Talk:Roundedness

Formants?
How can you tell from the formants (or whatever) of a vowel if it's rounded? —Felix the Cassowary 10:24, 2 September 2005 (UTC)


 * Yes, but there isn't a simple correlation, like there is for F1 and hight (high vowels depress the first formant). With rounded vowels, all higher formants are depressed. However, the effect is most noticeable with F2 for rounded back vowels, and with F3 for rounded front vowels. kwami 19:03, 2005 September 2 (UTC)

Endo- vs. exo-labial
I'd really like to see a source on this endolabial vs. exolabial rounding. 72.130.89.63 (talk) 04:12, 17 January 2008 (UTC)

I second the request on the source for the endolabial vs. exolabial with reguard to front and back rounded vowels.Kunoodle (talk) 05:20, 2 April 2009 (UTC)

Big crit
WellsTribute (talk) 10:57, 13 July 2009 (UTC)

This article has a number of embarrassing confusions in it:-

a. Is it about rounding in vowels, or rounding in consonants, or both? It announces vowels as its topic in the first line, but refers piecemeal to consonants. Since the Wikipedia article on Consonants deals with consonantal rounding, and says that this present article deals with rounding of vowels, consonantal rounding should be dropped here.

b. Is compression a type of rounding or not? The first sentence under "Types of rounding" says that it is, but the remark about Iroquoian labials only makes sense if it is not. The claim that compression is not a type of rounding is specifically made in the Wikipedia article on Vowels. It is not substantiated, however. It appears to rest on two papers by Mona Lindau from 1975 and 1978 (mentioned in Trask's Dictionary). Since it hasn't reached the mainstream in the intervening 30 years, it shouldn't be given airspace.

c. 'Protruded/non-protruded' is not the only dimension along which vowel-rounding can vary. Rounding can also be:

- 'less/greater' (the corners of the mouth can be drawn together to a lesser or greater degree);

- 'less open/more open' (the aperture can be narrower or wider).

(See O'Connor, "Phonetics", p. 37, and Laver, "Principles of Phonetics", p. 278.) See 'less/greater' dimension is covered in the article, but not the 'less open/more open' dimension.

d. The distinction between protruded and non-protruded vowels has a smaller place in the literature than the distinction between 'less open' and 'more open' vowels, and a smaller place anyway than this article suggests. It gets, for example, no mention at all in the following standard reference works:

- Catford's 'Practical Introduction'

- Clark, Yallop & Fletcher's 'Outline of Phonology'

- Crystal's 'Dictionary of Linguistics and Phonology'

- Crystal's 'Cambridge Encyclopedia'

- Ladefoged's 'Course in Phonetics'

- O'Connor's 'Phonetics'

- Pike's 'Phonetics'

and only a brief mention in:

- Catford's 'Fundamental Problems'

- Ladefoged's & Maddieson's 'Sounds of the World's Languages'

- Laver's 'Principles of Phonetics'

e. 'Endolabial/exolabial' are the wrong terms: they are not synonyms for 'protruded/non-protruded', and are not used as synonyms in any of the works listed above. They actually mean 'using the inside/outside of the lip', with no implication as to whether the lip is rounded or not. This is specifically stated by Catford ('Fundamentals', p. 146), and clear from Trask. (The Note to the current article makes it look as though Trask claims they're synonyms for 'protruded/non-protruded', but he doesn't.) Catford's use of 'exolabio-exolabial' (not simply 'exolabial') is surely a pedantic joke - he has in an earlier chapter set up 'exolabio-exolabial', 'endolabio-exolabial', 'exolabio-endolabial' and 'endolabio-endolabial' as possible descriptors of lip-action, whether rounded or unrounded.

