Philosophical Studies, Part II.

By A. E. Taylor

VI. St. Thomas Aquinas as a Philosopher

This paper was originally presented at the University of Manchester in 1924.

If an educated Englishman had been asked a hundred years ago who are the great original philosophic thinkers of the modern world, what answer would he have been likely to give? His list of names would, no doubt, have depended partly upon his personal preferences, but there are some philosophers whom he would have been sure to mention. He would certainly have named Descartes, Locke, and Hume, and almost certainly Francis Bacon, then all the more admired because the real character of his theories in logic was so little understood. A widely read man would probably have given the names of Leibniz and Spinoza, and the few who had any knowledge of German literature would no doubt have added that of Kant. It is almost certain that no mention would have been made of St. Thomas or any of the great schoolmen of the thirteenth century. The current estimate of them is indicated by the remark made in 1828 by Macaulay that “we extol Bacon and sneer at Aquinas.” If the same question were put today, there would still be individual variations in the answers, but there are some names which would be contained in them all, and I think it safe to say that among these would be that of St. Thomas. We are not today all of one mind in philosophy any more than our great-grandfathers were, and I do not know that it is desirable that we should be. But if we are not |225| all of us professed Thomists, we are all, I believe, agreed to recognise in St. Thomas one of the great master-philosophers of human history whose thought is part of the permanent inheritance of civilised Europeans and whose influence is still living and salutary. It is worthwhile to ask ourselves what is the real ground for the great difference between our own estimate of the worth of the Thomist philosophy and that of our great-grandfathers. Their depreciation of the Angelical Doctor, however unjust, presumably had some sort of reasons for its existence, and if we can discover them, we may learn a lesson which will be profitable in serving to keep us from repeating the same kind of mistake on our own account.

To indicate some of these reasons and to point out their inadequacy will be the modest purpose of the remarks I now proceed to make.

There are two qualities which we may fairly demand from the work of any man whom we are to recognise as a really great philosopher with a permanent importance in the history of human thought. In the first place the work must be original, and in the second it must be critical. When I say that the work must be original, I do not mean that it need be startling or revolutionary, but that it must be the achievement of genuine personal intellectual effort. The great philosopher must be one who has thought for himself and has thought hard. No mere skillful borrower or adapter, no mere eloquent exponent of the ideas of other men can permanently retain his place on the roll of honour of the world’s great thinkers. And by saying that the work must be critical, I do not mean that it must necessarily be chiefly devoted to criticism of other men’s thoughts, I mean that it must be something more than the construction of a brilliant but undisciplined speculative imagination. The great |226| philosopher cannot, indeed, have too daring an imagination provided only that its exercise is controlled by a profound sobriety of judgement, a massive common sense. Commonly we find the two gifts in an unequal combination. The daringly imaginative mind is apt to be deficient in sobriety of judgement, the emphatically sensible mind to be wanting in imaginative power. And perhaps, when this is the case, since the object of philosophy is the attainment of truth, the thinker of really massive common sense, even if his imagination is slower in its flights, really does more for philosophy than the dazzling but erratic and unsystematic speculator. The greatness of St. Thomas as a philosopher seems to me to lie in this, that his work combines high originality with an unsurpassed sobriety of judgement and sense for reality. To our great-grandfathers this statement would have seemed a paradox, and it is precisely because it would have struck them as a paradox that they could permit themselves, in Macaulay’s phrase, to “sneer at Aquinas.” May I attempt to state briefly what I take to have been their case?

Thomas, I think they would have said, is not one of the world’s great philosophers for the double reason that he is not original and that he is not critical. He is not original because his so-called philosophy is all borrowed and all borrowed from one man, Aristotle. And what he has taken from Aristotle is simply a framework of barren and verbal formal logic. He is content to treat every philosophical question as a mere matter of bringing the issue at question under the caption of some Aristotelian antithesis, like that of matter and form or potentiality and act. No doubt in effecting the reduction, he shows great skill in the construction of formal syllogisms and the multiplication of subtle verbal distinctions. But the syllogisms and distinctions do not |227| bring us one step nearer the real understanding of concrete fact, and that is why, as was assumed, the sciences made no real advance from the thirteenth century to the seventeenth. The scholastic philosopher might devote himself as much as he pleased to the elaboration of further and further deductions from arbitrarily assumed major premises, but so long as the premises were not tested by confrontation with empirical fact, all this a-priorist ingenuity was worse than wasted. In the early nineteenth century view, at least, what science needs is not formal logic but guaranteed empirical fact, first, last, and all the time.

And again, St. Thomas and his contemporaries are uncritical, and that in more ways than one. They are uncritical, it was held, in the first place in the selection of their great authority, Aristotle. It is an arbitrary thing to pick out this one man from among all the thinkers of the past and reject all the valuable lessons which might have been learned from other sources. It is uncritical, again, when the authority has been selected, to follow it with a blind trust in its infallibility on every subject on which it has delivered itself. And finally, it is uncritical in the highest degree to submit the conclusions of philosophical thinking to the constant overruling control of theological authority, as it was commonly assumed that St. Thomas and the other great schoolmen had done. This, I believe, is the substance of what would have been stated as the anti-scholastic case, and it must be allowed that if it could be sustained it would be enough to justify the men of a hundred years ago in their refusal to reckon seriously with the scholastic philosophies.

But the anti-scholastic case cannot really be sustained. In all that is most weighty in the indictment it rests upon complete misconception of the real situation |328| of thought in the thirteenth century and the work which was done by St. Thomas and his teacher Albert the Great. I will deal first with the charge of want of originality and then make some remarks about the other charge of want of critical judgement. I have not, indeed, anything novel to say, but on an occasion like this, when we are assembled expressly to “praise famous men,” it is worth while to remind ourselves of the facts which abundantly prove the real originality of Thomas.

If we are to put ourselves at the right point of view for appreciating that originality, we must begin by understanding quite clearly that the thirteenth century, like the seventeenth and the nineteenth, in their various ways, was not one of traditionalism, acquiescence in a heritage from the past, but one of restless and audacious innovation, and that to the eyes of contemporaries “Brother Thomas” was one of the most audacious of the innovators. To speak accurately Thomism, to those who were living at the time of its birth, was not the defence of a tradition but doubly a revolt against established tradition. It is, in the first place, of course, not true that the discovery of Aristotelian logic was the work of the thirteenth century. No doubt scholasticism in its latest days did degenerate into something like the substitution of the mere reduction of a problem to the technical terminology of Aristotle for the real examination of the facts. But if we are to make this a ground of attack against the first introducers of formal logic into the modern Western world, the scapegoats ought to be looked for in the twelfth and not in the thirteenth century. It was not the latter which wasted the best of its energies in unfruitful quarrelling about the old problem of Universals, nor were Thomas and Albert primarily formal logicians at all. Their Aristotle, |229| as I need hardly remind you, was very different from the Aristotle known to Peter Abelard. The interest of the thirteenth century was not in Aristotle’s logical doctrines but in the teaching of his Metaphysics, Physics, De Caelo, De Generatione, De Anima, and the rest of the treatises which were recovered for the Western world early in the century from the hands of Jews, Persians, and Moors, and the fascination of this literature lay precisely not in its logical form but in its matter. As M. Gilson has said, the recovery of the cosmological, physiological, and psychological work of Aristotle meant to the Western world the rediscovery of Nature herself as an object of investigation in her own right and by the “natural light of reason.” We must recollect that the world into which Aristotle’s metaphysics and science, as distinct from his formal logic, were reintroduced in the early years of the thirteenth century was one which for centuries had known only one work of real value on what we nowadays call science — Chalcidius’ fragmentary version of the Timaeus, and that the Timaeus, with all its profound insight into the character of scientific problems and scientific method, does not even aim at being a repertory of scientifically ascertained conclusions of fact. In the Aristotelian works rediscovered in the thirteenth century, Western Christendom was, for the first time since the closing of the ancient Hellenic schools by Justinian, being confronted with an encyclopaedia of the Natural Sciences as a fait accompli and put in a position to realise that the external world is something more than a useful key to the hieroglyphics of the Scriptures. There was, indeed, to be a period, some centuries later, when a rekindled interest in the remains of the pre-Aristotelian Greek thinkers would serve the same purpose of recalling men from fancy to fact, but we misread the facts of |230| history if we do not realise that the enthusiasm of the thirteenth century for Aristotle was prompted by the very same spirit as the protests of Galileo against the Aristotelian traditionalism of the universities of Northern Italy in the seventeenth century. Hence M. Gilson seems absolutely right in his contention that the return of Aristotle from his Oriental captivity opened for scientific thought an era not of subjection but of liberation.

This, however, is not all that has to be said about the attitude of Thomas toward the great Greek thinkers, nor the half of it. The next point I would raise is one on which I have already said a passing word — the point that, whatever Thomism in some later phases may have temporarily sunk into being, St. Thomas himself was no traditionalist but a vigorously independent thinker who impressed his older contemporaries as a daring innovator. In fact, as I said, he was doubly in revolt against tradition. The main tradition in Christian philosophical thinking, as you all know, had right down to the days of Albert and Thomas been the Platonic as derived through Augustine and Boethius; the substitution of Aristotelianism for Platonism as the basis of a specifically Christian philosophy was a revolution and a rather paradoxical revolution. The greatest admirer of Aristotle among you would probably be ready to allow at least that, on the face of it, Platonism impresses us as a deeply religious view of existence with the closest affinities to Christianity in its doctrines of Providence, moral judgement and retribution; on the face of it, again, Aristotle’s thought, at first at any rate, certainly looks to be what it has been called in recent years by one who has a right to speak on such matters, the least religious of the great philosophies. The task of finding a basis for a Christian philosophy precisely in the “First |231| Philosophy” of Aristotle was thus a singularly bold one from which any thinker without the highest intellectual independence must have flinched.

And if Thomas was thus a bold innovator on the Platonic tradition of earlier Christian thinkers, we must remember that he was equally an innovator in the interpretation he put on Aristotelianism itself. Here again there was a definite tradition confronting him, a tradition built up by men to whom, as his repeated references show, he felt that grave respect was due, Alexander of Aphrodisias, Avicenna, Averroes, and the tradition was decidedly in favour of a pantheistic and naturalistic interpretation of Aristotle which would have been fatal to the very cause Thomas had at heart, the creation of a system of thought which was to be at once reasoned closely out at every step and wholly true to the religious demands of Christianity. The complete victory won by Thomas over the Averroists and their naturalism must not blind us to the simple fact that philosophically it was Siger of Brabant and not Thomas who was following the lines of traditional Aristotelianism. In fact, the Aristotelianism of St. Thomas triumphed over that of the Averroists not because they were philosophically heretics and schismatics (they, in fact, stood for the main tradition of centuries), nor because it is demonstrably the more faithful interpretation of Aristotle’s own thought — a position on which one might argue forever without reaching a definite conclusion — but by its sheer merits. One may convince oneself on that point very readily by a simple study of that exquisite masterpiece of philosophical polemic, the de Unitate Intellects contra Averroistas. There is no pretence here that the issue at stake is decided in any way by mere exegesis of an authoritative text. The question is not what Aristotle |232| personally meant to say, but what is true. Even if the Averroists are right in their exegesis of Aristotle, it still has to be considered whether the doctrine presented, perhaps rightly, perhaps wrongly, as Aristotle’s is true or false. It is not concealed that the doctrine the writer of the essay believes to be true and holds to be compatible with the not very explicit utterances of Aristotle is in conflict not only with the teaching of Averroes but with the whole tradition of Aristotelian exegetes. There is no pretence that we have simply to discover what Aristotle meant, and that his opinion, once ascertained, must be taken as true because it is his.

The writer’s position is the very different one that it would not be at all surprising that Aristotle or any other philosopher should have fallen into an error. Our business is to satisfy ourselves by independent hard thinking, if we can, about the truth of the matter; if we subsequently find that Aristotle’s deliverances admit of being understood in a sense compatible with the truth, so much the better for him, but if they cannot be so interpreted, we have simply to say that, like everybody else, Aristotle was not infallible and has made a mistake. On the merits of the controversy it seems to me unmistakable that the Averroists receive, what is a rare thing in metaphysical controversy, a direct and crushing defeat which makes the reading of the essay a delight to anyone who can appreciate the art of “mental warfare.” That Thomas felt this, too, I should infer from the departure at the end of the essay from his usual calm impersonality of tone. I think I detect in the sentence in which the Averroists are invited if they really have anything worth calling an argument, to produce it and see what will happen to it, an innocent chuckle, so to say, over the completeness of a triumph which the Saint evidently recognises as due not to |233| superior acquaintance with Aristotelian texts but to honest personal clear-headed thinking.

The plain fact, indeed — though I am here anticipating for a moment — is that the Aristotelianism of Thomas, if we are to call it so, is not borrowed from anyone, except in the sense in which the most original of human minds may be said to “borrow” the suggestions it needs as the pabulum for its own thinking. It is a genuinely new systematic doctrine, indebted to Aristotle of course, and to many others besides, but owing its specific form and its systematic coherency to its deviser; the right name for it is not Aristotelianism but Thomism.

How completely the new philosophical doctrine revolutionised accepted tradition in the Christian West may perhaps be seen from two little facts doubtless familiar to all of you.

One is the fate of the famous a priori argument of Anselm for the existence of God, named by Kant the “ontological proof.” From the careful study of Father P. A. Daniels it appears that out of fifteen leading schoolmen of the thirteenth century who reproduce the argument, three, of whom Albert the Great is one, express no opinion on its validity, ten accept it, only two reject it, St. Thomas and Richard of Middleton. I think we may infer that the discredit into which the argument fell with philosophers and theologians, to whom one would have expected it to be specially wel- come from its simplicity and apparent finality, is wholly due to the vigour and originality of the famous criticism to which it was subjected by Thomas. It is highly significant that the moment Descartes attempted to revive the same line of thought in the seventeenth century, the critics of his Meditations raised the objection that his employment of it would render his philosophy suspect |234| in the eyes of theologians — the very class, as I say, whose natural bias one would expect to be in its favour. It is not for me, here, of course, to express any opinion of my own on the soundness of the famous argument; I want simply to make two observations about the rejection of it by St. Thomas. The first is that there could be no better proof of the independence of mind of a thinker whose principal task in life was the philosophical defence of religion than that he should have insisted on rigorously criticising and rejecting the very line of argument which promised to be the shortest of routes to Theism, and that in the face of the general consensus of the docti in its favour. The other is that it is a point of the highest interest that there are just two great philosophers who have independently, and on different grounds, rejected the ontological argument, Thomas and Kant, and that whenever you meet a philosophic thinker today who rejects it, you regularly find, when you come to scrutinise his ground, that he is moved directly by the considerations long ago urged by one or both of these great men. Kant and St. Thomas have often been, and still are, pitted against one another by their respective followers as natural antagonists; this makes it all the more significant that they are in complete agreement on this fundamental point, rather to the scandal of many of the so-called “idealists” of our own day, who would like to regard Kant as their legitimate ancestor. The secret explanation of the agreement I believe is to be found in the fact that both philosophers are really, in the best sense of the phrase, “critical realists” with all the realist’s distrust of brilliant speculation which has no solid root in a firm grasp of empirical fact.

The other example is equally significant, and to you it will be equally familiar. Thomas seems first to have |235| attracted attention as a thinker with an individuality of his own by his declaration that the philosophical arguments in vogue for an absolute beginning of the world’s history will not stand scrutiny and that, apart from a supernatural revelation, it must remain an open question whether the created world has not existed ab aeterno, a position in which he agrees with Rabbi Moses. In our own day the bias of metaphysicians seems to be preponderantly in favour of the Aristotelian tenet of the eternity of the world, and the fact stands in the way of our appreciating the originality of this declaration at its full merits. We have to remember that the philosophical bias of the contemporaries of St. Thomas was, quite naturally, in the other direction. Since no one who was an orthodox Christian thought of doubting that the world had had a beginning at no very remote date, the tendency was naturally to attempt philosophical demonstration of what was unquestionably accepted as a known truth, and living thinkers, including St. Thomas’ own teacher Albert, had professed to be able to supply the required proof. There is abundance of evidence that Thomas’ early declared and persistent denial of the worth of all these demonstrations was felt as something like a veritable scandal. It would naturally wear that appearance all the more that the rising philosopher would inevitably seem to be playing straight into the hands of the “naturalists” of the time, the Averroists. To appreciate the disturbance created, we have only to remember the alarm caused to well-meaning and pious men by the appearance of Darwin’s Origin of Species. We know the kind of thing which would have been said in the middle of the nineteenth century if a rising divine or Professor of Theology or Moral Philosophy had publicly declared that all the arguments by which |236| opponents of Darwin were trying to refute the doctrine of the origination of new species by natural selection are invalid and will not stand examination. But to the contemporaries of St. Thomas the assertion that there are no sound philosophical arguments against the creation of the world “from eternity” involved at least as great a revolution in traditional thinking as the doctrine of the origin of species by natural selection demanded of our grandfathers. It could have come from no one but a thinker of the most marked originality and determined courage, the last kind of man in the least likely to allow his judgement as a philosopher to be determined for him by extra-philosophical considerations or bias in favour of a venerable tradition.

These remarks lead me naturally to the part of my subject which happens, from the character of my own studies, to be most interesting to myself — the attitude of Thomas to the philosophical thought of the world’s past. The persons for whom Macaulay is speaking in the words I quoted at the beginning of this lecture supposed this attitude to be sufficiently described by saying that St. Thomas subjected human thought to the despotism of Aristotle. But the judgement is only intelligible when we remember that the accusation dates from a time when the serious revival of Aristotelian study in Europe had barely begun. The Aristotle of the “advanced thinkers” of the eighteenth and early nineteenth centuries was not the Aristotle of history as we have since learned to understand him, nor even Aristotle as seen through the eyes of St. Thomas himself, but Aristotle as falsely conceived by the epigoni of scholasticism in its feeblest and least original period. From our point of view such an estimate ought to be at once and for all impossible. It would be truer to say that what Thomas effected for the first time in history |237| since the expiry of neo-Platonic learning in the sixth century was a magnificent and original synthesis of past philosophical thought. He took his materials freely from the whole record of the classic past, so far as it lay open to him, and what he constructed out of them was no chaotic eclecticism but a coherent system welded into a unity by the presence throughout its details of a few great ruling principles, won by permanent hard thinking and held with the clearest consciousness of their implications. It is not true that he changed the existing philosophical tradition by dethroning one uncritically accepted authority and enthroning another. It would be much truer to say that he retained and built upon the thought which had been accessible to his predecessors, enriching and integrating it with the wealth of new matter made accessible to his own age by the recovery of Aristotle. Indeed, I do not know that it would be going beyond the mark to say that, for the first time in the life of the modern world he attempted something like a critical and thoroughly historical appreciation of past philosophy in its entirety. I cannot develop this thought so fully as I should like to do, but there are certain aspects of it upon which I should like to make a few remarks.

Aquinus’ knowledge is “full of lacunae, and Aristotle’s statements about his predecessors are often both obscure and vitiated by the defect, shared by Aristotle with Hegel, of inability to appreciate the ideas of others except with reference to his own system.”

To a student of Greek philosophy it is of great interest to note how frequent are the references in that great philosophical masterpiece, the Summa contra Gentiles, to the very thinkers about whom the men of the thirteenth century could have the least trustworthy information, the pre-Platonic men of science. It is curious how often the author goes out of his way to comment on the theories of men like Anaxagoras and Empedocles, and what earnest efforts he makes to understand them and to call attention to the truths they may be supposed to have been struggling to convey. |238| To be sure, many of his guesses at their meaning are such as we can now see to be unhistorical, but we have to remember the paucity of his information. In the main his knowledge about the earliest Greek thought is necessarily based on scattered notices in the text of Aristotle. He has thus to reconstruct under the double disadvantage that his knowledge is full of lacunae and that Aristotle’s statements about his predecessors are often both obscure and vitiated by the defect, shared by Aristotle with Hegel, of inability to appreciate the ideas of others except with reference to his own system. This, however, is a difficulty common to him with all writers about the beginnings of Greek science down to a very recent date. We have to remember that the work of collating the various sources of information, correcting the often grave errors in the traditional texts and establishing the historical affiliation of the sources, an indispensable preliminary to satisfactory interpretation, can hardly have been said to have begun before the publication of Diels’ Doxographi Graeci and the great Berlin edition of the commentators on Aristotle, well within our own lifetime, and is not even yet fully completed. Bear all this in mind, and then, comparing Thomas’ treatment of those of the early thinkers about whom the literature available to him contains a reasonable amount of information with Hegel’s treatment of the same subject in his lectures on the History of Philosophy, ask yourselves which of the two men, when proper allowance is made for superiority of the material available to Hegel, has used his material in the more objective and scientific way. I honestly do not think the advantage will be found to be on the side of the German.

A more important question is that of the relation of Thomas to Plato and to Aristotle himself. Here I am afraid I shall be bound to be a little more prolix. There is |239| a misleading but common impression that down to the middle of the thirteenth century the foundations of the philosophy of the Christian ages had been Platonic, but that at that date Thomas and Albert effected a sudden revolution by rejecting Plato for Aristotle. There is, of course, a certain truth at the bottom of this statement, but, like all summary statements, it is dangerously misleading unless it is very carefully qualified. In the first place it suggests the thoroughly unhistorical view that the philosophies of Plato and Aristotle are in principle opposed to one another, whereas the truth, as all careful students of both philosophers know, is that the differences between the two thinkers, important as some of them are, are comparatively few and are mostly on points which are quite secondary by comparison with the fundamental principles in which the two great philosophies agree. That Aristotle himself was clearly alive to this is manifest all through his work. The undisguised object of the standing polemic which he urges against the contemporary Academy is to suggest that, in spite of all differences in formulas, in spirit it is himself, not Speusippus or Xenocrates, who is the legitimate heir to the intellectual achievements of Plato. Nor do I think Plato himself would have resented this attitude. No doubt, if he could have returned to earth, he would have protested that much of the Aristotelian criticism rests on misconception of his meaning, the kind of misconception inevitable when a biologist attempts to follow the thought of a mathematician. But the philosopher who deliberately refused to compose a treatise on his own philosophy on the ground that philosophy is not a body of formulae or doctrines, but the living spirit of personal devotion to the pursuit of truth, would have been the last man to disown a successor on the strength of a difference in expression or opinion.

Much of Aristotle’s criticism rests on misconception of Plato’s meaning.

|240| The popular conception again completely misrepresents the facts about the actual achievement of Thomas. It is true that on a first perusal of him we are struck by the uniformity with which the familiar Aristotelian technical formulae recur in his treatment of all questions. His logical equipment, in particular, comes direct from the Topics and the Analytics. This is, in any case, what we should expect. There was, in fact, no other body of articulated logical doctrine with which to work. But when we penetrate to the matter which is presented to us in this vesture of Aristotelian logic, the case is altered. The Augustinian exemplarism, itself a direct derivation of the Platonic doctrine of Forms, is, as we know, an important and integral part of the Thomist philosophy. Intimately connected with it is the equally fundamental conception of causality as a process in which the effect imperfectly “mirrors” the cause, and this, with the great notion of the ladder of being, or scale of perfection, is wholly a legacy, through Augustine, from the neo-Platonists. It is, I think, safe to say that none of these conceptions could have been derived from the text of Aristotle unless Aristotle had first been read by the light of Platonism and neo-Platonism as mediated through Augustine.

Indeed, I should like to go rather further than this. It seems to me at least an ambiguous statement to say that Thomas directly opposes Platonism in metaphysics at all. He certainly opposes particular doctrines of the Platonici, but it is another question whether he can be said to oppose the doctrines of Plato. (I am speaking, of course, of the Platonic metaphysic; I do not refer to psychology and the theory of knowledge, where we all know that Thomism is definitely Aristotelian.) Before we can say that Thomas opposes the Platonism of Plato, we need to be quite sure what Plato’s own ripest thought |241| was, and to be sure on that point is a very difficult matter.

Plato has warned us that “his deepest thoughts are not to be found in his writings; he reserved them, in fact, for oral exposition to the students of the Academy.”

The trouble is that Plato has himself been at some pains to warn us that his deepest thoughts are not to be found in his writings; he reserved them, in fact, for oral exposition to the students of the Academy. Our knowledge of them has to be gathered in the first instance from the allusions of Aristotle, who notoriously treats as the distinctive doctrines of Plato propositions which are not to be found in the Platonic dialogues at all. Unfortunately Aristotle is most commonly content to make enigmatic statements without explaining what they mean, a task which would be superfluous while the original hearers of Plato were alive and active, and even when he gives an explanation we cannot be sure that it is one which Plato himself would have accepted as accurate. The dramatist’s gift of entering with ready sympathy into the inmost thought of another mind appears, indeed, to have been one with which Aristotle, with all his other gifts, was not richly endowed. The indispensable preliminary to real comprehension of Plato’s personal thought would seem to be a careful collection of all the notices of his oral teaching to be found in Aristotle and a confrontation of them with all the similar records of such notices by Xenocrates and other Academics as still survive in the commentaries of Simplicius and other men of learning in later antiquity. This preliminary work has only been systematically undertaken within the present century by the industry of M. Robin,1 and the task of interpreting the material he has collected for us cannot be said to have advanced beyond its bare inception. It will be time to raise the question whether Thomism and Platonism are really in fundamental antagonism when we have rediscovered, |242| if we ever do rediscover it, the real tradition of the first generation of the Academy. Meanwhile, I may perhaps be allowed to utter a word or two by way of caution.

If, as seems perhaps probable, Plato really meant to make the world of Forms which is the object of science something existing apart from God, independent of Him and above Him in the scale of perfection, it need not be said that Thomas completely rejects that part of his teaching, though the rejection is not a conscious repudiation of Plato, since no one in the thirteenth century understood Plato in this sense. So far as I have observed in my own reading, what polemic there is against the Platonici in the principal works of St. Thomas is principally connected with two points, (1) that Plato regards “natural species” as substances and (2) that the Platonici have said that the soul is in the body as the sailor is in his ship. The first of these statements about Plato is certainly borne out by well-known polemical passages in Aristotle, though it is interesting to remember that the precise meaning and the justification of these Aristotelian criticisms is still hotly controverted among special students of Greek philosophy. The second statement seems to an outsider like myself a little of a puzzle. The phrase about the sailor and the vessel does not actually occur anywhere in Plato and is not really well-chosen to illustrate the view about the relation between the soul and the body suggested even in the Phaedo. St. Thomas presumably took the phrase from the passage in the De Anima (413a9), where Aristotle says that the problem is one which will have to be considered. The curious thing is that the ancient commentators were all completely in the dark about the bearings of the observation; they are quite uncertain whose theory is alluded to, and even uncertain whether Aristotle |243| meant to express approval or disapproval. The one thing they do not suggest is that the reference is to Plato. Possibly, then, St. Thomas was mistaken in seeing any reference to Plato in the words. In any case the real attitude of Thomas to the great thinker who had inspired the philosophical thought of Augustinianism cannot be properly determined by reference to this rather special and limited polemic. It is much more significant, though it is too often forgotten, that Thomism incorporates in itself the whole of the Augustinian exemplarism. This really brings Thomas much nearer to Plato than Aristotle himself is, at least in his controversial moods. Aristotle often allows himself to speak as though the whole Platonic doctrine of Forms or Ideas had been “moonshine,” “empty words”; Thomas never says anything of the kind. In his theory the archetypal Forms have an important part to play; it is only that Plato was mistaken in supposing that they are directly accessible to our imprisoned intelligence. {Via Christa editor: Our “imprisoned intelligence” is captive to the doctrine of materialism, which pays little heed to our soul or Plato’s doctrine of recollection. Western education reinforces this failing at every turn.}

And what of the relation of Thomas to Aristotle himself? Is not this equally misrepresented when he is spoken of as an “Aristotelian” or a “follower of Aristotle” without further explanation? Clearly I think in using such unqualified expressions there is the danger of creating a thoroughly false impression. There is one sense in which Thomas is no follower of Aristotle nor of any other man. He never accepts a doctrine because it has been taught by a man with a famous name and an established reputation; what he accepts he accepts because he believes it to be true, or if not absolutely and certainly true, the nearest approximation that can be made to the exact truth. If then he, in many parts of his philosophising, follows Aristotle so closely, it is because he is convinced by the independent exercise of his personal thinking powers that Aristotle is on the right |244| lines. You will remember how explicitly this point is made in the annihilating critique of the Averroist doctrine about the “unity of the intellect.” We are told there, as plainly as we could be told, that our real concern is not with what Aristotle taught but with what is true. If the Averroists could succeed in showing that their own exegesis of Aristotle is correct, that would be so much the worse for Aristotle, but none the better for Averroes. The sanction of Aristotle is never an adequate refuge for error. One remembers, too, how in the commentary on the De Caelo2 we are explicitly warned against the very mistake made by the traditionalists of the time of Galileo. We may be content, we are told, to accept the scheme of the Aristotelian (or rather Eudoxian) astronomy, because “it saves the appearances,” but we must not insist that the machinery it assumes of the rotating spheres is real, “because it is quite possible that the appearances may equally well be saved by some other theory yet to be put forward.” That is, in modern language, we may use the current astronomy as a convenient descriptive theory on the strength of which eclipses and other astronomical events may be correctly calculated, but we have no right to treat the description as if it were an explanation. It is plain, I think, that if Thomas, who was, of course, well acquainted with the fact that a heliocentric astronomy had been taught in antiquity by Aristarchus and Seleucus, could have been confronted with the revival of the theory by Copernicus, he at least would have been ready to consider with an absolutely open mind the question whether the heliocentric theory does not “save the appearances” better than the geocentric.

