Re: [tied] The phonetic value of PIE *h3 and the 'drink' root.

From: elmerasdk
Message: 14214
Date: 2002-08-03

--- In cybalist@..., Miguel Carrasquer <mcv@...> wrote:
> On Thu, 1 Aug 2002 02:24:12 +0200 (MET DST), Jens Elmegaard
> <jer@...> wrote:
> >1. Alternation ó / zero depending on accent. This is practically
only seen
> >in reduplicated verbal categories
> There are some examples of ó/zero in the nouns as well (e.g.
> *pónt(e)Hs, *pn.tHés).

I took pains to present matters in the simplest way possible (without
deliberately telling lies), so I left out the
section "Lexical //o//", in part because I am not fully sure that
*is* the correct understanding, but this is one of the principal
cases in point. I see no reason at all for the root's having -o- in
*pónt-o:H2-s, *pn.t-H2-ós other than the mere fact that a root must
have *some* vowel timbre, and why would /o/ be excluded? That would
make /pont-/ the unconditioned form of the root concerned, and, being
a short vowel, its /o/ would alternate with zero. For this particular
lexeme, the analysis has the slight drawback that the underlying root
form cannot be used directly for Germanic *finth-i/a- (Goth. finthan,
Eng. find). However, that obstacle can be overcome by the assumption
of ablaut normalization in Germanic where e-a-zero-zero was
productive with roots of the structure CVRC-. I therefore draw no
inferences from pons, póntos, poNtI, hun concerning IE ablaut. But I
admit that the o-vocalism is kind of unusual, so if I could create a
rule to derive the o from an old e here, I would not hesitate to use

> >thus (1) perfect *kWé-kWór-e, 1pl
> >*kWe-kWr.-mé, (2) intensive *wr.-wórt-mi, 3pl *wér-wr.t-n.ti, (3)
> >causative aorist, athematic kind, 3sg *H1gi-H1gór-t > Ved. áji:gar
> >'awakened', as opposed to the thematic kind (old middle?) *wé-wkW-
> >> Ved. ávocat 'spoke'.
> The reduplicated aorist has o-grade? I see only evidence for
> zero-grade (aji:gar can be from *aji:gr.).

No, Indic /-ar/ is full grade; *-r. yields /-ur/; you are no doubt
thinking of Iranian which has /-ar/ also from *-r. And, from svap-,
the corresponding form is sís.vap (2sg inj.), it's no use making that
a zero-grade. I chose velar roots in my presentation in order to show
the lack of palatalization.

> >3. A related regularity has produced with /o:/ from stems
> >underlyingly long /e:/, as *pe:d- 'foot' => nom. *pó:d-s. All such
> >are animate and contain the nominative marker (or its effects)
> What about *wódr, *wédn(e)s?

I left that out too to avoid provoking the wolves. You may recall an
earlier discussion we had where I referred to my analysis of the o-
type of heteroclitics (and the like) as original collectives-turned-
singular (published in the Schindler memorial volume, Compositiones
Indogermanicae, Praha 1999). If the weak cases have -e-, we have to
depart from long -e:-; to get that to have o-timbre can be handled by
lengthening, and, hurrah, the old collective marker *-H2 lengthens.
In 'water' I then posit *wé:d-r-H2 (using, for the underlying form,
the Szemerényi-inspired rule that some suffixes are vowelless if the
stem ends in less than three consonants, as is here the case with /-
Vd-r/); in that form, the collective desinence will lengthen
producing *wó:dr-H2, and the cluster preceding the desinence will
subsequently cause shortening, the expected result being thus *wód-r-
H2. If the collective meaning is lost (how much water is "a lot"?),
there may be backformed an unmarked "singular" *wód-r., which, I
submit, is what we have. I therefore take *wód-r. to be a collective
form, while *yé:kW-r. 'liver' is the undisturbed singular of that
particular lexeme. The presence of the collective marker is confirmed
quite strongly by Skt. ásthi 'bone', which has -o- in most languages,
but also shows -a- (Welsh as 'rib', asgwrn 'bone'), and so, again,
must be based on *-e:- which is shortened in weak cases, producing
*H2ést- > *H2ást-, and lengthened before the collective marker to
give *H2ó:st-H2 > PIE *H2óst-H2/*H2óst-&2 (Skt. ásthi with aspiration
and schwa).

