Talk:Thai alphabet/Archive 1

Name of the Thai Script?
Does anyone know what the Thai name for the Thai script is? I assume this is what's in brackets at the being of the article. What's the name of the script transliterated to English? What does it's name mean (in a literal sense)? Or is it just called 'Thai script' in Thai?


 * See [Thai-to-English | Alphabet]. Also see my changes, unless someone reverted themLee 16:46, 7 March 2007 (UTC)

Thai script is not an alphabet
Thai script is an abugida (a.k.a. alphasyllabary), not an alphabet. How come such a glaring inaccuracy was made in the very name of the article, repeated all over it, but no-one in over two years of edits has spotted it? The most appropriate name for this article IMO would be "Thai script". I have no time to fix all the redirects that would arise from renaming this article myself, so please some sysop or someone else with more spare time do it, because such an inaccuracy does not belong in any serious encyclopedia. Uaxuctum 10:27, 7 Feb 2005 (UTC)


 * Probably because it depends on how narrow your description of "alphabet" is, and how much exotic grammatology terminology you expect casual readers to know. "Abugida" is a recently coined, not well known word. Typical dictionary definitions of "alphabet" are "The letters of a language, arranged in the order fixed by custom", and "A system of characters or symbols representing sounds or things". Both of these fit the Thai writing system. While I would hold that the current title is not wrong I agree that "Thai script" might be better, but perhaps the difference would be similar to the difference between "English alphabet" and "Latin script". &mdash; Hippietrail 12:49, 7 Feb 2005 (UTC)


 * I agree with Uaxuctum. This page should be moved. I propose it to be moved to Thai script. Most other articles about abugidas are named like this. Yenx 22:12, 16 November 2006 (UTC)


 * "Thai alphabet" is the name by which a great majority of people would look for it. Only the most narrow specialist would try to find it as "Thai abugida". Perhaps "Thai script" is a possible point of entry for a few people, so a pass-through from there would be right. As explained above the dictionary definition of alphabet easily covers the Thai writing system. &minus;Woodstone 20:03, 17 November 2006 (UTC)


 * I quite agree that this article should be moved to Thai script. In the Writing Systems WikiProject we are trying to come to consensus about names for articles about writing systems. It is likely that "Thai script" will conform to that model. -- Evertype·�&#65533;&#65533; 17:49, 29 November 2006 (UTC)


 * Are you also going to move "Latin alphabet" and "English alphabet" to "Latin script" (etc). I think the proposed move would only obscure the article. There is no compelling reason that it cannot be described as an alphabet. Only a few specialists would look for it as "script" and they most likely do not need the information anyway. &minus;Woodstone 20:41, 29 November 2006 (UTC)
 * There is a difference between the Latin script and the Latin alphabet. The Thai script is used for other languages besides Thai, too. And it's not an alphabet, it's an abugida. -- Evertype·�&#65533;&#65533; 23:27, 29 November 2006 (UTC)


 * As repeated several times before it's only not an alphabet in the most narrow specialistic interpretation (and even then only by a tiny margin). For common usage it certainly qualifies as an alphabet. &minus;Woodstone 20:49, 1 December 2006 (UTC)


 * Actually, I think it would be wrong to call it an alphabet: An alphabet is, in large, a system of writing in which every sound is represented with a letter in script, and the letters are read one by one after eachother (although digraphs and such occure). The Thai script however, has consonants placed on the line, and separate wovel symbols placed around these; some even placed before the consonant they are to be pronounced after. That there are no separate symbols for the vowel sounds, but that they have to stand at a consonant, also adds. This is simply not an alphabet. It is an abugida, but also a script. Moving it to "Thai script" seems to me the best thing to do, and I don't really see the controvercy, since most other abugidas and other forms of writing have articles named the same way. Yenx 23:44, 6 December 2006 (UTC)


 * I vote with the "move" crowd. Of course, there will have to be a redirect from "Thai alphabet" for people that prefer to use the term "alphabet" loosely. "Script" is a more generic term that encompasses all writing systems, whether alphabetic, syllabc, pictorial or whatnot. If a purist really wants a page for "Thai abugida", I would vote for putting the same redirect on it, leading to "Thai script". Calling writing systems "scripts" is a nice evenhanded way to have a label without getting into arguments over precisely which label is the right one. If we define an alphabet as having a 1:1 relationship between symbol and phoneme, then there would be no such thing as an English alphabet. Just about the only "alphabet" in the world would be Finnish. --Cbdorsett 10:43, 7 January 2007 (UTC)


 * The most prominent examples of other languages do not have an article "xxx script", but redirect to something more likely to be looked up by the reader. For example Chinese script redirects to Chinese character, Arabic script redirects to Arabic alphabet, Hebrew script redirects to Hebrew alphabet, Devanagari script (for Hindi) redirects to Devanagari. So (if only for consistency) I prefer to stick to Thai alphabet and redirect from Thai script and {{Thai abugida]]. &minus;Woodstone 11:12, 7 January 2007 (UTC)
 * I have no opinion on moving the page, but as to this 'alphabet' being an abugida, somewhere the article should state that Thai consonants each spell a syllable in their own write --incorporating its assigned tone plus the sound 'aw' when written alone, as when spelling its own name, and aw, uh, ah or some similar sound when used in a word without a (written) vowel of its own. That being said, I know of only four words of one letter ก็ ธ ณ and บ่ the latter being Lao dialect for Not so! or You don't say! Lee 13:44, 27 January 2007 (UTC)PawyiLee


 * The English alphabet is not an alphabet either. It's an abugida since initialisms and words like x-ray, g-string, and T-shirt have vowel sounds implicitly associated with the letters. The absence a vowel is indicated explicitly with a hyphen and/or capitalization. Isn't that the definition of an abugida?
 * If you're interested in moving some of the information, I would suggest moving the parts that are not specific to the Thai language to Thai script, and many of the parts that are specific to Thai to Thai phonetics or Thai phonology (and merging with other information from Thai language). Is there anything left over? Davilla 22:44, 13 July 2007 (UTC)


 * The Thai script is both an abugida and an alphabet. The Thai script is an abugida ONLY when it is used to write Sanskrit or Pali: each letter has an implied [a], and a virama (pinthu) has to be used to indicate when a consonant does NOT have the implied [a].  When the Thai script is used to write Thai, it is an ALPHABET: each letter has NO implied vowel by default, and an [a] or [o] are implied when a vowel is necessary but is not otherwise specified, and if a consonant is not followed by a vowel, then no special marker need be used because there is no implied vowel.  If the Thai script were an abugida, then ordinary Thai people would be able to read Sanskrit quite easily (even if they could not understand it), but the fact is that they cannot: when Sanskrit is written using the Thai script and is intended for layfolk to be able to read, all the short [a]'s have to be explicitly written in, and no special marker is needed if a consonant is not followed by a vowel (as you would expect for an alphabet):
 * อรหํ สมฺมาสมฺพุทฺโธ ภควา (Sanskrit passage with Thai script as an abugida) all consonants have an implied [a] unless "cancelled" by a pinthu; but
 * อะระหัง สัมมาสัมพุทโธ ภะคะวา (same Sanskrit passage written using Thai script, for the benefit of laypeople) consonants have no implied vowel and [a] must be supplied explicitly; also note that pinthu is not used because the default is that consonants have no implied vowel. --Gak (talk) 02:33, 16 December 2007 (UTC)


