Talk:Stress and vowel reduction in English

Unstressed full vowels

I propose to delete the final paragraph of this section, for several reasons. (a) It is about consonant contrasts, not vowel reduction; (b) it treats aspiration as an all-or-none phenomenon, whereas there is no scientific evidence that I know of that there is a categorical difference between the aspiration of /t/ in 'manatee' and its supposed absence in 'humanity', and (c) the argument on the potential loss of consonant voicing distinction is made to depend on a syllable-division strategy used by Wells which is little used by other experts in phonetics, most of whom give preference to a strategy based on the Maximal Onsets principle. If there is a consistent difference in aspiration it is probably due to the weight of the final syllable, not to the supposed position of the syllable boundary. I hope that is OK. Peter Roach RoachPeter (talk) 09:15, 1 April 2012 (UTC)

I see your point that it doesn't seem relevant. Though there might just be a different article to move it to. The one thing that would make it relevant to this article is the claim "One of the effects of vowel reduction is the partial loss of voicing distinctions in preceding consonants" which goes uncited. Is it untrue? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 15:50, 1 April 2012 (UTC)

It's hard to separate cause and effect here. But as far as I am concerned, the potential loss of voicing contrast is due to the absence of stress in words like 'humanity' and has nothing to do with vowel reduction or syllable boundaries. 'Manatee' has primary or secondary stress on the last syllable according to all the dictionaries I have checked, which is why the /t/ is more noticeably aspirated than that in 'humanity', where the final syllable is unstressed. Phonetically speaking, aspiration is not an all-or-nothing, on-or-off matter, but a more-or-less one. In answer to your question, there are some documented claims (which would take me some time to track down) that removing the aspiration artificially from /t/ can fool English listeners into hearing a /d/ (I have simulated this for teaching purposes myself, using computer editing - it's very easy), and that the American "flapped /t/" that occurs before an unstressed vowel can sound indistinguishable from /d/. I think the discussion of this topic could usefully go somewhere else, but I don't think it belongs in Vowel Reduction. Peter Roach RoachPeter (talk) 17:06, 7 April 2012 (UTC)

Anybody have a good idea of where it should go? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 22:41, 7 April 2012 (UTC)

Well, Kwami obviously feels it belongs right back where it was. If that's going to happen to edits I make, I don't think I can be bothered to carry on with this. I have been trying to write some new material on this topic (currently in rough draft form in my sandbox), bMut putting that in will mean deleting a lot of current stuff, and I can see that leading to more conflicts. Peter Roach RoachPeter (talk) 07:27, 8 April 2012 (UTC)

Sorry, I forgot about this discussion, and was going off the reasoning in your edit summary, which didn't hold up. You say If there is a consistent difference in aspiration it is probably due to the weight of the final syllable, but the final syllables have the same weight, at least in those dialects which do not distinguish the CITY vowel; they differ in that manatee has a full vowel and humanity a reduced vowel. You also say 'Manatee' has primary or secondary stress on the last syllable according to all the dictionaries I have checked, which is why the /t/ is more noticeably aspirated than that in 'humanity', where the final syllable is unstressed. But "secondary stress" is just the dictionary convention for a full unstressed vowel as opposed to a reduced vowel, so again, we're back to the difference being between due to vowel reduction. (The OED doesn't follow this convention: they don't mark a vowel as stressed unless it is actually stressed, but unfortunately they only list the final-stress variant of manatee.) The paragraph therefore fits right where it is. (Not to say it can't be improved.) — kwami (talk) 09:06, 8 April 2012 (UTC)

If you are prepared to take LPD and EPD as reliable dictionaries, you will find that both have some words and names ending in unstressed /i:/ (to take a few from the letter A, see Agee, Aintree, Albee, anomie, apogee) and some ending with secondary-stressed /i:/ like 'manatee', so it can't be the case that ' "secondary stress" is just the dictionary convention for a full unstressed vowel as opposed to a reduced vowel '. As far as I can see, your argument is that vowel reduction causes allophonic differences in consonants preceding the reduced vowel. To me this puts the cart before the horse - I claim that vowel reduction AND allophonic differences in preceding consonants are caused by the presence or absence of stress. So we're just down to a straightforward difference of opinion. Peter Roach RoachPeter (talk) 09:35, 8 April 2012 (UTC)

