Re: [tied] Re: Bader's article on *-os(y)o
> I can see *iddem reduced, even if irregularly, to /idem/ but the other
> way is logically problematic.
The source is not just id-em but also eod-em, ead-em. When final -d is last
on the ablative, the last two are reanalysed as eo-dem, ea-dem. This is the
real source of the -dem suffix.
The -em suffix is seen on a number of words in Latin (tamen, enim, autem
etc, and possibly the adverbs in -im). Pronouns, however, tend to have
the -dem suffix instead, although doublets do exist (tamen and tandem).
This indicates to me that the -dem suffix is secondary, the -em suffix