''pane'' and ''pain''

<<In much of the U.S., the /e/ isn't even diphthonged; it's a pure vowel.

American English has largely dropped the /eI/ diphthong and has dropped many other non-phonemic diphthongs as well.>>

Very true. I consistently produce /e/ and /o/ as monophthongs, especially when they're followed by consonants, so I should say in closed position. I've noticed that in my speech, in words like "hate" or "pane" or "boat" etc the vowels are exclusively monophthongal, while in open position they may either be monophthongal or may have a *very* slight diphthongal offglide (it's hard for me even to hear but it can be there). So, for me:

"sake" [sek]
"say" [se] or maybe [seI] in salient, stressed position in the utterance
"dote" [dot]
"doe" [do] or maybe [doU] in salient, stressed position in the utterance

Of course, this still doesn't mean my [e] or [o] sound identical to their "continental" equivalents, for example. Especially my [o], while being monophthongal, isn't cardinal [o] because it isn't very rounded and is somewhat fronted or centralized compared to its normal status as a back vowel. This is normal for my dialect.

Where I always do retain diphthongs are in the true, full, diphthongs in most varieties of English: /aU/ /OI/ and /aI/.

''I say all those word-pairs the same.''

<<I assume that I'm a minority in distinguishing ''pane'' and ''pain'', right?>>

''Yes. I think distinguishing between /e/ and /{I/ is unique to Wales, and maybe some parts of Scotland''

Lazar,

What about ''wine'' and ''whine''? Do you pronounce them the same?

For me they're different:

wine - /waIn/

whine - /WaIn/

Is distinguishing ''wine'' and ''whine'' as uncommon or less uncommon than distinguishing ''pane'' and ''pain''?

Sorry, Lazar. That was me that posted above. I accidentally typed your name.

<<What about ''wine'' and ''whine''? Do you pronounce them the same?>>

Yes, I pronounce them the same.

<<Is distinguishing ''wine'' and ''whine'' as uncommon or less uncommon than distinguishing ''pane'' and ''pain''?>>

Distinguishing between "wine" and "whine" is much more common in the US - there's actually a substantial minority of Americans who do so. I think the distinction is less common in the UK, again being limited to Wales and Scotland.

Mxsmanic, NAE has most definitely lost *phonemic* vowel length, but at least in the dialect here, there are clearly differences in vowel lengths, but these are primarily based on stress placement, syllabic structure, and whether there is a voiced consonant following a given vowel. Stress aside, a voiced consonant tends to lengthen a preceding vowel, and in particular lengthens it if it is in the coda of the syllable the vowel is in (but seems to still lengthen it if it is in the head of an immediately following syllable).

The interesting thing though is that consonants which are phonemically voiced do not seem to lose this effect even if they are devoiced, usually by assimilation another following unvoiced consonant, at least in the dialect spoken here in southeastern Wisconsin. The interesting effect of this is that if this devoicing becomes phonemic in the near future, it would also have the effect of reintroducing phonemic vowel length, albeit a different pattern of phonemic vowel length than that which existed historically in English.

For example, the spoken informal modal form "hadta" (/"h{dt@/) is distinguished in speech from a hypothetical word *"hatta" (*/"h{t@/) only by vowel length and because the /dt/ in "hadta" acts as a sort of geminate. This is because "hadta" is realized as ["h{:t_}.t_h@] (you could possibly write that as ["h{:.t_h:@], whereas *"hatta" would be realized as *["h{.4@] normally, and *[h{.t_h@] if stressed). If anything, the most important thing that distinguishes these two is vowel length; if I pronounce "hadta" with a short vowel, it just does not sound right, period, even though it is pronounced generally with just an unvoiced second consonant. Hence, it seems like these two forms are being distinguished in actual speech by primarily vowel length, even though the vowel length difference itself is not phonemic proper. Were the devoicing in this case to become phonemic, vowel length would also have to become phonemic as well.

Actually, now that I think of it, such becoming phonemic would require not only the introduction of phonemic vowel length, but also the introduction of phonemic *consonant* length, at least for plosives.

As for me, I pronounce ''pain'' and ''pane'' the same way /pen/.