196 THE PRINCIPLES OF SCIENCE 



to judge non-significant statistical data alleged to indicate an extra- 

 sensory perception (ESP) for which we can imagine no causal ex- 

 planation. 



The view of causality here proposed is, of course, kin to Leibniz' 

 principle of sufiicient reason. However, this principle can assume two 

 quite diflFerent variants. Leibniz himself adopted the cosmologic 

 variant that asserts something about nature, e.g., for each obserx^ed 

 eflFect there exists an assignable physical cause or causes. Far dif- 

 ferent is the methodologic variant that, to guide us, asserts a heuristic 

 maxim, e.g., for each observed e£Fect seek to assign a physical cause 

 or causes. A covert cosmologic implication attaches, we saw, to any 

 recommendation of policy for dealing with nature; but in this horta- 

 tory sense "sufficient reason" becomes the primarily methodologic 

 principle I equate with the conception of causality. 



Causality as heuristic max'tm. Planck considered causality 



... a heuristic principle, a signpost— and in my opinion, our most 

 valuable signpost— which helps us find our bearings in a bewildering 

 maze of occurrences, and indicates the direction in which scientific 

 research must advance in order to arrive at fruitful results. 



The "direction" is simply that in which, setting out from an observed 

 efiFect, we pursue the search for a causal mechanism competent to 

 produce that efiFect. Hume's long-standing demonstration that causes 

 cannot be assigned with certainty is, fortunately, much easier to ig- 

 nore than to refute. And, however far "true causes" may (or may not) 

 lie beyond our grasp, the search for such causes has pro\^ed im- 

 mensely rewarding. One might then say of the idea of causality what 

 Bacon said of alchemy: 



Alchemy may be compared to the man who told his sons that he had 

 left them gold buried somewhere in his vineyard; where they by dig- 

 ging found no gold but, by turning up the mold about the roots of the 

 vines, procured a plentiful vintage. 



Just so, we discover the conditioned reflex while seeking the cause(s) 

 of behavior. 



The word "cause" has a highly variable status in the scientist's 

 \'Ocabulary. Talking about his work in progress, he is apt to voice, 

 and act on, various speculations about the causes of certain observ- 

 able efiPects. But when his work is completed the word "cause" may 



