SECTION 23. GENERALISATION. 329 



166. (c) COMPREHENSIVE GENERALISATIONS. The 

 desire to understand the world turns men away from particular 

 facts, for they feel that the number of these is so multitudinous 

 that it is humanly impossible to collect, remember, or understand 

 them. Unfortunately, men have not advanced a short distance 

 beyond and insisted that trifling generalisations are scarcely 

 distinguishable from particular facts, and that a vast array of 

 unconnected generalisations and relations tends to confuse rather 

 than to enlighten. Countless modern experiments, and especially 

 observations connected with the mental and social sciences, 

 illustrate this inference. For the future, therefore, all narrow 

 generalisations which are not intended to fit into a larger 

 scheme already formulated, should be esteemed as trivial and 

 suspect, unless it is quite impracticable to reach extended 

 generalisations. Probably the greatness of great men has mainly 

 consisted in doing what the average of investigators would also 

 have effected if it were generally recognised that the purpose 

 of a scientific enquiry is to reach sweeping generalisations. 1 

 Not only, therefore, are graded generalisations necessary, but 

 the constant endeavour should be to attain to the broadest 

 possible generalisations, and this has become increasingly 

 possible through the accumulation of scientific facts and con- 

 clusions. If this be conceded, it follows that the present practice, 

 of individuals engaging in a large number of relatively restricted 

 and unconnected investigations requiring no wide outlook, 

 should be replaced, now that science is sufficiently advanced 

 to permit this, by their undertaking, as a rule, one or two 

 investigations covering extensive ground and occupying them 

 for practically a life-time. 2 



If it were merely a question of piecing together a great 

 number of petty generalisations into an imposing mosaic, no 



1 "Then, and then only, may we hope well of the sciences, when in a 

 just scale of ascent, and by successive steps not interrupted or broken, we 

 rise from particulars to lesser axioms; and then to middle axioms, one above 

 the other; and last of all to the most general." (Bacon, Novum Organum, 

 bk. 1, 104.) It is not improbable that training would transform all investi- 

 gators into expert generalises. 



2 "A. Comte signalait deja les inconvenients de la division du travail dans 

 1'ordre intellectuel. La specialisation a outrance dans le domaine scientifique 

 fait perdre au savant 1'habitude de la generalisation; en se cantonnant dans 

 sa petite sphere, il est prive de tout contact avec les autres spheres qui lui 

 deviennent de plus en plus etrangeres. Sans doute il augmentera plus 

 facilement, dans sa sphere, le nombre des connaissances precises. Mais la 

 science ne progresse pas seulement par 1'accumulation des petites decouvertes 

 eparses; c'est aussi et surtout a la coordination, a la synthese de toutes 

 ces verites isole'es qu'il importe d'apporter ses efforts; la science risque 

 de s'emietter en une multitude de petites sp^cialites si elle manque d'archi- 

 tectes, de tetes encyclopediques, d'esprits synthetiques pour rassembler, pour 

 coordonner les connaissances innombrables que les ouvriers de la pensee 

 ont accumulees de. toutes parts." (Paul Caullet, op. cit., p. 299.) The contrast 

 between the architects and the labourers seems a little strained, since in 

 science either the architects are simultaneously labourers and the labourers 

 architects or their work is of questionable value. 



