These considerations bring us to the province of psycho-analysis. There is much in the detail of psycho-analysis which I find fantastic, and not supported by adequate evidence. But the general method appears to me very important, and essential to the creation of right methods of moral training. The importance which many psycho-analysts attach to early infancy appears to me exaggerated; they sometimes talk as if character were irrevocably fixed by the time a child is three years old. This, I am sure, is not the case. But the fault is a fault on the right side. Infant psychology was neglected in the past; indeed, the intellectualist methods in vogue made it almost impossible. Take such a matter as sleep. All mothers wish their children to sleep, because it is both healthy and convenient when they do. They had developed a certain technique: rocking the cradle and singing lullabys. It was left for males who investigated the matter scientifically to discover that this technique is ideally wrong, for though it is likely to succeed on any given day, it creates bad habits. Every child loves to be made a fuss of, because its sense of self-importance is gratified. If it finds that by not sleeping it secures attention, it will soon learn to adopt this method. The result is equally damaging to health and character. The great thing here is the formation of habit: the association of the cot with sleep. If this association has been adequately produced, the child will not lie awake unless it is ill or in pain. But the production of the association requires a certain amount of discipline; it is not to be achieved by mere indulgence, since that causes pleasurable associations with lying awake. Similar considerations apply to the formation of other good and bad habits. This whole study is still in its infancy, but its importance is already very great, and almost sure to become greater. It is clear that education of character must begin at birth, and requires a reversal of much of the practice of nurses and ignorant mothers. It is also clear that definite instruction can begin earlier than was formerly thought, because it can be made pleasant and no strain upon the infant’s powers of attention. In both these respects educational theory has been radically transformed in recent years, with beneficent effects which are likely to become more and more evident as the years go by. Accordingly I shall begin, in what follows, with a fairly detailed consideration of the training of character in infancy, before discussing the instruction to be given in later years. CHAPTER II THE AIMS OF EDUCATION