Jens:
>> If it counted for something, I'd see an IE reconstruction. I asked
>> for it some posts back, you don't show it.
>
> The IE form would be *id-em. Anybody can do the maths.
Indeed. Math like this?...
m. i-dem + f. ea-dem + n. i-dem = Jens analysis is wrong
Either you're wrong or the following sample of a multitude of
links across the globe are wrong:
http://www.utexas.edu/cola/depts/lrc/eieol/latol-5-R.html
http://www.uark.edu/depts/latin/pronomina.html
http://unholypage.topcities.com/review/rev1_a-e.htm
http://mnatal.members.easyspace.com/mrnmisc/delectus/mrnmilan.htm
The first link shows all the forms of Latin 'idem' in case you
forgot them. I mean, how do you analyse /earumdem/ then? Is it
*earumd-em? :) I don't think so. Why would an inanimate *d spread
to all forms?! That paradox in itself terminates this discussion.
But what does Sanskrit say? There are many pronouns in Sanskrit
that end in -am (its own little innovation). So we have /vay-am/,
/yuy-am/, /ay-am/ and /id-am/. So that means...
Your credibility just went down the tubes. There is only *id in
Common IE.
= gLeN