View source for Template talk:IPAc-en

{{hatnote|For the discussion of the choice of symbols and the organization of diaphonemes, see [[Help:IPA/English]] and [[Help talk:IPA/English|its talk page]].}}
{{central|Module talk:IPAc-en}}
{{Permanently protected}}
{{talk header}}
{{FAQ|page=Help talk:IPA/English/FAQ}}
{{User:ClueBot III/ArchiveThis
|archiveprefix=Template talk:IPAc-en/Archive
|format= %%i
|age=4320
|maxarchsize=100000
|numberstart=1
|archivebox=yes
|box-advert=yes
}}

== Could we add a break option? ==

Occasionally we want a long transcription in a table or infobox with limited width. Currently in infoboxes it just goes off the right edge of the page. Could we have a break option, maybe BR instead of an underscore between words? (And presumably ,BR for ,_) [[User:Kwamikagami|— kwami]] ([[User talk:Kwamikagami|talk]]) 08:36, 15 May 2026 (UTC)

== Technical Request: Allow standalone "ɑ" as a valid diaphoneme input ==

It appears that short ɑ is presently blocked from appearing before a nasal, yet the sequence exists in U.S. English. Can this be repaired/enabled? [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 08:55, 23 May 2026 (UTC)

:There is no short {{IPA|/ɑ/}} in this diaphoneme scheme, so it's not totally clear what you mean. Could you give an example of something that's not currently possible that you want to do? — [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 13:07, 23 May 2026 (UTC)
::You state the problem well: no short /ɑ/ in the diaphoneme scheme. It's not currently possible to report the /ɑ/ of the stressed syllable of the pronunciation of ''[[chimichanga]]'', for example, in the diaphonemic representation. It's presently reported as /æ/ in the Chimichanga article, which is false phonemically and allophonically, misrepresenting the U.S. pronunciation (as would also reporting vowel length as in /ɑː/ at any level). (I haven't tried to correct the second vowel of ''chimichanga'', presently given as /ɪ/ rather than the normal short /i/; a less infelicitous representation, as [ɪ] can appear in fast speech.). I'm perfectly willing to use standard phonemics to read /ˌtʃɪmiˈtʃɑŋɡə/ in the Chimichanga article, but I fear that could ignite an unpleasant session of bickering. [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 15:52, 23 May 2026 (UTC)
:::I've added this pronunciation to the relevant page. [[User:Wolfdog|Wolfdog]] ([[User talk:Wolfdog|talk]]) 18:25, 23 May 2026 (UTC)
::::Appreciate the effort, seriously. But it still shows the stressed vowel as long, not /ˌtʃɪmiˈtʃɑŋɡə/. It's probably good to mark that, once it's achieved, as US, or perhaps North American English (i.e. including Canada), but where is the version with /æ/ normal? [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 19:52, 23 May 2026 (UTC)
:::::We use a [[Help:IPA/English|diaphonemic system on WP]] to broadly incorporate both American and British English, as much as phonemic overlap is possible. So, readers are meant to read "/ɑː/" as: /ɑː/ in (southern) British English and /ɑ/ in American English. This is also explained at [[WP:RHOTIC]]. I'm not aware of where /æ/ is the normal pronunciation. It sounds British to me, but also probably various people in the US use it too (see the chart [[American_and_British_English_pronunciation_differences#Single_differences|here]]). [[User:Wolfdog|Wolfdog]] ([[User talk:Wolfdog|talk]]) 20:13, 23 May 2026 (UTC)
:::::Oh, OK. There's a few things to unpack here:
:::::# This isn't a technical request; you seem to be asking to change the diaphoneme scheme.
:::::# You don't understand how [[diaphoneme]]s work (sorry to be blunt - I can see how it's confusing). Wolfdog explained above. It does seem kind of backwards to transcribe a US pronunciation using vowel length, which only applies to a {{em|UK}} pronunciation, but that's just how the system works, because it "would be" long in British English.
:::::# {{tqbm|where is the version with /æ/ normal?}} Have you checked sources? OED shows the pronunciation with /æ/ as British. I added a citation to the article.
:::::# "blocked from appearing before a nasal" is a red herring: if it doesn't exist, then it can't be "blocked".