According to Ladefoged, there is no such thing as secondary lexical stress. Once prosodic stress is accounted for, syllables are either stressed or not; if not, the vowel may be full or reduced. This results in the traditional 4-way phonetic distinction in English—primary (prosodic), secondary (lexical), tertiary (full), and quaternary (reduced)—though very few dictionaries indicate more than three of these, usually by conflating post-primary tertiary with secondary and pre-primary tertiary with quaternary. If that's just an opinion (though L had measurements to back it up), it's the opinion of an expert in the field, and so is still relevant for the article. — kwami (talk) 10:00, 8 April 2012 (UTC)

I have no disagreement with Ladefoged's analysis, and there is little difference between his account and the one I present in my own book apart from some terminological differences. The thing I am objecting to is expressed in the first sentence of the para in question: "One of the effects of vowel reduction is the partial loss of voicing distinctions in preceding consonants". This can only be interpreted as claiming that vowel reduction causes the effect on consonants. Ladefoged does not say that, nor does any other phonetician that I have ever read. I won't pursue this any further, though. Peter Roach RoachPeter (talk) 14:49, 8 April 2012 (UTC)

Well, that's something else entirely, and I agree with you. What would you do with it? (I tagged it as unbalanced for the time being.)

BTW, the article used to be at Unstressed and reduced vowels in English‎, so your proposed move was a restoration as much as anything. I think the current name is what you wanted? — kwami (talk) 18:36, 8 April 2012 (UTC)

I think we can just remove that first sentence and move the information elsewhere. I'm not sure where, though. English phonology? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:34, 9 April 2012 (UTC)

I think it belongs here, because IMO it's useful for a reader who's having difficulty distinguishing full from reduced vowels. /ʌ/ vs. /ə/ is particularly tricky, at least in my dialect, and having consonantal correspondences it helpful. Peter has drawn up some suggestions here. — kwami (talk) 20:16, 9 April 2012 (UTC)

I have made some more changes to the draft in my sandbox, referred to above, but I have now gone as far as I can with this topic. If it is still thought unsatisfactory, I will just abandon it. Peter Roach RoachPeter (talk) 19:43, 14 April 2012 (UTC)

Having read Kwami's latest comments on my draft, it's obvious to me that I am wasting my time on this. Leave the article as it is. RoachPeter (talk) 09:04, 16 April 2012 (UTC)

When you ask me for comments, and then practically shout that it's inappropriate for me to comment, simply because I point out that in one line you're presenting theory as fact, and when you make no attempt to discuss the issue, then yes, it does seem obvious that you're wasting your time working on a cooperative venture such as WP. — kwami (talk) 10:32, 16 April 2012 (UTC)

Whoa, Kwami. WP:BITE much? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:22, 16 April 2012 (UTC)

I suppose. But it's not like he's some kid. — kwami (talk) 16:26, 16 April 2012 (UTC)

And that justifies it how? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 17:29, 16 April 2012 (UTC)

Revised table of unstressed full vowels looks better, but I still don't see that a table is needed at all, given that all full vowels may be unstressed. Wouldn't there be just as strong a case for a table of stressed full vowels? RoachPeter (talk) 18:18, 15 June 2012 (UTC)

I agree; I was planning to eliminate the table anyway, when I get time to work further on that section. Victor Yus (talk) 15:42, 17 June 2012 (UTC)

Good luck! I hope you have more success than I had in trying to make sense of that section. RoachPeter (talk) 16:20, 17 June 2012 (UTC)

Much improved! If you need a ref for the 4-level approach, I think the first edition of the Longman Pronunciation Dictinary used this, but reduced to three for later editions (but I am away from my books at the moment and can't check). American treatments back in the Trager and Smith era (1950's) used to have 4 different stress phonemes. RoachPeter (talk) 18:57, 18 June 2012 (UTC)