 * You say "an [a] or [o] are implied when a vowel is necessary but is not otherwise specified", which sounds like there are implied vowels, just not in every possible position, but where defined by the rules on where a vowel is necessary. I think Hindi is similar, i.e. no vowel written at the end of a word usually means no vowel, and words with an implied final short a in Sanskrit become words with no final vowel in Hindi, e.g. Rama -> Ram, Arjuna -> Arjun, etc. --JWB (talk) 19:12, 16 December 2007 (UTC)
 * The parallel is not exact. Thai frequently drops a short terminal vowel even when the vowel is spelled out explicitly (e.g.,อุบัติเหตุ is "ubat-het" and not "ubat-hetu") and is therefore similar to Hindi in that way; however, in Thai, there is never a terminal [a] after a terminal consonant unless one is explicitly written, unlike Hindi when sometimes the final [a] is pronounced as written.  When I say that "a vowel is necessary", I mean consonants that are impossible to enunciate sequentially without an intervening vowel; example: คน has an implied [o] (and therefore pronounced "khon") because it is linguistically impossible in Thai for "kh" to be followed by "n" without a vowel.  Conversely, ปลา is read as "plaa" and not "palaa" because ปล "pl" is a legitimate consonant group in Thai, so no intervening vowel is required. The point here is that no vowel cancellation mark or conjunct or any other method is required to indicate that p is not followed by [a], whereas in a true abugida such a mark would be necessary.  In Khmer, (from which Thai is descended) consonant groups are denoted by using subscripted consonants, a method that is very similar to the use of conjuncts in devanagari: the Thai script dispensed with subscripted consonants when writing Thai because it is not an abugida.--Gak (talk)


 * A "true abugida" seems like a fairly a useless concept. Thanks for the info as it's very interesting. But what do you call a language that has a strict phonetic interpretation, like Spanish and Thai almost are, and which English and Mandarin are perfect counterexamples of? That's not a phonetic alphabet (as they're not 1-to-1), but does seem like a useful characterization. Davilla (talk) 19:18, 24 January 2008 (UTC)


 * Most of the scripts now in use in India are true abugidas, so I am afraid I disagree that the concept is not a useful one. --Gak (talk) 04:34, 4 May 2008 (UTC)

Transliteration
The transliterations of the letters is clearly intended to convey the sound when an English speaker says a word. However, because so many Thais will automatically convert some letter endings (even when reading english) I would always avoid some combinations when transliterating. The r/n conversion is particularly worrysome given the transliterations in the article.

For anybody not familiar with the way the language should sound there isn't any simple transliteration that will give the correct sound to an english speaker and keep Thais out of trouble (this is on the en site after all).

Personally I think that a comprise could be reached by replacing ors on the letters with aw. For example Kor kai would become kaw kai (although for this one letter I think a strong case can be made for using gaw gai or even gaw guy). --KayEss 19:19, 29 Mar 2004 (UTC)

I think none of the &#3620; (ror reu), &#3620;&#3653; (ror reuu), &#3622; (lor leu), &#3622;&#3653; (lor leuu) are consonant or consonant-vowel. They should be all vowels. Cheers, Art


 * They are in fact digraphs; Lao, which otherwise has a MUCH simpler 'alphabet', has 11 of them! Lee 15:49, 27 January 2007 (UTC)Pawyi Lee
 * On the contrary. In Thai they are used like consonants, but in Sanskrit they are vowels [ṛ] [ṝ] and [ḷ] [ḹ].--Gak (talk) 07:55, 5 January 2008 (UTC)

They are ligatures! Like the Œ and Æ formerly used in English, in Thai they are not treated as consonants or vowels but considered their own category of non-consonantal symbols. Nintala (talk) 07:22, 18 October 2009 (UTC)Only &#3620; a

The romanization blues
In roughly 30 years of dealing with the Thai language as an Anglophone scholar, I have yet to find a completely satisfactory, one-size fits all, system of romanizing the language. Well, of course not, it's impossible, unless the Thai government some day does what the Chinese did and invent an official system. Even then, it's bound to be inadequate for some purposes.

My preference is for giving the value of a character's sound in the International Phonetic Alphabet the first time it is used and using some less-obscure equivalent thereafter, to keep everyone from going cross-eyed. Accuracy is particularly important in an article with a linguistic subject. If the subject were Thai politics, what difference would it make? But when the subject is the Thai language, there is a lot to be said for precision.

As for ro-ruesi and company: in Thai they are consonant+vowel combinations. They correspond to now-extinct Sanskrit vocalic consonants. You know, you can write Sanskrit with perfect accuracy in the Thai alphabet, because it has preserved all the original characters, even when they are redundant in the Thai language. The Lao alphabet, which is essentially the same as the Thai alphabet, differs primarily in doing away with all the phonologically redundant characters. Goodbye ro-ruesi.

Might we in that case just put a footnote to the effect that the 'r' at the end of the words is not to be changed to an 'n' as is normal with Thai words?--KayEss 18:38, 27 Jun 2004 (UTC)

Other language and alphabet articles indicate the pronunciation with X-SAMPA symbols. This article should probably do the same. - Jim Henry


 * Because like the IPA, it means nothing unless you've already learnt the system? Markalexander100 02:59, 3 Aug 2004 (UTC)


 * I'm now attempting to hammer out a standardized Wikipedia format for romanizing Thai over at Manual of Style for Thailand-related articles based largely on this article, please join in. Jpatokal 07:24, 17 Sep 2004 (UTC)

Other symbols
I know that technically they may not be part of the alphabet, but we need to mention the other symbols of Thai script somewhere ( kind of thing). My suggestion would be to extend the tone marks table (which already includes mai taikhu) to be an "other symbols" table. If necessary we could rename the page "Thai script". Markalexander100 08:14, 9 Jul 2004 (UTC)


 * The vowel table doesn't expalin ก็, which means "also" or "too", and is a shortened form of กอ, not ก. 203.154.48.180 22:31, 13 July 2007 (UTC)


 * Some of the additional unicode symbols have been explained in a new sanskrit secction. --58.181.249.244 04:12, 15 July 2007 (UTC)

In fact there is an official romanized trascription of Thai
The official transcript of Thai to Latin letters is used in most of the official road signs for example. It has some basic problems, since it has no tones, no difference between short and long vowels or between the two kinds of "o"... but it's still much better than the strange "English like" ways people use too often (with the strange "r" everywhere, even there is no "r" sound"). Anyway you can see the official transcript and an alternative one on http://www.geocities.com/raz_h_h


 * It's the Royal Thai General System of Transcription you refer to, right? Well, we try to use that one whenever possible - see Manual of Style for Thailand-related articles. BTW: welcome to Wikipedia. andy 20:34, 28 Oct 2004 (UTC)


 * Yes, I refer to the "Royal" transcript. And by the way, even here in the "Thai" page they use this problematic "English like" transcript with all these strange "r"s. Most of the "r"s should be ignored... If it's my page I would update it as soon as possible ;-) The transcript offered on http://www.geocities.com/raz_h_h is close to the official one with the necessary additions, taking into account also the common problematic influence of English (avoids "ee", "oo" etc.) and with additional subscripts for easy reading by westerners...


 * The strange "r"s are the "Harvard r" and function like the "e" in "Like"; that is, they only affect the pronunciation of the preceding vowel.Lee 14:33, 27 January 2007 (UTC)PawyiLee

Does anyone has information about these 3 Thai symbols: &#3629;&#3642; &#3629;&#3662; &#3629;&#3661; ?
Apparently they're used in Sanskrit and Pali and they're not really Thai characters at all (according to my Thai wife).

