-(t)er

I borrowed another book at the library (oh no!). This time on
Hittite.

How come *s- and t- of *so- and *to- are considered to be
demonstrative stems, when they are also connectives? Wouldn't it be
more economical to see them as only connectives?

The fused morphs that originate in Wackernagel's position migrate to
other places with other meanings, but someone forgot to take them
apart again, thus *tom > Latin tum "then", but *tom > Germanic *Tan-
"him" (acc.)

Torsten