In general, as to this matter of the degree of dependence |245| of St. Thomas on Aristotle, there are, it seems to me, one or two things well worth saying even in a brief paper like this. (1) I suppose no one has ever made an extensive and repeated study of the Corpus Aristotelicum without feeling strongly that there were, so to say, two Aristotles. There was the Aristotle whose tendencies are to “naturalism” in philosophy and to detailed and specific research in the sciences, and there was the Aristotle who was “carried off his feet,” as Professor Burnet has said, in the Academy by the Platonic passion for the divine and eternal. The consequence of this manifest clash in the philosopher’s own soul between the “naturalistic” bent, presumably acquired in early education, and the “otherworld” tendency due to Platonism is a curious fault” which runs through all the chief Aristotelian treatises. Thus, just by way of indicating the presence of the “fault,” let me suggest the questions: (a) What is, according to Aristotle, the proper specific object of “first philosophy”? Is it “being as such,” or is it the eternal and absolute being, the “unmoved first mover”? (b) Is it really possible to fit on the few broken and enigmatic remarks of the De Anima about the imperishable “active intellect” to the general straightforward naturalistic account of the process of knowing and its presuppositions? Does not one feel of all the exegeses alike, “this may be, in point of fact, what Aristotle ought to have said, but I cannot convince myself that it is what he actually meant to say”? (c) Or again, does any of us believe that Aristotle has really succeeded in his Ethics in harmonising the view that the good for man is the special object of the science of “Politics,” and thus belongs altogether to the “active life,” with the other view which he springs on us at the end of his argument that our truest good can only be found in “contemplation”?

|246| On the face of things, Aristotle’s philosophy as Aristotle himself left it, is an imperfectly achieved attempt to hold together a secular or “one-world” and a religious or “two-world” view of things. The ordinary “naturalist” is content to see only the one world of the sensible and present; the “Platonist” is so interested in the “other” world of the unseen and eternal that his tendency is to come as near as he dares to treating “this” present world as a shadow or a bad dream. For a thoroughly critical philosophy the problem is precisely how to combine aright the two complementary attitudes of frank acceptance of the “secular present” and the noble “detachment” which refuses to accept it for more than it is worth. In Aristotle’s own philosophy, as it seems to me, both attitudes find their recognition, but they are not harmonised; they simply alternate. The general rule is that in any considerable work of Aristotle you start with what seems to be a thorough-going “empirical” “secular” or “this-world” attitude to the world and an avowed opposition to everything that is “transcendental.” But by the time you reach the end of the treatise, you find yourself landed in the full-blown “transcendental” — “the unmoved mover,” the “imperishable active intellect,” “the contemplative life,” without any clear indication of the way in which the transition has been effected. Here, as it seems to me, the so-called Aristotelianism of Thomas is much more thoroughly thought out and coherent than what I may call the Aristotelianism of Aristotle. The “this-worldly” and the “other-worldly” are not juxtaposed; the one is subordinated to the other in virtue of definite guiding principles clearly laid down and the relations of the superior to the subordinate are made logically transparent. To my own mind the clarity which is thus brought into the treatment of this supreme problem of |247| the relation of the eternal and the secular is the best proof of all of the genuine originality of the Thomistic thought and of its perennial significance for all generations of men. I may give as an instance of the quality I have in mind the treatment of the relation of the temporal and eternal good of man in the third book of the Summa contra Gentiles, as compared with the treatment of the relation between the life of practice and the life of speculation in the Nicomachean Ethics. I do not think there can be much doubt here which of the two philosophers show the coherency, lucidity, and assurance which mark the utterance of one who is really master of his theme.

“The Thomist philosophy is no mere Aristotelianism revised but a masterly synthesis of both Plato and Aristotle with one another and with Augustine, effected by original insight of the first order.”

For my immediate purpose, which is to urge that the Thomist philosophy is no mere Aristotelianism revised but a masterly synthesis of both Plato and Aristotle with one another and with Augustine, effected by original insight of the first order, it is particularly important to make two remarks.

As I have said, in the thought of Aristotle himself, when one tries to study it in a strictly historical spirit without preconceptions, there appear to be two conflicting strains, the naturalistic or positivist and the Platonist. Hitherto students have been content to note the fact without attempting to discuss the question of the historical development of Aristotle’s own mind. From the rise of the succession of expert Aristotelian exegetes in the first century of our era down to the dawn of our own twentieth century, the encyclopaedic “wor of the great thinker has only too successfully hidden his personality from us. The first serious systematic attempt known to me to reconstitute the features of the son of Nicomachus as a living personality with a history of growth behind it, has only just been made by Werner Jaeger in his fascinating Aristoteles, published |248| only last year.3 Many of Jaeger’s special results must, no doubt, be considered tentative, and I have elsewhere tried to show that in some cases he has fallen into definite errors. But in the main I believe he has made out his principal thesis, and that we can distinguish in all the principal works of Aristotle well-marked earlier and later sections, the general formula being that Aristotle starts on his personal career as a convinced and enthusiastic Platonist, passes through a phase in which he attempts to show that all that is most fundamental in the “other-worldly” Platonic metaphysic and ethics can still be retained when the doctrine of “Separate Forms” has been eliminated, and ends by approximating more and more closely to the cultivation of an almost purely empirical and positivist cultivation of the details of the special sciences. If this is so, I think we may safely say that it is Aristotle the Platonist rather than Aristotle the positivist who influences the thought of Thomas.

To illustrate by one or two examples. Jaeger has, I think, made it clear that there is a shift in the Metaphysics as it stands from the conception of “first philosophy” as the study of “being supreme and eternal” to the thought of it simply as the study of “being as such,” and the only reasonable explanation of the shift is that the parts of the work where the conception of “theology” as the crown of science is dominant come from manuscripts of the transition period, in which Aristotle is still concerned with the exposition of a real though remodeled Platonism, those in which “first philosophy” is equated with the science of “being as such” from those of the later and more positivist period. So with special reference to the famous book of the Metaphysics which expounds the conception of God as the “unmoved first mover,” I cannot resist the arguments used |249| by Jaeger to prove that in the main we are dealing with Aristotle in his earlier vein, while the chapter which introduces the fifty odd “unmoved movers” of the planetary spheres is a later addition in the interests of positive science, but wholly out of keeping with the tone of its context. As Jaeger says, from the point of view of this addition, God sinks from the position of being the one supreme and abiding source of all temporality to the position of something like a mere unknown cause of the diurnal revolution of the heavens, an astronomical hypothesis much on the same level as gravitational attraction in the Principia of Newton, or, if you prefer it, of the as yet undiscovered cause of gravitation of which Newton speaks in the concluding Scholium Generale.

Now there cannot be the slightest doubt that it is Aristotle in what seems to be his earlier vein, the Platonist Aristotle, who means so much to Thomas. One may doubt whether if the Metaphysics had been completely worked over from the point of view of the chapter on the planetary movers, so that God appeared throughout simply as the first member of a whole series of unknown causes of movements of rotation, the book would have had much fascination for him. Or I may make the same point in another way. I have sometimes asked myself what are, when all is said, the two or three leading conceptions drawn from Aristotle which are all-pervasive in the system of Thomas. I believe, though I am of course offering only my own personal impression, that we may reduce the list, if we confine ourselves to matters of absolutely first-rate importance, to two. One is the great conception of “potentiality” and the significance for every branch of science of the distinction between potentiality and act. I know that in modern times there has been a violent revolt against this |250| distinction and that many philosophies have made determined attempts to get rid of the notion of potentiality once and for all and to recognise nothing but actuality. The revolt was primarily intelligible and in many respects salutary. I take it, it would be generally admitted that there is some foundation for Francis Bacon’s complaint that Aristotle and his followers had corrupted natural science by this frigida distinctio. The schoolmen of the decadence really did tend to forget that the distinction never affords a sufficient description, not to say an explanation, of the specific detail of any process in nature, and for my own part I think to some extent they could plead the unfortunate example of Aristotle himself for their error. Since no philosopher is infallible, I should not be much surprised if Thomas himself can be proved to be sometimes a sinner in this way. But the history of biology or psychology, or any science which deals with objects which live and grow on characteristic lines, is enough to prove that in philosophy to ignore potentiality and treat the not yet actual as simply “what is not” leads to nothing but confusion and disaster. Insistence on the conception of the “potential” seems to me to be one of the most valuable features of the intellectual inheritance which we have to thank the great men of the thirteenth century for preserving for us. The other great fundamental principle which strikes me as all-important is the famous doctrine of the equivocity of “being” (or, if you like to put the point differently, of the irreducibility of the categories to any one supreme category), which is perhaps even more important for Thomism than for Aristotle himself, since it is the foundation of the whole theory of analogical knowledge as the means of escape from sheer ignorance of the supra-temporal in philosophy. It happened that some time ago I was reading simultaneously Eriugena’s |251| famous De divisione naturae and the Summa contra Gentiles, and naturally found myself led to make some comparisons. On one point, to be frank, I had to confess that Eriugena seemed to me to have an advantage; whatever one might think of his conclusions, one could not but be impressed by the extraordinary vigour of a speculative imagination to which one seemed to find no later parallel until one came down to Giordano Bruno in the full tide of the anti-scholastic reaction. Yet, on the other hand, quite apart from any consideration of theological consequences, one could not help feeling that, regarded simply as philosophical speculations, the speculations of Eriugena, like those of Bruno after him, were undisciplined and fanciful; one was dealing with the work of an imagination hardly controlled by sober judgement. The question this reflection suggested to me, as I thought by contrast of the magnificent sobriety of Thomas, was this. Was there any fundamental philosophical principle, unknown to John the Scot, which served to keep the thirteenth-century thinkers secure from the excesses of an unregulated imagination? If there was, was it a principle which they derived from their study of Aristotle and could not equally well have learned from Plato and the Platonists, who were as well known to a learned man in the ninth century as to the learned in the thirteenth? My own answer was that, so far as I could see, the principle wanted, a principle which could not well have been thoroughly learned except from Aristotle, was precisely that “being is predicated equivocally” or, in other words, that the categories form an irreducible plurality. This, as it seems to me, is the ultimate principle on which all the wild and dangerous philosophical Monisms must be shipwrecked, the safeguard of sane and sober critical thinking, the one indispensable form of “pluralism” which must |252| reappear in any philosophy with pretensions to be true.

Now both these principles belong as much to the earlier Aristotle as to the later. The distinction between potentiality and act, though best known to us from the prominence given to it in Aristotelianism, appears to have originated in the Platonic Academy and to have been the common property of the school. The immediate proof of this is that we find the distinction already employed without explanation, as something familiar, in a fragment of Aristotle’s lost Protrepticus which must have been written before the death of Plato, and can be seen from the remains preserved from it by later writers to have been an eloquent exposition of Platonism. Indeed we may fairly go a step further back. In the famous passage of the Theaetetus where Plato is making the distinction between knowledge in actual use and knowledge which we have acquired and can revive on occasion but are not actually using (Theaetetus 197 c) we have the very phrase that “in a sense we have none of these pieces of knowledge, when we are not using them”, what we have is the “power” of putting our hand on them. It is, I think, not unlikely that this observation is the starting point of the whole doctrine of potentiality and act, though, if it is, this of course in no way lessens our debt to Aristotle for his recognition of the fundamental importance of the distinction and its significance for the whole study of nature and of man.

The other great principle of the irreducibility of the categories is nowhere, so far as I know, expounded in Plato, and I do not feel sure that the later world would have arrived at it without great difficulty if it had not been able to learn the lesson from Aristotle. At the same time, it is fully in accord with the principles of Plato. Even of Xenocrates, against whom in particular, |253| the majority of Aristotle’s attacks on the Academy appear to have been aimed, we happen to know that he contented himself with the reduction of the categories to two, Substance and Relation. We may say then, I believe, that if the Thomistic metaphysics may be called Aristotelian in the sense that it is to Aristotle they are directly indebted for their most fundamental formulae, it is a misapprehension to regard them as Aristotelian in the sense that they are in principle anti-Platonic. They represent rather a rich and full synthesis and coordination into a systematic whole of the results won by the thousand years’ long travail of Greek philosophical thought, a synthesis only possible to a mind of the first order. It has been said that Kant, by far the most epoch-making of modern philosophers, aims at a synthesis of Hume and Leibniz, as imperfectly apprehended through Wolff. The remark was not, of course, meant as a disparagement of Kant’s originality; the suggestion is that the very achievement of such a synthesis is itself only possible to a thinker of the most absolute originality. But, by comparison with the Thomist synthesis of Plato, Aristotle, and Augustine, how comparatively incoherent and loose is Kant’s synthesis of Hume and Leibniz. The one has the unity of a great work of art, the other is by comparison an ill-constructed amalgam which visibly falls to pieces under the reader’s eyes.

In epistemology and psychology the case stands rather differently. Here the foundations of the Thomist edifice may really be called Aristotelian in the more special sense which I have just deprecated for the Thomistic metaphysics. And I confess that it is just here that my own difficulties about what seems to be St. Thomas’ teaching begin. I find it for example — perhaps the fault lies simply with my own failure to |254| understand — hard to reconcile the character of our knowledge in pure mathematics with the restriction of the functions of the intellect in man to the work of abstraction from what is given in sense, and again to satisfy myself that the theory of the part played by the “sensible species” in perception is quite consistent with itself, and with the surely sound conviction that the perceived qualities of things are real qualities of real things and not “psychical additions” to the reality. And yet, for all my difficulties, here again I feel convinced that if we would make progress to a sound theory, we cannot do better than go to school to St. Thomas. It is a striking sign of the times that contemporary philosophy is coming more and more to busy itself with precisely the problems which confront us in the Thomist epistemology and psychology of knowledge. The tendency is certainly spontaneous, since most of those who exhibit it in a very marked degree are pretty obviously quite unacquainted with all mediaeval philosophy. Yet it is undeniable that something like a return to the mental outlook of the thirteenth century is being forced upon us today in at least two ways. We have witnessed a widespread and vigorous revolt against the type of epistemological theory — ultimately derivable from Kant — which attempts to safeguard scientific knowledge against sceptical criticism by pronouncing the scientific characters of that knowledge, its universality and necessity, to be a “psychic addition” put into things by the human mind de suo. If we choose to define the very ambiguous word “Idealism” arbitrarily as meaning the doctrine that universality and necessity are “put into nature by the knowing mind,” we may fairly say that there are no “idealists,” in this sense, left today. The epistemologist of today, as compared with his predecessors of forty years or more since, |255| usually has the advantage of superior knowledge of some specific branch of science, and the influence of science, both natural and mathematical, is shown in this widespread conviction that a true theory of knowledge must treat knowing of all kinds from the outset not as a process of “creating” but as an adventure of discovery. We do not put “the categories” into Nature; we find them there. This view of the relation between the knowing mind and the Nature it knows, of itself, takes us back to the “critical Realism” characteristic of the philosophy of an age earlier than the unfortunate subjectivisation of the philosophical problem by Descartes.

It is significant also that, as the current literature shows, the reaction towards what I may call an epistemological realism has once more made actual a whole class of problems prominent in Thomistic philosophy but until recently too lightly dismissed by most later thinkers. When we have convinced ourselves that the apprehension of Nature is not, as Descartes too long led us to believe, indirect, through the mediation of “ideas,” but direct and first-hand, so that we immediately perceive the genuine qualities of real things, we are confronted, of course, with the difficult task of correctly explaining the facts which have been the stronghold of the theory of indirect or representative perception (facts about double vision, hallucinations, dreams, etc.), and of devising a really sound scientific account of the function of the processes which in the organism, though not in the mind’s apprehension, intervene between stimulation of the sense-organ and awareness of the apprehended quality. In a word we have back on our hands precisely the same questions which in the philosophy of Thomas are answered by the theory of the sensible and the intelligible “species.” As I have confessed, I have never |256| been able to feel sure that I quite grasp St. Thomas’ thought on this subject, though I do seem to discern that it is much subtler than the rather crude psychology apparently intended by Aristotle himself. But this at least is clear, that a theory of perception and perceptual knowledge which is to meet the acquirements of modern science will have to be something in its general character very much like that of Thomas. It will have to combine, as he at any rate meant to combine, the two complementary positions that our knowledge of the world around our bodies is mediated in fact by highly complicated processes of a very special kind, and that as knowledge it is direct, unmediated apprehension not of “ideas” or “images” but of actual physical reality. No one, so far as I know, among the great modern philosophers has ever seen more clearly than Thomas that the problem is precisely not to sacrifice one of these true positions to the other. Hence we may fairly say that the great task awaiting the epistemologist at this moment is no other than the task of providing us with an equivalent, expressed in terms of all that we know about the physiology of the brain and sense-organs, for the Thomistic doctrine of the part played in perception and perceptual knowledge by the sensible and intelligible “species.” For all I know, it is possible that the theory as it stands only needs to be translated into the language of modern physiological psychology, without further modification, to prove the very truth of which the epistemology of the present moment is so anxiously in search. Even if it should not be so, and modification as well as translation is called for, I am at least sure that the careful study of the Thomist doctrine on the subject is the best preparation for fruitful meditation of one’s own, and that the bad habit of beginning the study of so-called “modern” philosophy with Descartes, |257| in whom the epistemological problem is falsified from the first by “representationist” assumptions, is responsible for generations of mere fumbling in the dark which might have been escaped if the gentlemen of the eighteenth and nineteenth centuries had been willing to do less “sneering at Aquinas” and more study of him.


1^ L. Robin, La Théorie platonicienne des Idées et des Nombres d’après Aristote, Paris, 1908.

2^ ii. 17, forte secundum aliquem alium modum, nondum ab hominibus comprehensum, apparentia circa stellas salvantur.

3^ Jaeger, Aristoteles. Berlin, 1923.

Top ↑

VII. Francis Bacon

This paper was originally presented before the British Academy in 1926.

It is now three hundred years and a few odd months since the Easter Sunday1 on which Francis Bacon, Viscount St. Alban, ex-Chancellor of England, and author of the Great Instauration, passed from this temporal scene a disgraced and broken man. In his Will, executed some four months earlier, stand the words familiar to us, “for my name and memory, I leave it to men’s charitable speeches, and to foreign nations, and the next ages.” The appeal for charitable construction has been deprived by lapse of time alike of its force and its justification. Educated countrymen of Gardiner and Acton are no longer in danger of corrupting the story of our past into a deceitful legend for the greater glory, or shame, of a sect in politics or divinity. Time has also, and that none too soon, brought the opportunity for a sane and sober judgement on the services of Bacon to the cause which lay closest to his heart, the advancement of the sciences. The future to whose censure he submitted himself, as we may fairly think, has arrived. That the verdict of the “foreign nations” may not be wanting, the last year has brought us what I may perhaps call the most thorough, elaborate, and impartial study of the Baconian philosophy hitherto produced, the acute and learned monograph of Dr. Adolfo Levi, |259| Il Pensiero di Francesco Bacone.2 We should now be in a position to judge of Bacon’s services or disservices to scientific and philosophical thought in the only way befitting the historian of ideas, with severe and serene disregard of all questions affecting his character as a friend, a statesman, or a judge. These are matters for a different, it may be a higher, court: in this place and on this occasion, Bacon comes under our consideration as a thinker, not as a man, and it would be an impertinence to forget the distinction. With the rights and wrongs of his treatment of Essex or Peacham or Aubrey we have as little to do as the student of Shakespeare’s tragic art with the morality of the sorry intrigue half disclosed in the Sonnets.

We may doubt whether it would have been easy to pronounce a thoroughly impartial judgement even on this question of Bacon’s real merits as a natural philosopher before the opening of the present century. Science and philosophy, no less than politics, have their sectaries and partisans. Among ourselves, all through the first fifty or sixty years of the nineteenth century, the period which saw the two great editions of Bacon’s Collected Works — those of Basil Montagu and of Ellis, Spedding, and Heath — as well as Macaulay’s only too famous article in the Edinburgh — the dominant mood was one of uncritical and unhistorical magnification. What Pindar said of Hiero3 Englishmen of that age said in their hearts of their “national” philosopher; whatever fell from his lips was presumed to be great because it came from him. It was a point of honour, even with zealous Whigs like Macaulay, whose political bias led them to put the harshest construction on |260| all the acts of the Chancellor, to insist that the thinker had revealed once for all the one “true way” in science and philosophy, and that nothing was left for after ages but to tread meekly in his steps, in Verulami ponere ficta pedum pressis vestigia signis. Then, as might have been expected, followed the natural rebound from this idolatry, marked, for example, by De Morgan’s vigorous and mirthful criticism of Bacon and his fervent, Macaulay4 and Jevons’ piquant contrast between the false scientific method of Bacon and the true method of Newton,5 and culminating in the open contempt avowed by philosophers whose own conception of science had been chiefly inspired by Kant, a contempt expressed pithily in the complaint of Lewis Nettleship that Bacon, though Nettleship owns to “finding a good deal in him,” is “typical of the inferior English characteristics, a sort of swagger and ignorant independence.”6 A generation for which both the wave of adulation and the counter-wave of detraction are spent forces should be able to judge more historically and with more certain discernment.

It is not difficult, in the light of history, to understand the source of the perturbing influences which vitiate both the estimates to which I have referred. The supreme outstanding feat of the eighteenth century, so far as natural philosophy is concerned, had been the definite conquest of the mind of Europe by the Newtonian mechanics of the heavens. As developed and translated into the language of analysis by Lagrange and Laplace, the Newtonian gravitational astronomy appeared to be not only the last word of astronomical science itself, but the flawless model of explanatory theory to |261| which all genuine natural science ought to conform.7 The key to all nature’s mysteries, it was thought, must be found in the conception of “attractive forces” acting at a distance,8 and the work of the new century was expected to be the reduction of molecular physics and of chemistry to the Newtonian pattern. It is true that it would have been hard to prove that Bacon had exercised any special direct influence on Newton, but it is not surprising that no one should have been troubled by scruples on such a point in the first flush of the great triumph of the Newtonian ideas. What was undeniable and patent was that physical science, after remaining, to all appearance, stationary for so many centuries, had, in the hundred and odd years since the publication of the Principia, come triumphantly into its own, and seemed likely to extend its domain indefinitely by the mere process of applying the concepts of the Newtonian mechanics systematically in one field of fact after another. And it was not difficult to see where the Newtonian methods in science differed most obviously from those of the preceding ages of sterility. It was well understood that, as Dr. Whitehead has reminded us,9 science had won its resounding victories by a revolt against rationalism — the metaphysician’s characteristic demand for principles which completely justify themselves to the reflective intelligence — and a stubborn loyalty to “brute” fact for which no reason can be produced except that “we see it happens so.” The great outstanding triumph of brute fact over “rationality” was, indeed, just the peculiarly Newtonian element in |262| the Newtonian mathematical physics, the gravitation-formula. There is nothing self-luminous about the “law of the inverse square,” no superior rationality inherent in a system pervaded by that law rather than by another which might be a possible mathematical alternative:10 there is an apparent irrationality about the scheme which would have been resented by the typically rationalist intellect of the great thirteenth-century schoolmen and was actually resented by the equally rationalistic intellect of Leibniz, the most universal genius of the Newtonian age. To a thoroughly rationalist mind it was a rock of offence that the Newtonian celestial mechanics require as a first principle a formula which has no logical connection with the general Newtonian laws of motion and can only be supported by an appeal to brute fact. For the same reasons it is today an advantage of the “general theory of relativity,” in the eyes of a rationalist, that its laws of motion are so stated as to include the facts of gravitation and dispense with the need for a special gravitation-formula.11

Thus Newton’s famous hypotheses non fingo was naturally understood, all the more for its explicit allusion to the arch-rationalist Descartes, as the formal repudiation of rationalism, i.e. of the demand that natural science shall admit none but transparently self-evident premises.12 As everyone knew, two full generations before Newton, Bacon had been foremost in complaining of the scientific sterility of the mediaeval rationalism, |263| and preaching the doctrine that the one hope of progress lies in a true induction from observed and given fact. It was thus not without reason that the men of a hundred years ago saw in the Novum Organum the first plain enunciation of the programme of the return from rationalism and ontology to reverence for the given in all its inexplicability and apparent lack of justifying reasons.

This view was, no doubt, one-sided. The direct and immediate inspiration of Newton’s work came from a different quarter. It is Kepler and Galileo whom Newton has really to thank for the services without which there could have been no Newtonian mechanics of the heavens, and Kepler and Galileo, while fully sharing Bacon’s reverence for the factual and directly given, had understood, as Bacon never did, that mere acceptance of the given will never of itself give birth to science, but needs to be quickened by the fertilising influence of mathematics, the very type of strictly rationalistic thought. We have to allow here for the pardonable consequences of a failing most incident to man in all ages, national pride. The conqueror who had taken European thought captive was an Englishman; the more rationalistic philosopher whom he had dethroned in the universities and academies, a Frenchman. If it could be plausibly asserted that the precursor who had heralded the advent of the conqueror had also been one of our own countrymen, the story of the revival of science in the modern world would become a tale of the victories of the English intellect. How completely patriotic pride could obscure the facts for our grandfathers we may see from the confident air with which Macaulay lays it down that all great discoveries, in science, as in politics, are “ours”; France, the only rival whom the essayist deigns to mention, is relegated to |264| the humble position of interpreter of the English genius to an inferior world.13 The complacent pronouncement forgets Copernicus, Galileo, and Kepler, Torricelli and Pascal and Lavoisier, Maupertuis and Lagrange, and a host of other hardly less illustrious names, but the omission is characteristic of the writer’s time. The critics of the next generation naturally recoil from this fond exaggeration. A mathematician like De Morgan, really familiar alike with logic, with the text of Newton, and with the history of science — all subjects in which Macaulay was not truly at home — has no difficulty in showing that Macaulay misstates his facts, that the theorems of the Principia were not, and could not have been, reached by the induction described in the Novum Organum, and that Macaulay’s own conception of Bacon’s methods is so loose that, when he proceeds to give an example of their value, he gives us one in which they are misapplied to bring out an unwarranted conclusion. Later critics with their own minds full of the Critique of Pure Reason find it still easier to observe that Bacon leaves the fundamental questions of the metaphysician and the “epistemologist” untouched; we can understand why they go on to draw the conclusion that a writer with so little to say on the issues most interesting to themselves can only be called a “philosopher” by a stretch of courtesy. But it is for us to beware that in anxiety to avoid undue glorification we do not fall into undeserved detraction. Metaphysics and the theory of knowledge are not the only fields in which there is work for philosophy to do, and it is well for philosophical students in this country, after fifty years of preoccupation with the problems raised by Kant and |265| his continuators, to rediscover the external physical order itself as a fitting object of philosophical examination. Our professed philosophers have long enough been engrossed with the contemplation of mind; they may profitably go to school once more to the men of the seventeenth century whose most direct concern was with the world of natural objects, and it may well be that Bacon, in particular, has a valuable lesson to teach.

