> taliesin the storyteller wrote:
> > Basically, there are verbs that take can take subclauses as arguments,
> > here are two, seem and try:
> >
> >     1) "David seemed to leave"
> >     2) "David tried to leave"
> >
> > The first is a raising-verb, the second is an equi-verb.
> But I don't get what "equi-verbs" are supposed to be, unless it's
> "something that looks like a raising verb but isn't".

Contrast LIKELY (raising) and EAGER (equi):

   David is likely to leave.
   David is eager to leave.

"David" is a semantic argument of "leave" and of "eager" but not
of "likely". Likewise contrast CONSIDER (raising) and PERSUADE (equi):

  I considered David to be well-behaved.
  I persuaded David to be well-behaved.

"David" is a semantic argument of "(to be) well-behaved" and of
"persuaded" but not of "considered".

It still has not been settled to what, if any, extent there are
syntactic differences between raising and equi in English. But
from a conlanger's p.o.v. it is clear that whereas raising is
a grammatical quirk, which a conlanger can simply ignore (--
just say "That David leaves is likely" and "I considered that
David is well-behaved"), equi is something the conlanger does
have to find a way to deal with.

* * *

Livagian has raising, but as a grammatical construction that is
not an idiosyncrasy of lexical complementation. So for example
in addition to the raisingless

   ju      fmyhgh  li
   [begin [sleep  [me]]]
   "I begin to sleep", "That I sleep begins"

one can have

                 li   ju     fmygh
   [NULL-RAISER [me] [begin [sleep]]]
   "I begin to sleep"

where "li" is a raisee complement (a semantic nonargument) of
NULL-RAISER, and "fmygh" assigns its sleeper role to the
nearest higher raisee. One can also have:

    ju                  li   fmygh
   [begin [NULL-RAISER [me] [sleep]]
   "I begin to sleep", "'It begins me to sleep'"

As for equi, subject equi like "I want to sleep" is handled
as "I want that I sleep", analogous to how "I want you to
sleep" is handled:

   "I want you to sleep", "I want that you sleep":

    mahw  li   fmyhgh lu
   [want [me] [sleep [you]]]

                 li   moh   fmyhgh lu
   [NULL-RAISER [me] [want [sleep [you]]]]

    mahw  li                lu    fmygh
   [want [me] [NULL-RAISER [you] [sleep]]]

                 li   moh                lu    fmygh
   [NULL-RAISER [me] [want [NULL-RAISER [you] [sleep]]]]]

   "I want to sleep", "I want that I sleep":

                 li   moh   fmygh
   [NULL-RAISER [me] [want [sleep]]]

For obj equi like PERSUADE, there is a lexically-specific
construction whereby the predicate "persuade" in "I persuade
you to sleep" has three arguments, "me", "you" and "THE
PROPERTY OF BEING X SUCH THAT X sleeps" (capitalized elements
are phonologically null), and the lexical entry for "persuade"
specifies that the x variable in the 'outcome' argument is
bound by the persuadee argument.