We appear to have involved ourselves in paradox. On the one hand the presentation of external objects occurs only through those experiences which vary with the motor activity of the subject: on the other hand, the presentation of external objects occurs only through experiences which are independent of the motor activity of the subject. This seems a plain contradiction. But the inconsistency disappears when we consider that the kinds of experience which vary with our own movements may also vary in partial or complete independence of these movements. Thus the contradiction is removed, as many apparent contradictions are; the contradiction is present when we assert that A is at once B and notB; it disappears when we say that A is partly B and partly notB. The change in visual sensation, which occurs when we approach an object, may also occur when it approaches us; only in the second case does it constitute presentation of movement on the part of the object. In this simple instance the contrast may without serious inaccuracy be regarded as a contrast between change which is exclusively due to the animal's own activity and change in other respects similar which occurs in the absence of such activity. But for the most part the antithesis is of a subtler kind. The two factors cooperate so that the resulting change in part depends, and in part does not depend, on the varying position and movements of the organism. A given movement may always give rise to altered impressions from surrounding things. But the specific nature of the change is not entirely determined by the specific nature of the movements. On the contrary, the same movement yields varying results under varying circumstances. Opening of the eyes permits access to the light, but it does not determine the special nature of the optical stimulation received. Movement of the eye in a certain direction produces a sequence of optical impressions, but it does not determine what the impressions shall be which succeed each other, or in what order they shall occur. Similarly, the initiation of movement or of effort to move depends on the animal; but various and fluctuating external conditions determine whether a movement in a given direction shall be free or impeded, and if it is impeded what kind and degree of resistance it shall encounter. If the same motor activity always produced the same effect, there would be no such thing as adaptation to environment: the phrase would be meaningless. We may imagine the case of an animal able to command all the external conditions affecting its nervous system purely by its own initiative, so as to obtain any impression on any occasion merely by a suitable innervation of its muscles. Specific olfactory, optical, tactile, and other stimulations of the organs of sense would then be freely producible in the same manner as the specific impressions arising from the changing states of muscles, joints, and tendons, which accompany movements of the body and limbs. A certain mode of sniffing would always yield the smell of roses, and a certain movement of the eye would always yield the sight of roses, and so forth. Similarly, we may suppose that want of food or drink could be satisfied by merely going through the formal motions of eating and drinking, as at the banquet of the Barmecide in the Arabian Nights. Evidently, for such a creature as we have imagined, the external world would be virtually nonexistent. Such a creature would be a world complete in itself.
Our general result is as follows: (1) The presentation of external objects takes place through those experiences to which the subject must adapt itself if its action is to be efficient for the attainment of practical ends. (2) These experiences which correspond to external conditions and make possible practical adjustment must be of a kind which vary concomitantly with the movements of the animal. (3) But they must only do so in part. They yield effective guidance only in so far as they actually occur partially in independence of the movements of the animal. In other words, an external world exists for the animal only in so far as the same movement may give rise to different consequences, or different movements to the same consequences. We have now to apply this general principle to a special case of paramount importance. In general, the action which subserves the primary ends of animal life is effective only if and so far as it consists in or prepares the way for what we may call the direct manipulation of objects. It is difficult to find a better word; but it should be clearly understood that when we say manipulation we imply no exclusive reference to the hand. What is meant is all alteration or endeavour to alter the position, shape, arrangement, etc., of things, by direct putting forth of effort against resistance. All pulling, pressing, rending, tearing, combining, separating, breaking, bending, crushing, moulding, and the like, are included under this conception. It is obvious that in all such operations what lies in the power of the agent is only to make efforts in certain directions and in a certain order. The result varies with the nature of the material manipulated. To be effective the course of action must be constantly guided by varying experiences corresponding to the varying nature of the material. Now the paramount practical importance of the actual manipulation of objects constitutes it the ultimate and dominant test of what is physically real and what is not. The real size or shape of a thing is its size or shape so far as it has a practical bearing on actual manipulation. A man at a distance may look as small as a doll on the table; but he is not really so small, because if I went out and tried to pick him up and bring him away, I should not succeed. So the real size of a hole to an animal is essentially determined by reference to such questions as whether it can creep into the hole or not. This point is simple and obvious when once stated, but it is of the greatest importance for the whole psychology of perception.
We may now sum up. Perception of physical reality always arises in essential connexion with the experience of active movement. But the connexion is one of antithesis or contrast. Experience of active movement is not, as such, a presentation of external object. Only so far as the motor activity is limited or circumscribed in the attainment of the ends of animal life by varying conditions, is physical reality apprehended. The experiences which determine the adjustment of active movement to these conditions are, as such, presentations of the Not-Self, or external object. In the actual course of practical activity attention in the form of watching, searching, scrutinising, and the like, must be directed predominantly, if not wholly, to the external object. Only those experiences which determine adjustment have an objective reference. All else is ignored by attention. The psychological correlate of interorganic disturbance of the nervous system consists in animal appetites, emotions, instinctive impulses, and the like. These are merely changing states of the Self. In a sense they may be called passive, inasmuch as they arise and persist independently of the effects of motor activity. But in another and more important sense they are essentially active, inasmuch as they constitute the primary impulses or tendencies in which motor activity has its source. They do not of themselves contain any special contents of consciousness distinctive of external objects. In the actual course of practical activity they do not by their specific quality enable attention to dis criminate the conditions to which active movement must adjust itself. On the other hand, as being the source of activity, they constitute the interest which keeps attention alive and at work.
The case is different in the intervals of practical activity. Here attention not being preoccupied by external object may be directed to the Self and its states. The hungry lion deprived of the opportunity of satisfying its hunger, may attend to the hunger itself, instead of to its prey, and the like. It is very important, however, to note that in all primitive perception of the Self, it is not what is called the pure Self which is presented, but always the embodied Self. In other words, the apprehension of Self is always bound up with the apprehension of a particular external object. For the bodily organism has a twofold nature. On the one hand it behaves just as other external objects do. One hand can perceive and explore the other hand, or the eyes can perceive and explore other parts of the body, just as eye or hand can perceive or explore external bodies. If we select one sentient portion of the organism, the rest may be regarded as external object to it. So far the body is an external thing. On the other hand, the body as a vehicle of active movement and of senseperception, and the body in its peculiar and intimate relation to organic sensations, appetites, etc., belongs to the Self. Thus an animal's perception of its own organism constitutes a connecting link between the perception of Self and Not-Self.