> > Also, what theory do you subscribe to
> > as to the proto-Germanic 70-120?
>
> See
>
> http://groups.yahoo.com/group/cybalist/message/6791
>
> and the ensuing discussion between Miguel and me.
Good stuff. You should get it published. Curiously, there is a parallel in
Limbu of the "extra" 100 added to the tens. Limbu has 10, 20, 30, 40
as in the pattern 1-10 (!), 2-10, 3-10 and 4-10. But 50, 60, 70, 80, 90
are 5-100, ..., 9-100 whereas 100 itself is 100-1 (and presumably 200 is
100-2 and so on but the source _A Grammar of Limbu_ by G. van Driem does
not tell me). The tens with added units do not have this superflous 100
and just look like unit+unit i.e 53 is 5-3 modulo minorphonetic changes.
In fact, it seems to be quite common with numeral systems in that area
to have a "zero" with it's tens. The three I have looked at i.e Limbu
with an extra '10' or 'hundred', Tibetan with an optional element
meaning 'completed' and Kulung with a longer form of 20-90 than that used
in 21-99. Even though neither of these three can be shown to go back
etymologically to something meaning zero or empty (as with e.g Chinese)
they exhibit the structural property of being zero.
> (A curiosity: Crimean Gothic had <sada> '100', an Iranian loan.)
Speaking of which, which I bet has been discussed before, what do you
think of the origin of OCS sUto?
all the best,
Harald