From: Jens Elmegaard Rasmussen
Message: 22339
Date: 2003-05-28
On Wed, 28 May 2003, fortuna11111 wrote:
[...]
[On e/o]
> Very interesting. I had heard something of the kind from one of
> my Latin teachers. He even made parallels with the "ommmm"
> used for meditation by the yogi. This is of particular interest to
> me, because I have also done classical singing, where many of
> these purely phonetic issues arise. An interdisciplinary study
> may turn out helpful on the subject. There are many linguists, for
> example, who worked with singers on diction and interpretation
> and who may have lots of original ideas on the relationship
> between vowels and consonants. I can think of one linguist who
> worked with Toscanini and used to teach at the Julliard School.
> If you find this intriguing, look for something written by Evelina
> Colorni. It has nothing to do with comparative linguistics, but it
> may give you some ideas.
I will remember that when the occasion arises. Actually, even if I am
personally performatorily tone deaf (can't hum Old McDonald so that
anybody can recognize it), my main inspiration was an idea of singing: A
fuller vowel tends to be deeper, and a deeper tone tends to take on a
fuller vowel timbre. You know both worlds, so I ask you, isn't this right?
If it is, then I feel I begin to understand that voiced segments (which
have a lower pitch than voiceless sounds according to phonetics manuals)
can not only lower the pitch but also change the timbre of /e/ to /o/ (in
crude phonetic terms). And further, if the IE accent had a high tone (as
Indic and Greek grammarians say it did), then I also feel I begin to
understand that lack of it can cause a vowel to appear as /o/ instead of
/e/. Actually, it mostly appears as zero which is the normal result of
unaccented short vowels, but if the vowel is lengthened, the product is
apparently /o:/ and not /e:/: Endstressed Gk. limé:n 'lake' vs.
rootstressed ákmo:n 'stone', IE endstressed *p&2té:r 'father' :
rootstressed *swéso:r 'sister'. Therefore I suggest that the lengthening
produced by the nom. marker *-s (**-z) worked after there had already been
some reduction of old /e/.
>
> That sounds intriguing. It also brings me to the thought such a
> prestage of the language could have been influenced by other
> languages. Have you tried to look for parallels outside of IE?
Well, yes, sure I have, but I actually think this can be understood on its
own basis. If this is already in keeping with natural tendencies of the
use of the human speech apparatus it does not really invite an
interpretation in terms of outside influence, does it?
[...]
> By the way, I have not read any other theories on the same
> subject, so I cannot judge on your theory against all others. On
> the other hand, it must be a pity to leave your points out of the
> debate.
Thank you for saying so.
> I
> > On this list, my e/o rule has been accorded very wide
> acceptance. We
> > disagree a bit over the phonetic rationale; some say it's a
> quantum of
> > lengthening, and that a somewhat longer e became o; I am
> now more in
> > favour of a tonal interpretation, thereby approaching the old
> stand taken
> > by Hirt. It *is* at least an increase in sonority induced by the
> following
> > voiced segment by whatever exact avenue.
>
> That sounds fascinating as a thought. And leads me back to my
> comments above on possible overlapping with phonetics, why
> not even art.
I feel I'm genuinely being understood here. In fact I also feel I am being
understood even when colleagues give it a twist of their own, as when I
get Zwischenrufe that I should have written either "voice-induced backing"
or "a somewhat longer schwa". I wrote what I wrote in good faith in an
attempt to cover all interpretations, even some I would not choose myself.
Anyway these nuances are minor, and the common area of agreement looks
very wide.
> > That the e/o rule works only for the thematic vowel is a
> problem of its
> > own. I'll leave it be for the moment.
>
> Would you elaborate on that or have you already addressed this
> issue previously?
I have, but that shouldn't stop me. I would not be able to find the
relevant thread anyway. If you cut back in the Yahoo group list and find
mails with the title "The sectors of ablaut" you'll have it. I think it is
from sometime this spring, but I have a poor memory for time.
The main idea was that the normal ablaut rules apply to roots, suffixes
and desinences, but apparently not to stem-final vowels. That is begging
the question: why is that? My best guess is to regard it as a phenomenon
of sandhi, meaning that there once was a boundary of a special kind
between the stem proper and the inflectional endings of a word. A somewhat
similar alternation may be seen in Sanskrit sandhi when underlying /as/
develops to /e:/ in the interior of words (nédi:yas- = Av. nazdiiah-
'nearer', or weak pf. stem Skt. sed- = Av. hazd-), but is /o:/ at the word
boundary of the following word begins with a voiced sound (nalo na:ma). I
therefore take it that the IE inflectional endings we know were once
individual words, possibly enclitic and there so reduced as we find them.
That presupposes the prior occurrence of even older reductions about which
I do not even dare to speculate. It is largely the role accorded these
still older stages in the elucidation of the PIE facts that separates the
discussants on the list with me in an unusual part as the conservative
stick-in-the-mud.
The general alternative to this inference of a very high age of the
"thematic vowel rule" (and of the thematic vowel) is the superficially
sound-looking conclusion that it is a younger vowel. Younger simply
because it is retained even when unaccented, as in *bhéreti 'carries', and
even worse *bhérete 'you carry'. The argument is that since there are "too
many" vowels here some of them must be secondary, i.e. originated in a
later time when the ablaut reductions had ceased to operate. But if
younger vowels acted this way, oscillating between /e/ and /o/ in
dependency of the following segment, *all* vowels should do that, because
also old vowels were present in the late period and would also have been
hit by changes applying to vowels at that time. Moreover, since the
thematic vowel plainly influences the ablaut. e.g. by attracting the
accent from following segments which subsequently go into the zero-grade,
it simply must have existed before the Schwundablaut operated. Then there
hardly remains anything but to ascribe a sheltering effect to the
environment where it is found, which is stem-final position.
In fact practically all the long-range reconstructions of IE inflected
wordforms suggested on this list make them into older sequences of words,
so a special boundary is in a way agreed upon for the position between
stem and desinence. It is my contention that we can actually see it here.
> > You do right in asking directly. Much nonsense has been
> handed down
> > because teachers did not address questions, and the next
> questions were
> > not even asked.
>
> Yes, but I find this is often a problem in scientific debate. People
> are afraid to ask their questions because they are afraid of
> appearing stupid and uninformed. It is a "Catch 22" thing: how
> can you be informed if you never asked?
It is even more difficult to ask a clear question to a confused
presentation. I don't think Louis Hjelmslev, a national hero where I come
from, had many questions asked.
Jens