The diphthong thing

Tue Oct 3 21:23:29 UTC 2006

Thanks all,

I think we aught to put this one to bed.  And I surely appreciate you all
for your efforts in trying to convince me that long i is diphthong, which by
definition is two vowels interconnected.  A vowel is a phoneme, so it's two
phonemes, and the ones referenced are ah-ee.  I don't hear "ah" but a
different soud altogether which is long i.  It  may trail off into a y-glide
at the end slilghtly, but many vowels do that anyway as the mouth
transitions.  A y-glide is not a vowel anyhow, but a consonant.

I don't think that long a is a diphthong nor is the sound "awe".  These are
as heard in m-w.com and not just my dialect, though in fact it's mine as
well.

Tom Z

>From: "Arnold M. Zwicky" <zwicky at CSLI.STANFORD.EDU>
>Reply-To: American Dialect Society <ADS-L at LISTSERV.UGA.EDU>
>To: ADS-L at LISTSERV.UGA.EDU
>Subject: Re: ah/ awe
>Date: Tue, 3 Oct 2006 11:44:19 -0700
>
>---------------------- Information from the mail header
>-----------------------
>Sender:       American Dialect Society <ADS-L at LISTSERV.UGA.EDU>
>Poster:       "Arnold M. Zwicky" <zwicky at CSLI.STANFORD.EDU>
>Subject:      Re: ah/ awe
>-------------------------------------------------------------------------------
>
>On Oct 3, 2006, at 10:17 AM, Tom Zurinskas wrote:
>
> > Thanks Arnold.  However, I wasn't refering to my own dialect.  I was
> > refering to "right" as spoken in m-w.com.
>
>ah.  the pronunciation there is diphthongal, but with a very short
>offglide, because of the shortening effect due to the following
>voiceless stop.  listen to "ride", "rhyme", and "rye" for easier-to-
>hear examples.
>
> > The verb is one phthoung not two
> > as I hear it there.
>
>i don't really understand your use of the non-standard technical term
>"phtho(u)ng".  but the offglide is certainly there in m-w's "right".
>it's just not very long -- i'd guess about as long as the offglides
>for /o/ (in "goat") and /e/ (in "gate"), but i'm not about to fire up
>my Praat software just to demonstrate a point that's quite clear to
>my trained ears (and, i think, to almost everybody else with some
>training in phonetics).
>
>a genuinely monophthongal pronunciation of "right" would give
>something in between "rot" and "rat", and would be heard, out of
>context, as one or the other of these.  (in context, there would be
>clues as to the identity of the word, and this pronunciation for
>"right" would lead you to identify the speaker as a monophthongizer,
>probably a southerner.  meanwhile, /o/ and /e/ without their
>offglides are possible, and would lead you to peg the speaker as from
>the upper midwest.)
>
>the terminology here is tricky.  many phoneticians distinguish
>between "true diphthongs", which have two steady states within a
>single syllable nucleus, and "offglided" (or "onglided") vowels, with
>a short final (or initial) movement away from (or towards) a single
>steady state.  the english phonemes /aj aw oj/ are true diphthongs, /
>i e u o/ offglided vowels.
>
>the complexity comes from the fact that the phonetic realizations of
>phonemes in context can be quite various.  an unglided vowel like /E/
>can have glided realizations (as /E/ does for many speakers before /
>g/, as in "leg", but not before /k/, as in "lek").  a true diphthong
>like /aj/ can have merely offglided realizations, in shortening
>contexts.  and so on.
>
> >   Good to hear you say that it is entirely possible to
> > say long i as a one phthong.
>
>and if you do so, you'll sound like a southerner.
>
>perhaps the problem is in your understanding of what counts as a
>vowel segment (my guess as to what "phtho(u)ng" means for you).  a
>very elementary observation in phonetics/phonology is that a segment
>can be phonetically complex, with parts that differ in their phonetic
>properties: in the world of consonants, the textbook examples are
>affricates, in the world of vowels, diphthongs.  no one is claiming
>that an affricate like /tS/ (in "church", twice) is a sequence of two
>consonants, or that a diphthong like /aw/ (in "out") is a sequence of
>two vowels.
>
>complex segments are (sometimes) *transcribed* as sequences of
>symbols, mostly because insisting on unitary symbols increases the
>size of the symbol inventory needed (and challenges conventional
>keyboards).  even so, affricates are given unitary representations in
>some transcription systems: /C/ (well, c-hacek) for the "church"
>affricates.  we could perfectly well do the same for diphthongs,
>pressing, say "#" into service for the vowel nucleus of "out".
>
> >   I agree.  However I think that long i as a two
> > phthong would be rare.
>
>common as dirt.
>
> >   I hear in Australia "roit".
>
>yes, a diphthong, not too far from the one i have in "boy".  but this
>is just the australian counterpart of  american/british /aj/, with an
>initial centralized portion.  there's a question about the best way
>to transcribe it phonemically, but the australian version isn't
>relevant to the nature of the standard american vowel.
>
> >   Perhaps UK has
> > "rah-eat".
>
>this invented transcription looks like you're talking about a two-
>syllable version of the word, which i don't think *anybody* has,
>except is emphatic hyperarticulated speech.
>
> > But USA has "eye" as on phthong.
>
>good grief!  *listen* to the way the word "eye" is pronounced!
>
> > Perhaps you could point out some real saliant words with the two
> > phthong.
>
>my previous posting listed a number of words in which the diphthongal
>character of this vowel is quite clear.
>
>arnold (zwicky at csli.stanford.edu)
>
>------------------------------------------------------------
>The American Dialect Society - http://www.americandialect.org

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org