7 2 ARISTO CRA C Y AND E VOL UTION 



Book i fact which, when once attention has been called to it, 

 common sense will perceive to be as obvious as 

 it is important. 



It was just now observed that the same thing 

 can be great and not great, according to the things 

 with which we compare it. In the same way the 

 same statement may be true or not true, according 

 to the nature of the discussion on which it is brought 

 just as state- to Dear L e t us take as an example those familiar 



ments of 



averages statements of fact which are given in terms of aver- 



and classifica- T r i i 



tions of goods ages. If the vast majority of any given population 

 andrevdant vary in height between the limits of five feet six 

 pose n an P d U faise a d six feet, the statement that a man's average 



and irrelevant height IS from five feet Seven tO five feet eight 

 for another. 



would be a truth most important to the producers 

 of ready-made overcoats. But if half the population 

 were two feet high, and half rather more than nine 

 feet, to give the average stature as something like 

 five feet seven would be for the coatmakers the 

 most absurd misstatement imaginable, and would 

 lead them to make, if they acted on it, garments 

 that would fit nobody. 



Let us turn from the question of the truth of a 

 statement to the question of its mere relevance ; 

 and we can illustrate what has been said by an 

 example equally homely. In the transference of 

 goods by rail, these have to be sorted according to 

 bulk, weight, shape, fragility, perishability, and so 

 forth. In deciding which are to be sent by fast 

 trains, and which by slow, the primary question will 

 be that of perishability. When the perishable and 