In my own judgement he has, and if we are blinded to his real performance we are blinded by the very largeness and spaciousness of the design. Like others among the first-rate men of his marvelous century, Bacon has still before him the philosophical ideal — a legacy from the magnificent audacity of the great scholastics — of a single unified science embracing, in a single survey, and by a single method, the whole field of the knowable. We have been accustomed for more than a century to think of him chiefly in terms of that part of his vast project most fully represented by accomplished work, as the author of a treatise on scientific method, or, more narrowly still, as the author of a theory of the process of inductive generalisation. But it is necessary to remember that, from his own point of view, the elaboration of a new scientific method was only one part of the gigantic task of the revivification of science. Equally important were to be the survey of the domain of science and estimate of its derelict territories (only represented in Bacon’s actual output by the De Augments Scientiarum, an expanded Latin version of the Two Books of the Advancement of Learning, which had been originally designed independently of the main work on the Interpretation of Nature) and the improved Natural History, the great repertory of carefully ascertained and documented facts, for the actual |266| execution of which Bacon never succeeded in accomplishing anything of moment. If he could appear to plead his own claims on our grateful remembrance, he would certainly ask to be judged as the author, not of the Novum Organum, but of the Great Instauration, in which the theory of induction is but one part among others, not as the mere logician, but as the man who had “taken all knowledge for his province.”

When we judge Bacon from this point of view, many of the condemnations which the critics of the last half-century have passed on the defects of his theory of scientific generalisation will be seen to be of only secondary importance. The truth of the strictures has, indeed, often to be granted. We cannot deny, and Bacon himself often confesses, that his statements about the actual facts of nature are often gravely vitiated by a large admixture of falsehood due to acceptance from tradition of alleged facts for which there is no adequate testimony from known and competent observers — reliance on experientia vaga in the absence of a proper experientia literata. The reasonable attitude in such cases is not that of the detractors who condemn the men of the seventeenth century because they were not in a position to reject peremptorily the reports of an Aristotle, a Pliny, or an Albert the Great. It is more reasonable, as well as kinder, to recognise the open-mindedness which made Bacon and his fellows slow to dismiss as mere fable reports of strange things found in the best repertories of facts to which they had access, and to give them the fullest credit for their insistence that care must be taken for the future that all alleged observations shall be documented with a rigour unknown to earlier ages. Had Bacon and his contemporaries simply disregarded all accounts of strange things to be found in earlier authorities on the score of |267| insufficient documentation, some ancient delusions would have died less hard; so far there would have been gain to science. But, I take it, the gain would have been more than offset by loss, if all records of facts puzzling to the seventeenth-century English or French or Italian student had been summarily rejected on such grounds. (History would suffer in the same fashion from the summary rejection of everything for which there is not fully convincing contemporary documentary or inscriptional evidence.) The true scientific method is rather that professed by Bacon, provisional acceptance of what appears to have a body of respectable testimony behind it, even though the testimony has never been subjected to the severe controls of the laboratory, coupled with the determination that these severe tests shall be regularly applied for the future. We may fairly hope that, by this method, the experientia literata of the present may be made gradually to correct the experientia vaga of the past without any incurring of the very real risks attendant on sweeping scepticism. The errors which will be retained will only be retained for a time; radical scepticism about all observation not made under “laboratory” conditions would entail error of a much more widespread and durable kind.

So again, we have to admit the truth of the main unfavourable criticisms of De Morgan and Jevons on Bacon’s “induction.” It is manifest that Bacon was not — as so busy a man could hardly be — abreast of the greatest discoveries of his own age, and that he had no inkling of the truth stated by Galileo when he said that the script in which the book of nature is written is that of geometrical symbolism; that geometry is, so to say, the accidence of nature. There is no gainsaying De Morgan’s pithy contrast between Bacon’s and Newton’s views on the place of observation in natural science. |268| The great use of observations is to test theories, not to furnish them.

De Morgan is in the right of it when he criticises Bacon’s picture of the rapid progress of science in his imagined Atlantis (where one body of specialists are occupied for life in making endless random observations and experiments, which are then handed, over to an entirely different group, who are to elicit theories from this mass of miscellaneous observations), by remarking on the fewness of the actual experiments and observations which play any important part in establishing the results of the Principia. But defects like these, though they might be serious if we had to estimate Bacon solely by his contribution to the theory of scientific generalisation, become less considerable when we take the whole scheme of the Instauration into account and avoid the easily committed mistake of judging the projected whole by the part its author happens to have brought nearest completion. This is a mistake committed, as it seems to me, equally often by the eulogists and the critics of Bacon. The eulogists select for magnification just those features of Bacon’s doctrine which are most open to objection — his curious faith in the fertility of random uncontrolled experimentation, his rejection of the “anticipation of nature” by the method of hypothesis, his apparent conviction that mere empiricism, uncontrolled by “regulative ideas,” can be the foundation of assured and continuous scientific advance. The critics, alive to the vulnerability of all these positions, assume that to dispose of them is to dispose of Bacon’s claim to be remembered as a thinker of permanent significance.

Both parties might have been saved from misunderstandings if they had not apparently both begun and finished their study of Bacon with the Novum Organum. |269| It is not the least among the many excellences of the monograph of Mr. Levi that he follows a different and more promising path. The first three-quarters of his book are given to a full and careful study of Bacon’s conceptions about the scope of science, the classification of the sciences, and the general character of the “scientific view of the world”; it is only in the last quarter of the book that the author proceeds to examine the Novum Organum and its theory of generalisation. This is the philosophically right way by which to approach Bacon and his doctrine; when we adopt it, we discover that the real significance of Bacon for the development of European thought is independent of the merits and defects of his particular views on induction. We also win the right point of view from which to appreciate the special Baconian theory of induction itself, since we approach the method, as a reader who begins his Baconian studies with the Novum Organum cannot do, possessed of a real understanding of the question to which Bacon expected induction to provide the answer. Hitherto, as Mr. Levi says, the verdicts of both admirers and hostile critics of the “Baconian method” have nearly always been vitiated by the simple fact that the problem Bacon expected science to solve is different in character from the problems it has historically been the glory of modern natural science to have solved. He has received high praise on the ground that, as has been alleged, his method enables us to answer a question which he was not raising, and has been depreciated on the ground that the true method for dealing with this question is not that which he recommends. Until we are quite clear on the nature of the fundamental scientific problem as Bacon conceived it, both laudation and depreciation are premature.

Before we can either praise or depreciate Bacon as a |270| thinker with understanding, we need to put to ourselves the question which Mr. Levi has, with true philosophical insight, made primary for his discussion of the Instauration. On which side of the great dividing line drawn in science by the seventeenth century does Bacon fall? Is his right place with the latest of the men of genius who draw their inspiration from the mainly humanistic ferment of the Italian Renaissance? Or should he rather be counted among the earliest of the creators of the “modern scientific” conception of nature; the way of looking at the world about us which is at bottom common to Hobbes, Descartes, Spinoza, Newton, seemed a century ago to have been definitely established by Lagrange and Laplace, and is only now beginning to be severely criticised from within by the physicists themselves? On the right answer to this question, whether Bacon is the last and most eloquent of the men of the Renaissance or the earliest of the moderns, will depend our view of the character of the supreme scientific problem as Bacon understood it, and consequently, also, our view of the worth of his famous inductive method.

Naturally, I cannot undertake to reproduce here the substance of the elaborate argument by which Mr. Levi reaches his answer to this question; I can only express my conviction that the answer he reaches is the right one. Bacon is definitely among the first of the “moderns,” not among the last of the mediaevalists. The main proof of this is to be found in his insistence on the discovery of what he calls “forms” of “simple natures” as the last and highest problem for a true science. The natural world, as he sees it, appears to be a complex of bodies, each exhibiting a plurality of distinct sensible qualities, a colour, a smell, a taste, a specific hardness, a specific density, and the like. Each of these qualities |271| is what Bacon calls a “nature,” and his conviction is that the number of independent simple “natures” is strictly limited; they are perhaps not more numerous than the letters of an alphabet.14 What we call “things,” in their apparently infinite variety, are but the words of a language of nature, all reducible to arrangements of this small number of primary vocables. The problem of discovering the “form” of a simple “nature,” assigned by Bacon to that highest exercise of the scientific intellect for which he appropriates the name “metaphysic,” is then declared to be the problem of finding a “nature which is convertible with the given nature, present wherever the given nature is present, absent wherever it is absent, and is, at the same time, a specification of a more generic nature.”15 Thus to discover, for example, the “form” of whiteness is to discover a character which is always found where white colour is found, never found where white colour is absent,16 and is, moreover, a specific determination of some character more generic than itself. A further light is thrown on the real meaning of this conception of a “form” by Bacon’s famous and characteristic doctrine of the strict correspondence of knowledge and power. The “form” must be of such a kind that to know what it is is also to know how to produce the corresponding “nature” without limitation, wherever its production is physically possible. To know the “form” of white is, in Bacon’s view, equivalent to knowing how to produce a white colour in anything capable of exhibiting colour, that is, wherever there is a surface which can have any colour at all.17 The vast range of the transformations which knowledge of the “forms of simple natures” |272| would render feasible explains why Bacon chose for the practical art of the successive production of such transformation the name magic. It explains also why he held that knowledge of the comparatively few ultimate “forms” of such “natures” would enable its possessor to change the whole complex of the sensible qualities of a body, by superinducing one fresh “nature“” after another, so that the transmutation of metals dreamed of by the alchemists would be among the everyday operations of a properly informed and equipped “applied science.”18

If we remember that Bacon has also laid it down that all human skill can really effect in nature is to displace bodies, to move them to and from one another,19 we see at once — as, in fact, was already suggested by the conception of a “form” as a specification of a character more “generic” than itself — that the ideal Bacon has before his mind, though he has not realised the supreme position to which that ideal elevates mathematics as the key to science, is the correlating of all qualities in nature with configurations, or kinematical patterns. Clearly, if all the events of nature are to be expressible as syllables of one and the same language, the letters which compose the syllables cannot be colours or sounds or odours; the common measure of all natural events must be sought in characters which are themselves common to all events. This means that the common measure must be looked for in properties directly arising from the fact that all events fill a certain volume and duration and are bounded, spatially and temporally, in various ways by other events. When events have been, so to say, translated from the vernacular of the eye, the ear, the nostrils, the palate, the skin, into |273| nature’s “Latin,” they must all appear as space-time patterns. This is the thought which underlies the development from Galileo, through Descartes, Newton, Lagrange, and many another, to the full-blown “mechanical theory of nature” of our own days, though it is only of late that we have possessed a philosophical terminology which allows us to formulate the thought with precision. Accurately stated, it is the doctrine named by Dr. Whitehead20 that of “simple location.” Mr. Levi’s careful study makes it abundantly clear that Bacon held this doctrine no less than Galileo, though the lack of mathematical training, for which not Bacon but the Cambridge of his youth was to blame, prevented him from giving the thought the same lapidary expression. It is no accident that the one example Bacon has given us of the process of search for a “form” ends by the divination that the “form” of heat is a particular kinematical pattern, a specific type of motion in what we should call the molecules of a body,21 or that he should have more than once hinted that all the wealth of colours, sounds, fragrances, and other sensible characters of the world of daily life are a mental addition of our own to the bald reality,22 which consists of “indivisible bodies exhibiting motions conformable to a law,” and again, that the “forms” of the various “natures” are no more than the several clauses or sections of this law.23 Patterns of rhythmic movement across space and time the sole reality, everything else a mass of “psychic additions” embroidered on this monotonous ground by the human senses and imagination: these are the fundamental positions characteristic of the “mechanical” interpretation of nature |274| wherever it makes its appearance in man’s thought, whether in Democritus, in Lucretius, in Galileo, in Newton, or in our modern positivists of science. Bacon’s acceptance of these positions clearly stamps him as belonging to the beginnings of modern “naturalism,” not to the last stage of the confused “vitalism” characteristic of the men of the Renaissance, who were trying to move back from the elaborate qualitative physics of Aristotle and the Aristotelians to the cruder picture-thinking of the old pre-Socratic fathers of science.

The “mechanical” view of nature has yet several steps to take before its full implications will become clear. Before we can advance from Bacon’s position to Newton’s, Galileo and Descartes will first have to enunciate the programme of the geometrising of nature explicitly. Even the one vestige of an extra-geometrical reality which persists in Newton, the conception of “gravitational mass” as an ultimate property of his particles, with its consequence, the appearance of gravitation as an independent outstanding fact irreducible to the general laws of motion, will have to disappear, as it does in the kinematics of the “general theory of relativity,” before the complete reduction of physical science to geometry is accomplished. But Bacon has already taken the first step on the path to this goal; the taking of the rest is a mere matter of logical consequence. And yet, when we come to think of it, with all his lack of insight into the full consequences of his own assumptions, Bacon has kept true to the reality of things in a way which makes him, rather than the great physicists who were to succeed him, the most instructive guide for ourselves at the present moment. Beautiful and consistent as the “classical” kinematical interpretation of natural processes may be, needed as it was for the development of physics and chemistry, |275| it has just one defect, but one which is fatal to it as a philosophical account of the real. As Dr. Whitehead says,24 it is simply unbelievable; the more faithfully we keep to it, the farther away shall we get from anything we can regard as the final truth about the world. How unbelievable it is we shall see, if we will make two reflections.

(1) The kinematical view, taken seriously as a full statement of the truth, absolutely demands as an immediate consequence the doctrine, only hinted at by Bacon, but expressly avowed by Galileo25 and Descartes, that the sensible qualities of bodies, all, that is, which gives nature her apparent inexhaustible wealth and variety, all that is of primary importance for the life of animal organisms, must be simply unreal, fabricated by the mind as an unauthorised comment on nature’s text. Fully thought out, this must mean that the colour of the sapphire, the scent of the rose, and the like, belong, not to the sapphire or the rose, but to the mind of the sentient perceiver. It is the notorious fact that, throughout the animal creation, it is these characters, colours, sounds, smells, And the like, which count in the life of the organism. By their means creatures are attracted to their mates, directed to their prey, warned of the approach of their enemies; yet, on the theory, the characters which play this all-important part in the life of animated nature are purely illusory. The behaviour of the animal organism to its environment is based on awareness of something which is not there. In the nineteenth century it was fashionable to disguise the paradox by misstatement. The eye, it was said, apprehends vibrations “as” colours, the ear apprehends them “as” sounds, and so forth. The equivocation |276| should be too patent to need exposure. It should be clear that the eye apprehends no vibrations at all; it sees only colours: the ear hears not vibrations, but tones. If then, all that is real in the object apprehended is vibratory motion, it follows by consequence that sense-awareness is throughout awareness of things as they are not, and this should be incredible. For, in the last resort, it is from the very senses the argument would disable that we have learned of what it declares to be the only realities, shape, volume, situation, movement. An argument which is to discredit the senses as revelatory of the characters of real objects should make the characters treated by the “mechanical” theories as the key to the truth about nature as much of an illusion as all the rest. We only escape the admission by adroit juggling with words.

(2) For much the same reasons, the theory is doomed to fail us when we advance from the sciences of the in- organic to the sciences which study the manifestations of life itself. If the clue to the whole interpretation of nature is the reduction of all processes to changes of configurations of systems of given units which do not themselves change — whether we call the units “hard impenetrable corpuscles” or “point-events” makes no real difference — then, since sentient organisms and rational human persons are themselves components of the “kingdom of nature,” it is necessary that the explanation should work for organisms and persons, no less than for everything else. We are thus committed to all the difficulties which attend the attempt to work out a consistently “mechanistic” scheme in biology. These difficulties are sometimes met by the biologist with a retort which ignores the real problem. The defender of “mechanism” will point triumphantly to recent successes in the “synthetic” production of compounds |277| once believed to be obtainable only by the agency of living organisms, as proof of the needlessness of “vitalistic” hypotheses. As a defence of “mechanism” the retort misses its mark. On the face of it, the successful production of urea, for example, in a laboratory affords no adequate ground for expecting that the world will ever see a laboratory-made man, or even a laboratory-made rabbit. But the real reply goes deeper than this. The very plausibility of the general “mechanical view” of nature rests all along on the tacit presupposition that there remains something outside and independent of the nature which has been identified with kinematical configurations, viz. the sentient creatures which apprehend the configurations, and in apprehending them perform the miracle of translating them into colours, sounds, smells, savours. Without the sentient as an independent term the sensible world, in which we all move and have our being, could not even be an illusion, for there would be no one and nothing to be illuded. Either there is something real, not included in the nature which has been reduced to patterns of movement (viz. the sentient aware of the patterns), and then the reduction to kinematics is not applicable to nature as a whole; or else the sentient is, like everything else, a kinematical pattern and nothing more, and the sensible world cannot be so much as a misapprehension.

Since we cannot well deny the fact that there is sentience and a sensible world, we are thrown back on the former of these alternatives. The whole of nature, the whole of what there is for our observation, cannot be reduced to an intricate kinematical pattern. The living and sentient, at least, refuses to submit to the treatment, and the more we learn of the impossibility of drawing a hard and fast line between the sentient and |278| the merely animate, or again, between the animated and the inanimate, the more irresistibly we are driven to the conclusion that the patterns of a mathematical physics cannot be a final and complete account of the reality of any natural process, that it is the sciences of the organic to which we must go for our very interpretation of the patterns of the inorganic domain itself. This explains why a philosopher like Dr. Whitehead finds himself attracted by the comparative fluidity of Bacon’s thought and, in particular, why he has singled out a striking passage of the Silva Silvarum, where Bacon is anticipating the fundamental conceptions of a greater man still, Leibniz, as significant for our own age, on which the task of purging natural science from bad “materialistic” metaphysics is so urgently laid. “It is certain,” Bacon says, “that all bodies whatsoever, though they have no sense, yet they have perception: for when one body is applied to another, there is a kind of election to embrace that which is agreeable, and to exclude or expel that which is ingrate: and whether the body be alterant or altered, evermore a perception precedeth operation; for else all bodies would be alike one to another. And sometimes this perception, in some kind of bodies, is far more subtile than the sense; so that the sense is but a dull thing in comparison of it: we see a weather-glass will find the least difference of the weather in heat or cold, when men find it not …. It is therefore a subject of a very noble inquiry, to inquire of the more subtile perceptions; for it is another key to open nature, as well as the sense; and sometimes better.”26

Bacon is here preluding to Leibniz’s well-known doctrine of the reflection in the state of each “monad” at any moment of the simultaneous states of all other |279| “monads” and to Lotze’s contention that the interaction of two things implies that either “takes note of” the state of the other. From the rigidly “mechanical” point of view, all such language, implying as it does that the selective activity of the organism, so fundamental in biology, is only a special case of a character common to all natural agents, would have to be dismissed as a mere “anthropomorphic” metaphor. Yet we know very well that selection is something much more than a metaphor in the organic realm, and our inability to find a real lower boundary of that realm ought to make us careful of declaring the word to be a mere metaphor anywhere. It seems plain that we cannot escape Bacon’s shrewd conclusion that if selection were, in any domain of reality, only a metaphor, within that realm all bodies would be “alike” one to another. Historically, the remark may be accounted for by the consideration that the animation of nature at large was a common conviction of the Renaissance age, and that Bacon has not clearly seen how inconsistent the conviction is with the “mechanical” conceptions clearly foreshadowed by his own utterances. But precisely because he has not wholly committed himself to that radically erroneous identification of reality with a kinematical pattern which was to dominate the unconscious metaphysics of post-Newtonian men of science, his thought has a living significance which that of the reckless post-Newtonians has not. He is happily in the right where they were unhappily in the wrong; he sees, as we are learning to see, that a true philosophy of nature must be sought by trying to follow the categories of the organic sciences downward into the realm of the inorganic, not by treating categories which only suffice for the formulation of laws of the inorganic as a sole and sufficient clue to the structure of a unity which

|280| We might state the flaw in the “mechanical” theory once more in a more precise form: to do so will afford a natural transition to the brief consideration of Bacon’s characteristic doctrine of logical method with which I shall bring these remarks to an end. When the “minute philosopher” denies that selection is a universal feature of natural process, he is in effect maintaining that, in the domain studied by physics, the internal state or structure of anything which is a genuine unit is independent of its transactions with other units. Modification of the internal characters of A and B in virtue of the transactions between them is relegated to the organic realm, where, as we all know, or think we know, the transactions between organism and environment are correlated with profound modifications of the internal character of both. In the inorganic domain the component units of any system are, on the theory, to run through the whole series of their transactions with one another without any internal change, remaining all through simply self-same. What exhibits internal modification is to reveal itself by the very fact as not a real unit. The conception naturally took time to develop. Even Newton could speculate about what he called “old, worn” particles. But in the full-blown “mechanical” theory all such conceptions have disappeared; there are no “old, worn” atoms in the nineteenth-century kinetic theory of gases, and it is interesting to note that Sir John Herschel27 actually came very near accepting what Bacon means to treat as the impossible paradox of the complete “likeness” of bodies. He makes it an argument for Theism that one atom of a given chemical element is so exact a replica of any other that |281| atoms must be regarded as turned out from a divine factory. Such conclusions are direct consequences of the assumption that the “classical” mechanics is the universal clue to the structure of nature, since in mechanics motions, the only processes taken into account, are regarded as persisting unaltered and merely compounded with one another. This was the real reason why Leibniz rejected atomism as the last word about the constitution of nature. If it were the last word, he argued, sufficient knowledge of the present motion of a single atom should enable us to reconstitute the whole history of the physical world, whereas, in fact, there are always a countless plurality of alternative routes through space and time by which the atom might have reached its present state.28 That is, the kinematical states of the unit constituents of nature, as conceived by the “mechanical” theorist, are too wholly external to the constituents to admit of our thinking of the unity of nature as a whole as more than a mere resultant of transactions between the members of the aggregate of atoms. But such a unity of nature as a whole as is implied by a sound philosophy of the sciences must be something less superficial; it must be a real pervasive character, not an Aggregatzustand.

Leibniz thus reproduces Bacon’s thought when he requires us to substitute for the atom, as the genuine unit in things, the monad, which has an internal principle of development and is, at every moment, charged with its whole past and pregnant with its whole future. Unfortunately, Leibniz felt bound to make the development of the individual monad wholly self-contained, so that it is only its own past and future with which each is charged. Each behaves as though it were a universe to itself. Genuine interaction is thus replaced by the creative |282| activity of God, who has fashioned the monads in such a way that their various developments keep one tune and time. How, on this theory, Leibniz could so much as have come to suspect that there is anything in the universe beyond himself is a difficulty he never quite succeeded in explaining. Bacon’s thought is much more truly reproduced by Dr. Whitehead when he conceives of each of the monads of his own philosophy as weighted by the past and big with the future not only of itself, but of each and all the rest. The unity of pattern in nature thus gets adequate recognition as dominant of all partial sub-patterns. As Dr. Whitehead puts it,29 it is true that the living body may be regarded as a highly complex pattern of electrons and that electrons of the same kind are to be found in the lifeless bodies surrounding it; but the behaviour of an electron which enters into the pattern of a living body is differentiated by the fundamental fact that it belongs to that pattern and not to any other. In this way the nightmare of “epiphenomenalism,” which was threatening fifty years ago to make the selective action of the living organism superfluous, disappears, as the relation of organism to environment is made the clue for understanding the relation of the monad, the particle, or the electron, to its surroundings. Dr. Whitehead has worked out the thought with a power which is his own, but the germ of it is to be found, as he himself insists, in the passage already cited from Bacon. For the immediate future of scientific thought it was of the first importance that physics and chemistry should be developed as sciences of number, weight, and measure; relatively, therefore, the “materialistic” concepts which served as admirable instruments for the immediate purposes of physicist and chemist were entitled to their temporary |283| domination over the scientific mind. The developments of the biological and historical sciences since the middle of the nineteenth century make it necessary that our own age should revise its fundamental scientific conceptions by removing from them limitations imposed in the special interest of the extreme abstractions of chemistry and physics, and we may well pay our tribute of gratitude to the far-reaching vision of the man who so long ago gave the clearest hints of the line that revision will need to take.

Bacon’s peculiar logical theory, like his philosophy as a whole, has been both unintelligently belauded and unintelligently decried. In the days of laudation, Macaulay could persuade himself not only that Bacon, alone among logicians, had made a correct analysis of the process of induction, but that he had given a universal rule for sound reasoning, that Baconian induction is the one and only method of human thinking. The statement might have surprised Bacon, who is never tired of reiterating the complaint that the true method of science is something as yet unknown and undreamed of, and that his account of it is likely to be discredited precisely by its novelty. In the reaction from uncritical laudation, naturally enough, the Baconian rules were depreciated almost to the point of denying them all worth.

It had been foolishly asserted that the results of Newton’s Principia are illustrations of the power of the method. Men like De Morgan and Jevons, who knew what the methods of the Principia are, were naturally quick to point out that however Newton reached his conclusions, he did not reach them by any process like the construction of tables of presence, absence, and degree, and the routine performance of eliminations described in the second book of the Novum Organum, |284| and they were not slow to infer that, since Bacon’s method would not have served Newton’s purpose, it serves no purpose of any particular value. The true state of the case is different. As Mr. Levi has said, the tragedy of Bacon’s method is that it does provide solutions for a certain kind of question, but unfortunately not for the kind of question Bacon was most anxious to solve. We should therefore be judging too favourably of the Baconian induction by elimination if we were content to remark that it is in principle the method afterwards called by Mill the “Joint Method of Agreement and Difference,” and that we all know the usefulness of that method in biological and sociological inquiry. For Mill’s conception of the scientific problem itself was radically different from Bacon’s. Mill thought it the typical business of science to establish laws of uniform sequence between observed events, on which we may rely with sufficient confidence for practice. From Bacon’s point of view this is a task which science can execute, but it is not the main task. The supreme problem is to discover the “forms” of “simple natures.” Mill is satisfied with knowing the order of sequence between events; Bacon is asking after the formal structure of the events themselves. Unhappily his method of induction by elimination, though it will often give a reasonably adequate answer to Mill’s question, is quite unsuitable for the solution of his own.

We can illustrate this from any of the familiar examples used in text-books to exhibit the working of the method. Thus, let the question under investigation be concerned with the outbreak of a local epidemic. If we find that the houses which have been visited by the disorder, though they present no other discoverable feature in common, all get their milk supply from a particular dairy, while other houses which have not suffered seem |285| to have nothing much in common but the fact that they do not get their milk from this dairy, we shall be reasonably safe in suspecting that the milk of the dairy in question is the carrier of the disease. But even if the investigation should enable us to lay our hand on the actual offending cow, we should be no nearer knowing what Bacon would call the “form” of the disorder, the specific derangement of physiological function which is to blame for the symptoms. The “form” could only be identified properly on the strength of a correct theory of the infra-molecular physics of the human organism, and it is idle to imagine that such a theory can be elicited by comparison and elimination directly from facts taken over unanalysed from inspection through the senses, even if the senses have been fortified by the most elaborate instruments of precision. No procedure of mere generalisation from observed data could conduct us to the knowledge of the intimate structure of the patterns of nature called by Bacon knowledge of “forms,” and the advantage claimed by Bacon for his “induction,” that it functions mechanically and quasi-infallibly, thus making the interpretation of nature a matter for mere plodding industry, is therefore imaginary. So long as we are content, as on the whole Mill was, with mere detection of regularities of sequence among events directly accessible to observation, it is perhaps possible to approximate to this substitution of mere industry for insight. If a sufficient number of painstaking observers were employed in the construction of “tables” of presence, absence, and degree, it is likely that the tables would be reasonably full; mere accuracy in the recording of observations is a virtue which can be inculcated on the dull and laborious.

