Talk:Close-mid front unrounded vowel

From Wikipedia, the free encyclopedia

This article is part of WikiProject Phonetics, an attempt to build a comprehensive and detailed guide to phonetics and descriptive phonology on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks.

I am trying to pronounce this [e] sound accurately, but with my Canadian English dialect, I always seem to end up gliding it. I am trying to hold my tongue and jaw as still as possible when pronouncing it, but a glide always seems to make it into my pronunciation. Is there a technique I need to know to keep my mouth very still so as to pronounce [e] accurately?  Denelson83  00:04, 7 August 2005 (UTC)

[edit] Close-mid vs mid

It makes no sense. How can English have both close-mid and mid in the same words? Someone needs to fix it and considering I'd never heard about the mid front unrounded vowel before reading this page, I don't feel like the prime candidate for it. AEuSoes1 19:22, 6 December 2005 (UTC)

Different dialects have different pronunciations. kwami 22:49, 6 December 2005 (UTC)

That makes sense but it says RP has both in bed (and the page for the open-mid E also has bed as an example for GA). Is there any source for the mid vowel stuff in this and other articles? I think that might be necessaryAEuSoes1 00:04, 7 December 2005 (UTC)

No, I think you're misreading it. It says that Australian English has close-mid [e] in "bed" [bed], RP has it as the first component to the diphthong "laid" [leɪd], and RP uses a mid [e̞] in "bed" [be̞d]. You can draw up a table of realisations/correspondences thus (this uses the relevant info from this page and Open-mid front unrounded vowel):

Relisations of some front mid–like vowels in selected English dialects

	"led"	"laid"*
Closer than close-mid	NZE
Close-mid	AusE	RP, CanE
Mid	RP	GAmE
Open-mid	GAmE

* When a diphthong, the first element is denoted. NZE=New Zealand English, AusE=Australian English, RP=Received Pronunciation, CanE=Canadian English, GAmE=General American

I hope this clarifies things a bit, but it doesn't say anything the articles don't already.

—Felix the Cassowary (ɑe hɪː jɐ) 01:30, 7 December 2005 (UTC)

Ahh, you are right. I misread it. But I'd still like to see a source for some of that, especially the statement "the phoneme variously transcribed as /e/ or /ɛ/ is [e̞]." AEuSoes1 01:42, 7 December 2005 (UTC)

Which part, the phonetic realisation of the phoneme variously transcribed as /e/ or /ɛ/, or the fact that there are various transcriptions of the phoneme? If the latter, you merely need to contrast Received_Pronunciation#Vowels with this page by J. C. Wells. If the former, just about anything which discusses RP phonetics should be good enough... —Felix the Cassowary (ɑe hɪː jɐ) 11:59, 7 December 2005 (UTC)

I meant the latter part. According to Roca and Johnson in “A course in Phonology” (p 179) they set up “late” and “let” as minimal pairs so that the a in “late” is:

• [e] in Scottish English
• perhaps a little lower in Yorkshire accents: [e̞]
• diphthongized usually in GA, and generally in RP, where it is also a little lower ([e̞])
• lowered further (in addition to diphthongizing) in Cockney and in the southern hemisphere

the vowel in “let” is:

• usually [ɛ] in Scottish English and Yorkshire English
• slightly higher in GA and RP: [ɛ ]
• even higher in older style RP, but not as high as [e]: [e̞]
• diphthongized to various degrees in the American south.

So according to this source, a lowered /e/ and a raised /ɛ/ are not the same thing except in older style RP. We can also see in this Korean phonology vowel chart

But this same source explains that [e] and [ɛ] are cardinal vowels and that they don’t necessarily correspond to the real vowels of any natural language (p 126). My point is that it seems that having an /e/ or /ɛ/ that happens to be in the middle between the two is less significant than the article makes it out to be. I notice that for the example of NZE the diacritic is simply put in there without making a subsection on near-near close front unrounded vowels. I don’t think we need to take out mention of the mid-central vowel, but… well I’ll change the article to something closer to what I think it should be. If it turns out to be crap it can always be reverted and if it turns out great then I can extend it to the other vowel articles that mention the mid-vowel AEuSoes1 21:37, 7 December 2005 (UTC)

The two references to RP still strike me as confusing. What you've quoted above suggests that the [e̞] transcription only makes sense for old-fashioned RP, and that there isn't really much difference between the RP and GA vowels. I'd suggest removing RP bed from this page.--JHJ 17:35, 8 December 2005 (UTC)

All right, apparantly Kwami didn't like my edits. Considering he's the one who put the mid vowel stuff in there in the first place I suppose he's got a reason. I wonder what it is...