f. 'Vertically compressed' is an unhappy term - does it mean that pressure is applied from the sides, or from top and bottom? Trask uses 'horizontal compression' for non-protruded vowels; Catford ('Fundamentals') uses 'compressed vertically' for non-protruded vowels; and Lad&Mad use 'vertical compression' for protruded vowels. Laver, fortunately, is clear: non-protruded vowels have not only horizontal compression (in that the corners of the mouth are drawn in, creating rounding), but also vertical compression (in that the flesh displaced upwards and downwards by drawing in the corners is pressed back towards the teeth). Protruded vowels have horizontal compression (mouth-corners drawn in) but no vertical compression.

g. 'Endolabial/exolabial' have their meanings reversed in some (but not all!) Wikipedia articles, and on the bottom photograph in this article (both lower photographs show protruded lips, and are therefore 'endolabial' in this terminology.) The terms were also reversed on earlier photographs attached to this article, are currently reversed in Wiktionary, and reversed in many citations from Wikipedia on other web-sites, as shown by Google. The present article should include a note acknowledging this.

h. The claim that back rounded vowels are usually protruded and front rounded vowels non-protruded is not well-founded, despite appearing frequently in Wikipedia's articles on vowels. Catford states it briefly in 'Fundamental Problems', but produces no data; Clark, Yallop & Fletcher quote Catford, and again produce no data. None of the other standard works noted above mentions it - and if it were true, then it would surely feature in Ladefoged & Maddieson.

i. The claim that protrusion vs. non-protrusion can be phonemic is overstated. The Swedish case is far from certain - the two vowels in question also vary in length, position and off-glide. (See the discussion in Ladefoged & Maddieson; and note that Engstrand doesn't support the claim in his IPA Handbook article.) And the claim about Dutch is simply false. The IPA Journal has published analyses of six varieties of Dutch recently (1983(13/2), 1998(28), 1999(29), 2005(35/2), 2006(36/1), 2007(37/2)), as well as Gussenhoven's Handbook account, and none of them mentions such a distinction. Perhaps the contributor who entered this was confused by a reference to an endolabial (but not rounded) Dutch consonant in JIPA 1982(12/1). Given that the phonemic case is so weak, the unspecific nature of the term 'rounded' and the lack of an IPA diacritic are trivial.

j. A couple of minor points: (i) Note 4 says, "The convention is to use the asterisk," but this is not an IPA convention, and if (as stated) it's not done in the literature then it isn't a convention anyway. (ii) The opening of the second para. under 'Types of rounding' repeats material in the first para.

What I propose is this:

- remove references to consonants

- remove suggestions that compression (non-protrusion) is not a type of rounding

- add an account of narrower and wider apertures

- shorten the discussion on protruded vs. non-protruded

- replace 'endolabial' with 'protruded', and 'exolabial' with 'non-protruded'

- drop 'compressed'

- remove or soften the claim that back rounded vowels are protruded and front rounded vowels non-protruded

- remove or soften the claim of phonemic status for protruded/non-protruded

- clean up the account of what Trask says, and remove the 'endo/exo' reference to Catford

- add an acknowledgement of past errors

- add photographs (if I get permission) of narrow/wide aperture and protruded/non-protruded lip-positions

If there are no well-founded objections to the considerations I've put forward here, I'll start on the revisions in a week or so.

Sorry this is so long. WellsTribute (talk) 00:21, 29 June 2009 (UTC)


 * a. We should say something about consonantal labialization, since the two phenomena are so closely related. Rounded vowels labialize consonants, and labialized consonants round vowels. (In the Caucasus, Ndu languages, & central Australia, we find that phonetic [ku] and [ko] are phonemic and, for example; in Mandarin,  is [o] after labial consonants; in Vietnamese, final /k/ and /ŋ/ become  and  after /u/ and /o/, etc.) We should have a section on such interactions, just as we cover the connections between front unrounded vowels and palatalization. [I added this as a separate section.] If corrections need to be made, then it is in the other articles that do not properly describe this article.


