118 



SCIENCE 



[N. S. Vol. XXXII. No. 812 



" reasonableness," and the two elements of 

 reasonableness are identity-difference and con- 

 trol, validity and truth. " Every implication 

 is a subject-matter identical with itself, dif- 

 ferent from or exclusive of any other and, 

 taken together with its contradictory, ex- 

 haustive of the sphere of control in which they 

 are both found" (p. 283). There is no a 

 priori law or form of identity above and be- 

 yond the individuation of the object of 

 thought as such. Identity is of two sorts, 

 inner and outer, internal and external. The 

 latter is resolvable into recurrence and " sec- 

 ondary conversion " ; as to the former or inner 

 identity " we wait neither for recurrence nor 

 do we ask our neighbors. We find in the 

 immediately persisting and continuous mental 

 life the experience that enables us to call the 

 self identical" (p. 289). The present re- 

 viewer does not believe this to be true or 

 psychological. Self-identity in the subject of 

 experience is no more an immediate intuition 

 than the identity of the solar system is and we 

 arrive at the consciousness of the one by pre- 

 cisely the same dialectic steps as the other. 

 Perhaps Baldwin here means, however, that 

 self-identity seems to the hnower to come by 

 immediate intuition, but even this statement 

 seems to us to be a case of reading one's own 

 prejudices into the experience of other people. 



Class identity and singular identity difPer 

 only as a group of different objects with the 

 same meaning diifer from a group of recur- 

 rences of the same object with the same mean- 

 ing, and from the standpoint of community 

 these are the same thing. 



Identity in diiierence is prelogical; it be- 

 comes identity and difference when taken up 

 into the logical mode by judgment. This 

 process is called induction and issues in classi- 

 fication, ordination and definition. Hence the 

 two guiding threads of induction, says Bald- 

 win, are agreement and difference, giving rise 

 to the two fundamental methods recognized 

 by Herschel and Mill. The other methods of 

 Mill are variations on these two. Baldwin's 

 discussion of induction is weak for the follow- 

 ing reasons: Mill's methods are methods of 

 discovering sequences, and not coexistences — 

 an oversight that has characterized inductive 



logic ever since and one that characterizes 

 this discussion. Again, they are methods of 

 elimination and presuppose that nature is 

 composed of manifold kinds and causal agents 

 which have already been discovered and classi- 

 fied. These methods fail to describe the actual 

 method of scientific procedure, because they 

 assume that the objective content of judgment 

 and the original data with which the judgment 

 starts are one and the same thing. 



Mill's methods are largely deductive rather 

 than inductive, and the method by which Mill 

 arrived at his canons of induction is de- 

 ductive. They are deduced from an a priori 

 conception of the objects of nature to be 

 investigated. It will be evident, I think, that 

 the method of induction outlined by Baldwin 

 (pp. 30-4^7) is also deductive. 



The resulting judgment is one of implica- 

 tion rather than one of proposal or schematism. 

 The two great elements of validity are " uni- 

 versality of the necessary type " and the rela- 

 tion of dependence or ground. Such judg- 

 ments are reached by establishing the syn- 

 nomic force of judgments and the exhaustion 

 of the class-meaning by limitation (p. 312). 

 The former gives judgment its universality — 

 its necessity — and the latter, its rational char- 

 acter as logical ground. The treatment of 

 deduction is not only brief, it makes no at- 

 tempt to show the intimate connection between 

 deduction and experimentation in scientific 

 procedure. 



Chapters thirteen, fourteen and fifteen (part 

 four of the book) are devoted to The Dualisms 

 and Limitations of Thought. They are, for 

 the most part, a discussion of pragmatism. 

 They treat memory and thought as related 

 alike to reality; each can be acted on because 

 it is correct ; it is false to say they are correct 

 because they can be acted on. The criteria of 

 correctness are conversion (or social control) 

 and external control. Calling this view the 

 theory of knowledge through control, Baldwin 

 names the view of Dewey the theory of " con- 

 trol through luiowledge." The latter is "the 

 ' control ' of the Studies in Logical Theory 

 and other works of the Chicago school so- 

 called." " It is control of a personal sort " 

 (p. 349). Notwithstanding Baldwin's ex- 



