abstracted aspects of this life. It is by this abstracting, aided by analogies derived from the structure of the brain and nervous system, that psychology has drawn its pictures of complicated mental mechanisms-the mechanism of sense-perception, of memory, of emotion, and so on.

In spite of the abstractness of such psychology, and in spite of the objection that will be noted in the next paragraph, there is no likelihood that we shall ever dispense with this method of approach to mental life. These aspects of our experience are actual aspects, and these mechanisms, though they are the psychologist's mental constructs out of elements that are themselves constructs, have uses both theoretical and practical that correspond to the parallel constructs of physics and chemistry. If, however, anyone speaking in the name of psychology should suggest that mental mechanism is all there is to mental life, he would be convincing only to those whose analysis stops short of the primary empirical data.

The status of "states of consciousness" is in fact openly challenged in the name of psychology itself. The behaviorist movement, which represents the second of the three possible courses, says substantially this: Let us observe and experiment upon the movements of our own bodies and of animal bodies, making the least possible reference to accompanying consciousness. By noting outer acts we shall arrive at the most secure generalizations concerning the very life that traditional psychology has attempted to construe by treating it as a subjective phenomenon. Behaviorism in its extreme form declares that the assumption of

consciousness has never helped in the solution of any problem.1

Undoubtedly this movement arises out of a real need, and its influence upon psychology is almost certain to be wholesome. It is well that we are thus challenged to exhibit the actual data with which psychology works, and to see how much can be learned from bodily movements and physiological changes as such. But, while behaviorism is likely to be a permanent point of view, particularly in the study of animals, it is not likely to crowd consciousness out of psychology. How far-to take a prominent problem of behaviorism—can analysis of the learning process go without taking account of satisfactions and annoyances ?2 And, in general, has ?2 not some behavior meaning? From one point of view conversation, for example, is just behavior, that is, a set of co-ordinated movements of lips, tongue, vocal cords, diaphragm and intercostal muscles, facial muscles, eyes, hands, etc. Analysis of these movements will very likely help us to understand what happens when two men converse. But to ignore everything in conversation except such movements is to leave out the function of it, which is the interchange of meanings between persons. For his purposes the behaviorist may avoid discussing this function, but he should at least realize that thereby he chooses one among several points of view. Behaviorism, in short, represents simply a new division of labor within the field of psychology.

'J. B. Watson, "Psychology as the Behaviorist Views It," Psychological Review, XX (1913), 158-77.

2 E. L. Thorndike, one of the leading behaviorists, though he consistently endeavors to express our reactions as far as possible in terms of muscular and neural activity, makes much of "satisfyingness" as a factor in the formation of new connections.

Both of these types of psychological interest, thenthe "states of consciousness" type and the behaviorist type-necessitate a third type. The concrete experience out of which we abstract "states of consciousness" is the experience of being a personal self. Each sensation, feeling, or other "element" of structural psychology is simply a particular discriminable aspect of a self-realizing life. Sensations and feelings are not known to have any other kind of existence, and what other kind of existence they could conceivably have has never been explained.1 Now, since self-realizations are not less actual than sensations, but more so, and since much of our behavior is communication of self-realized meanings, we must have an empirical science of self-realizations, or, in short, of selves. This is psychology par excellence, because its data are the most concrete and the most distinctive.2

Several recent developments show how inevitable it is that sooner or later we should advance from a

'When we desire to represent the consciousness of lower animals, we invariably image to ourselves some fragment of our own self-realizing life. We are helped by our memories of dreams, and of the vague states between waking and sleeping. Our own memory gaps also help us to conceive lower degrees of organization than our own. Further, the instinctive and other automatic factors in our own life make it clear that adaptive response could be abundant even if there were little consciousness in the animal making the response. We have no experience, however, that enables us to construe mere atoms of a consciousness that is not in any degree an organized, self-realizing consciousness. -Behaviorism avoids the difficulty here involved by thinking of animals as if they had no sensations, or pleasures and pains.

2 Since 1900 Professor Calkins has contended that since states of consciousness are "facts-for-selves," psychology must be a science of selves as well as of states. See her article, "Psychology as Science of Selves," Philosophical Review, IX (1900), 490-501. How such psychology differs from the old "psychology with a [metaphysical] soul" will appear as we proceed.

psychology of states to a psychology of individual persons or selves.

1. Various branches of the science are obliged to take as their unit the self-realizing individual life. Abnormal psychology, for example, would be practically meaningless if it contemplated states of consciousness as such instead of individuals who vary from an assumed norm of self-realization. Similarly, child psychology is required to get the child's point of view, that is, to view experience from the standpoint of child selves. Folk psychology is in a parallel situation. Social psychology, too, turns attention to the self-realized relations of individual to individual.

2. Many influences-philosophical, theological, psychological-are focusing attention upon values as an aspect of experience. A "value" is anything experienced or thought of as satisfying, or the contrary. Here we are in the sphere of interests, preference, individual attitude, self-realization.

3. The attempt to relate psychology to biology has caused us to think of mental process as a part of active adjustment to the conditions of life. Here mental process comes to be thought of as mental function, which is mental action directed toward advantage, the furtherance of life-in particular, life that realizes its own improved state.

The standpoint of "function" emerges, in fact, in each of these fresh psychological growths. Here a functional psychology, or a psychology that recognizes the functional standpoint, is being created. Here belong the problems of the second type that appeared in our first chapter. Religious experiences have a mechanism,

to be sure, but they are occupied about ends or valueswhat Tagore calls "the realization of life." This is increasingly the case as we move upward in the scale of religions. At the summit of culture the character of each religion consists in its working conception of life's values, and the religious status of the individual is judged by his scrutiny, choice, and pursuit of ends. Accordingly, the psychology of religion may be expected to be predominantly functional. Therefore the idea of mental function needs to be carefully examined at the outset. We shall see that it is far less simple than one might suppose.2

What do we mean by "function"? As we use the term here it applies to living beings only. It signifies the part that any organ or process has in maintaining, reproducing, or improving the life of an individual or of

the

group to which an individual belongs. The function of teeth, for example, is to tear, cut, crush, and grind food, so that the digestive juices may reach all parts of it, so that it may be assimilated and built into living cells, so that the individual or group life may go on in strength.

I

1 Rabindranath Tagore, Sādhanā: The Realization of Life (New York, 1913).

2 The chief critical discussion of the relation of functional to structural psychology is J. R. Angell's "The Relations of Structural and Functional Psychology to Philosophy," one of the Decennial Publications of the University of Chicago, 1903, printed also in Philosophical Review, XII (1903), 243-71. The article has abundant footnote references. See also G. H. Mead, "The Definition of the Psychical," Decennial Publications of the University of Chicago, 1903. The first work on general psychology to be written systematically from the functional point of view was, I believe, J. R. Angell's Psychology, the first edition of which appeared in 1904. The clearest, most systematic discussion of the functional standpoint in the psychology of religion is chap. ii of E. S. Ames, The Psychology of Religious Experience (Boston, 1910).