WikiDiscuss

pc:
> 1> I doubt that the most common group is a singleton; the most common is
> surely no group at all but just and individual. But then I suppose that is
> what you meant;

Yes, I meant {lo pa broda}, with the proposed {lo}.

> quantifying into a singleton would make sense, though on with
> {pa} — and fractionals.

That would be:

PA pagbu be lo pa broda

That's a possible use for fractional quantifiers, though not
my preferred one.

> I again would say that the most common thing would
> be to count individuals, which I assume is what you mean. But it does not
> seem to me that that helps at all with the question of internal quantifiers
> as group sizes, since the analogy is not very good.

I understand Pierre does not object to using internal quantifiers
as group sizes. He objects to the use of the external quantifiers to
quantify over instances rather than over members.

I guess {PA broda} can equally well be understood as quantifying
over members of the group of all broda, or over instances of
a single broda.

(Re:mupli & cmima)
> 2> Yes, but these are rather complex. The first has a short form and the
> second does not, but I suspect the second is more common — or at least as
> common — as the first.

I presented some examples with instances of groups. Perhaps if
someone presented some examples of members of (generic) groups
we could get a better idea of what we are comparing.

> 3> {lo2} does not behave like a constant, since it is not one (it is not
> tranparent to any operation). {lo3} does behave like a constant (since it is
> one) but is abstract and relatively impervious (as described so far) to
> factual properties (though that could be changed fairly easily at this
> point).

If lo3 could be made more pervious to factual properties, then
we may be converging.

mu'o mi'e xorxes

__
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/