490 J. H. MUIRHEAD : first and most abstract of all the predicates by means of which the mind seeks to realise its ideal. The attempt to mark off a region of thinghood in the ob- ject which is unmediated by the subject naturally leads to the attempt to mark off an element of selfhood in the subject which is unmediated by the object, and we need not be surprised that Mr. McTaggart seeks to illustrate his abstract thing by the conception of an abstract ego. Facing the thing as the unity of its attitudes we have the subject as the unity of its perceptions. So far we are on well-known ground. Kant taught as much. But on Kant's view object and subject were both ultimately things in themselves and as- such unknowable. This new Kantianism makes a dis- tinction between them. The object as such is withdrawn from immediate presentation : it is only known from with- out, but the subject as known from within enters apparently immediately into consciousness. One can hardly believe that, Mr. McTaggart is really serious with this distinction, or means to assert that there is any knowledge of the self accessible to us which is not a knowledge of the world, any opaqueness in our knowledge of the w^orld which is not re- flected in our knowledge of the self. Yet abstractions die hard, and it may be worth while to restate the view upon this head, on which we are all, I take it, agreed, " except when we are supporting a thesis". We are all, I suppose, agreed as psychologists that the attempt to discover in the changing scene of feelings and cognitions a permanent identical content corresponding to the self is waste of time. The consequence of this admission for philosophy is not that there is no self (any more than the consequence of the astronomer's discovery that God was not visible through his telescope was that there was no God) but that it is to be looked for in another way. It is to be sought for at the end not at the beginning of our mental life,, in the extent and organisation of the contents of the mind, not in some needle's point of abstract consciousness. To know ourselves, therefore, is not to have access to some inner shrine of individual life but to understand the mode in which those contents are united to one another. Our guarantee for the unity of our own life is not any immediate consciousness of it but simply the fact that organised know- ledge exists. We may say, if we like, that the unity of the self is an idea or hypothesis by means of which we render the fact of knowledge intelligible to ourselves. But it would be truer to say that it represents one of the elements which
