"Mike Ruskai" writes:

On 13 Jul 2003 20:43:01 -0700, Richard Alexander wrote:

Al Klein wrote in message . ..

The definition of "scientific" doesn't include "testable".

I think we should at least settle this question; Can an hypothesis,
theory, principle, claim or statement be scientific if it is not
testable?

Depends on what you mean by "testable". For a theory to be scientific, it
must at least be falsifiable. Whether that's the same as "testable" or
not is mostly a matter of semantics.

Popper's theory of science demanded that a statement had to be in
principle experimentally falsifiable if it was to be considered
scientific. Being unfalsifiable was therefore the mark of the
unscientific.

This is now regarded as an oversimplification. For example, Lakatos, a
pupil of Popper, showed that there was a class of scientific
statements which were not falsifiable by an experiment. Instead they
were used to generate falsifiable hypotheses. Instead of being
abandoned because they had been falsified, they were abandoned when
they became unproductive generators of falsifiable hypotheses.

Unfortunately Popper's oversimplification caught the imagination of
science teachers and armchair philosophers, and its Procrustean view
of science is still believed to be the last word by many, whereas it
was only Popper's hypothesis of how science worked, which has since
been falsified :-)
--
Chris Malcolm +44 (0)131 650 3085 DoD #205
School of Informatics, Edinburgh University, 5 Forrest Hill,
Edinburgh, EH1 2QL, UK. [http://www.dai.ed.ac.uk/homes/cam/ ]