Re: Languages and Genes in Collaboration: some Practical Matters

--- Richard Wordingham <richard@...>
wrote:

> --- In cybalist@yahoogroups.com, "mkelkar2003"
> <swatimkelkar@...> wrote:
> >
> > --- In cybalist@yahoogroups.com, "mkelkar2003"
> <swatimkelkar@> wrote:
> > >
> > >
>

http://www.humis.utah.edu/humis/docs/organization_919_1166141662.pdf

> > >
> >
> > Don't we have our backs to the wall?! It is worth
> reading this passege
> > from Campbell over and over again.
> >
> > "The ancestor of English and Hindi did not begin
> to diversify into
> > separate languages
> > until some 5,000 or 6,000 years ago, but they
> share only some five
> > clear cognates on the
> > Swadesh 100-word list (Campbell and Poser in
> press).
>
> Note the words: *clear* cognates.
>
> > English and Hindi are "clearly related" because we
> the comparative
> > lingusits say so. They did not start diversifying
> until 4000 BCE
> > because we the comparative linguists say so. And
> yet the fact is they
> > have only 5 cognates on the Swadish list ...
>
> No. The others are just not so clear.
>
> thou - tu: - clear (but a bit dishonest, as 'thou'
> isn't actually in
> the English list.)
> who - kaun - not clear
> what - kya: - not clear

But if you see both pairs together, you notice the
wh-k parallel

> not - nahi~: - Or is this just coincidence?
> two - do: - clear
> horn - si~:g - not clear
> feather - par - looks clear - or am I wrong?
> tooth - dã:t - clear
> foot - pair - clear
> know - ja:nna: - not clear
> sun - su:raj - not clear
> star - (si)ta:ra: - clear
> warm - garam - not clear?

clear!

> full - pu:ra: - clear
> new - naya: - clear
> name - na:m - clear
>
> I score 7 honest clear cognates.
>
> There are another three I am not so confident about:
> eye - ã:kh
> (finger)nail - na:khu:n

Not unless you know Anglo-Saxon or another Gmc
language

> nose - na:k - Parallel stems?

but nark and na:k are definitely clear :>

I would have grouped words with similar consonants
together first and then taken a look
e.g. two next to tooth, etc.

>
> There's a crude matching technique that Alex
> Manaster Ramer used on
> Bengali and English. The statistical analysis -
> using a Monte Carlo
> techniques - showed that despite an apparently low
> number of cognates,
> the match was way better than random. The only flaw
> in the analysis
> would be that it would be misled by, say, a tendency
> to have /n/ in
> words for 'nose' and /m/ in words for mouth.
>
> Richard.
>
>

____________________________________________________________________________________
Looking for last minute shopping deals?
Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping