Talk:English phonology/Archive 4

/ts/ or /t͡s/?

Is there a set rule where the sequence of /t/ and /s/ is pronounced as the affricate /t͡s/? I've observed that it only occurs in a syllable coda (e.g. cats, pits), and not in an onset (e.g. tsunami, as pronounced by most Americans) or between syllable boundaries (e.g. sight·seeing, but not Pitts·burgh). 125.63.25.18 (talk) 03:41, 17 July 2017 (UTC)

It may be a phonetic affricate [t͡s], but it's not phonemic /t͡s/. Mr KEBAB (talk) 06:22, 17 July 2017 (UTC)

Still, there are rules governing whether it's pronounced with a [t] + [s] sequence. In the examples you gave, the syllable boundaries help govern the pronunciation. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:16, 17 July 2017 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on English phonology. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20120404203210/http://www.uiowa.edu/~acadtech/phonetics/english/frameset.html to http://www.uiowa.edu/~acadtech/phonetics/english/frameset.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 08:18, 21 September 2017 (UTC)

R-colored diphthongs?

Aren't there also two r-colored diphthongs?

• /aɪr/, which of course if no vowel follows is spoken as if it were /aɪɘr/ even in rhotic accents, but is spoken if a vowel does follow: fire [faɪɚ], firing [faɪrɪŋ],
• /eɪr/ probably in one word only, but in this very much so, namely, they're?

--2001:A61:260D:6E01:E085:47D:7919:7C07 (talk) 16:47, 18 December 2017 (UTC)

They're is pronounced [ðɛɚ] in General American. It may be regarded as a phonetic diphthong, but it's centering, not closing. Mr KEBAB (talk) 18:07, 18 December 2017 (UTC)

A diphthong contains two vowel qualities. [aɪ] counts as a diphthong, but [aɪɚ] does not. [ɛɚ] or [eɪ] could be a diphthong, but not [eɪɚ].

They're closer to being triphthongs, in that a triphthong contains three vowel qualities, but normally a triphthong consists of a glide followed by a vowel followed by a glide, like [jew] or [i̯eu̯]. [aɪɚ] in American English would have to be a vowel followed by two glides – [ajɹ] or [aɪ̯ɚ̯] – or a vowel followed by a glide followed by a vowel – [ajɚ] or [aɪ̯ɚ] or [aɪ̯ɹ̩]. None of those has the correct glide–vowel–glide structure that is characteristic of a triphthong. — Eru·tuon 23:01, 18 December 2017 (UTC)

Then again, the vowels of hour, fire, loir in some analyses of Received Pronunciation are sometimes called triphthongs even though they have the structure prominent vowel–less prominent vowel–less prominent vowel, and under that definition of a triphthong, American fire might count as a triphthong too. — Eru·tuon 23:06, 18 December 2017 (UTC)

A number of diphthongs in non-rhotic accents parallel to these R-colored diphthongs in rhotic accents are called "centering" (in contrast to rising and falling). That might be an important distinction. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 04:11, 19 December 2017 (UTC)

In certain American English dialects, the diphthongs /ɪə/ and /eə/ can be found in words such as ideas and rail, respectively.

I have a question about /ɪə/. In which dialects of American English is this vowel class not used for the "ea" in "idea"? Thank you.LakeKayak (talk) 22:49, 25 January 2017 (UTC)

Some dialects (maybe even most) definitely realize the "ea" in idea as [i:ə] (i.e. "eye-DEE-a"), with a longer middle vowel. Although maybe it gets closer to [ɪə] in rapid speech. Someone who is better with U.S. dialects will hopefully be able to provide more details. ¡Bozzio! 11:07, 27 January 2017 (UTC)

Note however that the article specifies the plural ("ideas"), and adding the /-s/ can definitely alter vowel length. ¡Bozzio! 11:09, 27 January 2017 (UTC)

@Bozzio: I did want to avoid examples of /ɪə/, being pronounced as [iːə]. This is a different production with the same realization. This is not a different vowel class being used altogether.LakeKayak (talk) 21:53, 27 January 2017 (UTC)

I may be under a wrong impression as a foreigner, but isn't idea sometimes mispronounced as if it were ideer by rhotic accents as it is so similar a sound to r-colored vowels in non-rhotic ones?--131.159.76.234 (talk) 18:24, 10 January 2018 (UTC)

See intrusive r and note that John C. Wells "considered it objectively part of Received Pronunciation" and that he remarked that "at least in RP, 'linking /r/ and intrusive /r/ are distinct only historically and orthographically'." Therefore, "though he noted that it was still stigmatized as an incorrect pronunciation" he did not consider it a mispronunciation. (Before Wells, Daniel Jones took a similar stance towards intrusive /r/ in RP, especially after /ə/ and /ɑː/.) Love —LiliCharlie (talk) 20:08, 10 January 2018 (UTC)

Yes, and intrusive /r/ in RP is probably the reason for which people from Northern Ireland and Scotland tend to say /aɪˈdɪər/ for idea. Perhaps that's what the anon is talking about. Mr KEBAB (talk) 20:40, 10 January 2018 (UTC)

Allophones of vowels

There are some problems with this section, principally in the statements made without citations. The one I am most concerned about is the issue of the difference in length between vowels preceding voiceless (fortis) consonants and vowels preceding voiced (lenis) consonants. The statement perpetuates the old fallacy that the difference consists of a lengthening of the vowel in the latter environment. It has been established scientifically for 50 years that the effect is in fact a shortening of the vowel by the fortis context. I intend to rewrite this bit; the problem arises more obviously in Vowel length#allophonic vowel length. RoachPeter (talk) 08:05, 10 April 2018 (UTC)

Now rewritten. RoachPeter (talk) 14:45, 12 April 2018 (UTC)

can comma-STRUT be considered allophones in RP?

@W. P. Uzer: I have a question about comma (unstressed) and STRUT (stressed). I have a source [1] that states that the two are non-contrastive also in BrE (and RP). Should we consider that also in this article?--Officer781 (talk) 15:47, 7 December 2017 (UTC)

I've no objection, if that source (or some other source that makes the same claim) is considered authoritative enough for Wikipedia. Though the opposite view (presumably Wells's among others) should also be noted, with reference to the "an equal vs. unequal" types of example that are offered for it. W. P. Uzer (talk) 16:26, 7 December 2017 (UTC)

@Officer781: You raise a good point: Why GA―and not all the other accents such as RP―"lacks a proper opposition between [ʌ] and [ə]"? Wells (1982: 121, 480–1) doesn't say. As I said here, /ə/ never occurs in stressed syllables so one can take any full vowel and call it an allophone of /ə/ that occurs when stressed, and allocate secondary, tertiary, or quaternary stress to it to justify such an assessment. Out of all full vowels, though, [ʌ] is nonetheless most defensible to pick as the stressed allophone of /ə/ because /ə/ is often as open as [ɐ] morpheme-finally,[2] overlapping with [ʌ] (as Lindsey points out in his post). But then Wells' (1982) specifying GA and not other accents is now even more perplexing, because he describes RP /ʌ/ as completely central while GA /ʌ/ as still remaining relatively back (which might have been true in 1982, but I suspect is now outdated except in varieties affected by the NCVS). NURSE, lettER and STRUT+/r/ aren't neutralized in RP unlike in GA, so such an analysis would still be able to account for the opposition of hurry/furry, as /ˈhəri/ vs. /ˈfəːri/.