The confrontation of the “tables” with a view to affecting eliminations, again, need not call for much |286| more than industry. So it is not surprising that the mere formulation of laws of sequence is, in actual fact, often accomplished by very commonplace persons. If the work of science went no farther than the detection of such regularities, it would have been no extravagant anticipation to hope, with Bacon, that all men, once provided with instruments of observation and taught rules for the elimination of the irrelevant factors from a problem, might be equipped fairly equally for the work of the scientific man. If science were the simple business it is made to look in those chapters of Mill’s Logic which present it as a mere matter of conforming to the four “inductive methods” — themselves merely slightly differing forms of Bacon’s exclusiva — there would be no good reason why all of us, except the “feeble-minded,” should not make ourselves a name in science. The trouble is that authentic science is so much more than this. If you have the true scientific spirit, you will not fancy you have penetrated very far into nature when you have merely qualified yourself to say, “As a general rule and where there are no interfering factors, A will be followed by B.” You will not be content until you see definitely why the “regular sequence” should he A–B and not A–C or A–D. To know so much, you need to have discerned the pattern of the event A, that of B, and that of the more complex event (A–B), and to see how the patterns of A and B demand, while those of A and C or A and D refuse, integration into a larger pattern.

Again, you will not even be content with a general scheme of this type. In actual fact the patterns A and B never occur in their typical abstract generality; we are confronted with a specific α and β which fall under the types of A and B, but have their own irreducible concreteness. And similarly α and β are not integrated |287| into the mere complex typical ( A–B), but into an (α–β) with its specific concreteness, and you want to account for just those concretions. Since it seems to be involved in the thoroughgoing unity of the world-process miscalled the “uniformity of nature” that the special concretion of the actual event is what it is because its total setting in the rest of “nature at the moment” is what it is, we readily understand why Dr. Whitehead should comment on Bacon’s method with the remarks that “induction has proved to be a somewhat more complex process than Bacon anticipated. He had in his mind the belief that with a sufficient care in the collection of instances the general law would stand out of itself. We know now, and probably Harvey knew then, that this is a very inadequate account…. I do not hold induction to be in its essence the derivation of general laws. It is the divination of a particular future from the known characteristics of a particular past…. Inductive reasoning proceeds from the particular occasion to the particular community of occasions, and from the particular community to relations between particular occasions within that community.”30 In other words, it is not generalisation, but the interpretation of the individual in terms of concepts transparent to the intellect, that is the never completely realised goal of a science awake to its task.

This is a just criticism of Bacon’s method of elimination, but it equally hits the weak point in the positivistic conception of science championed by Avenarius, Mach, and Pearson, as a mere labour-saving trick of summing up the “routine of our perceptions” in compact formulae. What we really ask of science is not to abbreviate the “routine” but to make it intelligible. For that reason, I should not myself regard Bacon’s |288| inadequate appreciation of the importance of precise quantitative determinations in science as the fundamental defect of his doctrine. It is true, to quote Dr. Whitehead once more, that it is just this which is the most readily recognisable difference in “tonality” of mind between Bacon and Galileo or Newton,31 and the characteristic he most conspicuously shares with the Aristotelianism he was anxious to dethrone. But we think inadequately of the mathematical method which has proved to be the true key to physical science if we make quantitative precision its primary recommendation. Mathematics, as Descartes long ago said, and as the modern creation of rigorously pure mathematics has abundantly proved, has not necessarily to do with quantity. What is distinctive of the mathematical sciences is not that they are concerned with quantity, but that they are in a special degree transparent to the intellect. In mathematical thought we deal, more truly than anywhere else, with “ideas” which — to use Hume’s terminology — are not merely “conjoined,” but “connected.” It is this substitution of connection for bare conjunction without “rhyme or reason” that a mathematical foundation brings into our study of the course of natural events. We can, indeed, never hope to see conjunction completely transformed into connection; the element of bare given fact never quite disappears from a science of nature, for the simple reason that the nature we are trying to understand is primarily given to our minds, not created by them. But it is the supreme task of science, as it advances, to reduce the element of mere “conjunction” in our apprehension of nature, “without limit.” The grave fault of Bacon’s proposed inductive method, as of all merely inductive methods, considered as an instrument of |289| science, is that it could do nothing to effect this reduction. However often and successfully we might employ it, it would leave us everywhere with conjunction not converted into connection on our hands. It might do much to make our forecasts of events more probable, and so to serve our practical utilitarian ends, but it would not advance our understanding of the world. That is, it would be useless for the discovery of “forms.”

Had Bacon been the mere vulgar utilitarian he has sometimes been taken to be, the perception of this defect, supposing it had been possible to him, need have caused him no concern. From the merely utilitarian point of view, we should praise Bacon, if we praised him at all, for the very limitations of his method, and regret that he retained so much of the old metaphysical leaven as even to raise the problem about the discovery of “forms.” But in truth it is in this other strain in his thought, the recognition of the discovery of “forms” as the true problem of the sciences and the identification of “forms,” so far as the physical sciences are concerned, with space-time patterns, that his real significance for living thought must be found. If the labours of Dr. Whitehead and other eminent thinkers who are engaged on the work of restating the first principles of natural knowledge in a form free from “materialistic” metaphysical assumptions should be crowned with success, it is not hard to name the great men of the past to whom the natural philosophers of 1950 will be found looking back as their intellectual ancestors. Foremost on the list of the forerunners of a philosophy of nature at once organic and mathematical should stand the names of the two great mathematical metaphysicians of the ancient and the modern world, Plato and Leibniz. Between them |290| as a connecting link, for all his personal want of mathematical equipment, might well stand Bacon, who found in Plato an anticipator of himself,32 as he was in turn recognised with generous appreciation by Leibniz.33 To be remembered as one, even were it the least, of such a triad would have been, I believe, glory enough to satisfy Bacon. Whatever were the failings of his character, and it is my business neither to exaggerate nor to extenuate them, excessive self-conceit does not seem to have been among them. Of himself we may well believe he would have wished us to say what he has said of the great men of an earlier time: “So let great authors have their due, as that Time, which is the author of authors, be not deprived of his due, which is, further and further to discover truth.”34


1^April 9 (O.S.), 1626.

2^ Il Pensiero di Francesco Bacone consider ato in relazione con le filosofie della natura del Rinascimento e col razionalismo cartesiano. Turin, 1925, Paravia & Co.

3^ Pyth. i. 87 εἴ τι καὶ φλαῦρον παραιθύσσει, μέγα τοι φέρεται|πὰρ σέθεν.

4^ Formal Logic 1, 216-24; Budget of Paradoxes (ed. 1915), i. 76-90.

5^ Principles of Science, 581-6.

6^ Works, i. 69 (from a letter of January 1880).

7^ J. T. Merz, History of European Thought in the Nineteenth Century, vol. i. chaps, iv., vi.

8^ We may recall the way in which the conception is made central for natural science in the Metaphysische Anfangsgründe der Naturwissenschaft of the convinced Newtonian, Kant.

9^ Science and the Modern World, chap. i.

10^ This explains why Leibniz complained that Newton, by appealing to gravity as an unexplained property of “matter,” was, in effect, reviving the belief in “occult qualities”: cf. Nouveaux Essais, Avant-Propos, p. 203 (Erdmann).

11^ É. Meyerson, La Déduction relativiste, 284-6; M. Schlick, Space and Time (E. Tr.), 44.

12^ The premises of geometry are, indeed, not self-evident. But it is clear that Descartes, from long familiarity with them, supposed them to be so. We may be sure that Euclid’s parallel-postulate was taken by him to be “evident by the natural light.”

13^ Essay on Horace Walpole: “the great discoveries in physics, in metaphysics, in political science, are ours. But scarcely any foreign nation except France has received them from us by direct communication.”

14^ De Augmentis, iii. 4 (Works, Ellis and Spedding, i. 566).

15^ Novum Organon ii. 4.

16^ Ibid.; De Aug. iii. 4 (E. and S. i. 566).

17^ Novum Organon ii. 5; De Aug. iii. 4 (E, and S, i. 568).

18^ Novum Organon i. 85.

19^ Novum Organon ii. 4.

20^ Science and the Modern World, 71-73, 84, 98.

21^ Novum Organon ii. 20.

22^ Valerius Terminus, ii. (E. and S. iii, 235); Novum Organon i. 41,

23^ Novum Organon i, 51, ii. 2.

24^ Science and the Modern World, 80.

25^ Galileo, Saggiatore (Opere, iv. 332); Descartes, Meditatio vi., Principia, i. 70.

26^ Silva Silvarum, cent, ix ad init. (E. and S. ii. 602).

27^ Quoted in Ward, Naturalism and Agnosticism, i. 100.

28^ Opuscules et fragments inédits, ed. Couturat, 522.

29^ Science and the Modern World, 215.

30^ Science and the Modern World, 63.

31^ Op. cit. 66.

32^ Novum Organon i. 105; De Aug. iii. 4 (E. and S. i. 565); Advancement, ii. (E. and S. iii. 355), Cogitata et Visa (E. and S. iii. 662).

33^ Confessio Naturae contra Atheistas, Erdmann, p. 45, “divini ingeni vir Franciscus Baconus”; De Stilo Nizolii, ib. 61, “incomparabilis Verulamius”; In specimina Pacidii introductio, ib. 91, “feliciter accidit ut consilia magni viri Francisci Baconi, Angliae Cancellarii, de augmentis scientiarum ad manus adolescentis pervenirent.”

34^ Advancement of Learning I. (E. and S. iii. 290).

Top ↑

VIII. Some Features of Butler’s Ethics

This paper was originally presented in 1926.

We are told on the best of all authorities that the last place where a prophet must look to be honoured is his own land. Whether the fame of {Joseph} Butler may be considered as illustrating this adage is a difficult question. From our great-grandfathers and grandfathers he certainly received admiration enough, but it is not so clear that their descendants of today rate him at anything like his real worth. To my own mind, at least, there is something niggling about a great deal of the criticism it has long been fashionable to bestow on the “philosopher of Anglicanism.” The explanation is presumably to be found in the domination of so much of the best philosophical work of the last generation by influences from Germany. When the ardent youth of our Universities were being taught, with abundance of persuasive eloquence, that in philosophy Kant and Hegel had made all things new, men were not likely to see much merit in the thought of the “home-keeping” wits of the eighteenth century, even if they condescended to study it at all, except in so far as, as in the case of Hume, some knowledge of it was necessary for the understanding of Kant. When the full history of the Germanic influence on our nineteenth-century literature comes to be written in a thoroughly critical spirit, a great deal will have to be done over the whole field in the way of rehabilitating reputations which have been unduly depreciated from |292| this cause; it will then, I believe, be acknowledged that one of the more unfortunate effects of the intellectual conquest of Britain by Germany has been the curious neglect of the very rich and valuable ethical literature which begins with Cumberland and Cudworth (or perhaps, taking dates of publication into account, one should say Samuel Clarke) and culminates in Richard Price.

We might, indeed, have expected that men like Green and Bradley would have found a kindred spirit in the author of the Sermons on Human Nature, but in fact Bradley, so far as I can recollect, shows no knowledge of the British rationalistic moralists, and in the Prolegomena to Ethics Butler receives only the barest incidental mention. It is too much, perhaps, to say that the attitude of the younger men who would probably have agreed to recognise Green as their spiritual father has usually been unappreciative where Butler is concerned; it is, at any rate, no more than the truth to say that their sympathy has been as imperfect as Charles Lamb’s with Scotsmen. The late Dr. Rashdall was in many ways more independent in his thinking and much more interested in our native traditions in ethics than most of the “men about Green,” as was even more manifest in the admirable lectures he used to deliver in Oxford some thirty-five years ago than in his writings; and from him I should have expected a higher measure of appreciation. Yet I know from our long-continued private correspondence that the unqualified admiration he accorded to Price was hedged about with all sorts of qualifications and reserves as soon as one ventured to claim it equally for Butler.

On the Continent Butler seems not yet to have received serious attention. Possibly the explanation may be that he has too much of the traditional English |293| Augustinianism to be altogether acceptable to the more religious thinkers of countries where the Thomist theology is dominant, while to the less theological the mere fact that he was a prelate and an apologist would be sufficient reason for not taking him seriously. Whatever the reason may be, in the one German manual of the history of philosophy where I remember to have seen his name, he was curtly dismissed as a divine who regarded “funk of hell-fire” as the principal reason for a virtuous life, while in France and Italy even his name seems to be hardly known. It was quite recently that a very eminent French philosopher wrote to me of Butler as an author whose name he had only just learned, and of whom he knew nothing beyond the fact that he appeared to have been a Bishop. For those of us who are persuaded that Butler is in fact a thinker of singular depth, as well as nobility, it is plainly a duty to show, if we can, that the comparative neglect into which his memory seems to have fallen is undeserved. It is particularly opportune to undertake such an Ehrenrettung, so far as the Sermons and their ethical teaching are concerned, at the present moment, since the present year is the bicentenary of their publication. The Analogy I may leave on one side, all the more that Dr. Broad has written of it so recently and so well.

I believe that most of the disparagement to which Butler’s ethical teaching has been subjected, so far as it rests on anything more than carping at words, arises simply from a lack of historical perspective. His utterances are judged, as no man’s utterances should be, without taking into account the circumstances of his life. Readers, especially young readers, are too apt to think meanly of the Bishop because he gave the world neither a completed system of metaphysical philosophy nor even a system of morals. They rightly feel that he |294| has said nothing on many of the issues which would have to be considered in a systematic treatment either of metaphysics or of morals, and they proceed to the wholly unjustified conclusion that since he has left so much unsaid, the permanent worth of what he has said must be slight. This is the line taken in most of the familiar unfavourable criticism of the Sermons in particular. Butler, we are told, has no systematic psychology of moral action. He “hedges” on the issue between reason and feeling as the source of our awareness of moral distinctions; he has no worked-out psychology of “conscience”; he does not face the problem created by the disagreements in the deliverances of the “consciences” of individuals and social groups; he ignores the fact that “conscience” develops; he does not discuss the question whether, even for myself, my own “private conscience” is an infallible guide; according to the fashionable interpretation of his language, he even leaves it in the end an open question whether “conscience” would not lose its authoritativeness if it could be shown to be in opposition to “self-love.” On all these grounds, his contribution to a definitive moral theory cannot be anything of much moment.

Now much of this criticism may be at once disqualified by one very simple reflection. Before we venture to say that any man ought to have done certain things which he admittedly has not done, and proceed to depreciate his performance because he has not done them, we need to know whether the circumstances in which his work was undertaken made it feasible for him to do these things, and if so, whether doing them would have been relevant to the purposes determined for him by his situation. It is no reproach to any man not to have done what his situation would not permit of doing, still less not to have done what his situation would have |295| made irrelevant to his purpose. In Butler’s case it must not be forgotten that his situation and his purposes were those neither of an academic Professor nor of a leisurely man of letters leading the “speculative life,” but those of an active and busy Anglican clergyman. As a preacher at the Rolls Chapel, and subsequently as the parson of a North Country parish, his work was primarily that of a practical “preacher of righteousness” and a Christian priest; as a prelate his time must have been taken up pretty completely by administrative cares and duties. At no time in his not very lengthy life had Butler the leisure to work out in detail a speculative system, and it is absurd to depreciate him on the ground that he never executed a task which he could not have executed without being false to the vocation he had chosen. Had Butler’s situation been like that of Locke before him or Gibbon after him, the case would be altered. But the performance would probably have been different too. For my own part I think it not impossible that in a situation of life which would have made devotion to pure speculation a moral duty or even morally permissible, Butler might have approved himself the greatest metaphysical intellect of his century. But that is, no doubt, a mere matter of personal opinion.

With reference to Butler’s ethical doctrine in particular, we must never forget that, except for the Preface to the Sermons and Dissertation appended to the Analogy, it is conveyed to us through sermons actually delivered in a London pulpit, and that the sermon is a literary genre determined by its special object. The preacher aims at producing a very specific result and at producing it by a spoken appeal to an audience with a specific psychology. Butler’s object, as appearsvabundantly from the text of his Sermons, was not to construct |296| elaborate and subtle psychological analyses, but to impress on a London audience the necessity of conducting their lives virtuously. Elaborate psychological refinements would have been badly out of place in sermons intended to dispose of the sophistries by which a fashionable audience tries to excuse itself for its neglect of what it really knows all the time to be its duties. For the immediate purpose what is wanted is what Butler undertakes to give, a convincing indication of the claim of the known moral law to absolute direction of our conduct. To a man of leisure sitting in his study, it is an interesting and a proper subject for discussion whether we apprehend that law by a “perception of the heart,” or by a “sentiment of the understanding,” or by both at once; the issue is irrelevant to the preacher dealing with a congregation whose excuse for misconduct is either that they do not apprehend the law at all or that, if they do, they do not see any obligation to regard it when they would prefer to ignore it.

So it is no fault of the Sermons that they disregard the apparent or real variations in the deliverances of “conscience.” We must remember that these Sermons are what they profess to be, actual discourses addressed to audiences of educated citizens of London in the last years of George I., a body with perfectly definite educational and social traditions. For the purpose in hand, it is irrelevant to dwell on minor differences between the moral convictions of individual members of the audience. It is quite certain that in all essentials Butler’s hearers would be agreed (as educated citizens of the same age, the same class, the same language, and the same historical traditions must necessarily be agreed), on the question what sort of conduct is right and what is wrong. The question Butler rightly reckons with as the one which will be raised is not what it is that “conscience” |297| tells us to do, but why we should do it if we are minded otherwise.1

All through the argument it is assumed that there are two points on which there is no dispute. We know what “virtue” is, and we are agreed that the reasonable and right course of life is to live “conformably to our nature.” The only point disputed between Butler and the opponents whom he wishes to convince is whether the way of living which is admitted to be “virtue” is or is not that which is conformable to our “nature.” Butler does not consider the possibility of conflicting moral codes and the grounds on which a choice could be made between them; and rightly not, because he knows that his audience will not justify their misdeeds, after the fashion of Shelley’s coterie in a later generation, by arguing that they are “virtuous” when judged by the standard of the “higher morality.” The early Georgian man of the world was too full-blooded a sinner to take that line of defence. To have foisted into an argument addressed to him reflections on the fact that the moral code of an Englishman of 1725 was not that of a Spaniard of the days of Philip II., nor of a contemporary of the First Crusade, nor of a Hottentot at the Cape, would have been the merest irrelevance. (Not that there would be any great difficulty in dealing with the suggested problem from Butler’s point of view.)

In principle these elementary reflections appear to dispose of all the usual criticisms made on Butler with |298| the single important exception of the allegation that in the remarks of the Preface on the “ambiguous case” in the ethical doctrine of Shaftesbury, and again in the well-known passage in the Sermons on the Love of our neighbour, he compromises the authority of “conscience” by admitting that it could not be maintained if there were ever any reason to think “conscience” and “self-interest” opposed. But I should like to make some more special remarks on the principal features of his moral teaching with a view to the clearing-up of what seem to me persistent misunderstandings about its real character.

(1) What is Butler’s real position on the questions which must be answered differently by a “rationalist” and by a “sentimentalist” in ethics? In the Sermons themselves he avoids taking up any definite position, and, as I have said, I think him justified on the ground that, for his immediate purpose, which is simply to insist on the authority of the moral law, a discussion of the precise way in which that law is apprehended would be irrelevant and disturbing. It does not follow that Butler had not a definite view, and the Preface to the Sermons, taken along with the explicit statements of the Dissertation, should make it fairly clear what that view is. It is true that even in the last-named work Butler does not expressly identify the “reflective faculty” by which we approve and disapprove our own conduct either with “reason” or with “feeling,” either with the “sentiment of the understanding“” or with the “perception of the heart,” but seems to hold, as is the fact, that moral approbation and disapproval imply both factors.2 But the real issue at stake in the controversy |299| between rationalist and sentimentalist can be stated in a way which involves no special psychological refinements, and when it is so stated, it becomes pretty clear what Butler’s attitude towards it is. I take it that no “rationalist” ever meant, by saying that moral distinctions are apprehended by reasoning or understanding, to affirm that moral approval and disapproval are purely unemotional intellectual states.3 In any man who asserted such a proposition it would be self-stultifying to appeal to our widespread human detestation of cruelty or ingratitude, or our sympathy with and admiration of the contrary virtues, as evidence for his theory. Yet such appeals are constantly made by “rationalists” no less than by “sentimentalists.“” The real question at issue is not so much that of the “”origin” of our approbations and disapprobations as that of their content.

The typical contention of all the writers who have been called “rationalists” is that, whatever may be the way in which we come by our moral convictions, what we approve or disapprove is always an objective character of our voluntary reactions to our situation, which can be identified by the intellect après coup. Or, in other words, the rationalists’ point is that the intellect can apprehend principles which are implied in and justify our approbations and disapprobations. The essence of the rival view, expressed most forcibly by Hume, is precisely that we cannot do this. Hume, for example, holds that we know what we approve and what we disapprove, but we can discover no rational principle at |300| the bottom of our verdicts of approval and disapproval. The issue, at bottom, is as unpsychological as the very similar issue in speculative philosophy about the presence or absence of a priori principles of interconnection in our scientific knowledge. To be a rationalist in your epistemology does not mean that you deny the difference between thinking and sensation, or that you hold that science draws no premises from sense-perception. It means simply that you regard the world as in principle through and through intelligible, as a system of “facts” or “events” interconnected by principles which justify themselves to the intellect on scrutiny. To be a rationalist in morals means in the same way to hold that the body of moral approvals and disapprovals which make up the content of morality is permeated by recognisable principles luminous to the understanding. This, and nothing else, is what, in fact, “rationalists” like Cudworth, whose ethical work has never been completely published and was wholly unpublished when Butler was preaching his Sermons, and Clarke, of whose work he had been a close student, had set themselves to show. And at this point I trust I may be pardoned if I make a brief digression from my immediate subject.

In the interests of clear thinking, I should like to enter a protest against attaching any undue significance to the common distinction between Intuitionism and Rationalism in ethics, or, what I suppose comes to the same thing, the distinction between Particular and Universal Intuitionism. I cannot believe that there ever has been a sane thinker who held either that all our moral approvals are “universal” or that they are all “barely particular.” In point of fact, in actual life we pass moral verdicts of both kinds. We approve or disapprove certain general ways of behaving, e.g. generosity, |301| gratitude, cruelty, revenge; we also approve a particular act as the proper act to be done in a certain concrete situation for which no accepted code of duties provides any rule at all. I cannot believe that writers like Clarke, who devote most of their space to the consideration of the broad general lines of right action, meant to imply, what none of them ever says, though they are sometimes criticised as though they had said it, — that our verdicts of right and wrong have been from the first couched in this “universal” form. I can see no reason to suppose that Clarke would have felt called on to dispute the sensible observation of Adam Smith, that our approvals and disapprovals are called out by individual acts, but express themselves in a form which “universalises” the character of the act evoking the verdict, as for instance, to take Smith’s own example, when we are revolted by a strikingly cruel act and say to ourselves, “I must never do that kind of thing.” On the other hand, I cannot believe that any sane “particular Intuitionist” ever supposed that when he approved or disapproved an act, his verdict was so “barely particular” that he could not indicate the character in the act at which it was aimed. And, of course, in singling out for attention any character of the concrete and individual, you are already “universalising.” Thus, as it seems to me, the attempt to make a distinction in principle between Rationalism or Universal Intuitionism and Particular Intuitionism is doubly objectionable. The distinction is not really applicable to the doctrines it is intended to classify, and it further rests on the importation into ethics of irrelevant and misleading psychological considerations.

Now, if the real ethical issue as between Rationalist and Sentimentalist has been rightly stated in what I have said, there can be no doubt where Butler stands. |302| He is definitely on the side of the Rationalists. This is, to begin with, indicated by the simple fact that he rests his case against his opponents on the question what kind of conduct is rightly to be called “natural,” when we have regard to the consideration that “nature,” in our ethical discussions, means “human nature,” and that human nature is a “system” of “principles of action.” The question here is not one of “psychological origin,” it is one of justification, and justification at the bar of the intellect. That is the bar at which you would have, if required, to justify the assumption that the “natural” rule of life is the rule it is good to follow; it is again the bar at which the further question what way of living is the “natural” life for men in particular must be argued. It is for intelligence to pronounce what is the natural life for man, as it is for intelligence to decide what is his “natural diet.” The reason in both cases is the same, that it is intelligence which supplies us with our standard, by revealing what is, as Butler likes to put it, in point of fact our “nature.” Butler, to be sure, gives a special name to intelligence as declaratory of the line of conduct conformable to our nature; he calls it “conscience.”4 But he is careful to define conscience simply as “the reflex principle of approbation and disapprobation,” the principle by which we judge approvingly or disapprovingly of our own conduct, and this ought to place his meaning beyond doubt.

|303| If we could feel any uncertainty on the point after reading the Sermons themselves, our doubts ought to be dissipated by the study of their Preface and of the Dissertation. It is significant that in the Preface Butler is careful to explain that he is fully in accord with the teaching of Clarke and has no strictures to make on his attempt to vindicate morality by reasoning from the “nature of things” in general.5 By his own avowal, the thoroughgoing rationalism of Clarke is formally the most satisfactory way of treating moral philosophy; his own more specific and restricted appeal to the nature of the human agent in particular has only been adopted because it has, for the purposes of the preacher, the advantage of being more readily apprehended by and making a more immediate impression on, the average congregation.6 So the rationalism of the Dissertation comes out unmistakably in one of its most important passages, that in which dissatisfaction with our situation is contrasted with censure of the conduct which has led to the situation.7 Butler is not arguing that, viewed as so much emotion, there is some special unique character in the uneasy feeling of the man who judges that he has, e.g., ruined his prospects by misconduct, which distinguishes it from the disagreeable feeling of the man who finds that he has been ruined |304| by accidents for which he is not accountable. (Suppose, the failure of his banker or the miscarriage of his ship.) He finds the relevant difference entirely in the intellectual judgement that my position has been caused by my own wrong-doing. And similarly, both in the Sermons and in the Preface to them, the reason why man, unlike the brutes, would be living an unnatural, and therefore an ex hypothesi wrong life, if he simply acted on the prompting of the “strongest impulse” is found in the fact that man has a power, which the brutes have not, of judging his own actions as good and bad. He is alive to a real character of his acts which has no meaning for a brute, and therefore it is unnatural that he should act as though he were not alive to it.

To criticise this straightforward Rationalism on the ground that Butler gives to understanding, in its practical function as the discernment of good and evil in our own conduct, the special name of “conscience” is surely to fall into the fault of mere carping at words. Aristotle, too, has his own special name for thought exercised upon questions of choice between acts and inspired by an “end” to be realised; he calls it definitely phronesis, by way of antithesis to the sophia which is the pursuit of purely speculative truth. Yet no one accuses Aristotle, as some men are not ashamed to accuse Butler, of the crime of accepting a “faculty psychology.” No one doubts that Aristotle’s meaning is that it is one and the same intelligence, whole and entire, which displays itself differently in the creation of science and in the creation of a system of morality. What reason is there to suppose Butler to mean anything else? In general, the accusation of belief in a “faculty psychology” is one of those criticisms which are too facile to be really enlightening. If the proof of |305| the charge is taken to be found in the mere fact that an author gives different names to the functions of thinking, understanding, and judging, according as they are employed upon a different matter, it will be hard for any philosophical writer who finds it necessary to achieve lucidity and brevity by the use of a technical terminology to avoid a conviction. If we further require, as we should, evidence that the accused has really committed the error of substituting a plurality of “faculties” for the single intelligent thinker as the agents in the mental life, most of the writers whom it is fashionable to condemn summarily as guilty of the fallacy of the faculty psychology will probably be entitled to an honourable acquittal. The word “faculty,” taken by itself, may be démodé, but is as innocent as the word “capacity,” or any other of the equivalents which the most determined enemy of Associationism finds himself quite unable to expunge from his vocabulary. As for Butler, his own language leaves it beyond dispute that when he speaks of “conscience,” he means to describe the fact that we are capable of judging, and habitually do judge, our own acts and characters as right and wrong, good and bad, and that such a judgement carries with it the conviction, which it is his object to justify, that we “ought” to give it effect in our actual “choices and avoidances.” It is a convenient thing to have a simple name for this capacity of judging our own conduct and character, and the name “conscience” is readily supplied by current popular language. To say, as I have had it said to myself by one distinguished moralist, “I should have had no quarrel with Butler if he had only used the word reason,” is to agree with him about the facts, but to make one’s self the pedant of a particular vocable.