:::::BTW I'm Canadian and I pronounce it the same way: {{IPA|/ˌtʃɪmiˈtʃɑŋɡə/}}
:::::— [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 20:44, 23 May 2026 (UTC)
::::::The lack of the length mark does not equal a phonetically short vowel, not even in a phonetic transcription enclosed in brackets (and this is a diaphonemic transcription, two levels of abstraction above that). A few people here seem to be confused about that. {{IPA|/ɑ/}} is often rather long in North America, ditto for {{IPA|/æ/}}. [[User:Sol505000|Sol505000]] ([[User talk:Sol505000|talk]]) 12:10, 24 May 2026 (UTC)
:::::::@[[User:Sol505000|Sol505000]] Sorry, what does the phonetic value have to do with this discussion? North American English doesn't distinguish vowels by length. — [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 13:18, 24 May 2026 (UTC)
::::::::No it doesn't, but we shouldn't be mixing up phones with phonemes. {{IPA|/ɑː/}} just means "whatever the vowel in "spa" is" in our transcriptions. {{tq|It does seem kind of backwards to transcribe a US pronunciation using vowel length, which only applies to a {{em|UK}} pronunciation, but that's just how the system works, because it "would be" long in British English.}} tells me you're also confused about this. The lack of phonemic vowel length in a language/dialect doesn't mean that the vowels are ''phonetically'' short. The short {{IPAblink|ɑ}} is definitely not the most common realization in North America, it appears mostly when the vowel is clipped by the following fortes or unstressed syllables (as in ''chimichanga'' {{IPA|/-ɡə/}}, but it's still not "perfectly short" as in Spanish), or just fast speech tempo (which clips everything). [[User:Sol505000|Sol505000]] ([[User talk:Sol505000|talk]]) 13:37, 24 May 2026 (UTC)
:::::::::I never said "phonetically short". — [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 14:18, 24 May 2026 (UTC)
::::::First, Wolfdog. Again, thanks. Yes, the glitch lies within the diaphonemic system when its strict application misrepresents or complicates (depending on point of view) what should be as straightforward as possible for a wide readership. Standard phonology of phonemes and (allo)phones is already challenging enough, without introducing the need to guess or somehow discover that /ɑː/ corresponds to /ɑ/ in the U.S.
::::::W.andrea, "{{tq|This isn't a technical request; you seem to be asking to change the diaphoneme scheme.}}" Technical request to "fix what's broke". I do understand to how diaphonemes work, and I can understand the attraction when they do work. But what does "work" mean -- elucidate, clarify? If so, when they don't work, as apparently in this case in which the genuine native-speaker short vowel ''cannot be displayed'' for readers, i.e. the representation obscures rather than clarifies, an adjustment seems to be in order. I appreciate the Pythonesque humor of '{{tq|"blocked from appearing before a nasal" is a red herring: if it doesn't exist, then it can't be "blocked"}}', but it's no help to the real-world quandary that it describes.
::::::If it's a mortal sin to adjust the diaphonemic scheme to allow the short vowel where it exists, perhaps a venial sin can be pardoned.
