A theory has often been maintained with regard to the consciousness; namely, that it supposes a relation between two terms--a subject and an object, and that it consists exactly in the feeling of this relation. By subject is understood the something that has consciousness; the object is the something of which we are conscious. Every thought, we are told, implies subject and object, the representer and the represented, the _sentiens_ and the _sensum_--the one active, the other passive, the active acting on the passive, the _ego_ opposed to the _non ego_. This opinion is almost legitimised by current language. When speaking of our states of consciousness, we generally say, "I am conscious; it is I who have consciousness," and we attribute to our I, to our Ego, to our personality, the role of subject. But this is not a peremptory argument in favour of the above opinion; it is only a presumption, and, closely examined, this presumption seems very weak. Hitherto, when analysing the part of mind, we have employed non-committal terms: we have said that sensation implied consciousness, and not that sensation implied something which is conscious.[23] The difference may appear too subtle, but it is not; it consists in taking from consciousness the notion of a subject being conscious and replacing it by the very act of consciousness. My description applies very exactly, I think, to the facts. When we are engaged in a sensation, or when we perceive something, a phenomenon occurs which simply consists in having consciousness of a thing. If to this we add the idea of the subject, which has consciousness, we distort the event. At the very moment when it is taking place, it is not so complicated; we complicate it by adding to it the work of reflection. It is reflection which constructs the notion of the subject, and it is this which afterwards introduces this construction into the states of consciousness; in this way the state of consciousness, by receiving this notion of subject, acquires a character of duality it did not previously possess. There are, in short, two separate acts of consciousness, and one is made the subject of the other. "Primitively," says Rabier, "there is neither representative nor represented; there are sensations, representations, facts of consciousness, and that is all. Nothing is more exact, in my opinion, than this view of Condillac's:--that primitively, the inanimate statue is entirely the sensation that it feels. To itself it is all odour and all savour; it is nothing more, and this sensation includes no duality for the consciousness. It is of an absolute simplicity." Two arguments may be advanced in favour of this opinion. The first is one of logic. We have divided all knowledge into two groups--objects of cognition, and acts of cognition. What is the subject of cognition? Does it form a new group? By no means; it forms part of the first group, of the object group; for it is something to be known.