Help talk:IPA/Thai

Untitled

Should [a:] be given as the or an alt trans. for mid tone? It is frequently not marked.

Should diphthongs end in [ə]? They don't quite reach [a], and that would have the benefit of making it clear that they're diphthongs.

— kwami (talk) 06:40, 8 September 2010 (UTC)

First remark above is too garbled to understand. (Now I see: "the or an".) I would support unmarked midtone (I'm never in favour of alternatives in these pages). The IPA handbook uses [a] in diphthongs; The RTGS also uses a; let's stick to the familiar usage.

Another question: should we list the long vowels (being simpler) first?

−Woodstone (talk) 07:01, 8 September 2010 (UTC)

We have both marked and unmarked tones in the articles, so until that is cleaned up we should have both here too. But there are only a few articles. — kwami (talk) 08:52, 8 September 2010 (UTC)

Changed the a-final diphthongs to ə-final, following Lao. This is also more intuitive from German etc, and it's doubtful the Thai vowels ever actually reach [a]. — kwami (talk) 20:05, 6 November 2010 (UTC)

What's that based on? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 20:21, 6 November 2010 (UTC)

What I've seen is phonemic /a/ but phonetic [ə] (or [ɛ, o] before [j, w]). We have that description for Lao, and our vowel chart (from the Handbook?) has the target as [ɜ] rather than [a]. — kwami (talk) 21:41, 6 November 2010 (UTC)

IPA diacritic for Lao low falling tone

Dearest all,
Where is this diacritic: a̭ (below the a) from? Is it an official IPA symbol? If so, would you please give me some references? I have attempted to replace it with a᷆ as it is an IPA symbol for low falling tone, but my edits have been reverted. The other symbol that I have seen in transcription for the Lao low falling tone is ȁ (http://sealang.net/lao/dictionary.htm), but we do not see it for Lao here on Wikipedia. If a᷆ is not to be preferred (I assume it's a fairly new IPA symbol; so, we don't see it in linguistic articles on Lao outside of Wikipedia), then I think ȁ should instead be used because it is at least an IPA symbol. However, I will stop reverting edits now as I do not see a point in doing so anymore.
Best,
--Alif Silpachai (talk) 07:47, 28 December 2013 (UTC)

The official IPA symbol is not used much because it's difficult to read; it also doesn't have the best font support (though maybe that's not the problem it used to be). Usually when those newer diacritics are required, phoneticians just switch to Chao tone letters. Since Thai tone is fairly simple, we went with diacritics, and this is the single case where the basic five don't suffice. A subscript grave accent was used in the IPA until 1989; since it was never reassigned, it would still be understood to have that value. That would avoid the legibility problem of the new symbol. The double-grave for bottom tone might also work. (It's not officially a falling-tone symbol, but it's not uncommon to apply symbols a bit loosely.)

In checked syllables, we could simply transcribe it as low, since it's the only low tone. — kwami (talk) 08:08, 28 December 2013 (UTC)

Thank you for your response. It would be difficult to write superscripted Chao letters here on Wikipedia like the way they do in linguistics articles. Personally, I think Chao letters should be used in phonetic transcriptions not phonemic. Alternatively, some linguists simply use numbers to indicate which tone it is: 1 = first tone, 2 = second tone etc. If the diacritic has been outdated for 24 years then I don't see a point in using it anymore. We should either use the new and proper diacritic (a᷆) or ȁ. How do I put this into a vote? --Alif Silpachai (talk) 08:35, 28 December 2013 (UTC)

Northern Thai?

We apparently have a separate Help:IPA for Northern Thai but this help page still seems to encode for Northern Thai. If we're already covering Lao and Thai together, I'm not sure if we would need a separate guide for Northern Thai. Thoughts? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 22:04, 22 December 2014 (UTC)

Now also an Isan Thai column was added. However there is no difference at all, except some omitted letters. This does not serve any purpose. The idea of this page is to show how each symbol is rendered in IPA. It does not matter if a variant of Thai does not use all symbols. I propose to revert to only one column for Thai. −Woodstone (talk) 05:13, 6 September 2017 (UTC)

