[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lojban] fancu



In a message dated 10/6/2001 6:50:56 PM Central Daylight Time, jjllambias@hotmail.com writes:


Could you give an exampleof this? Would not those that are
equivalent always be rephraseable so as to fit the matrix?


Yeah, but they might not be the one he knows, thinks of, etc. etc.  The intensional problem that extension-claim theory has. Almost every answer actually has an extension-claim equivalent, which somebody might think of, so set-of answers covers that case, but is not restricted to it.  We are ready for a wide range of possibilities in each case, not just the one.

<{noda
>kalma
>le zarci},  for example,

But I have always insisted that this answer _is_ included. It is
always the one that makes the extension analysis fail, because it
is not part of the extension of {le ka ce'u klama le zarci}.>

Well, surely knowing that the empty set is the extension of {le ka ce'uklama le zarci}  would count -- except for the intension problem.  That claim should be in the set of answers.

<>and -- perhaps related to that last bit -- {na'i},

That one I would probably exclude. Could you give an example?>

Classics: "Have you stopped beating your wife?" when you either don't have a wife or have never beaten her.  "How do you know the distance tothe moon?" when you don't know it.  And so on.

<I don't see how this follows from the purported omissions.
I think I tend to rely on the specificity of {le} to select
the acceptable answers, but in any case I am not at all sure
that my analysis is complete.>

Very often "Who" questions have a presupposition that someone does whatever.  When the answer is {noda broda}, it is also -- and more forcefully -- {na'i}.  When dealing with And, it turns out better not to use {le}, since he tends to make much of its idiosyncrasies -- which can be useful, but can also get you into trouble (who all has to agree on what is referred to and how detailed must that agreement be?)

<an array of propositional
>functions,
>rather than taking the whole as a function to indirect questions.

I now think it has to be the other way around. {makau} is a
dependent variable and {ce'u} the independent one, in a manner
of speaking.>

For the {dunli} -{frica} cases, it works nicely the way you had it before.  The new way is trickier on those cases, though it still comes down to the same sentence eventually.  I flipflop back and forth and the Logic texts have no guidelines at all.