"syllabicity"

[Rich Alderson]

On Fri, 16 Apr 1999, "Patrick C. Ryan" <proto-language at email.msn.com> wrote:

>From: <CONNOLLY at LATTE.MEMPHIS.EDU>
>Sent: Thursday, April 15, 1999 11:09 AM

>> Lehmann's book is a monument not only to structuralism, but also to
>> Neo-Grammarian notions of the 19th centuries -- both of these as the basis
>> for use of the laryngeal theory (in this instance, four additional PIE
>> consonants not recognized by the Neogrammarians) to explain some odd
>> developments in the Germanic languages.  Not surprisingly, the book is a
>> mess.

>It is profoundly irresponsible to label anything written by Lehmann as a
>"mess".  He is one of the preeminent IEists of the 20th century, and to
>cavalierly dismiss his work as "Neogrammarian", as if a label could discount
>his achievements and contributions, is tragically unjustified.

Lehmann is no more a god than is Szemere'nyi, Brugmann, Beekes, Watkins, or
Cowgill.  He falls down on certain issues, as do all the others, and gets some
things right, as do all the others.  But much of what he wrote 50 years ago--
and that is about how long ago it was written--is a mess.

However, Leo Connolly's use of "Neo-Grammarian" to describe his work is no more
dismissive than to so label the work of Carl Brugmann.  It assumes the system
reconstructed in Brugmann & Delbrueck, and attempts to explain exceptions in a
way which the _Grundriss_, Hirt, or Meillet would have approved (whether or not
they would have agreed with them).  Unfortunately, the _Junggrammatiker_ system
needed to be re-examined, and by not doing so Lehmann causes himself problems.

>> The Neogrammarians had realized that the vowels [i u] tend to alternate
>> with [y w] under conditions which no one has ever been able to specify
>> *exactly*;

>This is, in my opinion, totally misleading. The condition has been exactly
>specified, and in such simple terms, that hardly anyone, who does not have a
>predisposition to think that the latest fads in linguistics are the last
>word,  could not understand them: initial ['Y/WVC] is ['y/wVC]; initial
>[Y/WV'C] is [i/u'C]; ['CVY/WC] is ['CVi/uC];  [CV'Y/WVC] is ['Cy/wVC];
>[CVY/W'C] is [Ci/u'C]. Now, what was so difficult about that?

It's extremely simple.  However, what evidence do you have to back it up?  It
certainly does not appear in _Grundriss_ or Meillet.

>> since this was apparently his dissertation, he felt obligated to say this
>> within the framework dominant at the time: PIE [i u] were allophones of /y
>> w/, not "true" vowels,

>I do not know if Beekes had such an argument in *his* dissertation but in a
>book published as late as 1995, he is still asserting what Lehmann's
>dissertation asserted. So, though you may disagree, many eminent IEists
>still maintain that IE [y/w] are primarily consonontal. And, as any
>Nostraticist can assure you, IE [y/w] reflects Semitic [y/w]. If Nostratic
>[i/u] -- presuming they actually existed -- showed up as Semitic [0], you
>might have a talking point but they do not.

Read very carefully what Leo Connolly wrote:  Not that *y *w were not consonan-
tal, but that in the structuralist framework (alive and well in the 1940s) made
Lehmann choose one way or the other, and he chose to see *y *w as primary, and
*i *u as secondary.  That's a problem with structuralism, not with whether *y
and *w were or were not consonantal.

In structuralist terms, two phones in complementary distribution *must* be,
cannot *not* be, allophones of a single phoneme.  (Although a lemma requiring
something called "phonetic similarity" was inserted into the theory when it was
pointed out that in a pure framework, the English phones [h] and [N], as in
_hang_ [h&N], must be allophones of a single phoneme...)  Therefore, in the
prevailing structuralist framework of the 1940s, Lehmann *had* to define *i and
*u as allophones respectively of *y and *w.

>> just as PIE syllabic [M N L R] were allophones of /m n l r/.

>The syllabic status of [M/N/L/R] is a totally unrelated matter. These become
>syllabic when deprived of the stress-accent.

So the fact that all *six* resonants pattern the same is irrelevant?  That
*ey/oy/i parallels *en/on/.n by accident?  Then you disagree with Lehmann? What
of his god-like status?  Never mind, rhetorical questions.

