252 THEORIES AND MODELS 



No doubt the positivists' is an ideal science— in the sense that 

 "ideal" means the fanciful as contrasted with the actual. No scientific 

 theory has ever been re-constructed, much less created, in a rigor- 

 ously formal way; no useful report of observables has ever been 

 made free of all burden of inference; no fully productive scientist has 

 ever renounced all hope for explanation, or foresworn all speculation 

 about the true nature of a "real world" unobservable as such. Ap- 

 proach to the ideal— insomuch as necessary, insofar as possible— has 

 throughout the modern era been a routine part of the professional 

 common sense of scientists who have, however, been inspired and 

 guided by concerns and concepts positivism would outlaw. 



Bent on discovering the system of the real world, Copernicus did 

 not need positivism to teach him the wisdom of freeing observables 

 from their customary freight of inference. \\^holeheartedly accepting 

 the metaphysical assumption of a cosmos mathematically harmoni- 

 ous, Kepler did not require the tutelage of positivism to see the sig- 

 nificance of an apparently trivial discrepancy between theory and 

 observable. Not Galileo but Galileo's scholastic opponents preach the 

 supremacy of obsers^ables; and all this devout soul's trouble with his 

 church grew out of his stubborn refusal to accept what was urged 

 on him by friend and foe alike: the positivist view of scientific theory 

 as no more than economic description. Newton had not to be taught 

 the virtue of formal construction (virtue already amply apparent to 

 Euclid and Archimedes ) when— his thought suflFused by a metaphysi- 

 cal corpuscularianism required, for example, to gi\'e meaning to his 

 concept of mass— he made a universe with his concept of a gravita- 

 tional force. These heroes of scientific history did not become so 

 despite the fact they were not positivists: they could become so only 

 because they were not positivists. 



The manifold displays of the heuristic power of atomism are 

 founded on the willingness of scientists to suppose atoms real. Wil- 

 liam Wollaston, the highly gifted contemporary of John Dalton, had 

 a "properly" sceptical opinion of the status of the Daltonian atom. 

 Informed by the same strain of thought soon to produce positivism, 

 Wollaston proposed completely to bypass the difficult problem of 

 assigning relative weights to purely hypothetical atoms. 



. . . since the decision of these questions is purely theoretical, and 

 by no means necessary to the formation of a table [of equivalent 

 weights] adapted to most practical purposes, I have not been de- 