It is true that [ʌ] is more often regarded as a stressed allophone of /ə/ in GA than in RP, e.g. Merriam-Webster and Kretzschmar's work (OED, ODPCE, LDPCE). But in order to state "[ʌ] (stressed) and [ə] (unstressed) may be considered allophones of a single phoneme in General American", we need to state why, which I'm also trying to find. Thus Wells (1982) is insufficient as a citation to me. Nardog (talk) 00:44, 8 December 2017 (UTC)

I think I recall that Wells refers, somewhere, to examples of the an equal vs. unequal type (I seem to remember an ending vs. unending). W. P. Uzer (talk) 09:57, 8 December 2017 (UTC)

Hardly relevant discussion

But then, what if we'd write it down as [æn ɛndɪŋ] vs. [ʌnɛndɪŋ]. After all, words do have some separation in speech, and the schwa does have some coloring by the vowel it is written with. (Or is anybody supposing that national [i. e., "nashonal"], *nashenal and nashenil would sound exactly the same?)--2001:A61:260D:6E01:E085:47D:7919:7C07 (talk) 17:11, 18 December 2017 (UTC) An ending with unstressed an is [ən ˈɛndɪŋ] or perhaps [ɐn ˈɛndɪŋ], but not [æn ˈɛndɪŋ]. In normal speech, words aren't separated, at least not consistently. If you consistently pronounce an ending as [ən ˈʔɛndɪŋ] (with a glottal stop preceding /ɛ/), then chances are you're not a native speaker of English. the schwa does have some coloring by the vowel it is written with - that's definitely not true. Don't mistake orthography for pronunciation. Mr KEBAB (talk) 18:05, 18 December 2017 (UTC) As for /ə/ never occurs in stressed syllables, well, the phoneme as defined doesn't. And the other phoneme (as defined) that sounds alike is, by speakers and foreign language-learners alike, instinctively understood to be a different matter - only when we actually look at the sound, then /ɜ:/, which is most of the time stressed, does sound like [ɘ:], so much so that it is often written that way in older lexica (but very different from an [ʌ:]).2001:A61:260D:6E01:E085:47D:7919:7C07 (talk) 17:16, 18 December 2017 (UTC) But it doesn't sound alike. At least some scholars consider RP to have phonemic vowel length and /ə−ɜː/ is one of the short-long pairs. They always differ in length. Mr KEBAB (talk) 18:22, 18 December 2017 (UTC) Actually, you corroborate what I'm saying though I perhaps could have used more caution in expressing it. So, yes, the length makes the difference; but I was treating the same vowel, spoken long, and the same vowel, spoken short, to be the same vowel. The length of it is, of course, phonemically important (as is the stress!), which I treated instinctively as questions that are, while not unimportant, distinct from the "nature" of the vowel.--2001:A61:260D:6E01:E085:47D:7919:7C07 (talk) 19:21, 18 December 2017 (UTC)

@W. P. Uzer: Wells (1982) discusses what he calls the STRUT–schwa merger almost exclusively in the context of Welsh English and the "Near-RP", attempted approximation to RP by Northerners, and only very offhandedly with regard to General American, as cited above.

[S]tressed [ə] in STRUT seems to be particularly characteristic of northern Near-RP, with pronunciations such as cup [kəp], brother [ˈbrəðə]. Naturally, its use tends to imply the absence of any opposition between STRUT and the /ə/ of weak syllables. This is typically revealed by the lack of distinction between the strong and weak forms of but, does, must, us. Alternatively, northern Near-RP may have stress-sensitive allophonic variation between [ʌ] and [ə], which entails the appearance of seemingly weak forms even for words such as up and one, which have no RP weak forms – an alternation conveniently diagnosed by the pronunciation [ˈsʌmwən] someone (RP: [ˈsʌmwʌn]; other northern forms include not only ['səmwən, 'sʊmwʊn], but also [-wɒn]). (vol. 2, pp. 352–3)

One wonders, then, why he rules out (or at least doesn't mention) the analysis of the RP [ˈsʌmwʌn] as /ˈsəmˌwən/, even though the RP STRUT and commA may not phonetically overlap (which is rather a textbook instance of complementary distribution). Wells more or less answers this on his blog:

Some would claim that this is a non-issue, because STRUT is always stressed and schwa is never stressed. This argument might work if we define stress lexically, but it will not hold if by stress we mean a rhythmic beat in running speech.

Does this mean, though, it will, or may, hold true even if by stress we mean a rhythmic beat in running speech in GA? I really wish he had elaborated what he meant by "Even in GenAm it may well be considered that stressed [ʌ] and unstressed [ə] are co-allophones of one phoneme." Nardog (talk) 05:24, 21 February 2018 (UTC)

I presume it's because he believes that (most?) GenAm speakers don't make the kind of distinctions he refers to, e.g. between "an ending" and "unending", or "a large and tidy room" and "a large untidy room". I don't know how true this is - are there other sources that challenge this? W. P. Uzer (talk) 08:24, 21 February 2018 (UTC)

Segmentally, STRUT and commA in GA are said to exhibit little difference in height unlike in RP (e.g. Moulton 1990:133). So the use of ⟨ʌ⟩ for GA STRUT in itself may be just for comparison with RP and other accents, just like the use of ⟨ɜr⟩ for NURSE (see AoE pp. 480–1)—but it is still unfortunate that, while proclaiming the analysis of GA NURSE as /ʌr/ to be "logically unassailable", Wells states GA "lacks a proper opposition between [ʌ] and [ə]" only "usually", without revealing why exactly. Nardog (talk) 09:55, 21 February 2018 (UTC)

Even if they were segmentally identical, such pairs could still be distinguished in prosody, as in "...and tidy room" [ən ˈtaɪdiˈrum] vs. "...untidy room" [ˌənˈtaɪdi ˈrum]. But even if they were segmentally distinct, as in [ən ˈtaɪdi ˈrum] vs. [ʌnˈtaɪdi ˈrum], that still doesn't rule out the analysis of them as /ənˈtaɪdi ˈrum/ and /ˌənˈtaɪdi ˈrum/. Now that I think about it, whether STRUT and commA can be considered as comprising one phoneme must hinge rather on whether stressed and unstressed STRUT can ever contrast. Nardog (talk) 00:55, 23 February 2018 (UTC)

I suppose what Wells means by "it will not hold if by stress we mean a rhythmic beat in running speech" is that in RP e.g. someone can be unstressed and still be [sʌmwʌn], not [səmwən], but even then one can still make use of tertiary, quaternary, or even quinary stress, and attribute it to [ʌ] and regard it an allophone of /ə/. Stress, unlike pitch or tone, doesn't have a clear-cut definition so it eventually comes down to what features one is willing to ascribe to the concept of stress I guess. Nardog (talk) 17:41, 23 February 2018 (UTC)

General American differentiates between /ɑ/ and /ɔ/

@عربي-٣١: It seems that your edits on this page are predicated on General American being unable to feature the COT-CAUGHT merger. However, as you see on the General American article, which mentions "Cot–caught merger in transition", General American allows the speaker to decide whether to merge COT-CAUGHT or not. Merriam-Webster, while differentiating the two, does put a note for readers who merge the two. Also, the reason why I did not like you editing the Underlying Representation section is because the section is based off Lindsey (1990) "Quantity and quality in British and American vowel systems" whose argument for the underlying representation only works for cot-caught merging speakers, because the tense vowels form a natural class which factors into loanwords. By re-adding THOUGHT, you are negating the argument. I don't know too much about this because much of the work here was done by @Kbb2: and @Nardog:, so perhaps they could answer us on our doubts? Cheers.--Officer781 (talk) 02:36, 15 June 2018 (UTC)

