It must be freely admitted that there is a sort of circle here from which it seems impossible to escape.  In the order of efficient causes we assume ourselves free, in order that in the order of ends we may conceive ourselves as subject to moral laws:  and we afterwards conceive ourselves as subject to these laws, bjecause we have attributed to ourselves freedom of will:  for freedom and self-legislation of will are both autonomy, and therefore are reciprocal conceptions, and for this very reason one must not be used to explain the other or give the reason of it, but at most only for logical purposes to reduce apparently different notions of the same object to one single concept (as we reduce different fractions of the same value to the lowest terms).

One resource retrains to us, namely, to inquire whether we do not occupy different points of view when by means of freedom we think ourselves as causes efficient a priori, and when we form our conception of ourselves from our actions as effects which we see before our eyes.

It is a remark which needs no subtle reflection to make, but which we may assume that even the commonest understanding can make, although it be after its fashion by an obscure discernment of judgment which it calls feeling, that all the “ideas” [Footnote:  The common understanding being here spoken of, I use the word “idea” in its popular sense.] that comes to us involuntarily (as those of the senses) do not enable us to know objects otherwise than as they affect us; so that what they may be in themselves remains unknown to us, and consequently that as regards “ideas” of this kind even with the closest attention and clearness that the understanding can apply to them, we can by them only attain to the knowledge of appearances, never to that of things in themselves.  As soon as this distinction has once been made (perhaps merely in consequence of the difference observed between the ideas given us from without, and in which we are passive, and those that we produce simply from ourselves, and in which we show our own activity), then it follows of itself that we must admit and assume behind the appearance something else that is not an appearance, namely, the things in themselves; although we must admit that as they can never be known to us except as they affect us, we can come no nearer to them, nor can we ever know what they are in themselves.  This must furnish a distinction, however crude, between a world of sense and the world of understanding, of which the former may be different according to the difference of the sensuous impressions in—­various observers, while the second which is its basis always remains the same.  Even as to himself, a man cannot pretend to know what he is in himself from the knowledge he has by internal sensation.  For as he does not as it were create himself, and does not come by the conception of himself a priori but empirically, it naturally follows that he can obtain his knowledge