 * b. This is a theoretical point depending on one's definition of "rounding". It's hard to say either way, since few discussions include this parameter. Generally both endo- and exo- would appear to be subsumed under "rounding", but when those or similar terms are not used, there can be some confusion. Japanese /u/ is not infrequently contrasted with English /u/ by claiming it's not rounded, for example, even though it is clearly labialized. If Catford proves correct that back vowels tend to be endo and front vowels exo, then clearly both are covered by 'rounded' in the lit.


 * c. I'm not aware of more/less open independent of the openness of the vowel. Perhaps you can add s.t.?


 * d. The reason endo/exo takes up so much space is that it requires space to describe adequately. For the rounding of back vowels, we can simply say that English oo is rounded while ee is not, and for the rounding of front vowels, we can draw the analogy of back vowels, even if that's not entirely accurate phonetically. If this were Swedish WP, we could similarly simply say that Swedish y is endolabial while u is exolabial. Or, if our audience were bilingual in Japanese, we could say that English /u/ is endolabial while Japanese /u/ is exolabial. However, we don't have this advantage in English, and therefore require more space to cover the phenomenon.


 * e. There is certainly confusion on terminology here. If you can quote the appropriate passages from Trask, I would appreciate it.


 * f. I will incorporate this into the article. [never mind—you just edited]


 * g. I thought I had taken care of inconsistencies in other WP articles. Please let me know which ones are still a problem. However, I imagine that admins will object if we start making notes citing mirrors of old versions of this article: that would make the entire WP project a mess, and probably violates RS or self-ref guidelines.


 * h. Ladefoged simply neglects this issue, but that can't be used as evidence: They barely cover the topic at all, and they neglect other things which should be in their book, such as several laryngeal articulations besides glottal and epiglottal. Catford is IMO acceptable sourcing. But it would be nice to find additional coverage.


 * i. The Swedish off-glide is considered an epiphenomenon of the type of rounding, not vice versa. (In dialects where the /u/ is central, and in Norwegian, the offglide is also central: that is, it's (to make up an IPA transcription), which is unattested as a phoneme in the world's languages; if we wish to attribute the difference of y and u to the offglide, then we need to posit  and  to the Swedish & Norse inventories, but only as codas or diphthongs.) Swedish is merely the language where this comes closest to being phonemically contrastive, but as an English speaker, you certainly need to know the diff to pronounce Japanese correctly. I have no idea about the Dutch; I'll delete it. [never mind—you just edited]


 * j. The asterisk is an IPA convention. It can be used for anything that does not have an IPA symbol. Ladefoged uses it for Korean "tense" consonants, for example. The other points I'll take care of when I get back. [never mind—you just edited] —kwami (talk) 17:59, 13 July 2009 (UTC)

It also looks as though endo- and exo- are reversed in the photo, given our current definitions. kwami (talk) 01:47, 14 July 2009 (UTC)


 * And to make things more complicated, studies have shown that some vowels can sound rounded without actually being rounded :) —Preceding unsigned comment added by 208.104.45.20 (talk) 19:28, 23 October 2010 (UTC)

It won't do!
1. The change (on 16th July) from "compressed vowels are rare" to "compressed vowels where protruded would be expected are rare" misrepresents the literature. The rarity of attested compressed vowels makes the protruded/compressed distinction unimportant, and that information ought not to be suppressed. I've reverted the edit.

2. Endo/exolabial are not the words used in the literature for this phenomenon, and Wikipedia's job is to mediate between between the layman and the academic accounts. I propose to change them to "protruded/compressed" pretty soon.

WellsTribute (talk) 23:55, 16 July 2009 (UTC)


 * See how it is now. While I agree with your point, we can't very well cite a source that they are ubiquitous, and then claim that they're rare. What is rare are exolabial back vowels and a contrast in front vowels; exolabial front vowels is an open question. From what I've seen in discussion groups, it would seem that German ü is more like Swedish and Norwegian u than like their y, and that Swedish teachers emphasize that Swedish y is not like German ü. (Of course, it doesn't necessarily follow that Swedish u is like German ü.)
 * This is probably a good move. [done]
 * Again, doesn't the photo look backwards? Both from the descriptions and the diagrams in Ladefoged? kwami (talk) 00:12, 17 July 2009 (UTC)

Thanks for making the change, Kwami.WellsTribute (talk) 17:29, 17 July 2009 (UTC)

Answering request
kwami asked me for my help about identifying the three vowels in the picture. From the top, they look like respectively. The latter could be, but I'm not too sure about that.