This whole section seems rather strange to me; it seems to be giving quite a lot of space and weight to what is essentially one paper. If this is to be done, then certainly it should follow the argumentation given in that paper, but it should be made much more clear what is being done. Also I think there should be some examples to illustrate what is being said about loanwords, since that appears to be quite crucial to the argument. W. P. Uzer (talk) 08:11, 15 June 2018 (UTC)

I basically converted what was previously a section reserved completely to General American to one that was more equal to GenAm and RP. It was initially created because the original vowels section did not detail GenAm (see for instance [3]). I more recently considered removing the entire section altogether since we already explained the accents quite significantly (I defer the fate of the section to the users here). However, even outside that section عربي-٣١ removes "General" from the point about many (General) American speakers: [4] because presumably General American speakers cannot have the COT-CAUGHT merger. So this was the other point I'd like to raise.--Officer781 (talk) 08:37, 15 June 2018 (UTC)

Update. I have removed the section in question.--Officer781 (talk) 12:32, 16 June 2018 (UTC)

I was just going to suggest the same thing in alignment with W. P. Uzer. The section felt a bit overkill especially in relation to the rest of the article. I support this decision. Nardog (talk) 12:38, 16 June 2018 (UTC)

Table of consonants used in English

I propose that a series of changes take place for the table depicting the consonants in English's phonetic inventory. In it's current (as of Jan 24, 2019) form, the table has the following errors:

1) The Labial Approximant, /W/, has been placed into the position that a Velar Approximant would take up. This would suggest that the /W/ sound is formed with the mouth in the same shape as /g/ or /ŋ/. In order to fix this, it should be moved to its proper location among the labial sounds.

2) The Chart has no Labio-Dental column, and so the sounds /f/ and /v/ have no proper place in the table. Currently, they are listed as labial fricatives. Labial Fricatives exist, but English does not use them. The chart would be fixed if a new column "Labio-Dental" added. and then /f/ and /v/ moved there.

3) The chart has no row for Stops, and therefore The following sounds; /p/, /b/, /t/, /d/, /k/, and /g/ are wrongly listed as plosive/affricates. Labial Affricates, Alveolar Affricates, and Velar Affricates exist, but are not used in English. (except debatably /t͡s/). To fix this, there should be added a row for stops, and these sounds be moved into there.

4) In a past edit, I suggested that "fortis/lentis" should be changed to "unvoiced/voiced", however I now believe that I was mistaken on that suggestion.

If the table were to be changed according to my proposals, it would look like this:

A chart that represents the consonant sounds in English's phonetic inventory.

If anyone disagrees or doubts these ideas, I would also be happy to contact some linguistics professors at local universities to weigh in on the issue. I am confident that My proposed chart will withstand peer review. --Jonathansheehy1 (talk) 07:05, 25 January 2019

The question is whether we want a chart showing precise phonetic detail or one that merely distinguishes abstract English phonemes. As we aren't referring to any partcular accent of English phonetic detail seems to be out of place.

Labial is not the same as bilabial. Trask's A Dictionary of Phonetics and Phonology gives the following senses for the former: "... 2. adj. (of a segment) Articulated with the lips. 3. n. A segment so articulated, such as [p], [m] or [f]..." (7 senses altogether)

According to what has been labelled "plosive theory" plosives undergo three phases: approach–hold–release(=plosion). As English /p, b, t, d, k, ɡ/ do not always reach the release/plosion phase I prefer to call them stops rather than plosives. Love —LiliCharlie (talk) 17:13, 25 January 2019 (UTC)

I would put (plosives/)stops and affricates on separate rows, as I'm sure they normally would be in tables of this type. It's not as if there's any need for excessive compaction, particularly with one of the fundamental tables in this article. W. P. Uzer (talk) 17:44, 25 January 2019 (UTC)

I sort of like stops and affricates being in the same row, as they tend to have similar allophones (see English phonology § Obstruents), such as glottalized before another consonant and aspirated at the beginning of a word-initial or stressed syllable. — Eru·tuon 18:52, 25 January 2019 (UTC)

Thank you for the response. I hadn't known about stops and affricates following the same rules about when they are aspirated or unaspirated, that it interesting. That would make a good note to put below the chart, though I have to disagree that that would warrant putting them into the same category in the chart. I still think that the sounds are distinct enough that they should be separated. --Jonathansheehy1 (talk) 20:38, 25 January 2019 (UTC)Jonathansheehy1

Sorry to nitpick, but a stop with no audible release is still a plosive. I personally prefer the term "plosive" because it distinguishes oral stops from other occlusives, especially nasals, which are often referred to as nasal stops. But we are currently using "stop" at Stop consonant and in our IPA charts so I don't mind changing it to "stop" in this particular article until a larger consensus is made to prefer "plosive" site-wide. I also concur with W. P. Uzer's suggestion to split stops and affricates, and with LiliCharlie with respect to binding the labials together for the sake of brevity. Nardog (talk) 18:55, 25 January 2019 (UTC)

It can not hurt to be more precise. Its true that the current chart is very compact and looks nice, but in its current form, the chart can mislead people. For example, the chart places /f/ and /v/ in the same column as /p/, /b/, and /m/, which could lead people to thinking that these sounds are all made with the mouth in the same position (which, granted, they are formed with very similar mouth shapes, but still.)

That is a good point, I had not considered that "labial" was a category of sounds including both bi-labial and labio-dental sounds. --Jonathansheehy1 (talk) 20:38, 25 January 2019 (UTC)Jonathansheehy1

With this discussion in mind, I have devised another chart which now properly calls the bi-labial sounds as "bi-labial" rather than just "labial". Thank you everyone for your responses and critiques. I hope to hear more thoughts on this. --Jonathansheehy1 (talk) 20:38, 25 January 2019 (UTC)Jonathansheehy1

a chart depicting the consonants of English's phonetic inventory

I think that allowing the consonant to be compact is preferable. It's a phonemic chart and phonetic imprecision is generally okay, especially in an article that discusses the phonetic precision. The risk of English-speaking readers misunderstanding how to pronounce /f v m b v/ is pretty insignificant, especially when each character is a link to the relevant phone page. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 21:39, 25 January 2019 (UTC)

I concede that English-speaking readers will most likely not misunderstand how to pronounce the sounds in the chart, however people will be confused by inconsistencies in the chart. For example, according to the current chart, /p/ is a Labial Plosive, but clicking on the letter P redirects the user to the webpage for the voiceless bi-labial stop. It is reasonable to assume that this would confuse people, especially since the actual Labial Plosive can be found here on Wikipedia, and it is another sound on its own.

Even if making the chart more precise comes at the cost of making it bulkier and a little harder to read, It will be worth it to have a chart that is accurate to reality and consistent with the rest of Wikipedia. --Jonathansheehy1 (talk) 03:27, 26 January 2019 (UTC)Jonathansheehy1

1. I want /w/ to remain among the velars because it has the same potential of triggering an [ɯ]-like realisation of an adjacent /ə/ as the other velars. Also, its lip configuration (round and protruded, or endolabio-endolabial in Catford's terminology) is not like that of /p, b, m/ (which are typically mesolabio-mesolabial), but rather like that of /u(ː)/. BTW: The latter is placed to the right of /i(ː)/ on the vowel chart, and by analogy /w/ should also be placed to the right of /j/ on the consonant chart.
2. The bilabial occlusives /p, b, m/ often become labiodental when they precede another labiodental consonant, especially /m/ in words like comfort, triumph, symphony, but occasionally also the orals, as in hopefully etc. I see no reason to put the labials into two separate columns when they obviously form one natural class. Love —LiliCharlie (talk) 05:23, 26 January 2019 (UTC)

That is actually an interesting point. I had seen about a 50/50 mix when it comes to phonological charts, some list /w/ as a labial approximant, while others label it as a velar approximant, and until now I hadn't understood why. If I am going to be honest, I still am unsure where I stand about the placement of /w/, but if consensus rules it to be placed as a velar approximant, I won't disagree.

