈະ ຊະ ຖະ ທະ ຖະ ທະ ຜະ ພະ
ຂາ ຄາ ຈາ ຊາ ຖາ ທາ ຖາ ທາ ຜາ ພາ
ຂິ ຄິ ຈິ ຊິ ຖິ ທິ ຖິ ທິ ຜິ ພິ
ຂີ ຄີ ຈີ ຊີ ຖີ ທີ ຖີ ທີ ຜີ ພີ
ຂຸ ຄຸ ຈຸ ຊຸ ຖຸ ທຸ ຖຸ ທຸ ຜຸ ພຸ
ຂູ ຄູ ຈູ ຊູ ຖູ ທູ ຖູ ທູ ຜູ ພູ
ເຂ ເຄ ເຈ ເຊ ເຖ ເທ ເຖ ເທ ເຜ ເພ
ໂຂ ໂຄ ໂຈ ໂຊ ໂຖ ໂທ ໂຖ ໂທ ໂຜ ໂພ

Scholarly
ຂ ຆ ຉ ຌ ຐ ຒ ຖ ຘ ຜ ຠ
ຂາ ຆາ ຉາ ຌາ ຐາ ຒາ ຖາ ຘາ ຜາ ຠາ
ຂິ ຆິ ຉິ ຌິ ຐິ ຒິ ຖິ ຘິ ຜິ ຠິ
ຂີ ຆີ ຉີ ຌີ ຐີ ຒີ ຖີ ຘີ ຜີ ຠີ
ຂຸ ຆຸ ຉຸ ຌຸ ຐຸ ຒຸ ຖຸ ຘຸ ຜຸ ຠຸ
ຂູ ຆູ ຉູ ຌູ ຐູ ຒູ ຖູ ຘູ ຜູ ຠູ
ເຂ ເຆ ເຉ ເຌ ເຐ ເຒ ເຖ ເຘ ເຜ ເຠ
ໂຂ ໂຆ ໂຉ ໂຌ ໂຐ ໂຒ ໂຖ ໂຘ ໂຜ ໂຠ
Yena yena hi maññanti,
tato taṃ hoti aññathā.


In whatever way they conceive it,
It turns out otherwise.
(Sn. 588)
User avatar
Dhammanando
Posts: 6492
Joined: Tue Dec 30, 2008 10:44 pm
Location: Mae Wang Huai Rin, Li District, Lamphun

Re: Transliterating Lao Script Pali

Post by Dhammanando »

I forgot to mention, you also have the option of the older Lao Dam script.

ᨡ ᨥ ᨨ ᨫ ᨮ ᨰ ᨳ ᨵ ᨹ ᨽ
ᨡᩣ ᨥᩣ ᨨᩣ ᨫᩣ ᨮᩣ ᨰᩣ ᨳᩣ ᨵᩤ ᨹᩣ ᨽᩣ
ᨡᩥ ᨥᩥ ᨨᩥ ᨫᩥ ᨮᩥ ᨰᩥ ᨳᩥ ᨵᩥ ᨹᩥ ᨽᩥ
ᨡᩦ ᨥᩦ ᨨᩦ ᨫᩦ ᨮᩦ ᨰᩦ ᨳᩦ ᨵᩦ ᨹᩦ ᨽᩦ
ᨡᩩ ᨥᩩ ᨨᩩ ᨫᩩ ᨮᩩ ᨰᩩ ᨳᩩ ᨵᩩ ᨹᩩ ᨽᩩ
ᨡᩪ ᨥᩪ ᨨᩪ ᨫᩪ ᨮᩪ ᨰᩪ ᨳᩪ ᨵᩪ ᨹᩪ ᨽᩪ
ᨡᩮ ᨥᩮ ᨨᩮ ᨫᩮ ᨮᩮ ᨰᩮ ᨳᩮ ᨵᩮ ᨹᩮ ᨽᩮ
ᨡᩮᩣ ᨥᩮᩣ ᨨᩮᩣ ᨫᩮᩣ ᨮᩮᩣ ᨰᩮᩣ ᨳᩮᩣ ᨵᩮᩤ ᨹᩮᩣ ᨽᩮᩣ
Yena yena hi maññanti,
tato taṃ hoti aññathā.


