292 THE EVOLUTION OF SCIENTIFIC THEORIES 



do not even think of trying to derive the same results by the formi- 

 dablv complex techniques of general relativity. 



What is the status of a theory thus subordinated but not slain? A 

 respected physicist who is also a philosopher of science writes: "New- 

 ton's theory became false when Einstein's was accepted." What has 

 become "false" we would think must be discarded. Yet every day and 

 everywhere scientists continue to use Newtonian physics, and find 

 in it a tool of immense heuristic power. But when we speak of truth 

 or falsity clearly we speak not of science but of cosmology; and it is 

 solely as an element of cosmology that "Newton's theory became 

 false. . . ." In its own familiar realm, classical mechanics remains 

 completely valid as a scientific theory. Einstein himself saw this very 

 clearly when, with reference to his own work, he wrote: 



No one must think that Newton's great creation can be overthrown 

 in any real sense by this or any other theory. 



The ''problem' of dual description. Eddington asks us whether the 

 hard, smooth-surfaced table perceived by the man of common sense 

 is not "really" a thin haze of electrons and nucleons. We suppose 

 there can be but one correct description of the "real object": yet here 

 we are presented with two apparently wholly irreconcilable descrip- 

 tions, each of which makes a very strong claim to consideration. So 

 perplexed, we may begin to wonder whether both descriptions are 

 in some way illusory, whether any description can be aught but 

 illusory, and perhaps even whether there is any "real object" to be 

 described. These are surely absurd lengths to go in the face of a 

 pseudoproblem which arises simply from the co-existence of various 

 more or less subordinate theories. 



The situation is much clarified by the highly suggestive analogue 

 proposed by Toulmin. Consider, he says, the alternate representa- 

 tions of the same region ofiFered by two di£Ferent maps: a road map 

 and a physical ( i.e., topographic ) map. 



. . . many things can be mapped on the physical map which there 

 is no way of putting into the road map: this is a consequence of the 

 ways in which the two maps are produced, and of the comparative 

 poverty of the system of signs used on the road map. On the other 

 hand, given the physical map, one could produce a satisfactory road 

 map: all that appears on the road map has its counterpart on the 

 more elaborate map, even though in a dififerent form. But this does not 



