From: Robert Owen (rowen@technologist.com)
Date: Sun Oct 24 1999 - 23:47:32 MDT
Robert J Bradbury wrote:
> On Sun, 24 Oct 1999, Clint O'Dell wrote:
>
> >
> > When you try something 1000 times and it "accidentally" produces the same
> > result all 1000 times then it's not an "accident". No faith involved.
> > You still haven't shown me where faith enters science.
> >
I missed Clint's original post. But the question naturally arises of the
level of confidence one may have in the expectation that, if a fair
coin is tossed in the air and allowed to fall randomly 1000 times, the
result will be exactly 500 heads and 500 tails? Answer: very little.
How about 10,000 times, or 100,000 times? Can this increase your
level of confidence? So, Clint, why do you think that at some number
of tosses the result OUGHT TO BE exactly 50% tails, 50% heads?
Answer: because your BELIEVE in the "Laws of Probability". Can you
prove this belief rationally justified even though one cannot possibly
design an experiment to test these laws (hypotheses) empirically?
> Faith is involved if you "believe" it will produce a different
> result the 1001st time. :-) In science you trust your senses
> and rational thought. In religion you trust your beliefs.
The necessity for "faith", Robert, began with the formulation of the
scientific method itself: "Induction" was first formalized by Frances
Bacon in the 17th century. In his book, Novum Organum (1620/ 1952),
Bacon advised that facts be assimilated without bias to reach a
conclusion. The method of induction he suggested is the principal way
in which humans traditionally have produced generalizations that
permit predictions. But there is a problem with induction.
It is both impossible to make all observations pertaining to a given
situation and illogical to secure all relevant facts for all time, past,
present and future and for all places in the universe. However, only
by making all relevant observations throughout all time and space
could one say that a final valid conclusion had been made. This is
the problem of induction. On a personal level, this problem is of little
consequence, but in science the problem is significant. Scientists
formulate laws and theories that are supposed to hold true in all
places and for all time but the problem of induction makes such a
guarantee impossible.
Science has devised several workarounds, e.g. the use of statistical
methods in the analysis of data, despite limited sampling. There are
of course others. But after all is said and done, your assertion that
faith is required is valid provided you extend it to include belief in
your abstract methodologies of data processing as well.
Thank you for your balanced statement,
Bob
=======================
Robert M. Owen
Director
The Orion Institute
57 W. Morgan Street
Brevard, NC 28712-3659 USA
=======================
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:05:36 MST