Help talk:IPA/Conventions for English

Hi Aeusoes,

Sorry, your edits got lost in an edit conflict. You'd colored them dark red, a color which is scarcely visible on my screen against the blue background. Anyway, you were thinking of highlighting transcriptions which are different from the ones we use? kwami (talk) 09:53, 7 August 2009 (UTC)

Yeah. What we want is for someone who looks up e.g. emission at dictionary.com to easily find out here that dictionary.com is not good at indicating distinctions between ɵ and oʊ, or between ɨ ʊ, and ə. Probably even better than just having highlighted cells, we could have a different section for each of the 8 dictionaries that can be easily linked to in edit summaries or talk page discussion for quick reference. Such a conversation might look like this:

Person A: Why did you change my edit? The pronunciation of emission is /ɪˈmɪʃən/ that's what dictionary.com says.

Person B: See Help:IPA conventions for English#dictionary.com.

Person A: Ahh, I see now.

Person C: Person B, after seeing your logic and rhetorical skills, I would like you to have my baby.

In such sections, we can also point to which other dictionaries make the necessary contrasts. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 10:22, 7 August 2009 (UTC)

Yes, good idea.

What are you going to do with the baby? There's a reason I have that quote on my talk page! kwami (talk) 00:13, 8 August 2009 (UTC)

I added some highlighting. I think for maximum effect, we should minimize it: anything that affects multiple transcriptions that we can summarize (/r/, vowel length) would IMO be best left blue, so that the truly oddball correspondences are noticed. kwami (talk) 00:30, 8 August 2009 (UTC)

I noticed this page today, following a link added in Kwami's cleaning activities. It's a good one, but shouldn't we highlight the correspondences to the WP guideline instead of the deviations? I propose to use pale yellow, to avoid obscuring the symbols. And as suggested above we can absorb minor variants like vowel length, superscripts, diacritics. Doing this would show the logic of the WP system choices. −Woodstone (talk) 03:53, 25 April 2010 (UTC)

Well, the pink is scattered all over. I wasn't thinking of it as a way of convincing people of the validity of the WP system, but to highlight the conversions you should keep in mind if you're working out of dictionary X. kwami (talk) 04:53, 25 April 2010 (UTC)

Aussie

I'd thought that the Aussie IPA we have in some articles, which is credited to Macquarie Dictionary, used the same orthography as Macquarie. But it looks like Macquarie is only minimally distinct from RP. kwami (talk) 10:11, 7 August 2009 (UTC)

John Wells

Several editors use John Wells's blog for place names. Thus we should add a column for his conventions. kwami (talk) 10:43, 24 April 2010 (UTC)

I assume he uses the same conventions on his blog as he does in the Longman Pronunciation Dictionary, so I'd recommend we add a column for that rather than trying to divine a system from his blog. +Angr 11:37, 24 April 2010 (UTC)

Of course. You're right. Do you have it available? kwami (talk) 11:53, 24 April 2010 (UTC)

Alternatively, there's this list on his UCL page. Lfh (talk) 19:18, 24 April 2010 (UTC)

Kwami, I do have it available but I'm going to bed soon. I'll try to remember to do it tomorrow. Lfh, that list is mostly the same as in the dictionary, except that the dictionary includes symbols for GenAm as well as RP. The only difference I can see is that the dictionary uses ^ən and ^əl instead of n̩ and l̩. +Angr 21:01, 24 April 2010 (UTC)

I was going to bed too, or I would've looked further. kwami (talk) 21:40, 24 April 2010 (UTC)

Why is /r/ given as a phoneme as opposed to /ɹ/?

I've only heard of this being used in Scottish, and I don't even see an example of that in IPA_chart_for_English_dialects. I would guess that in 90% of cases the use of [r] in transcribing English sounds is flat out wrong from an IPA viewpoint. I'd like to see Wikipedia go by standard IPA and use [ɹ], rather than creating its own transcription for that sound; sure it would be more familiar to most English speakers, but considering that the transcriptions are given as phonemic rather than allophonic, I would think that these transcriptions are mostly used by people for whom the sounds of English might not be native, and would thus add an enormous amount of confusion. These conventions look fine in general; it's just the transcription of the English "r"-sound that sticks out like a sore thumb. I'd appreciate it if someone could let me know where I can talk about this (preferably on my talk page). Is there a talk page for the transcription of English somewhere? Theshibboleth (talk) 01:33, 13 May 2011 (UTC)

