Talk:English phonology

Vowel phoneme chart for General American

Currently, the vowel phoneme tables show RP as having central vowels for GOOSE, FOOT and a back vowel for START, but GA as having back vowels for GOOSE, FOOT and a central vowel for START. I question whether this is a real distinction: while the position of START in American English varies, I think the Northern Cities Vowel Shift is generally not considered to be default "General American". Would a front vs. non-front setup work better? I also am not sure ɑ ɜ ɔ fit well into the lax/tense binary, given that ɑ functions as the lax counterpart of oʊ in the context of correspondences like "trisyllabic laxing". Does anyone else agree that the GA table should be revised? Urszag (talk) 17:36, 11 January 2024 (UTC)

/x/ and glottal stop /ʔ/

I added the glottal stop /ʔ/ to the chart, but I also added /x/ to the chart in this article. While the glottal stop /ʔ/ and /x/ are not standard phonemes, I decided to be like a comprehensive linguist and added both to both articles. If you want to remove either one or both from either this article, this one, or both, pleade feel free to do so. But if we/these can't stop these marginal phonemes from continuing to occur in English, then I don't know who/what will. JordiLopezboy (talk) 23:11, 19 May 2024 (UTC)

GA Vowels

Is the open back vowel for GA [ɒ] or [ɔ]? The chart shows [ɔ] but links to [ɒ], and the footnote regarding the cot-caught merger makes no mention of the latter. Itstooslim (talk) 16:50, 28 June 2024 (UTC)

The phonemic symbol conventionally used to transcribe the (non-merged) GA THOUGHT vowel is /ɔ/, but actual phonetic realizations will vary from speaker to speaker among the different dialects that fall under the ill-defined umbrella of "General American". Many, of course, would pronounce it as [ɑ], and this increases with each successive generation. The symbols displayed in the chart for each dialect are the conventional transcription symbols, i.e. they are not intended to reflect the actual phonetic pronunciation when it differs from the conventional symbol. The links, however are meant to link to the article for the most phonetically accurate phone, i.e. the symbol displayed and the symbol linked to are not necessarily intended to be the same in cases when the conventional symbol differs from the current mainstream phonetic pronunciation. This is strikingly the case for Received Pronunciation, where about half the symbols displayed in the chart link to a different symbol (/ɒ/ links to [ɔ], /ɔː/ links to [oː], etc), as the conventional RP symbols are widely acknowledged to be phonetically inaccurate to modern RP pronunciation.

So the chart does not necessarily need fixing to either the displayed symbol or the article linked to. The sybmol displayed should definitely be /ɔ/, and the symbol linked to should be whatever is most phonetically accurate. In this specific case, I don't have a firm view on whether [ɒ] or [ɔ] is closer to the current median pronunciation of the THOUGHT vowel among GA speakers without the cot-caught merger. Likely some unmerged speakers have a realization closer to the former and other unmerged speakers have a realization to the latter. Indeed, the rise of the Low-Back-Merger Shift means that many merged young speakers now have a realization of their PALM-LOT-THOUGHT vowel in the [ɒ]~[ɔ] range. As far as I am aware, there has been no Geoff Lindsey-esque attempt to produce symbols that are phonetically accurate to modern GA. So I don't have a firm view on which article we should link to.

However, I will note hat the article itself is inconsistent, as the General America table links to [ɒ] but the RP and GA symbols table above links to [ɔ], despite the RP links going to the most phonetically appropriate articles rather than the articles for the transcription symbols. So, we need to decide which symbol is more phonetically appropriate to link to, and change one of the links to match. Offa29 (talk) 17:35, 28 June 2024 (UTC)

Move discussion in progress

There is a move discussion in progress on Talk:International Phonetic Alphabet chart for English dialects which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 15:48, 4 October 2024 (UTC)

Is there a "dialect-wide" "L[v]V-dropping" going on? (as in "lives">"lies")

Is there a "V-dropping" phenomenon that is systematically going on in English? A phenomenon of either the phone [v] being dropped, or [v] being dropped in a "L[vowel]V" combination. I've noticed this in television commercial ad copy, although I don't know if it's a "widespread" recognized change. The example that comes to my mind is where "lives" is pronounced closer to "lies". My dialect is (I think) midwest - Bloomington Indiana - whatever that is. Jimw338 (talk) 16:51, 23 November 2024 (UTC)

Unless you have a source this article may use, this topic belongs in Wikipedia:Reference desk/Language. Nardog (talk) 09:46, 24 November 2024 (UTC)

more accurate vowels, diphthongs and such

user @Nardog has requested to discuss my edits regarding RP vowels, and their lack of concensus. while i am unsure what the purpose of said discussion would be, i think it would be far better placed here than in an ever-increasing string of edit reversions. as such, i will reinstate my prior edits once more and ask that in future we discuss this here. if you are confused about my implementation, please see my citation, which i put next to the "FACE" lexical set (though i mistakenly said i put it by the "PRICE" lexical set, thats my bad). Wawapedia (talk) 09:23, 21 December 2024 (UTC)

"i think it would be far better placed here than in an ever-increasing string of edit reversions"

It would indeed, but as you are the one who wants to make a change to the established version of the page, the onus is on you to obtain consensus through discussion before making those changes.

The Geoff Lindsey symbols you wish to implement are indeed more phonetically accurate in most/all cases. Certainly no-one is going to dispute that the modern RP dress vowel is closer to [ɛ] than [e], or that its trap vowel is closer to [a] than [æ]. Some may debate on whether it is more appropriate to transcribe the diphthong glides with ⟨j⟩ and ⟨w⟩ rather than ⟨ɪ⟩ and ⟨ʊ⟩, but that is really a question of phonemic analysis rather than phonetic difference. For what it is worth, it is the Wikipedia convention to use vowel symbols rather than semivowels in such situations, even when we are giving Lindsey-esque phonetically accurate transcriptions of modern RP, such as in the table here.

