And remember that the written word is derivative from the
spoken language; lots of things have changed since the basics
of our orthography were set.
There are two reasons to hew to the standard tongue: mutual
easy intelligibility and petty-minded judgmental people. A
wide variety of American dialects allow the former; only one,
as codified by national news broadcasters and elementary
school teachers in the 1950s for use in the Ozzie and Harriet
world they thought existed satisfies the latter. I stand
corrected: "... satisfies the latter, if we take into
account dialectal idiosyncrasies of those doing the
judging." Few condemn their own dialect if they consider
it to bear prestige; they usually are unaware of the numerous
differences between their dialect and the standard, or the
areas in which the standard was never actually defined so they
fill in their dialectal traits by default and consider them to
be standard.
I don't know about Blitzer's pronunciation of
"Democrats'. He's from Buffalo (the great wiki tells
me--Tonawanda or thereabouts, specifically), so has vestiges
of the Northern Cities Vowel Shift (NCVS) no matter how he's
endeavored to extirpate them. I've heard the dialect. The
phonologist at U. Rochester considered it goofy (not her word
for it, I don't think), but she was from Massachusetts and
lacked rhotic vowels, so her subjective opinion has to be
taken with a few shakes of the ol' saltshaker.
There are two possibilities that come to mind to account for
your perceptions. I'll give the less-likely one first. The
second requires more background. [] is mixed phonology and
phonetics, not good practice but this needs no more help in
achieving great heights of pretentiousness; < > refers
to graphemes.
In the NCVS sound [I] shifts, but can't go up to [i] or down
to [e] without merging with them, so it moves back towards
schwa. The dialect avoids the pin ~ pen merger that so
bothered me in the Pacific Northwest (as well as the much more
common and therefore more confusing cot ~ caught merger). The
easy answer is that language change is messy, and to just
claim that perhaps the [E] in "democrat" has
undergone the same shift as the [I] in "pin".
"Easy" is usually wrong, but not implausible. It's
a stressed [E], which means it has to have a certain
"inherent" length in English; but trisyllabic words
have the first syllable shortened, which means the [E] would
start being perceived as [I], especially when followed by a
nasal. If it's perceived as [I], it can easily be an [I] for
vowel-shifting purposes, which means you'd get
"duhmocrats" (or "dihmocrats"). Slightly
clever an explanation, but still possibly wrong. It's growing
on me as I type, but I still don't quite buy it.
The harder explanation involves some background. Keep in mind
that vowels differ slightly in every token of a word--say
"dad" 20 times and look at the spectrogram, and
they'll all differ a bit. Take a set of words like "bed,
bet, beg, Beck, Bess", record them, and edit out the
final consonant. If the words are in a sentence, people might
not even notice that the final consonant's missing. If
they're not in a sentence and the speaker has the same dialect
as you, you'll be right 95% or more of the time in identifying
the words, with slight hesitation at most. I know; I've seen
it done and had the experiment done on me (being unfamiliar
with the speaker's dialect degrades performance sharply).
Vowels before a voiced consonant like "d" are longer
than before a voiceless one like "t"; they have
distinct features depending upon whether the consonant after
them is a velar ("k") or a labial ("p").
In some forms of AAEV, final consonants are dropped in some
words with no loss of intelligibility because the vowel length
is retained. We're sensitive to information that's there,
very much so, even if it's seemingly redundant; we're not so
sensitive to redundant information that's left out, even if we
usually rely on it. In one study looking at word pairs like
ham ~ hamster (lamb ~ lamppost, etc.), people decided which
was being uttered well before getting to the "m".
However, while every individual vowel is different, we're not
aware that they're different, at least where our native
dialect is concerned. Call it "categorical
perception", and ignore the controversy as to whether it
really exists: we "hear" vowels only after we've
"categorized" them, and vowels that don't categorize
nicely stick out (and we often mock them).
