Re: About 'zasti'

[Logical Language Group <lojbab>]

la rob. cusku di'e

> I feel quite uncomfortable with the presence of 'zasti' in
> the gismu list.
> IMHO, 'Exist(x)' is a notable example of a
> "pseudopredicate" producing a long series of inconsistencies
> and/or misunderstandings. As such, it should be banned in a "logical"
> language (by the way, this should have major consequences on the structure
> of Lojban-biased thought and could be of some relevance in testing the
> Sapir-Whorf hypothesis). What I mean is that, for example, the meaning of
> the verb 'to exist' in the two sentences
> 
> This desk exists
> 
> and
> 
> God exists
> 
> is only superficially the same. If 'Exist(x)' were a legitimate predicate,
> the well-known theological argument for proving the existence of God
> by the very definition of 'God' as 'the entity everything can be
> predicated about' would be trivially true!...

I think you have misunderstood or misremembered the so-called "ontological
argument".  It states not that God is the subject of every predicate
("God is ultimate evil"?  "God is blue"?) but that God has every desirable
property.  Since existence is better than non-existence, God must exist.

There are several reasons for thinking this argument illegitimate, even if
you think existence is a predicate, which I won't go into here,
partly because I'm not sure I can remember them correctly.

> As to the relation between 'zasti' and DA cmavo, consider for example the
> sentence:
> 
> Something does not exist                (1)
> 
> This sentence sounds pretty sensible (you can, for example, substitute
> 'Unicorns' for 'something' resulting in an admittedly true assertion).
> Furthermore, (1) is analogous to more innocent sentences like
> 
> Something is not blue                   (2)
> 
> which everybody would agree on.
> This parallelism is more evident when considering the formal logic
> translations of (1) and (2):
> 
> Ex[~(Exist(x)] = ~Ax[Exist(x)]  <---->  Ex[~Blue(x)] = ~Ax[Blue(x)]
> 
> The Lojban equivalent of (1) is (if I'm not mistaken :)
> 
> (su'o)da naku zo'u da zasti ( = (su'o)da na zasti) =

No.  "na" before the selbri means "naku" with widest scope, so
"da naku zo'u da zasti" collapses to "da naku zasti", as "naku" keeps its
scope.  This means "naku roda zasti" or just "roda na zasti",
"it is not the case that each thing that exists, exists".

> naku roda zo'u da zasti                                         (1.1)
> 
> which should be compared with the semantically "plain" Lojban bridi
> corresponding to (2):
> 
> (su'o)da naku zo'u da blanu ( = (su'o)da na blanu) =
> naku roda zo'u da blanu                                         (2.1)
> 
> >From this example, I would be induced to conclude that the best (unique?)
> interpretation for 'da' is the restricted one, even though no relative
> phrase follows:
> 'there is at least one x *in the universe of discourse*'.
> In fact, if 'da' conveyed the idea of real (physical) existence, (1.1) would
> result in a contradiction.

It is a contradiction, or at any rate always false, at least for proper x2
and x3 values of "zasti".  You can give them non-default values (for
knower and metaphysics respectively) that make "da naku zasti" sometimes true.

> >From the previous observation and the claimed equivalence between 'lo broda'
> and 'da poi broda', I would interpret 'lo broda' too in a restricted
> sense as:
> 'at least one, *in the universe of discourse, of those that are broda'.
> Since what the universe of discourse is usually depends on speaker's
> subjective attitudes or opinions, the restricted interpretation
> of 'lo broda' seems to me very close (apart from the difference in the default
> quantifiers and specificity value) to the possible interpretation of 'le
> broda' as 'all of those that are broda in (my) universe of
> discourse'.

But that is not "le broda".  "le broda" is "all of those to which I intend
to refer at this moment."  There is a difference between the universe of
discourse, which is pretty broad, and the object of my current intention
to refer.

-- 
John Cowan		sharing account <lojbab@access.digex.net> for now
		e'osai ko sarji la lojban.