It has become the orthodox view of physicists to-day, that the momentary state of a physical system does not determine its movement or development or behaviour, to follow; Nature is supposed to be such that a knowledge of state, sufficiently accurate for sharp prediction of the future, is not only unobtainable but also unthinkable. All that can be predicted refers to a large number of identical experiments, and consists in a definite statistics among all the possible developments to follow. The relative margin of indeterminacy (the 'spread' of the statistics) is large for a small system, for example, for an atom; but for large systems the margin is usually, though not necessarily, small, which makes it possible to account for the apparent determinacy of inanimate Nature.

When observed objectively in other creatures, free-will actions do not call for a special 'indeterminist' explanation any more than other events. When two persons (or the same person on different occasions) react differently under apparently the same conditions, we feel compelled to account for it, whether the reaction is a passive or an active one, by a real, though unknown, difference of conditions, including, of course, character and temporary disposition on the part of the reacting persons. A poet unrolling before us the objective picture of free-will actions is just as concerned about proper causation (here called motivation) as the classical physicist was for inanimate Nature.

On the other hand, when regarded as a fact of self-observation, free-will has quite a different standing from scientific experience. The two are, as it were, in different planes, which do not intersect. Self-observed free-will I would analyse into two facts. First, indeed, a prediction, but not based on previous experience, certainly not in the way in which scientific prediction is. If I am the actor, I just know what is going to happen, and that, apart from pathological cases, with the greatest amount of certainty which is ever met with in life. The second fact is a moral one. I feel responsible for what happens.

Now, it is true that this absolute prescience is a matter only of the very last moment before or when the action sets in, which it rather accompanies than precedes. Before that there is frequently doubt and even entire ignorance ('hesitation'). This antecedent period, together with the remarkable feeling of responsibility, entails the idea of choice between different possibilities for which a clue is sought in the modern views of physics. If that were right, it would mean either one of two things.

In my opinion the whole analogy is fallacious, because the plurality of possible events, in the case of an action under free-will, is a self-deception. Think of cases such as the following: you are sitting at it formal dinner, with important persons, terribly boring. Could you, all at once, jump on the table and trample down the glasses and dishes, just for fun? Perhaps you could: maybe you feel like it: at any rate you cannot. Then, which of the virtually possible events are to be called possible under the auspices of free-will? I would say, just the one that actually follows.

Against this view cases might be quoted where the decision is really difficult, serious, painful, bewildering, when we are down on our knees before the Almighty to forgo it. But in this He is inexorable!

The preceding remarks have been elicited by the first page of a highly interesting sketch by Prof. F. G. Donnan, "Integral Analysis and the Phenomena of Life" (Acta Biotheoretica, Series A, vol. 2, Pars I, 1936; Leyden: E. J. Brill), though not by way of contradiction. Prof. Donnan is not concerned with the question of free-will. His idea is that an organism is to be regarded as a 'historical' system, whose reactions at a given moment are not determined alone by its surroundings and by its momentary state, but also by what has happened to that organism during a certain previous period. This is a highly attractive view, and the mathematical treatment proposed by Prof. Donnan a very suggestive one - even if one should hesitate to agree with the view (which he considers essential) that some of the historical traces are not engraved in the momentary state otherwise than by modifying its reactivity.