Mike S., On 24/09/2012 19:49:
>
>
> On Mon, Sep 24, 2012 at 2:23 PM, And Rosta <and.rosta@hidden.email <mailto:and.rosta@hidden.email>> wrote:
>
> For the quantifier predicates, let's provisionally suppose there are three series (corresponding to the three types I mentioned), indicated by one of the two Cs, with the other C indicating the fraction, which might include:
>
> none [on the arguable grounds that this is more basic than negation]
> some
> some-but-not-all
> all
> most
> large proportion
> small proportion
> ?almost all
> ?almost none
>
> Do the latter two mean "many if not all" and "few or none"
> respectively?
No, but those are very useful too.
> I guess things like "want", "believe", "intend", "start" might warrant CCs too, i.e. on grounds of individual importantness.
>
> The first one in this class to get a CC would be "know", if going by
> frequency of usage in natlangs (IIRC). I will try to dig up my copy
> of the BNC frequency list later.
No, BNC frequency list is a list of lexemes not senses, and is of negligible evidentiary value for our purposes. There is no data I know of for frequency of senses, dearly tho one wishes there were.
> > I probably will whip together a shake-n-bake Lojbanically a-posteriori vocabulary
>
> What's a *Lojbanically* a-posteriori vocab? I'd like a handcrafted one, where patterning among forms reflects patterning among meanings.
>
> I am just talking about importing a bunch of predicates from Lojban, making sure they don't clash with each other, and at the same time start to study them and group them by their patterns.
Good for you, for putting the work in and making a start on that.
--And.