Re: [lojban] Re: mi kakne lo bajra

[Ian Johnson <blindbravado@gmail.com>]

There's a hierarchy here, and you have to sort of arbitrarily decide how it's structured. Physical concreteness (that is, being an object basically) is a good place to start. Splitting events off from the others because of their temporal concreteness is probably a good second step, which gives you 3 nodes at the 3rd level of the type tree, 2 of which are leaves. The separation continues after this of course, but I wasn't talking about that for purposes of this discussion.

And the idea that one argument of a predicate should be able to define the types of the others is interesting, but then you start talking about trying to implement a full-fledged type inference system in a spoken language if you want everything to really make sense. I don't know if that's reasonable (though it would be awesome if we pulled it off.)

mu'o mi'e latros.

2010/10/31 Jorge Llambías <jjllambias@gmail.com>

On Sun, Oct 31, 2010 at 6:55 PM, Ian Johnson <blindbravado@gmail.com wrote:
> The "event is concrete" concept is interesting, but I disagree with the idea
> of lumping events in with physical objects in predicate structure. If
> anything they should be a third type unto themselves.

Why only "third"? Do you lump numbers, sets, properties and
propositions into one type? Which predicates accept numbers, sets,
properties and propositions but not physical objects?

> As for the issue with {zmadu}, the problem that it is possible to compare
> two objects as long as they are of the same type, but that comparison is
> defined for different types as well.

Right, the x3, the property, determines the kind of comparison. It is
essentially the same case as "ckaji", where the x2, the property,
determines what can go in x1. And indeed any predicate with a property
place. The type of the x1 of "mutce" will be determined by the type
accepted by the property in x2.

mu'o mi'e xorxes

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/lojban?hl=en.