Re: [tied] tt/st/ss

On 2007-06-18 22:32, stlatos wrote:

> Why would root nouns be used for tools?

Becuse instrumental nouns are semantically akin to agent nouns (cf.
Eng. cutter, which may mean 'someone who cuts' or 'something that cuts').

> Why does it only happen with roots ending in -r?

I didn't say it did. You gave two examples with final *r, but there are
similar formations from other roots, e.g *se:to: 'ambush, place for
sitting' (OE sæ:t, ON sát, etc.).

> Why are words from the same roots with the same meanings found that
> end in *-tlo-?

Likewise, *se:to: has a meaning similar to Gmc. *seþ/tla-, Gk. hédra:,
etc. Why? Because IE derivational morphology is quite rich and virtually
synonymous nouns can be derived from the same verb root in a variety of
ways.

> Why aren't there any words that show another outcome of *rtl in
> Germanic?

See below.

> There are already at least 10 different outcomes of *-tlo- in
> Germanic (more if you believe in a tro/tlo-identity). Among all the
> IE languages, there are dozens more. If there were an example of
> -rþl- that proved a more regular outcome of -rtl- I wouldn't have
> made this claim.

Since I accept the hypothesis that a liquid in the root made the suffix
surface as *-tro-, I'd expect only *...r-tro-, yielding *-rþra-, as in
ON arðr.

> It seems that l>mora/r_ isn't too odd considering there are no other
> examples of rl.

No _other_ examples? That would mean there is no independent support for
such a change. Other sonorant clusters may develop into geminates but
don't cause vowel lengthening.

Piotr