378 THE REAL WORLD 



conceive. But such confirmations do, I think, add up to a problem 

 perhaps worthy the attention of those who maintain that such par- 

 ticles "exist only in our understanding." 



Some may wish entirely to deny that cross-fixes are significant, be- 

 cause there is some necessary concordance of theories and experi- 

 ments alike made by human scientists. But recognizing the argument, 

 one recognizes too that it is devoid of weight until hoiv such con- 

 cordance necessarily arises is revealed by a demonstration yet to be 

 produced and, in my opinion, most unlikely ever to be produced. For 

 consider just those most remarkable cross-fixes that involve the char- 

 acteristically different data, assumptions, and theories of different 

 sciences. In the 19th century relative atomic weights were derived by 

 chemists from chemical data and a number of special assumptions of 

 simplicity. In the 20th century substantially the same set of (aver- 

 age ) relative atomic weights was derived by physicists from data on 

 discharge tubes interpreted with the aid of a quite different body of 

 postulates. What are the odds against such success if chemical and 

 physical theories do not represent, each in its own way, some element 

 of reality? Or again, characteristically chemical data characteristi- 

 cally interpreted by 19th century chemists led them to impute certain 

 structural configurations to various organic compounds. In the 20th 

 century the physicist, using X-ray diffraction data and a different set 

 of theories and assumptions, is led quite independently to the assign- 

 ment of almost exactly the same structures. And so on. "Chance," 

 acceptable as explanation of one highly improbable event, is quite 

 insufficient to explain the many such as these. 



To argue that humanly invented and selected scientific principles 

 and theories in some measure depict the structure of the real world 

 is, perhaps, to ask acceptance of a miracle. But consider the force of 

 Hume's dictum: 



... no testimony is sufficient to establish a miracle unless the testi- 

 mony be of such a kind, that its falsehood would be more miraculous, 

 than the fact which it endeavors to establish; . . . 



I offer as testimony the discoveries that scientific theories of ever- 

 more extraordinary simplicity ha\'e immense heuristic power and 

 lead to truly remarkable cross-fixes. These are facts explicable, I 

 think, only on the basis of the miracle I propose. Denying this testi- 

 mony— e.g., regarding scientific objects as myth— one must suppose 



