Re: THEORY: Meanings of Verbal Accidents.

Hallo!

On Mon, 13 Aug 2007 20:10:43 +0100, R A Brown wrote:

> Elliott Lash wrote:
> [snip]
>
> > In any event, I think this is fairly common.  In Latin
> > for instance, the same set of morphemes can be either
> > active or passive, depending on the verb they are
> > attached to. This set of morphemes is usually called
> > 'passive' or 'deponent', but the deponent is really
> > just an active use:
> >
> >   deponent                   passive
> >  mori-or 'I die'       ~   am-or  'I am loved'
>
> Umm - I fail to see how "I die, I am dying" can possibly be considered
> _active_! Surely, if anything is passive, dying is.
>
> (I discount suicides from this - but they are a tiny minority, and then
> the reflexive 'I am killing myself' is perhaps more appropriate.)
>
> It has always seemed to me that Latin is rather more true to reality in
> giving the verb "to die" passive endings. I have yet to be convinced
> that 'morior' should be classified as a deponent verb any more that
> 'nascor' (I am being born) should be. The babe that gets thrust into int
> the world from its mother's womb s hardly the agent. At least with this
> verb English uses passive forms as well as Latin; yet, strangely
> anglophone Latin textbooks still list 'nascor' as a deponent - weird!
Right.  There is no agentivity to dying or being born.  These verbs are
considered "deponents" mainly for formal reasons: they inflect like
passives, but the corresponding "active" forms *morio and *nasco are not
attested (though they could be understood as 'I cause to die' and 'I give
birth', respectively - but the dictionary doesn't know them), at least not
in Classical Latin (Late Vulgar Latin, in getting rid of the deponent verbs,
innovated such forms).

We are getting into the murky waters of morphosyntactic alignment here.
Latin is a nominative-accusative language: semantic agency does not
really matter that much to morphological marking of argumants, but
rather "subject-hood" does.  There are plenty of stative verbs that take
active endings even though there is no action involved semantically.

(ObConlang: Old Albic does not commit such ridiculosities.  It is a
fluid-S language, where morphological marking follows the semantic
roles of "agent" and "patient" rather than "subject" and "object".
'I die' is _maraha_, with an ending that expresses a first-person
*patient*.)

... brought to you by the Weeping Elf