Re: [tied] Re: i-verbs in Baltic and Slavic

From: Miguel Carrasquer
Message: 44703
Date: 2006-05-25

On Thu, 25 May 2006 13:08:53 +0200, Piotr Gasiorowski
<gpiotr@...> wrote:

>On 2006-05-25 11:06, Miguel Carrasquer wrote:
>
>> On Wed, 24 May 2006 08:20:38 +0000, Piotr Gasiorowski
>> <gpiotr@...> wrote:
>
>> I wasn't suggesting there were problems. I was merely
>> ponting out that -itus is not something we only see in
>> causative verbs.
>
>However, one of your two examples (<domitus>) _is_ most likely derived
>from an original *-éje- verb!

Right: domh2-éye- > domáye- > doma:-. In <domitus>, it
looks like the causative suffix is indeed completely
deleted: *domh2-tós > domitus.

If I may indulge in a small syntactical excursion regarding
the causative passive: I was a bit confused by the concept
at first. If we take a transitive sentence like "X V's Y",
where X is the Actor and Y is the Object (and V is a verb),
the passive promotes the Object to Subject, and relegates
the Actor to Oblique ("Y is V'd by X"). The causative adds
another Actor, whereby the original Actor becomes Object,
and the Subject becomes Oblique ("Z makes X do V to Y").

If we combine Passive and Causative we really have two
possibilities. We apply the passive rule first (A -> obl; O
-> S) and then the causative rule (S/A -> O, O -> obl; + new
A):

pass caus
X (A) => (obl) => (obl)
Y (O) (S) (O)
Z (A)

(Z makes Y be V'd by X : causative of a passive)

Or we do it the other way around:

caus pass
X (A) => (O) => (S)
Y (O) (obl) (obl)
Z (A) (obl)

(X is made to do V to Y (by Z): passive of a causative).

>> I think Balto-Slavic -i"ti/-ýti and the Vedic future in
>> -ayi-s.yá- show that the causative morpheme, when athematic,
>> was *-(e)ih1-. The laryngeal was regularly suppressed in
>> the more common thematic forms (*-éyh1-e/o- > *-éy-e/o-),
>
>By the way, why are they so consistently thematic? The verb 'throw' isn't.

When it means "to do, make" it is consistently thematic (if
the term "consistent" can be applied to a single verb
(iyami/iyezzi) in a single language/branch
(Hittite/Anatolian)).

>> so
>> it wouldn't surprise me if a secondary zero-grade form /i/
>> was abstracted from that.
>
>Just in case you think I'm hostile to you idea, I'm not. It's an
>impressive analysis, even if it isn't the whole truth. I'm myself
>interested to see which of the competing explanations gets closer to
>explaining _all_ the branch-specific problems. The idea that Indic -p-
>may be a prefix is really brilliant and _almost_ makes me accept the
>compound analysis, except that there seem to be no traces of prefixation
>outside Indic (not even in Iranian). I'm planning to experiment with
>various compromise solutions and if any of them seems to work, I'll
>compare notes with you immediately :)

Please do. I dzie,ki.


=======================
Miguel Carrasquer Vidal
mcv@...