[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
Remember that for singular referents, "distributive" and "personal"
coincide. So for the singular "my going to the store" we get the
right result. The problem, of course, is how to recognize that this
{lo nu...} is being taken externally, as a node on the lattice, say,
rather than internally, as a bunch of trips, as it also is. In this
case, the predicate {rapli} gives it away, but I am sure there are
cases where the choice is not obvious from however much context you
want. Notice the other possibilities for making this point present no
problems of this sort: {mi muroi klama le zarci ca le cabjeftu} {mu lo
mi klama le zarci cu fasnu ca le cabjeftu}. Maybe {tu'a}'s role is to
mark nodal readings?
--- In jboske@yahoogroups.com, "Jorge Llambías" <jjllambias@...> wrote:
>
> On 10/10/06, John E. Clifford <clifford-j@...> wrote:
> > To say that my going to the store occurred five times this
> > week is just another way of saying five my goings to the store
> > occurred this week.
>
> Agreed.
>
> > It is useful to have two ways of saying this and,
> > in fact, the domain of discourse is different for the two (one of the
> > peculiarities discourse analysis revealed is that logically equivalent
> > expressions may generate different domains of discourse -- it is not
> > clear what the significance of this is).
>
> OK.
>
> > What is less clearly OK is
> > using the same predicates at different levels -- or tryng to use them
> > in the same way.
>
> To me this seems inevitable. For suppose we identify two levels and
> assign a different predicate to each level. The new predicates will
> each be subject to the same treatment as the original predicate was,
> and the process of differentiation would never end. (This is not to say
> that it is not useful to introduce diferentiating predicates whenever
> they are needed to make a distinction. What I think is impossible
> is to make the distinction obligatory for all predicates in all
contexts.)
Of course, I am really only concerned with the level of individuals of
the named sort and all the other layers above and below. But then the
peculiarity of individual predication comes into play. What effect
cases of actual collective predication has on all this is just unclear
at the moment -- especially distinguishing taking a node as an
individal predicated personally and a bunch predicated collectively
(but there has always been a suspicion that that collapse was not
going to work in practice.
> > I admit I would prefer that there be a separate term
> > for the nodal view -- if it is going to be used a lot. I don't really
> > see it being used a lot, so I don't much care then.
>
> Me neither. It's just a metalinguistic artifact anyway, not somehing
> that needs to be reflected in the object language. When the object
> language is used for some technical purpose then whatever
> technical terminology is needed can be defined, but it's not
> something that will be talked about in ordinary discourse.
I'm not sure what you are calling a metalanguage artifact; these all
seem to be points to be made in the object language, at least
occasionally (and not just wj=hen we are doing meta discussions).
> > I think we will
> > just use {lo broda} for any node on the broda lattice (when we don't
> > specify the node at all). And attribute all manner of things --
> > especially being a broda -- to it distributively.
>
> Consider:
>
> lo nu mi klama le zarci kei noi cacra li ji'i pa cu rapli li mu ca
le cabjeftu
> "My going to the store, which takes about an hour, was repeated 5 times
> this week."
>
> {rapli li mu ca le cabjeftu} is clearly not attributed
distributively, since
> it is not the case that each of my goings is repeated 5 times. What
> about {cacra li ji'i pa}? I would argue that each of my goings had a
> much more definite duration, and that it is only my going to the store,
> generically, which takes about an hour, and that this predicate is
> attributed personally to it. I don't know, however, if it makes any
> difference what kind of metalinguistic analysis we make. The only
> relevant question is whether we both would take {lo nu mi klama
> le zarci kei noi cacra li ji'i pa cu rapli li mu ca le cabjeftu} as
> describing the same situation or not.
The same situation as what? Oh, you mean that you and I agree about
what it means. I assume we do -- namely, what the other sentences
offered above say (less the non-restrictive relative clause).
I think we have done about as much as is useful with this. It is
still logically a crapola analysis of the situation of wants and we
are now moving again to a place where the two analyses cannot be
covered by a single formula (not that they could in other cases
already, but we haven't looked at other cases). So, while I have some
understanding of what you are driving at, I still think it is as
serious a mistake as I did when what you were saying was
unintelligible gobbledygook -- though now I know why.