In whatever way they conceive it,
It turns out otherwise.
(Sn. 588)
Richard W
Posts: 16
Joined: Thu May 20, 2021 12:17 am

Re: Transliterating Lao Script Pali

Post by Richard W »

Thanks for the suggestion. I forgot to check what Vinodh had been doing. His program converts "ທຸຕິຍັມປິ ທັມມັງ ສະລະນັງ ຄັດສາມິ" to "dutiñampi dammaṅ salanaṅ gad̈sāmi". I currently have that as (if I do it word by word) as "dutiyampi dammaṃ salanaṃ gassāmi", as currently displayed at https://en.wiktionary.org/wiki/ທັມມະ and https://en.wiktionary.org/wiki/ຄັດສະຕິ. I haven't yet updated the manual transliterations at the quotations - I'm pondering displaying both mechanical ('strict') and normalising transliterations. I've done that for Northern Thai Pali materials where a lot of words have unorthodox spellings.

Unfortunately, Vinodh doesn't cite any authorities. I think the reason that a lot of Lao text doesn't use ຢ for 'y' is that as onset consonants, the Tham equivalents of 'y' and 'ñ' are pronounced the same.

Tai Tham script isn't an option - Wiktionary transliterations have to be to the Roman script. I see that Phaya Maha Sena (Pouy)'s preference for round AA in Tai Tham hasn't been followed. And even if Tai Tham were allowed, the same question of standards would arise - so much written distinction has been lost in the popular Lao version of Pali.
Last edited by Richard W on Thu May 20, 2021 7:20 pm, edited 1 time in total.
Richard W
Posts: 16
Joined: Thu May 20, 2021 12:17 am

Re: Transliterating Lao Script Pali

Post by Richard W »

I've just noticed the dot above in "gad̈sāmi", which looked like a fleck of dirt when viewed in the font used for the forum on my machine. That's not IAST, so Vinodh's schme isn't even a candidate! :(
Richard W
Posts: 16
Joined: Thu May 20, 2021 12:17 am

Re: Transliterating Lao Script Pali

Post by Richard W »

cittaanurakkho wrote: Fri May 21, 2021 3:46 am How about this: https://www.unicode.org/L2/L2017/17106r ... r-pali.pdf
About twenty years too late. I'm relieved it wasn't opposed by the Lao government. Unless I've overlooked something in it, the proposal doesn't help with the question of how to transliterate Pali written using the repertoire of the standard Lao language into IAST. The proposal aims to help replace that system (or rather, those systems) by restoring the full repertoire. I did a Google search for the phrase "ນະໂມ ຕັສສະ ຠະຄະວະໂຕ" ("Homage to the Exalted One..."). I got three hits - two on Wiktionary and one in a PDF apparently explaining the new system. The new, or rather, restored system hasn't taken off yet. Windows 10 seems not to support it yet, at least by default, not even on Chrome.

My question isn't how to write Pali in the Lao script, or even how to read Pali written in the Pali script. The question is whether there is a standard way to transliterate Pali from the Lao script written using the standard Lao repertoire. The way is allowed to depend on the writing system used for the Pali. Full answers are all too likely to start with, "Well, this is how I do it...". As a general strategy, "Well, I wouldn't start from here..." is sound advice. However, I am addressing the issue of starting from a text written in the reduced repertoire of standard Lao, or more precisely, the Lao script of Unicode prior to Version 12.0. (I have seen cancellation marks used to confirm that there is no vowel between so sung (ສ) and the following consonant.)
cittaanurakkho
Posts: 66
Joined: Fri Dec 18, 2009 4:12 pm

Re: Transliterating Lao Script Pali

Post by cittaanurakkho »

