"dear" & "deer"

Here are my pronunciations for those words:

"bait/bate" [beIt]
"bet" [bEt]
"bear/bare" [bE@`]

"beet/beat" [bit]
"bit" [bIt]
"beer" [bI@`]

"matey" [meI4i]
"sat" [s{t]
"messy" [mEsi]
"Mary" [mE@`i]
"merry" [mEr\i]
"marry" [m{r\i]

Lax vowels like [I], [E], and [{] can never be followed by [r\] in the same syllable for me. Thus my intuitive syllabification of "serious" is [sI@`.i.@s] but my intuitive syllabification of "Sirius" is [sI.r\i.@s]. Likewise my intuitive syllabification for "Mary" is [mE@`.i], but for "merry" and "marry" it's [mE.r\i] and [m{.r\i].

"Standard" American English is actually pretty vaguely defined :) Also, even if it were, no one would speak completely "standard" anyway.

Yeah, you know what I meant, though.

<<That's interesting. Do you have two or three separate pronunciations for those?>>

I have a three-way distinction; [meri], [mEri] and [m{ri] for "Mary," "merry" and "marry," respectively.

<<Yeah, you know what I meant, though. >>

Actually, no I really didn't--that was my point. While many people may think their speech norms are "standard," that's not really the case. The case of pre-rhotic vowels is one of significant variation in the US, and even on national TV newscasts (arguably some of the most prestigious speech norms and registers) you'll hear obvious variation between different speakers (some are "Mary-merry-marry" merged, some aren't...going to other mergers, some are "cot-caught" merged, some aren't, etc.). There is no set defined "Standard American English" tho of course we tend to think of our speech norms and the dialects we're part of as what is "normal." A classic example of this may be found in the Northern Midwest, an area famous in sociolinguistics for having speakers who believe they speak completely "standard" (or "neutral") and "just like the newscasters on TV." However, any objective analysis and comparison shows that's not usually the case--the radical Northern Cities Vowel Shift has rendered such accents far different from most speech heard on TV. While this is an extreme example, much of the same could often be said for any region. For instance, as with any area, many people here in California think that Californian English is "neutral" or the norm/"standard," and while it is commonly heard on the media, many non-Californians would certainly beg to differ with the claim that Californian English is "neutral"--everyone's from somewhere.

The closest thing to what you're looking for is probably General American, which, while still somewhat vague, does have some clearly defined features and phonological phenomena.

Alright, "General American" then. Semantics. I know what you're talking about, though. I've read a lot about SAE and General American. I tend to refer to both as the same thing.

<<Alright, "General American" then. Semantics. I know what you're talking about, though. I've read a lot about SAE and General American. I tend to refer to both as the same thing.>>

Ok, well then going back to answer your question about the vowels in "standard" American English (and assuming that's General American), according to most definitions of General American, the vowels in "beer" and "bear" are lax. For instance, In detailing the phonology of General American Wikipedia shows [bIr\] for "beer" and [bEr\] for "bear."

http://en.wikipedia.org/wiki/International_Phonetic_Alphabet_for_English

<<I have a three-way distinction; [meri], [mEri] and [m{ri] for "Mary," "merry" and "marry," respectively.>>

Interesting--where are you from?

One thing that which is rather weird is that, while my dialect is "Mary"-"merry"-"marry"-merged, it not only merges such as /e/, but also in very informal speech in actual realization elision effects has made it so that all of [e], [E], [{] and [e:], [E:], and [{:] may be found before [r\], due to:

CVKX -> CVY

where X and Y are either /@`/ and [r\] or /l=/ and [5] respectively, where K is one of /t/, /d/, or /v/, and where, most importantly, the vowel in question is realized as if the elided consonant were still present, thus allowing it to be realized with an arbitrary vowel length and to break the pre-rhotic vowel phoneme restrictions otherwise present.

