400 CKITICAL NOTICES: question would only be effected ' by a change which should render some branch of experience formally impossible, i.e., inaccessible to our methods of cognition ' (p. 60). Similarly (p. 179) he asserts : ' The ground of necessity, we may safely say, arises from the mind '. Now that which is ' inaccessible to our-methods of cognition ' would seem only to mean that which we cannot, know ; it cannot imply that the judgments in question cannot be true. But apart from a proof that they cannot be true (in which case of course they cannot be known in the sense in which knowledge is distinguished from belief), it would seem that the only possible way of showing that they cannot be known must be- simply a psychological inquiry into the conditions that are necessary for the production of actual beliefs. The same Kantian fallacy is betrayed again very clearly on page 135, where Mr. Eussell asserts that ' what is purely intellectual cannot change '. Here- by ' what is purely intellectual ' must be meant either what is- distinguished by certain psychological marks, or what is already known to be presupposed in experience. But in the former case the fact that it cannot change can be only known empirically,, whereas in the latter ' intellectual ' simply means what is a priori, as presupposed in experience, and therefore cannot be used k> confirm this distinction. Mr. Eussell's professed deduction of his a priori axioms from a. necessary ' form of externality ' is therefore necessarily futile ; since by the necessity of the form of externality can only be meant that it is presupposed in actual experience, which he haa already shown his axioms to be. But there are two points in his actual argument for the necessity of such a form which seem to call for particular notice, as illustrating the psychological implica- tions which must be introduced in order to give a meaning to such a deduction. Mr. Eussell recognises (p. 184 foil.) that it is essential for his purpose to contend that, if knowledge is to be possible, there must be a diversity ' not only of conceptual content but of existence ' ; since a form of externality means a form of relation between several existents, and not merely between diverse contents. He has therefore to argue that the mere diversity of content,, which we may admit with him to be implied in any judgment, is itself impossible unless there be also a diversity of existents. His argument to prove this point seems to be purely pyschologicaL For he assumes as a premiss that ' knowledge must start from perception ' (p. 184). These words inevitably suggest that he is- concerned only with the genesis of knowledge a point of view which he has himself outruled as irrelevant to the discussion in hand. Unless he does mean this it would seem he must deny that any kind of cognition except perception is possible ; since otherwise what is necessary for perception would not be neces- sary for that other kind. Now it is hardly possible he would refuse to distinguish between e.g. a mathematical cognition and
