Wednesday, February 16, 2005

some contradictions are true...

SAINT of the day (some days ago) is Onesimus, ex-slave of Philemon.

CONTRADICTIONS. I can't believe I haven't seen any of Graham Priest's stuff before, well, except for his Logic: A Very Short Introduction. Only recently, I discovered his 'What's So Bad About Contradictions?’ (The Journal of Philosophy, Vol. 95, No. 8, Aug. 1998), a sustained and substantial brutalization of the Law of NonContradiction (LNC).

These are my reading notes thereof.

There are five reasons to abhor contradictions:
(1) Contradictions entail everything.
(2) Contradictions cannot be true.
(3) Contradictions cannot be believed rationally.
(4) If contradictions were acceptable, no one could ever be rationally criticized.
(5) If contradictions were acceptable, no one could deny anything.

GP considers (2) to boil down to an appeal to the Law of Noncontradiction (LNC).
G.P. accuses Aristotle of equivocation – (E) the slide between some contradictions are false and all contradictions are false. There's some further discussion of the problems with the Aristotelian arguments here.

Now there are 4 further arguments for (2),

A - Contradictions lack meaning.
1. Contradictions have no content, because they have no meaning.
2. therefore they have no true content.
Note that the classical logician is unable to support this argument; he believes that a contradiction implies everything, rather than nothing. So the contradiction has ‘total’ content.

But contradictions are meaningful utterances, else how could they be seen to be contradictions (and so, supposedly, always false) at all?

B - If contradictions are true, nothing could be meaningful.
1. Some claim is meaningful only if it rules out some other claim.
2. If A doesn’t rule out not-A, then A isn’t meaningful. (The minimum condition that A has to fulfill to be meaningful is to rule out not-A)
So, if claims of the form [A & not-A] are admitted to be meaningful, nothing at all is meaningful.

This suffers from the same fault pointed out in (E) above, viz. sliding between some X and all X.
The LNC requires only that some statements do not rule out their negations. Consider: the negation of 'nothing can be asserted to be and not to be at the same time and in the same repect' is 'something can be asserted to be and not to be at the same time and in the same repect'. However, It seems to be widely held that the proper negation is 'everything can be asserted to be ...etc.'

Also, the argument above relies on the claim that all statements rule out their own negations.
But then consider also: ‘Everything is true’. This rules nothing out and is clearly meaningful. Its negation, ‘something is false’, is also true.

There's more to follow, but this looks very interesting.

the EXPERIENTIALIST is back. Here.

the ENGAGEMENT is announced of a dear, loud, and quite mad friend. Congratulations!

No comments: