Don't touch my phonemes (was: minimal pairs ex: PIE e/o Ablaut)

[Gabor Sandi]

Response to the comments made on Sun, 15 Oct. 2000 by Robert Whiting
<whiting at cc.helsinki.fi>:

Overall, I detect in your contributions a concerted attack on the traditional
concept of "phoneme". Instead of being the smallest distinctive unit of speech,
a phoneme then becomes an abstract concept which, after the application of a
number of similarly abstract rules, is transformed into the speech sound we
actually hear.

I do not like the direction into which this will take us, because then a
marvellously simple tool of descriptive linguistics will disappear.

To illustrate my comments, let us look at the voiceless/voiced contrast of
intravocalic dental fricatives in English. If we regard the two sounds as
different phonemes, we have a very simple way to transcribe phonemically
"faith" and "bathe" as /feth/ and /bedh/, respectively (/th/ and /dh/ obviously
standing for the symbols theta and edh).

If I understand your approach, you would represent these two words something
like /feth + foreign (or - verb?)/ and /beth +verb/, respectively.  What on
earth is the advantage here? Shouldn't simplicity be an important criterion in
linguistics (or any other descriptive science)?

I do not dispute the fact that in languages there can be, and often are,
patterns in phoneme distribution that are correlated with morphological
factors, vocabulary origin and the like. There can also be marginal phonemes
that occur in a few words only. Indeed, phonemes may occur only in words
recently borrowed, e.g. /b/ and /g/ in Finnish. There must still be, IMHO, an
unambiguous way to denote such sounds without reference to non-phonetic
criteria. If we don't subscribe to this principle, there is no end to the
complications that ingenious linguists can introduce into the description of
languages.

Now for additional points related to your posting:

(RW)

>... we are indeed talking about the
>synchronic phonology of modern English.  About the second point,
>I am much less sanguine.  I think it is possible to describe how
>language works linguistically (some areas are easier than
>others), but I don't think that we are in a position to say what
>goes on in a speaker's head to produce language.  While it would
>certainly be nice to know, I think that the cognitive processes
>that produce language are beyond our reach at the moment.
>Speakers themselves don't know how they produce language, so you
>can't find out by just asking them.  So the only reality that we
>can get at is the language that speakers produce.

(GS)

I agree with this completely. We cannot get into people's heads, therefore -
until our knowledge of neurology improves drastically - we should consider the
brain as a black box. All evidence about language should be based on
observations of how language is used.

(RW)

>Historical grammar, when we have a written record, is based on
>empirically verifiable facts.  Synchronic grammar is based on a
>hypothesis about how native speakers produce their language.  A
>hypothesis is not a fact.  It is an explanation put forward to
>account for observable facts.  People tend to forget this and
>consider synchronic grammar a fact.  But whatever reality may
>exist in the speaker's head just can't be gotten at at the
>present time with our present knowledge.  In general, I agree
>that what we are trying to get at is the reality in speakers'
>heads, but it is a roundabout road that we have to take and we
>have to have a realistic picture of language before we are likely
>to get there.

(GS)

Has anybody really claimed that synchronic grammar is in some way real? It is
real only insofar as it is a human artifact, just as philosophy and esthetics
are, for example. Otherwise, synthetic grammar should be considered as just a
model of the rules that govern actual language use. If this model is
contradicted by observed language use, it is clearly false and should be
corrected.

(snip)

(RW)

>You ask what evidence I have that native speakers distinguish by
>rule the [th] of loan words from the [dh] of native words other
>than hundreds of examples and the fact that it produces a nice
>phonological generalization.  This seems rather like asking what
>evidence I have for Grimm's Law or Verner's Law other than
>hundreds of examples and the fact that they provide a nice
>phonological generalization.  What more can one ask of a theory
>(or hypothesis or law) than that it account for the observed data
>in a consistent and concise manner and that it provide a reliable
>generalization?  Nothing, really, except, of course, that it have
>a test for falsification.

(GS)

Do people question the validity of your generalization? Given that we know a
lot about the history of English, we can see that there is indeed a pattern,
and that, indeed, intervocalic /th/ is restricted to borrowed words. But it's
only because we know the history of English that we can say this - if we didn't
know it, there would be no objective criteria telling us that, say, "author" is
borrowed and "feather" is not. Marking "author" as +borrowed as part of the
proof that only borrowed words have intervocalic [th] sounds like a circular
argument to me.

To show how easy it would be to lead us astray with such arguments, let us
consider Indo-European /b/. This is widely known to be a rare phoneme in PIE,
and I would love to be able to show that it was actually absent from early
stages of the proto-language, and all its occurrences are in late loanwords,
assimilations and the like. *abol (apple) and *belos (strong) are among the few
words where *b may be reconstructed, so why not claim that they are borrowings?
I would be hesitant to do so, however, because there is no independent evidence
that they are borrowings, and the claim is easily seen to be made for the
specific purpose of proving something about the absence of PIE *b.

(snip)

Since I don't at all dispute that there is a pattern in the distribution of
th/dh in English, let me reiterate what I think is my main point:

For the description of a language, it is essential to have recourse to phonemes
as basic building blocks of speech. If two sounds are in actual or potential
contrast, they belong to different phonemes. This is true even if in most (or
even all) of their occurrences the choice as to which one occurs is predictable
based on some non-phonetic criteria.

Of course, if you manage to expropriate the term "phoneme" for your purposes,
those of us needing the concept for the original use of the term (say, in order
to show the pronunciation of words in a dictionary) will have to call phonemes
something else, won't we...

All the best,
Gabor Sandi