104 PART II. SOME IMPORTANT METHODOLOGICAL TERMS. 



in its general outlines, and that his educators should respect 

 this tendency in the child, and even be ruled thereby. This 

 doubly dubious analogy could have been easily ascertained to 

 be spurious by noting that the children of primitive folk pass 

 through the same phases as those of Western people, and that 

 the life of primitive and semi-civilised men is intrinsically that 

 of adults and not of children. Educators, moreover, are prone 

 to surmise this is the educator's fallacy par excellence that 

 whatever children are taught in a certain place and in a certain 

 connection, they will appropriately generalise to all suitable 

 places and suitable connections, erroneously ascribing a power 

 to the immature which the ideally trained thinker would envy. 

 And, of course, what is so largely true of the teaching profession, 

 is in all probability equally true of most other professions and 

 callings. 



Scholars, however, not infrequently indulge in the opposite 

 tendency of fixing limits. Bacon told us that "it is impossible 

 that air should ever be consistent, or put off its fluidity". 

 (Novum Organum, bk. 2, 33.) Comte confidently declared that a 

 sidereal chemistry is a chimera, and this on the eve of decisive 

 discoveries in this most fascinating domain of science. Bain 

 argues : "All assertions as to the ultimate structure of the par- 

 ticles of matter are, and ever must be, hypothetical. . . . That 

 heat consists of the motions of the atoms can never be directly 

 shown." (Logic, vol. 2, p. 132.) And very common is the assump- 

 tion that the stage of civilisation reached to-day will not be 

 appreciably excelled in the future, ignoring that since mankind 

 has made inconceivably great advances in the past, it is likely 

 to make inconceivably great advances in the future, advances 

 suggestive of a world as much ahead of our own day as ours 

 is ahead of early paleolithic times. Only a general training in 

 scientific method can save us from the two extremes, and place 

 us in a position to generalise warily whilst rejecting all hypo- 

 thetical limitations. 



We have reasoned throughout this Section as if in the process 

 of generalisation we commenced invariably with observing parti- 

 cular facts, and then generalised our observations. The actual 

 process of thought, however, is often far from being so free 

 from complications. As everybody is aware, an enquiry is 

 seldom wholly novel, and even beyond this lies the fact that 

 we start as adults with a colossal army of more or less con- 

 fused notions and generalisations at our disposal. From this 

 it is to be inferred that perhaps more frequently than not we 

 are scrutinising a series of facts which, to our knowledge, has 

 been previously examined and generalised by others. Con- 

 formably, we find, as a rule, generalisations to hand, and our 

 concern is not seldom to correct, remould, or replace them. 

 Indeed, the steady historic advance in reliable information 

 implies that we are mainly modifying and extending, rather 



