Re: Help: Zhyler ECM/Raising Verbs (Longish)

David Peterson:
> Zhyler has a whole bunch of suffixes.  For verbs, most of the suffixes
> are aspectual (finsishing something, starting something, ability, etc.).
> At least one, though, works as a (subject) raising or ECM verb would.
"want" looks to be control/equi, not raising. Offhand, the only English
example of a verb alternating between obj raising and subj raising that
I can think of (though maybe examples are copious & it's been too
long since I thought about complementation) is:

  She made him her husband ("She married him")
  He makes (her) a good husband

Control/equi strikes me as a much more challenging problem than
raising. E.g.
  He persuaded her to go.
  He longs to go.
The challenge is of course that an argument of the higher predicate
is an argument of its argument-sister, the lower predicate.

> The problem that arose was with sentences like "I want you to do x" or
> "I want the man to do x".  You can't just add the suffix to the verb,
> because it doesn't work if the speaker is different from the subject.
[...]
> Everything else I have no idea what to do.  Here are some of my ideas:
>
> (1) The direct object of the verb "to see" is the one who's wanted to
> do x.  This would be a kind of ECM construction, since the subject of
> the action would be in the accusative case.
[...]
> (2) The direct object of the verb "to see" is the whole sentence itself.
> So this would be like a subject raising construction: "I want [you to eat
> the food]" or actually "I want that [you eat the food]".  If this is the
> case, though, does the whole phrase get an accusative tag?  Each member
> of the phrase (note: Zhyler can double-case mark)?  Also, how can a verb
> be an object?
[...]
> (3) The whole seeing phrase is a relative clause about the subject of the
> embedded clause.  This would be like, "The I want him to man eats the
> food" (?).  I'm not even sure if this can make sense.
>
> (4) The embedded clause is the object of the verb "to see", but rather
> than being a verb, it's a verbal noun.
If you're keen to explore naturalistic weirdnesses, then you could
choose any of them, but logically (2) looks the most sensible. Or
at least, the most sensible thing would be to handle these cases in
the same way as you handle predicates like "know that, believe that,
forget that, insist that". As for where the case marking should go,
again there's room for naturalistic weirdnesses, but the logically
most straightforward option would be to make it the outermost
casemarker on the root of the subordinate clause (presumably the
verb).

--And.