She thinks the last one is pronounced 'ang' though. --KayEss 08:27, 3 Nov 2004 (UTC)

As a further note, she says that most Thai monks should be able to read them as the much of the scriptures are in Pali and Sanskrit. --KayEss 08:28, 3 Nov 2004 (UTC)


 * The important part are of course the tone marks, the &#3629; is only the silent consonant needed as tone marks cannot stand alone. According to the &#3629;&#3662; (Yamakkan) is an ancient punctuation mark, the Nikhahit &#3629;&#3661; is part of a vowel symbol, but if alone it is a Pali consonant, and Pinthu &#3629;&#3642; is a marker of the final consonant in Pali, as well as having the same function as the Yamakkan. andy 17:09, 3 Nov 2004 (UTC)


 * Thanks very much. However, do you understand exactly how these 3 marks work? From you reference I understand that the first one can be used to mark a cluster even when in normal thai it would include a short "a" in between, the second mark adds also the keeping of the original sound of the consonant without changing it to another (like ch->t) and the last mark adds another consonant bet I'm not sure what consonant exactly. Is this true? Do you have any more information about the additional sound for the last mark?


 * All I kow is what was written on that site, maybe you can find more with the character names and google - that's how I found that site at first. But my wife didn't know those characters either and only could say the probably come from Pali, so they don't appear in any normal written Thai today anyway (or maybe anymore?), except maybe texts written for monks. andy 22:37, 3 Nov 2004 (UTC)

&#3629;&#3661; is part of the vowel อำ, used in words like น้ำ.


 * I don't know about the use of these letters in writing Pali with Thai script, but phinthu (อฺ) is used for phonetic transcription of Thai in dictionaries, including the Royal Institute Dictionary. It goes under a letter to indicate that it is the first consonant in a cluster--i.e. โกรธ would be written phonetically as [โกฺรด] ([kroot]), which disambiguates its pronunciation. This sort of disambiguation is neccessary, since a word spelled โกรธ could theoretically be pronounced [โก-รด] or [กะ-โหฺรด] under Thai spelling rules.
 * As for yammakkan (&#3629;&#3662;), the only place I've seen it used is by Butterfly Book House (สำนักพิมพ์ผีเสื้อ), which uses archaic (and sometimes etymologically questionable) spellings of words, in an attempt to bring them back into currency. In some of its books it states on an introductory page that yammakkan indicates "half pronunciation" of the consonant below it, and appears to be used exclusively in transcribing foreign words. An example of its use is the dedication of the Thai version of Roald Dahl's Esio Trot. It reads: อุทิศแด่ โคล&#3662;เวอร์ และ ลู้ค. The English version reads "To Clover and Luke." I'm not sure what this symbol is supposed to indicate, and I suspect that nobody but the Butterfly editors really know, either. rikker 02:52, 26 September 2006 (UTC)


 * I know how อ๎๎ was used. It used to stand above the first consonant in a consonant cluster to aviod ambiguity. Thus, เพ๎ลา would always be phlao, and always เพลา pheelaa. It is now obsolete. The other two have been used only in Pali texts, to correspond to some marks in that writing, and effects the pronounciation in certain ways. I saw a great text about it somewhere on the Internet, but I can't seem to find it right now. Yenx 23:52, 6 December 2006 (UTC)
 * &#3629;&#3661; indicates anusvāra in Sanskrit and Pali. Thais find it impossible to nasalise vowels, so instead, they add a consonant m or ng to the end of the syllable.  &#3629;&#3642; indicates virama, which means that the consonant should be pronounced without its inherent vowel.&mdash;Gak 08:18, 15 July 2007 (UTC)

Typewriter story
I've heard the typewriter story about the two deprecated consonants, but is it just an urban myth? Is there a reference for it? --KayEss 05:53, 26 Nov 2004 (UTC)


 * FWIW, on my Thai computer keyboard about half the keys are used for four symbols, and about half are used for three, so there should be plenty of room. Mark1 05:58, 26 Nov 2004 (UTC)


 * Don't forget we're talking about the old mechanical typewriters - on a computer keyboard much more is possible than it was at that antique stuff :-) Well, the story is from http://www.nationmultimedia.com/specials/100Bangkok/index_nov02.php - maybe someone have to go to the National Museum to verify those two letters are really left out. IIRC it had one of those typewriters on display there, but that's a long time ago I visited it. andy 12:11, 7 Dec 2004 (UTC)

y or &#616;?
"–, sara eu,, in French "du" (short)"

"–, sara euu,, in French "dur" (long)"

The sounds in French "du" and "dur" are y, not. So which one is it? User:PlatypeanArchcow


 * I think the key phrase here is "very approximate equivalents".  The fact that we resorted to French for an approximation may indicate that this is a sound which is quite foreign to the European mouth. Are there closer matches with which readers might be familiar? (By the way, you can sign comments with four tildas).  Mark1 02:29, 5 May 2005 (UTC)


 * In Thai the corresponding sound is not rounded, so it cannot be /y/, choice is between and  both of which come close. Sources differ, but in my ear  is closest. &minus;Woodstone 13:43, 2005 May 5 (UTC)


 * The Royal Thai Institute uses ɯ.--Gak 17:28, 15 July 2007 (UTC)


 * So does the IPA handbook and by now most related WP. Note that you were responding to a remark more than two years old. &minus;Woodstone 17:39, 15 July 2007 (UTC)

align to Thai section of IPA handbook
Previously I had used the general table of the IPA to construct the best rendering of Thai in my ear, using a few other sources as well. I now have adapted to the specific section in the IPA handbook on Thai. Changes are mostly in the ch, oe and ue (RTGS) sounds. At the same time I changed the romanisations to RTGS. &minus;Woodstone 15:38, July 10, 2005 (UTC)

Old Thai
Should this article contain some information about the old Thai script as seen on the Ramkamhaeng stone thing? Maybe a graphic of it would be good to see. --Dara 06:40, July 11, 2005 (UTC)

Number of vowels
A change in the number of vowels from 28 to 32 was reverted without comment. Counting the forms (except the implicit ones) in the table in the article comes to 32. In many books on the subject there is little consistency, especially in which diphthongs are counted. The reverter should explain why inconsistency between table and count is preferred above a seemingly arbitrary choice. &minus;Woodstone 12:36, 24 December 2005 (UTC)


 * I think Heron's point was just that uncited changes in numbers are suspicious. Mark1 13:01, 24 December 2005 (UTC)

That's right, Mark1, that was my point. I also agree with Woodstone about the lack of consistency in other references. I'm not sure that we can get away with stating a vowel count, since any number would be arbitrary. Our table has 35, of which the first 4 are not normally included in vowel lists. I think we've missed one: เ–ือะ, the short version of เ–ือ. That makes 32. Another four of our vowels (เ–็ –, –ั –, เ–ียว and เ–ิ –) appear only in the most eclectic lists, and not in my favourites at http://www.learningthai.com/vowels.html and http://thaiarc.tu.ac.th/thai/thindex.htm, so we are left with 28 core vowels. thaiarc.tu.ac.th lists only these. The next most popular number is 32, as at learningthai.com, which adds the four obsolete characters ฤ (ror reu), ฤๅ (ror reuu), ฦ (lor leu) and ฦๅ (lor leuu). I'm not going to edit the table yet, as I'm not very good at Thai. I'll wait for some expert opinions first. --Heron 22:33, 26 December 2005 (UTC)


 * The four vowels you prefer to exclude are actually all quite common.
 * three (เ–็ –, –ั –, เ–ิ –) are omitted from many books because they are spelling variants that only occur in a closed syllable.
 * the fourth (เ–ียว) is a diphthong ending in ว, which (next to the ones ending in ย) are usually not included. The reason this particular one is often included by exception is that it is not pronounced as you would expect from its composition.
 * The first four of the list in the article (implied a (–), implied o (– –), pali a (–รร–), and ua (–ว–)) are either not written explicitly or written using a consonant symbol.


