* Saturday, 2011-09-10 at 10:43 -0300 - Jorge Llambías <jjllambias@gmail.com>:
> On Fri, Sep 9, 2011 at 9:08 PM, Martin Bays <mbays@sdf.org> wrote:
> >
> > The context here was meant to be that we are assuming that everyone is
> > loved by some chihuaua (in the sense
> > FA x:person. EX y:chihuaua. loves(y,x)),
> > and everyone is loved by some German shephard.
> >
> > It wouldn't follow in english that some dog loves everyone, nor that
> > some dogs love everyone, nor that some kinds of dogs love everyone, nor
> > anything else I can think of along the same lines.
> >
> > But it would follow that
> > {ro prenu cu se prami zo'e noi tci,uaua} (1),
> > and, unless I'm misunderstanding, your interpretation of that zo'e
> > makes the following true (in domains of discourse where (1) holds):
> > {da poi tci,uaua cu prami ro prenu} (2).
> >
> > I don't think this has an analogue in english, nor in any other rarbau
> > I know.
>
> Let's see. We have four sentences:
>
> 1L: ro prenu cu se prami su'o tciuaua
> 1E: Everyone is loved by some chihuahua.
>
> 2L: ro prenu cu se prami zo'e noi tciuaua
> 2E: Everyone is loved by chihuauas.
>
> 3L: zo'e noi tciuaua cu prami ro prenu
> 3E: Chihuahuas love everyone.
>
> 4L: su'o tciuaua cu prami ro prenu
> 4E: Some chihuahua loves everyone.
>
> We also have two domains of discourse:
>
> D1 = {lo prenu ku xi pa, lo prenu ku xi re, lo prenu ku xi ci, .., lo
> tciuaua ku xi pa, lo tciuaua ku xi re, ...}
> = {person_1, person_2, person_3, ...., chihuahua_1, chihuahua_2, ...}
>
> D2 = {lo prenu ku xi pa, lo prenu ku xi re, lo prenu ku xi ci, .., lo tciuaua}
> = {person_1, person_2, person_3, ...., chihuahuas}
>
> D1 and D2 are not the same domain. Sentences 1 and 4 "put us" in
> domain D1, while sentences 2 and 3 "put us" is domain D2. By that I
> mean that those are the natural domains in which to interpret those
> sentences without any more context. Do we agree so far?
Not entirely. I think 1E-4E could just as well be interpreted in the
union D12 of D1 and D2 - because a sentence involving "some chihuahua"
can't have the generic "chihuahuas" as an witness, and although
(as in Carlson) a predication involving "chihuahuas" is ambiguous
between being about the generic and about its
manifestations/stages/whatever, that doesn't mean the domain of
discourse has to be different for different interpretations.
You seem to be saying that D12 is an intrinsically unnatural domain for
lojban. That seems to be a difference from english.
> I think we are both in full agreement that sentences 4 entail
> sentences 1, but sentences 1 do not entail sentences 4.
>
> I claim that sentences 2 and 3 entail one another. At this point I'm
> not sure whether you agree with that or not.
I agree that 2L and 3L should be logically equivalent.
Probably 2E and 3E are too.
> I make no claims about any entailments between 2-3 and 1 or 4, since
> they have different natural domains of discourse. You are saying, if I
> unerstand you correctly, that 1 entails 2, and that 3 entails 4 (or
> that I am claiming that they do), and that therefore 1 entails 4 and
> we (or I) have a contradiction.
I think we agree:
D1 |= 1L
implies D2 |= 2L
implies D2 |= 3L
implies D2 |= 4L
(where "D |= S" means "sentence S holds in domain of discourse D")
So there's no actual contradiction; it's just that if I claim 4L and you
want to deduce something about D1 from it, you have two ways to proceed
- you can (i) assume I'm claiming D1 |= 4L, or you can (ii) assume I'm
claiming D2 |= 4L. By the above deduction, you can't deduce from (ii)
more than D1 |= 1L. (In general, you won't even be able to deduce that
much.)
This is the (weak) sense in which there's an ambiguity of logical form.
Now you say D2 is not a "natural domain" for 4L. That does sidestep
this issue. It also makes determination of what is and what isn't
a natural domain for a sentence a crucial part of the semantics of the
language... do you think there's a coherent general theory of that (say
for the small 'extensional' fragment of the language we've been
mostly considering in these threads, which excludes tenses and NU and so
on)?