Title of this article

I'd be grateful if someone with more experience of editing than I have could advise me. I am hoping to make some improvements to this article 'Vowel Reduction in English', but don't feel able to start on that until the relationship to the material in the article 'Vowel Reduction' is cleared up. The Vowel Reduction article quite correctly begins by pointing out that the term is ambiguous: it may refer to the phonetic process of reduction, observable under laboratory conditions as in the important work by Lindblom in the 1960's, or it may refer to the way in which one phonological vowel unit may take the place of another (for example, how the first syllable of 'photography' and the second syllable of 'photographic' come to have a schwa vowel). Hence the phonological aspect of vowel reduction is essentially about objects (reduced and unreduced vowels) and their incidence. The phonological study of vowel reduction could be historical (i.e. study of sound change), or it could be the type of theorizing found in Generative Phonology, where a surface reduced vowel is derived by rules from an underlying unreduced vowel (e.g. the vowels in 'photograph', 'photography' etc). So far, so good. However, when we come to the Vowel Reduction in English article, there is virtually nothing here that is actually about the process of reduction. What we do find is material on various reduced vowels of English, together with some information on the contexts in which they occur. If the title of this article were to be 'Reduced Vowels in English' it would be fine. Would it be possible to change this? I'm nervous of doing it myself. Peter Roach RoachPeter (talk) 10:51, 1 April 2012 (UTC)

Sure. Go ahead and move it. Or I will, if you like. — kwami (talk) 10:59, 1 April 2012 (UTC)

Yeah, the move seems fine to me. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 15:52, 1 April 2012 (UTC)

Thanks for the response. What i am proposing is changing the name of Vowel Reduction in English, and making the corresponding change in the reference to it in Vowel Reduction. My worry is whether this would break links from other articles. I haven't suggested moving anything. Would be glad if you would do the name change for me, if you are happy with that. Peter Roach. RoachPeter (talk) 20:47, 1 April 2012 (UTC)

Actually, because of the magic of Wikipedia, links to the old title are redirected to the new one so links aren't broken like that. At some point after a rename, robots (or magical fairies) go through and change the links to avoid redirects but that's just designed to reduce server demand. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 21:58, 1 April 2012 (UTC)

I would suggest retitling this article something like "Stress and vowel reduction in English", and merge in the English-related material from Secondary stress. These two topics are so much bound up in each other that it will be clearer and more useful to treat them both together. Also it seems wrong to have the secondary stress article, which (if it needs to exist) should be about a general linguistic phenomenon, being dominated by specifically English-related information. Victor Yus (talk) 10:44, 29 May 2012 (UTC)

Anyone object to doing this then? I've been tidying the secondary stress article, and have moved or copied the other specific-language information to the respective articles, but it's still dominated by a lot of detail about English, which ought to be moved out to an English-specific article, and because of the amount of detail I don't think it fits well (except as a summary) in English phonology; and so because of the close interconnection with the subject matter of this article, I find the most natural course of action to be to combine the two topics (stress in English, reduced vowels in English) into a single article. Agreed? Or not? Victor Yus (talk) 08:27, 5 June 2012 (UTC)

Sounds like a useful move to me. I'm worried in general by the way English tends to dominate articles on prosodic phenomena that don't have 'English' in the title. But I'm a bit uncertain about what's intended here - could you clarify? In the penult. sentence above, you say you want to combine 'stress in English' with 'reduced vowels in English'. Did you mean 'secondary stress in English' here? 'Secondary stress in English' is a tiny topic, but 'Stress in English' is big. RoachPeter (talk) 10:18, 5 June 2012 (UTC)

Yes, what I envisage is one comprehensive article covering the (interconnected) topics of stress and vowel reduction in English. Of course if that article itself got too big, then it could be split somehow, but based on the amount of information we currently have on these subjects on Wikipedia, I don't think it would be an excessive amount of material for a single page. Victor Yus (talk) 11:34, 5 June 2012 (UTC)

I like that approach. Would we need to title it "Stress and vowel reduction" in English or would "Stress in English" suffice? Neither title format is present for any other language. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:41, 5 June 2012 (UTC)

The thing is that in English there are distinctions made that, according to some approaches, are differences in level of stress, while other approaches interpret them merely as differences in level of vowel reduction. Calling the article just "Stress..." might imply a bias towards the view that vowel reduction is a stress difference. Victor Yus (talk) 14:41, 5 June 2012 (UTC)