AEuSoes1 23:50, 8 December 2005 (UTC)

Actually, I'm not sure what all your edits were. I kept the new ref, but you may have made other improvements in detail that I didn't see in the massive reorganization. (There was just a mass of red in the comparison view.) What I didn't like was conflating the close-mid and mid vowels, which are clearly distinct. This distinction is made in all six of the close-mid vowel articles, not just here. The mids could just as easily go in the open-mid article; they're here only because of tradition, not due to phonetics. kwami 01:38, 9 December 2005 (UTC)

Are they really that clearly distinct, given that the IPA doesn't provide different symbols and no language distinguishes them (according to the article)? I don't have a particularly strong opinion, but Aeusoes1 did seem to me to have a point. (I've re-removed the reference to RP bed, based on the discussion of let and late quoted from the source above, which suggests that there isn't much difference between RP and GA on this vowel.)--JHJ 08:46, 9 December 2005 (UTC)

I dunno. They are distinct enough that speakers of a language with a mid vowel realize that it's a different sound from a close-mid vowel. I can hear the difference, and neither occur as monophthongs in my dialect. Rather like the difference between [a] and [æ]: I doubt many languages distinguish them either, but you certainly know it if someone uses the wrong one. If we were talking about more marginal sounds, it wouldn't matter much, but lots of languages are described as having 'mid vowels' which may actually be close-mid or open-mid, and I think it's nice to clarify the situation a bit, so that people don't identify the 'mid' vowel in a target language with the 'mid' vowel in a language they already know. kwami 10:16, 9 December 2005 (UTC)

I generally agree that we should have both close-mid and mid separately covered. I think your second point is very close... I think basically it's that mid vowels are discussed often enough that they're encyclopaedic for that reason. Someone looking for stuff on mid-vowels should receive an adequate treatement on them, how they differ from and are the same as close-mid and open-mid vowels etc. As to the fact that they're distinct enough that you can hear it's a different vowel ... the Kiwi vowel denoted here as "[be̝d]" is noticeably distinct from my own AuE [bed], it's close enough in fact to my /ɪ/ "thet ef I was umutatung a Kiwi eccint, thin" I'd normally use my /ɪ/... That doesn't warrant it getting its own section, though. —Felix the Cassowary (ɑe hɪː jɐ) 12:42, 9 December 2005 (UTC)

I realized after the edit that it would appear as a mass of red. The main difference that I recall I made other than merging the two sections was that I said that the difference between mid and close-mid "may be" distinct to a speakers.

But I find it strange for Kwami to say "This distinction is made in all six of the close-mid vowel articles, not just here" when he himself is the one that included the distinction in all six articles around September 9th. I think that for Kwami, we need a bit more from an argument for keeping them separate in the articles than that he's separated them (also I planned to extend the merging to the other articles once I got the consensus kinks out). I think that if someone is looking for mid vowels, the fact that a search for "mid front unrounded" brings one to this page is good enough. I wanted to merge them for the very reason that JHJ brings up. If they're not important enough for the IPA to make a separate symbol then I would venture to say that Kwami ought to provide for some source that rationalizes his subsectioning. AEuSoes1 02:55, 10 December 2005 (UTC)

(moving discussion back from my talk page -kwami)

[...] there is no objective criterion as to what constitutes a separate vowel cross-linguistically. This is merely an issue of how we wish to present the subject. For me and at least one other editor, it makes sense to separate mid, close-mid, and open-mid vowels (all arbitrary categories) rather than conflating two of them. This is simply a case of aligning our description with the terminology used in the literature: since the IPA classification categorizes open, near-open, open-mid, mid, close-mid, near-close, and close vowels, we feel that our articles should cover the subjects of open, near-open, open-mid, mid, close-mid, near-close, and close vowels. kwami 21:50, 10 December 2005 (UTC)

I'm sorry it's taken me so long to respond. I've been having internet troubles. Anyway, you're right that there is no objective cross-linguistic criterion of what constitutes a vowel but we do have the IPA, which does not distinguish between close-mid, mid, and open-mid. There is a mid-central category for schwa but any IPA talk of "mid" is either close-mid, open-mid or both. For me and at least one other editor, it doesn't make sense to separate them the way they are separated. I really don't know why you separated them in the first place, especially because doing so makes it seem as though linguists generally agree to the "mid" or mid-central vowel distinction and that is not something that linguists agree on. If you've got some source that motivated you to do so, I think you ought to provide it as a reference.

AEuSoes1 00:37, 16 December 2005 (UTC)

As for me objecting that other articles make the distinction, my only point was that they should be consistant, not that they somehow proved my POV. Either mid and close-mid should be conflated in all articles or they should be distinct in all articles.