(2) To return to the series of more specific accusations |306| which have been leveled at Butler’s account of conscience, they may be reduced roughly to the following heads, (a) Butler does not deal with the problem of the historical growth of conscience; (b) he seems to leave us with no more certain and authoritative guide in action than the “private conscience” of the individual; (c) he does not explain in any systematic way what it is that conscience enjoins. Why Butler does not deal with these questions at length has, I think, been sufficiently explained. The reason is that he does not think that the great practical deficiency of his hearers is want of acquaintance with their duties; it is an insufficient sense of the obligation to take duty seriously. If we read the three Sermons on Human Nature in particular with careful attention, we shall find that they form an argument addressed to two classes of persons. The first class are those who, under the influence of a false theory of human nature, hold that egoistic self-seeking is the only possible motive of all human acts. “Virtue,” being ex hypothesi not self-seeking, is unnatural, in the sense that it is impossible. Anyone who bids us cultivate it is either a dupe or a designing impostor, a theory which it is not hard to identify with that of Mandeville; if indeed Mandeville was seriously intending to expound any theory at all, and not merely to amuse himself by cynical contemplation of the “seamy side” of human affairs. These opponents are met in principle by the argument for the disinterestedness of all the “particular passions.”

The virtuous man is as much, and no more, of an exception to their theory as the gambler who wastes a noble estate to gratify the passion for excitement, the rake who ruins splendid bodily and mental gifts in the service of Bacchus and Venus, or the vindictive man who throws away his life to gratify his grudge. |307| And it is undeniable that the gambler, the profligate, and the revengeful man are only too real types of humanity.

The second class of persons with whom Butler has to reckon have a theory of human conduct which is not that of Mandeville, but is common enough for all that. They are the type who admit the reality of the promptings of virtuous impulse, but see no reason why we should be virtuous when we do not happen to be strongly under the influence of these impulses. Their view, unlike the former, does not involve the confusion between “self-love” and the “passions.” They hold that the natural and proper course of life is to have no settled rule of action whatever, to let every impulse, good or evil, “take its turn as it happens to be uppermost.” If such irresponsible creatures were capable of enunciating a principle, the principle they would own to is that of “living in the moment and for the moment.” As the distinction between “self-love” and the “passions” is meant to supply the refutation of the first class of immoralists, so the subsequent argument for the supreme authoritativeness of “conscience” is intended specially for the benefit of the second. We see that this is so from the simple consideration that throughout this argument it is the “passions” in opposition to which the authority of conscience is asserted. There is no corresponding demonstration that conscience has “manifest authority” over the promptings of cool and reasoned regard for our own “interest.” We have to discover Butler’s view on this point by collation and comparison of passages which occur in several different contexts. No doubt this absence of a direct assertion of the supremacy of “conscience” over “self-love” can partly be explained by Butler’s avowed conviction that in the main “interest” and “duty” coincide; to disregard the second also means |308| disregarding the first.8 (Hence his emphatic statement that most men do not care too much, perhaps do not care enough, about their own interests; it is not considered pursuit of self-interest which is the worst enemy of practical morality.) But it remains a curious fact that Butler does not think it necessary to deal with a third possible view of the conduct “natural” in a human being, the view that it is neither abandonment to the passion of the moment, nor yet strict obedience to “conscience,” but the calculating pursuit of my own “greatest happiness on the whole,” the “rational egoism” of Henry Sidgwick. I do not see how the omission can be fully accounted for except by the consideration that Butler’s purposes are strictly practical and that he did not contemplate “rational egoists” as forming an appreciable part of his audience.

We can, however, perhaps collect the answer he would have given to the “rational egoist” from certain striking passages of the Dissertation. In the second Sermon, where the case of a man who rushes open-eyed on his own destruction is contrasted with that of a beast which does the same thing, for the purpose of proving that in man “self-love” (which has no existence in the beasts) is a “superior” principle by comparison with the “passions,” it is clear enough that Butler holds wanton indifference to our own “happiness” to be unnatural and therefore wrong. But the treatment given in the Dissertation to the question whether we are “more at liberty” to make ourselves miserable than to make others miserable, suggests that if he had been asked |309| why it is wrong to neglect your own happiness, his considered reply would not have been “because neglect is contrary to self-love,” but “because conscience forbids neglect.”9 His final view seems to have been that “self-love” is never, to use Kantian language, a source of imperatives. It is concerned with the formation of probable judgements of fact. Its utterances are all of the form “this will probably add to (or will probably detract from) my happiness.” The imperative “I ought to do what will make me happy and ought not to do what will make me miserable” is clearly assigned to “conscience.” It is in keeping with this that in the Sermons themselves conscience is always spoken of as the reflective principle, or the principle of reflex approbation and disapprobation.

“Self-love” is, of course, reflective, in the modern sense of the word; it is intelligent and “calculates,” as the “passions” do not. But Butler clearly means to use the word “reflective” or “reflex” in the strict sense. Conscience is said to be reflective, not simply because it is intelligent, but because its judgements are judgements about our own conduct; the agent is also the judge who turns his scrutiny upon his own act. Hence one of the most serious allegations against Butler (which shall be examined shortly), has, in my opinion, always been wrongly expressed. It is not true that Butler ever “coordinates“” self-love and conscience as principles of action; he never considers self-love as having a right to prescribe acts which conscience condemns. The real way to make the complaint intended by his critics |310| would be to ask whether Butler had not sometimes talked as though conscience might contradict itself by commanding regard for our own “interest” and yet also commanding some act which would be in conflict with that interest. The question is not whether there are two “authorities” who may disagree, but whether the one authority may possibly be inconsistent in its injunctions. All the “imperatives” are imperatives of conscience, but there may be a doubt whether any of them except the command to aim at my own happiness is wholly categorical.

I do not see that any serious ground to take exception to Butler’s principles is furnished by the evidence about the gradual growth of conscience or the differences between the conscientious convictions of individuals. It is important to note two points. Butler never speaks of “private conscience,” and what he ascribes to conscience is authority, not infallibility. Of divergences of moral judgement between members of his audience he speaks as trivial and secondary,10 and as I have argued, he was fairly entitled to take this view. It is a fact that persons of the same age and social tradition will “in the main” agree on the broad general question what kind of conduct is morally good. Those who talk of “conscience” as though it were a set of whims and fads confined to the individual are attaching far too much importance to the “conscientious objector.” Even his conscience is not active simply in “objecting.” He may have his personal fad, a belief that vaccination is sinful |311| or the like, but ninety-nine-hundredths of the things his conscience commands or forbids are precisely the things commanded or forbidden by the consciences of his fellows. And even in the hundredth case his disagreement with them is usually on a question of fact, not on a question of right. The anti-vaccinationist does not dispute his duty to protect his children from smallpox; he merely denies that vaccination is protective. His disagreement with the rest of us is on a point of medical, not of moral, theory. The vegetarian, for all his language about “wanton cruelty” to the animals, is always careful to insist that a vegetarian diet is more wholesome than a flesh diet. If he could be convinced that flesh-eating is desirable on dietetic grounds, he would apparently lose most, or all, of his moral scruples. Even the “objector” who refuses his services to his country in war-time commonly justifies himself not so much by denying the ethical principle that a society righteously defending itself may require those services, as by alleging that the facts about the special circumstances in which his country has entered on hostilities make the case one of unrighteous war. On the question what kind of war is righteous, he is often in complete agreement with everyone else.

There are ultimate moral disagreements, but they only cover a minority of the cases, and even they are, in the main, disagreements about axiomata media, not about principles. For example, there are those Englishmen who regard fox-hunting as inexcusable cruelty to animals, and those who do not. But both parties would probably agree in the propositions that some infliction of suffering on the lower animals is morally permissible, and that some is not. The point in dispute is only whether the amount of pain inflicted on the fox, when weighed against the enjoyment of the hunter, comes |312| under the head of the permissible. Butler thus appears to me right in treating genuine variations of moral judgement among members of the same group, with the same traditions, as a secondary matter. But, incidentally, we may note that by taking this line he makes it clear that he supposes the deliverances of conscience to be universal principles, e.g. the commands to practise generosity, gratitude, justice, and the other great typical virtues. He cannot have supposed that there would be anything like the consensus he assumes, if the question were whether a particular act, in a given complex situation, is the generous or just act to do. So his recognition of veracity as one of the things enjoined by conscience cannot be interpreted as committing him to the view that there are no circumstances in which “verbal misleading” can be right. This inference as to his meaning is strictly in keeping with the approving tone in which his Preface speaks of the moral theory of Clarke.

Again, I do not see why Butler need have been disturbed by the fact that the “conscience” of Englishmen of his own time differed widely from that of the heroes of the First Crusade, and both from the conscience of a Hottentot. As far as “historical development” goes, conscience would be in no worse case than any other form of knowledge. There is really no reason why we should defer more to the judgement of the Crusader or the Hottentot on a question of morals than on a question of science. If the Crusader regarded it as unquestioned truth that it is good to kill an “infidel” for being an infidel, he also, if he thought about the matter at all, presumably held that there is no land between the West of Europe and the East of China, and that the infidels whom he slaughtered worshipped Mahound. We do not regard ourselves as bound to doubt the competence of |313| the human intelligence because men have made these errors in geography and historical knowledge; why are we not equally free to admit that they have also judged wrongly in questions of morals? It cannot be fairly demanded of the believer in the authority of conscience that he should regard every opinion about right and wrong which has ever been entertained by any man as equally authoritative with every other. There are, however, two further remarks which ought to be made in this connection, as necessary to a reasonable interpretation of Butler.

The first is that there is a real danger of exaggerating the amount of disagreement between the “consciences” of society in different ages. Much which would be commonly put down loosely to “variations” in “conscience” does not really come under this head. Thus we disapprove strongly of the man who prosecutes the blood-feud, or, in other ways, attempts to right himself and his friends with the strong hand. In earlier ages and ruder societies the prosecution of the blood-feud had been regarded as a positive duty and a very sacred one. Yet it does not follow that the virtuous man of today who regards private revenge as absolutely prohibited really differs on the principles of right conduct from a man of the past who thought it a sacred duty to avenge the innocent blood with his own hands. The difference is, at bottom, a difference in the situation of the two men. In the one case there is an organised society which will make the punishment of the homicide its business, and will execute it with a certainty and an impartiality impossible where the repression of crime of this kind must be undertaken by private enterprise; in the other, unless a kinsman of a murdered man has the public spirit to devote himself to the prosecution of the feud, murder will often be committed with impunity by the |314| violent. The man who makes it a duty to pursue the slayer, and the man who, in happier times, makes it a duty to “leave the offender to the law,” may be entirely agreed on the question what justice is and both may have the same respect for justice. The difference between them may simply be that the methods for securing the execution of justice are much more developed in the society of the second than in that of the first, and this does not amount to a difference in the deliverances of “conscience.” Similarly, the great scholastics explained the polygamy of the patriarchs by saying that though the principles of the moral law which regulates human marriage are the same in all ages, there was a need in the world’s youth which there is not now for numerous offspring, in order that the earth might be brought effectively under the dominion of man. What would in us be only explicable as “licentious” disregard of the sanctity of conjugal ties was in the patriarchs respect for the moral law which enjoins the subjugation of nature by man. And it might not be hard to find other illustrations of the point.

If, as I have already argued, what Butler understood “conscience” to reveal was the great principles of moral conduct rather than the precise way in which they are to be applied to particular cases, considerations of this kind will make it clearer that he might have extended to many of the so-called variations of conscience from one age to another his remark about individual disagreements within the same community, that they are, after all, secondary.

However, it is, of course, clear that we cannot treat all the relevant facts in this way. We cannot, when all legitimate deductions have been made, deny that as intelligence develops there is real advance in delicacy of moral discrimination; men do become alive to obligations |315| of which they had formerly been ignorant. But here comes in the second of the considerations I wish to urge. Butler never says or suggests that conscience, in his own or any other age, is infallible. What he says is that it is authoritative. Now authority does not necessarily imply infallibility in moral matters any more than in others. A guide who is far from infallible may be the best guide I can get, though the distinction is so constantly forgotten in the controversies about the place of authority in theology. In matters of science, history, and the like we all understand the point well enough. A man who is not himself competent to form a personal opinion on a question of history or archaeology or textual criticism is expected to regard the general consensus of the acknowledged “experts” as authoritative; even a man who is qualified to form a personal judgement is rightly expected to allow great weight to such a consensus and not to reject it without being prepared to give convincing reasons for his dissent. This does not, of course, mean that the “experts,” when agreed, are infallible, or that anyone thinks them to be so. We all know that they may agree in a mistake and sometimes do so. But it means that in the main we are likely to be most often right, and right on the most important matters, if we follow their guidance. Of course, there are exceptions. For instance, a man who accepted the authority of the leading “experts” in Greek scholarship some seventy years ago would have had to reject a good many of the ripest of the Platonic dialogues as spurious and to declare the Homeric poems badly constructed pieces of patchwork. We know today that the “experts” of the fifties of the nineteenth century were wrong on the first point, and the conviction among Greek scholars themselves is happily spreading that they were wrong on the other point too. And yet, on |316| the whole, the man with little or no Greek scholarship would have done better to read the history of Greek literature by their aid than to venture on it without that help. His “…authorities” would have been very fallible indeed, but they would still have been the best he could get.

In principle the case seems to me to stand no otherwise with the authority of “conscience.”As I have tried to urge, Butler does not mean by “conscience” something peculiar and private to the individual man. He seems rather to have meant the body of moral convictions which is common to the “best” men of a society — a “public” rather than a “private” conscience. That he never, even by implication, claims infallibility for this body of convictions yields at least a presumption that he did not suppose it to be infallible. His teaching never requires us to hold that even the “best” morality of one age may not be improved on by a later. But, for all its possible deficiencies, it may still well be the case that the conscience of the best men in my society is the best guide I have and ought to be followed as such. If I find myself in sharp opposition to it on some question of conduct, I may be right, but, if I am a modest man, I shall be well aware that I am very likely to be wrong. And I shall think it a duty to discover what the best men I know think of my proposed line of action, and, if they condemn it, to make very sure indeed of myself before I venture to decide that, on the particular point in question, the “best” moral opinion of the age happens to be that which is incarnate in just one personality, and that personality my own. I can see nothing in any of Butler’s statements about the authority of conscience which would warrant the view that I am doing right in always acting simply on my own personal impressions, or that it is not my positive duty to attach the greatest |317| weight to the judgement of good men generally. Indeed, I believe it would be quite consistent with all he has to say, that a given man should recognise some other man as a better judge than himself on a specific class of moral questions (e.g. on matters involving a delicate sense of sexual purity), and consequently resolve in those matters to follow the other’s spiritual direction closely. Though, we must note, even in such subjection to spiritual direction, a man does not divest himself of ultimate responsibility for himself. The selection of the adviser is his own act and it is his business to select wisely.

I believe one might, without abandoning any claim made by Butler for “conscience,” go a step further. Probability, we must remember, is, according to Butler himself, the guide of life. Hence the recognition that, when all care has been taken to listen for the voice of the inward monitor, and to eliminate the possibilities of misjudgment due to unconscious bias, or to unfamiliarity with the kind of situation in which I am called upon to act, by humbly seeking the judgement of men better and wiser, or more experienced in the kind of affairs with which I find myself called upon to deal, I may honestly fall into moral error. The question will thus arise, not in connection with the broad principles of morality, but in connection with their application to the special situation, can I ever be absolutely certain, beyond all possibility of error, that the act the “reflective principle” pronounces right is exactly and perfectly right? It seems to me that absolute certainty is not attainable and that Butler’s moral theory does not require that it should be. We can be certain that some ways of acting in the given situation would be wrong because they would involve, e.g., injustice or ingratitude. But we cannot, in many cases, be |318| sure exactly what act would, e.g., be most fully in accord with the demands of justice or again of moral prudence.

Suppose, for example, that I am left as guardian and executor to the child of a friend or relative, with a very considerable latitude of action, neither the utmost scrupulosity nor the most careful consultation of the more experienced can ensure that I shall make no mis- take in selecting an investment or a school for my ward. And the situation is only typical of those which are arising every day for every man whose notion of right action is anything more than mere abstention from the ways of behaviour which good social opinion has definitely classed as “not respectable.” The problem recurs for a magistrate every time he has to decide for himself what penalty to inflict for an offence, for a citizen every time he is called on to exercise the franchise.

It does not seem to me that in cases of this kind we can ever expect to be absolutely certain of the “path of duty.” We have always to make the reservation which Prince Florizel makes in Stevenson’s tale when he decides to throw the Rajah’s diamond into the river, “God forgive me if I am doing wrong, but its empire ends tnight.” I doubt if any of us who has ever had to take a momentous decision in the course of his life has been able, in the act of taking it, to avoid uttering the prayer “God forgive me if I am deciding wrong.” But, at the same time, I cannot see that this possibility in any way diminishes the authority claimed for conscience, when we have honestly done our best to make sure that the decision taken has been dictated by conscience. To demand more certainty than this would be to require that “conscience,” when due pains have been taken to enlighten it, shall have not only authority but |319| infallibility, and this is a claim which Butler never advances.

Indeed, it is just in not advancing this claim that Butler appears to me to show a decided advantage in sobriety over that other great moralist of duty, Kant. Kant does more than once make the extravagant assertion that no honest man can ever be in doubt about the path of duty, because he can always discover that a proposed course of action is wrong by simply applying the test of universalising its “maxim.” But manifestly this test is hopelessly unsatisfactory. Not to dwell on the point that there seems to be a great deal of very bad morality which would stand the Kantian test quite successfully, it is obvious that at the best the test will only secure you against “flagrant sin.” It would, for instance, have been of no use whatever to any British citizen anxious to know how he ought to use his vote at the last General Election. Before voting for the Labour candidate or the Conservative candidate a man certainly ought to ask himself whether he would be doing right by voting for either, or again by abstaining from using his vote, but no “universalising” of any maxim would decide that very pertinent question. All you can say is that I ought to use my vote, having regard to the general issues at stake, the particular choice between candidates in my own constituency, the probable consequences of the return of each of the contending parties to power, in such a way as to do the most good and the least harm. It is certain that, however I vote, I shall be voting for a party which, supposing it to be victorious, will do some good and also some evil, and I have in the first place to judge which party is, on the whole, likely to do most good and least evil. And the problem is usually still more complicated. I may be satisfied that on the whole it is desirable that one of the parties should |320| come into power, and equally satisfied that it is desirable that it should be sobered and balanced by not having too large a majority over its rivals, and in some cases this, I take it, would be a valid reason for thinking it my duty to vote against “my own party.” And there is still a third complication created by the fact that I may be convinced that one of the candidates between whom I in particular have to choose is a man of such outstanding character and ability that his services ought not to be lost to the Legislature, irrespective of the question what his party is.

To make the illustration more precise, let us suppose I am satisfied that on the whole the Labour party stand for the best interests of the inhabitants of the country as a whole, and therefore desire to see that party returned. I have to set against this desire the consideration that it is very undesirable that they or any other party should be able, from the weakness of the Opposition, to do pretty much what they like for four or five years. Such complete domination of the situation is very likely to mean that they will effect that very part of their programme which I regard as bad and unwise, as well as that on the strength of which they have my approval. Or, alternatively, it may mean that the discipline and vigilance of the party will suffer; they will be free to develop internal differences and to quarrel among themselves, and the very measures on the strength of which the party has my general approval will be less effectively carried out. And finally, I have to ask myself whether, supposing one of the rival candidates to be a man of exceptional character and ability, it would not be better that the majority of “my party” should be diminished by a defeat in this constituency, where their candidate is a man of no special qualifications, than that the exceptionally valuable man should |321| be lost to the counsels of the nation. I am not using my vote conscientiously unless I balance all these considerations against one another, and it is not an easy thing to decide what relative weight should be given to each of them. Any moral theory which professes to provide a short cut to the right decision in such a case stands self-condemned by its own pretensions. It thus seems to me proof of the judiciousness of Butler’s mind that, with all his insistence on the authoritativeness of the “reflex principle,” he never advances for it the claim to infallibility. To use the valuable distinction of Hutcheson, at best one can secure “formal rightness” in one’s conduct by obedience to conscience; there can be no guarantee of “material rightness.”

(3) I have left to the end the most serious imputation which has been brought against Butler as a moralist; I mean the charge that in the end he runs away from his own professed position by admitting that the obligation to obey conscience depends on the coincidence of the path of duty with the path of self-interest. So far as I can discover, the case for his critics rests on two passages, one paragraph from the first Sermon on the Love of our Neighbour, and the section in the Preface to the Sermons which discusses Shaftesbury’s problem of the “sceptic not convinced of the happy tendency of virtue.” There could, of course, be no ethical objection to what is clearly Butler’s view, that in point of fact a man does always promote his own truest interest by the strict practice of virtue. It is possible to hold that the fact is far from certain; but to believe it to be certain is no derogation from the claims of virtue, so long as you also hold that, though virtue does always make for our personal happiness, it is to be pursued, not for that reason, but because it is virtue. In Aristotelian language one may perfectly well hold that virtue and interest |322| “are the same,” in the sense that they are inseparably conjoined in fact (the virtuous act is always also the act which is “to my interest”), but that their εἶναι or esse formale is not the same, just as according to Aristotle the truly virtuous life and the truly pleasant life are the same, though virtue is not pleasure. Thus the mere fact that Butler, like all Christians, holds that virtue in the end always coincides with interest has no relevance to the charge we are considering. We may therefore confine our attention to the two specific passages to which I have referred.

Now if we had only the one sentence from the Sermon to go upon, we might, I think, have to admit that Butler has perhaps fallen into inconsistency.11 He certainly does appear there to be saying in so many words that we could not regard “benevolence” or any other conduct as morally obligatory unless we were convinced that it is also to our interest, and if he means this to be an expression of his own opinion, he is unsaying what |323| he himself had said about the “manifest authority” of conscience; the famous phrase “had it power as it has manifest authority, it would absolutely govern the world” becomes meaningless, if it is true that we need to satisfy ourselves that an act will be for our interest before we can pronounce it obligatory. Even so, it would be fair to urge that no one is ever absolutely consistent with himself in his every utterance, and that an author’s convictions ought to be judged by the tenor of his work as a whole and not by a single phrase which contradicts his own emphatic and repeated declarations. The reasonable thing would be to discount the isolated sentence as not really faithful to the writer’s convictions.

But this is not all that may be said. It is almost certain that the words complained of are not intended to convey Butler’s own opinion at all. They are a sarcastic concession to the audience he wishes to carry with him. Many of them, we can hardly doubt, would hold that, when all is said, there is force in the point of view put brutally by a character in Pickwick, “damn hurting yourself for anyone else,” and Butler may mean no more than that, even if you grant this principle, “benevolence” is not “hurting yourself.” Even if a man agrees with Lowten, you want him in practice to perform kind acts, and you may fairly urge that, after all, even on his own principles there is every reason for doing so, because he will not be hurting himself but very much benefiting himself. And there is one consideration in particular which, as I think, goes a long way to show that Butler’s language in this particular passage is meant to be no more than a dialectical concession to the audience. We may reasonably suppose that if he has taken special care anywhere to adapt his language closely to his thought, it will be in the Preface, |324| in which he is at pains to explain the precise drift of the discourses which follow. Unlike the Sermons themselves, the Preface is not a spoken address from the pulpit aiming at making an immediate impression; it is a literary production addressed to readers, composed at leisure to be studied at leisure. And the very passage critical of Shaftesbury, which is made part of the count against Butler, contains an express statement flatly inconsistent with the sentence we are now considering. There Butler claims to be proving that the “utmost scepticism” about the coincidence of virtue and happiness of which we can conceive leaves the obligation to the strictest practice of virtue unaffected. Now he cannot have meant both to affirm this and also to affirm that no obligation is binding unless we are satisfied that it is to our interest to recognise it. If one of these utterances represents his convictions, the other cannot. And there can be no doubt which of the two passages may be most reasonably taken to represent the author’s deliberate view. The criticism of the “ambiguous case” in Shaftesbury’s system is the one elaborate and express discussion of the problem which Butler has given us; the remark in the discourse on Love of our Neighbour is made at the end of a discussion proceeding on different lines, and has much more the character of an incidental obiter dictum. We must therefore conclude either that Butler has by a mere oversight allowed himself to make an incidental remark which is not really consistent with his considered position and forgotten to correct it, or, more probably, that the remark was not intended as more than a temporary concession to the prejudices of an audience.

There remains the actual discussion of the case put by Shaftesbury, the case of the “moral sceptic.” Butler’s determination of the case has, I believe, been |325| sometimes misunderstood from neglect of the dramatic form in which it is given. If we examine it carefully we shall see that the argument has two stages which must be carefully discriminated. The misconception comes in when it is supposed that Butler discloses his own hand in the first stage of the answer. The following analysis will make the point clear.

Butler begins by asserting that it is Shaftesbury’s neglect of the notion of obligation which led him to the view that the case of the “sceptic” is “without remedy,” i.e. that you cannot expect a man to see any reason for acting virtuously if he doubts whether it will be to his interest to do so. He then supposes the defenders of this position to retort that it makes no difference to the situation to “take in the authority” of conscience, since, they urge, even if you admit it and grant that there is an obligation on the side of virtue, still there is also a counter obligation to pursue my own interest. The immediate answer is that, even if this is granted, the conclusion they wish to draw would not follow. It would only follow that, in this statement of the case, there are two contrary obligations which cancel one another, not a “formal obligation to be vicious.”12 That is, if you leave the authority of conscience out of account, it might be fairly argued that a man who thinks he would promote his happiness by a vicious act is absolutely justified in being vicious. If you admit the authority of conscience, then, even on your own theory that there is another independent obligation to pursue my own happiness, there is at least as much justification for preferring virtue to interest as for preferring interest to |326| virtue. (It is true that Shaftesbury had never expressly said that the sceptic would be formally “bound” to be vicious in the supposed case, but it is part of Butler’s fairness that he wishes to state the case against virtue, which he has to meet, as powerfully as he can, and is therefore at pains to point out that, on the principles of a “morality without obligation,” a man would not merely be free to be vicious, if he thought vice to his interest, but would be acting in a way such a morality must pronounce wrong if he were anything but vicious.)

The all-important thing to be observed here is that the conclusion that the sceptic would “be under no obligation at all” is not Butler’s. The conclusion rests on premises which he rejects. “Granting this to be so” means no more than “allowing this to be so, for purposes of argument, though I deny it is so,” dato ma non concesso. Hence Butler continues by denying the relevant premise; “the obligation on the side of interest does not remain." His own solution of the problem is exclusively contained in the sentences which follow on this denial.13

|327| It may, however, be said that, even so, his own appeal is to interest. For does not his argument that “no one can be certain that vice is his interest” amount merely to the suggestion that perhaps vice does not pay? This is, I think, a complete misconception, as may be seen from the absence of any reference to what would be the central point if the argument were an appeal to interest, the magnitude of the disagreeable consequences which may, for all we know, attend on vice. The argument sometimes supposed to be Butler’s must take this point into consideration. Fully stated it would run thus. Even if there is only a very trifling probability that “all tales are true” about hell, still, if they should be true, the pains of hell are so terrific that a wise man will not run even a distant risk of suffering them. But Butler is wholly silent about this question of the. “value of the expectation” of future suffering. The contrast he makes is simply the contrast between the uncertainty that vice will be to our interest and the absolute certainty that the moral law carries its authority with it. The point is simply that this authority is “certain and known.” It is because of its certainty that Butler goes on to say that the authority of “conscience” absolutely destroys all the weight of the appeal on the other side to considerations of interest, which “would have been of real force” but for the known certainty of moral obligation. Had Butler held the position ascribed to him by some of his critics, he would have had no right to express himself thus. His argument should have been that even if the probability that Vice will lead to misery appears slight, yet the magnitude of the misery to which, according to the theory of future rewards and punishments, it does lead is so overwhelming that it must be imprudent to take the risk that the theory may prove true. The probability |328| of “damnation” may be low, but the “expectation,” which is what a wise man would be guided by in deciding whether to run the hazard, is another matter. As it would be reasonable to risk a penny on the tiniest of chances of winning a “million,” so it would be reasonable not to incur the faintest risk of the penalties of damnation.

