324 C31E.\TI\X SCIENCE 



Ptolemaic conception. The root of his scepticism? Faith, of course- 

 faith in the Pythagorean conception of a cosmos mathematically 

 harmonious. 



Consider now the last of our four pairs. At the price of doubly 

 contra\ening the principle of polarit>', we may conceive scientific 

 knowledge as generated bv the endless renewal of this cvcle: 



Imagination 



Concepts, /^ \ Facts, 



hypotheses, y observations, 



and theories \. y and experiments 



Logic 



When the empirical results do not agree \\-ith what deduction from 

 our h\potheses has led us to expect, we try to imagine some suitable 

 alteration of our original body of ideas— and so initiate the first stage 

 of the next generative cycle. The indissoluble connection of fact and 

 h\pothesis, already so much emphasized, is conceded at least im- 

 plicitly by proponents of Method. But I uish now to argue, in direct 

 opposition to Method, that primacy (if it attaches to either one of 

 this pair) belongs far more to h\pothesis. which first permits us to 

 grasp fact, than it does to fact which may then suggest some further 

 hypothesis. 



The primacy of hypothesis. Within the context of aesthetics, Wolf- 

 flin remarks. "All pictures owe more to other pictures than thev- do to 

 nature." Within the context of science, Darwin maintains, "Without 

 the making of theories I am con\-inced there would be no observa- 

 tion." \s for experiments, we saw in Chapter \T how they are con- 

 cei\ed, designed, observed, and interpreted on the basis of prior 

 hypotheses and theories. Now a companion insight of DarvWn's fur- 

 ther suggests that only certain experiments (or observations) are 

 worth making. 



How odd it is that anyone should not see that all observation must be 

 for or against some view if it is to be of any service! 



If "for" the theoretic view, it helps confirm it; if "against," it helps 

 supply the impetus for the next generative cycle. But I think a rather 



