[lojban] "a bear runs" (Was; Re: Re: Again {xorlo} and Wiki.)

[John E Clifford <lojban-out@lojban.org>]

Trying to sort out various ways to say "a bear
runs" and various things that these might mean --
and various situations in which they might
correctly be used.
Notes: Remember that {cu bajra} covers the range
of tense/modal/whatever and so may mean (by
context or in intention) anything from
"necessarily always must run" (and/or "always
necessarily must run") to "possibly sometime
could run" ("sometime possibly could run"), with
a tendency, perhaps, to focus on particular
tenses of actuality and the modes of generality
and ability.
Since the modi are unexpressed, their scopes --
over one another and over scoping items in the
sumti are also undetermined: "one gnerally" and
"generally one", for example, can be sorted at
best by context -- and the example is given
contestless to allow the full range to be
explored.
In what follows, I take the predicates applied to
 {lo} expressions to be predicated
distributively, those applied to quantified
expressions individually.

A. {pa da poi cribe cu bajra}
"There is exactly one thing in the universe which
is a bear and runs"
Assuming the universe is known, this has only the
uncertainty of {cu}.  It is false if there are no
bears, or if none of them or more than one of
then runs. 

B. {(su'o) da poi cribe cu bajra}
"At least one thing in the universe is a bear and
runs"
Like the previous one except that it is true if
more than one bear runs as well as when exactly
one does.

C. {lo cribe cu bajra}
"The bear type embraces running"
Just what "embraces" means is vague, ranging from
"includes as part of its essence" to "does not
absolutely exclude." And what even these mean in
terms of bear tokens (actual bears) is perhaps
even less clear. Toward one end, from "essential"
to "includes in it core" (approximately), the
sentence would hold even if there were no bears,
being entirely a matter of relation between
concepts, definable (we can hope) in a suitable
system for dealing with concepts without
reference to manifestations of those concepts. 
The same is true at the other end.  In between
are a number of stations which (so far as I can
tell reading up on intensional system) can be
adequately defined only with reference to the
behavior of tokens (however this reference may be
disguised -- as it is very thoroughly in
Navya-Nyaya, say).  The basic way station is that
at least one token actually has the property, one
bear actually runs.  Beyond that fact, the actual
number never seems to matter, but varies pretty
freely, often under the influence of the (covert)
modi.  The typical focuses in conversation seem
to be the upper, conceptual, one   (which may
hold even if there are no bears), generalization,
and particular cases, which latter two seem to
require bears but no particular positive number
that run -- though moore for generalization,
generally, than for particular cases. (I know, we
have such cases as "Bears generally run and they
are now" where the numbers are irrelevant
altogether, but generally ...) All of these are
false if there are bears but none of them runs,
but there being more than one running bear does
not falsify the sentence. 

The above is the xorlo explanation.  For CLL and
the intermediate position we have slightly
different explanations, though almost the same
extentional applacations.
"Some of all the bears run."
In CLL, this is almost directly equated with B
and holds in the B cases.  In the intermediate
position, this reads as "There is a bunch of
bears that run" and reduces to B again, though
slightly less directly.  It also holds in the B
cases. In both of these older understandings,
there is a stylistic distinction between C and B,
in that C is more natural outside the particular
cases, for generalities and possibilities and the
like.  This is partly a matter of scope
considerations, I think, although the effect of
this can be argued both ways.  In any case, this
stylistic difference persists in xorlo, with more
obvious 
justification.

D. {lo pa cribe cu bajra}
xorlo: "The one-bear type embraces running" 
Most of what was said just above applies here as
well.  Now, of course, we automatically exclude
any property (if there are any) that require two
or more bears, say, but, taken distributively,
still allows what a bear can do/be alone. This
focus on single bears restricts the generalities
allowd, then, somewhat: bears run in packs (well,
they don't but take this as an example of a
pattern) but one-bears do not.  That aside, the
effect of this is much the same as of C. It would
seem that {lo pa cribe} could be used, like {lo
cribe} even when there were no bears, though this
is arguably misleading. When there are several
bears, using this expression does not mean that
there is only one bear in all the universe that
runs, rather it says that there is at least one
instance of one-bear that runs; there may be
several such instances without contradicting the
extentional part of this claim.
CLL: "Some of all the bears , of which there is
in fact only one, run" = "The only bear there is
runs".  Obviously this will be false if there are
more or fewer than one bears in the universe. We
get to one bear by restricting the number of
bears, not by restricting the number of them we
pick.  This is a very restrictive usage and was a
major reason for the (unofficial) change to the
intermediate position.
Intermediate:  "A one-membered "bunch" of bears
runs." This reduces to "One bear runs," but
without denying that other bears run; that is, it
is not A (as the CLL form happens to be). It
rather focuses on one of these and ignores what
others there may be.  That is, it picks out some
one-bunch without denying that there are other
one bunches that also fit.  It also does not
restrict the number of bears (except exclude
zero).  As quantifiers go, the operative one is
"some," selecting from one-bunches of bears, not
"one" selecting from bears.  This is analogous to
the way that xorlo works at the intensional
level.

E. {pa (lo) cribe cu bajra}
CLL: "One (token of the type) bear runs"
Which comes down to "One (actual -- as determined
by context, of course) bear runs." External
quantifiers bring xorlo back to extensional talk
from intensional, though the extension is still
defined as that of the earlier type expression. 
Notice that this sentence is false or meaningless
when there are no bears and false when more than
one bear in the whole world runs. That is, this
now reduces to A. 
The CLL version reduces to A also, "One out of
all the bears."
The intermediate position, however is "One out of
some bunch of bears runs," allowing that ther may
be other running bears -- in other bunches.  That
is, there is still a "some" between the one and
the all, here. Like, {lo pa cribe}, this also
points to a potentially identifiable and thus
pronominalizable particular one (and bunch, too).
Of course, the other version of E do this as
well, but not the other versions of D.