I disagree that IPA talk of "mid" is 'either close-mid, open-mid or both'. Sometimes it's just mid. The basic point is that the IPA has a label "mid". People will expect "mid vowels" to be mid vowels, and separating them out is a way of clarifying when they are and when they aren't. I don't follow what you mean by "linguists" not agreeing on a "mid" vowel distinction - any linguist will tell you that vowel height is a continuum phonetically, though structural phonologists may debate how many heights there are in a featural system. It's true that the IPA has set up cardinal vowel positions, but this is just to define the grid. Over a third of IPA vowel symbols don't occupy cardinal positions.

Per SOWL, Danish for example is a language which contrasts close-mid and mid vowels. However, it doesn't have open-mid vowels, so they're transcribed /e ε/. That means that the /ε/ in Danish is similar, and perhaps closer, than the /e/ in Croatian, which is mid or on the open side of mid.

Amstetten Bavarian is a language which distinguishes close-mid, mid, and open-mid vowels, or something very close to them. However, the symbols for the near-open and open vowels are available for the open-mids (/æ ɶ ɑ/ are a third of the way between /a/ and /i u/, which makes them cardinal open-mids!), which means that the open-mid symbols can now be used for the mid vowels. So when we do get a language which contrasts the three mid vowel heights, that's hidden by using symbols for other heights to represent them. (Actually, /ε œ ɔ/ may be closer to close-mid, with /e ø o/ even higher, but you get my point.)

We can certainly make it clearer that the IPA does not enshrine mid vowels with dedicated symbols. And if the phrase "mid vowel" weren't so commonly used, I'd have no trouble lumping them together as you'd like. However, symbols are commonly used in whatever way is convenient for setting up phonemic contrasts, without paying much attention to how they actually sound. For example, I just came across a representation of voiceless [j] as /ç/. Presumably that's because it's hard to add an under-ring to <j>, but we'd certainly want to make it clear in this case that /ç/ is not a fricative. Given that <e ø o> are commonly used for mid or open-mid vowels, and that they may be called mid vowels even when they're close-mid, I think we'll prevent a lot of confusion by separating out the examples of true mid vowels. Again, if people didn't refer to 'mid vowels', there'd be no need to do this. kwami 01:59, 16 December 2005 (UTC)

All right, further examination of the IPA chart does indeed classify schwa as "mid" (the article on mid vowels classified it as "mid-central" sucks to me for believing a Wikipedia article) but what I mean about linguists not considering "mid" a vowel I mean that in the front and in the back they do not considered even "true" mid vowels as separate vowels but as realizations of more cardinal vowels. So your Danish example would be considered to contrast between open-mid and close-mid but the open-mid vowel would be raised. The source that I put in the article has two different vowel realizations in English that generally fall in the "mid" range between open mid and close mid but they are considered phonetically different and realizations of both open-mid or close-mid. My changes that were reverted still kept in the explanation of "true" mid vowels and the diacritic that indicates them. I don't think readers will be too confused that way. AEuSoes1 03:18, 16 December 2005 (UTC)

AEuSoes1 03:18, 16 December 2005 (UTC)

[edit] Hungarian example seems wrong

The hungarian example of "hét" seems wrong. Its pronounciation is closer to /ɪ/; it certainly is a very different vowel from the other examples given, and it's definitely not a 'front vowel'.

The Hungarian article in the IPA Handbook (by Tamás Szende of the Hungarian Academy of Sciences) has é as a mid front vowel, very different from what you're describing. kwami 05:54, 12 December 2005 (UTC)

I dunno, I'm just a native speaker of Hungarian and Dutch (the first example word). The only thing I know is that the Dutch example of 'één' has a starting sound that is very different from the Hungarian é in 'hét', so maybe the Dutch example is misqualified. In pronouncing the two words, there is definitely a difference in backness, with the Dutch example being more to the front.

But again, I'm no expert. Are there any online sources for the vowel phonology charts for different languages? --128.32.198.103 18:30, 12 December 2005 (UTC)

Wikipedia's got some stuff... Compare the first image (Hungarian) with the second (Dutch). Seems to suggest that they have roughly the same frontness, but the Dutch is substantially closer; the Hungarian vowel actually appears to be [ɛː] (with Hungarian /ɛ/ being [æ]). —Felix the Cassowary (ɑe hɪː jɐ) 01:39, 13 December 2005 (UTC)

The Hungarian vowel is plotted right where Dutch [ɛː] is. However, we have to be careful comparing charts: we don't know different the scales are, or whether there were gender differences in the subjects who were measured. But going by the chart, it certainly seems that Magyar is mid to open-mid. kwami 02:03, 13 December 2005 (UTC)

Are these vowel distributions measured off of the sonic spectra from a single person? --Sander Pronk 18:17, 13 December 2005 (UTC)

The Hungarian is a man in his 50s with an academic background. The Dutch article doesn't say. They should be averaged values, of course, but not everyone goes to that much trouble. kwami 19:34, 13 December 2005 (UTC)