Re: [lojban] About plural 'ro'

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

----- Original Message ----
From: Jorge Llambías <jjllambias@gmail.com>
To: lojban@googlegroups.com
Sent: Wed, April 21, 2010 1:53:13 PM
Subject: Re: [lojban] About plural 'ro'

On Wed, Apr 21, 2010 at 2:57 PM, John E Clifford <kali9putra@yahoo.com> wrote:
>
>> [But the question is, how do these referents relate to the referring expression.]
>
> We can (and do) specify that, but why should that be relevant to how
> they then relate to other referents?
>
> [[I'm not sure what you mean.  What other referents are involved?  Referents of what?]]

Referents of another sumti. In general what we do is make claims about
the relationships between the referents of one sumti and the referents
of some other sumti. "mi tavla do" makes a claim about a relationship
between the referents of "mi" and the referents of "do". As a special
case, we make claims about the referents of a single sumti only. In
either case, why should the way the referents of each sumti came to be
referents of that sumti be at all relevant to the relationship we want
to claim between those referents and the referents of another sumti,
or relevant to the claims we may want to make about just those
referents?

It seems to me that two independent processes are involved: one is to
get to the referents of each sumti, and another, different process, is
the claims we may want to make about those or about how they relate to
the referents of some other sumti.

[[[Well, presumably, the connection is mediated by the predicate 
involved, 'tavla' say.  But notice that the issue here is about 
quantification, not about the relation of one referring espresions or 
what it refers to to another.]]]

>>> [Yes, because plural quantification, taken alone is clearly wrong.]
>>
>> If you say so. But I don't see what's wrong with it. It is what it is,
>> neither right nor wrong.
>
> [[As an attempt to represent natural language expressions, it fails to give the right results in crucial cases, as you have pointed out several times.]]

Which is why I say we should not be using plural quantification to
represent those crucial cases. The natlang each/every/all should not
generally be represented with a plural universal quantifier.

What I don't see, is why you say we should use it anyway, and make
changes somewhere else (especially since it is not even clear what
those other changes would be).

[[[Well, you want plural reference and thus plural quantification, so, to satisfy you (it doesn't seem to be working) we need some other modifications.  Maybe the ones suggested are not the right ones, but randomly (I don't see any rules here yet) choosing to use one kind of quantifier or another doesn't seem to help much.]]

>> The referents of a plural term can
>> instantiate a singular variable, one at a time. The referents of a
>> plural term can instantiate a plural variable, one or more at a time.
>
> [['AxFx, therefore Fa' is a case of instantiation.  If 'a' is a term with plural referents, and x a singular variable, then the premise is true but the conclusion false, since only one referent of a, not all of them are Fs.  Or rather, there is nothing inherent to prevent this, whihc is quite enough.]]

Right, the corresponding inference when you have a singular quantifier
'Ax' and a plural term 'a' is:

"AxFx, therefore if a is one, Fa"

[[[Where did the "if a is one come from?  It is not anywhere in the formula above.  And, of course, it won't help in the reverse case, even if it does here.  But it doesn't of course, because a ios not one but plural ex hypothesi]]]

>> This discussion has nothing or very little to do with xorlo, as far as
>> I can tell. It started because I said "ro" has to be (it is most
>> convenient for it to be) the singular universal quantifier, and you
>> apparently think it ought to be the plural universal quantifier.
>
> [[If you want to have plural reference, yes.]]

But just saying it doesn't make it convincing.

I gave examples for why I think a singular universal quantifier is a
must have. I don't see much use for a plural universal quantifier,
since in the rare cases when we do want to express what it expresses
we can do it via singular quantification:

    ro lo su'o mei be lo broda cu brode
    "Each of the at-least-one-somes of broda, is brode."

It is a little longer than a single word, but for the rarity of its
use plenty enough.

Alternatively, or additionally, we could have a plural universal
quantifier in addition to "ro". Perhaps something like "roro" (or
something better thought out), or even a new cmavo.

What I don't see as a good idea is redefining "ro" as a plural
quantifier, because its use as singular quantifier is very common and
needed.

[[[Agreed, it is needed, but it is incompatible with plural reference, without some further work.  So we also need a plural quantifier.  And then a way to tell which to use.  There are two ways of thinking about plural reference, one as a sort of conjunction, the other as a sort collaboration (e and joi or some such).  If you use the first sort, then singular  quantifiers are no problem, but you can't do collaborative references. If you use the second sort, then collaborations are easy, but singular quantifiers no long work. Suppose a piano (or something) ways a ton and a half.  Then, "all the people who carried the whatever will be rewarded" is going to be either either vacuously false or vacuously true, even though the piano got carried (if we are going by singulars) or either nobody gets a reward or everybody gets several (if we go by plurals).  Since both plural reference of at least the latter sort and plural quantification (which follows) seems
 useful, some trick is needed.  And one that can be reliably and consistently used,]]]

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.