870 



SCIENCE 



[N. S. Vol. XXXVII. No. 962 



Perry's damaging concession. Pitkin, for ex- 

 ample, declares that " the realist can not 

 count his case won " until he has shown " the 

 complete independence of all things thought 

 of." He feels obliged, therefore, to hold that 

 even hallucinatory objects, and errors of all 

 sorts, are " no less independent of cognition 

 than true propositions are." That secondary 

 qualities, illusory presentations, and the like, 

 can all be, without contradiction, conceived 

 as objective and independent, Montague, Holt 

 and Pitkin alike are concerned to show; and 

 the only imaginable reason for their under- 

 taking to justify this paradox is an accept- 

 ance of the view that consciousness is in tio 

 case constitutive of any perceived content, 

 that it is always and absolutely a relation 

 which does not create its terms nor modify 

 their other relations. 



But for this more rigorous construction of 

 the relational theory, what evidence is offered ? 

 It, if established, would prove neo-realism's 

 case, as I have admitted; but by what argu- 

 ment is this all-important premise itself to be 

 established? Of general and positive argu- 

 ments for it there are, so far as I can see, none 

 in the book. Direct empirical evidence is, 

 once more, unattainable, in the nature of the 

 case, and is not attempted. What we are 

 given is merely a series of attempts to show 

 that the theory is not absurd, that the general 

 and unqualified assertion that no perceived 

 datum ever does or can depend upon its rela- 

 tion to a perceiver for its existence or any of 

 its attributes or its other relations, is not the 

 extravagance which it at first appears to be. 

 Even if these attempts be regarded as suc- 

 cessful, they could not, by a rigorous logician, 

 be regarded as establishing the conclusion de- 

 sired. There are many propositions which 

 are not absurd which are also not true. The 

 battle for the relational view of consciousness 

 can hardly be won by purely defensive tactics. 

 But are even those tactics successful? To 

 this question we must now turn. 



2. Science, I have said, as well as popular 

 thought, is still, as a rule, accustomed to 

 think of some of the content of experience as 

 existing " merely subjectively." The whole 



distinction between appearance and reality — 

 in the ordinary, empirical sense — which sci- 

 ence has found so indispensable has usually 

 taken the form of the supposition that certain 

 data of perception, — e. g., the secondary quali- 

 ties of matter, illusions, dreams, hallucina- 

 tions — -can be explained away as having their 

 being only in " minds " and by virtue of 

 minds, as being only in so far as, and only in 

 the sense that, somebody is conscious of them. 

 This, then, is the way of thinking and of 

 speaking which the new realism (in so far as 

 it treats its relational theory of consciousness 

 as a universal proposition) invites us to give 

 up. It therefore proposes a radical revision 

 of widely current preconceptions. The im- 

 portant question to raise concerning it, then, 

 is this : Can we, while maintaining realism, 

 completely dispense with the idea of " subjec- 

 tive appearance," of " mental representations " 

 of objects, can we hold without self-contradic- 

 tion that what things seem they also are, and 

 that the entity present " in consciousness " 

 whenever we perceive or think of an object is 

 just the original, simon-pure object itself, un- 

 transformed, unduplicated and untransposed ? 



That their doctrine stands or falls with the 

 answer to this question, the authors vei'y 

 frankly acknowledge. " The crucial problem," 

 says Pitkin, " for the new realism is the prob- 

 lem of error (in all forms). And the acutest 

 critics " of the doctrine " urge that its fatal 

 flaw is the acceptance of the full ' objective ' 

 nature of illusions and errors and its simul- 

 taneous refusal to put illusory objects, with 

 all their colors, shapes and behaviors, identi- 

 cally in the very space and time in which they 

 immediately belong. H the charge is true it 

 is deadly." To meeting this type of objection 

 the papers of Montague, Holt and Pitkin are 

 chiefly devoted. 



Unfortunately space is lacking here for an 

 adequate analysis of these highly ingenious 

 and rather involved pieces of argumentation; 

 that examination must be attempted else- 

 where. For the present it must suffice to observe 

 that these three vsT-iters are unable to agree 

 upon any one " solution of the problem of 

 error " in terms which shall be consistent with 