 * Much of the difference seems to be caused by whether to count by spelling or by sound. For some vowels the way they are written in a closed syllable is different form how they appear in an open syllable (the table shows this by an added dash (–)). Here a list of correspondances following the same order as the main table:


 * – (implied a) is a way to denote short a (–ะ) in the first syllable of a composite word
 * – – (implied o) is the way to denote short o (โ–ะ) in a closed syllable
 * –รร– is a way to write short a (–ะ) in a closed syllable (in words of pali origin)
 * –ว– is the way to write ua (–ัว)  in a closed syllable
 * –ั – is the way to write short a (–ะ) in a closed syllable
 * เ–็ – is the way to write short e (เ–ะ) in a closed syllable
 * เ–ิ – is the way to write long or short oe (เ–อ or เ–อะ) in a closed syllable


 * เ–ียว is not pronounced like เ–ีย+ว (ia-w), but like io, (but RTGS is not consistent here)


 * The obsolete characters ฤ, ฤๅ, ฦ, ฦๅ are included in the consonant table
 * The first two symbols are not obsolete, such as ฤๅษี (hermit) and อังกฤษ (English)151.204.238.125 (talk) 07:15, 18 October 2009 (UTC)


 * So from the 35 vowels in the table in the article, if counted by sound (and not counting the regular diphthongs eding in ว or ย), there are 28 vowels. If counted by (explicit) spelling there are 33 (or 32 if เ–ียว is discounted).


 * We might put these conclusions in the article. &minus;Woodstone 12:12, 27 December 2005 (UTC)

Thanks for explaining, Woodstone. For the record, I don't "prefer to exclude" anything; I would rather we list all possible vowels, but also explain why some of them don't appear in other people's lists. As you yourself have demonstrated, it is impossible to state the number of vowels without qualification. --Heron 12:28, 27 December 2005 (UTC)

I have added 11 diphthongs; there are still about 5 more rare diphthongs and one alternative spelling in closed syllables. I am considering to expand the lead in the vowel section to explain the system better. &minus;Woodstone 15:45, 28 December 2005 (UTC)


 * In the views of a Thai speaker who never studied linguistics, it seems inconcistencies arise in part from the function of some diphthongs/tripthongs which in traditional Thai language (or at least what is taught grade school anyway,) are considerd to be vowels with closing consonants (namedly, ว and ย). For example, whereas "–วย sara uai" is listed in the article at the moment, there is no such thing as "sara uai" in normal usage of the Thai language, because it is considred to be sara ua (สระอัว) with ย as an end consonant.
 * To illustrate, If all the vowel combinations were to be arranged into the 32 traditional vowel sounds with the "default" form preceding, it should look something like


 * short vowels
 * long vowels
 * สระเดี่ยว
 * -ะ, -ั- (includes -ัย)
 * -า, -า- (includes -าย and -าว)
 * -ิ, -ิ- (includes -ิว)
 * -ือ, -ื-
 * เ-ะ, เ-็-
 * เ-, เ-- (includes เ-ว)
 * แ-ะ, แ-็-
 * แ-, แ-- (includes แ-ว)
 * โ-ะ, --
 * โ-, โ--
 * เ-าะ (doesn't exist in closed syllable)
 * -อ, -อ- (includes -อย), -ร (when ร is the closing consonant)
 * เ-อะ (doesn't exist in closed syllable)
 * เ-อ, เ-ิ-, เ-ย (ย is considered the closing consonant) and in few cases, เ-อ- (includes เ-อว)
 * สระประสม
 * -ัวะ (doesn't exist in closed syllable)
 * -ัว, -ว- (includes -วย)
 * เ-ียะ (doesn't exist in closed syllable)
 * เ-ีย, เ-ีย- (includes เ-ียว)
 * เ-ือะ (doesn't exist in closed syllable)
 * เ-ือ, เ-ือ- (includes เ-ือย)
 * สระเกิน
 * -ำ
 * ใ-
 * ไ-
 * เ-า
 * ฤ
 * ฤๅ
 * ฦ
 * ฦๅ
 * }
 * -อ, -อ- (includes -อย), -ร (when ร is the closing consonant)
 * เ-อะ (doesn't exist in closed syllable)
 * เ-อ, เ-ิ-, เ-ย (ย is considered the closing consonant) and in few cases, เ-อ- (includes เ-อว)
 * สระประสม
 * -ัวะ (doesn't exist in closed syllable)
 * -ัว, -ว- (includes -วย)
 * เ-ียะ (doesn't exist in closed syllable)
 * เ-ีย, เ-ีย- (includes เ-ียว)
 * เ-ือะ (doesn't exist in closed syllable)
 * เ-ือ, เ-ือ- (includes เ-ือย)
 * สระเกิน
 * -ำ
 * ใ-
 * ไ-
 * เ-า
 * ฤ
 * ฤๅ
 * ฦ
 * ฦๅ
 * }
 * ใ-
 * ไ-
 * เ-า
 * ฤ
 * ฤๅ
 * ฦ
 * ฦๅ
 * }
 * เ-า
 * ฤ
 * ฤๅ
 * ฦ
 * ฦๅ
 * }
 * ฤ
 * ฤๅ
 * ฦ
 * ฦๅ
 * }
 * ฦ
 * ฦๅ
 * }


 * The implied a (-) and ro han (-รร = un; -รร- = a) are usually treated as different concepts since they are inherently borrowed from pali/sanskrit. Also, the name of a vowel is the same regardless of its form e.g. -ะ and -ั- are both sara a (สระอะ) whereas mai han-akat is the name of the symbol itself. Similarly, เ-อ, เ-ิ-, เ-ย and เ-อว are all sara oe (สระเออ). Another point to note is -ว- is considered a vowel form in the same way -อ is, since ย, ว and อ all serve as vowel symbols (so ว is a vowel symbol in -ัวะ, -ัว and -ว- but in the other mentioned instances are treated as closing consonants.)
 * I am not saying the arrangement should be changed. One point I would like to point out, however, is that the names sara uai, sara oi, sara ai (for ัย and -าย), sara ao (for -าว), sara iu, sara eo, sara iao and sara aeo are not existent in the Thai language.--Paul C 13:59, 2 January 2006 (UTC)


 * Hi Paul C, thanks for the excellent structuring of the vowels above. It makes complete sense to me (as a very imperfect Thai speaker). We should work this into the article. &minus;Woodstone 14:10, 2 January 2006 (UTC)


 * 28 or 32 vowels, there are only 15 symbols. In So Sethraputra dictionaries the order of these symbols is |อ |ะ | ั |า | ำ | ิ  |   ี  |   ึ     ื  |  ุ  |  ู  | เ | แ  โ  |  ใ | but check a dictionary's preface to see what order it uses [if any]. Confusion arises because until recently vowels were considered mere symbols Lee 17:35, 7 March 2007 (UTC)

