which bear vitally upon the question of Freedom and which tend to obscure the issue. Of these, the foremost is that once we have acted in a particular manner we look back upon our actions and try to explain them with particular reference to their immediate antecedents. Here is where the mischief which gives rise to the whole controversy has its origin. We make static what is essentially dynamic in character. We call a process a thing. There is no such "thing" as Freedom; it is a relation between the self and its action. Indeed, it is only characteristic of a self IN ACTION, and so is really indefinable. Viewed after the action, it presents a different aspect; it has then become historical, an event in the past, and so we try to explain it as being caused by former events or conditions. This casting of it on to a fixed, rigid plan, gives action the appearance of having characteristics related to space rather than to time, in the real sense. As already shown in the previous chapter, this is due entirely to our intellectual habit of thinking in terms of space, by mathematical time, rather than in terms of living time or la duree. Another point which causes serious confusion in the controversy is the notion that because, when an act has been performed, its antecedents may be reckoned up and their value and relative importance or influence assigned, this is equivalent to saying the actor could not have acted in any other way than he did, and, further, that his final act could have been foretold from the events which led up to it. It is a fact that in the realm of physical science we can foretell the future with accuracy. The astronomer predicts the precise moment and place in which Halley's comet will become visible from our earth. It is also a fact that we say of men and women who are our intimate friends: "I knew he (or she) would do such and such a thing" or "It's just like him." We base our judgment on our intimate acquaintance with the character of our friend, but this, as Bergson points out, "is not so much to predict the future conduct of our friend as to pass a judgment on his present character--that is to say, on his past." [Footnote: Time and Free Will, p. 184 (Fr. p. 140).] For, although our feelings and our ideas are constantly changing, yet we feel warranted in regarding our friend's character as stable, as reliable. But, as Mill remarked in his Logic: "There can be no science of human nature," because, although we trust in the reliability of our friend, although we have faith in his future actions, we do not, and can not, know them. "Tout comprendre c'est tout pardonner." To say that, if we knew all the conditions, motives, fears, and temptations which led up to the actions of another, we could foretell what he would do, amounts to saying that, to do so, we should have actually to become that other person, and so arrive at the point where we act as he did because we are him.