Re: comments on CONN.TXT

[Jorge Llambias <jorge@PHYAST.PITT.EDU>]

And:
> 7. 13.7 "An imperative sentence [lio] is true if the command is obeyed".
> The discussion is potentially misleading. It should be explained that
> one could obey the command not only by bringing tea, but also by instead
> making this not be coffee. Therefore it is only an approximate and
> misleading translation of the English.

I'm glad you think that. The fact that {do} is an argument of the
command is purely circumstantial, that's why I think that an attitudinal
is much better than {ko} for imperatives. In other words, "act so as to
make it true that X happens" does not need {do} to be an argument of X
in general.

> 10. 14.16. (a) Is there a non-connective version of fahu? As in
> "the two men love their respective spouses".

Unfortunately no. Since I have no use for {le'i} and co., I might
start using them for that function.

> (b) How can one tell
> which series matches up with which? Is it glorked from context?

I would say so.

> 11. "The place structure of "casnu" is: the mass x1 discusses/talks about
> x2 so the x1 place must be occupied by a mass". This seems nonsensical,
> unless the x1 of casnu is merely arbitrarily required to be a mass. So
> far as I can tell, only gunma's x1 needs to be a mass.

{casnu} is supposed to mean something like "x1 interchange opinions
about x2". It doesn't make much sense to have a single person casnuing,
unless they are talking to themselves. The restriction to being a mass
is nothing but that, I think.

> 16, 19.3 "mi jinvi le du'u loi jmive zvati gi'onai na zvati la .iupiter"
> why don't {jmivezvati} form a tanru?

They do. It should be {cu zvati}.

> How come {la .iupiter} is a
> sumti of both {zvati}?

It isn't (explicitly). It should be {zvati gi'onai na zvati vau la iupiter}
for that.

> How come {gionai} conjoins the two {zvati}
> rather than, say, {duu} or {jinvi} with the second {zvati}?

Because there's no terminator. To conjoin with {jinvi}, it should
be {kei gi'onai}. You can't conjoin using GIhA under LE.

Jorge