::::::UK pronunciation is now specified, and it's apparently accurate whether construed as phonemic (as the convention / / likely suggests to readers with Intro to Linguistics under their belt), or diaphonemic (which fewer readers will understand) -- welcome coincidence. That leaves a diaphonemic form for what is now the first transcription much less necessary (if necessary at all). Very simple: mark what is now the first transcription as US, and replace what's there with straightforward phonemic /ˌtʃɪmiˈtʃɑŋɡə/. Voilà. The vowel that very much does exist in your and my and millions of others' /ˌtʃɪmiˈtʃɑŋɡə/ is there, all readers know which is US and which is UK, and non-expert readers with a bit of IPA are informed clearly and accurately, without having to grapple with the long vowel that describes neither the US /ɑ/ nor the UK /æ/. In the words of Google's AI, "Wikipedia editors now use the generic {{IPA}} template for custom or narrow strings that do not fit the strict IPAc-en diaphoneme rules." [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 15:55, 24 May 2026 (UTC)
:::::::{{IPA|/ɑ/}} does not mean "not phonetically long" (just as {{IPA|/ɑː/}} doesn't necessarily mean "not phonetically short". Also, our {{IPA|/i/}} doesn't mean "always/typically short" but "{{sc2|KIT}} or {{sc2|FLEECE}}", so that's another argument against your proposal. Millions of Brits say {{IPA|[ˈhæpɪi]}}, ditto for South African/Welsh {{IPA|[ˈhæpiː]}}) and the vowel is clearly long(-ish) most of the time in the US and Canada. It's the outcome of the father-bother merger. The vowel ({{IPA|/æ/}} in this case) is also clipped in British English due to the following unstressed syllable (called "rhythmic clipping" by [[John C. Wells]], who gives the triplet ''man'' {{IPA|[mæːn]}} vs. ''manage'' {{IPA|[ˈmæˑnɪdʒ]}} vs. ''management'' {{IPA|[ˈmænɪdʒmənt]}} as an example of the {{sc2|TRAP}} vowel that gets progressively shorter). Maybe not as much as in the US, but still. <s>Also, English bans {{IPA|/ɑː/}} from ever occuring before {{IPA|/ŋ/}} in native words, so there isn't really anything to compare that word to.</s> It is ''not'' a different phoneme to the one in ''spa'', let alone a different diaphoneme. You're introducing a lot of confusion here by mixing up phonetic vowel length with the application of the length mark in (dia)phonemic transcription. I also find it rather dubious that the length of {{IPA|/ɑ/}} in "Chimichanga" matches that of {{IPA|/a/}} in the Spanish pronunciation of the word. It's almost certainly somewhat longer. [[User:Sol505000|Sol505000]] ([[User talk:Sol505000|talk]]) 16:38, 24 May 2026 (UTC)
::::::::The point is clarity for readers. This wins the prize for illustrating lack of clarity, though it might be comforting to non-native speakers struggling to get their /i/s and /ɪ/s right: "our /i/ doesn't mean "always/typically short" but "<u>KIT or FLEECE</u>". For the item in question, ''chimichanga'', we have "English bans /ɑː/ from ever occuring before /ŋ/ in native words", yet we have a transcription system that demands representing /ɑː/ occuring before /ŋ/. Clarity for readers? The point of creating an encyclopedia of use to any and all readers in the first place? This would come much nearer that goal than the present version: "Wikipedia editors now use the generic template for custom or narrow strings that do not fit the strict IPAc-en diaphoneme rules." [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 18:17, 24 May 2026 (UTC)
:::::::::Is ''chimichanga'' a native word? Exactly. [[User:Sol505000|Sol505000]] ([[User talk:Sol505000|talk]]) 18:29, 24 May 2026 (UTC)
::::::::::Nativized, yes, especially in the Southwest, but elsewhere as well. True enough that you'll find hordes of certified gringos ordering [tʃimiˈtʃaŋɡa]s in the U.S., again in the Southwest especially, where Borderland Asymmetric Anglo-Bilinguals are thick on the ground. But you'll also find many anglos ordering the nativized [ˌtʃɪmiˈtʃɑŋɡə] in the same restaurants, perhaps at the same tables. Depending on the company, even the same folks who sometimes order a [tʃimiˈtʃaŋɡa]. [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 23:38, 24 May 2026 (UTC)
:::::::{{tqbm|In the words of Google's AI, "Wikipedia editors now use the generic  template for custom or narrow strings that do not fit the strict IPAc-en diaphoneme rules."}} I'm not sure how that adds to your point, but it is correct: you can use the generic template {{tl|IPA}}, e.g.
:::::::{{demo-inline|<nowiki>{{IPA|/ˌtʃɪmiˈtʃɑŋɡə/}}</nowiki>}} (which is tagged correctly as IPA in the HTML)
:::::::However, that level of precision isn't required here, plus it doesn't link the key for more info ([[Help:IPA/English]]) and doesn't include the hover text {{tq|'a' as in 'father'}}.
:::::::It's important to remember that [[WP:Wikipedia is not a dictionary|Wikipedia is not a dictionary]], so for articles that aren't about linguistics, [[MOS:PRON|pronunciations]] are given mainly for convenience (at least as I see it). The source (OED) has a more precise transcription, without the short marker.
