Re: Root reversal in Dravidian

Talked to my friend who's a native Tamil speaker/slavicist/linguist, and she
says that the transliteration here is
"messed" up.   The word for "attic", of course, should be immediately thrown
out, but additionally, apparently
the Wikipedia is wrong on the transliteration of "look": the /n/ should be
retroflex, whereas the /n/ in "peep"
should not be.   So it looks like the only one there is "gather" and "join",
and even though those could be seen
as semantically similar (depending, of course, on the *exact* definitions of
"gather" and "join" [i.e., what if by
"to gather", they mean "to collect", and by "to join", they mean "to sign up
for something"?]), one pair is hardly
enough to base a pattern on.   In other words, I'd say that what's on the
Wikipedia is *highly* suspect.

Also, with the palindromes, the first two mean "mom" and "dad", and it seems
to be rather common to have
something that either looks like reduplication or a palindrome for pet names
for "mom" and "dad".   My friend
(Vrinda) is her name sites Hebrew as an example where the words for "mom" and
"dad" are palindromes.   It
also happens in Korean, where the words are, as it turns out, exactly the
same: ama "mom"; apa "dad".   (I'm
unclear about the consonant.   It might be the so called "superaspirated" /p/
there.)

[Interesting aside: The word for "dad" in Georgian is "mama", and the word
for "mom" is "deda".]

Interesting idea, though.   I once toyed with the idea of having consonant
reversal indicate plurality.   You'd
probably have a bunch of words where you didn't know whether it was a
singular dog or a plural house, for
example, but it might be neat to see fully done out.

-David
*******************************************************************
"sunly eleSkarez ygralleryf ydZZixelje je ox2mejze."
"No eternal reward will forgive us now for wasting the dawn."

-Jim Morrison

http://dedalvs.free.fr/