(In fact, I *would* like to claim that 1L logically implies 2L, because
I would still like to analyse {zo'e} (but not {lo}) as in the subject
line of this thread. But that's beside the point.)
> I think that the only way you can move from 1 to 2 is by changing your
> domain of discourse, so there is no logical entailment there.
>
> > I'm not convinced that it's a good idea to have as a frequent element of
> > the domain of discourse an individual "chihuahuas" which gerkus and can
> > be a value of {da}... but this is kind of a separate issue. Here I'm
> > just saying that there shouldn't be one which gerkus and also pramis
> > everything which any chihuahua pramis.
>
> In the same domain of discourse? I agree. As I've been saying, mixing
> generics and their instances in the same domain of discourse is not
> impossible, but it requires extra work. If chihuahuas are dogs, and
> Spot is a dog, and Pichichus is a dog, that doesn't mean we have three
> dogs there. We can't just add Spot, Pichichus and chihuahuas together
> to get three.
Yes, but you are allowing, as in D2 above, an individual which gerkus
and also pramis every prenu which any chihuahua pramis. That isn't much
better, from my point of view.
There's also the obvious issue of handling {ro gerku cu se prami su'o
chihuahua}, for which the analogous D2 would indeed contain Spot,
Pichichus and chihuahuas.
> >> >> "I love buying things, but then I never know where to put them."
> [...]
> >> If the objection is that the statement is too coarse grained for your
> >> taste, that you prefer statements that are more precisely nuanced,
> >> that's fine, but that doesn't mean that the coarse grained statements
> >> violate any logical rule.
> >
> > No, my objection is that the english pronoun "them" has a more
> > complicated anaphoric meaning which is being lost by working only with
> > the generic "things".
>
> OK, but consider these two points:
>
> (1) "I love bumping into John, but then I never know what to talk
> about with him."
>
> Would your analysis of "him" also require that it picks up the same
> stages of John that I bump into? If so, this has nothing to do with
> "them" referring to a generic, since the same type of issue would
> arise with John-him. If not, why not?
Maybe... I'm really not sure. If these sorts of questions had easy
answers, I wouldn't feel the need to study Lojban ;)
> (2) "I love buying things, except when they are too expensive."
>
> Would your analysis of "them" in the original sentence be the same as
> the analysis of "they" in this sentence? If you use a different
> analysis, it seems to me you are making pronouns much too ambiguous.
I guess I'd use the same kind of analysis - i.e. one in which the second
'they' does end up referring to individual things. Here, it would end up
as "I love it when I buy something which is not too expensive".
> The way I would capture what you claim is being lost by the generic is
> something along these lines (which I'm not claiming is anywhere close
> to a formal theory as I'm presenting it):
>
> It is part of the meaning of "buying" that when you buy things you end
> up having them, and it is part of the meaning of having things that
> you can put them places, so there is a natural semantic connection to
> be made when talking of buying things and of putting things in places
> that we are talking of the same things, the same manifestations of
> things. But I don't think this connection is hidden in the pronoun.
> It's the same as with the John case, it is part of the meaning of
> someone bumping into someone else that they have to be in the same
> place at the same time, and when two people are in the same place at
> the same time they can talk. So there is a natural connection to be
> made when talking about bumping into someone, and of talking with
> someone, that the same stage of that someone is involved in both
> cases. But I don't think that this connection is hidden in the pronoun
> either, it follows from the type of situations involved. If the next
> sentence used the same pronoun with a predicate that didn't have to
> involve the same stages, the referrent of the pronoun would not
> change.
>
> For (2) the connection to be made goes through a different route:
> buying things requires things to have prices, and things having prices
> mean things can be expensive, so again there is a natural connection
> to be made there that the manifestations of things involved in the
> buying of things excepted from being loved are those manifestations
> that are expensive. But I don't see how that can be adjudicated to the
> pronoun.
So it's just a matter of context-hinting forcing the second predication
about 'things' to refer to specific stages/manifestations rather than
things-in-general? Maybe you're right. Since the result is the same
whether we derive it via complicated anaphora or via context-hinting,
it's hard to distinguish.
Do you know if there's anything in the formal semantics literature on
this?
Martin
Attachment:
pgpzeqyImpCB1.pgp
Description: PGP signature