My argument for Libertarianism as against Determinism will not then seek to show any more, in the last resort, than that freedom is a 'necessary' illusion. Even if the conviction of freedom be proved to be an ineradicable element in human experience, it must still be admitted to be in a sense 'illusion' (since, to give only one reason, the concept of freedom with which we work carries an inherent reference to an objective world so characterised that its ultimate reality cannot be accepted). But on the other hand I shall try to show that Determinism is an illusion not merely 'in the last resort' or 'metaphysically,' but also from the standpoint of human experience. It is Determinism, not Libertarianism, that is the 'mere' illusion, the illusion corrigible by fuller thinking. That, at least, is the case which we shall endeavour to make good in the sequel.

And the second remark is this. It is not all loss that one should be debarred by one's metaphysical doctrine from attempting a 'metaphysical' justification of freedom. On the contrary it may be urged - and the argument seems to me unanswerable - that for those who believe that the Whole is intelligible, and that therefore an ultimate or metaphysical justification is obligatory, no justification of freedom is or can be possible: or, at any rate no justification of the kind of freedom which is a postulate of the moral life. The burden of the 'coherent Whole' lies like a dead-weight upon all endeavours after a metaphysical justification of freedom. When 'the shouting and the tumult die' the simple question is put,

'Could any man, on your theory, have acted otherwise than he in fact did'? And to this question only one answer is possible on the doctrine of an intelligible Whole, an answer which to the plain man means the straight denial of freedom. Only where the oppressive 'sense of the Whole' is absent can it be said that the problem of freedom has life and meaning. For a Rationalist metaphysic the solution - an adverse one - is strictly speaking pre-determined.

This brings us to a topic, however, upon which somewhat ampler discussion is necessary. I hold not only that the Idealist doctrine of freedom is false as a statement of fact (resting as it does upon a false metaphysic), but also that it is utterly irrelevant if offered as a contribution to the problem of freedom. The freedom which Idealism defends is not the freedom about which the plain man is concerned, and (if this should seem to the reader more important) is not the freedom which constitutes the traditional problem of philosophy. That problem arose in the interest of moral responsibility. It is the freedom necessary to vindicate moral responsibility that has been for so many centuries a main battle-ground of the philosophers and the theologians. And this freedom the Idealist doctrine not merely fails to defend, but, by the very character of the freedom for which it argues, must be held by implication to repudiate.

For what is the freedom that is required to validate moral responsibility? There is surely no great mystery about it, if only we are prepared to approach with open minds. Limiting our considerations, for simplicity's sake, to responsibility for wrong-doing, it is quite evident that the passing of moral censure upon a person, in implying the belief that the person ought to have acted otherwise than he in fact did, implies also the belief that the person could have acted otherwise than he in fact did. It is possible, indeed, that the 'judger' may have cause to believe that the person has so habituated himself to the form of conduct in question that he is now virtually incapable of acting otherwise. But if he does so believe, he yet only passes censure upon the person because he assumes that such a state of affairs has been brought about by past acts of will which 'could have been otherwise'.

This is not a matter, however, upon which elaborate argument is really in place. Other conditions there are, of course, implied in the conception of moral responsibility - conditions, e.g. such as Bradley has set forth in the first chapter of Ethical Studies. But that the condition just stated is also indispensable (although not included in Bradley's list) it is only possible to deny if one is also prepared to deny that the moral 'ought' has meaning without the assumption of a 'can.' Blame and praise are alike utterly pointless if we believe that the person could not have acted otherwise than he did. In other words, the conception of moral responsibility loses all significance unless it can be maintained that there is in rational beings a real capacity for alternative action.1

Now the freedom for which Idealism contends is certainly not a freedom which involves a capacity for alternative action. It is not even neutral with respect to it, for it is inextricably bound up with a metaphysic for which any such capacity is unthinkable. Ought it not then to be acknowledged frankly that Idealism's 'freedom,' instead of vindicating moral responsibility (as we not unnaturally expect of a philosophy which professes to champion 'freedom'), is definitely inconsistent with it? It would at least minister to clarity of discussion if Idealists would emulate the candour of Spinoza, and avow openly that on their view moral praise and blame should be replaced by mere congratulation and condolence.

1 I am aware, of course, that a number of attempts have been made, in recent ethical literature especially, to justify the union of moral responsibility with Determinism. But I must frankly confess that the arguments advanced appear to me to be so obviously of the nature of special pleading that to traverse them in detail is mere waste of time and labour. One can understand, and in a measure sympathise with, the attitude of mind in which these attempts originate. As a rule they are due to the writer's natural reluctance to admit that moral responsibility is a mere fiction, working along with his firm conviction that a freedom of 'open possibilities 'has long been proved to belong to the realm of mere mythology. But however keenly one may desire to dissever moral responsibility from a freedom of this suspect type, their indissoluble connectedness really seems to be one of the few perspicuously plain truths of human experience. To be morally responsible means, I suppose we can all agree, to be the legitimate object of moral praise and blame. But to impute praise or blame to a person if you know that that person could not possibly have done otherwise (whether in his present act or in those of his previous acts which condition the present act), is a proceeding which so flagrantly violates the fundamental nature of our moral consciousness that not even the authority of many of its champions should prevent us from calling it nonsensical. The common practice of moral judgment in mankind generally, wherever that judgment is critical and considered and not, as so often, thoughtless and all but mechanical, is flatly inconsistent with it. One could, of course, select examples from the hasty and semi-automatic 'moral judgments 'passed in everyday experience which might seem to lend to it a certain plausibility. But from this field one could select examples which would support almost any doctrine. Reflective moral judgment, the judgment that is deliberately concerned to judge justly, is the only kind of judgment which really counts in this matter.