That is a good point. If you value organizing sounds according to how they interact with eachother more than simply manner of articulation than that is a better way to categorize them, and since I value the manner of articulation, my chart optimizes that. There really is no objective standard that can say one is better than the other.

What if we compromised? It is possible to make a chart that will have all the labial sounds; /m p b f v/ in a "labial" category, with a note that describes the behavior of this natural class, and then further subdivide the labial sounds into bi-labial and labio-dental. The resulting chart will be a bit bulky, but it will then accurately categorize the sounds according to both the phonological behavior and manner of articulation.

chart depicting English's consonants

--Jonathansheehy1 (talk) 21:08, 26 January 2019 (UTC)Jonathansheehy1

@Jonathansheehy1: When you say "the actual Labial Plosive can be found here on Wikipedia, and it is another sound on its own", what are you referring to? — Eru·tuon 06:00, 26 January 2019 (UTC)

@Erutuon: This - https://en.wikipedia.org/wiki/Voiceless_bilabial_affricate --Jonathansheehy1 (talk) 19:05, 26 January 2019 (UTC)Jonathansheehy1

That's not a plosive. That's an affricate. Nobody is going to be confused by the moderate level of phonetic imprecision that comes from combining columns in a consonant chart. Moreover, as a page on phonology, an important factor is how phonemes are classed together and this can involve grouping together sounds with similar places of articulation. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 19:32, 26 January 2019 (UTC)

Well, that's embarrassing. I had assumed that Plosives and Affricates were just different words for the same thing. (I kid you not, I believed that they were the same thing because on this very chart they are grouped together.)

Either way, I think there has been a lot of useful discussion here, and even if the chart never gets changed there are some interesting points that people have brought up that could improve the English Phonology article. --Jonathansheehy1 (talk) 21:08, 26 January 2019 (UTC)Jonathansheehy1

Voiceless palatal stop

In some English dialects, the voiceless palatal stop [c] is considered an allophone of [k] and used before [y] and front consonants, but the article does not note this. — Preceding unsigned comment added by 2601:647:5300:670:A1A6:A1A3:6B75:63D8 (talk) 02:44, 29 May 2019 (UTC)

We would want a source that states this and even says which dialects do this. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 03:15, 29 May 2019 (UTC)

I believe that the article on the voiceless palatal stop lists two sources. — Preceding unsigned comment added by AnUnnamedUser (talk • contribs) 22:36, 31 May 2019 (UTC)

Wells' Notation

Re recent edits (ɒ): I don't have a copy of Wells. Does he really use ɒ instead of ɔ? The OED uses ɔ. This always confuses me because I use both vowels in different words ("awe-ore split"?) Dbfirs 11:58, 20 July 2019 (UTC)

See Wells's Phonetic symbols for English page which shows the symbols he used in the Longman Pronunciation Dictionary and other publications. Love —LiliCharlie (talk) 12:19, 20 July 2019 (UTC)

Yes, he does. Wells inherited a system from A.C. Gimson whereby the vowels on RP always have a qualitative difference, rather than merely a difference of length. He uses ɒ for LOT and ɔ: for THOUGHT. Epa101 (talk) 13:09, 20 July 2019 (UTC)

Thanks. The OED actually gives both as alternatives for cloth. I now see what the edits were intending. Does anyone else have an "awe-ore split"? Dbfirs 13:24, 20 July 2019 (UTC)

@Dbfirs: These edits reflect the way RP vowels are most commonly pronounced by younger speakers. /ɒ/ is [ɔ] and /ɔː/ is [oː], just as /eə/ is [ɛː] and /æ/ is [a]. Kbb2 (ex. Mr KEBAB) (talk) 05:40, 22 July 2019 (UTC)

That explains why I am confused! We have Wikipedia IPA (1950s RP?), Modern RP (BBC English), younger British speakers who seem to have picked up Estuary English vowels, Northern vowels, and various dialect pronunciations, all different. Dbfirs 07:18, 22 July 2019 (UTC)

@Dbfirs: Here, the only difference is between phonemes and allophones. The RP phonemes /ɒ/ and /ɔː/ are realized as [ɔ] and [oː], the same as German /ɔ/ and /oː/. Older speakers of RP have more open norms for /ɒ/ that are closer to cardinal [ɒ], but the open-mid realization of /ɔː/ has been obsolete since the 1930s. RP /ɔː/ is higher and more rounded (closer to [oː]) than cardinal [ɔ], which is actually understood as a variety of LOT in RP (but it passes for native THOUGHT in General American). Kbb2 (ex. Mr KEBAB) (talk) 07:23, 22 July 2019 (UTC)

@Dbfirs: What's also confusing is our recording of the open-mid back rounded vowel. It's nowhere near cardinal [ɔ] but more like [o]. Cardinal [ɔ] sounds hardly any different from the cardinal [ɒ], that's why the two very rarely contrast in world's languages. Kbb2 (ex. Mr KEBAB) (talk) 08:35, 22 July 2019 (UTC)

@Dbfirs: (Sorry for bothering you). The recordings of the open-mid back rounded vowel and the open back rounded vowel have been replaced. Can you tell me if my message makes more sense to you now? Kbb2 (ex. Mr KEBAB) (talk) 11:40, 23 July 2019 (UTC)

Thanks for the re-recordings. Yes, I think it makes more sense now, but I'll listen again when I have more time. Dbfirs 06:30, 25 July 2019 (UTC)

I think those words have been pronounced differently in most traditional Northern English dialects. At least in WR Yorkshire, awe would be ɔ: whereas oar would be ʊə(r) - although this distinction is sharply recessive now. However, if you mean whether any dialect has ɒ: for one but ɔ: for the other, then I'm not aware of any. Epa101 (talk) 16:37, 20 July 2019 (UTC)

@Epa101: AFAIK, no English dialects contrasts [ɒː] and [ɔː]. There are Welsh and Northern English dialects that contrast [ɒː] (THOUGHT) with [oː] (GOAT), though, and I'd expect a similar contrast in conservative Irish English.

I'm actually not aware of any language in which there's a genuine /ɒ/-/ɔ/ (with the latter being genuinely open-mid, not higher - in the latter case ⟨o⟩ is a more optimal symbol) contrast based on height alone, without additional differences in roundendess or length (including diphthongization). Geordie may feature such a contrast (with [ɒː] being used for PALM and [ɔː] for THOUGHT) but it could be disappearing, with THOUGHT merging with close-mid monophthongal GOAT [oː]. Kbb2 (ex. Mr KEBAB) (talk) 08:53, 22 July 2019 (UTC)

Thanks for the response. If you look at the British Library recordings, some of them (e.g. Ossett) have GOAT as /ɔ:/ and THOUGHT as /ɒ:/. In most transcriptions of this area (e.g. Courtney, Petyt, Wells), the distinction is written as GOAT /o:/, THOUGHT /ɔ:/. Epa101 (talk) 21:04, 22 July 2019 (UTC)

@Epa101: They don't necessarily mean different things.

