368 



BELL SYSTEM TECHNICAL JOURNAL 



The recognized difference between these two situations leads us to 

 make the following simple postulate: 



Postulate 1. All chance systems of causes are not alike in the 

 sense that they enable us to predict the future in terms of the past. 



Hence, if we are to be able to predict the quality of product at least 

 within limits, we must find some criterion to apply to observed vari- 

 ability in quality to determine whether or not the cause system pro- 

 ducing it is such as to make possible future predictions. 



Perhaps the natural course to follow is to glean what we can about 

 the workings of unknown chance causes which are generally acknowl- 

 edged to be controlled in the sense that they permit of prediction within 

 limits. Perhaps no better examples could be considered than those 

 which influence length of human life and molecular motion, for it often 

 appears that nothing is more uncertain than life itself, unless perhaps 

 it be molecular motion. Yet there is something certain about these 

 uncertainties. In the assumed laws of mortality and distribution of 

 molecular displacement, we find some of the essential characteristics 

 of control within limits. 



A . Law of Mortality 



The date of death always has seemed to be fixed by chance even 

 though great human effort has been expended in trying to rob chance 



UJ 



O 

 < 



UJ 



> 



«>- 



t^ 



< CD 



^m 

 ±o 



z 

 o 



I- 

 u 

 < 

 a. 



45 50 

 AGE IN YEARS 



Fig. 1— Law of mortality — ^law of fluctuations controlled within limits. 



of this prerogative. We come into this world and from that very in- 

 stant on are surrounded by causes of death seeking our life. Who 

 knows whether or not death will overtake us within the next year? 



