Caw-fee

<<Kirk and Lazar,

Here's a map about the cot-caught merger http://www.ling.upenn.edu/phono_atlas/maps/Map1.html

The red dots show areas where the merger occurs and the blue dots have the distinction.>>

Yes, that's a pretty useful map. I've seen it before but thanks for linking it :) Clearly, the strongest non-merging areas are the Northern Cities, significant parts of the South, and the coastal ribbon from NYC (and a bit north) to Washington, DC.

<<Kah-fee sounds kinda weird.>>

Not to us 40-50% of Americans who say it that way ;)

<<Yes, "caw-fee" is the pronunciation that about 50-60% of Americans use. But there can be subtle regional variations in the sound represented by "aw" - a c-c unmerged "coffee" might come out, for example, as [kOfi] or [kQfi].>>

Very true. Good point. Also, in the NCVS, "aw" is [A], even tho it's not merged with the sound in "top," which is a fronted [a].

Personally, I have something in between [A] and [O] for "coffee" and I'm "cot-caught" merged.

its cawt,cawfee,bawt,fawt, ..... haht,naht,swaht,....EU-gene,ore-GON,to me...sorry folks.

It's LAW [la], TALK [tak], DAWN [dan] but DOLLAR [dol@r] for me
I'm from CA

<<I don't say cawffee, cawpy :p
I believe coffee (cawffee) is some kind of regional pronunciation (East&South) [General American normally excludes these regions :))>>

<<Actually, Arizona HQ, probably more than half the country does not merge "cot-caught" and those who don't pronounce then say "coffee" with a different vowel than "cot.">>

That's interesting, Kirk. For me, ''coffee'' doesn't share the same vowel as in ''cot'' /kQt/ or ''caught'' /kOt/. ''coffee'' has the vowel in ''cloth'', ''cost'' and ''off'' for me i.e. /kQ:fi/.

>>Very true. Good point. Also, in the NCVS, "aw" is [A], even tho it's not merged with the sound in "top," which is a fronted [a]. <<

Depends on not only the degree of the NCVS, but also the subpattern of it in question; for example, while I do have a moderate degree of NCVS in my own speech, /O/ is left unaffected by it, and /A/ is not that strongly shifted forward towards [a] in it either.

The Eastern New England dialect often uses [a] for its "father-star" phoneme.

<<The Eastern New England dialect often uses [a] for its "father-star" phoneme.>>

Yeah like the classic Bostonian [k_ha:] for "car." :)

<<Depends on not only the degree of the NCVS, but also the subpattern of it in question; for example, while I do have a moderate degree of NCVS in my own speech, /O/ is left unaffected by it, and /A/ is not that strongly shifted forward towards [a] in it either.>>

Yeah, that's true. I was thinking of full chain-shift NCVS speech, but I actually did think of your speech when I wrote that because I knew your /O/ was still more /O/ than /A/. In some of your audio recordings I've heard before your /A/ sounds fronted towards the direction of [a] to me, but not to the extreme I've heard in the most progressive NCVSers.

<<That's interesting, Kirk. For me, ''coffee'' doesn't share the same vowel as in ''cot'' /kQt/ or ''caught'' /kOt/. ''coffee'' has the vowel in ''cloth'', ''cost'' and ''off'' for me i.e. /kQ:fi/.>>

Yeah, that makes sense, as you're not a "cot-caught" merger.

<<It's LAW [la], TALK [tak], DAWN [dan] but DOLLAR [dol@r] for me
I'm from CA>>

CA as in Canada or California? Anyway, the influence of postvocalic /l/ --> [5] does have the effect of turning /A/ --> [O] for some speakers, so that's not too surprising.

>>CA as in Canada or California? Anyway, the influence of postvocalic /l/ --> [5] does have the effect of turning /A/ --> [O] for some speakers, so that's not too surprising.<<

At least in my dialect /A/ simply does not exist before /l/ except when said /l/ is intervocalic; all in cases where /A/ would have been present historically in such positions /O/ is used in my dialect. Note though that this isn't simply the influence of realizing /l/ as [5], as *most* instances of intervocalic /l/ are realized as [5] in my dialect, with the few instances of such which are realized as [l] practically always being between two front vowels (usually high front vowels at that), and most often (but not always) having the preceding front vowel also being preceded in turn by an alveolar consonant.

That's interesting, as I believe I always have [A5], and never [O5], no matter the position (whether intervocalic or final).

<<That's interesting, Kirk. For me, ''coffee'' doesn't share the same vowel as in ''cot'' /kQt/ or ''caught'' /kOt/. ''coffee'' has the vowel in ''cloth'', ''cost'' and ''off'' for me i.e. /kQ:fi/.>>

<<Yeah, that makes sense, as you're not a "cot-caught" merger.>>

I don't think you really understood me. For me, ''coffee'' does not share the same vowel as either ''cot'' or ''caught'', but has the vowel in ''cloth'', ''soft'', and ''gone'' i.e. /kQ:fi/. Thus, ''coffee'' is not */kQfi/ for me, but nor is it ''cawfee'' */kOfi/.

<<At least in my dialect /A/ simply does not exist before /l/ except when said /l/ is intervocalic; all in cases where /A/ would have been present historically in such positions /O/ is used in my dialect.>>

Kirk and Travis,

For me, the sequence /Al/ appears to be quite rare mostly occurring in foreign loan words. I'm from Northern England.

<<The Eastern New England dialect often uses [a] for its "father-star" phoneme.>>

I myself have /a/ for my ''cat'' phoneme which is quite common here in Northern England.

<<and they get mad when I say Q-pon instead of cooopon (coupon)>>

As for ''coupon'', I myself say /kupQn/ for that. I never heard Q-pon before.

<<I don't think you really understood me. For me, ''coffee'' does not share the same vowel as either ''cot'' or ''caught'', but has the vowel in ''cloth'', ''soft'', and ''gone'' i.e. /kQ:fi/. Thus, ''coffee'' is not */kQfi/ for me, but nor is it ''cawfee'' */kOfi/. >>

So you have three phonemes, with "don" /dQn/, "dawn" /dOn/, "gone" /gQ:n/ - is that right?

Where in northern England are you from? I'm also from northern England and just have /Q/ (the "cot" vowel) in "coffee", ''cloth'', ''soft'', and ''gone''.

I guess fawdy p'cen a'Americans say i' wei'd