Richard W wrote: Fri May 21, 2021 8:35 am
cittaanurakkho wrote: Fri May 21, 2021 3:46 am How about this: https://www.unicode.org/L2/L2017/17106r ... r-pali.pdf
About twenty years too late. I'm relieved it wasn't opposed by the Lao government. Unless I've overlooked something in it, the proposal doesn't help with the question of how to transliterate Pali written using the repertoire of the standard Lao language into IAST. The proposal aims to help replace that system (or rather, those systems) by restoring the full repertoire. I did a Google search for the phrase "ນະໂມ ຕັສສະ ຠະຄະວະໂຕ" ("Homage to the Exalted One..."). I got three hits - two on Wiktionary and one in a PDF apparently explaining the new system. The new, or rather, restored system hasn't taken off yet. Windows 10 seems not to support it yet, at least by default, not even on Chrome.

My question isn't how to write Pali in the Lao script, or even how to read Pali written in the Pali script. The question is whether there is a standard way to transliterate Pali from the Lao script written using the standard Lao repertoire. The way is allowed to depend on the writing system used for the Pali. Full answers are all too likely to start with, "Well, this is how I do it...". As a general strategy, "Well, I wouldn't start from here..." is sound advice. However, I am addressing the issue of starting from a text written in the reduced repertoire of standard Lao, or more precisely, the Lao script of Unicode prior to Version 12.0. (I have seen cancellation marks used to confirm that there is no vowel between so sung (ສ) and the following consonant.)
As I understand it, sometime you have two words in Pali (e.g. dhamma and damma) which were transliterated into one word of standard-Lao ທັມມະ. And you want to transliterate ທັມມະ into roman-Pali. There are standards for Lao into Roman: such as Royal Thai System of Transcription (RTSG), or Library of Congress, or IPA, or other. The transliteration of ທັມມະ would be something like: thamma or tamma. Yet in either case thamma or tamma is not standard roman-Pali for dhamma or damma. That may be sufficient if you just want to show how dhamma & damma was spelt in standard-Lao script.

There aren't any transliteration standard from standard-Lao to Pali because standard-Lao do not have sufficient number of consonants to express Pali fully (which need 32-33 consonants). It's just like trying to use standard English alphabet to write Pali, which is not possible unless one use trick like Velthuis notation. Lao could have develop something like standard-Lao-Velthuis notation, but they don't.
Richard W
Posts: 16
Joined: Thu May 20, 2021 12:17 am

Re: Transliterating Lao Script Pali

Post by Richard W »

cittaanurakkho wrote: Fri May 21, 2021 11:50 am There aren't any transliteration standard from standard-Lao to Pali because standard-Lao do not have sufficient number of consonants to express Pali fully (which need 32-33 consonants). It's just like trying to use standard English alphabet to write Pali, which is not possible unless one use trick like Velthuis notation. Lao could have develop something like standard-Lao-Velthuis notation, but they don't.
There was a system with dots below (rather like the way adopted in the Roman alphabet), but I don't know how well U+0323 COMBINING DOT BELOW works with Lao in Windows-infested areas. Working with Unicode 12.0 and later, I've encoded them as PALI VIRAMA rather than with U+0323.

Transliterating as though the Pali were Lao might work, so long as tone indications were kept. I'll keep this option in mind. Dropping tone would merge 3 sets of stops per position, which would not be good.

I'm not surprised there's no standard, but I couldn't be sure. Thanks for everyone's input.
User avatar
vinodh
Posts: 45
Joined: Tue Jan 13, 2009 8:00 am
Location: Hamburg, Germany
Contact:

Re: Transliterating Lao Script Pali

Post by vinodh »

Richard W wrote: Thu May 20, 2021 7:10 pm Thanks for the suggestion. I forgot to check what Vinodh had been doing. His program converts "ທຸຕິຍັມປິ ທັມມັງ ສະລະນັງ ຄັດສາມິ" to "dutiñampi dammaṅ salanaṅ gad̈sāmi". I currently have that as (if I do it word by word) as "dutiyampi dammaṃ salanaṃ gassāmi", as currently displayed at https://en.wiktionary.org/wiki/ທັມມະ and https://en.wiktionary.org/wiki/ຄັດສະຕິ. I haven't yet updated the manual transliterations at the quotations - I'm pondering displaying both mechanical ('strict') and normalising transliterations. I've done that for Northern Thai Pali materials where a lot of words have unorthodox spellings.

