<mounts pulpit-- it's Sunday after all>
> > --- In [log in to unmask], John Cowan <jcowan@R...> wrote:
> > >But others don't. I, for example, pronounce "r" with alveolopalatal
> > >articulation: the tip of my tongue is behind but not touching my
> > >lower teeth, while the blade of the tongue approximates my hard
> > >palate.
> > Finally! A description of my "r." I've been sitting here these past
> > few days saying, "rrrrrrrrrr," & trying to figure out where my tongue
> > is. This is the description that works, only to add that the sides
> > of the blade are in contact with the inside of my upper teeth. I
> > don't think I've every heard an American use a retroflex "r."
Well, close inspection and introspection are the only ways, sometimes, to
figure out what's going on in our mouths :-)) My only revision to the above
(pl.) would be to add: the tip of the tongue is in mid-air and is probably
curled up/back _to some degree_.
Is our /r/ retroflexed? Try these experiments (valid at least for Murkins):
Say "cod" or "hod"-- notice where the tongue tip ends up. Now say "card" or
"hard". Unless I miss my guess, you will see that the "d" is now articulated
noticeably further back. You will also feel, I think, that the d of cod etc.
involves more of the tongue blade, whereas the d of card is more apical
Same with "lodge" vs. "large", no? (And needless to say, same with final -t
and [-tS] words.)
Or contrast "cotter" and "Carter" or "lodger/larger"-- 2-syl. words work
better, since we tend to drawl monosyllables, leading to an intrusive [@^].
IMO the experiment should show that our /r/ is indeed retroflexed, in that
it shifts the POA from alveolar to post-alveolar, even if not to the hard
palate where true retroflexes are articulated.
Just for fun, if you can isolate those apical post-alv. t's and d's, try
using them in words where no r is involved. I guarawntee, it won't sound
right-- almost Indian. (One of the characteristics of Indian Engl., as Prof.
Catford liked to point out-- with a killer imitation(1)-- is their use of
retroflexed t/d in place of our alveolar ones.)
[k&l`'k6t`6 'k4IkIt` kl6b]
Not to confuse the issue, but note that [r\] (IPA inverted r) is not
precisely specified as to dental/alv/post-alv. in the chart at
http://cassowary.free.fr/Linguistics/cxschart.png -- note too that it's the
symbol used for "consonantal" (pre-vocalic) r, as in the ex. [r\i.&kt]
"react" under "Suprasegmentals". No argument there, I think.
It's post-vocalic, semi-vocalic r that's the problem. In close phonetic
transcriptions, it is often indicated as a _modification_ of the vowel, esp.
of [@]/ (stressed/unstr. resp.), so ['k_hIl@^], [b3^d] "killer, bird".
Since [@^, 3^] are unitary vowel sounds (no transitional movement of the
tongue as in "are, ear"), it suggests that in other post-voc. environments
they're functioning as glides and producing a sort of diphthong-- just like
superscript _U and _I in e.g. [a_U, a_I]. Thus in a close phonetic
transcription of "card", perhaps we should write [k_ha_@^d]; if we drawl
the word, we get almost 2-syl ['k_ha.@^d], just as drawled "cloud" will come
out [kla.ud]. Obviously, to an audience of Engl. speakers it isn't
necessary to be that precise-- consequently the various shorthand ("broad
phonetic") variants [br\=d] (or even [br=d] though that's bad phonetics, OK
as phonemic), [kIlr\=], [kar\d].
Note that in non-rhotic dialects, [@^] simply loses its retroflexion but
survives as an [@] offglide, as in "beer" [bI_@], NYC-ese "sure, shore"
[SU_@] (the vowel is actually somewhere between [U] and [o]) or compensatory
length (Tristan's Australian [bI:], RP "court" [k_hO:t]-- quite on a par
with the dropping of the glide-[j] is Southern US, "I" [A:], "fire" [fAr\].
Sally Caves wrote:
> I just can't duplicate what John is describing and still pronounce "car"
> way I do it. So there's no curling up of your tongue tip towards the roof
> of your mouth? It stays behind your lower teeth? Is there any curling at
> all, John? When I try to duplicate that, without the curl, I get not only
> sound that changes the quality of my "a," but an "r" that sounds like
> with "r-coloring," ...(snip) Maybe these distinctions are so subtle that
> it's hard
> for others to hear it when they aren't listening for it.
Your last sentence is the operative one :-))))
> I'm saying "car" now and holding it. The back of my tongue drops down
> the back vowel). The sides of my tongue are half way between my upper and
> lower back teeth, and the tip is turned up behind the alveolar ridge and
> pointing towards the hard palate. Let's try it with a front vowel, "ear":
> tongue rises in the mouth to accommodate the front vowel. Back and sides
> the tongue are touching the back teeth. To get the "r" ("ear" is
> a kind of diphthong for me), the whole tongue drops slightly, but not as
> as in "car," and tip of tongue curls up behind the alveolarpalatal region
> point at the palate. If I raise the tip of the tongue, it touches that
> tickly part of my hard palate that arches up and away from the post
This IMO is an entirely accurate description, and matches my own.
> But everybody's mouth is different. Mine is long and narrow (which is why
> had to have such extensive orthodonture: lots of teeth yanked because of
> over crowding)
Ho ho, you too, eh?
> Maybe that's the problem. We're assigning parts to the mouth, but every
> mouth is different.....(snip) > I guess I'm frustrated that I don't
> completely grasp where these areas in my
> mouth are: "post alveolar, alveolar palatal, and retroflex region.
True; from teeth to velum is a continuum; and the tongue is infinitely
mobile. Everyone's mileage differs.
> been entrenched in thinking that retroflex means the curling of the tongue
True; but that entails some slight backing too. Cf. ____ vs. ____/
>Those Americans who bend it up and back, which is what I think some of
> you are describing in using the term retroflex approximant, do exist, but
> associate that "r" with certain parts of the south, or parts of the
"Stage" or parodied Irish too, now that I think of it. Yes, some retroflex
more than others..........
> What we need in CXS is a better representation of the variations in the
> American "r." Judging from what I've heard, these sounds have been
Sometimes I wonder if those 19th C. French/German/British phoneticians
weren't just a bit flummoxed by Engl./Amer. "r" when it came to devising the