Peter Isotalo 12:40, 25 July 2009 (UTC)


 * Thanks, Peter. It would appear then that the endo- and exo- got reversed, as I suspected. We should then relabel the pic, perhaps with 'compressed' vs. 'protruded'. kwami (talk) 20:10, 25 July 2009 (UTC)

Roundedness as distinguishing for back vowels in English
The article only lists Vietnamese as a language for which roundedness is a distinguishing characteristic for back vowels. But isn't it one for English? 24.136.232.57 (talk) 04:51, 1 December 2009 (UTC)
 * Not as clearly. RP, for example, uses rounding to distinguish /ʌ/ from /ɔ/ and /ɒ/ from /ɑ/, but there is also a length distinction (the second vowel in each of those pairs being longer than the first). In General American, on the other hand, /ʌ/ is much farther forward than /ɔ/ (and some varieties of American English don't have /ɔ/ at all), and /ɒ/ and /ɑ/ have merged. +Angr 06:51, 1 December 2009 (UTC)

What dialects of English use vowel rounding?
I've been speaking English my whole life, and I don't think I've ever used vowel rounding in it. I can't really figure out the descriptions though. If some dialects use vowel rounding, then either it's an important split or its loss is an important merger. 96.231.17.131 (talk) 16:11, 5 November 2012 (UTC)

Why is this and every other article on phonetics on Wikipedia written to confuse?
Whenever I encounter an article referencing pronunciation on Wikipedia I end up on an endlessly recursive quest trying to understand the arcane explanations and descriptions, that are always defined in terms I do not understand which in turn are explained in the same obscure fashion. No-where, it seems, on the whole of Wikipedia is an accessible explanation of the basic principles and framework for understanding linguistics, phonetics and pronunciation that does not assume prior familiarity with the intimate details of the subject.

I don't struggle as much to understand articles on such abstractions as relativity, mathematics, art, music or even theology. It seems there are those who use language and those who study it, and the former must be kept from understanding why and how they say what they say so that the latter can study them.

In this article I need to understand labialisation, front vowels, back vowels and how to determine the height of a vowel just to understand the first introductory paragraph. And this is one of the less opaque articles on phonetics.

Quite simply, in my view linguistic articles on Wikipedia are consistently the most poorly explained; they are written as if for a phonetic textbook, not an encyclopaedia.

PLEASE! Can someone who does understand all this gibberish (and to me it is gibberish until I can find explanations that use plain English rather than being self-referential) tackle this and other pages with a linguistic bent and ensure that each one defines its core material in a way that doesn't need further research to understand it. 23:35, 30 December 2015 (UTC) — Preceding unsigned comment added by Stub Mandrel (talk • contribs)
 * It's not poorly explained, it's your just fault for being too lazy to click on the linked articles. It's impossible to paraphrase every basic technical term for laymen, it would make the article bloated. I never had help from anybody either, but the more I read, the clearer it became. There's nothing to complain about. --2.245.80.116 (talk) 19:39, 3 August 2016 (UTC)