Punctuation
The everything2 page has some interesting stuff about the punctuation marks used with Thai script. Someone who knows about it should probably work it into this article. Marnanel 03:42, 23 November 2006 (UTC)

Vowels on Thai language page
The IPA transliteration of some vowels on the Thai language page differ from those found here. Namely, โ–ะ –ะ and -ั differ. The discrepancies lie in the inclusion of a glottal stop for the first two, and the disputed back/front pronunciation for the latter two. Wikky Horse 05:20, 20 April 2007 (UTC)

Sanskrit and Pali
The lengthy section on Sanskrit and Pali is very confusing, as it gives other values for some letters than Thai. It does not belong in this article. It would be better off in a separate article, on the development of various scripts. &minus;Woodstone 19:25, 14 July 2007 (UTC)


 * The Thai script is used to write Sanskrit and to write Pali, but I think it would be wrong to put this into the Sanskrit article, for example, because sanskrit does not itself have a writing system. Each abugida in India has its own set of characters which are used to write sanskrit/pali, and Thai, as an indic alphabet, is no different.  The section also answers the question above about what the additional symbols on a Thai keyboard (and in the Thai unicode table) are for.  I also disagree that this belongs in a separate article on development, because this system of writing Sanskrit is currently used throughout Thailand and is neither dead nor obsolete.  Also, this transcription system for Sanskrit words is currently used for names (both of people and of places): e.g., ศรีนครินทร์ (pronounced in Thai as "See-naka-rin") is written in Thai street signs as Srinagarindra (i.e., the name is transcribed using the Sanskrit values for all the letters and not the Thai values). There is a one-to-one correspondence with most other Indic alphabet, which means that Thai newspapers usually transliterate Indian names directly, thus giving them pronunciations most Indians would find unrecognisable. --Gak 04:17, 15 July 2007 (UTC)


 * The more the Sanskrit/Pali section grows, the more inappropriate it becomes as part of an article on the Thai alphabet. It gives all kinds of pronunciations that are simply non-existent in Thai. It should definitely be split off to another article. How about starting one on "History of the Thai alphabet"? &minus;Woodstone 17:54, 15 July 2007 (UTC)
 * But this is nothing to do with the history of the Thai alphabet: the section on Sanskrit and Pali talks about current usage. If you don't know the Sanskrit equivalents of the Thai letters, then it is impossible to see how "Mahatma Gandhi" can be transliterated into Thai as "มหาตมะ คานธี".
 * Then choose a better title. As it stands now, many of the pronunciations given are simply incorrect for Thai and thus confusing to the reader of this article. &minus;Woodstone 18:42, 15 July 2007 (UTC)
 * 1. The article is about the Thai script, not the Thai language. Why should it matter that the values listed are not for Thai?  The Thai script is descended from a script not originally used to write Thai: why should it surprise you to discover that it is also used to write other languages, any more than it should surprise you to discover that the Latin alphabet is used to write English, which is not descended from Latin?  In any case, the section is clearly titled Sanskrit and Pali: why should there be any confusion?  I disagree that this information belongs in another article, because the Thai script is not used outside of Thailand, and words borrowed from Sanskrit/Pali mix freely with native Thai words and are part of everyday speech.  You cannot put this information into the Sanskirt or Pali articles, because neither language has its own native script: the local script is always used with local adaptations.  That the Tamil script may be used to write Sanskrit is discussed within the article on Tamil script; likewise the article on the Sinhala alphabet discusses the specific alterations needed in that script to write Sanskrit and Pali.  Why do you feel that the article on the Thai script should be any different from articles on other Indic scripts?  This article is about the script, not the language.  2. The values are only theoretical in the sense that Thais use the Sanskrit values to romanise Thai, but still pronounce the words using the Thai values for all the letters (including the tones).  Hence, Bangkok airport is spelled "Suvarnabhumi" in English, but is actually pronounced "su-wan-na-poom".  If you feel that the write-up needs to be re-written to make that clearer then perhaps that is a valid criticism.  3. I agree that having two sets of values for each of the Thai letters is confusing, but that is not my invention: it is a fact of life in Thailand.  You only have to walk around Bangkok to realise that both romanisation schemes are in current use.  The Sanskrit values are used when transliterating personal names (certainly true of every single member of the Thai Royal family).  The King's name is spelled "Bhumibol" and never "Phoomiphon": of course this is confusing, but tell me where else in wikipedia is this explained?  Road names (particularly roads named for people) are likewise written out using the Sanskrit values (I gave you the example of "Srinagarinda"), as are numerous place names.  Have you tried driving around Bangkok?  Half of the street names are Sanskrit or Pali.  This is not some kind of esoteric exercise in obscurity:  this is one of the ways the Thai script is used in day-to-day life in Thailand, and also one of the issues not explained by many Thai language books for foreigners.--Gak 15:09, 16 July 2007 (UTC)
 * As you state above, many Thai words are descended from Pali, but they are always pronounced with the Thai values for all letters. So speaking of Pali values for the Thai script is incorrect. What you call the "values" are actually the classical form of romanisations of those Pali words, which are neither Pali nor Thai. Furthermore in romanisations in Thailand you never see any small dots under letters, nor are there any retroflex sounds in Thai, nor any aspirated voiced plosives. The romanisations you indicate are by no means universally applied for Pali descended words. Very often they follow the RTGS system. As a conclusion, this section has nothing to do with pronunciation of the Thai alphabet or with the Thai language, nor with the Pali/Sankrit languages using the Thai script, since that only becomes visible as soon as you go to show the romanisations, which are as you correctly state are "theoretical". Perhaps a better place would be in the article on Thai transcription, in a section explaining an alternative way to romanise Thai words of Pali/Sanskrit origin. &minus;Woodstone 22:17, 16 July 2007 (UTC)
 * I have rewritten the section to take into account your comments. There is not just an issue of transliteration, so making it a subsection of the Thai transcription article (which does not exist: it redirects to the Royal Thai General System article) would be inappropriate.  I would still argue that the information on Sanskrit belongs in this article and not a separate article, if only for consistency with every other article on Indic scripts in Wikipedia.--10:34, 17 July 2007 (UTC)


 * I just noticed that the article on Sanskrit contains a sentence transcribed in many scripts, but not the Thai one. Nor does it mention that Sankrit can be written in the Thai script. Is that an omission or a reflection of reality? Could you add some information there? —Preceding unsigned comment added by Woodstone (talk • contribs) 13:25, 16 December 2007 (UTC)
 * The sentence on the Sanskrit page can be easily transcribed into the Thai script: "ศิโว รกฺษตุ คีรฺวาณภาษารสาสฺวาทาทฺตตฺปรานฺ" and should be legible to any Thai monk (although he will not be familiar with it, because the quote comes from the Hindu scriptures and not the Buddhist scriptures). If you want to make it legible to the average Thai person, you have to write in all the short [a]s: ศิโว รักษะตุ คีรวาณะภาษาระสาสวาทาตัตปะราน.  If you want to paste in the Thai script and change the caption in the Sanskrit article to reflect this addition then please be my guest. --Gak (talk) 16:09, 23 December 2007 (UTC)