Unfortunately, Vinodh doesn't cite any authorities. I think the reason that a lot of Lao text doesn't use ຢ for 'y' is that as onset consonants, the Tham equivalents of 'y' and 'ñ' are pronounced the same.

Tai Tham script isn't an option - Wiktionary transliterations have to be to the Roman script. I see that Phaya Maha Sena (Pouy)'s preference for round AA in Tai Tham hasn't been followed. And even if Tai Tham were allowed, the same question of standards would arise - so much written distinction has been lost in the popular Lao version of Pali.
Probably, I'm a bit too late.

I did some research on the same topic a whole ago. And found nothing official. Everyone seems to have their own standard. I try to normalize the different set of rules based on the Google hits I got for a particular transliteration (not very scientific. But it was some way to measure how wide a particular word is transliterated)

I took some notes to document how it widely written. You can see it here.

https://aksharamukha.appspot.com/describe/Lao

Also, the two dots above d. IAST sadly doesn't have any mechanism to transcribe non-Brahmic consonants. So, I had to innovate a way to indicate those. If you could suggest anything more appropriate, that would be great.

V
http://www.virtualvinodh.com

Buddhists Texts in Brahmi Script : http://www.virtualvinodh.com/brahmi-lipitva

yo dharmaṁ paśyati, sa buddhaṁ paśyati
One who sees the Dharma, sees the Buddha

na pudgalo na ca skandhā buddho jñānamanāsravam
sadāśāntiṁ vibhāvitvā gacchāmi śaraṇaṁ hyaham

Neither a person nor the aggregates, the Buddha, is knowledge free from [evil] outflows
Clearly perceiving [him] to be eternally serene, I go for refuge [in him]
Richard W
Posts: 16
Joined: Thu May 20, 2021 12:17 am

Re: Transliterating Lao Script Pali

Post by Richard W »

I'm afraid I don't trust Eisel Mazard's account of pronunciation - there's too much that disagrees with other sources, and he seems generally confused. I wouldn't endorse it, as you could be taken to.

In your chart, you associate the consonant 'ch' with a greyed out ຈ CO. I think rather it should be associated with a greyed out ສ SO SUNG. What has happened with Lao is that after the Thai consonant shift, /tɕʰ/ has merged with /s/, as happened in Northern Thai.

ຮ HO TAM for standard 'r' isn't something I had anticipated - this is a case of the Lao lexicon interfering with Pali, something Mazard complains about for Thai. I think I'll have to transliterate it the same as ຣ RO. That will be no problem if the writing system of the text being cited normally writes 'r' using ລ LO. I got very confused by the superfluous RO in ສັບພຣະພຸດທານຸພາເວນະ (sabbrabuddānubāvena, "by the power of all the buddhas" until I realised that some (all?) Laotians write ພຣະພຸດທ... ("Buddha") rather than prescriptively correct ພະພຸດທ... ("Buddha") when writing Lao, and this habit had carried over into the Pali text!

I simplified my task by only aiming to transliterate Pali. I've also noticed that IAST is not reversible with Unicode - tall v. round AA is one issue (not for Lao), and a failure to note 'kinzi' (superscript NGA as part of a CCV stack) are immediate limitations. (IAST worked fine with Unicode 5.0 in these respects!) I therefore convert clusters of coda stop consonants plus onsets into Pali clusters. I also convert seeming -ny- to -ññ-, and at the next edit I will restore -nl- to -ll-. I resolve the merger of 'ṅ' and niggahita by transliterating ງ NGO to ṅ before velars and ṃ elsewhere. My code for all this is in wiktionary Module:pi-translit.

I dug up what seems to be the original (1894) specification of IAST. It seems that we should be using ḻ instead of ḷ. I don't think that's going to happen now.
Post Reply