Is "spread" always synonymous with "unrounded"?
I've always been perplexed at how English front vowels, at least when enunciated, are produced with noticeably different shapes of the lips compared to, say, [ɑ] or [ə], yet most texts only make distinction between rounded and unrounded. Sure enough, Cruttenden (2014:15) defines "spread" lip position as the lips being "held sufficiently far apart for no friction to be heard, yet remaining fairly close together and energetically spread ... taken up for vowels like that in see when energetically enunciated", and refers to the lips being "held in a relaxed position with a lowering of the lower jaw ... taken up for the vowel of sat" as "neutral". I think this should be incorporated into the article, but I don't know if I'm well enough equipped to do so myself right now and need more research. Nardog (talk) 14:52, 9 July 2017 (UTC)
 * Spread close front vowels are unrounded, dunno about the rest. I know this: the lower rounded vowels are, the less rounded they are in comparison with the close . Maybe something similar applies to unrounded vowels. I'll try to look for a source, I think Collins & Mees wrote something on that. Mr KEBAB (talk) 13:42, 11 July 2017 (UTC)
 * Thanks, I appreciate that. I'm also curious if Cruttenden means the same thing by "close rounded" and "open rounded" as endolabial and exolabial. On the surface, it seems like he's merely talking about the vowel height, but the same descriptions could very well be given for protrusion/compression. In other words, I wonder if there's such a thing as a compressed open vowel. I'm sure roundedness is always just a matter of degree and descriptions like protruded and compressed are arbitrary measures anyway, though (I also bet no language contrasts them either).
 * I also feel, at least in GA, sat suits his description of "spread" better than that of "neutral". This obviously might not be the case in modern RP, which is what the book mainly talks about. Nardog (talk) 14:31, 11 July 2017 (UTC)
 * Probably not. He describes French as 'close rounded' and English  as 'open rounded'. This is perfectly in line with what I said about the amount of rounding getting gradually lower as the vowel height decreases. I'd also like to add that  are probably an exception to that rule and typically have rounding that is more typical of the cardinal  (correct me if I'm wrong though, but this seems to be in alignment with what the SOWL says).


 * I'm not sure, but the protruded-compressed dichotomy may not really apply to rounded vowels lower than open-mid (among which only is in any way common). Starting with the close-mid height, it's not phonemically relevant in any language.


 * Protruded-compressed contrasts are very rare, but they exist. Swedish and Norwegian contrast close front spread, compressed and protruded vowels. This contrast is the clearest in Central Standard Swedish, in other dialects it may manifest differently (a recent change in Sweden is that are retracted to a central position and are pronounced with voiced alveolar friction, whereas  stays front). Danish and conservative Western Norwegian dialects don't have this contrast because they didn't undergo the  vowel shift. The protruded close front vowel in Swedish/Norwegian is IMO an attempt to maximize the  contrast, no matter how  is realized (central or front). Obviously, the situation is complicated by the fact that these sounds are commonly diphthongized (with various outcomes, depending on the region) and the fact that some Norwegian dialects merge  with  or  with  (short or both short and long). But when you say 'Norwegian', many people normally first think of Oslo Norwegian (or 'Urban East Norwegian'), in which  is a clearly maintained contrast.


 * That may be true. However, modern RP must be similar in that regard. What you're thinking about is maybe a more northern, which does have a slightly different lip shape since it's central (the open central unrounded vowel sounds to me like a retracted and raised cardinal , but that's another story I won't bore you with). Mr KEBAB (talk) 16:01, 11 July 2017 (UTC)

Apparently unrounded vowels can also be divided into two groups (spread and neutral) much like rounded ones, according to the IPA Handbook. So there are four classifications for the lip position of a vowel in total. Maybe this article should be moved to Lip position or something. Nardog (talk) 03:36, 1 August 2017 (UTC)
 * Maybe you're right, I'm not sure. 'Rounded and unrounded vowels' would also work, but as I said - I'm not sure whether this article needs to be moved at all. Mr KEBAB (talk) 16:27, 1 August 2017 (UTC)
 * Rounded and unrounded vowels might be a pretty good name actually. I've always thought the current name sort of fails to acknowledge that the article only concerns vowels, because "roundedness" can also refer to labialization (and possibly many other things in general). Nardog (talk) 17:36, 1 August 2017 (UTC)
 * AFAIK modern sources rarely speak of "rounded consonants", but you can encounter such terms, sure. In that case, I'd say I'm in favor of renaming the article. Mr KEBAB (talk) 18:27, 1 August 2017 (UTC)