Re: THEORY: A possible Proto-World phonology

As someone who is not a fan of Nostratic at all, I'll give give a couple quick
little comments.  I'm primarily a Native Americanist (but a recent one)

Danny Wier wrote:
>And regarding the Nostratic theory in general: I do accept at least some of
>it, even to the point where I can call it a hypothesis instead of a theory,
>for these reasons, expressed in these parallels:
>
>1) the three-way voiced-voiceless-ejective stops/affricates, found in
>Afro-Asiatic and Kartvelian, and altered in Indo-European -- somewhat
>preserved in Korean as well, if it is indeed an Altaic language
Also found in "Na-Dene"[*], Wakashan, and Salishan families.  Probably a
lot of
others, so there is no good reason for a connection here.  Not that unusually,
typologically speaking.

>2) common pronouns and case endings in Indo-European, Uralic and Kartvelian
This is an interesting point.  But I'm quite happy with an IE - Uralic
connection.  I don't know anything about Kartvelian, so  I have no opinion
there.

>3) animate-inanimate gender in IE (Anatolian) and Dravidian
Found in some of the Na-Dene languages.  In fact, Eyak and Haida plural
marking
is dependant on whether a noun is animate or inanimate in very complicated
ways.  This kind of gender split is not unusual.

>4) evidence of "laryngealism" in IE and Kartvelian and a possible link to
>the "laryngeals" of Afro-Asiatic
>5) ergativity in Kartvelian is also attested in early IE
Ergativity is also found in the Wakashan languages.  I think Salishan too, but
I'm not sure.   Ergativity in IE is debated, of course.  A few of my friends
studying IE don't believe it, but I don't know their reasons.  IE has also
been
claimed to be active-stative.

>6) vowel harmony in Uralic and Altaic, and ablaut in IE
Also not uncommon in the world.  You're also making quite a leap to connect
vowel harmony and ablaut, IMHO.

>7) IE k^/k/kw, g^/g/gw, g^h/g/gwh may have resulted from neutralization of
>vowels to /e/ (which of course mutates to /o/ and zero due to stress shift
>and other factors); the original CV groups must have been ke/ka/ko, or
>ki/k@/ku (and likewise for g and gh).
Don't know anything about this area.

>8) cognates (or possible cognates) I've found in Germanic and Semitic
You've found words that follow all the phonological rules leading back to
Proto-Germanic, proto-IE, Proto-Semitic, and then have regular correspondances
between the latter two?  Maybe it's just me, but I've always be skeptical of
those types of connections, unless they are overwhelming.

>9) masculine and feminine genders found in Afro-Asiatic and IE
Once again, common in the world.  I also find it curious that you've mentioned
two different gender systems for what is supposedly one group.

>10) VSO word order found in Celtic and Semitic
There are only six logically possible orders of S,V, and O; three of which are
hardly ever found.  All three common orders are found in IE, so word order
could be grounds for a connection with any family in the world.  In short,
hardly indicative of anything at all.

Most the things you've mentioned are typological, and can occur completely
independently in any number of languages.  This is hardly good grounds, IMHO,
for serious scholarly proposals.  Now conlanging, that's another story.  I was
really amused when I first read about Dolgopolskij's Nostratic "hunting" story
-- can't imagine a better use of artificially creating a proto-language.

Just my two skeptical cents worth.

Marcus

[*]  I quote "Na-Dene" because I don't believe it is really a family.  I've
done some investigation into the question, and I think the entire idea is
misguided.  Mainly because there are virtually no possible cognates in the
"core vocabulary".