Isan Thai is basically Lao written in Thai script, so there may be an argument for a separate column, but I think it can be handled with footnotes. The only real difference in Isan that's relevant to this page is that <ร> and <หร> are /l/ or /h/ (eg. รำ = /lám/; รัก = /hāk/) and <ญ> and <หญ> are /ɲ/ and ฉ, ช, ฌ (/tɕʰ/) aren't used. Tone contours are also different, but I think that's already handled sufficiently on this page. So it's a little more than "some omitted letters" since a few letters have different values, but that can easily be indicated with footnotes. Northern Thai, on the other hand, is perhaps different enough to warrant the current situation of being a separate IPA Help page.--William Thweatt ^TalkContribs 23:18, 6 September 2017 (UTC)

@WilliamThweatt: In the case of Northern Thai, do you mean differences in spelling? Mr KEBAB (talk) 09:30, 9 September 2017 (UTC)

Correspondence between IPA tones and Thai/Lao tone marks (Mai ek/tho/etc.)

This article gives the IPA tones present in Thai and Lao, but it fails to show which tone marks (called mai ek/eak, mai tho/toh, mai tri/dtree/ti, and mai chattawa/juttawa/catawa) correspond to such tones. According to this external link on Thai tone marks, mai eak is low2, mai toh is falling3, mai dtree is high4, and mai juttawa is rising5; no tone mark means mid tone, I assume. Probably the same applies to Lao. It would be useful to add those tone marks to the tones table in this article. Thank you. 24.139.84.34 (talk) 16:52, 17 July 2016 (UTC)

It is much more complicated than that. The tone marks don't "correspond" to tones. The tone is an integral part of the spoken word, just as important as the consonants and vowels. In writing, the tone of a spoken word is indicated with combinations of high/mid/low class consonants, vowel length, final consonants and, sometimes, tone markers. The tone indicated by a particular spelling depends on all 4 of these factors, not just the tone marker. Furthermore, Thai tones are different than Lao tones. In fact, different dialects within both Thai and Lao have different tone contours and even totally different tones. Central Thai has 5 tones, Chiang Mai Thai has 6 tones, most Lao and Isan dialects have 6 tones, while some have 7 and Lao in Luang Prabang has 5. So, in central Thai, a mai ek with a low class consonant indicates a falling tone, but a mai ek with a mid class consonant indicates a low tone. In most dialects of Lao, a mai ek with either a low or mid class consonant indicates a mid level tone. There is no one-to-one correspondence between tone marker and spoken tone. See Thai language, Thai alphabet#Tone, Lao language#Indication of tones and Isan language#Vientiane Lao Dialect for more info and references.--William Thweatt ^TalkContribs 20:33, 17 July 2016 (UTC)

Move discussion in progress

There is a move discussion in progress on Help talk:IPA which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 16:18, 15 July 2017 (UTC)

Consonant positions

The page should, but doesn't, indicate the different symbols representing each consonant when it is in different positions (initial, medial, final); it currently shows the symbols representing the consonants when they are in initial positions only.

For example:

• The consonant "ข" is represented by [kʰ] when it begins a syllable (e.g. "ขัน" = [kʰǎn]), but is represented by [k] when it ends a syllable (e.g. "นัข" = [nák]).
• The consonant "ญ" is represented by [j] when it begins a syllable (e.g. "ญี่" = [jî:]), but is represented by [n] when it ends a syllable (e.g. "ธัญ" = [tʰan]).
• The consonant "ช" is represented by [tɕʰ] when it begins a syllable (e.g. "ชุน" = [tɕʰun]), but is represented by [t] when it ends a syllable (e.g. "นุช" = [nút]).

--Miwako Sato (talk) 19:44, 19 July 2018 (UTC)

The page is meant to describe the sounds denoted by each Latin symbol. As a bonus it also lists all Thai letters that it can represent. There is no need to specify final positions separately, because the appropriate Latin symbols are chosen when in final position. So /k^h/ is never used in final position. Or when ญ occurs in final position, /n/ is used to represent it. −Woodstone (talk) 02:09, 20 July 2018 (UTC)

I think those references're from brahmi writing system. Juidzi (talk) 15:01, 27 February 2020 (UTC)

"อย" is this a consonant?

surprised to see "อย" as a consonant.