But when the argument is stated in this way, it becomes a variant of “Pascal’s Wager”; it is differentiated from Butler’s reasoning in two ways. The magnitude of the penalties which the sinner may expect if there is a moral government of the world becomes the pivot on which the argument turns, and the consideration which is crucial with Butler, the intrinsic authority of the moral law, ceases to be relevant. In an argument intended simply to prove that the faintest probability that there may be a hell makes it imprudent to run the risk of going to hell, if a hell there be, it is superfluous and irrelevant to raise the question whether the moral law has an intrinsic authoritativeness or not, just as it would be irrelevant if one were captured by brigands and threatened with torture, to ask for a sight of the warrant empowering the infliction of the tortures. It should be noted, as further proof that Butler’s argument is not in principle that of the Wager, that he goes on to lay stress on the point that what renders a man justly liable to punishment is not fore-knowledge of the penalty but the transgression of a known law. Obviously this is no more than the truth. It would be no relevant defence on the part of an alien in this country standing his trial for murder to plead, what might be the truth, that he had not known that the penalty for murder by our law is death. He knows, at least, that murder is prohibited, and, consequently, by violating the prohibition he renders himself justly liable to whatever |329| penalty, known or unknown to him, the law assigns to the crime of murder. Similarly it would be no defence to a charge of any kind to urge, even with truth, that hitherto the administrators of the law have winked at the offence in question, and that it has been committed by others with complete impunity. This might (or might not) have some force as a plea after conviction for mitigation of sentence; as a defence it would be worthless. But the fact that Butler calls attention to this analogy of itself shows that the magnitude of the transgressor’s “expectation” of misery, the central consideration in every form of the prudential argument for virtue, is not intended to come into account at all in his own theory. His point is simply that we know with certainty that we have specific moral obligations, and that uncertainty whether neglect of them will affect our “interest” does nothing to destroy this certainty of their reality. Their reality would, in fact, remain, even if I could be certain, as I cannot be, that vice will do nothing to diminish my happiness, or even that a vicious act will augment it. If you grant the intrinsic “authoritativeness” of the moral law, even certainty that I should gain by violating it would not in any way affect the other certainty that I ought not to violate it, any more than certainty that I shall be rewarded by the Government for committing a convenient crime would make the commission of the crime a lawful act.


1^ Sermon III. “Let any plain honest man, before he engages in any course of action, ask himself, Is this I am going about right or is it wrong? Is it good or is it evil? I do not in the least doubt but that this question would be answered agreeably to truth and virtue, by almost any fair man in almost any circumstance. Neither do there appear any cases which look like exceptions to this; but those of superstition and of partiality to ourselves. Superstition may perhaps be somewhat of an exception; but partiality to ourselves is not; this being itself dishonesty.”
  Dissertation. Nor is it at all doubtful to the general, what course of action this faculty, or practical discerning power within us approves, and what it disapproves.

2^ Dissertation. “It is manifest great part of common language, and of common behaviour over the world, is formed upon supposition of such a moral faculty; whether called conscience, moral reason, moral sense, or Divine reason; whether considered as a sentiment of the understanding, or as a perception of the heart; or, which seems the truth, as including both.”

3^ Butler’s at first sight singular phrases “sentiment of the understanding,” “perception of the heart” are perhaps chosen deliberately to make this point. It is hard to believe in Whewell’s “emendation” of them to “perception of the understanding,” “sentiment of the heart.” Whewell should have remembered Pascal’s Le coeur a ses raisons, [And I believe there is truth in the contention that the more explicit recognition of “sentiment” in the Dissertation is due to study of Hutcheson.]

4^ Preface, “reflexion or conscience or approbation of some principles, or actions, and disapprobation of others.” “Conscience, or reflexion, in the nature of man plainly bears upon it marks of authority over all the rest”; “that natural authority of reflexion”; Sermon II. “that particular kind of reflexion which you (i.e. Butler) call conscience”; “a superior principle of reflexion or conscience in every man … which passes judgment upon himself.” Dissertation, “a capacity of reflecting upon actions and characters, and making them an object to our thought”; “our perception of vice and ill-desert arises from, and is the result of, a comparison of actions with the nature and capacities of the agent.”
  Passages like this show that “conscience,” “reflexion,” “thought directed on the quality of our own acts and characters” are, for Butler, strictly synonyms.

5^ When Butler speaks in the Preface to the Sermons of a way of treating the subject of morals which “begins from inquiring into the abstract relations of things” and concludes that “vice is contrary to the nature and reason of things,” the language shows that he is thinking in the first instance of Clarke, though he may also mean to include Wollaston, to whom he refers later on as a “late author of great and deserved reputation.” Cudworth’s Eternal and Immutable Morality was published too late to be referred to in Butler’s volume.

6^ Preface. “The first seems the most direct and formal proof, and in some respects the least liable to cavil and dispute.”

7^ “This approbation and disapprobation (sc. of a ’due concern about our own or others’ happiness’ and of the neglect of it) are altogether different from mere desire of our own or of their happiness, and from sorrow upon missing it…. In one case, what our thoughts fix upon is our condition … in the other our conduct.”

8^ The most unqualified statement is that of Sermon III. ad fin. “Conscience and self-love, if we understand our true happiness, always lead us the same way. Duty and interest are perfectly coincident.” But here Butler is speaking from the pulpit and is free to assume the teachings of the Christian faith about God and immortality. In the Preface,i> where he is considering things as they appear without the light of faith, he has, of course, to speak without this confidence.

9^ Dissertation. “It should seem that a due concern about our own interest or happiness and a reasonable endeavour to secure and promote it, which is, I think, very much the meaning of the word prudence in our language; it should seem that this is virtue, and the contrary behaviour faulty and blameworthy; since, in the calmest way of reflection, we approve of the first, and condemn the other, conduct, both in ourselves and in others.”

1^ Sermon II. “The appearance there is of some small diversity amongst mankind with respect to this faculty, with respect to their natural sense of moral good and evil; and the attention necessary to survey with any exactness what passes within, have occasioned that it is not so much agreed what is the standard of the internal nature of man, as of his external form. Neither is this last exactly settled. Yet we understand one another when we speak of the shape of a human body. So likewise we do when we speak of the heart and inward principles, how far soever the standard is from being exact or precisely fixed.”

11^ Sermon XI. “It may be allowed, without any prejudice to the cause of virtue and religion, that our ideas of happiness and misery are of all our ideas the nearest and most important to us; that they will, nay, if you please, that they ought to prevail over those of order, and beauty, and harmony, and proportion, if there ever should be, as it is impossible there ever should be, any inconsistence between them; though these last two, as expressing the fitness of actions, are real as truth itself. Let it be allowed, though virtue or moral rectitude does indeed consist in affection to and pursuit of what is right and good, as such; yet, that when we sit down in a cool hour, we can neither justify to ourselves this or any other pursuit, till we are convinced that it will be for our happiness, or at least not contrary to it.”
  The phraseology here, “it may be allowed,” “if you please,” “if there ever should be, as it is impossible there ever should be,” “let it be allowed,” makes it, to my mind, almost certain that the position is meant to be one which Butler himself does not accept, but is content to assume for the purpose of reasoning with an audience who regard “self-love” as the one rational rule of conduct. He had said at the beginning of the discourse, “there shall be all possible concessions made to the favourite passion, which hath so much allowed to it, and whose cause is so universally pleaded; it shall be treated with the utmost tenderness, and concern for its interests.” We must therefore expect a touch of “irony” in the sermon. It is monstrous to discuss the incriminated passage without taking into account Butler’s sarcastic words about the indulgence he proposes to show the “favourite passion” of his auditors.

12^ Preface. “But does it much mend the matter, to take in the natural authority of reflexion? there would indeed be an obligation to virtue, but would not the obligation from supposed interest on the side of vice remain? If it should, yet to be under two contrary obligations, i.e. under none at all, would not be exactly the same, as to be under a formal obligation to be vicious.”

13^ Loc. cit. “But the obligation on the side of vice does not remain. For the natural authority of the principle of reflexion, is an obligation the most near and intimate, the most certain and known; whereas the contrary obligation can at the utmost appear no more than probable;... and thus the certain obligation would entirely supersede and destroy the uncertain one, which would yet have been of real force without the former.” (The important words here are “utterly supersede and destroy,” which show that Butler is not thinking of anything in the nature of a “gamble in futures”.)
  “Take in then that authority and obligation, which is a constituent part of this reflex approbation, and it will undeniably follow that though a man should doubt of everything else, yet that he would still remain under the nearest and most certain obligation to the practice of virtue.” “Though men should be ignorant of or disbelieve any authority in the universe to permit the violation of this law; yet, if there should be such authority, they would be as really liable to punishment, as though they had been beforehand convinced, that such punishment would follow.”
  Thus “hell” does not come into the argument, since Butler holds that the reasoning would be equally valid if no one had ever heard of “hell,” if “through stupidity” mankind were “universally ignorant” on the subject.

Top ↑

This paper was originally presented before the University of Cambridge as the Leslie Stephen lecture of 1927.

When the University of Cambridge did me the high honour of inviting me to deliver this lecture, I could not but feel that the invitation, in some sort, determined for me the subject of my discourse. The character of the field of letters in which Sir Leslie Stephen achieved such well-deserved renown made it plain that my theme must be the life or doctrine of a British philosopher, by preference a philosopher of that eighteenth century which Sir Leslie Stephen himself had done so much to illustrate by his writings. To the occupant of a Chair of Moral Philosophy in the University of Edinburgh the philosopher could not well be anyone but David Hume, the most famous, even if not the greatest, of all Scottish metaphysical thinkers and, with the single exception of Walter Scott, the most distinguished man of letters whose life has been closely connected with the city of Edinburgh. The philosophical thought of Hume, as a whole, is no topic for the discourse of an hour, but that space of time may be profitably, and I trust not un-entertainingly, spent on a consideration of the side-issue raised by the once notorious tenth section of the Enquiry Concerning Human Understanding which deals with Miracles. Historically, this essay is interesting by reason of the scandal it created, and was perhaps intended by its author to create. To it, and not to the Dialogues Concerning Natural Religion, which were |331| not published during Hume’s life, our philosopher owed the ill-repute he enjoyed alike with the orthodox Presbyterians of Aberdeen and the High Anglican Churchmen of Oxford, the set of Johnson and Wesley; to it, in the main, he owes the admiration of modern “anti-clericals,” who have regarded him as a hero of militant “free-thought.”

The entertaining feature of the situation, to the reflective mind, is that this reputation, for good or ill, is quite undeserved. It is as certain as anything in biography can be that Hume was, in point of fact, no anti-clerical zealot, but an amiable and easy-going man of the world whose chosen social circle consisted largely of the “moderates” among the Edinburgh Presbyterians. The members of that circle, to be sure, were not men of the faith which removes mountains and conquers the world. But they valued established beliefs as a bulwark of comfort, peace, and social order, and would have been the last persons to sanction a violent and wanton attack on any set of doctrines which serve to keep the “vulgar” in their place and to guarantee the “thinking” minority against disturbers of their ease and leisure. They may fairly be presumed to have understood that Hume’s assault on the “bigotry, ignorance, cunning and roguery” of that considerable “part of mankind” who profess belief in the miraculous meant very much less than it looked, on the face of it, to mean. They would be confirmed in the suspicion, if they entertained it, by observing that, for the purposes of the Enquiry, the whole section is superfluous, while the known character of Hume makes it impossible to account for its presence as the irrelevancy of the fanatic who has got, for the time being, upon his dangerous topic. The irrelevance is, indeed, so manifest that it seems best explained by the suggestion of Hume’s |332| learned editor, Mr. Selby-Bigge, who supposes that the philosopher’s motive in the assault was a simple craving for notoriety at any cost. The “learned world,” as we know, “said nothing to the paradoxes” of Hume’s Treatise; he was determined that it should say something to them in their amended version, and none too scrupulous about the methods by which publicity was to be ensured. Hence the combined violence and irrelevance of a section which was at least certain to get the Enquiry talked about, as it very effectually did.

A similar explanation can, as I hope to show, be given of the apparently strange logic of Hume’s argument. On the face of it, there would seem to be something amiss with reasoning which proceeds from the principle that “a wise man proportions his belief to the evidence” to the conclusion that in a vast, if none too well defined, field, the “wise man” will simply refuse to consider “the evidence” at all. We cannot be surprised that Hume’s admirer, Huxley, should have been much more perturbed by reasoning of this kind than his unfriendly critic Green, whose objection to the argument, indeed, does not go further than to urge that it does not come with the best of grace from the mouth of our professed sceptic. If we look more closely, I believe we shall see that the reasoning, interpreted in the light of Hume’s professed general doctrine, certainly proves something, but something very different from what Hume had suggested by his boast of having discovered an argument which “will, with the wise and learned, be an everlasting check to all kinds of superstitious delusion.” What has really been proved, as Hume himself says at the end of the whole discussion, would certainly not destroy the faith of the average Presbyterian of the eighteenth century, though it should leave a thinking Presbyterian dissatisfied; as Hume does not say, it |333| should also leave the serious believer in science at least equally concerned. The “academic” or “sceptical” philosopher of Hume’s own type is, in fact, left in the position of an amused spectator of the conflict between two irrationalities. We miss half of Hume’s irony unless we understand that it is meant to hit not only “dangerous friends or disguised enemies to the Christian religion,” but also “dangerous friends or disguised enemies” to Newtonian science. I trust I need not say that I do not myself regard amused detached contemplation of either Christianity or natural science as a right attitude in a rational man. But it is an attitude very characteristic of the century of so-called “good sense,” and none the less likely to be the secret attitude of David Hume, that it is hard to say which would have been more enraged by it, if he had understood it, John Wesley or T. H. Huxley.

It is desirable, before we go any further, to set out the steps of the argument we are to consider as briefly and baldly as we can. If we try to do so, it will be found, I think, to fall into eleven successive propositions, which I will state in order, with a minimum of exegetical comment. We begin (1) with the proposition, taken from Archbishop Tillotson, that the evidential value of the testimony of others, so long as it is considered as the testimony of others and nothing more, is always inferior to that of our own senses, a thesis I should not be too ready to concede myself without a great deal of qualification, since in many cases I should put vastly more confidence in the report of a trained observer than I should in my own eyes. (2) We are next told that this general principle is now to be applied to “accounts of miracles and prodigies found in all history, sacred and profane.” “Miracle” is here apparently equated with “prodigy,” and neither word receives any definition. In the course of the argument, we shall find two other |334| incidental definitions of “miracle” introduced. Unfortunately, the definitions are not coincident, and to adopt either seriously obscures the nature of the reasoning. For the present, so far as can be seen from the context, “miracle” and “prodigy” both mean any very unusual and unexpected event, anything which, to use a definition I once heard given by a divine, makes one say “O!” (3) We now take it into account that some events are found in experience to be “constantly conjoined”; other conjunctions are more variable. Here we must, of course, remember that it is part of Hume’s general metaphysical doctrine that there is no “necessary connection” between events. All events are separate, and there is nothing in the character of any event which demands that it should be continued in one way rather than in another. All our information is that certain types of event have been found, we do not and cannot know why, to be continued in certain ways and not in others. (4) A wise man, then, will always “proportion” his belief in any statement about a succession of events related to have taken place “to the evidence.” This is interpreted to mean that he will count the number of “instances” in which a conjunction of the kind in question has occurred, and the number of instances in which one member of the conjoined pair of events has occurred without being continued by the other. It is on this counting that the wise man will base his judgement of the probability of the alleged narrative.

(5) Further, all inference is founded solely on “our experience of the constant and regular conjunction” of events and on nothing else. Among the events covered by this principle are the true and false statements made to us by others. Hence the principle applies, among other things, to our inferences about the trustworthiness or untrustworthiness of human testimony. (6) Hence, |335| when we are offered testimony to alleged facts which “partake of the extraordinary and the marvelous,” there is a “conflict of opposite experiences.” We have, in fact, to balance our knowledge that the unusual and unexpected sometimes happens against our knowledge that informants sometimes, from various causes, speak falsely, whether intentionally or not. This is why it was reasonable in “the Indian prince” to be incredulous when he was first told of the freezing of water in our northern winter, though it would have been unreasonable to persist in his incredulity if he had had numerous independent reports to the same effect from informants with no motive for deception. The general result, so far, then, is that the strangeness of an alleged sequence of events diminishes the probability of its reality, except when the testimony for the sequence is so strong that it would be still stranger that all the witnesses should be mistaken or untruthful.

(7) There is one case in which the value of testimony is not merely diminished by the “conflict of experiences,” but actually reduced to zero. This is the case when the event testified to is a “miracle,” that is to say, a violation of the “laws of nature,” or, what comes to the same thing, of “the common course of nature.”This would be the case with testimony to the resuscitation of a dead man, since such a conjunction between death and subsequent resuscitation has “never been observed.” There is “uniform experience against it.”

Here we observe that the reasoning is affected by the introduction of a disturbing element, a new definition of “…miracle.” We have been concerned with the general question of the worth of testimony to the unusual and unexpected in genere, and it has been assumed that we know nothing of “laws” or “patterns” in nature which prescribe one continuation of a course of events rather |336| than any other. The only difference permissible according to our fifth proposition is that between a familiar and an unfamiliar continuation, and the only conclusion to which we are really entitled is that it is natural to be incredulous when we are told of the unfamiliar, as the Indian prince was mistakenly incredulous about the freezing of water. The shift from the unfamiliar to the “contrary to uniform experience” is confusing and unjustified, but indispensable to the further development of the argument, and directly causes the inconsequence on which Hume’s critics have remarked. It is quietly forgotten that, on the premises, there cannot be said to be “uniform experience” against the resuscitation of the dead man or any other sequence of events. At best I have only a uniformity within the range of my own experience to urge; a narrator who professes to have seen the resuscitation is actually appealing to his own experience as the foundation of his story. Thus, unless I am to assume that my own personal experiences are the standard of the credible — and if I do assume this, there is an end of all correction of expectations — it is a petitio principii to say that there is “uniform experience” against any event to which any man claims to be able to testify, and there is no basis for any distinction between the “miraculous” and that which is merely unfamiliar, and therefore startling, to a particular person.

The paradox that the principle that “belief should always be proportioned to evidence” justifies, in a certain class of cases, rejection of testimony without examination, is only established by the illegitimate device of changing a fundamental definition in the course of the argument. A footnote which introduces a further distinguo seems to show that Hume is a little uneasy about this procedure. It admits that the testimony may, after all, really justify belief in a fact which |337| seems “miraculous,” i.e. a fact to which my experience affords no analogy, though it does not justify belief in the “miraculous” character of the fact. This leads to a fresh definition of a “miracle” as “a transgression of the laws of nature by a particular volition of the Deity or by the interposition of an invisible agent”. Apparently, then, in spite of what has been said, Hume would admit that there may be adequate testimony to the resuscitation of a dead man. There might be evidence which would require us to admit the fact, though none which would establish the point that the fact was due to a “particular volition of the Deity.” (The reference to other “invisible agents” appears to be a needless rhetorical amplification.)

This new definition of “miracle,” sprung on us in a footnote, seems to be the most unfortunate feature of the argument. It obviously disposes at once of all which had apparently been secured by the appeal to the inviolability of “laws of nature,” since it permits us to accept as facts the sequences of events which that appeal was intended to rule out, provided only that we do not profess to have proved by testimony that the facts have “a particular volition of the Deity” as their cause. In this version of the matter, there is no story of legend or folk-lore which a sufficient number of testimonies might not require us to accept as a genuine account of facts, provided only that we eliminate all reference to “the Deity” or “invisible agents.” It is not clear how a principle compatible with this position is to be of any use to “the wise and learned” as an “everlasting check to all kinds of superstitious delusion”; yet it is clear that this, and not the much more radical procedure of dismissing a whole class of alleged events without scrutiny of the testimony for them, is the only consequence really compatible |338| with the principles on which Hume’s polemic professedly rests. The upshot, after all, is no more than the statement, which might be made about all testimony to anything, that no testimony can establish a fact unless the falsity of the testimony would be more unlikely than the unreality of the alleged fact. This is virtually admitted when Hume proceeds to formulate his eighth proposition.

(8) In no actual case of a reported “miracle” do we find the testimony to be of this strength. This is, of course, itself an allegation about a fact, and we observe that before we are entitled to make it we must tacitly surrender what appeared to have been secured by the appeal to “inviolable laws.” If the weakness of the testimony is to be a relevant consideration in determining the credibility of the supposed event, we must examine the testimony before we can pronounce it to be weak. The proposed dismissal of testimony without examination has thus led us to nothing, and ought to be eliminated from the reasoning as a superfluous complication. How completely the ambiguity of the word “miracle” has vitiated the argument is shown when Hume goes on, in this connection, to urge it as a grave objection to stories of alleged marvels in the career of Mohammed that Plutarch, Livy, and Tacitus record marvels as occurring in an earlier age at Rome. This is meant to suggest the objection that the marvels, if they occur, must be regarded as “evidences” in favour of a religion, but evidences in favour of incompatible religions may be considered as destructive of one another. Plainly, this consideration, whatever it may be worth, has no bearing on the value of the historical testimony, considered simply as testimony to the actual occurrence of an unusual fact. It concerns not the fact, but the “theological” interpretation of the fact. Thus it appears |339| to be an authenticated fact that the first volley of the firing party told off to execute the Bāb severed the cord by which the victim was secured without doing him any injury. Hume’s reasoning would require us to discount the excellent testimony to the fact, on the plea that similar stories related in the Acts of Christian martyrs may be regarded as so many testimonies against the occurrence of the fact in the case of the Bāb.

(9) We now reach what is meant to be the conclusion from all the considerations so far put before us. No testimony to a “miracle” has ever amounted to a probability, much less to a proof, and even if the testimony, in any case, did amount to proof, it would be opposed by another proof derived from the “very nature of the fact which it would endeavour to establish.” Here, as it seems to me, confusion of thought reaches a maximum. To know that it is a fact that no such testimony has ever amounted to a probability, we must, of course, have examined the amount and character of the testimony, and thus have made it a duty to do the very thing Hume originally proposed to show superfluous. It is at least difficult to understand the suggestion that there could be a proof of any proposition if there were also a proof of a second and incompatible proposition, unless the remark is meant as a mere rhetorical amplification of the statement that proof of the “miraculous” is not forthcoming. The final appeal to “the very nature of the fact,” which proof of the “miraculous“” would “endeavour to establish,“” takes us back again to the position which has just been incidentally abandoned, that there is a whole class of cases in which testimony may properly be dismissed without examination, and this, if justified, makes the weakness of the testimony, the very point on which Hume is specially insisting, irrelevant.

|340| (10) We next have a curiously qualified retractation of this attempted return to the abandoned position. The occurrence of “violations of the usual course of nature” may actually be proved by testimony, but they cannot be so proved as to make them “the foundation of a system of religion.” Since we have never been told exactly what is meant by a “religion,” this admission is not very enlightening, but it is illustrated by a singular example. There might be testimony which would not merely make it probable, but actually prove, that the whole earth was covered by a mysterious darkness for the first week of January 1600. If the testimony were abundant and of good quality, we should have to accept this as a fact, and look for an explanation, though our present knowledge of science suggests none. But no testimony whatever could give us any ground to believe that Queen Elizabeth died on January 1st of that year, was buried, reappeared again and resumed the government in the following month. And “if this miracle should be ascribed to any new system of religion,” we might dismiss the testimony without examination. I confess I cannot see on what ground Hume makes any distinction between the two cases he has, with notable bad taste, been pleased to imagine. If it is true that experience is our only criterion, the two imagined sequences seem exactly alike in being wholly startling and unfamiliar, and I should have thought that the same kind and amount of testimony might serve indifferently to accredit either. There is as much or as little precedent for one as for the other. And, again, it is hard to understand why testimony to either, seeing that on Hume’s own theory it would have some initial value, however slight, should lose all that value merely because the belief in the event had led ex post facto to the appearance of a “new system of religion.”

|341| Presumably, the thought in Hume’s mind is that the partisans of a “system of religion” already in existence, especially of one fighting its way to victory over its rivals, are likely to be unduly disposed to believe in, or even to invent, stories of marvels which recommend it, and that this diminishes the evidential value of their testimony. This, no doubt, is a truth, though not a specially novel one. But he appears to be using this obvious reflection in a surely illegitimate way to destroy the whole value of testimony, even outside testimony, to marvelous events which have preceded and caused the “new system of religion.” Where the “system” has actually been called into being by antecedent belief in the occurrence of a “miracle,” manifestly the fact that the belief has had this historical effect cannot in any way diminish the antecedent probability of the truth of the belief, except in so far as we might have reason to suspect the adherents of the “new system” of successful suppression of evidence which would tell against the truth of their story. This would be a relevant consideration, but one which could only be taken into account on the assumption that we scrutinise testimonies and do not dismiss them unexamined.

(11) We come at last to the surprising and famous volte-face with which Hume ends his essay. There is nothing in what we have heard, in spite of obvious allusions in the worst of taste, which plays into the hands of opponents of “our most holy religion.” For that is founded altogether on faith, not on reason. The Scripture history is full of incidents which are infinitely improbable. The testimony for them, judged by the principles which have been laid down for our guidance, amounts to nothing at all, and we might therefore suppose that the benefit promised at the beginning of the essay to the “wise” would prove to be the abolition of |342| Christianity. But Hume reminds himself at the close that since, on his own showing, “the Christian religion … cannot be believed by any reasonable person” without a miracle, anyone who does believe “is conscious of a continued miracle in his own person, which subverts all the principles of his understanding,” and the arguments against “miracles” must obviously fail when addressed to a person who has the actual present experience of one in himself. The lameness of the conclusion, when it is compared with the vaunting of Hume’s exordium, has naturally given rise to the current view that it is a mere piece of mockery — a transparent substitute for the true ending, like the finale of a Euripidean tragedy on Dr. Verrall’s interpretation of the poet. That there is mockery in the language is undeniable. The question I propose to raise is whether, after all, the conclusion, satisfactory or not, is not that which follows from the reasoning on Hume’s principles, and the violent contrast between exordium and peroration itself a part of the mockery.

If we are to find any coherence in Hume’s argument, we must, in the first place, eliminate an ambiguity which can hardly be accounted for except by the writer’s determination to make a sensation, the ambiguity of the term “miracle” itself. A “miracle” may mean either of two very different things; it may mean simply an unusual and arresting event, an event “out of the common course,” or it may mean an event, not necessarily particularly unusual, which is held to disclose, as most events do not, the direct activity of God as author and controller of events; a miracle may be either a mere “wonder” or it may be a “sign,” and, as everyone knows, in the language of theology no event is called a “miracle” unless it combines both characters. But the second is, from the theologian’s point of view, |343| the more important. St. Thomas,1 for example, reckons among minor “miracles” such events as the relief of a “fever” by the offering of a prayer, a process which he must have believed to be illustrated by daily occurrences, and Dante is following the same doctrine when he appeals to the story of the opportune cackling of the Capitoline geese as evidence that “miracles” prove the divine selection of the Roman people for universal empire.2 Manifestly it is clear that there are two quite distinct, though connected, questions which need to be carefully distinguished: (1) What sort and amount of evidence is needed to justify belief in the reality of an unusual occurrence? (2) Whether such occurrences, if there is evidence for them, can rightly be employed as proof of the control of events by a divine purpose? It is one question whether there can be adequate evidence of the occurrence of events “out of the common course,” another what the evidential worth of such events as testimony to the doctrines of a “religion” may be. The first question belongs to inductive logic, the second to theology, and nothing but confusion can come of the attempt to treat the two questions as one.

