Sent: Friday, May 18, 2001 5:43 PM
Subject: Re: [nostratic] AA-IE
I do not feel qualified to discuss the
validity of Proto-Dravidian or Proto-Afroasiatic reconstructions. I can offer
comments mainly on IE forms. Other things being equal, one would expect IE
data to be of superior quality. Unfortunately, there is no really usable "PIE
dictionary" for informed amateurs; root inventories like Pokorny's are of
limited value -- they don't reflect the current state of PIE studies, and
contain a high proportion of questionable reconstructions, inaccuracies and
patently false data. Cross-family comparisons often draw from doubtful or
outdated sources and indiscriminately mix figments with reliable forms. Your
list is no exception, I'm afraid. Now as for details:
(1) Reflexes of PIE
*g^enh1-/*g^onh1-/*g^nh1- (this is the correct form) 'beget' are not
attested with the meaning 'make love'.
[PCR]
I assume because you are not a native
speaker of English, you are confused.
I have never asserted that *g^en- meant
'make love', just fornication; it means 'beget', 'fertilize an ovum with
sperm'. I hope you are not asserting that it does not mean that because you
would be dead wrong!
And I believe there are forms in IE
languages which require *g^en-.
[PG]
PIE had other specialised terms for
the latter. Sumerian has a number of terms for 'child', the most important of
them being <dumu>. What is your **gin-na based on?
[PCR]
I have explained that in a previous
message to the list.
[PG]
The Sumerian verb corresponding
semantically to PIE *g^enh1- (meaning 'bear, give birth, beget, fashion,
create', etc.) is <tud> -- no etymological relation.
[PCR]
You mix apples with oranges: first, there
is 'beget', then 'bear'. And frankly, I doubt whether tu(d) means 'beget'. The
archaic sign depicts a 'seed with a sprout'.
I suspect that tu(d) is ultimately
cognate with *do:u-, and simply means 'give', i.e. 'produce'.
[PG]
(2) The PIE words for 'crane' are
considered to be deverbal, from *gerh2- (~ *grah2-) 'make a loud cry'. Similar
terms for noisy birds are very common (Baltic/Slavic *k^orh2-k-ah2 'magpie',
various names for 'crow' or 'raven'), so the 'crane' word is probably of
imitative origin. Such items should be used with great care, if at all, in
distant comparison.
[PCR]
Yes, one must be careful. But, I will
say, that I consider Nostratic had a heavy bias towards nominal forms. And
again, Arabic kurki:y-un, 'crane', shows no sign of a
'laryngeal'.
[PG]
(3) [Why separate two homophonous and
semantically overlapping etyma?] PIE has a good number of roots meaning
approximately 'turn', 'twist' or 'plait', 'weave', but *ger- is not
particularly conspicuous among them. In fact, I doubt if it can be regarded as
authentic. What particular forms justify this reconstruction?
[PCR]
It is the cognates from other languages,
which cannot represent *g^ that incline me in this direction. But there is
also Old Indian gárta-H, 'wagon-seat'.
[PG]
(4) Very loose semantics, plus a
questionable equation within IE. Gk. kephale: matches Toch. A s'pa:l formally
and semantically, but the Germanic 'gable' word may be unrelated.
[PCR]
Yes, perhaps not as strong.
[PG]
(5) PIE has *gerbH- 'carve, notch' plus
several "scratchy" roots like *skrebH-, all of them no doubt
onomatopoeic
[PCR]
Ridiculous!
and thus of little use in distant
comparison. As for <zrák>, why cite only a Czech word if the root in
question is found everywhere in Slavic and a precise reconstruction is
possible? The original meaning of the word is 'sight' (the semantic
development as in German Gesicht), derived from the Slavic verb *zIr-E-ti
'look, see; appear, be visible', from *g^Her- 'shine' (Slavic *zorja 'light in
the sky'). I think you lump together unrelated items here.
[PCR]
I think the idea is rather 'become
visible by being scraped'.
Pat