JOHN STUART MILL'S PHILOSOPHY TESTED. 



85 



The second point to be considered is the fright- 

 ful loss and disappointment you prepare for your 

 child if, as is most probable in these days, he be- 

 comes convinced that the doctrine is founded on 

 insufficient evidence. It is not merely that you 

 have brought him up as a prince, to find himself 

 a pauper at eighteen. He may have allowed this 

 doctrine to get inextricably intertwined with his 

 feelings of right and wrong. Then the overthrow 

 of one will, at least for a time, endanger the other. 



You leave him the sad task of gathering together 

 the wrecks of a life broken by disappointment, 

 and wondering whether honor itself is left to him 

 among them. Leave him free of this doctrine, 

 and his conscience will rest upon its true base, 

 safe against all storms : for it is built upon a rock. 

 Then he can never reproach you with raising hopes 

 in him which knowledge is fated to blast, and 

 with them, it may be, to blast the promise of his 

 life. — Nineteenth Century. 



JOHN STUART MILL'S PHILOSOPHY TESTED. 



By W. STANLEY JEVONS. 



III. — The Experimental Methods. 



MY last article on Mill's Philosophy treated 

 of what ought to be, or rather necessarily 

 is, the basis of all reasoning processes — the Re- 

 lation of Resemblance. It was shown that Mill 

 first of all dismisses this relation as a minor, or 

 even a doubtful matter of fact, or as " still another 

 exceptional case ; " that he then unintentionally 

 makes it the pivot, nay, almost the substance of 

 the reasoning process, as treated in the book on 

 Induction ; yet that, in a later chapter of that 

 book, he returns to the subject of Resemblance 

 as if it had so far been passed over, and finally 

 comes to the conclusion that Resemblance is sel- 

 dom regarded as the subject of science. 



From the base let us proceed to the pillars of 

 Mill's logical edifice. These are the celebrated 

 Methods of Experimental Inquiry — the Method 

 of Agreement, the Method of Difference, the Meth- 

 od of Residues, and the Method of Concomitant 

 Variations ; to which may be added, as a kind of 

 corollary, the Joint Method of Agreement and 

 Difference. Mill's exposition of these methods is 

 considered perhaps the most valuable part of his 

 treatise, and much of the celebrity of the book 

 is due to this part. Many people, indeed, whose 

 reading in logic has not been extensive, think 

 that these are Mill's own methods, that he in- 

 vented them. Any one at all acquainted with the 

 history of logical science knows, of course, that 

 this is not the case ; nor did Mill ever claim that 

 it was. Francis Bacon set forth the methods, 

 excepting perhaps that of Residues, in the second 

 book of the "Novum Organum," vaguely, no 



doubt, but with substantial correctness. Taking 

 the nature of heat to exemplify the mode of in- 

 vestigation, he firstly enumerated " Instances 

 agreeing in the Nature of Heat," nearly, if not 

 exactly, corresponding to the Method of Agree- 

 ment. Next came " Instances in Proximity want- 

 ing the Nature of Heat," by means of which the 

 Method of Difference, or the Joint Method of 

 Agreement and Difference, was brought into play. 

 The Table of Degree or Comparison in Heat 

 forms a rude application of the Method of Con- 

 comitant Variations. 



Sir John Herschel, again, described these 

 methods with great clearness, and in a manner 

 which I have always preferred to that of Mill. 

 Three of the methods are stated on pages 151 

 and 152 of his admirable " Discourse on the 

 Study of Natural Philosophy," and the " Method 

 of Residues" is given on page 156. Mill has 

 amply acknowledged his indebtedness to Herschel 

 in several places and ways, and there is not the 

 slightest fault to find with him in that respect. 

 The question is whether Mill, in adopting and 

 formulating the methods anew, and incorporating 

 them into his supposed system of logic, has done 

 better than his predecessors. I shall proceed to 

 show that this is not the case ; on the contrary, 

 he has misinterpreted both the foundation and 

 the results of these methods. On some other 

 occasion I shall have to point out that, in treat- 

 ing them, he has positively confused together an 

 experiment, which is a material operation, with 

 the generalization by which we pass from the re- 



