320 



THE POPULAR SCIENCE MONTHLY.— SUPPLEMENT. 



instruments of science, these propositions must 

 have corresponding importance. Perhaps, how- 

 ever, we shall derive some light from the con- 

 text ; reading on a few lines in the description of 

 the Method of Difference, 1 we find Mill saying 

 that— 



" In the Method of Agreement we endeavored 

 to obtain instances which agreed in the given cir- 

 cumstance but differed in every other: in the 

 present method (i. e., the Method of Difference) 

 we require, on the contrary, two instances resem- 

 bling one another in every other respect, but dif- 

 fering in the presence or absence of the phenome- 

 non we wish to study." 



It would really seem, then, as if the great Ex- 

 perimental Method depends upon our discovering 

 two instances resembling one another. Here re- 

 semblance is specified by name. We seem to 

 learn clearly that Agreement must be the same 

 thing a3 Resemblance ; if so, Difference must be 

 its opposite. Proceeding accordingly to con- 

 sider the Method of Difference, we find its re- 

 quirements described in these words : 2 " The 

 two instances which are to be compared with one 

 another must be exactly similar, in all circum- 

 stances except the one which we are attempting 

 to investigate." 



This exact similarity is not actual identity, of 

 course, because the instances are two, not one. 

 Is it, then, resemblance ? If so, we again find 

 the principal subject of Mill's Logic to be that 

 which he relegated to section 2 of chapter xxiv. 

 If we proceed with our reading of Mill's chapter 

 on the " Four Experimental Methods," we still 

 find sentence after sentence dealing with this re- 

 lation of resemblance, sometimes under the very 

 same name, sometimes under the names of simi- 

 larity, agreement, likeness, etc. As to its appar- 

 ent opposite, difference, it seems to be the theme 

 of the whole chapter. The Method of Difference 

 is that wonderful method which can prove the 

 most general law on the ground of two instances ! 

 But of this peculiarity of the Method of Differ- 

 ence I shall treat on another occasion. 



Perhaps, however, after all, I may be misrep- 

 resenting Mill's statements. It crosses my mind 

 that by Resemblance he may mean something 

 different from exact similarity. The Methods of 

 Agreement and Difference may require that com- 

 plete likeness which we should call identity of 

 quality. It is only fair to inquire, then, whether 

 he uses the word Resemblance in a broad or a 



1 Same chapter, second section. 

 = Same chapter, third section, third paragraph, 

 fourth line. 



narrow sense. On this point Mill leaves us in no 

 doubt ; for he says distinctly, 1 " This resemblance 

 may exist in all conceivable gradations, from per- 

 fect undistinguishableness to something extremely 

 slight." 



Again, on the next page, while distinguishing 

 carefully between such different things as nu- 

 merical identity and indistinguishable resem- 

 blance, he clearly countenances the wide use of 

 the word resemblance, saying, 2 " Resemblance, 

 when it exists in the highest degree of all, amount- 

 ins; to undistinEfuishableness, is often called iden- 

 tity." It seems, then, that all grades of likeness 

 or similarity, from indistinguishable identity down 

 to something extremely slight, are properly com- 

 prehended under resemblance ; and it is difficult 

 to come to any other conclusion than that the 

 agreement and similarity and difference treated 

 throughout the Experimental Methods are all cases 

 of that minor relation, seldom considered the 

 subject of science, which was postponed by Mill 

 to the second section of the twenty-fourth chap- 

 ter. 



But the fact is, that I have only been playing 

 with this matter. I ought to have quoted at 

 once a passage which was in my mind all the 

 time — one from the chapter on the Functions and 

 Value of the Syllogism. Mill sums up the con- 

 clusion of a long discussion in the following 

 words : 3 



" We have thus obtained what we were seek- 

 ing, a universal type of the reasoning process. 

 We find it resolvable in all cases into the following 

 elements : Certain individuals have a given attri- 

 bute ; an individual or individuals resemble the 

 former in certain other attributes ; therefore they 

 resemble them also in the given attribute." 



All reasoning, then, is resolvable into a case 

 of resemblance ; the word resemble is itself used 

 twice over, and, as I shall hereafter show, the 

 word attribute, synonymous with properly, is but 

 another name, according to Mill, for resemblance. 

 It is true that this quotation is taken from the 

 second book of the System, not from the preced- 

 ing part of the third book to which Mill referred 

 as not having treated of resemblance. But this 

 can hardly matter, as he speaks of the universal 

 type of the reasoning process, which must include, 

 of course, the whole of the inductive methods 

 expounded in the third book. 



But, in case the reader should not be quite 

 satisfied, I will give yet one more quotation, taken 



i Book I., chapter iii., section 11, ParaGraph 4. 



s Same section, fifth paragraph, third line. 



s Book II., chapter iii., section 7. at beginning. 



