Privacy or uniqueness is no more obstructive to understanding than is change, and, like change, has no peculiar applicability to mental states as matter of knowledge. Privacy or uniqueness applies to physical objects of knowledge in essentially the same way as it applies to mental states. Mere accessibility is, in principle, common for all objects of knowledge, to all subjects.[149] But there is a special reason why the subject of the state is particularly disqualified, as compared with others, for knowing his state immediately, _i. e._ intuitively; namely, that, at the time of the existence of the state, when, alone, it could be known intuitively, he is mainly occupied with another object of knowledge, the object of the state in question. You do not, then, know a mental state best by living it, or rather _in_ living it; your knowledge of it is just then at its worst, since you are then preoccupied in knowing something else. The state, as an attribute of the subject, is clearly one of the subject’s relations, and, so, conceptually distinct from either term. It cannot be at once a knowledge and the object of that same knowledge. Bergson’s treatment of the conscious state conceives it in just that way--as if the relation were itself one of its own terms, the object. Knowing a mental state can only mean understanding it. It is not a concrete datum, like the sky, but an abstraction from the relationship in which the subject and the sky function as terms. One does not intuitively know the subjective process of blueness, in looking at the sky; one knows the sky in that sense, but the process only conceptually, by reflection. Is it any less an authentic object of knowledge? Is it not itself--is it any symbol of itself?--which you name and define and talk about and understand? The practical significance of saying that one felt and now remembers a feeling is not that the feeling is what one ever felt. Feeling Number One is not an object for feeling Number Two, neither during Number One nor afterward, in reminiscent feeling. So far as the reminiscent state is another intuition, its object is the same as that of the intuition remembered--so far. But to be reminiscent, a conscious state must reflect upon, or refer to, a conscious state distinct from itself. This reflective reference is a conceptual co-element together with the intuitional character of the reminiscent state. So far as the memory is reflective, consciousness is oriented toward the original state itself as a fact, a process, conceptually distinguishable from the object of it. It is thus only _so far as conceptual_ that subjective processes can be objects of knowledge, or, in short, be known. But if so, Bergson is wrong in two essential points: in denying that subjectivity can be objectified, and in affirming that knowledge of subjectivity is immediate (_i. e._ non-conceptual) or intuitive.