GOAT as /oː/ is probably more appropriate. To the best of my knowledge, open varieties of GOAT and FACE typically don't reach the open-mid height but are something in-between, like Spanish or Romanian /e/ and /o/. It's thus better to write them with ⟨eː⟩ and ⟨oː⟩, per the Handbook of the IPA. Those who write them ⟨ɛː⟩ and ⟨ɔː⟩ do so because they're more open than close-mid, rather than because they're open-mid.

The transcription of THOUGHT as /ɔː/ is one of the most annoying things about English phonology. In English, ⟨ɔː⟩ denotes a vowel that can be as close as close-mid (in RP, Australia, New Zealand) or as open as near-open (in General American, Northern England English, Welsh English). For such a variation ⟨oː⟩ is a more neutral symbol, and the symbols ⟨ɔː⟩ and ⟨ɒː⟩ should be reserved for accents in which THOUGHT is genuinely open-mid or lower.

I've never understood Gimson's decision to write THOUGHT with ⟨ɔː⟩ instead of ⟨oː⟩ when he writes DRESS with ⟨e⟩, which is about the same height. RP /ɔː/ has always (well, since the 1930s or perhaps 1940s or 1950s) been acoustically closer to [oː], with the amount of rounding it has. Kbb2 (ex. Mr KEBAB) (talk) 09:36, 23 July 2019 (UTC)

Ah, I get it now. It's because he wrote GOAT with ⟨ou⟩ and (in weak syllables) ⟨o⟩. That's hardly any justification for keeping ⟨ɔː⟩ after dropping both of the former symbols in favor of ⟨əʊ⟩ though... Again, given the amount of rounding cardinal [o] has, it's almost a perfect fit for RP /ɔː/ and recommending to use German/French/Italian/Portuguese /o/ as "second best option" for this vowel should've been standard practice in Gimson's Pronunciation of English since the very beginning. I don't get why they still recommend [oʊ] as a good enough substitution for GOAT, it's harder to learn to properly contrast it with THOUGHT.

There's another downside to writing ⟨ɔː⟩, which is that it can make non-native speakers think that because that vowel is written as it is (with a special IPA symbol ⟨ɔː⟩, rather than the ordinary ⟨oː⟩) it must then be open-mid, akin to German/Portuguese /ɔ/, and it isn't. Kbb2 (ex. Mr KEBAB) (talk) 13:06, 25 July 2019 (UTC)

Aspiration of English stops

Recently I made a minor edit to the conditions under which English stops are aspirated. This was reverted with the statement that my edit requires a reference or citation, which is fair enough. However, the person doing the reversion made some unusual claims about English syllabification, and referred to a web page by Wells. Following the link I find material by Wells which was published in a volume without peer review, and which is extremely out of touch with the typical standard analyses of English in linguistics. I will take the time to find the sources I need, but I caution that this analysis of English by Wells should never be the basis for material on Wikipedia as it is simply too far from the mainstream of generally accepted viewpoints. Safulop (talk) 07:09, 12 February 2019 (UTC)

OK, the appropriate reference to support this would be Iverson and Salmons 1995 "Aspiration and laryngeal representation in Germanic," Phonology 12:369-396. They say pretty much exactly what I wrote in my addition that got reverted. This article has been cited hundreds of times and is considered standard in the field. The idea from Wells that the [p] in "upper" and so forth is not syllable-initial is not supported by any literature other than, perhaps, Wells.Safulop (talk) 07:46, 13 February 2019 (UTC)

Here is the relevant quote:

And since the influential metrical work of Kiparsky (1979), the two main contexts in which aspirated stops are found in English — at the beginning of a word or at the beginning of a stressed syllable - have been unified into a single environment, i.e. the initial position of a stress foot. p. 374

I don't think we need to mention feet at all. Perhaps the desired wording here is "aspirated word-initially (except in clusters beginning with /s/) and at the onset of a stressed syllable." — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 16:42, 13 February 2019 (UTC)

Are unstressed word-initial plosives aspirated? Why not just say "at the onset..."? (Also "beginning" instead of "onset" might be better; then we can do away with the caveat about clusters beginning with /s/.) Nardog (talk) 17:14, 13 February 2019 (UTC)

I think "beginning" would be better. Otherwise, the exception to aspiration in clusters beginning with /s/ needs to also apply word-internally, as in distinct /dɪˈstɪnkt/ and askance /əˈskæns/ (in which the st and sk clusters have an unaspirated stop). — Eru·tuon 19:59, 13 February 2019 (UTC)

Yes these stops are aspirated, albeit somewhat lightly, in word-initial position when unstressed. Note the difference between the /p/ sounds in "potato" (somewhat aspirated) vs. "gaping" (not at all aspirated). Safulop (talk) 08:45, 14 February 2019 (UTC)

The problem is that there is no way to correctly state the conditions for aspiration without invoking feet. What is currently stated on the page is wrong, and your alternative suggestion here is also wrong because it fails to account for the aspiration that sometimes occurs at the beginning of unstressed syllables word-medially, as with the [t] sounds in the middle of "opportunistic" and "Mediterranean". I don't see why we can't just use the peer-reviewed scientific literature which I already cited above instead of doing all the research over again here on the talk page. This is not how Wikipedia is supposed to work. We shouldn't be hashing through the problems on this page that scholars already consider to be solved. We should simply repeat what they say about it, cite it and move on. Safulop (talk) 23:24, 27 August 2019 (UTC)

I agree the current wording is inadequate, though not so much because it doesn't mention feet as because there are competing approaches to syllabification, as the article explains later.

You (and Iverson & Salmons) are talking about the phonological definition of "aspirated". The article is talking about the phonetic definition. Is the VOT of [t] in opportunistic so much longer than in sty as to be considered aspirated in the sense the latter isn't? I feel like it's so much better to just list the VOT of the fortes in different environments (if we can find such a study) than to just discuss which ones are "aspirated" and which ones are not (although mentioning which ones are traditionally considered "aspirated" wouldn't hurt).

While we're at it, if you're so well versed in metrical feet, will you be interested in augmenting Flapping and Foot (prosody)? We recently had a discussion about feet here. Nardog (talk) 00:25, 28 August 2019 (UTC)

Distribution of clear and dark /l/

The article claims that in Australian English (among others), /l/ is always dark. I have an Australian accent, and as far as I can tell, I use both allophones of /l/. The description of Received Pronunciation matches how I talk: /l/ is clear before the vowel in a syllable and dark after the vowel in a syllable. Replacing a clear /l/ with /w/ makes the word sound wrong, but replacing a dark /l/ with /w/ is almost unnoticeable, i.e. "load" and "woad" sound very different and "hill" and "hiw" sound almost the same. If the three main types of Australian accent (Broad, General, and Cultivated) are given the values 0, 5, and 10 respectively, my own accent would be about 7 on this scale. Perhaps /l/ is always dark in the Broad accent? If someone can find some research to point to about the pronunciation of /l/ in at least these three types, that would be good. Danielklein (talk) 05:12, 7 May 2019 (UTC)

At kindergarten in Sydney in the early 1990's we were taught that /l/ is pronounced as what I would later learn is called a dark /l/. But I realised right away that this was incorrect. Having studied more about these things in later years I now know why. In the General Australian accent we only use dark /l/ at the end of words (or perhaps syllables?), e.g. as in "purple". At the beginning of words we always use a clear /l/. The claim in this article is wrong. 1.126.108.147 (talk) 01:34, 15 October 2019 (UTC)