I never heard of อ used to change the definition of a consonant. rather it is used as a vowel.

Looks like an error to me. but my Thai is too rudimentary to have a definite view.....Jazi Zilber (talk) 18:05, 20 October 2019 (UTC)

Yes, the two letters together represent /j/. The อ in this case is called "อ นำ" and the digraph is used to make a mid-class /j/ consonant (ย by itself is low-class) in four common words: อย่า, อยาก, อย่าง, and อยู่ much the same as ห is used to make high-class digraph consonants out of low-class letters.--William Thweatt ^TalkContribs 19:06, 20 October 2019 (UTC)

Confusing footnotes

The footnotes are largely incomprehensible. Under [h], I'm told "For the Southern Thai, allophone with x." What does this mean? When are editors to transcribe with [h] and when with [x]? Why isn't [x] in the table? The same for [w]. "For the Central Thai, replaced by ʋ in Capital accents. For the Southern Thai, replaced by ɴ." Are we encoding for Central, Capital, and Southern accents? If so, why not do like it's done at Help:IPA/Catalan or Help:IPA/Portuguese? This is a bit of a mess. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 22:04, 30 October 2019 (UTC)

It looks like some editors mistake this page for an explanation of Thai phonology, including all dialectic variants. Just as we do for English, one way of diaphonetic transcription is intended, and automatic phonetic variations are not indicated. Much of info from earlier attempts has already been moved to the main namespace. Here we should go back to just one (the official) dialect.−Woodstone (talk) 14:37, 31 October 2019 (UTC)

Purpose of page

Once again there is creep in this page to incorporate a host of Thai dialects. This page is in help-space. It is not meant as a guide to distinctions between Thai dialects. The purpose is to show a transcription for use in WP articles to represent Thai words. As such it should give simply one sign for each phoneme. Just like in the comparable English page, each sign is diaphonemic and is derived from the main dialect(s). So all doublings in the table should be removed. In my opinion also the footnotes should go, because they only serve to create confusion. The information on Thai dialects might find a place in article-space. −Woodstone (talk)

This page should be reverted back to the version where only one variety is explained. There are not even any sources cited for the clear-cut definition of these dialects, and if we are to list every variants possible for each phoneme, I'm afraid we have to add [θ] for /s/, [x] and [kx] for /kh/, etc., as well. --Potapt (talk) 00:42, 4 February 2020 (UTC)

Thai lanna & lao language, can hear ⟨x⟩ sound. But central thai and southern thai are not. Juidzi (talk) 05:31, 28 February 2020 (UTC)

Ps, Thai lanna is Kham keuang language. Juidzi (talk) 05:32, 28 February 2020 (UTC)

I wasn't talking about Northern Thai/Kham Mueang though. --Potapt (talk) 14:40, 9 March 2020 (UTC)

Please watch my talk in Help:IPA/Lao

See something on my talk. Juidzi (talk) 09:25, 28 February 2020 (UTC)

https://en.m.wikipedia.org/wiki/Special:MobileDiff/942998364 Juidzi (talk) 09:28, 28 February 2020 (UTC)

About vowel "◌ิว"

I think it should not use the writing "io", because it's sound like "i(a)o" in chinese.

These are better writing choices 1. iu 2. ew 3. iew Juidzi (talk) 10:39, 9 March 2020 (UTC)

In the romanization system established by the Royal Institute of Thailand, the spelling <io> is used to represent the sound [iw]. It isn't intended to reflect the Chinese pronunciation. --Potapt (talk) 14:38, 9 March 2020 (UTC)

Yes! but you should becareful. Juidzi (talk) 13:58, 11 March 2020 (UTC)

About what? --Potapt (talk) 20:46, 11 March 2020 (UTC)

Writing of thai. Juidzi (talk) 07:56, 12 March 2020 (UTC)

Why? Care to elaborate? --Potapt (talk) 13:58, 12 March 2020 (UTC)

