THEORIES AND MODELS 239 



position of elements sometimes superficially syncretic. Thus, for ex- 

 ample, with an electronic wave function we compute the probability 

 of finding a corpuscular electron in a certain region. We speak and 

 think then in terms of waves and particles quite obviously not the 

 waves and particles of everyday experience, but somehow related 

 to them. The relation we conceive with the hierarchic model, one 

 example of the function of which is described by Hutten with an apt 

 and vivid metaphor. 



Physical theory developed by gradually overcoming the limitations 

 of the original model; but this does not mean that . . . the original 

 model is completely abandoned. Some connexion with the original 

 model is retained; the model is re-designed and, thus, refined in order 

 to agree better with our latest experiments. What remains is the 

 wave equation, together with a minimum intei-pretation in terms of 

 experience upon which the application of the equation rests. It is 

 like the grin and the Cheshire cat; the picture of the cat has receded 

 into the background, but knowing that there was once a cat we un- 

 derstand that the residual phenomenon may be intei-preted as a grin. 



THE "SUPERFLUITY" OF THE PHYSICAL MODEL 



Working with a purely formal system, the logician seeks assurance 

 that its various premises are self-consistent. In like fashion the scien- 

 tist desires assurance that his theoretic premises are mutually com- 

 patible. If Nicod is right, much the same procedure will be brought to 

 bear in both cases : 



The discovery of one system of meanings satisfying a group of 

 axioms is always logically very important: it constitutes the proof that 

 these axioms do not contradict one another; and this is the only 

 known proof of consistency. 



If all the primitive concepts of a scientific theory have experiential 

 denotations, we may approve the theoretic premises simply as well- 

 established colligative relations. But when some or all of these con- 

 cepts lack such denotations, the self-consistency of the premises can 

 be examined only in the context of the theory as a whole. Consider 

 for example how Planck's constant, appearing in the premises of a 

 quantum theory, also appears, with other "fundamental constants," 

 in many derivative relations ( those referring to blackbody radiation, 

 photoelectric eflFect, emission spectroscopy, and so on). The theoretic 



