From: Patrick Ryan
Message: 42895
Date: 2006-01-12
----- Original Message -----
From: "tgpedersen" <tgpedersen@...>
To: <cybalist@yahoogroups.com>
Sent: Thursday, January 12, 2006 7:57 AM
Subject: Re: [tied] PIE suffix *-ro - 'similar-with'
<snip>
Now if the PIE *-n-, *-t-, *-nt-, *-r- suffix is actually a *-Nd-,
that would set it apart among the other PIE glosses and morphemes.
Semantically it fits beautifully with those East Asian V..n "thing
V'ed" suffixes. An extra piece of circumstantial evidence is the
phonetic and semantic fit between the East Asian 'instrument'-prefix
construction s-V vs. the Semitic causatives of the form s-V and the
PIE verbs with s-mobile and sometimes suitable semantics, eg. Engl.
melt : smelt.
Cf. the fate of Romance '-able' in English.
Torsten
***
Patrick:
I think you are definitely on the right track here, Torsten.
I suppose almost all of us have struggled with puzzling out with
significance the PIE s-mobile prefix (not many of those) might have had.
I suppose the earliest language had 'verbs' which did not formally
distinguish between transitive and passive usage, this being indicated by
the number of adjuncts: one = passive; two = transitive.
It looks to me as if one method of narrowing the meaning to transitive in
situations where the mere transitive activity was desired to be indicated
without specifying an object was to append a pronominal or demonstrative to
suggest it; alternatively, one that indicated an agent.
The data you have collected indicates a preferential inclination to regard
it as an agent affix.
In my opinion, this is also the indication provided by Egyptian causatives
in s-; I would expect Egyptian *z- if an object was being indicated.
I am fairly certain that that accounts for some instances of PIE s-mobile
like the example you give: 'smelt'. The question is: is that an agent or an
object indication?
However, this does not seem to work well for some. And here, I believe, the
preferred explanation should be an 'energeticus', which, to make our work
more difficult, in PIE is also *s-.
***