396 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



on method or on subject matter; in the former case reference 

 must be made to such methodological features as the dif- 

 ferences between rational and empirical sciences, experi- 

 mental and non-experimental sciences, statistical and indi- 

 vidual sciences, quantitative and non-quantitative sciences, 

 etc. ; in the latter case reference must be made to such meta- 

 physical problems as the classification of the categories, 

 "levels'" of reality, dependence and independence of the 

 sciences, etc. The solution to the problem of human freedom 

 presupposes that generalization has already been made as 

 to the existence or non-existence of causal determinations in 

 the inorganic realm. Problems of the nature of reality are 

 based either upon logical considerations pertaining to the 

 character of scientific symbols and the general nature of 

 scientific explanation, or upon metaphysical considerations 

 reflecting the interrelations of the sciences. Thus the solu- 

 tions to speculative problems approach a higher degree of 

 adequacy to the extent to which they are based upon careful 

 analyses of the more critical problems. The strict nature 

 of this dependence will be brought to light in the illustra- 

 tions to be given in the following chapters. 



In the third place, since the inference in the case of each 

 of these problems is to a realm which is extra-scientific, 

 there is reason to doubt whether the criteria of scientific 

 verifiability are strictly applicable. Just because the infer- 

 ences extend out into that wider realm in which man reacts 

 morally, religiously, and esthetically, one may not be justi- 

 fied in estimating the results in the cold light of reason. 

 Broad suggests that conclusions in speculative philosophy 

 are almost certain to be influenced by the condition of one's 

 liver and one's bank account. Many of the inferences are 

 clearly little more than rationalizations of beliefs held upon 

 grounds other than those afforded by science. Eddington 

 suggests that we must avoid the "tendency to use 'reality' 

 as a word of magic comfort like the blessed word ' Mesopo- 

 tamia'" ; x we are continually confusing "reality' with 



1 Nature of the Physical World, p. 327. 