It is also clear on the face of it that Hume’s essay begins as a discussion of the first question, and that the introduction of the second is a piece of irrelevance. The conclusion which was to deliver the “wise” from “superstition” once and for all should have been that events sufficiently out of the daily routine are so inherently improbable that any counter-probability suggested by the amount of testimony in their favour may be dismissed as infinitesimal. This, if it could be established, would make the raising of the theological issue superfluous. Or, at most, all that it would be relevant to urge would be the subsidiary consideration that the theological convictions of |344| witnesses of the alleged events are likely to be a source of antecedent bias. Even this reflection should, in strict logic, be dispensed with, if it can be independently established that no amount of testimony, biased or not, has any weight, if only the event testified to is sufficiently strange. Now if Hume had confined himself to this contention, his argument would have gained in force and coherence: there would have been none of the perplexities introduced by the repeated change of the initial definition of a “miracle,” nor would the whole reasoning have been imperiled, as it is, by the damaging concession that a departure from “uniform law” may after all be accepted as a fact, if the testimony is abundant and good enough, or by the arbitrary decision that testimony might be sufficient to prove the occurrence of a world-wide darkness of a week’s duration and yet not deserve any examination if it were offered in favour of a "resurrection." We do less than justice to Hume’s acuteness if we imagine that he was not alive to the havoc made in his argument by this confusion of the issues, and it is only reasonable to suppose that the irrelevance is due to a purpose. Without the sensational attack on the theologians, the main argument would probably have attracted no particular attention from anyone; the Enquiry would have been as little talked about as the Treatise, and Hume was, above everything, determined that he would be talked about.

What remains to be considered, then, is the question what Hume’s main argument amounts to when all the irrelevance has been removed, and it is taken simply as an argument about the value of testimony to the strange and unprecedented. Even when taken in this way, at first sight the argument is bound to seem curiously wrong-headed. We are apparently told that assent should always be based on a careful weighing of the |345| direct and indirect evidence on both sides of a question, and this is, then, strangely enough made the ground for asserting that, if only an alleged occurrence is unusual enough, we need not weigh the evidence produced for it. We may confidently dismiss our witnesses unheard. If this were really Hume’s meaning, it is clear that his conclusion and his premises would be oddly at variance, and it is not surprising that a disciple like Huxley, who was anxious to maintain the conclusion, but to maintain it with some appearance of logic, should have restated the case in a way which abandons Hume’s main contention. With Huxley, the main weight of the conclusion is made to rest on the assertion that in actual fact the testimony which has been produced to the occurrence of "miracles" has always been dubious or inadequate. Whether this is the fact or not, it is at least an appeal to what is itself ascertainable fact, and can only be made after that very scrutiny of the testimony in the particular case from which Hume promises to deliver "the wise." If we are careful, however, to remember certain fundamental positions of Hume’s general philosophy, we shall see that, before we can get at the real meaning of his argument, we have to translate its terms into the language of Humian scepticism, and that when we do so, the reasoning proceeds to a conclusion which is valid enough, though quite unsensational.

We have to bear in mind, to begin with, Hume’s peculiar doctrine about the nature of assent or belief, a doctrine proclaimed loudly enough both in the Treatise and in the Enquiry. A belief, we must remember, is explicitly asserted to be simply "a lively idea" associated with a "present impression"; it is only in respect of its superior "liveliness" that a belief differs from a fancy, or assenting to a proposition from merely |346| framing it. "I believe this," we are told in so many words, means only "I feel at this moment a strong propension to consider things in this light," and it is because belief means no more than this, that Hume professes to believe his own metaphysic, though, as he ingenuously says, he forgets about it and virtually denies it whenever he mixes in the social life of his fellows. While he is meditating alone in his study, though at no other time, he feels a strong propension to consider things in the fashion of the Treatise, and this, of itself, is believing, for the time, in the Treatise and its doctrines. It is another of Hume’s convictions that, though beliefs have causes, they never have sufficient justifying reasons, and that in the sequence of events, there are "customary conjunctions," but never discernible "connections." There is no pattern in the course of events in virtue of which any event should require one continuation into the future rather than another. Each event has its own special character and all events are loose and separated. There is no discoverable reason why any one event should not be "conjoined" with any other whatsoever as its sequel. What events are followed by what we discover only by experience, and experience is a purely passive awareness of sequence. It follows that, when we hear from Hume of "uniform experience" and of "laws of nature," uniform experience can mean no more than a type of sequence which has been regular so far as our own recollection and that of the other persons with whom we are in communication, goes. (Indeed, it is plain that it is only by the courtesy of an opponent that Hume is entitled to bring in the reference to the experience of any one but himself. A second person who makes a statement to me about his experience may always be mistaken or untruthful, and, in the end, it will be my own personal |347| acquaintance with the sequence of events by which I have to judge whether or not he is a credible witness.) Similarly, an "inviolable law of nature," when we translate the words into their equivalent in Humian metaphysics, means no more than a type of sequence which, so far as my own recollections, eked out by those of any other persons in whom I rightly or wrongly put credence, go, has been uniform.

Next we have to remember that the same considerations must be applied in interpreting the statement that our knowledge of the sequences in nature depends on "customary experience." This does not mean, and in the Treatise Hume is at pains to make the point clear, that there is any discoverable reason why there should be a fixed routine in the course of events, and a routine which each of us can discover from the small fragment of the sequence open to his personal observation. There is no logical ground for expecting that the sequence of events will conform to any rule; any one event might perfectly well be succeeded by any other. All that is meant is that it is an inexplicable, non-rational tendency of the human mind to expect that the usual will happen and that the unusual will not. Repetition gives rise to a subjective "association of ideas" and therefore to expectation; in fact, the probability of an event, as is explained in the Treatise, means neither more nor less than the degree of "vivacity" with which some one imagines, and therefore anticipates, it. "Customary sequence" is thus only a cause, not a justifying ground, for our anticipations. To put the point quite simply, all our judgements are judgements, true or false, about actual fact; there are no judgements of value, neither a logical ought nor a moral ought. Just as "this is right" means only "the disinterested spectator, as a fact, contemplates this with |348| pleasure," so "this is probable," "this is certain," means only that "actual observers expect this with a less or greater degree of confidence."This universal reduction of all propositions to statements about actual occurrences is the most characteristic and important peculiarity of Hume’s whole philosophy. The ideal before his mind is the same which inspires the chief work of Avenarius, the elimination from all assertions of every element which is not "pure experience," simple record of the event without interpretation or valuation. The problem he has not faced is that directly suggested by the Critique of Pure Experience, the question whether intelligence, as distinguished from mere insignificant reaction to stimulus, would not vanish completely from a community in which the ideal had been realised.

Let us, then, restate Hume’s main argument in the terms of his own philosophy. It will be seen, I think, that it amounts to this. It is a fact, and a fact of which no explanation can be given, that repetition gives rise to "associations of ideas," and that the strength of these associations depends on the frequency of the repetition. The more often I have seen A followed by B, and the less often I have seen A occur without being followed by B, the harder I find it to believe that A will occur, or has ever occurred, without being followed by B. Similarly, if I have never seen A followed by C, I shall not imagine C, and consequently shall not expect C to occur, as a continuation of A. This is the ultimate causal explanation of habits of thought and expectation. They are, in fact, prejudices without logical value, as the Treatise declares in deliberately provocative language, but the strength of the prejudice may be, as a fact of my mental makeup, invincible. This, and no more, is all that Hume is entitled, on his own principles, to mean |349| when he talks of the inconceivability of the violation of a uniform law of nature. Properly speaking, there are no laws of nature to be violated, but there are habits of expectation which any one of us, as a fact, finds himself unable to break through. Since, again, we discover, also as a fact for which we can give no reason, that it is "customary repetition" which appears to be the foundation of these habits, it is also the fact that we expect a certain sequence with the greater confidence the more familiar its type is to us, and that, if an alleged sequence of events is sufficiently startling and contrary to our expectations, we feel a stronger propensity to consider the statements of witnesses as mistakes or lies — things of which we have some experience — than to accept them as true. We find it easier to imagine vividly that our witness is deceived, or is deceiving us, than to imagine the events he describes. The conclusion to which we are led is thus, like the premises from which we started, a simple proposition of fact. Men find the unusual hard to imagine in proportion to its unusualness, and there is a point, for any one of us, when the difficulty amounts to a psychological impossibility. When this point is reached, the unimaginable event is a "miracle."Thus the conclusion to which the argument is leading us is really the assertion of fact that a "miracle" is not believed in by anyone, being ex vi termini simply an event so unexpected that we find it easier to imagine the falsity of the testimony than to imagine the occurrence of the event.

Now, if Hume had ended his essay at this point, it could have brought him no notoriety, and it would not have proved anything which could well be disputed. The pious and the impious, the orthodox and unorthodox alike, might well have agreed that there are some things which any man does find incredible. The trouble |350| is that men differ so much from one another in the matter of what, in particular, each finds incredible. What we want to know, if we are to write or read history, is not what a given man finds credible or incredible, but what we ought as rational beings to pronounce credible or incredible, and the philosopher of "pure experience" can give us no guidance here. Hume, for example, indicates that in his opinion a resurrection from the dead is flatly incredible, no matter what the apparent testimony for it may be, whereas a week of worldwide darkness is credible, if there is enough testimony for it, though both events seem equally to baffle our powers of rational explanation. Any one of us who does not feel the special "propensions" of Hume on this point, may obviously ask him to justify his position, if he can, by showing that there is some difference in principle between the two cases. What is more, any man who finds any event whatever not wholly unimaginable may pertinently ask Hume why that particular event should be placed in the category of those which may be disposed of as unreal without examination of testimony. It is this obvious reflection which both justifies and demands the apparently inconsistent concession to the Christian believer with which Hume has seen fit to end his essay. The tone of persiflage is manifest throughout the paragraphs, but the whole is not persiflage; there is a real concession which is absolutely demanded by Hume’s own radical "positivism."

The main result really reached had been that, in point of fact, men find it hard to believe the marvelous, and if the marvel is sufficiently astounding, they refuse to believe. In plain language, this ought to mean that no one ever does believe in the reality of a sequence of events quite unlike the routine of his customary experience. |351| But this conclusion would be manifestly false in fact. For Hume could not deny that, to take the most familiar example, there were many sincere orthodox Christians in his own society and that they did, as a fact, believe in the reality of certain events to which their customary experience afforded no analogy. It might be argued, though not very reasonably by an avowed irrationalist, that the orthodox have no right to hold their convictions; but the fact that they did hold them, and hold them with assurance, was beyond dispute, and it has to be shown that this fact is itself consistent with Hume’s own principles. Otherwise the principles would themselves be completely discredited as leading, by logical necessity, to a conclusion false in fact. This explains the serious motive for the apparently paradoxical concessions which Hume proceeds to make. There would be a real paradox, if it were part of Hume’s case that "customary experience," and it only, provides a rational justification for our beliefs about the course of events. But his position is that none of these beliefs have any rational justification; belief, as the Treatise says, is more of the nature of sensation than of the nature of reason. The only problem that can be raised about our beliefs is that of their cause, and Hume has insisted strongly on the point that all we can say about the cause of belief is that, as a fact, it is commonly produced by "customary experience." We could not say that it cannot be produced in any other way, for to say so would be to go beyond the limits of the factual. And the existence of a single genuine believer in the unusual is empirical proof that, as a matter of fact, "customary experience," though the commonest, is not the only cause of belief.

Now Hume has been careful to protest against any assumption that there is a rational connection of any |352| kind between customary repetition and belief. That the repetition of a certain sequence of events should make us expect its recurrence with the greater confidence, the oftener the repetition has been noted, is, according to him, a fact of human nature which we can observe, but for which we can allege no justifying reason. (It is on a par with the gambler’s belief in the continuation of a run of luck, a belief neither more nor less rational than the opposing belief, also found among gamblers, that the "luck must change".) It is thus only what might be expected — if in a world where there are only "conjunctions" we can talk of what may be expected — that "custom" should be only the customary, not the invariable, cause of belief. And, since there is no discoverable connection between repetition and belief, you cannot tell the man who confidently believes in the reality of the unparalleled that his belief is unreasonable. He finds in himself, as Hume takes care to put it, "a determination to believe what is contrary to custom and experience." But, on Hume’s own showing, to believe what is contrary to custom is neither more nor less reasonable than to believe what is conformable to custom. It is less usual, and we have no right to say anything more. This is, in fact, what Hume does say, in more provocative language, when he speaks of the Christian believer as having a standing "miracle" within himself, which makes him proof against all reasoning against the miraculous. The strict meaning of the statement is simply that in this case the believer’s "propension to see things in a certain light" has causes which are not those of other men’s belief, nor of his own beliefs about the majority of things. Stated thus baldly, Hume’s conclusion would be at once proper to a consistently "sceptical" philosophy and quite acceptable to the vast majority of the orthodox. In fact, what would be shocking |353| to any man who was deeply religious as well as orthodox would be the suggestion that the motives of credibility in matters of religion are of the same order as those involved in believing or disbelieving a newspaper report of travel or exploration. Why the statement that this is not so should have been couched in language which was certain to create scandal can hardly be explained, except by Hume’s resolution to attract notice at all costs.

There is another side to the matter which demands a word or two. From the position of his own sceptical philosophy, Hume is not entitled to regard the belief of the man who "has the miracle in himself’’ as inherently more or less unreasonable than his own. But there can be no doubt in what light Hume was himself "strongly inclined" to view all such matters, and we can quite understand that he would contemplate the orthodox, whose "propensions" are so unlike his own, with a detached amusement. To his eminently secular mind they would be no better than entertaining oddities. But there is another party whom Hume must have found equally entertaining for precisely the same reason, the militant rationalist who assails orthodoxy in the name of science. Rational science goes outside the limits of a philosophy of pure experience in much the same way as dogmatic theology. The greater part of Hume’s Treatise had been devoted to an attempt to demolish the foundation-stones of a rational science of nature. Necessary connection, the permanence of substance, the extra-mental reality of the physical world, are all dismissed as superstitions. To use more modern language, no "constants" are left in nature, unless we can give the name of a "natural constant" to our own inexplicable and unjustified prejudice in favour of the usual and familiar. From this point of view, the man of science, |354| who builds on the "uniformity of nature," and the divine, who appeals to the immutable attributes of his God, are alike constructing dogmatic theories on a basis of "extra-belief." Spinoza, the typical assertor of natural necessity, and the theologians may both be beaten equally effectually with the same stick, since both go beyond the limits of fact in precisely the same fashion, by "feigning" that there is real "connection" where they should have been content to record mere "conjunction." It follows that "our most holy religion" does not stand alone in being founded on a "faith" which has no foundation in reason. The same faith is also the foundation of our most admirable science. The only difference which could be pleaded as giving an advantage to science is that the predictions of the scientific man, up to the present, have largely been found to be verified by the course of events; the predictions of the theologian, which concern an unseen order, are naturally incapable of this verification. On closer scrutiny the apparent advantage is found to vanish. For we have no means of knowing that the theologian’s predictions, too, may not get their verification; the predictions of science are never verified more than approximately, and we do not know that the course of events may not change its character at any moment in a way which would deprive them of even approximate verification. That they have proved trustworthy up to the present is no more than a curious and inexplicable "happy coincidence."

Hume’s light-hearted irony has, then, a double edge. As he says, in effect, the Christian believer is staking his all on a mere promise which, for all we can prove, may remain eternally unfulfilled. As he does not say in the context, but in effect urges all through his polemic against rationalist dogmatism, the convinced believer |355| in science is equally staking everything on a promise of the same kind to which the course of events may give the lie at any moment. The believer in God and the believer in an order of nature both have the same consciousness of a "miracle" within their own breasts. That either should excommunicate the other for their common guilt of "extra-belief" is the feature of the situation which the "academical" philosopher finds entertaining. The "naturalist" who derides his neighbour’s "groundless" anticipations of the joys of Paradise forgets that, on Hume’s showing, his own anticipation that the sun will rise tomorrow is equally groundless. If only one half of this condemnation of all anticipations of the course of things as alike unintelligent is actually expressed in the essay, the reason must be that notoriety was to be got by an attack on the Church; an attack on the Royal Society would pass unregarded.

“There is no such thing as the probability of a given event, there are only probabilities relative to a given set of premises, and such a set of premises inevitably includes some metaphysical assumption about the structure of the world as a whole.”

As a contribution to logic, Hume’s essay is thus an attack not so much on the credibility of "miracles" as on the validity of induction. His point, if one works it out, is that all inference from the present occasion to anything beyond itself presupposes some metaphysical theory of the structure of the world as a whole, whereas the world revealed to us by "experience" has no structure; it is a mere series of loose and separate incidents. Now this is half the truth. Even if it were the case, as Dr. Broad has so brilliantly shown that it is not, that scientific induction can be reduced, as Hume tries to reduce it, to a mere application of the theory of Probability, the difficulty would still remain. For — to appeal again to a principle which has been made specially clear by the work of Cambridge philosophers — there is no such thing as the probability of a given event, there are only probabilities relative to a given set of premises, and such a set of premises inevitably includes some |356| metaphysical assumption about the structure of the world as a whole. Hume himself makes one such wholly undemonstrable assumption of the utmost importance when he rests his whole philosophical edifice on the proposition, admitted by himself to be incompatible with other equally indispensable postulates, that "all our perceptions are distinct existences, and the mind never perceives any connection between its ideas," in other words, that all events are loose and separate. This assumption, once made, ought to dispose not merely of all dogmatism, theological or scientific, as Hume intended it should, but of every attempt to regard any one event as more or less probable than any other. In principle it excludes all inference from the particular occasion to anything beyond itself, and therefore leads direct to the denial of the very possibility of science, even when the pretensions of science have been reduced to the most modest proportions. If the doctrine is to be carried out regardless of consequence, even the most abstract pure logic and mathematics cannot escape condemnation. For if there is really no means of transcending the particularity of the particular occasion, even my statement that two and three make five can be no more than a record of the event that I have, on this occasion, counted with this result; I can have no guarantee that the result will be the same the next time I perform the counting. All I am entitled to say is that at present I feel a "propension" to consider the matter in that light.

It would be of no avail to introduce a distinction between particular data of sense and "universals" which pervade the data, in the hope of preserving, at any rate, scientific knowledge of the inter-relations of "universals," for, if all events are loose and separate, there can be no warrant for assertions about "pervasive" elements in them. The attempt at analysis itself already |357| implicitly transcends the supposed disconnection of events. Dr. Whitehead thus seems to me wholly in the right of it when he insists on the point that the "induction" indispensable to science is not to be regarded as a process of generalisation but as one in which we "divine" some characteristics of a particular future from the known characteristics of a particular past, except for the slight oversight by which he has spoken of the conclusion inferred to as though it must be "future." No one, I take it, knows better than Dr. Whitehead that the inference may equally well be from one particular past occasion to another equally past. But, of course, the very admission that it is a legitimate procedure to divine any of the characteristics of one occasion from what we know of another occasion demands the surrender of the assumption that "occasions" are simply loose and separate, or in other words, that "experience" is mere awareness of a series of disconnected "events" which are only externally "conjoined." I cannot dwell long on the point, but I would only urge that in principle this must mean that a scientific view of nature must be much more like that of Leibniz than like that of Hume. So far from being merely "conjoined" as earlier and later, events must have a pattern in virtue of which each brings with it traces of all that has gone before and is big with all that is to come. And to admit so much is to admit that, in the end, the course of events as a whole has a supreme pattern which appears, with the needful modifications, as the dominant factor in determining the patterns of its parts. What the dominant pattern is, in its detail, we naturally cannot say, since we never see more than fragments of it. But we can, on the supposition that such a pattern is really there, make "divinations" which are more or less true to the main scheme, and these |358| divinations, without which inference could not advance a single step, are the (commonly unconscious) metaphysical presuppositions which guide us in our judgements of probabilities. They furnish premises without which we should have no logical justification for pronouncing any one anticipation of experience more or less probable than any other. Were it true that "connection" is, as Hume supposed, not "divined" but simply "feigned," induction would not even be what Earl Russell once called it, a method of making "plausible" guesses. It would be guessing without any method, and there would be no sense in calling one guess any more plausible than any other.

The very existence of a dispute between the divine and the man of secular science about the reality of "miracles" is only possible because, unlike Hume, the parties are both, at bottom, rationalists in their conscious or unconscious metaphysics. In a purely non- rational world, any one occurrence is just as much or as little of a "wonder" as any other. You may say, with Hume, that it is only from customary experience that we can derive the expectation that an event will have a given continuation, or, with the most supranaturalistic of Mohammedans, that the only reason why anything happens is that "Allah almighty disposes it so"; the formulae are different, but their sense is identical. The distinction between the "ordinary" or "normal" and the "astounding" or "miraculous" (in the etymological sense) can only be made on the basis of a metaphysic which recognises a real connection of events by a coherent and all-pervasive pattern. Only where this is a common metaphysical dogma does it become justifiable to raise the question whether better testimony is required to establish the unusual and surprising.

If by a miracle we mean simply an unusual occurrence, |359| it may then become a mere problem in the estimation of opposing probabilities to determine whether the reality of such an occurrence is credible. From this point of view there could be no question of dismissing testimony unexamined as intrinsically worthless, but it would be possible to accept the general principle that the antecedent unlikeliness of the facts testified to make it reasonable to demand exceptionally abundant and weighty testimony. The one criticism one would feel inclined to pass upon Hume’s version of this principle would be that he forgets that the probabilities with which we have to reckon in matters of history are mostly not of the kind which admit of exact mathematical expression. A wise man is regularly determined in his decision to credit human testimony by "imponderables" to which no calculus can assign probability-coefficients.

The real issue, in the case of "miracles" appealed to as of significance for religion, is not the bare antecedent probability of unusual events. The unusual event gets its significance as a "miracle," in the religious sense of the word, from the conviction that it is an event in which the character of a divine purpose underlying the whole course of events becomes exceptionally transparent; it is a "sign" of the mercy, the justice, the power of God. It follows at once that our whole attitude towards the credibility of miracles is profoundly affected by our ultimate metaphysical position. The problem cannot even be discussed with any profit between two parties of whom one is a theist and the other an atheist or a pure agnostic. For they will differ profoundly about the nature of the pattern which binds nature into a connected system. On any rationalistic hypothesis, startling and singular events must be reasonably expected to occur from time to time, and there can, so far as I |360| can see, be no means of saying in advance how startling the surprises which the course of events contains may prove to be to us, who are familiar, after all, with so small a fragment of the whole. But, in an atheistic or neutral metaphysical scheme, there would be no reason to expect the surprises to wear any special character, or to be distributed in any special way over space and time. We should expect them to make their appearances as simple "freaks." If our philosophical world-scheme is definitely theistic, the case is altered completely. For we shall then conceive of the pattern of events as a whole not merely as providing a connection between them, but as providing a connection which is intelligent in the sense that, like the structure of a symphony, or a well-lived life, it exhibits the realisation of an end of absolute value. We should, thus, antecedently look for the "singularities" in nature and history to exhibit a special kind of concentration, exactly as the surprises in the construction of a great piece of music or the conduct of a life of wise originality exhibit the same concentration. The intelligence of the great musician or the great statesman shows itself neither in unbroken adherence to an iron routine nor in wild eccentricity. It reveals itself in the way in which conformity to routine, where there is nothing to be gained by departure, is combined with bold and original departure from routine because the situation makes the demand for it. A theist, conceiving of the pattern of events in the light of such analogies, will thus reasonably regard it as to be expected that surprises of a certain kind and in certain historic connections, surprises which contribute, so to say, to a plan, "worthy of God," should occur in history and that others should not, exactly as one would expect some kinds of musical surprises in a newly discovered symphony professing to |361| be by Beethoven, but would emphatically not expect others. Thus the difference in ultimate metaphysical outlook between a theist and a non-theistic philosopher would make a difference between the two sets of initial premises relatively to which each estimates the probability of certain events. It is not in the least unreasonable, for example, in a convinced theist to be satisfied with evidence for the resurrection of Jesus Christ which would not satisfy him of the resurrection of his next-door neighbour, since he may well ascribe to the resurrection of Christ a unique spiritual value for the whole history of the human race which he could not ascribe to the resurrection of his neighbour. We should be misconceiving the whole issue if we did not bear in mind that what is affirmed by the Christian creed is not simply the resurrection of some man, but the resurrection of just this one man and no other.

This is what I meant when I said, a little while back, that the motives of credibility to which the religious man, who is also a thinking man, appeals are of a kind quite other than those which Hume takes into consideration. The determining factor in leading him to believe in certain "miracles," assuming for the moment that he professes a religion which includes this belief, is his underlying conviction that the plan of the world is dominated by certain absolute values, and that these events are the most striking and transparent examples of the dominance of just these values. They are like the comparatively few critical moments in which we find a revelation of the inmost character of a friend and from which we then proceed to interpret the whole of his more ordinary behaviour. It follows, of course, that the whole question of the reality of "miracles," in any but the bare etymological sense of the word, is secondary to the much graver question of the legitimacy of a |362| theistic interpretation of life. The alleged occurrence of miracles cannot itself be rationally made a premise for the argument for Theism. Two theists of different creeds, provided they agree in attributing a certain character to God, may discuss the question whether the "miracles" of one of the different creeds are or are not worthy of the character of God. It would be idle to ask any man to accept Theism in any form on the evidence of any kind of "miracle," since, unless he already admits what you are seeking to prove, the most irrefragable evidence that the facts to which you appeal are genuine facts would establish no more than the hardly disputable conclusion that strange things do sometimes happen. It is thus not surprising if, as Francis Bacon remarks, miracles have been wrought to convince idolaters, but none to convince atheists.

The point I want to make, then, is this. The problem of Hume’s essay, as Hume himself states it, is vitiated by illegitimate simplification. When we are dealing with testimony to a startling event which claims also to have the value of a "sign" from the unseen, we have two questions, not one only, on our hands. There is, of course, the preliminary question, which arises whenever we have to decide for or against accepting testimony, the question of the quality of the testimony, the intelligence and bona fides of the witnesses, the nature of the agreements and disagreements between their reports, the presumption that these reports are independent, and the like. When we have satisfied ourselves, if we succeed in doing so, that our evidence is unexceptional in all these respects, there still remains two real questions, not to be disposed of by these antecedent considerations: (1) Is it more likely that the most unexceptional witnesses should fail us in this case or that the event to which they testify is a fact, and (2) if it is a fact, is it |363| merely a puzzling fact, or has it the value of a "sign"? Our answer to both questions, I hold, is legitimately influenced by our metaphysic. If our metaphysic is definitely theistic, it will be rational to regard such "signs" as likely to mark the course of history; if it is anti-theistic or neutral, the same expectations will not be, for us, rational. Whether it is rational to be prepared to acknowledge "miracle" as a feature of the historical process is thus a question which depends on a prior question: Is it irrational to be theists in our metaphysics?

As I have said, it would involve an obvious circle in our reasoning if we alleged the occurrence of miraculous events as the ground for adopting a theistic metaphysic. If a theistic interpretation of the course of events is to be justified, the justification must be based on the cursus ordinarius of nature. Our metaphysic, if it is to be more than an idle play of fancy, must be a response of thought to the full concrete reality of the world in which our life is set. And I think it follows that we cannot expect to arrive at a metaphysic of any great worth so long as we confine our contemplation to the domain of formal logic, or epistemology, or even of experimental science. We and our fellows are ourselves a part of the world to be interpreted, and we have no right to assume at the outset that we may not even be its most significant part. The material for interpretation is supplied not only by the natural sciences of the laboratory, the observatory and the field, but by the whole history of man with his ideals, his achievements, his failures, his self-condemnations, his hopes and his fears. For the tissue of life is inextricably woven of all these strands, though we are tempted, in an age of un- avoidable specialism, to forget the fact when we retire to our studies or our laboratories and concentrate our |364| attention on the artificial and poverty-stricken extract from the wealth of the real world which we call our special "subject." Like the Ephesians of the time of Heraclitus, we retire each into a poor little private world of our own, and forget the "common," and this, as that great man said, is to live like men half-asleep. Indeed, it is so to be logicians, or chemists, or historians or ethnologists that we forget to be men. The fault is not wholly our own, and it cannot well be escaped by our generation, though we may be permitted to imagine with innocent envy the possible happier lot of our successors, if some great social and economic change should simplify this intellectual problem by leading to the destruction of the masses of accumulated misapplied "erudition" which are our nightmare. But it is our fault if we make no attempt to escape complete subdual to that in which — for our sins — we have to work. The final verdict on the question whether Theism is not a legitimate, or, it may be, a necessary feature of a metaphysic which can "give account" of the real world in all its real fullness could be expected neither from an age of specialists like our own, nor from an age of gentlemanly loiterers, like the literary coteries of the eighteenth century. For a voice which might speak with compelling authority we have to look to a society which has an intensely rich and full life of its own and yet is not mastered by it but masters it, "sees it steadily and sees it whole." For my own part, I think I know what the verdict of such a society would be. I cannot, of course, expect that all my hearers should be of my mind. What kind of response one makes to life will, no doubt, for better or worse, depend on the sort of man one is for good or bad. One will respond with a different metaphysic according as one thinks that "we are such stuff, As dreams are made on," or μέγας ὁ ἀγών, οὐχ ὅσος |365| δοκεῖ. But we can all make it our purpose that our philosophy, if we have one, shall be no mere affair of surface opinions, but the genuine expression of a whole personality. Because I can never feel that Hume’s own philosophy was that, I have to own to a haunting uncertainty whether Hume was really a great philosopher, or only a "very clever man."


