THE EVOLUTION OF SCIENTIFIC THEORIES 273 



brings one species of opinion to triumph, in the struggle with its 

 competitors, if only it displays superior heuristic power. 



One or a few men grasp a daring new idea. If it is to survive they 

 must at once make it count for something: well has it been remarked 

 that when skating on thin ice one must keep moving. Let us suppose 

 the new idea heuristically powerful enough to lead those who enter- 

 tain it to a few notable discoveries. These draw the attention of 

 others who aspire to discovery and, adopting the idea at least provi- 

 sionally, some of these recruits may also profit by it. Those most suc- 

 cessful in this way will be favored in the competition for posts of 

 influence, particularly in the universities. Wherever they are or be- 

 come established, their successes will make them especially attrac- 

 tive to students who also aspire to be discoverers. From the many 

 who compete for the privilege of working with the successful investi- 

 gator, he will select what seem to him the most promising men of the 

 coming generation. Gaining command of the new theory at the out- 

 set of their careers, these students of superior ability can, through 

 their own discoveries, still further enhance the theory's luster. 



Meanwhile the die-hard adherents of the old theory make few 

 comparably notable discoveries: the heuristic power of their ideas 

 will by now be largely spent. As time passes such a resistor of con- 

 version may come increasingly to acquire the reputation of a crank 

 —which cannot but still further decrease his capacity to attract able 

 students. And even his students will become familiar with the new 

 theory, through the publications in which its partisans report their 

 unmistakable successes. Given the evidence of those successes, the 

 students may adopt the theory despite their master's opposition. At 

 a slightly later stage the new theory finds its way into textbooks: 

 however conservative the authors of these may be, they can hardly 

 fail to mention a view that provokes important discoveries the texts 

 cannot ignore. And more vigorous statements of the new theory will 

 be given in texts written by its partisans, whose own successes go far 

 toward ensuring wide use of their books. As the terminology of the 

 new theory thus gains general currency it is even more firmly estab- 

 lished—in precisely the manner Joseph Black had in mind when he 

 noted the subversive effect of the new chemical nomenclature La- 

 voisier and his cohorts had so designed 



. . . that the very denominations of the different objects should imply 

 the doctrines of their theories; so that, by using this language, it 



