Re: [tied] IE *de:(y)- 'bind'.

New counts:

I had not anticipated the interest that would be attached to the
numerical facts about the percentage of CVC roots among root
presents and aorists, so I just took the LIV I had at hand, which
turned ut to be the first edition. It also seems I did not count
very accurately. I apologize for any inconvenience thereby caused.
Using the revised edition and counting more attentively, I now get
the following:

A. A root present is formed by: 158 roots (153 in the list proper +
5 in the list of Wurzelstativ with zero grade).

B. An acrostatic (acrodynamic, lengthened-grade) root present is
formed by 64 roots (52 in the list proper + 12 in the list of
Wurzelstativ with full grade).

It is a matter of ultimate doctrine whether one will like to pool A
and B.

There are 28 CVC roots in A (none in the Wurzelstative), 16 in B (22
if Wurzelstativ is included).

C. There are 409 roots forming a root aorist. 86 have the shape CVC.

The percentage of CVC roots among the root presents is 17.7 (28 out
of 158); it is 22.5 if the acrostatics are included (50 [28 + 22]
out of 222 [158 + 64]).

The precentage of CVC roots among root aorists is 21.0 (86 out of
405).

It may be called a matter of very delicate taste whether there is a
significant difference in probability between the two guesses that a
given CVC root forms a root present or a root aorist. If acrostatics
are excluded there is indeed some greater probability that it is an
aorist (21.0 against 17.7); if acrostatics are included the balance
tips to a slight majority in favour of a root present (22.5 against
21.0).

I for one find it very hard to ascribe any importance to these
figures.

The concept of the IE root underlying the entries in LIV is the same
as in Pokorny's IEW, except that laryngeals are added.

Jens