tlhIngan-Hol Archive: Tue Dec 26 09:21:34 2000

Back to archive top level

To this year's listing

[Date Prev][Date Next][Thread Prev][Thread Next]

RE: Grammar Highlight Each Day (bo-)

> -----Original Message-----
> From: Jiri Baum [mailto:[email protected]]
> Sent: Sunday, December 24, 2000 2:28 AM
> To: [email protected]
> Subject: Re: Grammar Highlight Each Day (bo-)
>
>
> > >what did you think {luHoHwI'} means?
>
> Alan Anderson:
> > The only thing I can possibly interpret it as is "they who kill
> him", but
> > it gets there through the odd literal meanings of "one who
> > (they-kill-him)" or "(they-kill-him)er".  It's a "givation" sort of
> > thing, where it has only one conceivable meaning, even if it sounds
> > completely wrong.
>
> DaHoHwI' sounds to me like ``agent of your killing'' - though I can't
> decide between a causative and a performative agent.

You are missing ghunchu'wI''s point. In English, you don't call a "waiter" a
"he-waits-on-you-er". For that, you'd say {Dutoy'bogh nuv}. You are trying
to turn a verb with {-wI'} into a relative clause. This suffix was never
built to do that. I suggest with some degree of confidence that you will not
find any example in canon of Okrand using any prefix on a verb with {-wI'}.
I suggest that it is not that a prefix is optional. I suggest that no prefix
is allowed under any conditions. If you want a relative clause, use {-bogh}.
If you want to indicate the doer of an action as a simple noun, use {-wI'}
and omit any prefix.

> That is, does the DaHoHwI' make you kill, or does he kill for you?

I think you may be stumbling into the reason that prefixes are not allowed.
There is only one noun involved with a verb with {-wI'}. That noun is not
the subject or object of the verb. It is merely a noun which is named after
the action that the noun performs as part of its duty or function. The
action gets {-wI'} applied to it and this becomes a label for the noun.

If you want to describe the interaction of multiple nouns (hence requiring a
prefix), then you want a relative clause, not {-wI'}.

> It would have to be a very strange context, either way.

It is gibberish.

> Jiri
> --
> Jiri Baum <[email protected]>
> As we all know, real error messages have two parts:  a message code, and a
> return code. Ideally, the message code is hexadecimal, the return code is
> octal, and the manual explaining the error messages uses decimal. --r.h.f

SarrIS

Back to archive top level