WikiDiscuss

Fredegisus (I looked it up finally) is getting a
workout here, along with several other guys.

1. {ta broda no da} just denies in a simpler way
(because the negation is already confined) {ta
broda su'o da}.

2. {li no} does refer to something, the number
zero, so that {ta broda li no} cannot be true at
the same time as {ta broda no da} is.

3. {li no} does not refer to the empty set
(except in very strange situations — mirroring
arithmetic in set theory, say), but the empty set
is something, too, so {ta broda le nomei} can't
be true at the same time as {ta broda no da}

4. Put another way, neither {li no} nor {le
nomei} refers to nothing.

5. Capless bottle have usually been discussed
not in terms of {botpi fo no da}, things which
are incomplete bottles because they lack lids,
but rather in terms of {botpi fo zi'o} a totally
different predicate which, however, does seem to
be about things like bottle except that they do
not definitionally have lids — vases maybe. It
does not mean that its referents are bottle
without lids, since they are not bottles at all
in Lojban, not botbi but botpi fo zi'o. And they
are certain not bottles for which nothing is a
lid ("and a damned poor lid it would be too" as
Fred might say).

6. {zi'o} doesn't refer to nothing either, since
it doesn't refer at all but just plugs a place in
a predicate, taking it out of play. (the answer
to "To what does {zi'o} refer?" is {na'i}, the
presupposition of the question — that {zi'o}
refers — is false).

7. And of course no other expression refers to
nothing either, since there is no nothing to
refer to — and worse, no expression can even
purport to refer to nothing.

(Sartre's book would be Lojbanned, roughly but
literally, as {le nu zaste ku e le nu na zaste}.)

One of the upshots of all this is that sentences
with "nothing" or "no" or other denial words in
them need special treatment insofar as certain
kinds of formulae will not work for them: they
can't be internal quantifiers for example,
general constructions of numeric quantifiers
can't cover them within a single formula with the
other numbers (this is also true of "at most"
formulae — although, in fact, the numeric forms
of all these can be defined recursively).

More to the present point, {zo'e} of course can't
refer to nothing, but also {no da} is not a
permissible substitute for {zo'e}. This is
because replacing {zo'e} with {no da} would not
merely (indeed, not at all) specify an object, IT
WOULD ALSO NEGATE THE SENTENCE, that is not only
fail to specify but change the sentence about as
much as possible. This is not an appropriate
thing for specifiction to do.