On 2007-01-01 14:52, raonath wrote:
> This claim has always bothered me, as is the related claim that
> the aorist shows more irregularities than the present. This is
> patently false for Sanskrit where the most irregularities are
> seen in the root present, which far outnumber root aorists.
As for PIE, Meier-Brügger's statistics, based on the LIV data, should
perhaps be taken with a small grain of salt, but nevertheless he finds
the following numbers of reconstructible PIE verb stems in each of the
relevant types:
ordinary root presents: 139 (102 "Fälle sicher")
acrostatic (Narten) root presents: 46 (31)
root aorists: 392 (265)
The two subtypes of root presents counted together are significaltly
less numerous than e.g. nasal presents or *-éje/o- causative/iterative
stems, not to mention the *bHéreti type and presents in general. By
contrast, root aorists account for ca. 67%-78% of all aorists (and the
sigmatic type accounts for most of the rest).
Piotr