100 PART II. SOME IMPORTANT METHODOLOGICAL TERMS. 



given instance, will always hold true of similar instances in 

 the future. An examination of the nature of the General will 

 demonstrate the untenability of such a position, and it will also 

 indicate that often mere intensity or repetition of an experience, 

 rather than reasoning, accounts for a conclusion. 



Let us consider an experience. An acquaintance is ascribing 

 to me to-day mistaken, though flattering, motives concerning 

 a certain action of -mine. This had also happened yesterday, 

 without my then launching any generalisation. To-day I say 

 to him that both yesterday and to-day the ascription to me of 

 the particular motives was presumably due to his assuming on 

 those occasions that he himself would have been prompted by 

 such motives in those circumstances. Here I reasoned from 

 one particular to a second particular. I now begin systemati- 

 cally to extend. My acquaintance, I think, (1) sometimes, (2) fre- 

 quently, (3) generally, (4) practically always, (5) invariably, attri- 

 butes to me certain motives, because he would, in such a situa- 

 tion, be himself actuated by such motives. I further extend 

 this to (6) some, (7) many, (8) very many, (9) most, (10) all 

 people. I continue to extend this to (11) some, (12) many, 

 (13) very many, (14) most, (15) all ideas and sentiments which 

 men possess. I venture* on the broad generalisation that (16) 

 people presume others to be and do what they are and do 

 themselves. And, finally, I vaguely surmise that (17) in the 

 Universe like assumes like. 



39. The ambiguity involved in the conception of the 

 generalising process would be removed if we were to employ 

 the term Extension instead of the term Generalisation, for ex- 

 tension naturally suggests an indefinite number of stages, 

 whereas generalisation tends to direct attention to one stage 

 only. For instance, scrutinising the nature of the sensations, 

 I assume that special and general memory (vide 19) would 

 be required in viewing the inrushing tide (which 1 am now 

 watching) a second, third, fourth, or /zth time; I then reason 

 to similar objects seen, to sight generally, to sound, and to all 

 the senses. Thus, again, formulating a charter of liberty which 

 refers to individuals, I extend the charter methodically to politi- 

 cal and other groups, to humanity as a whole, to the inner 

 life, as well as to art, to all other human activities, and, con- 

 ditionally, to the whole animal creation. Or noticing that 

 3 2 1 is divisible by 2, 4 2 1 by 3, and so on, I generalise the 

 formula to n~ 1, and further to n m 1 divisible by n 1. (Sig- 

 wart, Logic, vol. 2, p. 212.) Clearly, the first generalisation in 

 any of these cases did not psychologically compel the last, and 

 historic progress is registered in extending generalisations to 

 undreamt-of realms, and to spheres which had been at first 

 expressly excluded from a generalisation. Prof. Creighton justly 

 remarks: "A conception, or mode of regarding things, which 

 has proved serviceable in one field is almost certain to dominate 