However, phonetic accuracy is not the only consideration in selecting a set of phonemic symbols, both generally and for this specific Wikipedia table. It would be more phonetically accurate to write ⟨ɹ⟩ rather than ⟨r⟩, but we don't do that, nor do most sources.

We also have to consider transcription simplicity and established convention. In this case, there isn't much of a difference in simplicity in most cases (though /ɪj/ and /ʉw/ are a bit more unwieldy than /iː/ and /uː/), but the symbols used on the prior version of the page (which mostly originate from A. C. Gimson in 1962) are, in most cases, the long-established standard symbols for transcribing RP, still used in most sources. The Lindsey symbols which we first proposed in 2012, have gained traction in certain circles, particularly with the rise of his YouTube channel in the last few years, but they still do not have anything like the widespread currency that the Gimson symbols have, so much so that Lindsey himself still uses almost all the Gimson symbols when he teaches the UCL summer course.

We do have one difference in our symbols from Gimson: our use of ⟨ɛː⟩ rather than ⟨ɛə⟩ or ⟨eə⟩ for the square vowel. So evidently we're not 100% committed to maintaining every Gimson symbol. But what you're proposing would be a far greater deviation.

Don't get me wrong, I personally dislike the continued widespread use of the phonetically outdated Gimson symbols and would be very happy if the Lindsey system or something similar to it replaced them as the widespread standard transcription for RP, but that hasn't happened. I acknowledge that it is awkward how on the present page we have links showing one symbol (the established Gimson one) but links to a different page (the one for the sound represented by the more phonetically accurate symbol) -- for example our link diplaying as ⟨e⟩ links to the article for [ɛ]. This has confused editors in the past and is a rather confusing to the uninitiated, I admit. So I'm not totally against making some changes. But I think it may be difficult to justify and obtain consensus for making such a massive overhaul.

If there is one symbol where I think a change is most justified, it would be to change the dress symbol from ⟨e⟩ to ⟨ɛ⟩, as that would avoid the rather odd situation we have at the moment where we have a table row that gives different symbols for RP and GA dress vowels, which are pronounced essentially the same (if anything, the median RP dress is probably slightly more open, the opposite of what the symbols would suggest). This already has some adoption in more limited reforms that Lindsey's like Clive Upton's reform that is now used in many Oxford University Press dictionaries. But as ⟨e⟩ is still the most widely used symbol for the RP dress vowel, even that might be hard to justify if our primary aim is to reflect common notation. Offa29 (talk) 12:25, 22 December 2024 (UTC)

if the issue is the lack of consensus of more accurate symbols, how would one go about obtaining that? is there some certification of concensus which may be obtained? is it not the case that, given that you agree that my proposed revision is more accurate, we already have consensus?

if the issue then lies in that most people still use the Gimson symbols, what can be done about that but to stop using them ourselves in the hopes that others follow suit? overall im unsure of the sensibility of worrying about consensus over correctness

thanks for your reply Wawapedia (talk) 08:12, 28 December 2024 (UTC)

(also, for what its worth, i think using glide symbols in the cases of /ɪj/, /ʉw/, /ej/ and the like makes more sense insofar as that they avoid having to invent the concept of "glide insertion" in situations where they appear before a vowel.) Wawapedia (talk) 08:16, 28 December 2024 (UTC)

You seem unaware that, when it comes to phonemic transcription, "[t]he IPA does not provide a phonological analysis for a particular language, let alone a single 'correct' transcription, but rather the resources to express any analysis so that it is widely understood" (IPA Handbook).

Not only do Gimson's symbols for those vowels continue to be used by virtually all dictionaries of British English that use IPA ([1][2][3][4][5][6]), they are alive and well even just in works published in the last few years: [7][8][9][10]. And we give university presses and leading academic publishers more weight than one author's book or YouTube video.

Wikipedia doesn't lead; it follows. Lindsey's notation might indeed be widely accepted one day. But until then, we have nothing to "correct". Nardog (talk) 09:12, 28 December 2024 (UTC)

why would i be unaware of that. dont patronise me.

so youre telling me that your intention is to have a page which presents information which is inaccurate and out of date on the basis that thats whats popular? in the section you linked, it says to wait until things have been reported by "reliable sources", have i not done that? do i need more than one? do i need ones of a specific medium? though such a thing was never specified?

i really dont see the utility in upholding misinformation on the basis that thats what everyone else is doing, that seems quite unintuitive and if anything irresponsible. Wawapedia (talk) 20:30, 28 December 2024 (UTC)

Presumably we're talking about phonemic (rather than phonetic) transcriptions for RP, right? The phonemic symbols here go by certain time-worn conventions, which (a lot more arbitrary that precise phonetic symbols) were selected for various reasons: simplicity or accessibility to newcomers, popularity/tradition per reliable and long-respected sources, and (admittedly to a lesser degree) rough phonetic accuracy. But for an even greater degree of phonetic accuracy, Wawapedia, you can still use Lindsey's symbols in square brackets on the page or link to Lindsey's notation at Received Pronunciation. Phonemic transcriptions in slashes by definition cleave less to phonetic exactness than transcriptions in square brackets. And both systems can (and already do) co-exist on the page. (Note: Obviously, Lindsey is trying to introduce a new phonemic system, which is your point I believe, but it's still just one of many used by scholars, and as others have pointed out still a minority or non-mainstream system whose phonetic accuracy is somewhat irrelevant to the question of which phonemic system we choose.) Wolfdog (talk) 21:08, 28 December 2024 (UTC)