SECTION 20. STUDIES PREPARATORY TO ALL INVESTIGATIONS. 161 



"The second part of these manuals is more provoking. This 

 part is supposed to represent modernity and science, as the 

 first part obviously reflects antiquity and verbalism. Broadly 

 speaking, we have here a more or less close reflection of John 

 Stuart Mill's treatment of induction. Proceeding practically on 

 the assumption that discovery is due to innate and unana- 

 lysable capacities possessed only by the favoured few, the 

 contention of the authors generally is that inductive logic is 

 only concerned with proof. Even so, however, its object, it is 

 said, is not to impart practical knowledge enabling the student 

 to 'prove all things', but to comprehend the principles on 

 which the scholar of acknowledged scientific eminence proceeds. 



"From the viewpoint of the man of science the inductive 

 logics exhibit a painful misapprehension of the nature of science. 

 Whilst investigators move in a world where practical certainty 

 is infrequent and theoretical certainty is almost altogether 

 absent, our logicians discuss, e.g., the philosophy of induction 

 and the precise meaning of the term hypothesis, from a purely 

 speculative and perfectionist standpoint. The great matter, if 

 one may indulge in a sweeping statement, is words, words, 

 words, and the means of illuminating the words is by words, 

 words, words, and the total result is still words, words, words. 



"However, the man of science may not be competent to 

 appraise at their true value these discussions aiming at theo- 

 retical certainty. What, then, of the considered tests relating 

 to legitimate induction as they appear in the text-books? In 

 this matter one is astonished to note that whilst science has 

 during the last eighty years progressed by leaps and bounds, 

 Mill's rather incomplete analysis of the scientific process has 

 become more and more attenuated, till almost only its bare 

 shadow remains in the more recent books. His five Canons 

 are dutifully quoted and a few words are said in explanation ; 

 but the pages devoted to observation, generalisation, and deduc- 

 tion, have dwindled to a diminutive rump. 



"Of course, our logicians may retort that the object of their 

 inductive logic is not practical. The difficulty, however, is to 

 discover what purpose, in that case, these manuals are supposed 

 to serve. Are we to assume that the few rules presented on 

 the various aspects of methodological procedure are the only 

 rules, or the most important rules, to be abstracted from the 

 best practice of men of science? They are certainly neither 

 the one nor the other; in fact, one's heart sinks when one 

 meditates that not one of the very able writers on the methods 

 of science appears to have made an actual study of the scienti- 

 fic process in their own labours, or even in those of men of 

 science. Had they done so, they would have much- restricted 

 the discussion of terms, only given a page or two to the prob- 

 lem of theoretical certainty, greatly extended the rules, and, 

 with each decade, exhibited an increasing superiority over 



11 



