[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] go'i: repeated referents or just sumti?
la tanatos cusku di'e
Actually, the cardinality is part of the description. Last paragraph of
page 131:
Note that the inner quantifier of "le", even when exact, need not be
truthful: "le ci nanmu" means "what I describe as three men", not
"three of what I describe as men".
I certainly agree that it is not "three of what I describe as men",
that's "ci le nanmu", but "what I describe as three men" is too
vague. It really means "each of those I describe as three men".
Of course I may be misreading that. {ro le ci nanmu} may be more
accurately translated as, "each of a group of individuals, each of which
I describe as a man, and in number what I describe as at least one,...".
That's fine by me, it doesn't really matter that much whether
the inner quantifier is veridical or not as it is incidental
information.
It's a problem that too many of the examples
of pro-sumti in The Book always assume single referent, which makes it
harder to see what the pro-sumti mean as bound variables.
Indeed. Lots of scope and quantification issues are just not
present with single referents.
Is {mi prami
mi} "each of us loves himself", or "each of us loves each of us"?
In my view, it is "we (as a whole) love ourselves (as a whole)".
I take {mi}, {do}, etc. to be masses, so they have a single
referent and no complications. The Book says they can be taken
both ways, sometimes as a mass, sometimes as individuals, depending
on context.
That may have been why I wanted to use ko'a as referring to the group in
mind the same way mi refers to the group of speaker(s). There is
precedent for pro-sumti referring to groups of individuals.
Yes, that's possible, but {lei mlatu goi ko'a} does just that.
I think I'd like {re le mlatu goi ko'a} to not quantify or bind ko'a,
leaving that until the first appearance of ko'a as a sumti, as if were a
separate variable from the unnamed one being quantified by re, but being
restricted to the same group of individuals; and for {re le mlatu ku goi
ko'a} to have ko'a be the same as the unnamed variable quantified by re.
I think that's more complicated than it need be. You can always
force the first case by using an explicit quantifier on ko'a, so
it's not much of a gain to have both possibilities from the goi,
and it is a loss because you are forced to use a lot of ku's.
Ah, I don't think I've ever run across requantification before I
starting learning Lojban, so all I have are natural language
assumptions.
Requantification is a Lojban aberration. I wouldn't think you'd
run across it anywhere else. But in ordinary language we do very
similar things. When we say "some people do this, some do that,
and others do so and so", "some" and "others" mean "some people"
and "other people", so we are in a sense requantifying from the
same set ("people"), but obviously not just from the first "some
people".
I mean for {my} to be a separate variable that is restricted to the same
group of individuals in mind as {re le mlatu}. So that {re le mlatu cu
catlu my} would mean "two cats in mind look at each cat in mind".
But that's easy to say explicitly: {re le mlatu cu catlu ro my}.
It is much more convenient to have {re le mlatu cu catlu my} to
mean "two of the cats look at themselves". In this case you could
also use {vo'a} for that, but in other cases it can be more
difficult.
Since
we have a particular group in mind it seems useful to be able to
introduce new variables restricted to that group, without aliasing them
to other variables.
I agree, and that's exactly what happens when you introduce
a new quantifier. But in the absence of a new quantifier, the
best choice is to keep the original binding.
mu'o mi'e xorxes
_________________________________________________________________
Chat with friends online, try MSN Messenger: http://messenger.msn.com