sight is reconciled with liberty by being converted into sight. For when we see an act we feel none of that impulse to declare it coerced which we feel when we suppose ourselves to foresee it. The lower stage to which the difficulty is supposed to be brought down lies in the reception of the proposed mystery in the Being of God; and it is contended that, under certain circumstances, to accept certain mysteries is more reasonable than to reject them. This, of course, is stated by way of illustration and explanation, not by way of argument. And explanation cannot be thought superfluous when it appears that so distinguished a metaphysician as Mr. Mill is, or was, ignorant of any difference between divine and human foreknowledge, so far as concerns the bearing of foreknowledge on the freedom of the will.* pro As I do not propose to recur to this subject after once quitting it, I will here state the general grounds upon which, in common with many others, I get over this difficulty. They are these:-I hold the difficulty to be only one form, and not the most puzzling, of the insoluble problem of the co-existence of Infinite and Finite; that is to say, it is perly an argument against Theism, not against Freewill. And I shall frankly avow, with Dugald Stewart, that, if compelled to choose between the two, I should elect (quod abominandum) to deny the Prescience of God rather than Freewill. This, it will be noticed, is the first of the three possible courses laid down by Hamilton. If any one thinks it ridiculous under these circumstances to believe both in God and Freewill, he might spend a profitable hour in counting the names and examining the reputations of those who think otherwise. Here we might conclude ur account of this tedious business; but there is a particular argument which is worth noting for the sake of the picture which it affords of a controversy in the sere and yellow leaf. I mean that reproduction by Copleston of Cudworth's fallacy, which was referred to above. In the Preface to his Enquiry into the Doctrine of Necessity and Predestination, Copleston tells us that the argument of his treatise was suggested to him by some observations of Mr. Dawson of Sedbergh; and it is a curious illustration of the confusion pervading the subject, that the arguments of Dawson, which Copleston supposes to have the same scope as his own, should be levelled at Fatalism, while Copleston's are aimed at Determinism. This appears on the surface when they are examined critically. Dawson's reasoning, as reported by Copleston, *Logic, 5th ed., vol. ii. p. 414. Works, ed. by Hamilton, vol. vi. p. 398. runs to the following purpose. It is the tendency of the opinion of Necessity-that is, of Fatalism, for the statement is not true of Determinism-to discourage individual exertion, so far as it is believed to be true. Now suppose Fatalism to be true and known to be true; then all activity would come to an end; for it is the aim of activity to modify that universe which, by hypothesis, is incapable of modification. Hence the result of a perfect assurance of the truth would be an universal state of repose. But this is contrary to the analogy of nature, by which it appears that, in general, intelligence and activity increase together, and that intelligence increases with knowledge. Therefore the hypothesis of Fatalism, upon which the conclusion is founded, must be false. Such is Dawson's argument, when put into the strongest form of which it is capable. Copleston's arguments to prove that Fatalism (he, of course, calls it Necessity) tends, so far as it is believed, to discourage exertion, are cogent, but scarcely needed. After this preliminary we may sum up his main position in a single sentence of his own. "It appeared to me that the moral consequences of the hypothesis in question might also be pursued; for the notion of a moral agent gifted with mental powers, the improvement of which naturally tends to the weakening or the extinction of moral principle, is an absurdity."* Now it is quite plain what he would say. He means to say that, if we improve our understanding so far as to arrive at a knowledge, not of Fatalism but of Determinism, this will oblige us to acknowledge that the vulgar notion of moral desert is a fantastic notion; that is, in his language, it will destroy all moral principle in us; and that it is absurd to suppose the co-existence in us of intellectual and moral powers, only in order that the improvement of the former may be the destruction of the latter. Hence he infers that the hypothesis of Determinism, the doubtful premiss from which the conclusion springs, must be false; just as in Dawson's argument it was inferred that the premised Fatalism must be false. Here we have another emphatic witness, unconsciously borne by a man of the highest ability, to the tenacity of the vulgar notiɔn of moral desert. But in its formal statement, as an attempt to bolster up a theory of the will, the argument is a fallacy. pleston, like Cudworth, assumes the real point at issue in order to prove a pretended point. If the vulgar notion of moral desert be indisputably a real notion, that was quite * An Enquiry, etc., Preface, p. vii. Co enough for Copleston's purpose: he was not | stage, in which the attempt to decide will called upon to prove any theory of the will be made, is the last that remains; and it is at all. He could have but one motive for also the most arduous. constructing a theory of the will, namely, that he might be able to deduce from it the reality of the vulgar notion of moral desert. Therefore, to assume the reality of the vulgar notion in order to construct a theory of the will, was to beg the question in dispute. It yet remains that something shall be said about the Calvinistic doctrine of Predestination, because very erroneous opinions are current about its bearing upon the question before us. People are apt to conclude that, because Predestination is manifestly opposed to Freewill, therefore it must have some connection with