We do use [ɹ] on WP for GA and RP. However, such details are not terribly relevant in a phonemic transcription, so we follow the general practice in the literature of transcribing it /r/. This is partly a concession to readers who already have difficulty with the IPA, and in any case I doubt there are many English students out there who aren't aware that English /r/ is not a trill. This topic comes up every once in a while, but there's never been any real move to change the consensus arrived at when draughting IPA-en. — kwami (talk) 02:35, 13 May 2011 (UTC)

Yeah, using /r/ in broad transcription of English is well established practice. Jones, Gimson, Wells, Ladefoged, and Kenyon & Knott all do it. I think it would seem very pedantic of us if we didn't. —Angr (talk) 07:34, 13 May 2011 (UTC)

I'm on the verge of conceding this point--the two .edu links I checked for English transcription (The University of Arizona and the University of Iowa) give the sound as /r/; however I will point out that I am a linguistics student at UCLA (where Ladefoged once taught) and I have been taught in all my linguistics courses to transcribe the English r-sound as /ɹ/. I suspect that transcription of r as /ɹ/ is now or will soon be the standard in academics.

At a global level English r is from the sites I visited most often transcribed with /ɹ/ and Look for instance at [1] as well as the multiple translations of this page on other Wikipedias -- most of which give the phoneme as /ɹ/.

Also regarding "English students" knowing that English r is not a trill, that may be true, but only because they are students; Before I studied French or Spanish I had assumed that their r's were /ɹ/, and there is no reason to suspect that people from other language backgrounds would not assume that orthographic r had the same sound in English that it does in their own language--and what of speakers of languages that do not have an r-sound at all?

I was looking over WP:name and it appears that English language sources, as opposed to international standards, are favored, so I am going demonstrate support from English sources of /ɹ/ as standard. Here is an exercise from UCLA: [2].

Alright I must begrudgingly concede the point. I went over a number of journal articles, not in a particularly scientific manner, but nevertheless I could not find any support for the use of /ɹ/: [3], [4], [5], [6].

Nevertheless I certainly will use /ɹ/ in my own work, and doing so does not seem to me pedantic; doing otherwise suggests that [r] is the mother sound for /ɹ/ in various English dialects--this was what I had assumed when I saw this page, but I quickly learned otherwise. Theshibboleth (talk) 02:54, 14 May 2011 (UTC)

I assume the global sites you visited use ‹ɹ› because they're intended for a non-English audience, and they wish to remind their readers that it isn't [r]. But we don't have that aim. We don't need to tell English speakers that they don't trill their ars, and meanwhile we do have a problem with English speakers freaking out over all the odd letters in the IPA without us adding more of them. — kwami (talk) 05:07, 14 May 2011 (UTC)

Doesn't RP use a true /r/ as well as /ɹ/ and /ɚ/? For instance, isn't the middle consonant in marrow /r/? I guess it could be considered a flap or tap, /ɾ/, rather than a trill, /r/, but it seems to me to be closer to the Scottish /r/ than to the Japanese /ɾ/. I believe RP also uses a true /r/ in initial position, at least in very formal speech and song. (The old American "public-speaking" accent, still used by some preachers, also employs /r/: find a recording of Martin Luther King or Franklin D. Roosevelt, and note how he says "Europe".) If RP uses a true /r/, as I would say it does, isn't that a good reason to use /ɹ/ for the alveolar approximant? J. D. Crutchfield | Talk 21:29, 13 March 2014 (UTC)

Vowels (reduced)

Shouldn't the bottom line of the table for reduced vowels have the barred u, not barred ʊ , to match the barred i three rows above? RoachPeter (talk) 06:49, 10 July 2012 (UTC)

Yeah. That's how it's listed at WP:IPA for English. I've fixed it. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:32, 10 July 2012 (UTC)

What is æː in Wikipedia convention?

