Re: [lojban] No title, since the subject will have changed by the time it gets there

[John E Clifford <kali9putra@yahoo.com>]

The gadri difference is merely because these particular bunches seem to be 
especially common (in this discussion, anyhow) and so might use the separate 
mark -- another way to intimate generality, perhaps.  I actually do want to mix 
extensions across possible worlds, though perhaps not willy-nilly.  Extensions 
in other worlds are just sets with members and can then be dumped, for some 
purposes at least, into a bunch with other objects from other sets.  I don't see 

the problem (unless, of course, you hold that there is only one set of ultimate 
referents, which get recycled in all worlds, perhaps with radically different 
properties).  But I am still unclear why intensions keep coming up whenever {lo} 

phrases are mentioned (but not with the very similar {le}, for example).  I know 

that properties, in one sense (pun?), are just functions from worlds to sets of 
thing in the relevant worlds, but I don't see what properties have to do with 
the matter here.  We want things in the extension of a given term, or, perhaps 
its extensions in many worlds, but these are just matters in each world, not 
some function on worlds that I can see (of course, given all the worlds, these 
would "contain" the function involved, but that is not obviously relevant 
here).  The long metaphor that follows is somewhat opaque: why do extensions 
derive from intensions, if they do? what does not interacting with things 
outside their  "state of affairs" (which is? was der Fall ist?) mean (they 
clearly can be referred to together in the same sentence, as witness the 
implicit appeal to cases where thing have hearts but not livers in your next 
remark)?   I don't have just an extension in a given state of affairs (I think, 
since I am not sure what that involves exactly), I have the extension of "broda" 
or "cinfo" or whatever, a matter totally within the given world, and what the 
intension is built up from.  I like that you allow that a domain of discourse 
may contain many possible worlds, since that is where I am too, but I don't see 
that as involving intensions  particularly -- nor do I see xorlo as forcing the 
issue.  Your next paragraph seems backward to me.  I was not suggesting we 
needed a different gadri for intensions from extensions, since they are already 
distinct in the predicates involved (well, they should be, but Lojban is a 
little slippery here, not being very specific about just what {ka} means).  You 
are given the extensions in various worlds, the bears or lions or whatever, and 
from that you can construct the transworld functions, if you want -- or just use 
the extensions you need, without the abstract step.  Since the difference 
between intentions and extensions is between different kinds of things, the 
place to do it is in the predicates, where things get sorted out.  And that is 
where Lojban does it (as much as it does do it).  There is very little to do 
here with tense and mode and aspect and mood and whatever else you want to throw 
in, except that most of these things (maybe all) do in fact involve alternate 
possible worlds (or at least past or future ones to this one).  But the 
involvement is in totally different ways.
I am trying to keep the semantics of Lojban as simple as possible.  One aspect 
of this is avoiding intensions as long as I can, and I don't see xorlo as 
requiring me to introduce them.  Even for generalizations beyond the current 
crop of whatevers.  What have I missed in the past 45 years?




----- Original Message ----
From: maikxlx <maikxlx@gmail.com>
To: lojban@googlegroups.com
Sent: Thu, November 24, 2011 9:19:17 AM
Subject: Re: [lojban] No title, since the subject will have changed by the time 
it gets there

On Thu, Nov 24, 2011 at 9:32 AM, John E Clifford <kali9putra@yahoo.com> wrote:
> Well, sorta.  A maximal bunch is a bunch (L-set, plurality) in a single domain
> of discourse that contains all the relevant objects (the current ones almost
> always, the past ones often. the possible ones occasionally -- probably only 
>the
> possible ones from proximal worlds).
>
It's not clear to me how that would work or how you'd want it to work,
and I doubt the issue could be sorted by marking distinctions using
the gadri.*  You do want to refer to a bunch, which I think is roughly
equivalent to an extension, both to obtain a generic reading at a
given world-time index when desired and to allow quantification when
desired.  However, you do not want to mix extensions across possible
worlds willy-nilly.  Intensions, which do span possible worlds, do so
in a well-defined way, and they have totally different properties than
do extensions.  Intensions "contain" the extension (possibly empty) of
some thing-a-ma-jig at every possible state of affairs.  In other
words, intensions are functions that say, "give me a possible world
and time, and I'll give you exactly the extension that you're looking
for."  This is the closest thing to a mathematical formalization of
"meaning" that I have yet encountered.  Extensions are myopic entities
that unable to interact with anything outside their own state of
affairs, and they have no way of determining the intensions they are
related to or are derived from.  We notice this when we notice that
there are many possible states of affairs in which the extensions of
{se risna be da} and {se livga be da} are identical.  If all you have
is a bunch (i.e. L-set or C-set of entit(y/ies) in a given state of
affairs), how would you know which intension they came from? However,
by going to the intension, by expanding the domain to include possible
worlds in which {se risna be da} and {se livga be da} are not
identical, then we are able see _precisely_ what the "meanings" of
these two phrases really are.   Thus typically I conceptualize a
domain of discourse as naturally containing not just one, but
virtually _countless_ possible worlds when an intension is invoked,
and I see an intension invoked whenever xorlo is used.

Largely, this is not a problem.  As I indicated, I doubt that we need
to distinguish intensions from extensions on the gadri because, again,
in most cases* as soon as you specify or glork the world and time your
working with, the intension will automatically give you the extension
you need.  Most of the time, context will give you the world and time,
but you can also be explicit.  If you apply {ca'a}, then you shift
into the actual world and get a set of actual {nanmu}.   When you
posit a fictional and counterfactual state of affairs, you posit
exactly a world with the characteristics that you have given it.  If
there is any lack of clarity in all of this, it's due to the
"headlinese" style of prose that Lojban licenses.  Ultimately,
distinguishing intensions from extensions is the responsibility of the
aspectual and modal system in Lojban, not to the gadri.

*Except for possibly two cases: specificity, which I suspect is
already covered by {lo / le}; and true kinds (i.e. what I would
somewhat tentatively and arcanely call "non-generic idealizations of
intensions"), which arguably should be marked insofar as they do
disallow any existential reading.  In the latter case, we seem to be
dealing with the intension per se rather than as a function back to
extensions.

-- 
You received this message because you are subscribed to the Google Groups 
"lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to 
lojban+unsubscribe@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/lojban?hl=en.

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.