I believe the OP is concerned (and rightly so) "io" is a terrible way to represent –ิว in English. The "o" is misleading (there is no sound even close to standard English "o" in this diphthong) and has the potential to confuse the pronunciation with เ–ียว (the "Chinese" sound referred to above). A much better alternative would be "iu" or "iw". In fact, prior to 1999, RTGS used "iu". However, I would point the OP to our Royal Thai General System of Transcription article and note that, despite the name, the Thai government system is more of a transliteration (letter-for-letter) and not meant to be a phonetic transcription. See the Thai language Royal Government publication here. It was changed so that treatment of final "ว" is consistent and always transliterated as "o".--William Thweatt ^TalkContribs 19:09, 12 March 2020 (UTC)

Thanks! William for explain it. I'm very concern, and it have to becareful. Juidzi (talk) 13:34, 15 March 2020 (UTC)

Add some approximations

"ก" can hear a sound like "k" in "sky", and the "ค" group can hear a sound like "k" in "key". Juidzi (talk) 16:10, 24 March 2020 (UTC)

We already have English approximations for those sounds, scan and can, respectively. Adding additional examples is an unnecessary redundancy. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]}

Why do those examples are unnecessary? Juidzi (talk) 02:37, 26 March 2020 (UTC)

So I think the alphabet "ก" & "ค" group're with "k" than "c". And I add those examples, "ก" can hear a sound like "k" in "sky", and the "ค" group can hear a sound like "k" in "key" or "kine". Juidzi (talk) 03:01, 26 March 2020 (UTC)

• 're present with Juidzi (talk) 03:01, 26 March 2020 (UTC)

(Again!) So I think the alphabet "ก" & "ค" group're present with "k" than "c". And I add those examples, "ก" can hear a sound like "k" in "sky", and the "ค" group can hear a sound like "k" in "key" or "kine". Juidzi (talk) 03:02, 26 March 2020 (UTC)

Note for "ฑ"

"ฑ" this alphabet/letter, by an initial consonant is have 2 sounds. 1) “[tʰ]” when it appear in the live syllables. 2) “[d]” when it appear in the dead syllables.

You have to note this, because it's important. Juidzi (talk) 06:48, 26 June 2020 (UTC)

I've replied at Talk:Thai script#About the alphabet/letter "ฑ". --Paul_012 (talk) 15:51, 29 June 2020 (UTC)

Thanks! Juidzi (talk) 01:24, 3 July 2020 (UTC)

Finding some English approximation for “[ɯa]”.

If you need some English approximation, I have choices for you. -Estonian: õa -Turkish: ıa Juidzi (talk) 12:53, 20 December 2021 (UTC)

Those aren't English. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:18, 20 December 2021 (UTC)

Ƶ§œš¹ Why can't we type it? Juidzi (talk) 06:04, 21 December 2021 (UTC)

Okay! I find some English approximation for “[ɯa]”, now. It's "to a (weak sound)". Juidzi (talk) 16:35, 22 December 2021 (UTC)

I search it on Cambrid English dictionary. Juidzi (talk) 16:37, 22 December 2021 (UTC)

Ƶ§œš¹ I see this link https://en.m.wikipedia.org/wiki/Engrish and don't forget to press/click it. Juidzi (talk) — Preceding undated comment added 09:18, 14 January 2022 (UTC)

That seems like an adequate approximation, but I don't think we need to say "weak sound." I don't think it will be clear to readers what that means. I don't even know what it means. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 16:02, 14 January 2022 (UTC)

Ah! for example: Japanese english "you“[ju:]”", they say "ユー“[jɯ:]”". Juidzi (talk) 11:04, 15 January 2022 (UTC)

Missing diphtong ɯːa เือะ

The diphtong ɯːa เือะ is mssing. It is in the vowels letter chart row 6 line 1 which you find everywhere in Thailand 49.228.57.2 (talk) 04:27, 8 April 2023 (UTC)

Of course it should be เ ือะ in Thai script. I had written a circle made from dots beneath the [ɯː] ◌ื, but for unknown reasons it did not reproduce here in WP. I try again: เ ◌ือะ 49.228.57.2 (talk) 11:46, 8 April 2023 (UTC)

The short diphthongs เอียะ, เอือะ and อัวะ are not listed in the IPA manual. They're phonetically the same sounds as their long versions, only followed by a glottal stop. Not sure if they should be mentioned here, as it's not really the purpose of this page (as an IPA guide) to explain the Thai vowel symbols, as opposed to the Thai language and Thai script articles. But I can see how their omission could be confusing. --Paul_012 (talk) 16:01, 8 April 2023 (UTC)

I am irritated now. You talk about short diphthongs, but with mentioning ɯːa เ ือะ I am claiming that a LONG diphtong is missing.

Yes, of course, all vowel combinations should be listed here, for the sake of completeness. I am a beginner’s learner, and I want to know the whole system. What would be the reason for skipping this? Could you name it?

And could you do me a favour and provide the long versions for the short diphthongs เอียะ, เอือะ and อัวะ both in Thai letters and in IPA, just here in the comments. Thank you!

Are there other vowel combinations missing in the list? If yes, which ones would that be? 49.228.57.2 (talk) 04:16, 9 April 2023 (UTC)

You might be confusing the signs. เ◌ือะ is the "short vowel" (diphthong) [ɯaʔ], which is paired with the "long vowel" เ◌ือ [ɯa] in Thai grammars. However, despite the Thai categorization, these "short"/"long" pairs aren't really a phonetic contrasting pair with differing vowel lengths—there are actually no [ɯːa], [iːa] or [uːa] sounds. The actual difference is that the "short" เ◌ือะ always ends with a glottal stop, while เ◌ือ without a closing consonant forms an open syllable. It's the same with the other two, so the pairs are as follows:

"Short"	"Long"
เ◌ียะ [iaʔ]	เ◌ีย(◌) [ia] (not [iːa])
เ◌ือะ [ɯaʔ]	เ◌ือ(◌) [ua] (not [uːa])
◌ัวะ [uaʔ]	◌ัว (or ◌ว◌) [ɯa] (not [ɯːa])

The reason that this page isn't the best place to explain all this is that it's only meant to be a quick-reference guide to the IPA symbols used in Wikipedia. It's not supposed to be part of Wikipedia's encyclopedic content. The content should ideally be covered in the articles about the language and script, in the Main (article) namespace. --Paul_012 (talk) 17:13, 10 April 2023 (UTC)

Thank your for the tables and the clarification. But all textbooks I have seen so far make the difference between short and long vowels, i.e Colloquial Thai by John Moore & Saowalak Rodhchue and many thers I conculted. Why don’t they know better?

Of course we should never rely on their traditional local phonology patterns. If we did like that we had only a handful of diphthongs which is not true (as I had to find out myself) and this was very confusing for me. 49.228.249.14 (talk) 11:39, 19 April 2023 (UTC)

จ is voiced affricate in Thai language

Comparing in https://wiki.riteme.site/wiki/Affricate#Sibilant_affricates

There are consonant shift that have inverted many consonant in modern Thai language. Notably Buddha, which in Thai pronounce as Puttha. The same go for จ and ฉ That was swapped in writing and pronunciation. จ is completely difference than ช and is much more similar to d͡ʑ than t͡ɕ

And the closest sound to that in english pronunciation is something like Jar and Jump ThainaYu (talk) 08:08, 7 April 2024 (UTC)

"More accurate"

@L5boat Please explain why you think your version is "more accurate". "Cut" is an appalling bad approximation of [ä] for speakers without the FOOT-STRUT split. [ɛ] is closer to the DRESS vowel that the TRAP vowel in the overwhelming majority of English accents. "Story",i.e. the FORCE vowel is a good approximation of [o] in the overwhelming majority of accents of English, and it is the same vowel as "boat" and "go" in Scottish English with the very important advantage of applying to speakers of all accents of English, not just Scottish English. [ɔ] is closer to the vowel of CLOTH than THOUGHT in the accents that distinguish them, most of which have THOUGHT closer to [oː] than [ɔː]. GOOSE-fronting is a very well attested and described change occurring in many, many, many dialects of English, but it generally doesn't happen before /l/, making a pre-/l/ example the best way to illustrate [u]. Offa29 (talk) 11:07, 13 November 2024 (UTC)

User: Offa29 Why did you write this post in the Thai language page rather than the Telugu page? అ is the first letter of the Telugu alphabet. It represents a schwa sound (literally the first thing taught to kids in kindergarten). There is no "æ" sound in the Telugu language, like the vowel in "cat." L5boat (talk) 16:58, 13 November 2024 (UTC)