Break-out by region of vowel table
I seriously wonder if there is any added value to breaking out the vowel table to English speaking regions. Anyway, there should not be any IPA-like symbols in those columns, as there is a separate IPA column and that cannot depend on English regions. Also, the original column (NE US) should at least be kept intact, to ensure continuity. &minus;Woodstone 09:24, 15 July 2007 (UTC)
 * Vowel sounds vary greatly from country to country, but I am inclined to agree that only one English-speaking region should be selected and used for all the tables (for the sake of uniformity). If there is any doubt by other English speakers as to the correct pronunciation, then that is what the IPA column is there for.&mdash;Gak 15:40, 15 July 2007 (UTC)

Implied vowels
IPA symbols in square brackets. In this article, implied vowel for an open syllable is given as [a] and that for a closed syllable as [o]. In fact, in spoken Thai, both [a] and [ə] are common as the vowel for an open syllable: e.g., ถนน is usually [tʰanon], but สบาย is usually [səbaːj], not [sabaːj]. There is even more variation for the closed syllable, which in addition to [o], may also take [ɔ] (e.g., คน is often kʰɔn) or [u] (e.g., ฝน is usually pronounced [fun]).--Gak 00:36, 18 July 2007 (UTC)
 * Unstressed implied /a/ can be reduced to a schwa in informal speech, just like in many other languages. That does not really change the phoneme. The difference between short /o/ and short /ɔ/ in Thai is minimal, as is reflected also in the RTGS system (both "o"). It is very difficult to decide which one occurs in which word. By convention, the implied "o" is denoted as /o/. I never heard ฝน (rain) pronounced /fun/. &minus;Woodstone 20:11, 18 July 2007 (UTC)
 * And สยาม is pronounced si-yam, while พญา is read as pi-ya. No sign of a schwa there? --Gak 07:30, 24 July 2007 (UTC)

Thai keyboard
Need help: does there exist a Thai keyboard layout that allows for the arrangement of Thai letters on the same keys as the corresponding English letters? I've downloaded one like this for the Arabic alphabet, which some hacker made, which has the Arabic letters on the same keys as the English letters in the QWERTY keyboard, and it's very useful in that I don't have to learn a new keyboard layout and go through all 30+ keys to find each letter. The only difference is that sometimes there are 2 or 3 letters for the same sound (as in Thai), so sometimes you have to use the shift key to get the less common letters for the same sound. Thanks, Badagnani 01:22, 13 October 2007 (UTC)


 * I have never seen such a keyboard. It would be practically impossible to construct, since apart from some doublings that could be solved with the shift key, there are 4 p's, 8 t's, 4 s's and 4 k's. On the other hand there are letters that stand for sounds that do not have a letter in English, such as the one expressed by "ch". Furthermore, mapping of the redundant letters c, q, x is not obvious. Note that – even worse – the Thai digits on the standard Thai keyboard do not match the Arabic/Eglish positions of some of those digits. &minus;Woodstone 12:20, 13 October 2007 (UTC)

Correct IPA transcription
The thai language is often not transcribed into roman letters according to IPA. Consonants are transcribed as devoiced when in reality they are voiced. For example, ko kai is pronounced as a g (IPA), not as a k as indicated. The Thai capital in Thai is written "Krung Thep" but pronounced as grung tep. Additionally, r's are often put in place meant to be silent as a speaker of british english would speak them. For example, the Thai dish, larb is pronounced lob. This creates a problem. In the table of consonants listed, the transcription to IPA is as the letters are normally written in roman letters but not as they are pronounced. —Preceding unsigned comment added by 67.100.132.234 (talk) 22:34, 29 November 2007 (UTC)


 * This is a recurring comment caused by the way many tourist pronunciation guides use transcriptions. The consonant ko kai is not voiced, just as to tao and po pla are not voiced. These are all not aspirated. The confusion comes from the fact that ko kai does not have a voiced counterpart, as are do dek and bo baimai for the other two. See more explanation here. In the IPA column the "r" for vowel lengthening does not occur. You see it in the "variants" column, which shows several of the many ways trancriptions occur. &minus;Woodstone (talk) 04:02, 30 November 2007 (UTC)


 * Are you sure ko kai is devoiced? I live in a thai community and they insist that the consonant it voiced and that it's acutally the tourists pamphlets that have it devoiced causing confusion.  It certainly sounds voiced and I've been taught to voice it.  I've even checked it in Praat and it looks voiced  —Preceding unsigned comment added by 67.100.132.229 (talk) 00:10, 15 December 2007 (UTC)


 * Quite sure. Confirmed by the "Handbook of the International Phonetic Association" (ISBN 0-521-63751-1). Much confusion is caused by the distinction between voicing and aspiration. The plosives can be voiced or unvoiced and aspirated or unaspirated. In many Indian languages all four combinations exist, in English (in initial, not clustered position in a syllable) only the combinations voiced/unaspirated and unvoiced/aspirated occur. Therefore English native speakers have great trouble pronouncing or hearing an unvoiced/unaspirated consonant like ko kai (to tao, po pla). The confusion is compounded by the fact that in Thai the voiced/unaspirated combination does not exist for the gutturals (unlike do dek, bo baimai). To my knowledge, there is no difference between Isan and standard Thai in this respect. The pattern is as follows, giving IPA and some Thai spelling:


 * {| class="wikitable" style="text-align: center"

!  || colspan="2" | unaspirated || aspirated !  || voiced || unvoiced || unvoiced ! dental ด|| [t] ต|| [th] ท ! labial บ|| [p] ป|| [ph] พ ! guttural not present in Thai || [k] ก || [kh] ข
 * [d]
 * [b]
 * [g]
 * }


 * The voiced guttural does not occur in Thai. Ask your Thai friends to exercise the relations in the table and judge again. The Thai letter ก is pronounced unvoiced and is thus systematically best represented by "k". This is consistent with both IPA and the Royal Thai Institute guideline. &minus;Woodstone (talk) 13:18, 15 December 2007 (UTC)


 * ก ko kai is not voiced in standard Thai (the central dialect), but is voiced in some parts of the north and northeast (and possibly in the far south, but I am not sure). Where are you in Thailand?  Are you sure your friends are not speaking Thai with a regional accent? There are many differences in pronunciation especially in the north and northeast: for example, in the north, all consonants are unaspirated. --Gak (talk) 16:19, 23 December 2007 (UTC)

Possible error for Pali "t"?
Is ฅ really used for the Pali "t", or is it supposed to be ต? Echalon (talk) 21:13, 25 February 2008 (UTC)
 * It's a typo. Well spotted. --Gak (talk) 04:37, 4 May 2008 (UTC)

Introductory paragraph
The intro paragraph seemed a bit too cluttered and detailed, so I moved portions of it the appropriate sections. For information that didn't fit anywhere else, there's now an Orthography section. Wikky Horse (talk) 03:14, 4 July 2008 (UTC)

Thai alphabet - use in advertising and magazines
The Thai alphabet used in advertising and on the covers of magazines is very different to that used in written text. It contains western-looking letters such as backwards C's and U's, and a slightly misformed S. An article showing the correspondence to the normal representation form would be very useful. —Preceding unsigned comment added by 89.245.175.252 (talk) 10:58, 19 October 2008 (UTC)


 * It's a style of writing that removes the small circles from the characters. Otherwise, it is still the same alphabet. Because this style is not the official one, it probably doesn't exist as a different set of characters in Unicode and would be difficult to place in a table. Wikky Horse (talk) 02:15, 29 October 2008 (UTC)
 * Many manuals teaching Thai include photos of Thai writing used in advertising and magazines; would a gallery of same work for this article? Pawyilee (talk) 13:42, 30 October 2008 (UTC)