I think that's right. When you look at all the places in Wikipedia where English stress is mentioned, it's clear that there are very different views about what English stress actually is, and it will be important to see that all properly argued views are represented. RoachPeter (talk) 16:01, 5 June 2012 (UTC)

[i] and [u] and related vowels

I'd like to suggest tidying up by merging material on [ɪ] and [ɨ] with more recently added material about the happY vowel [i], and adding material on the corresponding [u] vowel (inflUence) to the part on reduced back rounded vowels. There would then be just one piece on reduced front unrounded vowels and one on reduced back rounded vowels. I will do this if nobody objects. RoachPeter (talk) 09:19, 7 June 2012 (UTC)

Seems fine. — kwami (talk) 18:09, 7 June 2012 (UTC)

I've had a go at the editing I proposed, and hope it reads better. I have yet to add the works referred to into the Bibliography, but will do so shortly. RoachPeter (talk) 19:11, 8 June 2012 (UTC)

Looks good. Can you explain one thing to me: you write that happY is being analyzed as a neutralization of KIT and FLEECE in final position, which makes sense; but how do these accounts address the vowel in, say, hurrIEd (which I believe is distinct from the possible KIT in horrId)? Is this no longer considered to be the happY vowel, or are we relying on the morpheme boundary to explain the tensing? Victor Yus (talk) 10:06, 9 June 2012 (UTC)

Hi, this is quite a complicated matter, and I hesitated about going into it at length in the article. The present situation arises from two sources: (1) the observation that many younger speakers were using a longer, closer [i]-like vowel for final unstressed vowels previously transcribed with the KIT vowel, which led to the term happY-tensing (we could call that a sociolinguistic issue), and (2) the need of dictionaries to represent both possible pronunciations without needing to print the word twice (which is essentially about economy of space, a matter of great importance to dictionary publishers). HappY-tensing could be represented without inventing new symbols, simply by saying that speakers may use either the KIT vowel or the FLEECE vowel, but that increases the number of pronunciations to print in a dictionary. The first person to use the convention of using the [i] symbol as a KIT/FLEECE archiphoneme was Gordon Walsh, of Longman publishers, for the Longman Dictionary of Contemporary English in the 1980’s, but he never gets any credit for it because he wasn’t an academic. Once the idea of a single symbol that’s neutral between KIT and FLEECE caught on, its use spread to many dictionaries and textbooks. John Wells has pushed it further than others for the Longman Pronunciation Dictionary, so that in the 3rd edition (2008) it’s used not only for ‘hurry’ but also for ‘hurried’, on the gounds that the vowel doesn’t change when the suffix is added and a speaker is free to pronounce with KIT or FLEECE. The Cambridge English Pronouncing Dictionary does the same. Both dictionaries give only the KIT vowel for the last syllable in 'horrid', which is of course a single morpheme. However, the Cambridge dictionary doesn’t agree with Wells when he uses this vowel preconsonantally in unstressed prefixes such as be-, de-, pre-, re- and de-. Cambridge continues to use the KIT vowel for these. Do you think I should try to put in something brief to explain the convention more fully? RoachPeter (talk) 10:54, 9 June 2012 (UTC)

Sure, that might be useful, though my question was more about the phonological interpretation than about the dictionary convention. If hurrIEd has [i] and horrId has KIT, and these are distinct, then how can [i] be analyzed as being an archiphoneme that includes KIT? Victor Yus (talk) 11:29, 9 June 2012 (UTC)

OK, the honest answer has to be that the theoretical justification for [i] has not been argued in detail, to my knowledge. I think it can be said that the archiphoneme [i] (or |i|) does not include the KIT vowel, it takes its place in certain contexts. The idea of the archiphoneme has a respectable history in phonology (perhaps the most exhaustive analysis is Tsutomu Akamatsu’s ‘The Theory of Neutralization and the Archiphoneme in Functional Phonology’ (1988, 550pp)); the Wikipedia treatment of it in Neutralization and archiphonemes is clear and thorough. In the case of [i], Wells says ‘… the opposition between FLEECE and KIT is in effect suspended in weak syllables …’ (1982, p.166). I think we have to say that [i] only occurs in unstressed syllables, either before a vowel or at the end of a morpheme. Minimal pairs are hard to come by, but if we compare ‘gelid’ and ‘jellied’ (far-fetched, I agree), then the former must have the KIT vowel and the latter the happY [i]. We could not have [i] in ‘gelid’ because the vowel is neither prevocalic nor morpheme-final. RoachPeter (talk) 14:48, 9 June 2012 (UTC)

