Now it is obvious that the relation of causality, understood in this second way, does not involve the necessary determination of the effect by the cause. History indeed proves it. We see ancient hylozoism, the first outcome of this conception of causality, explained the regular succession of causes and effects by a real _deus ex machina_: sometimes it was a Necessity external to things and hovering over them, sometimes an inner Reason acting by rules somewhat similar to those which govern our own conduct. Nor do the perceptions of Leibniz's monad necessitate one another; God has to regulate their order in advance. In fact, Leibniz's determinism does not spring from his conception of the monad, but from the fact that he builds up the universe with monads only. Having denied all mechanical influence of substances on one another, he had to explain how it happens that their states correspond. Hence a determinism which arises from the necessity of positing a pre-established harmony, and not at all from the dynamic conception of the relation of causality. But let us leave history aside. Consciousness itself testifies that the abstract idea of force is that of indeterminate effort, that of an effort which has not yet issued in an act and in which the act is still only at the stage of an idea. In other words, the dynamic conception of the causal relation ascribes to things a duration absolutely like our own, whatever may be the nature of this duration; to picture in this way the relation of cause to effect is to assume that the future is not more closely bound up with the present in the external world than it is in our own inner life. [Sidenote: Each of these contradictory interpretations of causality and duration by itself safeguards freedom; taken together they destroy it.]