From: fournet.arnaud
Message: 58297
Date: 2008-05-03
----- Original Message -----
From: "Francisco Antonio Doria" <doriagen@...>
>If I correctly recall, Greenberg proposed an unifying scheme for
>African languages.
==============
I suppose this is not the place to discuss African languages,
so I will focus on the methods of Greenberg and his followers.
I once bought a book about Uldeme, a language assigned by Greenberg and Co
to Chadic PAA.
I read it and when I put back the book on the shelf,
I wondered why this language could be classified as PAA,
when nothing in Uldeme looks like anything that vaguely reminds of PAA.
Recently I discussed with one of Greenberg's Followers.
I will not give his/her name, for charity's sake.
I wrote this :
>> I have previously expressed serious doubts that Uldeme (Wuzlam) is part
>> of PAA.
> > I'm still interested in knowing what are the particular reasons this
> > language deserves to be within PAA.
> > I must state that I definitely disagree.
> > http://www.livingreviews.org/wals/languoid/lect/wals_code_uld
> > I'm interested in your opinion about this particular language
> > in connection with the issue of the "limits and perimeter" of PAA.
I got this as an answer :
> thanks for expressing your opinion about Wuzlam of the Cameroons.
> I have no opinion about Wuzllam at all. I have never heard of it at all.
> Do you know where some real data on Wuzlam can be found?
> It may turn out to be an important language. Who knows?
> Merci!
> PS If it is Chadic, then the odds are that it belongs to the Central
> branch.
> Guessing by location. [I trimmed the post for clarity]
In other words, Greenberg and Co have no idea what Uldeme/Wuzlam looks like.
But this language has been assigned to the Biu-Mandara branch of Chadic.
They have invented a new method : Mass-comparison _without_ data.
As regards anything these guys have (claimed to have) done,
I would warmly advise the utmost prudence.
I think Amerind is not even worth contempt,
Eurasiatic is in my opinion wrong.
And PAA and African languages amount to classification without data.
Unifying scheme ?
Wow.
As a matter of fact, if you ever tried to muddle thru 300 pages of data,
you would know that to classify languages when one has no data is really
cake.
Data are the cause of trouble in that field.
That's today's bad news.
I will refrain from using my favorite 4.letter words starting B* S* or J*.
Arnaud
===========