I see the minimal pair taxis/taxes in comments above. Or is that something else? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:12, 9 June 2012 (UTC)

Good one - I had forgotten that example. It works the same: the reduced vowel in 'taxis' is morpheme-final, so gets the happY vowel, while the equivalent vowel in 'taxes' is morpheme-initial, not final, and gets KIT. RoachPeter (talk) 16:39, 9 June 2012 (UTC)

Candid vs. candied might be another one. So given these pairs, it seems there's quite a reasonable case to identify [i] positively with FLEECE rather than KIT. But then we have this manatee vs. humanity thing, also mentioned here, which complicates things further... Victor Yus (talk) 18:35, 9 June 2012 (UTC)

Weak form and strong form

What should be done about this article Weak form and strong form? Like the previous version of secondary stress (and even more so in fact), it seems to relate entirely to English. Looking at Google Books and Scholar, it seems clear that this (the linguistic) use is not the primary use of these terms anyway. I propose merging it into this article, since that topic is intimately connected with the present one (two). Victor Yus (talk) 07:03, 19 June 2012 (UTC)

I think merging as you suggest would be a good idea - and filling it out with a bit more information. RoachPeter (talk) 14:50, 20 June 2012 (UTC)

OK, I have now done this. If you have more information you'd like to add, please go ahead. Victor Yus (talk) 13:01, 25 June 2012 (UTC)

Looks good. Just one concern - the case of 'our' as a weak form. You rightly separate it from the main weak form words, but I think readers could still get the impression that it's one of a special group of words. Unless you have a citation, I feel that any weak or reduced pronunciation of 'our' is really just a fast speech phenomenon. Unstressed 'hour' would reduce in the same way, wouldn't it? (e.g. In 'Half an hour over the limit') RoachPeter (talk) 18:04, 26 June 2012 (UTC)

I guess so, though I seem to remember reading somewhere that the smoothing of unstressed our is more consistent than that of other words, and even extends to dialects that don't generally exhibit much smoothing (probably somewhere in Wells, though I don't have any reference to hand). Victor Yus (talk) 11:41, 27 June 2012 (UTC)

You might want to check out Lisa Lavoie's 2002 JIPA article "Some influences on the realization of for and four in American English." — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:44, 27 June 2012 (UTC)

Distinctions between reduced and unreduced vowels

Before the recent improvements, the last para of this section was marked as possibly biased to a paticular point of view. This was because I wanted to remove the para, principally on the grounds that it is about consonants, not vowels. It also makes an argument that is based on one single, non-standard view of syllable division. When I tried to remove the para, Kwamikagami insisted on putting it back in. I would at least like to put the warning flag back on it. RoachPeter (talk) 14:58, 20 June 2012 (UTC)

How is it biased when we're explicit that it's the POV of certain people? — kwami (talk) 20:39, 20 June 2012 (UTC)

Are we? All I see is "some linguists." What's the contrasting view? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 23:49, 20 June 2012 (UTC)

And then we say Wells. — kwami (talk) 06:32, 21 June 2012 (UTC)

Right, we give the views but don't position them within a larger context of what other linguists think about the same phenomena. If it's a POV issue, we might want to do that. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:48, 21 June 2012 (UTC)

I should make it clear that the "biased" flag was not put in by me, and my argument was not principally about bias. It was that this para is not about vowels (reduced or unreduced) but about allophonic variation in consonants, and consequently not directly relevant to the article. RoachPeter (talk) 09:05, 21 June 2012 (UTC)