 * It's always the same alphabet, but printed in different fonts, just like these exist for the latin alphabet. There are 4 major font styles:
 * the standard one with small circles on most letters
 * a classic one in cursive style, with many added flourishes
 * an old variant where serifs are used instead of the small circles
 * a modern variant where the circles are omitted ( in cases where ambiguity would arise, in the less frequent one has a small bar to replace the omitted circle )
 * These fonts also exist for computer rendering, but I'm not sure if copyright regulations allow to add images of some representative letters in the various fonts to WP. &minus;Woodstone (talk) 15:53, 30 October 2008 (UTC)


 * The website f0nt.com (note that's a zero, not an o) freely distributes Thai fonts of their own creation (retaining copyright, though). It should be straightforward to get the site owner's permission and post an image showing some letters of the alphabet in several fonts, right? --rikker (talk) 16:23, 30 October 2008 (UTC)
 * ARTWORK is what we mean, not fonts. For instance, what looks like the number "17" rendered with simulated Chinese-character-strokes spells "เจ". The word means vegetarian, though you'd have to know that to be able to read it. Stylized representation on what looks like a squared-away yellow traffic sign identifies a vegetarian restaurant or food shop. I have a picture of one that I'll upload when I get a round tuit. Another example is "USA" that can be altered artistically to "นรก", which means "Hell"; "อเมริกา" can be altered so as to simultaneously spell "America". I think "นรก" as "USA" was unintentional: the poster advertised a blood-and-guts movie that I don't think had anything to do with the USA; but it stopped me in my tracks when I saw it. Unfortunately, I did not have my camera with me, nor do I have an example of "อเมริกา" looking like "America". I just bring these up to emphasize that we are talking, not about fonts, but artwork. Pawyilee (talk) 06:33, 31 October 2008 (UTC)
 * I think the original question was probably just a matter of fonts; seems they think that there's a different "Thai language" used on different signs. I see what you're referring to now, though. Breeze detergent also looks like USA, and they design the Thai Harry Potter logo to look very similar to the English letters, etc. That would definitely make for an interesting article. --rikker (talk) 10:55, 31 October 2008 (UTC)

is uploaded at Wikimedia in Category: Rendered texts in the script of Thailand: it has no introductory paragraph setting forth what it's for, but, seems like a good one for what we've been talking about. Nowadays, Wikimedia asks that category pages have intros. Would you (or other editors) write one? To say it serves as repository for stuff like this, or direct it to where it really belongs? (I also added mine to Signs in Thailand | Religion in Thailand | Culture of Thailand ). Pawyilee (talk) 14:39, 3 November 2008 (UTC)
 * To get back to the original objection, backwards C's and U's, and a slightly misformed S: backwards C is likely ว, and U, likely บ, but could be ข; a slightly misformed S is likely ร but could be ธ. An upside down U is almost surely ก. Assuming they are sans serif. With a character like W, however, sans serif is likely พ but serifs can change it to ผ or ฟ or ฝ or even ฬ. At f0nt.com, the zero is a red ๐ but stands in for อ. เ่จ gives just one example of Thai being made to look Chinese: It can also be made to look Lao. Both pix and context are needed, and even then can confuse even a native! That's why I think a gallery would be a good idea (if someone renders them to Rendered texts in the script of Thailand !) Pawyilee (talk) 15:28, 3 November 2008 (UTC)


 * I noticed the same thing with signs like these: We don't need to worry about copyright with non-vector images of fonts. If s.o. has a font like those in the photo, we can use a .png image of a Word document.


 * Oh, here's the "USA" detergent: kwami (talk) 13:05, 21 December 2008 (UTC)

inherent vowels
Something I'm not clear on: if you write CC, the inherent vowel is /o/. When is it ever /a/? When you write CCV? kwami (talk) 11:13, 20 December 2008 (UTC)
 * I think it has to do with usage, and local usage at that. Though not directly related to your question, if you have So Setaputra's dictionary, look up เจ้า : he gives the one spelling but in his phonetic re-spelling gives จ้าว as the proper pronunciation for some usages, which he specifies. I mention this as an extreme case that I encountered while roaming the dictionary; only a real scholar can answer you properly. Pawyilee (talk) 18:05, 20 December 2008 (UTC)
 * Generally, the implicit vowel is /o/ in a closed syllable (pronounced CVC) and /a/ in an open syllable (pronounced CV), but there are a few cases (of Pali origin only) where CV has /o/. For example ถนน (thanon, road) consists of syllables ถ (th, implicit a) and น	น (n, implicit o, n). &minus;woodstone (talk) 21:34, 20 December 2008 (UTC)
 * Nice example. I'll add it to the article. kwami (talk) 00:20, 21 December 2008 (UTC)

More generally, how precise is the IPA here? /a/ is described as the vowel in hut, and /o/ as the vowel in boat. (Looks like this was partially addressed above, but it could use more comprehensive coverage.) kwami (talk) 00:25, 21 December 2008 (UTC)


 * The /a/ and /o/ are close to pure IPA (like the Italian vowels). Hut and Boat are poor examples (but neither /a/ nor /o/ exist in English). Both inherent vowels are short (length is phonemic in Thai). In fast speech, the /a/ can sound more towards /ɑ/ and /o/ towards /ɔ/, but they are never diphthongised. A curious phenomenon is the interaction of implicit vowels with the vowels that are written around the initial of their syllable. For example is written around its initial, symbolised by a dash, as เ–าะ. The word /tɕʰapʰɔ/ (only) is written เฉพาะ. The explicit vowel's split extends over the implicit vowel's initial. &minus;woodstone (talk) 11:25, 21 December 2008 (UTC)


 * I know some Indonesian scripts do that, but I didn't realize Thai did. (I "learned" Thai from the Bangkok yellow pages. Not the most fruitful introduction, but it did allow me to follow street signs.) I'll add that is as well.


 * I assume the circumfix vowels only encompass an inherent vowel which precedes them? Can they encompass more than one? Also, the glottal stop in short vowels, does it only occur word-finally?


 * Where do the high-mid-low terms come from? The low-tone consonants never trigger low tone, and the high-tone consonants are never high, but often low. "Class 1, 2, 3" would seem to be less confusing.


 * Thanks, kwami (talk) 11:33, 21 December 2008 (UTC)


 * Yes, a circumfix vowel only encompasses a single additional preceding consonant. Note that this way of writing circumfix vowels blurs the difference between initial clusters and an encompassed syllable with implicit /a/. The glottal stop occurs only syllable final (or initial, written as zero vowel อ). It is weak, but audible in careful speech. The terms high-mid-low are used in all textbooks, never any other terms. I do not know the Thai terms or the origin. &minus;woodstone (talk) 12:14, 21 December 2008 (UTC)

After reading the article circumfix, I'm not sure the word is correctly used here. Here we do not have a morpheme around another one, but a spelling device. This feature concerns not the language, but the script. &minus;woodstone (talk)


 * I'm not sure if that's considered sloppy or not. "Subfix" is only used for orthography, AFAIK, and is even used in chemical nomenclature. (It's easier to attest than the others, because it's not swamped by morphological usage.) In our Wylie transliteration article, we say, "In Tibetan script, consonant clusters within a syllable may be represented either through the use of prefixed or suffixed letters, or by letters superfixed or subfixed to the root letter (forming a "stack")." With Maya glyphs (e.g. in the Outline Dictionary of Maya Glyphs and Glyph Studies), 'prefix', 'postfix', 'superfix', 'subfix', and 'infix' (the latter when one glyph is embedded inside another) are all used. I think such usage is fairly common. kwami (talk) 14:02, 21 December 2008 (UTC)

