[From: Bruce Nevin (Mon 93104 15:44:57 EDT)]
( Bill Powers (931005.1000 MDT) ) --
It looks to me as though you're distinguishing between the
_objective_ difference between spin and sbin and the _perceived_
difference.
Come again? Have we suddenly gained a direct line to Boss Reality?
The instructions were to substitute two sounds, which Rick perceives as
different (and contrasting, but let that rest) into a context that
differs only in having s before it. I asked Rick if that was hard to do.
I asked him, if so, why? What stops him from controlling a perception (p
vs. b) that he is able to control without that preceding s?
if Rick hears no difference between spin and
sbin, he can't follow your directions.
The question is, why can't you hear a difference?
The method you outline is useful. More combinations of phonemes are
possible at morpheme boundaries than are possible within a morpheme. (It
is this that underlies Harris' method of identifying morpheme boundaries
by a rise in next-successor phoneme count, which I have outlined before.)
More than this, the pronunciation of phonemes within morphemes may be
systematically different from the pronunciation of the same phonemes when
separated by morpheme boundary. Compare the tr cluster in "nitrate" vs
"night rate". This difference--the exceptional phonetic features
associated with the phonemes in one context but not the other--may be
considered the way the boundary itself is pronounced, its phonetic
"shape".
But attend to your own pronunciations:
[sb] fleece bin
[sp] spin
[sp`] fleece pin
You are controlling a pronunciation of [p] that is neither [b] nor [p`]
in the word spin. Ipso facto, you are perceiving it. Now pay attention
to this hitherto unconscious perception--a perception that you have been
controlling with great reliability, I know, because I have heard you
speak on Dag's video recording.
Here's some help making it "objective". cut a 1" strip across the end of
an 8.5 x 11 sheet of paper. Press one end against your forehead, in such
a way that the strip hangs across the tip of your nose and the other end
is suspended in front of your mouth. Then say the words
pin
bin
Keep alternating these words, and observe the affect of the air emerging
from your mouth on the paper. Call one effect "puff" and the other "no
puff" (or big puff, little puff, or aspirated, unaspirated, or whatever
seems to fit your perception of the affect on the end of the strip of paper).
Then say the above three utterances. Or, better, substitute the
following, so as to get rid of the distracting effect of the f:
mouse bin
spin
mouse pin
(Maybe you're in a store that sells computer accessories.) Compare each
in turn, pairwise, with bin and with pin. Identify the affect of each in
the same way, according to the size of the affect of egressive air on the
position and movement of the strip of paper. (It is possible to
distinguish the effect of the preceding s from the effect of the
following consonant.)
By this test, is the sp of spin most similar to the "sb" of mouse bin or
the "sp" of mouse pin?
(It is also possible, by attending to subjective perceptions, to observe
differences in the relative timing of lip opening and the beginning of
vibration felt in the larynx. Bell, Sweet, and others got pretty good at
this sort of thing before there was instrumentation for it. I'll bet you
can find a way to check this out with your sound spectrogram program on
your PC. This gives another kind of comparison.)
Now, substitute the p of "mouse pin" into "spin", observing the effect
on the strip of paper to be sure you have it right. Can it be anything
but "spin" (with a funny spin on it, so to speak)? Try using this
pronunciation in phrases, like
It went into a sPin.
Now try
It was sPinning around.
I find this difficult to produce, whereas the normal pronunciation is easy:
It was spinning around.
Do you find this so too? Is there a clue here? Or is this just an
unfamiliar sequence, which with practice we might be able to do easily,
as with learning a foreign language, perhaps?
What made me think of this was the way Kirk Sattley taught me how
to pronounce a Russian consonant which I can't spell. It's the
shch sound in fish chowder.
Or getting an Anglo to say "pot o' coffee", and then helping her to
recognize that the first two syllables are like the first two syllables
of Spanish "Paraguay". Yes.
Looks to me like learning a higher level of control for
manipulating simple phoneme sounds.
Rather, manipulating sounds below the level of phonemic perception.
Phonemes are not sounds. They are relations of contrast between
utterances, or perhaps between syllables. The sounds that are contrasted
can then be taken as representing the contrasts and as locating them
within utterances (or within syllables). The p of pin contrasts with all
the other sounds. That is, omitting vowels (and glottal stop), something
like this:
s sh h
_ t ch k
b d j g
m n
l
w r y
These symbols are arranged in rows and columns for a reason. All the
elements in a row or in a column have some sound-feature or
pronunciation-feature in common.
Alternatively, then, the lip closure of p b m (and the lip rounding of w)
contrasts with the neutral lip position of the other sounds, the
voicelessness of s sh h p t ch k contrasts with the voicing of the other
sounds, and so on, and those features (which define a matrix organization
of the segments listed above) are the phonemic elements that represent
the contrasts. If the phonemic elements are features, then p is not a
phoneme, it is a simultaneous bundle of phonemes (phonemic elements).
There are other possibilities that have been explored, different ways of
representing the contrasts between utterances, and locating them within
utterances. It is entirely possible (even likely) that different
speakers of the same language have developed different ways of
representing the contrasts between utterances, so that they can control
their pronunciations in such a way that different utterances are
recognizeably different to others and those others can reliably recognize
what they intend to say, and so that they can control their perceptions
of others' speech as recognizable utterances in a useful way. Learning
an alphabetic representation and learning to read may come easier to some
than it does to others for this reason, for instance. People to whom it
came hard tend not to be strongly represented in scholarly publications.
These are all representations. What is fundamental is what they
represent: the contrasts between utterances (or between syllables).
In English, there are no syllables that begin with two consonants, where
the first is s and the second is one of {b d j g}. In "mouse bin", etc.,
the s is at the end of one syllable and the b is at the beginning of the
next. Thus, there is no contrast between utterances (or syllables) that
can be represented by p vs. b, t vs. d, ch vs. j, or k vs. g, or by the
feature voiced vs the feature voiceless, or the feature aspirated vs. the
feature unaspirated, or any of the other ways of representing the
contrasts between similar utterances (syllables) without the preceding s.
Now what is interesting, and the reason why I directed Rick's attention
to his perceptions of these things, the reason why I am still eager to
see his report of what he observes--what makes this especially
fascinating is that we make ourselves unconscious of certain perceptions
that clearly are available to us. We put those perceptions out of reach,
so to speak. In the course of everyday talking in English, we control
the perceptions necessary to produce three kinds of bilabial non-nasal
stops (i.e. leaving out m)-- the p of pin, the b of bin, and the "p" of
spin. Those are obviously controlled perceptions, obviously "in reach"
for us to control. If we are asked to substitute either of the former
for the last, or the last for either of the former, we draw a blank, or
we produce the utterance without the substitution and swear that the
substitution has been made.
I should think this would be interesting to you, too, Bill. But you have
to notice it first.
Bruce
bn@bbn.com