Determinism. But this is a mistake. An enumeration of the points in which Predestination differs from Determinism will show how little they have in common; and my object in introducing Predestination is only to show that it never needed to be introduced. (1.) Doctrines of this kind draw much of their significance and colouring from the premisses from which they are deduced. But Determinism and Predestination spring from premisses which lie quite in separate regions of thought. We have seen what are the chief premisses of Determinism. The chief premisses of Predestination are found in what is accepted as a Divine revelation. (2.) The Predestinarian is obliged by his theology to admit the existence of a free will in God. And as a matter of fact, though this is perhaps unnecessary, he does admit the existence of a free will in the devil. Nor is the existence of a free will in man inconceivable to him, though it is denied as being incompatible with the exercise of a free will by God. Thus he ignores the destructive criticism passed by the Determinists on the definition of a free will put forward by the Libertarians. (3.) But the final consideration, which puts a great gulf between the Determinist and the Predestinarian, is this, that the latter asserts the reality of the vulgar notion of moral desert. Even if he were not obliged by his interpretation of Scripture to assert this, he would be obliged to assert it in order to help out his doctrine of Eternal Reprobation. Many eminent Determinists have, on the contrary, openly avowed their belief in the doctrine of Final Restitution, upon no other ground than their denial of the reality of the vulgar notion of moral desert. Here I bring to an end the second division of my undertaking, which is concerned only with the statement of arguments, not with the decision between them. The third If it were the only question, whether Freewill or Determinism shall ultimately prevail in the world, a Libertarian might rest secure in the vitality of his opinion. Nor would this security necessarily indicate any unworthy desire to triumph in the stupidity of the vulgar. There must be some reason for the existence of a fact; and, when Determinists prove that all reason is against the common belief in Freewill, their success has rather an ominous appearance. The later Determinists, particularly Mr. J. S. Mill, are well aware of their obligation to account for the fact of the vulgar belief upon their Own principles; and we shall presently examine the account given by them. But there is something else for which they still have to account. How does it happen that men of candour and metaphysical acumen are found to adhere to the old prejudice of Freewill, after fully examining all that has been said on the other side? Or will it be denied that any Libertarian is both acute and candid, besides being well informed! We shall see presently whether the arguments and explanations of the Determinists are such as to bear out this denial. The question now before us is not whether the Libertarian may feel pretty sure that he will always have the general voice of the world on his side, but whether Determinism has been proved true upon grounds of reason; and I do not mean to shirk this ques tion. I believe that fatal gaps can be pointed out, both in the argument for Determinism, and also in the explanation of the origin of Freewill. But if this much can be made out, that will be enough for my purpose; for Freewill is so evidently the doctrine in possession that to discredit the proof of Determinism is to establish the proof of Freewill. The feeling of the reality of the vulgar notion of moral desert is so strong and so universal that it is prima facie its own justification: the burden of disproof lies with those who impugn it. Hamilton was doubtless right in thinking that, if Freewill and Determinism could be put on the same level in point of antecedent inconceivableness, the latter must go to the wall. Coming now to an actual criticism of the arguments of the Determinist, I follow my three-fold division, and have therefore these three separate points to consider :-(1.) Since the strength of Freewill lies in the tenacity of the vulgar notion of moral desert, therefore the corresponding argument against Determinism lies in showing that it can give no satisfactory explanation of the origin of this notion. (2.) Since the strength of Determinism lies in its analysis of the facts of volition, therefore the corresponding argument in favour of Freewill lies in shewing (a) that this analysis is incomplete, (3) that the assumption of Freewill supplies what is needed to complete it. (3.) The third stage should be the consideration of the antagonism between Prescience and Liberty; but upon this head I shall say no more. If the other two points be sufficiently handled, this third may be left, in the present state of the controversy, to take care of itself. In dealing with the other two, I shall take first the question of the completeness of the Determinist analysis of the facts of volition, postponing to it the more difficult question which stands first, namely, the sufficiency of the Determinist account of the origin of the vulgar notion of moral desert. For the former question can be considered at once; but the latter will require some preliminary examination of the Doctrine of Inseparable Association. (I.) In considering the Determinist analysis of the facts of volition, it is to be observed that I have two separate points for which to contend :-(a) That the Determinist analysis is incomplete, which by itself would suffice for my purpose. (6) That the assumption of Freewill supplies what is needed to complete the analysis, which is more than sufficient; for, if it could be made out, it would amount to a proof positive of Freewill, whereas I have only pretended to offer a proof negative, drawn from the destruction of the proof positive for Determinism. (a) That the analysis is incomplete appears to me to follow from these considerations. If