Dear Glen,
I must apologize, if because of my poor English I failed to express my
position clear enough.

[Glen]

> Alexander:
> >It's a pity that both you and the opponent have a _geographically_ >based
> >preliminary classification of the Nostratic families ("Steppe >languages"
> >and "East Nostratic" respectively).
>
> I base my classifications on the shared grammatical features I outlined,

not

> geographical positions. You didn't read carefully enough if you missed my
> points on grammatical commonalities between these languages.

I didn't have a thought, that your classification was based _mainly_ on
extralinguistical data. However you brought into the world something more
valuable than a product of pure liguistic analysis - basing initially on
linguistics you took then into consideration geography, history, archaeology
... etc. and suggested a scheme or a model of spreading languages in a
large part of Eurasia. A model is not only a result of handling row data. It
is also a tool which allows to obtain new information. For example, if some
parallels between IE and Uralic are not clear enough, we accept those of
them which are supported with Altaic material and deny those which
contradict to reliable facts already established from IE-Altaic comparisons.
In other words, after we accepted a model our comparativistic material (I
mean the material we actively use) became more rich and less pure, it is
influenced by our hypothesis (if the model was erroneous, the material in
the focus of our attention became "worse" than it was before).

IMO, when comparing Uralic & IE versus Uralic & Dravidian we need to use
arguments not influenced by our models. That is why I suggested to "forget"
them again for a while _at this stage_ of analysis. Otherways having once
accepted a hypothesis we risk to stay a slave of it forever.

Glen, please don't take this as a hint that your arguments are not pure
enough. These were just methodological considerations.

[Glen]

> Alexander returned with:
> >Just a thought, perhaps a very naive one.
> >Look:
> >Dravidian is odd;
> >Altaic is very odd;
> >Afroasiatic is rather odd too;
> >[...]
>
> Everything is apparently odd to you.

Not to me - to different linguists. I'm not a linguist, I try to approach
the ethnogenetic problem from other side (archaeology, mythology, ecology,
paleobiology etc.). I need to buy at the philological market a conception
which I could trust and relay on. Up to now the "classical" variant by
Illich-Svitych looks most attactive for me (there are strong extralinguistic
considerations in support of a model basing on it). Unfortunately, I'm not
so erudite linguistically to judge whether its arguments are strong enough.
That's why I'm interested in opinions of other linguists and in comparing it
with alternative variants.

[Glen]

> Interesting how you can critique me on
> that erroneous perception that I'm basing my classification on

geographical

> positions alone, yet, you turn around and do something much worse by
> describing languages in unexplained, impressionistic terms and using this

as

> a basis for further speculation.

I have a great lot of speculations (Like many of us, I believe. Any
internally incontradictive hypothesis basing on the deficiency in facts
available at the moment is a speculation, isn't it? After new facts arrive
it becomes either theory or rubbish). But in this case it was just a
question, a though aloud.

[Glen]

> I agree that IndoEuropean is the most widely studied to the detriment of

the

> other families, but I still fail to understand what "odd" is supposed to
> mean in your usage. Altaic and Dravidian philology have already been
> "invented", [...]

I'm glad. Are there in those philologies any new categories which are not
applicable to IE languages, which loose sense there?
It's not polemics, I just wonder.

Alexander