There are no formally codified regulations of what makes a scientific theory acceptable; but the generally accepted criteria are two: It must explain all relevant observations, and it must be contradicted by none.
The reason why the second criterion, the absence of contradictions, is more important is that failure to explain everything may be tolerated, but contradiction cannot. For example, much of macroscopic mechanics fails to explain what goes on inside the atom. But that is a limitation, not a contradiction. If somebody pulled himself up by his own bootstraps, that would be a contradiction (of the theorem that the center of mass cannot be shifted by internal forces).
The requirement of no contradictions is very exacting: A thousand confirmations of a theory make the theory look better and better, but they do not irrefutably prove it; yet a single contradiction is enough to kill it-once it is clear that the contradiction cannot be explained.
The case of heat illustrates this. It is only some 150 years that we know it to be a form of energy. Up to the early 19th century heat was thought to be an all-pervasive fluid (associated with the combustion-causing substance called "phlogiston") that flows from hotter to colder bodies until their levels even out.
It was a pretty and very successful theory. It explained virtually everything then known about heat. Heat transfer, conduction, convection, radiation; heating values and capacities, why you kept warm under a feather bed and froze in the open; and practically everything else.
Until an "Oh-yeah?" man came and asked an embarrassing question: If you rub two pieces of ice against each other in freezing weather, why do they melt? Where does the fluid come from? The contradiction remained unexplained, for in those days they had no Ottingers, Weavers or other politicians to meddle in science and to get themselves elected by phlogging the phlogiston. And so it died, killed by the ice test (and some other cases of heat generated by friction, notably in boring metal for guns).
|
Vol. 11, No. 1
Newsletter: Access to Energy Newsletter Archive Volume: Issues Issue/No.: Vol. 11, No. 1 Date: November 29, 2004 11:04 AM Title: Ten years
Copyright © 2004 - Access to Energy Newsletter Archive
|