Do you have a source? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 15:19, 15 October 2019 (UTC)

In the context of English phonetics "clear" doesn't necessarily mean not velarized, it means not vocalized and less velarized (or not velarized at all) than the coda /l/. The standards for calling the various allophones of /l/ "clear" are rather loose, in some dialects that are claimed to feature a distinction between clear and dark /l/ there's what to my ears sounds like an allophonic distinction between weakly and strongly velarized /l/. General American is a perfect example of that, compare "like" [lˠaɪk] with "ill" [ɪlˠˠ]. Australian English is probably no different (though "like" is obviously [lˠɑɪk] in Australia), except "ill" can be pronounced as something very close to [ɪw]. This is sporadic in America. Kbb2 (ex. Mr KEBAB) (talk) 15:28, 15 October 2019 (UTC)

"How to pronounce English" listed at Redirects for discussion

An editor has asked for a discussion to address the redirect How to pronounce English. Please participate in the redirect discussion if you wish to do so. 1234qwer1234qwer4 (talk) 23:26, 8 March 2020 (UTC)

ŋɡ vs ŋ

There is no mention of the variant pronunciations ŋɡ vs. ŋ in the same words e.g. /ˈɪŋɡlənd/ vs. /ˈɪŋlənd/ and /ˈɪŋɡlɪʃ/ vs. /ˈɪŋlɪʃ/ compared to the apparent lack of variation in the pronunciations of some or most words with the phoneme ŋ, which apparently always have either ŋɡ or ŋ, never both, as in the words finger and singer.

I actually ended up here because the same problem is on the pages Italian phonology and Italian orthography, which explain how words like piangere are pronounced but not words like pianga, ŋɡ or ŋ. Neither does Wiktionary... --Espoo (talk) 07:47, 5 May 2020 (UTC)

I haven't heard that with England, but I have heard people pronounce hanger with an epenthetic g. I haven't seen sources that talk about this, though. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:42, 5 May 2020 (UTC)

We seem to have more than one issue raised here. The first case is one of free variation - it is certainly possible for the cluster /ŋɡl/ to be simplified to /ŋl/, though not common. I notice it in singing (perhaps to maintain continuous voicing), for example Bud Flanagan's performance of "Who do you think you are kidding, Mr Hitler" when he sings "if you think old England's done" with /ɪŋlənz/ - (see https://www.youtube.com/watch?v=Jvr6X054xLY ). However, there is in addition the case of distributional constraints on /ŋɡ/ and /ŋ/. Here there are rules governing where the cluster /ŋɡ/ occurs (in words such as 'finger' /fɪŋɡə/) and where /ŋ/ (e.g. 'singer' /sɪŋə/). Minimal pairs exist, such as 'hanger' (object for hanging things on, /hæŋə/) vs. 'hangar' (place for storing aircraft, /hæŋɡə/), or 'longer' (a person who longs, /lɒŋə/) vs. 'longer' (comparative of 'long', /lɒŋɡə/). These distinctions are not so regularly observed nowadays - for example, almost everyone pronounces 'dinghy' as /dɪŋi/ whereas the rules predict /dɪŋɡi/. The rules for such cases are well known, and I am sure they must have been spelt out somewhere in Wikipedia, but I have not been able to find them. I could perhaps add something to this article on the subject. RoachPeter (talk) 14:18, 6 May 2020 (UTC)

American realization of the Strut vowel

The article on General American claims that the realization is nearly the same as in RP, which is supported by the accompanying image on the page. The article on Open-mid back unrounded vowel, however, claims that the realization is closer to the RP Nurse vowel. Meanwhile, this very page page does not mention what the realization of Strut is in American English (even though everyone seems to agree it is not actually an open-mid back unrounded vowel). If I had to guess, I think the confusion might be because it is always unrounded, while the near-open central vowel does not specify roundedness. -MToumbola (talk) 13:34, 7 July 2020 (UTC)

ts and dz

Since it might not be clear, Broman178 (talk · contribs)'s recent addition of the claim that sequences of /t/ +/s/ and /d/ + /z/ are "phonetically realised" as alveolar affricates is not backed up by the source given.

The source itself, Picard (1987), does discuss the sounds in question in the page range, but it says "English is usually recognized as having two affricate consonants" (/tʃ/ and /dʒ/) and that these are "truly a single phonetic unit in English". He then says that sequences of /t/ +/s/ and /d/ + /z/ are not usually presumed to be affricates, but advocates that they should be considered single units because "they can occur in exactly the same phonetic environments" as /tʃ/ and /dʒ/.

In addition to the inaccurate reasoning (/ts/ and /dz/ do not appear word-initially, for example) and the clearly controversial nature of the claim that the source does make (which means we can't insert it without some sort of qualification), the source does not claim that these sequences are realized phonetically as affricates. This claim likely won't be backed up by sources and I'm quite certain that most similar sources (introductory phonetics textbooks) will say the exact opposite. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 15:21, 3 October 2020 (UTC)

I'm very disappointed and frustrated that my edits were all reverted but I can understand your reasoning Aeusoes1 (talk · contribs) on this, however, I think it would be possible to add those back in if a better source is found which actually supports mention of that because in my opinion, its absurd to suggest that /ts/ and /dz/ are not affricates at all (if they are spoken quickly in English they certainly do become affricates [ts] and [dz] even though most phonetic sources would consider it a stop plus a fricative). I won't add this back in for now but I will carry on looking until I do find a source that supports it and maybe then I will reinsert it. Until then, I would suggest removing mention from the Voiced alveolar affricate/Voiceless alveolar affricate that they occur in English lads and cats respectively or reword those sentences because there is no source to support them either. Broman178 (talk) 19:02, 3 October 2020 (UTC)

Good call. I've modified the wording accordingly.

My suggestion as far as finding a source that makes the claim that you would like and is able to substantiate it is to look at phonetic studies. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 19:14, 3 October 2020 (UTC)

/ts, dz/ are indeed phonetically affricates, as are /tr, dr, tθ, dð, pf, bv/ (see e.g. Cruttenden 2014:186ff, Roach 2009:39–40, Wells 2008:15), but they are usually not counted among English affricates because what differentiates an affricate from a sequence of a plosive and a fricative, or a plosive with fricated release, is mostly phonological. Ladefoged & Maddieson (1996:90) write:

It is not always easy to say how much frication should be regarded as an automatic property of a release; some places of articulation seem to be often accompanied by considerable frication. ... At the other extreme, a combination of a stop and fricative that both happen to have the same place of articulation do not necessarily form an affricate. Phonological considerations must play a part in any decision as to whether a stop and a following homorganic fricative is to be regarded as an affricate which is a single unit, or as two segments (or two timing slots), forming a sequence of a stop and a fricative.

