Re: MD Quality and information theory

From: Graham Wyatt (graham@gpwyatt.co.uk)
Date: Tue Jan 01 2002 - 12:31:55 GMT


Hi All,

RICK:

> ---"Knowledge" is an opinion that has withstood all previous challenges,
> though with no guarantee it will always do so in the future. Once
> successfully challenged, it is no longer knowledge.

GRAHAM:

Certainty sounds a little strong, but I don't think that you can get away from it. You can't say that you know something if you
allow any uncertainty. If you do, what you hold to be true is simply an opinion. This is taking me to the conclusion that most of
what we think of as knowledge is only, in fact, opinion. At the moment, with one possible exception, I have the feeling that
knowledge (in the sense that I mean it) does not exist. It is an ideal that we must always try to move towards, but perhaps we can
never reach.

The possible exception that I am thinking of are mathematical proofs. Please tell me if I am wrong (I am no mathematician), but my
understanding is that a mathematical proof is only accepted if it is irrefutable. Pythagorus's theorem will always be true - no
matter what end of the universe you exist in.

ROG:
As one may expect of a fan of Pirsig and James adopting an idea called
Evolutionary Epistimology, which is an offshoot of Popper, I would be a
strong supporter of Erin's view that we are in search of the most certain
rather than totally certain knowledge. Popper strongly encourages the
falsifiability of beliefs, and that one must ensure one avoids dead ends
where a belief cannot be self negated. Interestingly, Popper's ultimate
argument rests on something similar to James' pragmatism -- that the value of
a belief is in the results it produces (including its correspondence with
experience and other beliefs).

GRAHAM:

As I said to Erin, I think we are in agreement. If certainty is unobtainable, then we have to look at what remains. To me, the whole
point of reducing uncertainty is that it increases value. A belief with a low level of uncertainty has high value because it
provides a closer match to reality (which is itself perhaps unknowable) and is consequently more useful. Beliefs with a high level
of uncertainty can persist for long periods of time, but they are less useful, and their divergence from reality will eventually
reveal itself.

But I also think that acknowledging that these beliefs are simply opinions also has value, because it makes it much easier to
abandon them when something better comes along. Hanging on to beliefs that have outlived their usefulness can be a dangerous
practice (using the sacrifice of virgins as a method of climate control, when it would be better to modify agricultural techniques
and investigate methods of water conservation is an example that comes to mind).

ERIN:

So after reading your email again I would like to apologize.

GRAHAM:

No need - you were quite right to object. It was a sloppy example and I really should have said that "Uncertainty is reduced to
almost zero". I need to brush up on my communication skills, and I see that I need to be impeccable here. Also, I am a little
excited at the moment, and I suppose I should apologise for that and ask you to allow me some room to ramble . I have been thinking
about this stuff for close on 25 years, and this is the first time I have ever really discussed it with anybody.

Graham

MOQ.ORG - http://www.moq.org
Mail Archive - http://alt.venus.co.uk/hypermail/moq_discuss/
MD Queries - horse@darkstar.uk.net

To unsubscribe from moq_discuss follow the instructions at:
http://www.moq.org/md/subscribe.html



This archive was generated by hypermail 2b30 : Sat Aug 17 2002 - 16:01:45 BST