@Offa29: You seem to be changing a large number of these help pages. I don't think you should be doing this. I think these help pages should be edited by editors who have some familiarity with the languages in question. I think you fundamentally misunderstand how IPA vowel symbols work. The sound represented by [ɛ] in language A may not be the same as the sound represented by [ɛ] in language B. This is why the vowel quadrilateral exists. This is all explained in the IPA handbook. Your changes do not help people to pronounce Thai words correctly (which is the purpose of this page). For example, while it is true that "Cut" is an appalling bad approximation of [ä] for speakers without the FOOT-STRUT split, in what variety of English does the vowel in cat sound like the vowel in คำ [kʰām]? (I notice you are involved in a similar dispute with L5boat regarding Help:IPA/Telugu.)

@Nardog, IvanScrooge98, Paul 012, Potapt, and WilliamThweatt: Can you offer any advice in relation to this matter? Khiikiat (talk) 14:56, 13 November 2024 (UTC)

This is a problem that resurfaces every now and then on these help pages. There’s a reason these are called “approximations”: the English language does not feature all the sounds of the world’s languages. Best we can do in my opinion is use the approximations that are closer for the largest possible number of English speakers and/or mark a specific dialect for the approximation which we’re gonna use. Now, I’m not an expert of Thai phonology so I’m not sure which English vowel best describes Thai /a/. ~ IvanScrooge98 (talk) 15:05, 13 November 2024 (UTC)

I am well aware of IPA symbols work. I am particularly well aware of how the IPA symbols typically used for English differ often differ wildly from the "canonical" IPA vowel chart and other languages using the same symbols. You know nothing about me and my level of knowledge and understanding of anything. Don't make assumptions, don't be condescending. Do I have familiarity with Thai? No. But for drawing up a list of English examples, it is also important to have familiarity with the many, wildly varied in their pronunciation of vowels, accents of English.

I'll park the discussion of [ä] and [ɛ] for the time being and move onto more cut-and-dried, in my opinion, examples, which you have so far not responded to.

1. On what basis do you claim that "moon" and "shoot" are more accurate srepresentation of Thai [u(ː)] then "cool"? It is well-established that the GOOSE vowel in English is considerably more fronted than [uː] in many, many dialects of English, including modern Received Pronunciation and many modern varieties of General American, and shifting ever more fronted in many cases. The exception to this is when the vowel is followed by /l/, due to the backing effect. With accents that don't have GOOSE-fronting, "cool" has the same vowel as "shoot" and "moon", so the example is just as good. In accents that do have it, "cool" is a better example. On what basis do you claim otherwise?
2. On what basis is an example that only works in Scottish English better for [o] than "story", a word that has the same vowel as "boat" and "go" in Scottish English anyway, and a quality close to [o] in the vast majority of accents of English, unlike "boat" and "go"? The overwhelming majority of English speakers from outside Scotland will not be linguistically aware enough to be able to visualize the vowel specifically in Scottish English and how it differs from their own. "Giving an example that only applies in Scottish English is not helpful to speakers of any other English dialect.
3. Giving "shoot" for [u] and "goose (Australian)" for [ɯ]. Those words have the same vowel in Australian (and all other varities of) English. What on Earth is a speaker of Australian English supposed to think the difference between [u] and [ɯ] is? And again, giving an example that only applies in Australian English is not helpful to speakers of any other English dialect (in any case, many other varities of English have a GOOSE vowel quality similar to that of Australian English).
4. "Lawn" for [ɔː]. Contrary to the IPA symbols typically used in the transcription of English, the THOUGHT vowel is typically closer to [oː] than [ɔː] in accents that distinguish it from CLOTH (e.g. modern RP, most other forms of English English, Australian and New Zealand English). In accents that have CLOTH-THOUGHT merged (e.g. all North American accents), it is generally closer to [ɔ], but a CLOTH word would work just as well for those accents anyway, and better for most other (i.e. most non-North American) accents. On what basis do you claim otherwise?
5. Giving "boy (Australian) " for [oːj] and "boy" for [ɔːj]. What on Earth is a speaker of Australian English supposed to think the difference between [oːj] and [ɔːj] is? Not to mention that there are many other dialects of English, such as modern RP and most forms of Southern England English, and New Zealand English, with a vowel in "boy" similar to that of Australian English, closer to [oːj] than [ɔːj], so it is not at all helpful to speakers of these dialects either. Surely we should either specify dialect(s) for [ɔːj] as well as [oːj] (and probably note which other dialects are closer to [oːj]), or just give "boy" as the example for both on the grounds that it is the closest available vowel to both in all dialects of English.

