From: Don Winterstein (dfwinterstein@msn.com)
Date: Sat Oct 04 2003 - 04:56:10 EDT
Walter Hicks wrote in part:
"...It is indeed fair to put scientific theories to the
falsifiability test. If there is data to disprove the universality of a scientific notion,
then it must be taken seriously. A scientific theory must be right 100% of the time - not
"most" of the time."
No scientific theory is always right. Disproving a theory also is often difficult or impossible. A theory that predicts precise quantitative results can be disproved easily if and when experiments don't yield the predicted results; but in practice it's often possible to modify or reinterpret theory so that it comes into agreement with data. It may be practically impossible to disprove some accepted theories that don't predict quantitative results.
We would like our theories to be always right, but in practice we demand only that they embody the best available unifying principles. By "best" we mean most elegant, most widely applicable, etc. Newton's theory of gravity was not "right" about the orbit of Mercury, so physicists made ad hoc assumptions to bring it into agreement. Nevertheless, for a long time Newton's was the best gravity theory available and as such was widely if not universally accepted. Theories of less quantitative sciences (such as geology) at times cannot elegantly account for particular observations, but we accept them because they're the best we have.
Long ago, as I was making the transition from physics to geophysics, a crusty old geophysicist explained to me the difference between physicists and geophysicists: "You physicists actually believe your models," he said. At the time I wasn't sure what he meant, but it's long since become clear: Our theories are not reality, and we put ourselves at risk if we start to believe they are.
Don
This archive was generated by hypermail 2.1.4 : Sat Oct 04 2003 - 04:53:08 EDT