I don't claim to have knowledge about whether the descriptive claims in the paragraph are accurate, or are contadicted by other sources, but I don't think it's appropriate to add a POV flag when we already note that the views are those of certain people. (Of course if contradictory views exist, then they can be added.) I also don't think this can be considered irrelevant to the topic, as it addresses the important subject (which is not currently well explained) of what the real differences might be between the vowels we elect to call "reduced" and those we call "full". The different effects that the two sets of vowels (are claimed to) have on syllable division, and/or/hence on consonant allophony, would provide evidence of a real distinction between them, and hence is highly pertinent. Victor Yus (talk) 09:22, 21 June 2012 (UTC)

It does seem better contextualized than it was before. I haven't seen any suggestions for where it might be moved to. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:48, 21 June 2012 (UTC)

Unreferenced/POV

(Moved from User talk:Smyth)

FYI, if you feel like the section you tagged at Stress and vowel reduction in English has POV concerns, you might want to consider using {{POV-section}} instead of {{unreferenced-section}} and being more overt about your concerns in the talk page. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:27, 14 July 2012 (UTC)

I was actually doing two independent things there: marking the first section as unreferenced (because it is), and rewording the second section to avoid terms such as "note" and "report" which imply that their content is true. I have no remaining POV concerns about the current version of either section. – Smyth\^talk 14:33, 14 July 2012 (UTC)

Now Kwamikagami has reverted me, saying "these are the empirical findings". Clearly they are not undisputed, otherwise we would have only one viewpoint in the article rather than three. In those circumstances, we shouldn't use loaded words which imply that this particular viewpoint is the correct one. – Smyth\^talk 01:27, 15 July 2012 (UTC)

I agree with part of your POV. However, L&M report empirical findings. "Note" and "report" are the proper terms for that. That doesn't mean that they are accepted by anyone in particular. — kwami (talk) 01:36, 15 July 2012 (UTC)

It looks like Kwami has added sourcing. Is that satisfactory? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 03:29, 15 July 2012 (UTC)

My POV? I don't have one, I know little about this subject. But as I see it, the form of the article is currently as follows:

"Some linguists believe X. However, other linguists note that Y is a better theory, and report that the alleged X does not exist."

Is that really an acceptable way of describing the debate? – Smyth\^talk 12:00, 15 July 2012 (UTC)

I'd say it can. In reference to the four wording changes in this edit, here are my two cents.

• Phoneticians vs. some phoneticians: If we're describing a scholarly debate, we probably shouldn't include references to how prevalent a viewpoint is if we're not actually familiar with such figures. Technically, such a claim would be uncited anyway. In the past, I've changed uncited additions of "some" or "many" to "a number of" but this is semantically identical to making no reference to number at all.
• noted vs. argued: Either is fine. If we're going to be pedantic, these phoneticians are taking note of a possible analysis, not necessarily advocating that it is true of people's internal grammars. Still, "argued" would probably make it clearer that this is a scholarly position.
• report vs. state: Since the claim in question is a fact that's come from the result of scientific research, saying "state" unnecessarily calls into question the results of the research. That's not our job.
• their vs. this: Either is fine. The former reinforces the viewpoint as one held by people and the latter depersonalizes it. For our purposes, we might want to go for "this" which matches usage earlier in the paragraph.

These are really more of a wording issue than POV. The paragraph is so short and the alternate views are given equal coverage so that no reasonable reader is going to be untowardly skewed for or against this analysis. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:27, 15 July 2012 (UTC)

In the language I've grown up with, incite and insight, as well as increase in both senses, have the same exact stress

So I'd like a reference for the claim that they have different stress patterns, and a clarification on in which dialects they have different stress patterns. 173.66.211.53 (talk) 20:03, 9 February 2013 (UTC)

All you need to do is look them up in a dictionary.

