I can neither doubt nor dispute that sensation is, in part, a psychological phenomenon, since I have admitted, by the very definition I have given of it, that sensation implies consciousness. We must, therefore, acknowledge those who define it as _a state of consciousness_ to be right, but it would be more correct to call it the _consciousness of a state_, and it is with regard to the nature of this state that the question presents itself. It is only this state which we will now take into consideration. It is understood that sensation contains both an impression and a cognition. Let us leave till later the study of the act of cognition, and deal with the impression. Is this impression now of a physical or a mental nature? Both the two opposing opinions have been upheld. In this there is nothing astonishing, for in metaphysics one finds the expression of every possible opinion. But a large, an immense majority of philosophers has declared in favour of the psychological nature of the impression. Without even making the above distinction between the impression and the act of cognition, it has been admitted that the entire sensation, taken _en bloc_, is a psychological phenomenon, a modification of our consciousness and a peculiar state of our minds. Descartes has even employed this very explicit formula: "The objects we perceive are within our understanding." It is curious to see how little trouble authors take to demonstrate this opinion; they declare it to be self-evident, which is a convenient way of avoiding all proof. John Stuart Mill has no hesitation in affirming that: "The mind, in perceiving external objects, can only take notice of its own conditions." And Renouvier expresses the same arbitrary assertion with greater obscurity when he writes: "The monad is constituted by this relation: the connection of the subject with the object within the subject."[15] In other words, it is laid down as an uncontrovertible principle that "the mental can only enter into direct relations with the mental." That is what may be called "the principle of Idealism." This principle seems to me very disputable, and it is to me an astonishing thing that the most resolute of sceptics--Hume, for example--should have accepted it without hesitation. I shall first enunciate my personal opinion, then make known another which only differs from mine by a difference of words, and finally I will discuss a third opinion, which seems to me radically wrong.