WEQATAL vs. QATAL statistics, cleaning up assumptions, 90% regularities

Henry Churchyard churchh at usa.net
Wed Aug 30 12:24:07 EDT 2000

> Date: Tue, 29 Aug 2000 10:19:23 -0500 (CDT)
> From: "Henry Churchyard" <churchh at usa.net>

> The term "phonemic" has a highly specific technical meaning within
> phonological theory: it refers to something which is not predictable
> from the surrounding purely phonological/phonetic context.  (In
> classic 1940's/1950's American structuralist linguistics, things
> were considered non-phonemic only if they could be redundantly
> predicted from strictly surface phones and junctures in a
> "bi-unique" way; but with the rise of generative phonology, an
> extended definition of "non-phonemic" has come into use, in which
> things are also considered non-phonemic if they can be redundantly
> predicted from more abstract phonological representations, such as
> synchronic underlying representations.)

Oh, something I perhaps should have mentioned:  In recent phonological
writings, the word "distinctive" is actually used more often than the
word "phonemic", but "distinctive" and "phonemic" come down to pretty
much the same thing.  (Probably "distinctive" is used because "phonemic"
has such strong associations with old structuralist phonemes, which are
no longer really used in modern phonology, at least in their original
"bi-unique" form...)

Mr. Buth is using "phonemic" in its original sense in the passage
below (while in discussing Hebrew stress patterns in my last post, I
used "phonemic" in the broader new sense which is less commonly used;
you can substitute the word "distinctive" there, if you prefer).

>>> Date: Tue, 22 Aug 2000 09:36:21 -0400
>>> Author: yochanan bitan-buth <ButhFam at compuserve.com>
>>> Subject: Re: WEQATAL vs. QATAL statistics, cleaning up assumptions

>>> stress in Hebrew is phonemic, as everyone agrees: E.G. SHAva 'she
>>> returned versus shaVA 'she is returning'.  And stress has many
>>> conditioned euphonic changes like in pausal forms or with nasog
>>> aHor.  But moving the stress to the end of a 1s,2s suffix
>>> conjugation verb qatalTI/qatalTA is not one of the euphonic
>>> changes.  However, with veqatalTI and veqalTA stress moves to the
>>> end.  It is not phonologically motivated.  It is a different
>>> morphological class.

>> From: Rolf Furuli <furuli at online.no>

>> When it is phonemic, we have the following situation: Two words are
>> written similarly, and they are differentiated by the position of
>> the stress.  This differentiation is consistently made in *all*
>> cases of the word in question.

I'm not sure in what sense you're using the word "phoneme" above; in
classic 1940's/1950's structuralism, there's the principle of "once a
phoneme, always a phoneme", so that if stress positioning is
unpredictable in some cases in a language, it's phonemic everywhere in
the language -- while in more recent phonological theory, stress is
phonemic (or "distinctive" if you prefer) if it cannot be predicted
from the surrounding phonological context.  So according to neither
definition of the word would non-pausal stress be excluded from being
"phonemic" because forms belonging to the same fully-specified
paradigm slot have a different stress in phrasal phonological
contexts.  Anyway, according to your proposal, you would have to
ignore contrasts which are not affected by pausal phonology (such as
the shaaBHAA vs. SHAAbhaa contrast mentioned by Buth above) if they
were sometimes neutralized by the Rhythm Rule stress shift (i.e.
_n at siga_ / _nasog 'ah.or_), which doesn't make too much sense.

>> However, nobody will claim that pausal forms are phonemic, yet they
>> are stressed differently than the same forms in context.

No-one will claim pausal forms are "phonemic" (in the recent sense,
not the classic structuralist sense), because the pausal forms do not
in fact show an unusual stress-positioning -- it's the _non-pausal_
forms which show the unusual stress-positionings, and yes, I certainly
would claim that non-pausal stress-positioning can be phonemic (in the
recent sense, i.e. "distinctive").

It's true that the non-pausal stress-positioning in the stress-shift
off a light penult (which creates alternations such as that between
pausal qaaTAAluu and non-pausal qaatLUU) is predictable from the
surrounding phonological context in the majority of cases (in a
complex way, which has to take into account both the location of a
word within its phrasal environment, and the presence of word-final
-nuu, -nii, -huu, -haa and -naa, onto which stress generally doesn't
shift); but even in this case stress is not fully phonologically
predictable (since it's when -nuu, -nii, -huu, and -haa are pronominal
suffixes and -naa is a verb suffix that stress doesn't shift onto
them, while stress also doesn't shift onto word-final -uu in segholate
nouns with underlying CVCw stem shape -- and there's no way to acount
for all this in strictly phonological terms without morphological
information) -- and certainly the non-pausal stress-positioning in the
exceptional form _hin at NUU_ `here we are' (the one form in the language
in which stress does shift onto a word-final 1st. person plural -nuu
suffix; see Malone _Tiberian Hebrew Phonology_ 1993, p. 99) is

And in the WEQATAL and WAYYIQTOL stress-shifts, the shifted non-pausal
stress positioning must be phonemic/distinctive because it's
unpredictable from the surrounding phonological context (the w at -/waC-
prefix does not provide a valid purely-phonological conditioning
factor for stress-shift, as I've explained before -- and in any case,
w at - and waC- prefixes occur attached to many forms which do not show
stress shift).

I think you should really let phonology determine what is and is not
phonemic or distinctive, instead of trying to override phonological
considerations with high-level semantics (just as you haven't let
morphology determine morphology, but override it with abstract
semantic theoretical models).

>> I also suppose you agree that there is not a single example of
>> WEQATAL where the syntax or context would not allow us to
>> interprete the WEQATAL as WAW+QATAL (conjunction+verb).  In other
>> words, there is not a single example of WEQATAL where the WAW
>> *must* be something other than a simple conjunction.

Actually, the "non-compositional reference" of WEQATAL is exactly what
I showed in the statistical analysis, as explained in my posting of
Aug. 27th.  Syntax is not affected because such "non-compositional
reference" does not affect a word's status as a verb with a prefixed

Henry Churchyard   churchh at usa.net   http://www.crossmyt.com/hc/

More information about the b-hebrew mailing list