I'm trying to update the template at Blaa to {{IPAc-en}}, which is given as /blæː/ under {{IPA-en}}, but I'm not sure how to change the æː sound over to Wikipedia's conventions. Is it /eɪ/? Is this an omission, or is it just a problem with that page? 0x0077BE [^talk/_contrib] 19:14, 2 November 2014 (UTC)

I think the problem is that æː isn't a phoneme of RP English. I would think the most likely substitutions for æː would be /ɑː/ or (less likely) /ɛə/. RoachPeter (talk) 08:41, 3 November 2014 (UTC)

You might be interested to watch this: https://www.youtube.com/watch?v=6ugLNB0THLQ RoachPeter (talk) 11:52, 3 November 2014 (UTC)

Namespace

I think this article would find a better home in the main namespace. Aside from references to Wikipedia's transcription system, the content is pretty encyclopedic. Nardog (talk) 23:12, 19 June 2017 (UTC)

I don't know about that. There's constant reference to the Wikipedia transcription system and how to compare the systems of various dictionaries to our own. That doesn't make for good namespace content. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 18:06, 21 June 2017 (UTC)

I'm talking about the comparison tables though. Maybe, as opposed to moving the entire page, the tables can be copied to International Phonetic Alphabet chart for English dialects or somewhere. Nardog (talk) 01:06, 24 June 2017 (UTC)

Actually, there is a similar table at Pronunciation respelling for English#International Phonetic Alphabet. So that section might be expanded in accord with this page. Nardog (talk) 01:19, 24 June 2017 (UTC)

Yes, you have a good point. The tables themselves can, if anything, be copied to main space. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:06, 24 June 2017 (UTC)

Inherent stress in monosyllabic words?

Currently, the article says that “[s]tress need not be included in the notation of a monosyllabic word since it is self-evident”. I seriously doubt that. I would rather think that there are some monosyllabic words that have inherent stress and others that do not, mainly function words such as “the” or “of”. Sources? --mach 🙈🙉🙊 20:37, 20 September 2017 (UTC)

I can't think of any unstressed monosyllabic words that aren't also function words, a closed class that readers are expected to be familiar with anyway. Is there likely to be an instance of such a function word being transcribed in isolation? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 21:15, 20 September 2017 (UTC)

Those function words have the schwa, which never occurs in a stressed syllable anyway. Stress is always a matter of degree, so a monosyllabic word by itself cannot be said to either have stress or not have stress (it can be said, however, to have either a full vowel or a reduced one). And when we transcribe with our diaphonemic notation the pronunciation of a word (which also usually happens to be a proper noun or a technical term), 99% of the time we are transcribing the word as a word – not as part of an utterance. And whenever we describe the pronunciation of a word as a word, not just on Wikipedia but just in general, we are usually describing how the word would be pronounced when said in isolation – otherwise the second and fourth syllables of pronunciation need to be described as having the same level of stress (see Stress and vowel reduction in English#Degrees of lexical stress).

In any case, omitting stress in monosyllabic words is already the practice in LPD, CEPD, RDPCE (ex. ODPCE), and many lexical dictionaries (the only exception AFAIK is Merriam-Webster, which uses /ˈə, ˈər/ to notate the STRUT and NURSE vowels), so we would need a justification to go the other way. Nardog (talk) 10:18, 21 September 2017 (UTC)

"WL"

What does "WL" stand for? — Preceding unsigned comment added by 103.210.34.129 (talk) 07:52, 11 September 2020 (UTC)

Ditto. Does anyone know? Wolfdog (talk) 15:03, 28 March 2023 (UTC)

@Wolfdog: Using WikiBlame I found that the "WL" column was added in this edit, with an edit summary mentioning Windsor Lewis. His work is mentioned in the reference section. -- John of Reading (talk) 15:20, 28 March 2023 (UTC)

Aha! That's very useful; thanks! Wolfdog (talk) 15:21, 28 March 2023 (UTC)

Oh-oh, that was me … though I see the article did not exist back then (relieved). Thanks for adding the link. --mach 🙈🙉🙊 17:57, 28 March 2023 (UTC)

Consonants

Dear Editors,

On account of the fact that the pre-vocalic 'r' of rich is sounded in all varieties of English, and that the /ɹ/ symbol is not encountered in dictionaries of the language, and indeed rarely, if at all, outside technical literature on the subject, I humbly request that a slight modification be implemented within the 'consonants' section of the discussed help page, so as to provide it with the following appearance:

Consonants vary little between dictionaries. The ones which do are those in the words:

char /r ~ (r) ~ ∅/;

which, /ʍ ~ hw ~ (h)w ~ w/;

and new, /juː ~ uː/.

Wikipedia editors have decided to go with /tʃɑːr/, /hwɪtʃ/, /njuː/ for these words.

A few dictionaries, such as dictionary.com, use "/y/" for /j/, which is at odds with the official IPA usage, which defines /y/ as close front rounded vowel (as in French tu or German über), and appears as such in transcriptions of French and German, as well as some dialects of English.

Yours faithfully,

Maciuf (talk) 11:07, 19 April 2022 (UTC)