:::::::{{tqbm|But what does "work" mean -- elucidate, clarify?}} I mean that they're abstract. A phonemic system basically applies to only one variety, so it can be precise, but a diaphonemic system specifically applies to {{em|multiple}} varieties, so it has to make compromises, and that's what's happening here.
:::::::Lastly:
:::::::{{tqbm|Technical request to "fix what's broke".}} I interpreted "technical request" as "change a minor detail" but you're asking to overhaul the transcription system for this diaphoneme.
:::::::— [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 17:47, 24 May 2026 (UTC)
::::::::Just to mention, another option for the generic template is:
::::::::{{demo-inline|<nowiki>{{IPA|en-US|generic=yes|link=yes|ˌtʃɪmiˈtʃɑŋɡə}}</nowiki>}}
::::::::— [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 17:58, 24 May 2026 (UTC)
:::::::::Thanks very much for putting all this together. I've just been called away, but I'll see what I can do with the chimichanga article asap. Here's hoping a straightforward phonemic rendition of the US version isn't shot down immediately for not being diaphonemic. Cheers, and thanks again. [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 18:41, 24 May 2026 (UTC)
::::::::::No, I'm saying don't do that, even though it's possible. — [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 19:36, 24 May 2026 (UTC)
:::::::::::So just leave it with no guide to the US pronunciation. Good grief. [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 20:55, 24 May 2026 (UTC)
::::::::::::It already has the US pronunciation, just in diaphonemic form. — [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 21:33, 24 May 2026 (UTC)
:::::::::::::Exactly...the vowel length marker of which misleads the reader who might know some basic phonology but hasn't delved into or even heard of diaphonemics (and with the / / phonemic convention present, said reader has no way to suspect what's up other than basic structural phonemics). [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 23:44, 24 May 2026 (UTC)
::::::::::::::It's only going to confuse ESL readers, because native speakers should recognize it as the {{sc2|spa}} vowel and pronounce it however they normally do. And even then, lengthening isn't even phone'''t'''ically wrong, as Sol explained. Yes it's phone'''m'''ically wrong, but hopefully Ling 101-level folks should know that you need to understand the phoneme scheme to understand any particular symbol, and that's what the link and hovertext are for. Suffice to say, yes it's not perfect, but it's as good as we can do while still being consistent. — [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 18:34, 26 May 2026 (UTC)
::::::::::::Not to fuel any fires, but I think we're speaking to a larger recurring issue which is: "what is the lesser [least] of several evils"? IPA will confuse plenty of people. Within that: a phonemic system will confuse plenty of people. A diaphonemic system will confuse plenty of people. The respell system will confuse plenty of people. Etc. This issue comes up again and again. For every conceivable transcription system, editors have expressed (sometimes righteous) bafflement and will only continue to do so. (I wonder how many times I've battled with Brits about our diaphonemic /ər/, with them wondering how a Brit *should just assume* to interpret that as nonrhotic /ə/. Their annoyance is not unfounded.) However, whatever systems we have now are based on a general community consensus, I suppose (which however can always change). Not sure what more there is to decide here. (And as W.andrea says, if one insists on workarounds, they do exist.) Thanks all, anyway, for keeping it collegial. [[User:Wolfdog|Wolfdog]] ([[User talk:Wolfdog|talk]]) 21:59, 24 May 2026 (UTC)
:::::::::::::I think that {{angbr IPA|ɑː}} is far more commonly used in reputable sources for pronunciation that use the IPA (LPD, CEPD, you name it) than {{angbr IPA|ɑ}}. Both mean the same ''phoneme'' and that's what we should focus on. Phonetic vowel length is strongly influenced by what follows the vowel in ''both'' the US and the UK. Here, the difference is clearly phonemic, but not in the way described by the OP. In the US, the word in question uses the {{sc2|SPA}} vowel in the stressed syllable, whereas in the UK it has the {{sc2|TRAP}} vowel, presumably due to the lack of pre-nasal [[ae tensing]]. It's a common and predictable pattern described as early as 1990 by Dr [[Geoff Lindsey]]. {{angbr IPA|ɑ}} for the diaphoneme has the massive disadvantage of looking almost like {{angbr IPA|ɒ}}, and almost no modern source uses it for IPA transcriptions of British English. I think disposing of the length mark is out of question here. [[User:Sol505000|Sol505000]] ([[User talk:Sol505000|talk]]) 22:04, 24 May 2026 (UTC)
