330 PART V- WORKING STAGE. 



harm would ensue ; but almost invariably daylight only emerges 

 when we have proceeded far in an investigation, and accord- 

 ingly fractional generalisations, more often than not, are largely 

 erroneous, and cannot be profitably combined. It is of decided 

 consequence, therefore, that the investigator should not relax 

 his efforts until he has virtually reached the heart of his 

 subject, which is but reaffirming our suggestion that men and 

 women should devote virtually a life-time to solving one prob- 

 lem to some extent. In view of the confusion, and even 

 apathy and antipathy, arising from the publication of a long 

 series of disjointed, necessarily crude, gravely misleading, and 

 often contradictory essays, there would be an enormous saving 

 of time and effort if concentration of interest and activity on 

 a large scale became universal in science. From another direc- 

 tion, then, we reach the conclusion that the watchword for 

 all investigators needs to be to aim at comprehensive generali- 

 sations. 



In matters of practice the above is admirably illustrated by 

 two parallel lines of enquiry conducted during the last genera- 

 tion. On the one hand, psychologists and physiologists have 

 published a considerable number of monographs bearing on 

 industrial efficiency and increased productivity, and relating to 

 such subjects as expenditure of human energy, fatigue, and the 

 like. The practical effect of these isolated and exceedingly 

 limited efforts has been virtually negligible. On the other 

 hand, efficientists in America, imperfectly equipped scienti- 

 fically, decomposed certain industrial processes into their ele- 

 ments and recomposed them on strictly economic lines, taking 

 account, of all noteworthy accompanying circumstances. Their 

 interest lay in the totality of the process, and their practical 

 success was almost instantaneous and far-reaching. The very 

 comprehensiveness of their enterprise saved them from innumer- 

 able fallacies and removed countless complexities. 



Whilst extremely wide, but unexplained, generalisations fre- 

 quently convey little meaning to the student, and are therefore 

 of small value, 1 it is of material importance in actual scientific 

 work to posit them sometimes in order to be clear as to the 

 final goal. 



Thus, suppose we observe that comparatively hot water 

 issuing in a thin and weak stream from a tap feels relatively 

 tepid to the hand placed immediately underneath. Noting the 

 principle implied, we may then, without graded generalising, 



"Bacon has judiciously observed that the axiomata media of every 

 science principally constitute its value. The lowest generalisations, until 

 explained by and resolved into the middle principles of which they are the 

 consequences, have only the imperfect accuracy of empirical laws; while 

 the most general laws are too general, and include too few circumstances, 

 to give sufficient indication of what happens in individual cases where the 

 circumstances are almost always immensely numerous." (Mill, Logic, bk. 6, 

 en. o, 5.) 