Imagine a trapezoidal vowel space with four corners:
"ee", "ae" ("ask"),
"aw" (low and back) and "u". Every vowel
is between those four--the <u> in "but" stands
for something higher than [a], more front in the mouth than
[o], more back than [E]. If I were to record myself saying
different words with the various English vowels and map them
(there's software for helping out), you'd see that very rarely
does one of my [E] come out as high as any of my [e] or
anything like my [ae]; my vowels are fairly distinct, with
some weirdness for [a], schwa, and the [U] in
"wood". (I've been a lab subject for enough student
projects to know this, and to know everybody has some vowels
that overlap a little). I can draw nice lines around all the
[E] and identify an [E] region, and do the same for an [ae]
region, etc., etc. Every realization of a vowel in a given
region is equivalent, for most purposes. If you did it, your
vowel chart would come out about the same, unless you have
some mergers or splits in your dialect. And if I had people
where I grew up have their vowels mapped, and people that
speak like you map their vowels, we'd see that my peers' vowel
mappings look almost identical to mine, and your peers' are
nearly identical to yours. We each share a dialect with our
peers, in other words. Now, if you instrumentally produce a
series of a 100 vowels from [i] to [ae], going from
"beet" to "bit" to "bait" to
"bet" to "bat", you could fairly reliably
categorize every one of the 100 vowel sounds. You'd find that
every vowel that you identified as an [i] just happened to
fall in your [i] ("ee") region, and the same for
[I], [e], [E], and [ae]. But our regions wouldn't match up
perfectly. I bet we'd disagree on at least 10-15 of the vowel
sounds. I also bet that you and I wouldn't hear any
difference in our regions until we looked at the vowel space
mappings, because we'd be using context and meaning to help
determine what the slightly not-quite-right vowels are; I
could be wrong, there are those who hear and are aware of
fine-grained differences. Usually the more you hear people
with other dialects, the less sensitive you become to hearing
differences, unless something makes the difference salient,
but sometimes it works the other way.
Some people here seem to think of the word
"Democrat" as some sort of icon and, when normal
English-language processes act upon it, it must be defended in
the name of ideo-linguistic purity lest it be besmirched and
violated (and need an abortion). There are also words and
sounds that serve social functions, serving to identify social
groups that are looked down upon. (But that's being
redundant, isn't it? "Democratic" and its forms are
shibboleths on DU). These markers are very salient, and we're
often sensitive to them.
One more bit of useless but important trivia. Some Americans
have two schwas, some (most?) have only one. I have two,
call them "uh" and "ih", and we're all
pretty much the same; we don't have any wriggle room with our
schwas. Those with only one schwa fall into two groups:
Those that usually pronounce it "uh", but can allow
it to drift towards "ih" in some words, and those
that do not ever allow it to move forward towards
"ih". When I speak, people with only one schwa have
a choice: they can hear both my schwas as "uh", or
they can hear me saying schwa when I use a back schwa, and [I]
(as in "bit") when I use a front vowel. If they go
the latter route, my dialect is hard to understand, and
usually those people maintain a consistently back schwa. They,
fortunately, are rare. They usually consider only their
dialect to be normative; they are wrong, schwa usage in the US
isn't standardized, it's not a marker of class or geography,
there was no old class distinction that it reflected, and so
it's been under the radar for the most part. Some older
descriptions only admit one, but those also insisted on
"It is I" and rejected the genderless singular
"they" that over 20 generations of English speakers
have found so useful. You know the kind of prescriptivist:
they didn't like what the contraction of "am + not"
had become, so in their wisdom they rendered a slightly
aberrant paradigm even more defective and odd: "he is,
he's; I am, I'm; he is not, he's not, he isn't, isn't he?; I
am not, I'm not, I XXX, aren't I?".
That's preamble. Here's what I think is happening. I'm
pretty sure recall hearing a wider back/front range for [E] in
the W. New York area than I was comfortable with--I kept
hearing front schwas where I expected [E]. I think Blitzer's
backed his [E] into what you consider schwa territory (which
I'd probably consider front-schwa territory), and you're
interpreting it as a schwa, an "uh" sound. I
suspect you have only one schwa, or so don't like phonological
change in "democrat" that you take a front schwa as
a back schwa to make the sin in his dialectal pronunciation
abound.
Me? As long as I understand him well, I don't have much use
for social and class markers. I find creating and maintaining
them annoyingly divisive.
Printer-friendly format
Email this thread to a friend
Bookmark this thread
(1000+ posts)
Send PM |
Profile |
Ignore