::::::::::::::That's absolutely tangential, but I don't think nasalisation is decisive here. That's how the ''foreign'' or ''Italian A'' gets redistributed in other words too, e.g. ''mafia, pasta, taco''. – <i style="text-transform:lowercase">MwGamera</i> ([[User talk:MwGamera|talk]]) 14:31, 25 May 2026 (UTC)
:::::::::::::::I know, but it does play a role. {{IPA|[ɛə]}} or {{IPA|[eɪ]}} (before {{IPA|/ŋ/}}) sounds nothing like {{IPA|[a]}} or {{IPA|[ɑ]}}. The only connection between these is the letter used to write them in orthography. In the UK, there's no parallel problem as the first two diphthongs would be interpreted as {{sc2|SQUARE}} and {{sc2|FACE}}, respectively, rather than {{IPA|/æ/}}-the-phoneme. [[User:Sol505000|Sol505000]] ([[User talk:Sol505000|talk]]) 14:37, 25 May 2026 (UTC)
:::::::::::::Wolfdog, you're a scholar and a gentleman, and you speak much sooth. I'm trying to do just one thing: keep (inevitably complex) matters as clear and straightforward as possible, i.e. add no unnecessary complexity. (You're right that W.andrea says that if one insists on workarounds, they do exist. Note that he also says don't do it, exactly what I suspected would happen if I did edit the chimichanga article to give the genuine U.S. English phonemic structure.) [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 23:54, 24 May 2026 (UTC)
:I think I finally know what the OP means. The primarily stressed vowel in ''chimichanga'' can sometimes get shortened if the speaker aims at a pronunciation that's closer to Spanish. I heard more or less the same thing done to the name "Jan", pronounced {{IPA|[jɑn]}} when the speaker repeated the name after the person, and after that just said {{IPA|[jɑːn]}}, with the expected long vowel. I wonder if we can get [[WP:RS]]'s to describe this at [[General American]] and/or [[English phonology]], because it's definitely interesting. The question is: does anybody keep those unusually short vowels consistently? I think we'd need a [[WP:RS]] for that, because it seems to me that this is just a mixed English-Spanish pronunciation. You can get all sorts of mixing with L2 speakers of English whose native language is Spanish, for example. [[User:Sol505000|Sol505000]] ([[User talk:Sol505000|talk]]) 13:23, 26 May 2026 (UTC)
::Can't swear to this under oath, but I'm reasonably certain that my AmE version of ''chimichanga'' has no lengthening of the stressed vowel. It ''can'' have length, it seems, if it's hyperstressed, "No, no. I want a CHIMICHANGA, not a plate of tacos!", but that can also have short vowel. And that can be true of "Tom" (short) vs. "That's not Bill! That's his twin brother TOM!" (can be long or short), etc. Ditto with the /i/ of more firmly nativized ''pizza'', it seems; it can be lengthened, but [i] sounds more normal to me. Again, no oath, but even though I never bet on anything, I'd be tempted to wager a nice sum that the short vowel is the norm in these, and that it can "get lengthened" occasionally ("Let's get a p[i]zza." A P[i:]ZZA? We had p[i]zza just yesterday"). An interesting case is that of monolingual AmE speakers trying to emulate an Italian's string of stressed vowel before a geminate consonant. The Italian native speaker's stressed vowel is invariably short in the closed syllable. Something of interest (exotically foreign?) is apparently heard (and misperceived) by the anglophone, with results such as the infamous ['a:ni] rather than ['anni] in giving age: ''ho trent' ''['a:ni] 'I have thirty anuses' rather than ['anni] 'years'. -- But none of this great fun dissolves the problem of the poor /ˌtʃɪmiˈtʃɑŋɡə/ in Wikipedia, alas. [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 19:37, 26 May 2026 (UTC)
:# On the surface level, the issue being described here, as W.andrea, Wolfdog, and Sol505000 have all mentioned, is with the interpretation of diaphonemes. From a diaphonemic standpoint, the length mark in {{IPA|/ɑː/}} is just a symbol, not a definition of the realized length of the sound by all speakers. Even within phonemic transcription, that is somewhat the case.
