WikiDiscuss

---

wrote:

>
> pc:
> > > A predicate, as I'm using it, is the thing
> that
> > > takes
> > > arguments and, with the arguments, forms a
> > > sentence
> > > (or formula). {naku} negates a sentence, or
> > > turns
> > > a sentence into another sentence. {na'e}
> > > changes one predicate into another
> predicate.
> > >
> > I have no problem with the distinction you
> are
> > using, but you seem to think it is an
> absolute
> > one. In fact, in Lojban at least, every
> > predicate is a sentence and far and away most
> > sentences are predicates.
>
> I'm afraid we are speaking different languages.
>
> The distinction I make is between "bridi" and
> "brivla",
> where a bridi is the kind of thing that can
> have truth
> values, which a brivla cannot have.
>
OK. But in a given context, how do you tell
which it is? {mlatu}, for example, standing
alone might be either and it is not obvious to me
that {na mlatu} clarifies the issue.

> So you need something
> > more than that mark to explian the difference
> > between the uses of {na} and {na'e}. I
> suppose
> > you want a scope one, but then you have to
> allow
> > that {naku} at least sometimes has only a
> > predicate in its scope and so must be
> modifying
> > it.
>
> {naku} always has a full bridi in its scope:
> it takes a bridi and returns a new bridi.
>
Now I do think we are in different languages, the
interesting question being what "scope" means.
You have amintained often enough what I would put
(and thought you had put as well) that the scope
of {naku} is everything to the right of that
occurrence in a sentence. I don't really believe
you have changed this view (it would seem to me
to redouble the confusion that using {naku} was
meant to alleviate), but I don't know how you
would put it now.
Of course, if we start with a sentence (as we
usually do), then {na'e} converts one sentence
into another as much as {na(ku)} does, though
perhpas the locus of its action is often narrower
than that of {na}. Notice that {na} is not
restricted to things that are true or false,
since {lo na broda} is an OK construction — an
another case of predicate scope.

> {na'e} never acts on a bridi. It takes a brivla
> and returns a new brivla.
>
And thus a sentence and a sentence.
>
> > > {na} is the same as {naku} in this regard.
> It
> > > may
> > > differ in the order in which it operates
> with
> > > respect
> > > to other bridi operators (quantifiers and
> > > connectives).
> >
> > Is that the whole difference?
>
> Yes.
>
> > > The only controversy about {na} has been
> about
> > > where
> > > an equivalent {naku} would occur. I don't
> think
> > > up
> > > to now anyone had suggested that {na} and
> > > {naku}
> > > differed in anything but scope.
> >
> > But scope is just what the problem is here,
> when
> > is the scope the sentence and when (in cases
> > where the distinction is useful) a predicate?
>
>
> The scope of na/naku is always the bridi, never
> the brivla.
>
I guess I need to know what "scope" means to you
and how you would describe the difference between
{mi naku broda} and {mi na'e broda} — without
prejudging the issue of "scope" in your sense.

> If
> > you want that {naku} is just sentential
> negation
> > wherever it occurs, then the same problem
> arises
> > in figuring out what is the sentence it
> negates.
>
> That's what the reduced form is for. In that
> form it
> is very clear which sentence it negates.

I am now not at all sure you can do what you want
to do with this: how would you differentiate
between {mi na klama ta} and {mi naku klama ta}
if both negate the whole sentence? Or do all
terms have to be fronted? The difference seems to
be betweeen {mi zo'u naku 1 klama ta} (with "1"
to be lexed any of severaal ways)and {naku mi
klama ta} but then, as expected, the sentence
which {naku} negates is different from the one
{na} does (though at least cases like this seem
to fit easily into algorithms). Indeed, {naku}
negates a preicate , {klama ta} into which an
external term has been inserted, only
incidentally different from {na'e klama}
logically.

>
> > It is usually not just what is left when the
> > {naku} is dropped, as that will typically get
> > quantifiers (and tenses and existence
> conditons
> > and so on)wrong.
>
> Right. That's why I'm doing the reduced form
> exercise.
>
> I suppose that the distinction
> > you want is about length of scope not actual
> > scope: {na(ku)}takes as long a scope as it
> can
> > get within a sentence, {na'e} takes just the
> next
> > complete structure, typically a brivla or a
> > marked tanru — and with {bu} apparently a
> sumti.
>
> {naku} operates on a sentence. The reduced form
> shows
> which sentence.
>
> {na'e} operates on a brivla (or tanru). It
> never operates on
> a sentence.

This seems to be a logically irrelevant
distinction, but at the best of times it does not
show that {na(ku)} does not modify preeicates,
which is the crucial point at the moment.

Which brivla it operates on is
> already obvious
> from the parse, so there is no need for
> transformations here.
>
> {na'ebo} operates on a term, also obvious which
> one.
>
> > Is {na bu} a possibility to contradictorily
> > negate a term?
>
> Nope.

The parser accepts it (with an inserted BOI).

> > Probably not but {ko'a klama naku
> > ko'e} is and seems to function like {na'e}
> but
> > with some mysterious additional meaning: a
> goes
> > to someplace other that b, such that not
> going to
> > this place would be going to b, which makes
> sense
> > in some restricted cases anyhow.
>
> {ko'a klama naku ko'e} is simply {naku zo'u
> ko'a klama ko'e}.
> There are no other bridi operators to interact
> with {naku}
> here.
>
Ah, here with have a fundamental disagreement
(probably noted already above); I would say it
comes from {ko'a zo'u naku 1 klama ko'e} at
least, and, indeed, needs to get the {klama} left
of the {naku} as well: it is not just that "a
goes to b" is false (which might be the case if a
didn't exist, for example, or didn't go anywhere)
but that it is false because a goes somewhere
else, that where absolutely unspecified (unlike
{na'e} which implies a range in mind or so).