Words in -og

Among Americans who distinguish cot and caught, some words that originally were pronounced with cot are now with caught. I'm interested knowing in which ones. From what I can tell, 'dog' is almost always /dO:g/ for non c-c merged Americans, but other words in -og are variable. How does everyone pronounce these words:
hog
frog
fog
smog
log
bog
clog
cog
flog
jog
Prauge
slog

or any others in -og that you can think of. Thanks.

Do you mean "Prague ... and are you sure that it even belongs here? What you're looking for is a subset of what might be called the CLOTH words.

http://en.wikipedia.org/wiki/Lexical_set#Standard_lexical_sets_for_English
http://en.wikipedia.org/wiki/Phonological_history_of_the_low_back_vowels#Lot-cloth_split

<<Do you mean "Prague ... and are you sure that it even belongs here?>>
Yes, sorry that was a typo. I pronounce it /prA:g/, but I'm c-c merged, so I was wondering if anyone pronounces it /prO:g/. I know in the UK it's pronounced /pr{g/.

I have [Q] in:

hog
frog
fog
smog
log
bot
flog

I have [a] in:

cog
jog

I have [A] or [Q] in:

clog
slog

I'm from Michigan. Here are my pronunciations:

"hog" /hAg/
/frAg/
/fAg/
/smAg/
/lAg/
/bag/
/clag/
/cAg/
/flag/
/dZag/
/prag/
/slag/

"clog" and "cog" were typos. They should be /klag/ and /kag/ and yes they rhyme.

Travis, do you have a 3-way distinction between Q, A, and a for these words? Also are you from an NCVS city? It seems weird for someone to have both [a] and [A] as possible realizations of the LOT lexical set.

There're all [O] (except for "Prague": [pr\6:g]) i.e. the same phoneme as RP's /Q/ but raised in my Aussie accent for me i.e. no LOT-CLOTH split.

>>Travis, do you have a 3-way distinction between Q, A, and a for these words? Also are you from an NCVS city? It seems weird for someone to have both [a] and [A] as possible realizations of the LOT lexical set. <<

I do speak an NCVS-affected dialect, but the thing is that my /a/ phoneme (GA /A/) has [A] as an allophone before /L\/ (GA /l/), free variation between [A] and [Q] as allophones after /L\/, and has free variation between [a] and [A] as allophones before /R/ (GA /r\/) not before a fortis obstruent (where then it is always [V]) or after /R/. This is probably due to /L\/ and /R/ normally being purely dorsal sonorants in my dialect and consequently coloring /a/ before or after it more than anything else.

<<Travis, do you have a 3-way distinction between Q, A, and a for these words? Also are you from an NCVS city? It seems weird for someone to have both [a] and [A] as possible realizations of the LOT lexical set.>>

I have the NCVS and have [A] for the "thought" vowel and [a] for the "lot" vowel.

Travis, where are you from? And Guest, I'm just curious, do you ever find that non-NCVS people confuse your PALM/LOT for their TRAP? (ie. you say 'stock' and they hear 'stack').

I'm from Milwaukee, Wisconsin. Even though the other question was directed towards Guest, in my case I do have to say that I haven't had any real problems with other individuals confusing my PALM/LOT with their TRAP, but I have had individuals confuse my TRAP with their DRESS, and I will sometimes confuse other individuals' TRAP with my PALM/LOT or other individuals' PALM/LOT with my THOUGHT, in particular with individuals under the influence of the California Vowel Shift.

Damn! Now there's a lot-cloth split?

Wow, confusing -_-. Why don't all you back-Easterners just become cot-caught merged like the rest of us? It simplifies things quite a bit.

<<Damn! Now there's a lot-cloth split?>>
Certain words that formerly had the LOT vowel have shifted to the THOUGHT vowel. It mostly only remains in the US. It occured before voiceless consonants in commonly spoken words. Many speakers also have it before -g, and in the words 'gone' and 'on'.