The description is for standard English. Most of our dialects diverge from the standard in one way or another. — kwami (talk) 02:23, 10 February 2013 (UTC)

I'd like to know what dialect of English you speak. Where are you from? In my dialect (roughly a middle class Boston dialect), they are stressed differently, that is, the same way as described in the article. Or is it that you are pronouncing the words in isolation? Trying using them in a sentence. Bostoner (talk) 17:38, 5 June 2014 (UTC)

Please not "schwi" again

Please see the discussion of "schwu" in topic 4 above. "Schwi" is fine as a joke term for informal use, but I am not aware of any use of it in serious publications on English pronunciation. If there is one, please cite it. The risk is that if these terms are used seriously, people will come to regard the name "schwa" as having something to do with the 'palm' vowel and this would be wrong. The fact that there is a redirect doesn't make it OK to continue to use this term, unless with a warning about its use. RoachPeter (talk) 13:22, 18 May 2014 (UTC)

You're free to add a warning that this is not an "official" technical term in linguistics, but more like a jocular informal term like haplogy or, closer to home, NZ English DRESS rising. I wouldn't know either way; I'm not an Anglist. But as long as there is a redirect, the reader who is brought here through this redirect has a right to be informed what the hell a schwi is. --Florian Blaschke (talk) 14:15, 20 June 2014 (UTC)

DRESS raising, as funny as it sounds, is perfectly formal. See Lexical set#Wells Standard Lexical Sets for English and Relative articulation#Raised and lowered. — Peter238 (v̥ɪˑzɪʔ mɑˑɪ̯ tˢʰoˑk̚ pʰɛˑɪ̯d̥ʒ̊) 19:40, 16 February 2015 (UTC)

Normal Stress Rules in English

I came to this article looking for the "normal" stress rules in English. Yes, stress is phonemic in English, but when an English speaker encounters a word he has never seen before, at least in writing, which admittedly is secondary in language, he usually knows how to stress it. I just tested this by looking up a random word in the Oxford American Dictionary. I found the words febrifuge and its derivative febrifugal. I had no problem figuring out the stress, and the phonetic transcription confirmed this, so there must be some rules that apply in at least many cases. Bostoner (talk) 17:47, 5 June 2014 (UTC)

There are patterns, but no 'rules'. The more words you know, the more likely you will be able to predict the stress of a novel word, but you'll never be 100%. — kwami (talk) 02:39, 10 June 2014 (UTC)

There is a '"normal" stress rule' in English... it's called the 'Normal Stress Rule', imaginatively enough. But it's something of a Holy Roman Empire case, being more of an occasional guideline for vowel reduction than advertised. Quick summary: antepenultimate syllable, with a whole hatful of other 'rules' that contradict it (such as verb/noun distinctions), and some ad hoc exceptions on top, as per. Not clear why there's not an article on Wikipedia on it. The topic is touched on initial-stress-derived noun, and there's an article on the paper The Sound Pattern of English which itself goes into Chomskyesque detail and precise on the topic. 109.255.211.6 (talk) 01:09, 1 September 2020 (UTC)

phonetic indication of reduction of unstressed vowels

How does one indicate that the sound of u in "gerund" is not the same as in "up"? I can understand that some kinds of English don't have a schwa here, but the transcription /ˈdʒɛrʌnd/ would mislead most nonnative speakers to pronounce this word and many others incorrectly. Wells is claimed as a source for this misleading transcription; what exactly does he say about the pronunciation of unstressed vowels in general and in this case?--Espoo (talk) 16:51, 19 November 2017 (UTC)

"your" and yod-coalescence

Does "your" retain the initial /j/ after yod-coalescence? I think that the usual pronunciation of Does your child go to school? is [dʌʒɔː tʃaɪəld ɡəʊ tə skuːl], at least in the UK. Sol505000 (talk) 12:08, 2 September 2020 (UTC)

Interesting question! You might not have a reply from the usual linguistic suspects as this may not be either the most-visited or the most-relevant page. You could try at wP:LING or phonological history of English consonant clusters. However, here's my 'game amateur' attempt in the meantime.

My understanding is that by the very definition of yod-coalescence, if it occurs, then it elides the /j/ -- otherwise it would be yod-colouring, or some such malarkey.

Whether it does occur I think is largely determined by prosodic stress. If there's stress on the first word, or emphatic stress on the second, I don't think you can have coalescence. But if the emphasis is elsewhere, one might end up with a weak-form does (or possibly your, confusing myself trying to which out which is the more likely), and thus something like: [ˌdʌʒə(r)] or [dəˌʒɔː(r)].

Hope that's at least somewhat helpful, if only in summoning others to put me straight! 109.255.211.6 (talk) 22:40, 4 September 2020 (UTC)