While correcting the tone from a dictionary, I saw I had the wrong spelling. After some digging I found the transposition goes further: after an implicit open syllable, if the second syllable starts with a nasal or liquid, the tone is determined by the preceding consonant class. &minus;woodstone (talk) 14:31, 21 December 2008 (UTC)


 * Are diacritics then used if the tones of the first and second syllable are not both compatible with the tone class of the initial consonant, or does this reflect a constraint of Thai phonotactics? kwami (talk) 14:50, 21 December 2008 (UTC)


 * Diacritics can still be added. They are placed on the second consonant. The behaviour is the same as for initial consonant clusters (which can only have /l/, /r/, or /w/ second). It looks almost as if the /a/ is just there to break the cluster. &minus;woodstone (talk) 15:08, 21 December 2008 (UTC)


 * But the inherent vowel is not predictable, is it? kwami (talk) 15:27, 21 December 2008 (UTC)


 * Correct. In case of an allowed cluster, it is unpredictable if an implicit /a/ is inserted. There are other cases where the syllable division is not predictable. One complicating factor is that sometimes a final consonant of one syllable (written once) doubles as the initial of the next syllable with implied /a/. In that case the syllable is not written at all! &minus;woodstone (talk) 16:12, 21 December 2008 (UTC)


 * I'm a native speaker of Thai and found the discussion interesting but hard to follow. I'm wondering if the article is based on a certain linguistic frame of reference, for I'm finding it difficult to compare to what set of rules I learned in school.


 * For instance, the aforementioned usage of initial consonant clusters is known in Thai by the concept of อักษรนำ, which includes ห นำ and อ นำ. (Initial clusters with /l/, /r/, or /w/ are known separately as คำควบกล้ำ.) My sixth-grade Thai textbook gives the following treatment of the concept:
 * Defined as a group of two consonants accommodated by the same vowel.
 * Placement of vowels acts as if the two were one entity; above- or below-line vowel marks and tone marks are placed over or under the latter consonant.
 * With the exception of ห นำ and อ นำ, both consonants are voiced sounded, with the first assuming the /a/ vowel sound, but with reduced fullness (อะกึ่งมาตรา).
 * The first consonant may be any class. A high consonant may lead any consonant class, but a middle or low consonant may only lead a low consonant.
 * อักษรนำ may act as closing consonants. In such cases, only the first consonant serves as the closing consonant, while both serve as the leading consonants of the next syllable.
 * If the combination is that of a high or middle consonant leading an unpaired low consonant (อักษรเดี่ยว, referred to above as nasal and liquid consonants, which are rather Pali/Sanskrit terms and not generally used to describe Thai characters), the latter assumes the tone properties of the former.
 * Otherwise, the following consonant assumes its default tone properties.
 * For ห นำ and อ นำ (strictly, ห นำอักษรเดี่ยว and อ นำ ย), only the following consonant is voiced sounded.
 * อ leading others acts the same as other regular อักษรนำ


 * This example treats the implied /a/ (คำไม่ประวิสรรชนีย์) as being mostly the same as อักษรนำ. In actuality, there is considerable controversy on the classification of such language rules, which were mostly retrospectively created, as can be seen in this blog post, and definitions vary. Some sources define อักษรนำ differently as having a low consonant as the following one, and classifies the rest as คำไม่ประวิสรรชนีย์. Whether or not the reduced /a/ amounts to a separate syllable (พยางค์) is also debatable.


 * Back to kwami's question, though: from a Thai linguistic point of view, inherent vowels may be governed by the aforementioned rules of อักษรนำ or คำไม่ประวิสรรชนีย์ (where the /a/ vowel sound is implied for the first consonant), or สระลดรูป (reduced vowel forms). Most syllables written as CC are considered to have the /o/ vowel in reduced form (กก is pronounced /kok/). /ɔː/ may also assume this reduced form when ร is the closing consonant (as in กร, /kɔːn/) and in the word บ่, and ัว appears with a closing consonant in reduced form as -ว-, though in this case ว acts as the vowel mark. I understand that the first two concepts are also treated as สระลดรูป in terms of syllable construction, but I'm not sure.


 * For example, the following words are written as CCC but have different rules governing their pronunciation:
 * ถนน is an อักษรนำ with the vowel /o/ in reduced form, and is pronounced /tʰà nŏn/.
 * ขจร is also อักษรนำ, but with the vowel /ɔː/ (determined by ร as the closing consonant), and is pronounced /kʰà tɕɔːn/.
 * สรณ, a direct loanword from Pali, is คำไม่ประวิสรรชนีย์, has the vowel /a/ and is pronounced /sà rá ná/ (in actuality it is more like /sa ra ná/, although this is impossible to transcribe in Thai).
 * --Paul_012 (talk) 18:29, 21 December 2008 (UTC)

That is a lot of food for thought. My Thai is limited to simple daily life topics and I can certainly not read grammar books. Do I gather correctly from the above that the term "cluster" is used also for cases where an /a/ is inserted? Usually it stands for several consonants pronounced together. One statement above I do not understand: both consonants are voiced. In the example ถนน, the ถ does not sound voiced. &minus;woodstone (talk) 09:14, 22 December 2008 (UTC)


 * I simply meant sounded rather than voiced, sorry for the confusion. The two are treated as different concepts, but I don't know what they are called in English. --Paul_012 (talk) 10:19, 22 December 2008 (UTC)

Vowels closed with ย or ว
User:Paul 012 has reworded all vowels combined with ย or ว. This aligns with the proposal above at talk:thai alphabet proposed by User:Paul C (I just noticed he is probably the same person). It might be a good idea to restructure the whole table to reflect this. We could add one or more column to indicate the possible closing additions and their pronunciations. What do you think? &minus;Woodstone (talk) 21:16, 29 December 2008 (UTC)


 * Yes, that was my earlier signature (same user, though). My opinion is that unless the current structure is based on a specific convention outlined in the RTGS or other widely-used source, it should be preferable to follow the Thai conventions of 32 vowels and diphthongs. I'm not sure about the status of ฤ, ฤๅ, ฦ and ฦๅ among contemporary Thai linguists, but have heard that there is a trend to no longer consider them as vowels, as they don't function as such in Thai. --Paul_012 (talk) 09:24, 30 December 2008 (UTC)


 * ฤ, ฤๅ, ฦ and ฦๅ are sanskrit vowels. In Thai they function as consonants and are treated as such in dictionaries. I support reverting to the Thai convention. --Gak (talk) 01:02, 7 February 2009 (UTC)

Indic script warning
Thai is an indic script, but as noted in the article itself, it is written using visual and not logical order, so the warning about indic script support is not relevant and has been removed. --Gak (talk) 01:01, 7 February 2009 (UTC)

"Imagine Peace" wiki page needs help
Help is needed to edit the Thai image on Imagine Peace Tower to show the proper characters in Thai script. Photograph of this panel is at: http://commons.wikimedia.org/wiki/File:Imagine_Peace_1.jpg and the characters used should reflect this exactly even if some would translate the English phrase "Imagine Peace" differently. Thanks. Irv (talk) 16:16, 16 July 2009 (UTC)


 * Never mind. The script in question is Tibetan.  My error. Irv (talk) 10:38, 19 July 2009 (UTC)