There are more points I could make about other examples, but in the interest of keep the discussion down to a manageable number of topics at any one time, I will leave it at these five points for now. Offa29 (talk) 22:18, 14 November 2024 (UTC)

I agree that the English approximations could be improved. I tried to improve them with this edit, but you immediately reverted it. Your comments above are not about the edit that I made. Khiikiat (talk) 03:01, 15 November 2024 (UTC)

Because that edit was not the version on the article at the time of writing the comment, as you reverted it back to the older version. But I'll comment on them now. Some of the changes in your edit were I agree were better than the original examples, but some were worse in my opinion.

• I agree that "come (RP)" is preferable to "cut" for [ä] since it is not misleading to speakers without the FOOT-STRUT split. I still question whether it is the best possible example we could give, and plan to resume that discussion later, but since we agree that it is better than the old version, I will reinstate it onto the article for now.
• "Bed" for [e] hinges on what we do for [ɛ], so I will withold discussion of it for now, beyond the observation that there is not an ideal approximation for [e] in most accents of English.
• Re: "bin" versus "happy". "Happy" is problematic because of its variation in accents without happY-tensing, some of which have it as open as [ɛ], though the large majority of accents of English now have happY-tensing. On balance, I think "bin" is probably better, even though the quality is a bit different from [i] in most accents of English, so I'll reinstate it.
• "Bomb" for [o] I do not agree with. It is particularly ill-suited for North American English, where the LOT vowel is typically an unrounded [ɑ] even a fronted [ä], verging on [a] in accents with the Northern Cities Vowel Shift. North American accents with the Low Back Merger Shift may have a rounded LOT vowel in the [ɒ]~[ɔ] range, but this is still considerably more open than [o]. Irish English (outside Ulster) also typically has an unrounded LOT vowel. In most other accents outside North America the LOT vowel is rounded, typically around [ɔ], though some accents (e.g. Northern England) may have a more open [ɒ]. Even the [ɔ] typical of modern RP, Australian and New Zealand is still pretty far from [o], and those accents have other vowels which would approximate [o] more closely.
• "Bought" for [ɔ] I disagree with. It is better for than "lot" for some accents (e.g. American English without the cot-caught merger), but worse for others (e.g. modern RP/Southern England, Australian, New Zealand English). I neglected to do this in my initial edit, but the best solution for [ɔ] is to use a word from the CLOTH lexical set, such as "off". This effectively means "the LOT vowel for most non-North American accents, the THOUGHT vowel for American (and some Irish) accents", which is the best solution for both types of accents, neatly presented in a single word. The majority of IPA Help pages with an [ɔ] phoneme already do this, and have done so for some time.
• I agree that not having any example for [ɯ] is better than "goose (Australian)" since that is just confusing and the Australian "goose" vowel is, in any case, rounded, fronted, and diphthongal. I think we should try to give some sort of example where possible, but that example has to be actually helpful and this wasn't. I will reinstate the blank cell of your edit.
• I disagree with just "go" replacing "go (Scottish English)" for [oː]. Having an example that applies only to Scottish English is obviously not ideal, but it is at least accurate. In most other accents of English, the GOAT vowel is not [o], and it is very often nowhere near [o]. It can range from [əʊ] (RP and Midland American English) to [əʉ] (Australian English) to [əʏ̯]~[ɛʏ̯] (Southern American English) to [äʉ̯] (New Zealand English) to [ɛʉ̯] (Scouse, East Midlands English) to [ɵː]~[øː] (Hull/East Yorkshire English), and more. The range of possible variations in the realization of the GOAT vowel in English is so wide that it should, in my opinion, never be used as an example for anything without a caveat specifying which dialects' GOAT vowel is being used as the example. The large majority of accents of English have a FORCE vowel close to [o], so a FORCE word like "story" is by far the best approximation. In Scottish English, the FORCE vowel is just the GOAT vowel preceding /r/, so it works just as well for Scottish English.
• I agree with not using "house (Australian" and "cow (Australian)" as the examples for [ɛw] and [ɛːw], since it is not a close approximation. Some dialects of English have a GOAT-vowel that may be suitable, though I admit it is not ideal even in most of these dialects so I will save discussion of this for now. I will reinstate the blank cell from your edit.
• I agree that having just "boy" for both [ɔːj] and [oːj] is better than the confusing "boy (Australian)" versus just "boy", so I'll reinstate it for now.
• I agree with having a blank cell for [ɤːj], which was what I had in my edit. "millefeuille" has three different pronunciations in British English. There is no good English approximation for [ɤːj] that I can think of.