This allows, in theory at least:

[si:r\] : "seeder", from /"sid@`/
[mir\] : "meter", from /"mit@`/
[sIr\] : "sitter", from /"sIt@`/
[ber\] : "baiter", from /"bet@`/
[hE:r\] : "header", from /"hEd@`/
[bEr\] : "better", from /"bEt@`/
[b{:r\] : "badder", from /"b{d@`/]
[b{r\] : "batter", from /"b{t@`/
[5ur\], "looter", from /"lut@`/
[fUr\] : "footer", from /"fUt@`/
[bor\] : "boater", from /"bot@`/
[5O:r\] : "lauder", from /"lOd@`/
[mAr\] : "modder", from /"mAd@`/

I couldn't think of words that could produce [U:r\] or [Or\] via such or normally, so hence I did not list such above.

This is one of the few elements of my own dialect that I do tend to try to deliberately downplay when in doubt, simply because it often confuses people who are not familiar with hearing it; many, especially non-native speakers, don't expect [5Er\] to be "letter" and [nE:r\] to be "never", for instance. Note though that it is not consistent at all in the first place and seems to vary in usage and frequency from speaker to speaker, and from word to word (for instances, common shorter words like "never" seem to be most often affected); however, my own speech seems to be more progressive in this regard than that of many here.

I'm from a suburb of Philadelphia.

Mine (northern England) are phonologically similar to Lazar's, but phonetically a bit different:

"bait/bate" [be:t]
"bet" [bEt]
"bear/bare" [b{:(r\)]
"beet/beat" [bit]
"bit" [bIt]
"beer" [bI@(r\)]
"matey" [me:tI]
"sat" [sat]
"messy" [mEsI]
"Mary" [m{:r\I]
"merry" [mEr\I]
"marry" [mar\I]
"Sirius" [sIr/i@s]
"serious" [sI:r/i@s]

The /r/ sounds in the last few all seem the same to me - the difference is in the preceding vowels.

<<I'm from a suburb of Philadelphia.>>

Oh ok--that does make sense as that is "Mary-marry-merry" non-merged territory. Does anyone there merge them?

Travis, your transcriptions are interesting--as far as I can tell I don't have such consonant deletion even in my most informal speech.

Thanks for your transcriptions, JHJ--very interesting! :)

<<Travis, your transcriptions are interesting--as far as I can tell I don't have such consonant deletion even in my most informal speech.>>

Same goes for me - I would never delete /t/, /d/, or /v/ in those positions.

>><<Travis, your transcriptions are interesting--as far as I can tell I don't have such consonant deletion even in my most informal speech.>>

Same goes for me - I would never delete /t/, /d/, or /v/ in those positions.<<

One note is that there is an intermediate stage where there is the deletion of them, but *not* the devocalization/desyllabicization of the following /@`/ or /l=/, which are still realized as [@`] and [5=] respectively. Furthermore, it involves the *replacement* of the deleted phoneme with [j] or [w], if what preceds it is a mid or high tense vowel which is front or back respectively. Likewise, this may happen at points where /t/, /d/, or /v/ are in intervocalic positions where *any* unstressed vowel may be following them, which may result in [j]/[w] epethensis *or* diphthongalization of the two newly adjacent vowels.

And why is all of this happening? My guess is that an intermediate phonological value of [4] or [v] is probably "vowel-like" enough to be easily elided in very informal speech where plenty of other elision is already occurring. However, at the same epethensis with [j] or [w] (which one may consider to be simply having an increased degree of sonority and being potentially rounded by a preceding vowel rather than being elision proper) or diphthongalization may result when the adjacent vowels come into contact with each other, *or* compensatory "hardening" of a following rhotic vowel or syllabic consonant may occur; of course, because this is happening phonologically and is not expressed at the phonemic level, its realized results may effectively break restrictions which are imposed on phonemes proper with respect to their allowed positions.

One thing though that must be clarified about such is that such in practice generally only happens frequently to phonological words which are unstressed, are at the morphological level either "short" words or have the host word (in phonological constructions containing clitics) being "short" words; consequently, phonological words like "never", "it'll", and "that'll" tend to be affected far more heavily than non-"short" phonemically two-syllable phonological words.