276 THE EVOLUTION OF SCIENTIFIC THEORIES 



sound. That is, the theoretic ideas so selected will be generally just 

 those that best seize the opportunities presented, and best support 

 the demands imposed, by the conceptual and empirical situations at 

 a time and place. And these viable ideas are the correct choice, 

 whether or not they prove timelessly valid. 



Rejecting the opinion of the alchemists, Lavoisier holds the chemi- 

 cal elements immutable. We find them transmutable. But retrospec- 

 tively it appears that Lavoisier's was the only conception on which a 

 viable systematic chemistry could be founded in his time. Proust 

 denies, and triumphs over, Berthollet's opinion that genuine com- 

 pounds are at least slightly variable in their proportions. We now 

 see some merit in Berthollet's opinion. But was not Proust's view ab- 

 solutely essential for the first conception of a chemical atomic theory 

 that could function in the early 19th century? Similarly, I fail to see 

 how that theory could then have been created save, as it was by 

 Dalton, on the presently-rejected assumption that the chemical 

 elements are uniquely characterized by invariant atomic weights. 

 Newton's concepts of space and time we now question, but for the 

 18th and 19th centuries the selection of those concepts was abso- 

 lutely correct. Our present scepticism of those concepts is indeed 

 very largely the result of developments to which they themselves 

 have led us. In each case we see an earlier judgment of natural selec- 

 tion revised or even reversed. In each case, however, the earlier judg- 

 ment did single out the most powerful heuristic tool for the time. 



THE LIFE CYCLE OF A SCIENTIFIC THEORY 



The broad outline of scientific advance develops from the cyclic rise 

 and fall of scientific theories, each of which determines the general 

 direction and extent of such progress as will be made until its suc- 

 cessor is born. Kroeber finds that "The Configurations of Culture 

 Growth" can effectively be conceived in terms of a cycle of creation- 

 exploitation-exhaustion of cultural patterns. Science is itself a cul- 

 tural phenomenon, and the life cycle of scientific theories is not with- 

 out analogy to the cycle of specialized cultural patterns of which 

 Kroeber writes : 



The very selection which at the outset is necessary if a distinctive pat- 

 tern is to be produced, is almost certain later on to become a limita- 

 tion. ... It seems to be historically almost as difficult to reconstitute 

 a pattern fundamentally, or to widen greatly the scope of a growth, as 



