It should be remarked, however, that he tried, at least, to insist that we know the external world directly.  We may divide realists into two broad classes, those who hold to this view, and those who maintain that we know it only indirectly and through our ideas.

The plain man belongs, of course, to the first class, if it is just to speak of a man who says inconsistent things as being wholly in any one class.  Certainly he is willing to assert that the ground upon which he stands and the staff in his hand are perceived by him directly.

But we are compelled to recognize that there are subdivisions in this first class of realists.  Reid tried to place himself beside the plain man and failed to do so.  Hamilton (section 50) tried also, and he is not to be classed precisely either with the plain man or with Reid.  He informs us that the object as it appears to us is a composite something to the building up of which the knowing mind contributes its share, the medium through which the object is perceived its share, and the object in itself its share.  He suggests, by way of illustration, that the external object may contribute one third.  This seems to make, at least, something external directly known.  But, on the other hand, he maintains that the mind knows immediately only what is in immediate contact with the bodily organ—­with the eyes, with the hands, etc.; and he believes it knows this immediately because it is actually present in all parts of the body.  And, further, in distinguishing as he does between existence “as it is in itself” and existence “as it is revealed to us,” and in shutting us up to the latter, he seems to rob us even of the modicum of externality that he has granted us.

I have already mentioned Herbert Spencer (section 50) as a man not without sympathy for the attempt to rehabilitate the external world.  He is very severe with the “insanities” of idealism.  He is not willing even to take the first step toward it.

He writes:[1] “The postulate with which metaphysical reasoning sets out is that we are primarily conscious only of our sensations—­that we certainly know we have these, and that if there be anything beyond these serving as cause for them, it can be known only by inference from them.

“I shall give much surprise to the metaphysical reader if I call in question this postulate; and the surprise will rise into astonishment if I distinctly deny it.  Yet I must do this.  Limiting the proposition to those epiperipheral feelings produced in us by external objects (for these are alone in question), I see no alternative but to affirm that the thing primarily known is not that a sensation has been experienced, but that there exists an outer object.”