370 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



their reference to the property of causal efficacy, or in their 

 reference to the property of temporal asymmetry. But the 

 former of these properties is not a datum, and the latter is 

 not required hi the statement of a law. For example, "in 

 the extension of a wire by weight or other force (Hooke's 

 law), the force has popularly been called the cause of the 

 extension, and the extension then figures as the effect of the 

 force, the idea being that with the force in operation the 

 extension takes place, and that without it the extension 

 would not take place. . . . Now there may be a certain 

 advantage to this popular version: certainly, it is an in- 

 stinctive stand taken in everyday life. But on careful exam- 

 ination it must be confessed that the view has little value 

 for physics. Thus, in the example of Hooke's law, as far as 

 the law itself is concerned in its symbolic form, it is just as 

 sensible to call the extension the cause and the force the 

 effect as vice versa. . . . All that the laws state is a relation 

 among symbols which represent well-defined operations in 

 the laboratory, and no notion of precedence or antecedence, 

 or dynamic enforcement is involved in them." l If. as 

 Hume maintained, causal efficacy cannot be discovered, it 

 cannot appear in the law as an empirical datum. Further- 

 more, even granting that something like "compulsion" is 

 discoverable, certainly it cannot be measured and hence 

 cannot occur as a variable in the mathematical statement of 

 the correlation. On the other hand. if. as has just been seen, 

 laws of succession can be reduced to functional relations in 

 which tune occurs as an independent variable, then even 

 the unique before-after character of cause and effect correla- 

 tions is lost. One can infer just as readily from an effect and 

 a time to a cause as from a cause and a time to an effect, 

 since each is the ground of which the other is a consequent. 

 The sole differentiating features of causal laws are therefore 

 lost, and laws of this type may be replaced by statements of 

 functional relations. 



This presumed reducibility of causal laws to functional 



1 R. B. Lindsay and H. Margenau, Foundations of Physics, p. 18. 



