empirical concepts, but that is in accordance with the nature of the human soul, which has never hitherto rested content with experience alone. Everything new in the human mind proceeds from speculation. Mental development proceeds by way of speculation, never by way of limitation to mere experience. I realise that my views are parallel with those of Bergson, and that in my book the concept of the libido which I have given, is a concept parallel to that of “élan vital”; my constructive method corresponds to Bergson’s “intuitive method.” I, however, confine myself to the psychological side and to practical work. When I first read Bergson a year and a half ago I discovered to my great pleasure everything which I had worked out practically, but expressed by him in consummate language and in a wonderfully clear philosophic style.
Working speculatively with psychological material there is a risk of being sacrificed to the general misunderstanding which bestows the value of an objective theory upon the line of psychological evolution thus elaborated. So many people feel themselves in this way at pains to find grounds whether such a theory is correct or not. Those who are particularly brilliant even discover that the fundamental concepts can be traced back to Heraclitus or some one even earlier. Let me confide to these knowing folk that the fundamental ideas employed in the constructive method stretch back even beyond any historical philosophy, viz. to the dynamic “views” of primitive peoples. If the result of the constructive method were scientific theory, it would go very ill with it, for then it would be a falling back to the deepest superstition. But since the constructive method results in something far removed from scientific theory the great antiquity of the basic concepts therein must speak in favour of its extreme correctness. Not until the constructive method has presented us with much practical experience can we come to the construction of a scientific theory, a theory of the psychological lines of development. But we must first of all content ourselves with confirming these lines individually.