it were both complete and accurate, then we should be able, by means of a complete and accurate synthesis of its elements, to construct a true representation of the facts of volition, that is, of the will itself. But this is not the case. Whenever I contemplate the analysis and attempt to construct out of its elements a representation of the will, the result of my synthesis is not the will of which I am conscious to myself, but something else. This other something is not monstrous or impossible to be conceived; it is a kind of will which I can easily imagine to exist within me, but which in fact does not. Hence I infer, being convinced that my synthesis is both accurate and complete, that the analysis is either inaccurate or incomplete, or else both. But, as I do not impeach its accuracy so far as it goes, the conclusion lies against its completeness. These are the grounds on which I conclude that the Determinist analysis is incomplete. Of course there is an obvious objection to this. It might be suggested that I have blundered over, the synthesis. And it may be so. But I have taken great pains, both in gathering the materials and also in putting them together; and the result has been what I have stated. I admit that it is of the highest importance to inquire what other people think of my statement; and if it shall so turn out that what I have said is contradicted by the common voice, I shall be ready to believe that I have fallen into error. But the sequel will contain some further indirect reasons which strengthen my present belief that the synthesis is sound. Another objection is equally obvious. It might be asked why I do not directly point out the error in the analysis, instead of inferring it from the result of the synthesis. I might reply that, in a matter of this sort, with which language is ill fitted to deal, leave must be given to choose that course which is the more easily followed, rather than that which is the more direct. But I am able to make some approach to the direct indication required; and the previous indirect process of inference was placed first only as a matter of convenience in arrangement. Here is the direct indication. Hitherto I have been speaking only of my waking volition. But also when asleep I am often in some sense conscious of exercising volition of some sort; and I retain an image of these sleeping volitions in memory when I am awake. Now the waking notion which I have of my sleeping will, exactly resembles the result of my synthesis aforesaid. Therefore I infer that the Determinist analysis would be a complete account of my sleeping will, but that it is an incomplete account of my waking will. And to the demand that the error in the analysis shall be pointed out I reply that the error lies in omitting to take account of the difference between waking and sleeping volition-that the analysis is a complete account of the latter, but an incomplete account of the former. (3) As to the second point: it follows from what has been said that the difference between the real will of experience and the fictitious will of Determinism is the very thing needed to turn Determinism into Freewill, by making it compatible with the reality of the vulgar notion of moral desert. For the volitions of my sleeping will do not subject me to the consequences of the reality of the vulgar notion, while the volitions of my waking will do. On contemplating the two, I seem to be made clearly aware that the 62 enough for Copleston's purpose: he was not called upon to prove any theory of the will at all. He could have but one motive for constructing a theory of the will, namely, that he might be able to deduce from it the reality of the vulgar notion of moral desert. Therefore, to assume the reality of the vulgar notion in order to construct a theory of the will, was to beg the question in dispute. It yet remains that something shall be said about the Calvinistic doctrine of Predestination, because very erroneous opinions are current about its bearing upon the question before us. People are apt to conclude that, because Predestination is manifestly opposed to Freewill, therefore it must have some connection with Determinism. But this is a mistake. An enumeration of the points in which Predestination differs from Determinism will show how little they have in common; and my object in introducing Predestination is only to show that it never needed to be introduced. (1.) Doctrines of this kind draw much of their significance and colouring from the premisses from which they are deduced. But Determinism and Predestination spring from premisses which lie quite in separate regions of thought. We have seen what are the chief premisses of Determinism. The chief premisses of Predestination are found in what is accepted as a Divine revelation. (2.) The Predestinarian is obliged by his theology to admit the existence of a free will in God. And as a matter of fact, though this is perhaps unnecessary, he does admit the existence of a free will in the devil. Nor is the existence of a free will in man inconceivable to him, though it is denied as being incompatible with the exercise of a free will by God. Thus he ignores the destructive criticism passed by the Determinists on the definition of a free will put forward by the Libertarians. (3.) But the final consideration, which puts a great gulf between the Determinist and the Predestinarian, is this, that the latter asserts the reality of the vulgar notion of moral desert. Even if he were not obliged by his interpretation of Scripture to assert this, he would be obliged to assert it in order to help out his doctrine of Eternal Reprobation. terminists have, on the contrary, openly avowed their belief in the doctrine of Final Restitution, upon no other ground than their denial of the reality of the vulgar notion of moral desert. stage, in which the attempt to decide will be made, is the last that remains; and it is also the most arduous. If it were the only question, whether Freewill or Determinism