> >4. There is the "thematic vowel", i.e. vowels in stem-final
position which
> >regularly behave in a way of their own. Being independent of the
> >they have no zero-grade alternants, but show up as /o/ before
> >segments (and surprisingly also before the nominative marker which
I then
> >take to have been earlier voiced), and as /e/ before voiceless
> >(and zero). The dependence upon the environment is generally
> >but for reasons beyond my comprehension common opinion has
apparently only
> >got as far as "o before sonant", covering cases of following m, n
(t), w, y
> >and r, while disregarding *-od, *-odhi, *-obhi and *-o- + vowel,
and of
> >course the difference between 2sg *-es and *-os. The
> >rationale is probably again one of tone: voiced sounds are spoken
on a
> >lower tone than voiceless segments.
> Strictly speaking, voiceless sounds are toneless (tone is carried by
> the vibration of the vocal chords). There are two kinds of
> voicelessness: that of /?/ (vocal chords closed), and that of the
> other voiceless (especially voiceless aspirated) consonants (vocal
> chords too open to vibrate). A transition to /?/ may be
anticipated a
> rise in pitch of the preceding voiced segment (e.g. a vowel), while
> transition to a normal voiceless (aspirated) consonant may cause a
> fall in pitch. A transition to silence can in principle occur in
> either direction (although I have the impression that it tends to be
> more towards -V?# (-> rising tone) than towards -Vh# (-> falling
> tone)).

All sounds have *some* frequency, often composite, and
voiceless "noises" are of higher main frequency than voced consonants
and (especially) vowels. Transition to silence may be anticipated by
deactivation of the voicing, this making the final part of the vowel
voiceless which gives it a markedly higher main frequency. So silence
acts like voicelessness, at least in some languages. I observe that
the relevant prestage of PIE was one such language. Another is
present-day Icelandic.

> >Why would we have IE *dhor-, *tog-, *bhor-, but
> >*k^ubh-, *bhug-? I have investigated the matter at length and
found that a
> >given root-structure type consistently uses either the form with -
o- or
> >the form with zero in such formations, thus also *tomH1-áH2 (Gk.
> >*tois-áH2 (Lith, tiesà) like *g^onH1-éye- (OE cennan, Ved.
janáyati) and
> >*tois-éye- (Lith. taisýti).
> I can see no difference in principle between a syllable structure -
> and a structure -ewg (except for the voicing of teh final
> Do you have an explanation?

It is a descriptive fact of many, many languages that the phoneme /s/
has a combinatory ability far exceeding that of the other phonemes.
English has words beginning with str-, but not ktr- or ltr-; you
have /nekst/ ("next"), but not *neklt or *nekpt; you have marks, but
not *markp or *markw. That is why I chose to treat /s/ as a
phonological category of its own, a strategy that proved quite
justified by the facts I found.

> > A second confirmation is the comparable loss of a laryngeal in
> >environment oC_C (perhaps only oR_C) in the position after the
root, a
> >rule that works only if the -o- is the one we're talking about:
> >are Gk. pórne: 'harlot' and tólme: (mostly, but secondarily, tólma)
> >'patience' from the roots *perH2- 'trade' and *telH2- 'endure,
> >These are derived from verbal nouns in *-men-, which would be IE
> >*pér&2-mn. and *tél&2-mn.; Greek has the compounded form with zero-
> >apó-pra:ma 'subletting' (*-pr.H2-mn.) and the animate
telamó:n 'carrying
> >strap' which both demand these forms to have existed. From these
> >IE *pórnaH2, *tólmaH2 are derived just like *tog-áH2 from the root
> >noun) *teg- 'cover'. The underlying forms are *perH2-men-é-H2 and
> >*telH2-men-é-H2 plus the addition of the -O-. By the zero-grade
> >this gives *pOrH2-mn-é-H2, *tOlH2-mn-é-H2, i.e. monosyllables in
> >again, the *consonantal* -O- caused so much crowding that the
> >of the root segments were lost.
> > A third support is offered by the same forms in the reduction
of the
> >suffixal *-mn- to *-n- (where labiality precedes somewhere in the
> >already) or *-m- (where there is no other labialized consonant),
an event
> >otherwise known only from cases where the suffix *-m(e)n- is
itself in the
> >zero-grade and is on top of this preceded by zero-grade segments
only (as
> >must have been the case in the old instrumental of man-stems in
> >which are found to end in -ma: or -na: with only one nasal).
> > A fourth support is also offered by these forms, viz. the
> >of the -ó- in derivatives from heavy roots like these (CeRH-) as
> >to unaccented -o- in parallel derivatives from light roots (CeC-),
> >e.g. Gk. kormós 'stump vs. tórmos 'hole' from kérma and trêma
> >respectively. The underlying forms are here *kér-mn => *ker-men-ó-
s and
> >*terH1-men-ó-s plus the -O-, i.e., after ablaut, monosyllabic
*kOrmnós and
> >*tOrH1mnós. Again the -O- caused the laryngeal to vanish, but its
> >presence has triggered the -O- to be syllabified so early that the
> >was a vowel in this type when the initial accent rule operated,
> >*tórmos, but still for a while monosyllabic *kOrmós which only
became *kormós
> >at a time when the initial accent rule has ceased to operate. I
> >collected some 75 examples of these derivatives, some forty or so
> >languages that can show the IE accent, and there is a complete
match with
> >unaccented -o- with light roots as against accented -ó- with heavy
> >Not all of these derivatives end in *-mo- or *-no-; some end in *-
> >*-wo- or *-to-. Examples are Lith. bal~sas 'voice' (with laryngeal
loss as
> >opposed to the root of bìlti 'start talking'), Skt. sárva-/Gk. hól
> >'all' from the root of OIr. slán 'healthy', Lith. vietà, acc.
> >[acc.-class 2] = Germ. Weide (PGmc. *waitho:) from the root of
Lith. výti
> >'hunt'.
> I think the evidence is sound, but there are a few things which are
> puzzling. I can understand the laryngeal (and one of the two
> being eliminated in a cluster like *pRrh2mn-, and I can understand
> the syllabification of *R (or alternatively, the insertion of a
> svarabhakti vowel [à la *pontHs -> *pontVHs?], or, alternatively,
> blocking of zero grade) in a super-heavy cluster like this, but it
> seems strange that *both* would happen (as apparently is the case in
> porn-).