Nardog (talk) 01:49, 4 October 2020 (UTC)

Good point. — If we present sequences of English phonemes as single segments on non-phonemic grounds the following appears to be the logical consequence: As it is impossible to tell where syllable breaks are without knowing how English syllables are structured phonotactically, we might also have to present random sequences of vowels as diphthongs, triphthong, and tetraphthongs. That's certainly not what we want and what helps users. Love —LiliCharlie (talk) 03:21, 4 October 2020 (UTC)

Those sources are pretty good in my opinion, makes we wonder whether we can include these in this article (and those two affricate articles) and readd what I edited. I must say while the source I added didn't fully support my edits as Aeusoes1 explained above, that source still listed in the table that /dz/ and /ts/ could be affricates ([5]). Anyway, I might just point out that the whole definition of an affricate is that its a stop (plosive) followed immediately by a fricative so the notion that the stops + fricatives/dz/ and /ts/ in English do not fit into that is in my opinion ridiculous (even if these affricates are not main phonemes in English, they are certainly allophones of d+z combined together, especially in fast speech) and the phonetic studies which support they are not affricates are wrong (even phonetic studies can be incorrect, especially with the way they transcribe ɹ as r, and ɐ as ʌ in English). The affricate article even explains that "An affricate is a consonant that begins as a stop and releases as a fricative, generally with the same place of articulation (most often coronal). It is often difficult to decide if a stop and fricative form a single phoneme or a consonant pair". If I were to define it myself, I would consider affricates as consonant pairs rather than a single sound because they are essentially two consonants combined together although if I were to put that without a source, it would border on original research. Now if the sources Nardog has included above aren't good enough, I will look at phonetic studies myself to see if any support /dz/ and /ts/ being affricates in English and include it back in this article and the two affricate articles sometime later . Broman178 (talk) 09:06, 4 October 2020 (UTC)

I do want to say that I haven't put forth that phonetic studies have argued that English /ts/ and /dz/ are not affricates in English. I haven't looked into it. I do recall that, in Russian and Polish, a distinction is made between affricates and stop + fricative sequences, and the distinction manifests primarily in duration (so that, in Russian, a sequence of /t/ and /s/ is pronounced twice as long as the affricate /ts/). The issue of segment duration, of course, may be more fluid or slippery in rapid speech, but that's where phonetic studies can help us to be sure.

Also, since I'm apparently going to be the person to parse cited sources uncharitably here, Roach (2009) does not say that those sequences are phonetically affricates, at least not in that page range. He says that, under a certain loose definition of affricates, they would be included. But he doesn't seem to advocate that definition. He also doesn't say why (I can't access chapter 13, so I don't know what his justifications are). — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:02, 4 October 2020 (UTC)

That /ts, dz, tr, dr, tθ, dð, pf, bv/ (or at least most instances of them, where they don't span across multiple stems as in outrage) are indistinguishable from affricates from a purely phonetic point of view is, as far as I can tell, uncontroversial. See also Catford 2001:108, Abercrombie 1967:148, Laver 1994:366, Wells 1982:48, Wells 1990. Collins & Mees 2013:86–7 are pretty explicit. It is whether to call them, or count them as, affricates that can potentially vary depending on your view.

The affricate vs. stop–fricative contrast in Polish lies, as you said, in the duration of the fricative component, which means the so-called stop–fricative sequences are indeed, phonetically speaking, affricates (see Zagórska Brooks 1964). Nardog (talk) 09:56, 6 October 2020 (UTC)

I don't know if this will be any use, but since I have been (tentatively) quoted I thought I should provide chapter and verse. The relevant bit of Chapter 13 of 'English Phonetics and Phonology' can be read (I hope) at https://www.peterroach.net/affricates.html. There is no simple answer to the question of the phonemic status of affricates, but there is very little reason to treat /ts/ and /dz/ as single phonemes. RoachPeter (talk) 09:14, 6 October 2020 (UTC)

Nobody is arguing to treat /ts/ and /dz/ as single phonemes. The debate is about whether they are "phonetically" affricates. Nardog (talk) 09:21, 6 October 2020 (UTC)

Sorry, I just assumed this question was part of the larger issue of phonemic status. Whether or not /tʃ, dʒ, ts, dz/ are phonetically affricates seems a bit of a non-question to me. What's the alternative? That there are cases where the initial plosive is released and the fricative is then articulated as a separate segment? I suppose I am coming into this discussion in the middle and not getting the point. RoachPeter (talk) 09:41, 6 October 2020 (UTC)

For context, this is the material added by Broman178 and removed by Aeusoes1. I also don't understand the point of Aeusoes1's contestation. We already mention the affrication of /tr, dr/ after all. Nardog (talk) 09:57, 6 October 2020 (UTC)

As I said in the beginning of the section, I removed the content because the source used to cite it does not make the claim that it's being used to corroborate. Irrespective of the truth of the claim, we should generally avoid this.

I appreciate RoachPeter coming in and providing us the content of chapter 13 (I almost pinged you to ask for your input). I also appreciate Nardog providing additional resources to look into. My understanding of affricates is perhaps overly influenced by the research I've done on Russian, which makes a distinction between affricates and stop+fricative sequences. Perhaps I'm naive here, but I'd assumed that, if a language makes such a distinction, that the field of phonetics would do so as well. If I'm wrong, so be it, though it certainly makes it pretty meaningless to say that e.g. the /ts/ in cats is "phonetically" an affricate, as well as the qualification that this occurs "especially" in rapid speech, since the lack of such a distinction would make all sequences of a stop and homorganic fricative affricates. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 15:05, 6 October 2020 (UTC)

So how about discussing why /tʃ, dʒ/ are considered separate phonemes (but /ts, dz, tr, dr, tθ, dð, pf, bv/ aren't), perhaps in the Obstruents section? That would address all of our concerns AFAICS. Nardog (talk) 01:03, 7 October 2020 (UTC)

It couldn't hurt. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 01:07, 7 October 2020 (UTC)

(/t/-) affrication

How relevant is the affrication of alveolars (and other) stops for present-day English prestige variants, especially RP? It was quite promininently featured in the original 1962 edition of Gimson's An Introduction to the Pronunciation of English, and is still mentioned in the 2014 edition of Gimson's Pronunciation of English. Even the FL-advice to use [tˢ] and [dᶻ] for /t/ and /d/ to avoid dental articulation is still there, so apparently Cruttenden does not think of this as guiding learners to speak Mockney. I am aware that strong affrication is associated with basilectal regional speech, but since we mention t-glottalization – which per Buizza & Plug (2012) is less frequently heard than /t/-affrication in non-basilectal BE speech – shouldn't we mention (/t/-)affrication here too? The question boils down to: WP:DUE or not? A special ping to @RoachPeter, since Roach (2004) is mentioned in Buizza & Plug's article. –Austronesier (talk) 12:19, 6 March 2021 (UTC)

@Austronesier I'm afraid I haven't read the article, and can't get access to it immediately. Will try to do so and respond. RoachPeter (talk) 15:10, 6 March 2021 (UTC)

Underlying form of "longish"

Is /ˈlɒnɡəst/ really the underlying form of "longish" (and identical to that of "longest")?2806:102E:18:2767:8932:70F0:4FC6:CEAA (talk) 20:15, 6 March 2021 (UTC)

I would also like for a more qualified person to comment on this, it doesn't seem right at all to me. Then again, I'm not a linguist.--Megaman en m (talk) 00:07, 7 March 2021 (UTC)

Sorry, this was just a stupid error, now corrected. Thanks for pointing it out. RoachPeter (talk) 09:09, 7 March 2021 (UTC)

"How to pronounce English" listed at Redirects for discussion

A discussion is taking place to address the redirect How to pronounce English. The discussion will occur at Wikipedia:Redirects for discussion/Log/2020 July 14#How to pronounce English until a consensus is reached, and readers of this page are welcome to contribute to the discussion. –LaundryPizza03 (dc̄) 00:24, 14 July 2020 (UTC)

"/h/ may not be a phoneme in H-dropping dialects. "

Is this accurate? There's no reference, and it doesn't seem right - even in h-dropping dialects, wouldn't /h/ be preserved in intervocal positions such as the word "alcoholic"? — Preceding unsigned comment added by Yaldi5 (talk • contribs) 18:45, 12 September 2021 (UTC)

I agree this is an issue. It is widely claimed that there are English accents with no /h/ phoneme (I can add some refs), but the 'alcohol' example needs examining - I don't know how this would be pronounced in an h-dropping accent, nor do I know how many similar words there are. RoachPeter (talk) 10:27, 3 December 2021 (UTC)

There are a couple of interesting examples here: https://www.youtube.com/watch?v=BEkCp7nelf4&t=108s

To my ear, the phrase "an alcoholic" sounds close to "a now colic", although there are other examples where the same speaker pronounces the "h" distinctly.

