Re: sumti categories

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

la lojbab cusku di'e

> sumti places are explicitly NOT limited to one category.  Pragmatically, it
> may be rare to use a category 'error', but we don;t want to forbid it.

I agree that any rule would be just pragmatic, but since different categories
give different meaning to the gismu, to know the category is useful in order
to understand the meaning of the gismu. For instance, {djica} means
"x1 wants x2", where x2 can only be an event. If this was not explicited in
the list, I might think that an object was also allowed. On the other hand
{spuda}, "x1 responds to x2", explicitly allows both object or event in x2,
i.e. it's polysemous (is that the right technical word?). In this case, there
is no sumti raising because when using an object we are using the second
meaning of the gismu.

> I rather like being able to think of "la lojbab" as an event - it had a
> beginning, and presumably will have an end ...

Of course, events can be given names. But just because the "object" called
lojbab, and the "event" called lojbab have the same name, it doesn't mean
they are the same thing. The only problem in this case, is that if someone
says {spuda la lojbab}, we don't know whether they mean a response to lojbab
the person or to the event that you call lojbab. (In the first case, it would
be a response to some action of lojbab's, in the second, a response to the
whole event that had a beginning and presumably will have an end.)


> Likewise other concrete sumti can be treated as abstracts, and if we were
> SURE that sumti raising is NOT taking place, the semantic usefulness of
> being able to cross between categories could be Sapir-Whorfianly important.

For some gismu (like spuda), sumti raising is allowed, for others, (like
djica), it isn't. This means that we can never be sure if sumti raising is
taking place unless we have the gismu list handy. The only way that the
place can force a category on the sumti that fills it is if the place has a
strict category assigned, otherwise, the sumti will rule, modifying the
meaning of the selbri to suit the type of sumti.

Jorge