The facts are the ones to decide, whether we understand them or not;
but that is only in principle, this case is not beyon comprehension.
I considered the alternative options very carefully and found none to
work; you are in effect taking me back to considerations and even to
preliminary theories I once entertained, but gave up because the
facts did not fit them. *pónt-o:H2-s/*pn.t-H2-ós is not on a par with
tórmos/kormós or the vocalism of the causative *mon-éye-ti which acts
by a completely different set of rules, this being my whole point (my
only point, in a sense). "Blocking of zero-grade" is not an option,
for that would not explain the vanishing of laryngeals and the
reduction of -mn-, events that demand articulatory crowding of
consonants. The syllabification of the retained -R-'s (if now we call
them that) will have to wait until zero-grade has been reached and
the result is one massive array of consonants (mostly a very long
monosyllable). In *that* intermediate stage, which may have been
shortlived, given its complex nature (which, however, is still not as
extreme as that of Modern Georgian or Itelmin), laryngeals were
weakened, but to begin with not lost entirely: In *pRrH2mnéH2, the
first H2 was reduced (let me call the product h2 with lower-case h);
in a sense, laryngeals in that environment were on Death Row, waiting
for the inevitable execution. Now, in *pRrh2mnéH2, or already
*pRrh2néH2 in case -mn- was processed earlier, which we can't know,
the -R- was syllabified in a *relatively* early period which comes
now, changing the form to *porh2néH2. In a comparable derivative from
a light root, as *kRr-m(n)-é-z, the was no such vocalization yet. At
some unknown point in time after the early round of infix
syllabification the weakened laryngeals were lost (h2 > zero). Now,
while we are still waiting for the second round of infix
syllabification (in lighter sequences), the INITIAL ACCENT RULE is
born, changing *por(h2)néH2 to *pórneH2, but not affecting *kRrméz.
Then, finally, we have R > o (and thematic vowel to e/o depending on
voicing), this giving *pórneH2 and *kormóz, whence PIE *pórnaH2 and

> Perhaps, then, one of the two phenomena has a different
> explanation than that of heavy clustering. Was there something
> *R itself that caused laryngeals to disappear? Did the retraction
> the stress have something to do with the loss of the laryngeal (or
> viceversa: maybe there are some clues in the later development of
> Balto-Slavic accentology?).

I utilized the evidence of Balto-Slavic accents to the complete
despair of my critics (including the evaluation board of my thesis in
the mid-1980's). Taking the evidence collected by Illic^-Svityc^ and
his school, and going on in the same vein, I found that BSl. showed
the exact same picture as Greek. I am not above learning, but I do
not accept the insinuation that I disregarded Balto-Slavic. On the
contrary, that is the position to which my critics are now clinging.
I'm not gonna mention names, but all published attempts at
criticizing my results have fallen flat on their faces once BSl. was
considered. There are even some who are trying to work out new rules
for BSl. accents so that they do *not* fit the impression gained from
the other languages by the Moscow School and found confirmed by my
investigation of the IE o-problem, that's how low some "scholars"
stoop. And, since very few IE-ists know anything about BSl. accents,
these attempts are not without some initial success.

I have one basic message to Diachronic Phonetics, i.e. Don't get
blinded by what I would call "Quantum Phonetics". Most phonetic
changes are gradual, and, where languages are actually recorded over
an extended period of time, processes are hardly ever simple.
Therefore, the loss of laryngeals does not have to entail a switch
from one pronunciation with laryngeals at full blast to one with no
trace of them, there may very well have been weaker laryngeals spoken
in the time it took for the process to reach completion. I have
recently treated Norse and West Germanic i-umlaut, finding it
possible to actually combine the obvious and striking correspondences
by making the process a gradual one, so that Proto-North-West-
Germanic had cases of "half-an-umlaut" on many of its vowels, and
weakened-but-not-yet-lost unaccented vowels, so that later dialects
were free to act differently on the same basis, and the shared
intermediate stage offers a natural basis for all the later special
conditionings of umlaut observed. But here again, I am on sacred
ground where scholars perhaps do not even *want* problems to be
solved and go away.