Message-Id: <199512241157.NAA12486@xiron.pc.helsinki.fi>
Date:         Sun, 24 Dec 1995 11:54:29 +0000
Reply-To: ucleaar <ucleaar@UCL.AC.UK>
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: ucleaar <ucleaar@UCL.AC.UK>
Subject:      Re: du'u continued
To: Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 2234
Lines: 46

ar> >> Here goes again: {duu} is used in a range of 2-place predicates,
ar> >> each of which relates a proposition to the sentences that express
ar> >> the proposition. This is tantamount to a 1-place predicate, just as
ar> >> "cousin of John" is effectively a 1-place predicate - the x2 of
ar> >> "cousin of John" is of course John, so it seems a bit pointless to
ar> >> have "le se cousin-of-John" to refer to John.
jc> >Having read what we said very carefully, I think that you are right
jc> >and I am wrong, but I am by no means certain.
lb> If du'u is a one-place predicate then "da du'u broda" has some
lb> properties - is it a relation, a sentence or what?

The x1 of duu is always singular in extension & is determined solely
by what is in the bridi bit (duu ... kei) and not by what is in the x2.

"da duu broda" is like "da se cousin-of-John" = "da belongs in the
one member class of things that are John"; it means "da belongs in
the one member class of things that are the bridi "broda".

So if {da duu broda ije de duu broda} then {da du de}.

Another analogy: duu is analogous to a 2 place-predicate "x1 is the
cardinality of x2 such that x2's cardinality is 4" ({da xuu 4 kei de}),
just as {da duu broda kei de} is "da is the proposition expressed by
de such that de expresses the proposition {broda}".

> I'm not sure what a "class of sentences" is, alas, but I am sure that
> "da" is NOT a class, though all possible values of da might be members
> of a class.

That seems right.

> Now perhaps the argument is that we have no non-self-referential way to
> fill in the x1 of "du'u bridi" i.e. we are forced to say "le du'u bridi
> cu du'u bridi", and hence the sumti place is meaningless.  Actually
> rather, we are converting a grammar structure that is a bridi into a
> particular kind of a sumti, which is what all the "LE+NU" constructs do.

NU converts a bridi into a selbri. LE converts a selbri/sumti tail into
a sumti. My objection to {duu} is that it is pointless for it to be
a selbri. Whenever I used {duu} I was always irked that the choice of
gadri was utterly irrelevant; I tried using {li}, but ended up deciding
that in {kuau} (which is zabna glico in being like _that_) lay the best
remedy.

---
And