"Alcohol" → "a(w) k(uh) o(w)" → "a(w) ko(w)". Jean-de-Nivelle (talk) 15:18, 3 December 2021 (UTC)

I'm not a proper linguist and have no experience with these /h/-dropping accents except trying to fake one, but I'd suggest that ⟨alcoholic⟩ would become something like /ælk(ə)ʔɑlɪk/, with a glottal stop (/ʔ/) taking the place of /h/. -- Categorically Not (🗪) 07:52, 25 December 2021 (UTC)

/oː/ is 'close'?

It's strange that in the vowel tables about RP and Australian, /ɔː/ = /oː/ are grouped with the close vowels. In the Australian table it's even stranger, because /eː/ is classified as a mid vowel, but /oː/, in spite of having the same height, is placed in the 'close' row. In the RP table, too, /ɔː/ could quite naturally be grouped with /ɒ/ as mid vowel and designated as its long counterpart, as the spelling suggests. It certainly isn't of the same height as the other 'close' vowels /iː/ and /uː/ = /ʉː/. It's almost as if the table is treating a vowel shift of the Swedish-Norwegian type, with /oː/ > /uː/, as a complete fact, when it's only a possible future development.--178.249.169.67 (talk) 21:24, 9 February 2022 (UTC)

Missing diphthong eə in RP pronunciation table

Why there is no "eə" as in "bear, care, where" in "Diphthongs" section of "Received Pronunciation" table?

Artur Severi Bilov (talk) 20:51, 23 February 2022 (UTC)

It's represented as the monophthong /ɛː/ following recent sources. Nardog (talk) 09:07, 27 February 2022 (UTC)

References

Adding tables to the phonotactic section

Given that English has a big selection of onsets, and that many of the groups described currently are too big (/tl/ for example fits "Stop plus approximant other than /j/", but I'd say its not an allowed onset), what do you think about showing this information in a table? The cool numel (talk) 19:16, 29 June 2022 (UTC)

English Received Pronunciation (RP) as described by the vowel chart is not accurate to modern pronunciation.

The symbols used are what is prescribed in pronunciation dictionary and taught to learners or RP, but this is done only out of common practice and not actually descriptive of RP or standard southern British English.

Should the chart instead be changed to the following as described by linguists like Dr Geoff Lindsey.

Mono	front		Centeral		back
	short	long	short	long	short	long
close	ɪ	(ɪ:)	ɵ	(ɵ:)
mid	ɛ	ɛ:	ʌ	ə:	ɔ	o:
open	a					ɑ:

Glides	w	j
ɪ		[ij]
ɛ		[ɛj]
o:		[oj]
ɑ:		[ɑj]
a	[aw]
ɵ	[ʉw]
ə:	[əw]

*some dialects replace these with [ijə] and or [ʉwə] and are therefore only marginal phonemes.

please watch Geoff's video on youtube titled "Why these English phonetic symbols are all WRONG" (wikipedia wont let me put the link) for if you want more information on the symbols of RP Jan Sosu (talk) 20:06, 14 October 2022 (UTC)

@Jan Sosu: Has anyone followed up on this? What's the consensus? Even though modern English IPA is still far from widespread, it seems worth at least mentioning on the page. Not to mention that it does make more sense than the traditional symbols for diphthongs. Udon Nomaneim (talk) 17:53, 1 April 2024 (UTC)

RP table

@Great Mercian: /ɒ/ is literally in the mid row. /ɔː/ is in the close row and links to [o], /ʊ, uː/ are in the central column and link to [ɵ, ʉ], and /æ/ links to [a]. This follows Curttenden (2014), the second source cited for the table, who says /ɒ/ is "nowadays nearer to C. [ɔ] than C. [ɒ]" (p. 126). As the diagrams at Received Pronunciation#Vowels show, the qualities of RP vowels have diverged, and even some of the conservative qualities (such as /ʌ, e/) don't match the values defined on the IPA chart to begin with. We have articles on Open back rounded vowel, Open-mid back rounded vowel, etc. but not on "Vowels transcribed with ⟨ɒ⟩", "Vowels transcribed with ⟨ɔ⟩", etc. So whenever one is linking phonemic symbols to articles, the only obvious choice is articles about the typical qualities of the phonemes. Nardog (talk) 22:01, 7 December 2022 (UTC)

Have you even tried to sound out the vowels? like, in the bathroom or casual conversation? although I don't believe we have the same dialect just try, like "I thought about it a lot". I honestly had to read your message thrice to even try and get it (can you link me the source please?). you know what, fuck it, I propose there's a Near-open back rounded vowel and it's used in West Midlands English! maybe then the people of this encyclopaedia will finally be happy. Great Mercian (talk) 02:07, 9 December 2022 (UTC)

Schwa not stressed?

"/ə/ does not occur in stressed syllables" is not qualified enough. It's true for Wikipedia's diaphonemic system, but only true for the phonological system of dialects in which kit and strut don't merge with schwa, and that means it's not true in prominent native accents like New Zealand English (kit merges) and General American according to some (strut merges). The statement could be qualified with "in dialects in which kit and strut don't merge with schwa", or is there some way to characterize the dialects where the statement is true based on region (southeast England for instance)? — Eru·tuon 17:06, 21 December 2022 (UTC)

FWIW I still have the same questions as those I asked in /Archive 4#can comma-STRUT be considered allophones in RP? almost five years ago. A recent YouTube video by Geoff Lindsey seems to have brought the idea of the STRUT merger in AmE to popular consciousness, but in it he takes for granted AmE doesn't contrast STRUT and commA without explaining why that's the case or why it doesn't apply to other varieties even though, as I understand it, STRUT is often [ʌ~ʌ̟] in AmE and the traditional definition of /ə/ is that it's a weak vowel so any full vowel can be said to be the stressed allophone of it. Recent edits to Schwa are clearly prompted by the video and likewise take the merger for granted, and I think it needs some work. Nardog (talk) 18:27, 21 December 2022 (UTC)

Phonotactics\Syllable structure\Other onsets

In this section a bit has been marked as dubious and very understandably so: it purports to be listing additional syllable onsets which English speakers produce only in loanwords, however many of the examples are names of places or people which I would not consider to be loanwords at all, nor do I find people's attempts to correctly pronounce a name as noteworthy of inclusion. Of the proper nouns listed, I would propose removing all except maybe Dvorak, as it is widely used to refer to the keyboard layout, and, as such, conveys meaning beyond simply the Czech surname. Anditres (talk) 03:45, 18 May 2022 (UTC)

What I noticed was marked as "dubious" in this section is the fact that the onset in "Kjell" is pronounced as /kdʒ/. I can actually see why this one in particular was marked for discussion; /dʒ/ (the English pronounciation of "j") may not reflect the pronounciation in the source language (which I would presume pronounces it as /j/). This also brings in the question: how often would one hear English speakers pronouncing "Kjell" with /kdʒ/ compared to /kj/ or whatever it is? Uptherial (talk) 23:35, 7 March 2023 (UTC)