:# On a lower level, it ought to be recognized that this template has specific parameters to suggest interpreting from a phonemic position, not a diaphonemic one; or at least, from a highly abstracted position, to a less abstracted one. As soon as a regional parameter is applied, such as in {{tlx|IPAc-en|US}}, we are no longer suggesting the same diaphonemic scheme as without a parameter. I think that there is a point to be argued that adding a regional parameter is still using ''a'' diaphonemic scheme, but a regionalized one rather than a universal one. Accordingly, then, we have several levels of abstraction here: English diaphoneme > regional diaphoneme > varietal phoneme > realized phone.
:# If we recognize this difference in abstraction that the regional parameter suggests, then I actually somewhat agree with Barefoot through the chollas here. If the template is provided as {{tlx|IPAc-en|US}}, then why not make it display the {{IPA|/ɑː/}} diaphoneme without the length mark? Since, as has already been recognized in this conversation, AmE has no phonemic vowel length distinction, and it is widely acknowledged that the AmE (dia)phoneme is therefore transcribed as {{IPA|/ɑ/}}, without a length marker, as are all AmE (dia)phonemes which correspond to long vowels in other regional groupings of English.
:# I think there is further argument for this regional switching given that, for reasons that frankly dumbfound me, we include a vowel in the diaphonemic scheme that arguably violates the principle of what a diaphoneme is! We define {{IPA|/ɜː/}} as occurring only in some dialects, and have a written requirement that it must be denoted with a regional parameter because of this, yet for whatever reason don't enforce this on the template level. Realistically, given the written requirements we ask for in its usage, {{tlx|IPAc-en|ɜː}} without any regional parameter should result in an error, perhaps even if the US parameter is included (though some AmE speakers might still use this vowel in some overlapping circumstances). We treat the system as being dialectally neutral, yet provide a parameter for a sound which we explicitly acknowledge is in fact not.
:# To the point of regional handling of both {{IPA|/ɑ(ː)/}} and {{IPA|/ɜː/}}, yet another issue arises: for further reasons that dumbfound me, this template treats UK and US English with some kind of exclusivity. From an ENGVAR standpoint, we recognize 17 distinct regional varieties of written English (per [[Module:Engvar/doc#List of English varieties]], and not including Oxford & IUPAC), many of which also correspond to distinct phonological conventions. Why does this template allow for special treatment of only two of those varieties? I recognize the technical burden of supporting all of them, which would undoubtedly be compounded several-fold if we provided such an option for display switching based on region, but the point here is the internal inconsistency.
:To summarize my points:
:* We treat usage of this template as if it should always be used within a neutral diaphonemic system,
:* except for where we provide the ''option'' to diverge from that system to a regional level,
:* except for where we include sounds within the system that theoretically are ''required'' to be denoted with a region,
:* except for the fact that we don't even include the majority of regional varieties that we recognize within ENGVAR.
:It is no wonder that editors get confused with how this template should properly be used, as it contradicts itself on multiple levels. It is always diaphonemic, except for all of the times that it isn't. ~ [[User:Oklopfer|<span style="color: #e3308a">oklopfer</span>]] ([[User talk:Oklopfer|💬]]) 00:11, 27 May 2026 (UTC)
::Very well said, indeed. Thank you. [[User:Barefoot through the chollas|Barefoot through the chollas]] ([[User talk:Barefoot through the chollas|talk]]) 17:03, 27 May 2026 (UTC)
::{{tqbm|Why does this template allow for special treatment of only two of those varieties?}} Because GenAm and RP are the most widely-spoken and widely-recognized. I'm Canadian myself and I would say we really don't need to include Canadian because most Canadian accents are more similar to GenAm than some other American accents are (e.g. [[Texas drawl]], [[Bawlmer]], Bernie Sanders).
::As I said above, {{strong|you can get more precise if you need to}}, but it's just not necessary here since the article isn't about linguistics. Wikipedia isn't for teaching people to speak English, and it's not a dictionary. Pronunciations (on non-linguistics articles at least) are provided mainly for convenience.
::That said, great point about "several levels of abstraction".
::— [[User:W.andrea|W.andrea]] ([[User talk:W.andrea|talk]]) 13:49, 31 August 2026 (UTC)