underlying schwa

[Mark_Mandel at DRAGONSYS.COM]

Larry <be262 at scn.org> writes:

>>>>>
Consider the word "calcule" (/kalkju:l/) which appears in the 2041 edition
of Webster's. We find the following entry: "1. a decision based on a
calculation (in the figurative sense), a consideration of various
possibilites; 2. a choice or decision, made against seemingly strong odds,
but turning out to be the correct one in hindsight".

When considering *possible* words in English we may be able to prove that
there can be no underlying schwa. Supporting evidence might be found in the
history buried in (ore revealed by) English spelling: every schwa is
represented by one of the 5 vowel symbols, and we will find cognates or
antecedents in other languages that have non-reduced vowels in the exact
same position where we find schwa in English.

     [Jorge Guitart had written:]
> What is the better candidate for the UR of X where X is the morpheme
> common to each alternant--shwa or something else?

Contrary to what I suggested in my first message, I would now suggest that
the one example (the "calcul-" words) may not be useful evidence to support
an underlying schwa because it is merely accidental that there is no word
(any longer, yet) from that group where a stressed /ju:/ sound occurs
between the second "c" and the "l".
<<<<<

In all this discussion I have seen little or no mention of the influence of
spelling. I had it ground into my head very early in my linguistic career that A
PHONEME IS NOT A LETTER and so on, in the effort of overcoming the layperson's
tendency to treat the written language as the true or underlying form, and I
suppose that many other linguists had the same experience. But I fear that this
early drill leads us to ignore the powerful effects of literacy, especially in
rare words that are encountered only by the highly literate and through the eye
rather than the ear. Certainly SPE pretended that English-speakers somehow
develop a huge store of essentially diachronic knowledge, some of it based on
highly arcane lexa, without any influence from the written word.

Could it just possibly be the case that the so-called "underlying" stressed
/ju:/ in these groups of words, many of them freshly coined and never learned
aurally, is something that we learn and maintain in our knowledge of the
*written* language?!

   Mark A. Mandel : Senior Linguist and Manager of Acoustic Data
         Mark_Mandel at dragonsys.com : Dragon Systems, Inc.
 320 Nevada St., Newton, MA 02460, USA : http://www.dragonsys.com/
                     (speaking for myself)