Re: PIE Word Formation (2)

--- In cybalist@yahoogroups.com, "Brian M. Scott" <BMScott@...> wrote:

>
> At 5:57:48 PM on Friday, March 31, 2006, Richard Wordingham
> wrote:
> > I think Patrick is right in thinking that they are not
> > examples of dissimilatory loss. Both words are naturally
> > subject to reversible syncopation. Iin the first case, you
> > have library > libr'ry (syncopation) > lib'ry
> > (degemination) > lib'ary (desyncopation). Does the second
> > example occur in rhotic dialects?
>
> Yes, though not nearly so often as 'liberry'.

Note that any accent in which (unlike RP but like most varieties of US
English) <library> has a full second vowel, "liberry" can't be due to
syncopation plus epenthesis and must be a genuine example of
dissimilatory loss. Note also the common dropping of the firt /r/ in
words like <surprise> ("supprise") in rhotic accents.

Other typical examples:

Lat. cri:brum, proprius > Sp. cribo, propio (exactly like *h2r.g^-ró-

> *h2r.g^ó-)

PWGmc. *breura- > *beura- 'beer'

PIE *k^r.h2-sr-Vn- (Lat. cra:bro:) > PSl. s^Irs^en-

-- and in PIE itself there are at least two certain examples, where
the dissimilation is visible both in Indo-Iranian and in Old Irish:

*tri-s(o)r- > *tis(o)r-

*kWetwr-s(o)r- > *kWet&s(o)r-

Piotr