There's a few more of your edits I didn't get round to discussing but I don't have any more time now so I'll save them for later.

I will pose two questions for now:

1. Why do you oppose "story" for [o(ː)]?
2. Why do you oppose "cool" for [u(ː)]?

Offa29 (talk) 12:56, 15 November 2024 (UTC)

1. I am opposed to story for [o(ː)] because I think it will lead to a great deal of confusion. Story is commonly transcribed with [ɔː]. Some examples: oed.com, oxfordlearnersdictionaries.com, dictionary.cambridge.org, collinsdictionary.com. I simply do not think that the vowel in story sounds like the vowel in ลม [lōm] or the vowel in โลก [lôːk]. In Thai: An Essential Grammar, David A. Smyth gives Ron as an English approximation for the Thai vowel represented by [o] and go as an English approximation for the Thai vowel represented by [oː].
2. I am opposed to cool for [u(ː)] because many speakers of English have l-vocalization. Smyth has book for [u] and coo for [uː].

Khiikiat (talk) 13:48, 16 November 2024 (UTC)

1. As you have said yourself, the IPA symbols conventionally used to transcribe one language do not necessarily match up with other languages using the same symbols. You yourself have argued against using bet for [ɛ] on these grounds, so I am surprised to see you making this argument now. Our aim is to give examples that accurately reflect pronunciation, not that reflect conventional IPA transcriptions (the conventional IPA transcription for British Received Pronunciation is widely acknowledged to be outdated, see the writings of Geoff Lindsey, one of his proposed symbol changes is /oː/ for the vowel traditionally transcribed as /ɔː/. Is it your view that the pronunciation samples of story here are not close to Thai [o(ː)]?
2. Compare the UK recordings of cool to shoot. Which sounds more like Thai [u(ː)]? (The US recordings don't have much difference, but many other dialects of English, including many US dialects, having GOOSE-fronting as well, and even for the US recording example "cool" would be no worse than "shoot"). The effect of L-vocalization on the preceding [uː] vowel generally isn't too great in my experience; its main effect is on the /l/ itself.

The two questions I posed here are genuine, not rhetorical. Offa29 (talk) 23:39, 18 November 2024 (UTC)

1. Is it your view that the pronunciation samples of story here are not close to Thai [o(ː)]? Yes, that is my view. Thai [o] sounds a bit like RP long, Thai [oː] sounds a bit like RP low, and Thai [ɔː] sounds a bit like RP law. Perhaps story could be used for Thai [ɔ]. (It is worth noting that Thai people themselves perceive Thai [oː] to be the Thai equivalent of the GOAT vowel. For example, the Thai Wikipedia article about Soho is entitled โซโฮ [sōː.hōː].)
2. Which sounds more like Thai [u(ː)]? I don't think there is a great deal of difference between cool and shoot in the recordings. I would use a word like zoo or who for Thai [uː]. I think the presence of the l complicates things unnecessarily. No doubt there are some varieties of English in which there is less GOOSE-fronting in cool than there is in shoot, but there are also varieties of English that have a cool–call merger. With regard to Thai [u], I think I would follow Smyth and use book.

Khiikiat (talk) 23:32, 19 November 2024 (UTC)