shall ultimately prevail in the world, a Libertarian might rest secure in the vitality of his opinion. Nor would this security necessarily indicate any unworthy desire to triumph in the stupidity of the vulgar. There must be some reason for the existence of a fact; and, when De- The question now before us is not whether Hamilton was Coming now to an actual criticism of the arguments of the Determinist, I follow my three-fold division, and have therefore these three separate points to consider:-(1) Since the strength of Freewill lies in the te nacity of the vulgar notion of moral desert. Here I bring to an end the second division of my undertaking, which is concerned only with the statement of arguments, not therefore the corresponding argument against with the decision between them. The third Determinism lies in showing that it can give Of course there is an obvious objection to this. It might be suggested that I have blundered over, the synthesis. And it may be so. But I have taken great pains, both in gathering the materials and also in putting them together; and the result has been what I have stated. I admit that it is of the highest importance to inquire what other people think of my statement; and if it shall so turn out that what I have said is contradicted by the common voice, I shall be ready to believe that I have fallen into error. the sequel will contain some further indirect reasons which strengthen my present belief that the synthesis is sound. no satisfactory explanation of the origin of | are the grounds on which I conclude that the this notion. (2.) Since the strength of De- Determinist analysis is incomplete. terminism lies in its analysis of the facts of volition, therefore the corresponding argument in favour of Freewill lies in shewing (a) that this analysis is incomplete, (3) that the assumption of Freewill supplies what is needed to complete it. (3.) The third stage should be the consideration of the antagonism between Prescience and Liberty; but upon this head I shall say no more. If the other two points be sufficiently handled, this third may be left, in the present state of the controversy, to take care of itself. In deal ing with the other two, I shall take first the question of the completeness of the Determinist analysis of the facts of volition, postponing to it the more difficult question which stands first, namely, the sufficiency of the Determinist account of the origin of the vulgar notion of moral desert. For the former question can be considered at once; but the latter will require some preliminary examination of the Doctrine of Inseparable Association. (I.) In considering the Determinist analysis of the facts of volition, it is to be observed that I have two separate points for which to contend :-(a) That the Determinist analysis is incomplete, which by itself would suffice for my purpose. (3) That the assumption of Freewill supplies what is needed to complete the analysis, which is more than sufficient; for, if it could be made out, it would amount to a proof positive of Freewill, whereas I have only pretended to offer a proof negative, drawn from the destruction of the proof positive for Determinism. (a) That the analysis is incomplete appears to me to follow from these considerations. If it were both complete and accurate, then we should be able, by means of a complete and accurate synthesis of its elements, to construct a true representation of the facts of volition, that is, of the will itself. But this is not the case. Whenever I contemplate the analysis and attempt to construct out of its elements a representation of the will, the result of my synthesis is not the will of which I am conscious to myself, but something else. This other something is not monstrous or impossible to be conceived; it is a kind of will which I can easily imagine to exist within me, but which in fact does not. Hence I infer, being convinced that my synthesis is both accurate and complete, that the analysis is either inaccurate or incomplete, or else both. But, as I do not impeach its accuracy so far as it goes, the conclusion lies against its completeness. These But Another objection is equally obvious. It might be asked why I do not directly point out the error in the analysis, instead of inferring it from the result of the synthesis. I might reply that, in a matter of this sort, with which language is ill fitted to deal, leave must be given to choose that course which is the more easily followed, rather than that which is the more direct. But I am able to make some approach to the direct indication required; and the previous indirect process of inference was placed first only as a matter of convenience in arrangement. Here is the direct indication. Hitherto I have been speaking only of my waking volition. But also when asleep I am often in some sense conscious of exercising volition of some sort; and I retain an image of these sleeping volitions in memory when I am awake. Now the waking notion which I have of my sleeping will, exactly resembles the result of my synthesis aforesaid. Therefore I infer that the Determinist analysis would be a complete account of my sleeping will, but that it is an incomplete account of my waking will. And to the demand that the error in the analysis shall be pointed out I reply that the error lies in omitting to take account of the difference between waking and sleeping volition-that the analysis is a complete account of the latter, but an incomplete account of the former. (3) As to the second point: it follows from what has been said that the difference between the real will of experience and the fictitious will of Determinism is the very thing needed to turn Determinism into Freewill, by making it compatible with the reality of the vulgar notion of moral desert. For the volitions of my sleeping will do not subject me to the consequences of the reality of the vulgar notion, while the volitions of my waking will do. On contemplating the two, I seem to be made clearly aware that the |