63. "If I wish," he says, "to know by experience the numerical identity of any external object, I apply my attention to that which is constant in the phenomenon, to which all the rest is referred, as a determination to its subject; and I observe the identity of the subject at the time in which the determination changes. I am an object of the internal sense, and time is only the form of this sense; I therefore refer all my successive determinations, and each one of them in particular, to that which is numerically identical, in all time, that is, in the form of the internal intuition of myself. Hence the personality of the soul ought only to be deduced or concluded as a proposition perfectly identical with consciousness in time; consequently, this proposition is valid _a priori_, because it does not really announce any thing else than that in all the time in which I am conscious of myself, I am conscious of this time as a thing, which is a part of my unity. This is the same as to say: All this time is in me as individual unity, or rather, I am in all this time with numerical identity." It would have been desirable if Kant had shown why the internal sense of the numerical identity may be expressed by the proposition; all this time is in me as an individual unity, or in this other; in all the time in which I am conscious of myself, I am conscious of this time as a thing, which is a part of my unity. It is true that the numerical unity is perceived in the diversity of time; but it is not true that we are conscious of time as a thing which is a part of us. He is treating of the consciousness of self, as it is found in the greatest part of mankind, who, far from considering time as a thing which is a part of themselves, regard it as a sort of vague extension or succession in which they and all that is variable exist.