In a structuralist framework, if two processes appear to be the same, they must
*be* the same, and you cannot separate *y *w from *n *m *r *l in this way.

>> Furthermore, though the laryngeals were unambiguously consonants in PIE (his
>> view and mine, though others differ), the attested IE languages often have
>> vowels where there were once laryngeals.  The Neogrammarians had posited PIE
>> Schwa in just such places.

>I do not dispute that 'laryngeals' were consonantal in Nostratic but by
>Indo-European, I believe their consonantal had been lost except for Hittite.

Your belief has nothing to do with the evidence.  As pointed out elsewhere by
Miguel Carrasquer Vidal, there is evidence for *consonantal* reflexes of one or
more laryngeals in non-Anatolian languages; Germanicists have long argued for
the presence of consonantal laryngeals in Germanic (and Lehmann is included in
this group).  All Indo-Europeanists who accept laryngeals (and this is very
nearly all of them by now) accept that they were consonantal in PIE.

[ snip ]

>> But I simply do not remember his ever making any use of "syllabicity" beyond
>> the obvious ones: /m n l r y w/ had syllabic realizations between
>> non-syllabic segments

>> [ Moderator's comment:
>>   The "syllabicity" in question is in the final chapter of the book, in the
>>   discussion of the stages of pre-IE vocalism leading up to the vowel system
>>   seen in PIE, by which I mean that reconstructible from the daughter
>>   languages in Neogrammarian fashion.  The earliest stage which he posits is
>>   one in which there is *no* phonemic vowel at all.  I was charmed by the
>>   notion for years as an undergraduate, but then I learned more phonology.
>>   --rma ]

>Rich, I would be interested to know what phonological principles you believe
>Lehmann's "syllabism" violates?

This is the reason it has taken me a week to get around to responding to this
posting, and not wearing my moderator hat to do so.  It's a very large question
with an answer unlikely to satisfy the questioner.

First, you must understand that I follow David Stampe's "Natural Phonology", as
outlined in his dissertation and other works, and in the works of his students.
Most important in this context is the work by Patricia (Donegan) Stampe on the
phonology of vowel systems.  Natural Phonology is process-oriented and requires
that both lexical representations and derivations always be pronounceable; it
is thus distinguished from Chomsky & Halle's _Sound Pattern of English_-style
generative phonology, in which underlying (lexical) representations of English
reproduce the Great English Vowel Shift.  (See, for example, her dissertation,
_The Natural Phonology of Vowels_, available as an Ohio State Working Paper in
Linguistics, No. 27 I think.)

On this basis, as an undergraduate I began an examination of the monovocalic
analysis of Indo-European 25 years ago.  I started out with what I considered
the most interesting analysis of IE vowels, that of W. P. Lehmann, and looked
for parallels in other languages (as the only way to demonstrate that an analy-
sis is valid is for it to explain not only historical but synchronic phenomena
in more than one language).  This led me to look at Abkhaz, Abaza, Ubykh, and
Kabardian, all of which have very large obstruent systems and very small vowel
systems.

Natural Phonology is, as well as process-oriented, constraint-oriented and
hierarchical:  The presence of certain phonological entities entails the
presence of others.  Thus, vowel systems are constrained:  Certain kinds of
vowel system are more stable than others, and unstable vowel systems rapidly
turn into stable systems by either eliminating contrast or by adding contrast.
In addition, processes which are not repressed may increase distinctions
between vowels in the system (long vowels may become tense, for example, or a
distinction in palatality vs. labiality may arise as in Arabic short /a/ vs.
long /a:/ = [&] vs. [O:]).

There is no such thing, in the world's languages, as a system with only one
phonemic vowel (and _a fortiori_ no such thing as a language with none).  The
smallest phonemic vowel inventories yet found are in the languages I named
above--and the analyses which shrink them to two vowels may violate one or more
of the axioms of natural phonology.

Thus, Lehmann violates a major principle when he asserts that any stage of
Indo-European lacked a phonemic vowel:  If a phone is present in a language, it
has a psychological status in the lexicon, and while it may alternate with
other sounds in the language because of morphological rules or unconstrained
processes, it cannot be denied phonemic status.

								Rich Alderson