[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
xorxes:
> > > But the sumti of {roi} is for the interval over which the
> > > number of instances repeat, not for the duration of the event.
> >
> > Okay. Should that sumti be quantified at all, then? It seems
> > to me that, logically, it should be something like
> > {lo'e du'u ke'a djedi li ze}.
>
> It won't be quantified from the point of view of the tag.
>
> mi klama le zarci paroi le pavdei
> ije mi klama le zarci paroi le reldei
>
> -> mi klama le zarci ge paroi le pavdei gi paroi le reldei
> -> mi klama le zarci paroi le pavdei e le reldei
> -> mi klama le zarci paroi ro le re djedi
I was thinking not of cases like "n times on Monday" but
rather of "n times per x" cases: it's x that I'm suggesting
shouldn't be quantified.
> > > It doesn't mean that if we allow
> > > overlapping minutes, I agree. Is that the objection?
> >
> > That wasn't my objection, but in fact it is quite a good objection!
>
> Actually, not so good, because (assuming neverending uniform
> rotation) every minute will have one rotation, even though the
> rotations will start at different points for different minutes.
Only works for regular distribution, though.
> > But what I meant was that when I say "I clean my teeth twice
> > a day [or: twice every day]", I don't mean that during every
> > day I clean my teeth twice. For example, before I was born I
> > didn't clean my teeth. Now of course Grice means that this
> > usually wouldn't be a communication problem, but on the whole
> > I would prefer that what is said is what is meant.
>
> We can say {paroi ro le mentu} to restrict to the relevant minutes.
Not a proper solution, though, because it leaves too much to
be glorked.
> We could also say {paroi lo'e mentu} and see all repetitions as one.
This is better, but it doesn't guarantee a "once per day" reading.
> > > > If so, then you could formulate {roi} as {fi'o ra'inrapli be li
> pa
> > > > fo'a}, and prove your point using that reformulation.
> > >
> > > Yes, that's good, though I think I want to keep the quantifier
> > > vis-a-vis the rest of the terms in the bridi.
> >
> > Just to help me get my head round your proposal/argument, could
> > you essay such a reformulation using fi'o, for my benefit?
>
> I don't think I can, because I want the number to keep its
> quantifier nature for whatever follows. I want:
>
> <Q2> roi <Q1> broda <Q2> brode
>
> to export the quantifiers to the prenex in order Q1, Q2, Q3,
> so that <Q2> roi acts as a normal selbri towards <Q1> broda,
> but at the same time has scope over the following terms.
Okay. This is okay for things like "On each=Q1 day of my
holiday, I went to the museum twice=Q2".
But I'm still looking for a better way to say "During part of my
holiday, I went to the museum twice a day". Or "For a brief
period, my heart was beating 100 times a minute".
--And.