1^ Summa contra Gentiles iii. 101.

2^ Monarchia, ii. 4.

Top ↑

X. Knowing and Believing

This paper was originally presented before the Aristotelian Society in 1928, and subsequently reprinted in its Proceedings.

We are all accustomed in everyday life to draw a distinction between some things which we know and other things which we only believe, but do not know. But what is the precise difference we mean, or ought to mean, to mark by this distinction? Is there a real difference in kind between the act, or attitude, of knowing and that of believing? If there is, is this difference one which the psychologist could detect by an examination of the two acts as such, independently of any consideration of the intrinsic characters of their objects, the scitum, that which is known, and the creditum, that which is believed, or is it only by explicit references to the natures of scibile and credibile that the correlated acts of knowing and believing can be discriminated? The question is not, of course, whether that which is known to one mind may not be merely believed by another; it is manifest that there is much which he or you may know, but I can only believe without knowing. It is whether there are certain things which, from the nature of the case, are capable of being known, whether a given mind knows them or only believes them, and others which, again from the nature of the case, can be believed but cannot be known by any mind — or at any rate any human mind. Plato, as I presume we all know, teaches emphatically that to know (ἐπίσταςθαι) and to believe (δοξάζειν) are radically distinct |366| intellectual attitudes, each with its own class of appropriate objects, so that there are two distinct domains, in either of which truth is attainable; the domain of that which is, in its own nature, adapted to be known, the eternal, and the domain of that which is, not from any incidental disqualification on our part, but inherently, incapable of being known, and can only be believed, the temporal. Only the immutable and eternal can be known in the proper sense of the word; of the temporal we have, at best, only "true belief or opinion," though it is not denied that belief or opinion can be true. This is why, in the symbolic language of the Timaeus, "true discourse" about matters of sense-perception is called "opinion" and "belief" (πίστις, δόξα), and assigned to the "circle of the Other," but true discourse about the non-sensible is named "understanding" and "knowledge" (νοῦς, ἐπιστήμη), and given to the "circle of the Same" (Timaeus 37b-c). Aristotle also — though he never, I believe, expressly denies that the temporal can be known — appears in principle to agree with this Platonic view, since he does deny that "sense" can yield knowledge, and demands that the name shall be restricted to conclusions which can be demonstrated with formal logical necessity from premises which are themselves materially necessary, and by insisting that such premises must be strictly reciprocal as well as universal, in effect, makes knowledge and demonstration impossible outside mathematics.

“How does knowing differ from opining and believing? By being vision.”

On this view, then, objects which properly belong to the domain of the scibile, may accidentally be objects of mere belief to a particular mind, like the propositions of Euclid to the mind of the distinguished Oxford man of the story who "could not say that he had precisely proved them, but flattered himself that he had made them appear highly probable."

|368| But objects which properly belong to the domain of belief can never become matter of knowledge for any mind. This consideration seems to me to make a difference between the position of Plato or Aristotle and that of the mediaeval philosophers and theologians who depend so closely on Aristotle for their general logical and metaphysical doctrine. At first sight, indeed, the great schoolmen, with their insistence on the indispensability of a faith which is not knowledge, might seem to be retaining the principle of the Platonic distinction and merely reversing Plato’s estimate of the relative worth of belief and knowledge. But the actual facts are hardly as simple as this. The theory was, indeed, that to us in our present condition on earth the most momentous of all verities, the specifically Christian affirmations about God, are matter of a belief which is not knowledge, but, in the phrase of St. Thomas, "in a sense midway between knowledge and opinion." But this was regarded as incidental to our present earthly state; in our "fatherland" in glory, it was taught, we shall know what are at present to us the "mysteries of faith" with the same direct apprehension of their truth that we now have of the truth of the principle of Contradiction; thus it was not thought to be an inherent consequence of the character of these credenda that they are not now objects of knowledge. Though it is admittedly true that no one can simultaneously know and believe the same truth, yet, it was said, there is no more difficulty in understanding how the triune nature of Deity can be at once known to the beatified and believed by us on earth, than in understanding how a geometrical theorem can be at once known by the mathematician and believed by the layman. On the other side also, it was held to be a consequence of divine omniscience that God knows, in the |369| full sense of the word, all those "contingent" truths about matters of fact which fall outside the limits of demonstration and remain matter of belief for the created mind. "Future contingents," in particular, being contingent, cannot be demonstrated, and being future cannot be known through experience. Hence no created intelligence, not even that of a seraph, can know what I shall be doing five minutes hence, but this has been eternally known by God.

Thus, the position, I suppose, comes to this: whatever is in its own nature an object of knowledge may be also, per accidens, to some mind the object of a belief which is not knowledge. There are some truths which can never be matters of more than belief to any created mind, if only for the reason that no such mind knows all "contingents." Since, however, they are all said to be known to God, their admitted unknowability to us does not quite constitute them into a domain distinct from that of the inherently knowable, any more than "truths of fact" are made unknowable for Leibniz by his theory that the knowledge of them would demand an infinite analysis which no mind but that of God can complete. It will only be, to use a scholastic distinction, quoad creaturas, not simpliciter, that there is a realm of credibilia which are not scibilia, though it might possibly be said that this account requires to be modified in the light of the principle that the identity of knowledge in God and in the creatures is only one of analogy. However that may be, it is interesting to observe that it is precisely the great Greek thinkers — whom it is fashionable in some quarters to depreciate for their "intellectualism" — who most explicitly refuse to make the domain of the knowable coextensive with that of the actually true.

The modern world has not taken kindly to this |370| Platonic doctrine of the two radically distinct domains; its tendency is rather to assume that whatever is true is inherently knowable though circumstances may prevent us for a time, or even permanently, from knowing it. But if we reject the distinction of domains, the question is obviously at least suggested whether we can still retain a radical distinction between the acts or attitudes. If whatever can be an object of either may also, with suitable conditions, be an object of the other, we may fairly ask why we should be at the trouble to assume two mental attitudes or activities, when one would possibly serve our turn equally well. We have a reasonable motive for considering whether one of the two terms may not be the genus of which the other is a further specification. May it not prove either that knowing is a specific way of believing, or believing a peculiar way of knowing?

Clearly it would not be promising to treat knowing as the generic term and believing as a subvariety of knowing, for the obvious reason that many, if not most, of men’s actual beliefs appear to be false, and it seems preposterous to admit that anything which is false can be known. But it is tempting to regard believing as a genus, and to define knowing as believing a proposition which is, in fact, true. From this point of view there will be no difference between the mental attitudes or acts of believing and of knowing. To know will be to believe a proposition which happens to have the peculiarity of being true; the act, or activity, of believing such a proposition, considered as an act or activity of the believing mind, will be in no respect different from that of believing what is false, since the difference between the true and the false will be a difference of some kind in what is apprehended, not the manner of its apprehension. It should follow that a psychologist who keeps |371| strictly to his subject has not to concern himself with knowledge. He is within his province in treating of belief as a distinctive mental attitude and discussing the various ways in which it is generated or destroyed; but he strays over the borders of that province if he ventures to discriminate between beliefs which are true, and others which are false, since true and false are "extrinsic denominations," just as he would be violating the same frontiers if he undertook to deal with the distinction between virtuous habits and vicious. The sharp discrimination thus effected between logic or epistemology and psychology is a real attraction to many minds. It is true that we get into difficulties when we go on to ask whether such a view really allows us to believe in the possibility of "epistemology" as a tertium quid which is neither quite the same thing as logic nor quite the same thing as psychology. But we may ignore this consideration for the moment, as the possibility of "epistemology" seems problematic on any theory of the relation of knowing to believing. What I wish to ask is whether this undeniably attractive view will really work within its own limits.

Obviously, if we are to identify knowing with believing that which is in fact true, we shall need to be careful to distinguish believing itself from some attitudes of mind with which it is often confused in careless speech. None of us would say that a man knows a proposition which he merely holds to be either more likely to be true than to be false, or even much more likely to be true than to be false, though we sometimes talk of "believing" in these cases. Thus I myself think it more likely that Plato wrote the Epinomis, and overwhelmingly more likely that he wrote the seventh Epistle, than that he did not, but I hope I should never say that I know either of these things. If I were betrayed in the |372| heat of argument into making such assertions, I should expect to be properly checked for the careless temerity of my language. Again, I think there are good grounds for holding that Plato did not write the First Alcibiades, and should in ordinary conversation, say without hesitation that I "believe" it to be unauthentic. But suppose it were discovered tomorrow that the real name of the author had been preserved in some newly discovered scrap of papyrus, then I should all along have been holding an opinion which was true, but on any theory about knowing, it would be an abuse of language to say that I had "known" all along that the dialogue is spurious. If we are to do reasonable justice to a doctrine which identifies knowing with believing what is true, we must at least add that it is not every degree of assent to a proposition that amounts to belief. We shall have to say of belief in general what St. Thomas says of belief in the articles of religious faith, that it is not mere cogitatio cum assensu, but cogitatio cum fixo assensu, thinking of the proposition accepted with firm assent, as distinguished from the imperfect and wavering assent which amounts to simple opinion. And this consideration at once leads to an important question. St. Thomas, whose purpose in making the distinction is to determine the precise character of the assent to the Christian creeds required for salvation, does not tell us how firm assent must be, to remove it from the level of mere opinion to that of belief, but other utterances make it clear that he does not regard a man who merely thinks the statements of the creeds more likely to be true than not, or even considerably more likely to be true than not, but still subject to a real possibility of falsity, a "believer." The inferiority he admits in belief, as contrasted with the vision anticipated in a future life, lies not in any uncertainty of conviction but in the |373| absence of clear and distinct apprehension. The difference between the states of probation and of glory, as he conceives it, is not that the same verities are held in the one problematically, in the other with assurance, but that in the first they are perplexing and mysterious, in the second they will be obvious and self-evident; in both they are taken to be certain. I take it, however, that in a more mundane connection he might have been willing to regard convictions which, though firm, fall short of such absolute assurance, as fairly entitled to be called beliefs, and whether St. Thomas would have admitted this or not, I feel sure that it ought to be admitted.

What, for example, is the status of my conviction that William Shakespeare, the player from Stratford, was the author of Hamlet and Othello? It is not quite that of the unshakeable assent to articles of faith, not to be disturbed by any accumulation of appearances to the contrary, demanded by St. Thomas of the Christian believer. I can conceive evidence which would be destructive of the conviction. And yet it would misrepresent the facts to say that I only entertain the proposition "William Shakespeare wrote Hamlet" in the same fashion in which I entertain the proposition ’William Shakespeare did not write any part of the Two Noble Kinsmen.’" That is a personal opinion, of which I am ready to admit that it may quite well be mistaken; that I myself wrote the sentence I have just read is, to me, matter of knowledge. That William Shakespeare wrote Hamlet, I should say, is neither merely the one nor quite the other, and I can only distinguish it from both by saying that whereas I think, or opine, that Shakespeare did not write any part of the Two Kinsmen, and know that I am writing this address, I neither opine nor know, but believe that Shakespeare |374| wrote Hamlet. There is thus, as it seems to me, a state of mind which is neither wavering and uncertain opinion, nor yet knowledge (even on the view we are now examining that absolutely assured conviction is knowledge, when that of which I feel the assurance is true), and for this state of mind I can find no other name in our language but belief. On the theory under consideration the only intrinsic character which discriminates opinion, belief, and knowledge must be the degree of assurance and confidence with which assent is given to a proposition. But if it is admitted that any distinction can be made in this respect between a belief which does not amount to knowledge and a mere opinion, there is the difficulty that it is quite impossible to say what degree of assurance is required to convert opinion into belief. We clearly must not say, unless we mean to deny manifest facts, that any opinion becomes a belief when it is held with any degree of assent whatever going beyond the mere admission that it may possibly be true. It would be monstrous to say that I believe in the guilt of the accused in a sensational trial for murder the moment I concede that the very imperfect and fragmentary "evidence" obtainable tells rather for conviction than for acquittal. On the other hand, it is equally impossible to deny the name of belief to all convictions which are held subject to the possibility of doubt and correction, for this would be to reduce the practical convictions by which we govern our daily lives to the status of mere opinions. But the conviction shown, for example, by a life assurance company, when it accepts me as a "first class life," clearly has a status very different from the opinion of one of the directors about the authorship of Junius or the authenticity of the Zinoviev letter which played a part in the General Election of 1924, or the probable |375| issue of the coming General Election of 1929. And there is a further special difficulty about such a position if it is combined with the definition of knowledge as a belief which is, in fact, true. For from the combination of the premises that all opinions, and only they, which are held without any qualification of the recognition that they may be mistaken are beliefs, and that a belief which is true is knowledge, it would follow that I know any proposition whatever which I happen to hold very confidently, whenever that opinion happens to be a true one. If, for example, I very stubbornly hold that Philip Francis wrote Junius, for no better reason than that Macaulay said so, then, if tomorrow documentary proof of that fact should be discovered, I should have known today who was the author of Junius; and this seems quite unreasonable. However hard it may be to say precisely what the distinction to be made is, a sound theory will need to distinguish not two terms only, belief and knowledge, but three, opinion, belief, knowledge, where the theory in question only enables us to discriminate two.

The impossibility of identifying knowledge with confident belief of what is true in particular, is well illustrated by the example Plato has selected in the Theaetetus, the effect of advocacy on a jury. Skillful advocacy will frequently lead a jury to pronounce with complete confidence on a question of fact where the evidence is patently incomplete. Indeed, we may imagine a case — and I do not think it would be hard to produce actual instances of such cases — in which such evidence as there is all points in one direction, and yet the jury are induced to return an unhesitating verdict in the opposite sense by clever and eloquent but wholly irrelevant appeals to sentiment and prejudice. And in a case of this kind it may well happen that such a |376| finding is in accord with fact; the available evidence may have pointed unmistakably in one direction and yet have been misleading. I might myself, for example, be brought to trial for an alleged offence in some country where Englishmen happen to be very unpopular, and there might be evidence tending to establish my innocence, but none of my guilt which could stand examination. I might then be convicted on the strength of a moving appeal by the prosecution to the local dislike of the English, and it might be the case that I had actually committed the offence laid to my charge, though the evidence of my guilt was worthless. Or I might be an innocent man with a formidable mass of incriminating evidence against me which could not be shaken in cross-examination and no evidence in my favour, and might owe my acquittal to a pathetic but irrelevant insistence by my advocate on the distress of my invalid wife and helpless babes. In such a case it would be an abuse of language to say that the jury who found the correct verdict knew me to be guilty or innocent, though they were undeniably led to believe with complete conviction something which, in fact, was the truth. And, as against the view that the differentia of belief may be found in the peculiarity of being attended by some doubt or uncertainty, some admission that the thing believed may, after all, be false, I would add that if the appeal to sentiment or prejudice were sufficiently adroit and eloquent, every member of the jury might leave the box without a shadow of hesitation in his mind, and yet it would be monstrous to call this unqualified conviction knowledge.

It seems to me, in fact, that what those who make it a differentia of belief to be entertained with a conscious residual uncertainty really have in mind is opinion rather than belief proper, and even so, they |377| are not, I think, distinguishing opinion and belief quite accurately. Thus, to take an example, A. B. is accused of the murder of C. D. and acquitted because the evidence for the Crown proves to be wholly in- adequate. I may know that the evidence is inadequate, and yet may hold either that it is overwhelmingly probable that A. B. did not wrongfully cause the death of C. D., or that there is no more than an "off-chance" that he did not. In the first case I should say I believe in A. B.’s entire innocence; in the second that I do not. I believe the Crown has not made out its case, and I may, in addition, opine that A. B. was wholly blameless, or I may not. The difference is one which any one of us would recognise in practice, and it seems to me an important difference The difference between opining and believing is thus a psychological one. It need not necessarily have a logical ground. I may, for instance, know that all available evidence supports one conclusion, and yet be absolutely firm in my belief that, in spite of the evidence, a different conclusion is the true one. Whether this is an admissible frame of mind or not, it is one with which we are all familiar. We can believe, and believe intensely, in the direct face of evidence, and it is generally held that there are cases where it is a duty to do so, e.g. that Othello ought to have believed in the loyalty of Desdemona against any evidence that could have been amassed. But we cannot believe in this fashion against demonstration or against self-evidence. The difference between opinion and belief thus seems to me to be properly psychological. A conviction, false or true, held with a certain degree of confidence which we cannot exactly define, is a belief. But the difference between belief and knowledge is of another kind. When we know, we are indeed absolutely certain; but we may be, |378| and often are, absolutely certain where we do not know, but merely believe. The difference between believing and knowing is thus not primarily psychological, not to be detected by any mere examination of the act or process of knowing or believing; and, again, if what we have said so far is sound, it does not lie in the mere fact that what is known is true, since we often believe what is true without knowing it. Yet it seems clear that there is a difference. What then is it?

It is natural enough to attempt to answer the question in the way in which it is at once answered by one of the interlocutors in the Theaetetus, by saying that what makes the difference between believing and knowing is that when one knows a proposition one can "give an account of it," or, more precisely, one can produce demonstration of it. Where one cannot do this one merely believes, but does not know. One half of this statement, that knowledge and demonstration are coextensive, if knowledge is taken in the fullest and strictest sense, is, of course, exactly the teaching of Aristotle, and would, I suppose, be generally admitted so far as this, that we cannot be said to know any truth which can be demonstrated, unless we are acquainted with, and could, if called upon, produce the demonstration. If, for example, I cannot demonstrate that the series of prime numbers has no last term, I can hardly be said, without justification, to know the proposition. I may have known it once, at school or college, but at present, until I have recovered the lost demonstration, I can hardly be said, except by courtesy, to know the proposition. The case is still clearer if the proposition asserted is one of which I admittedly have never had any demonstration. Thus suppose I were now to assert the proposition "there is a consistent geometry in which the total number of points is less than twenty"; if the |379| only justification I could produce for the statement turned out to be that I had once heard my distinguished colleague, the Professor of Mathematics in the University of Edinburgh, assert it, I might fairly be told that I do not know the proposition I am affirming; I only believe it, though, in view of my colleague’s eminence, I do well to believe it. What I really know is not that there is such a geometry, but that I once heard my colleague say that there is such a geometry. If Russell spoke truly when he said that in the time of Kant no proposition of geometry had ever been correctly demonstrated, we may fairly say that no geometrical truth whatsoever was known by Kant and his contemporaries, though this leaves it possible that they believed a great many such truths. It is another question whether the negative half of our statement (in which it diverges altogether from the doctrine of Aristotle), "where we cannot demonstrate, we merely believe" can be sustained. As both Plato and Aristotle saw, the identification of knowledge proper with awareness of demonstrations at once raises the question about the status of the "first principles" of demonstration, the ultimate major premises of knowledge. Do we know these premises or do we not? Obviously if we know them, it cannot be by demonstration. If we said that no one knows the conclusion of a demonstration unless he has first demonstrated its premises and the premises which have been used in demonstrating them, we should be making all knowledge depend on an infinite analysis, and thus, in effect, denying that a "creature" can know anything. If we say that the ultimate premises are not known but only believed, we should have to admit that this defect in our cognition of the premises affects the whole superstructure which is built on them, and so, once more, there would be no genuine knowledge.

|380| We do not really escape this difficulty by pointing out, what is true enough, that in mathematics, the type of strictly demonstrative knowledge, it is not the ultimate postulates, but only the implication of further consequences by the postulates, that is formally asserted. The difference, we are sometimes told, between the modern philosophical mathematician and his less cautious predecessors is that the predecessors asserted both a conclusion, such as the Pythagorean theorem, and the various more ultimate postulates from which the demonstration of the theorem is drawn; their more philosophical successor is content to assert only that the theorem is a consequence of the postulates without committing himself to any further assumption about the truth of either postulates or theorem. This is true, so far as it goes. As a mere device for securing a necessary subdivision of intellectual labour it is an excellent arrangement that the mathematician should be set free to develop the implications of a set of premises without being called on to concern himself with asking from what quarter his premises have come, or how they can be justified. In acting so, the modern philosophical mathematician is, after all, only carrying further a procedure begun by Euclid himself when he substituted the colourless names of "common notions" and "initial demands" (κοιναὶ ἔννοιαι and αἰτήματα) for the older designation of unproved premises as "axioms," obviously with the intention of excluding controversy about the source and justification of these initial premises. But no device of this kind can obscure the central fact that there are undemonstrated principles which the most cautious of mathematicians cannot avoid asserting, at least tacitly. He must, at least tacitly, assert all the principles, whatever they may be, of a logic of implication itself. However true it may be that one asserts |381| neither the Pythagorean theorem nor the postulates which virtually define an "Euclidean space," but only the consequence of the first on the second, still, in asserting as much as this, one is assuming as known the structural principles of the logic of implication. If these are not known, I cannot know that one specific conclusion rather than another "follows from" a given set of premises. If they are merely believed, then I cannot know, but can at best believe, that any given consequence "follows from" the premises from which I assert it to "follow." Indeed, the profession even of a complete scepticism about the possibility of attaining knowledge, if it claims to be a reasonable scepticism and not a personal "fad," involves the claim to know that "nothing can be inferred with certainty," and he who says as much as this is plainly claiming by implication to know what conditions would be required to make inferences valid.

The point is luminously illustrated by the notorious crux in which Mill involved himself in his Logic. Nominally Mill’s account of inference is intended as an analysis and justification of the procedure of the experimental sciences. But in view of his belief that "induction" is a method, and in the end the sole method, of demonstration, his book becomes a theory of demonstration, with the peculiar characteristic that it is committed to justifying demonstration from the point of view of pure experience, that is, without presupposing any ultimate principles of demonstration. It is no incidental and removable oversight, but a result dictated by the nature of Mill’s undertaking that, in the end, all knowledge is made to repose on "induction by simple enumeration," a procedure for which Mill has really no justification to offer except the old one of Hume that it is quasi-instinctive and has some remarkable and |382| quite unaccountable successes to its credit. (And even this allegation is only made on the strength of the very curious assumption that the results of such men as Galileo and Newton have, in fact, been reached by reliance on "enumeration.")

I hope I need not say that I have not the slightest quarrel with the doctrine that the prima axiomata are disclosed to us by "induction," if only the nature of the induction is rightly understood. The all-important point is not that this induction is not proof of any kind. We do not "prove" such an "axiom" by an enumeration of particular instances; we see it, or as Aristotle says, "recognise" it in the particular instance, and this recognition is direct and immediate. (This is the explanation of the fact, found so mysterious by Mill, that one "observation" is sometimes sufficient, whereas, in other cases, thousands may be inconclusive. The collection of the thousand is not the "induction," it is preliminary to it.) It is the clear perception of this point which gives Aristotle’s brief statement about "induction" at the end of the Posterior Analytics its immense superiority over the accounts of our modern "inductive" logicians, who have been seduced by Bacon into mistaking "induction" for a method of demonstration. In intention, I suppose, I do not differ much from the position which writers like Bosanquet really mean to adopt, but I feel that they have not quite shaken themselves free from the notion that "induction" is proof, and are thus involved in the insoluble difficulty about "circular" proof.

It seems to me, then, that it is quite impossible, unless we take refuge in a personal scepticism which does not even profess to be able to offer any rational justification for itself, to get away from the familiar Aristotelian doctrines that we possess knowledge, and knowledge which is direct and immediate, not mediated through |383| inference, and that knowledge is something really different from belief though the difference is one which cannot be detected by the psychologist. Neither knowledge nor belief can be treated as a special subvariety of the other with a differentia of the kind an observational or experimental psychologist can identify. (I may remark in passing that I know, of course, that so far as terminology goes, Aristotle restricts the name knowledge (ἐπιστήμη) to awareness of the conclusion of demonstration as guaranteed by its premises, and speaks of the immediate apprehension of premises and principles which cannot be demonstrated by another name, νοῦς, or as the schoolmen said intellectus. But the point which is important for my purposes is that νοῦς or intellectus, like knowledge, is sharply discriminated from opinion and belief (δόξα and ὑπόληψις), and held to have a still higher "epistemological" status than demonstration, knowledge of proved conclusions as guaranteed by their premises.) It is this recognition of an immediate knowledge as genuine knowledge, and indeed as the type of complete and perfect knowledge, which I am concerned to defend.

For a full defence, no doubt, it would be necessary to examine the position of a logician like Bosanquet who believes himself able to admit the reality of knowledge which is something more than a belief in that which happens to be true, and yet to deny that there are "unproved premises of proof." I cannot undertake such an examination because, no doubt by my own fault, I have never been able to understand what Bosanquet’s position really comes to. I suspect that he may partly have been influenced by the consideration that the formal principles of deduction (for example, the principle of the syllogism in Barbara) do not themselves appear as the actual premises of deductions of |384| which they are the principles. At most, however, this would only go to prove that the unproved and unprovable implications of a deduction should not in strictness be called its premises, not that there are no such principles. And it has also to be remembered that the general principles of the logic of implication are not the only unproved truths presupposed by the demonstrations of the sciences. Each of them has also, as Aristotle insisted, its own special postulates (οἰκεῖαι ἀρχαί), and these may and do figure as actual "premises" for example, the "fifth postulate" will occur as an actual premise, though the principle that "if equals be added to unequals, the wholes are unequal" will perhaps not. As far as I understand Bosanquet’s view on the general question, he seems to hold — and this is why I suspect that I must be misunderstanding him — that all knowledge involves a circle. Our conclusions are guaranteed by the consideration that they follow from certain ultimate presuppositions, and these presuppositions, in turn, are guaranteed by the fact that they lead to these conclusions. If this is really what is meant, it should seem that at any rate the principles of a logic of implication are themselves exempt from the asserted dependence on conclusions; if they are not, how can I know that an alleged conclusion does or does not "follow" from the premises from which it is alleged to follow?

If, as is perhaps more likely, all that is really meant is that what we commonly call our "existing scientific knowledge" consists, outside the domain of pure mathematics, of propositions which have never been and never will be demonstrated, and do de facto acquire a higher probability from their "consilience" with fresh records of the independently observed, this seems irrelevant as an objection to the Aristotelian account of the |385| "undemonstrated principles of demonstration." It is, of course, to be remembered that the recognition of a real difference between knowing and believing does not carry with it the consequence that I am never mistaken when I think I know. That I sometimes suppose myself to know when I do not really know is a fact having the same sort of signification as the equally familiar facts that I sometimes suppose a demonstration which really involves a fallacy to be cogent, that I sometimes suffer from hallucinations of the senses, and that my memory is sometimes at fault. What the fact really shows is merely that there is no psychological criterion by which we can infallibly discriminate knowledge from belief, any more than there is such a criterion for the discrimination of sense or memory from imagination.

What I am concerned to suggest, then, is this. If it is really impossible to get away from the positions that knowing and believing are different in kind and that some knowing is immediate, there must be something wrong with certain conceptions of complete or perfect knowledge which have been popular in our own lifetime. Neither inference nor judgment can be the type of the most profound and thorough knowledge. For the inferentially known is always mediately known, known as Aristotle says, "on the basis of a previously existing knowledge," and it is in the right of this knowledge previously existing that the inferentially known can claim to be known. Even judgment, though it is an abuse of language to identify it with inference, or inference with it, not only tends, as Bosanquet used to say, to "expand into inference," but tends always to provoke the demand that it shall be justified by exhibiting it as the conclusion of a process of inference, though in the case of the "unproved principles," the demand can only be met by refusing to meet it. It is never |386| transparently absurd to ask about any judgment offered for our acceptance, "What are the grounds for it?" even though the answer may have to be that no other judgment or judgments can be produced as grounds for that I have made. This is notoriously not the case with perception. When I have justified an assertion about present matter of fact by saying that, for example, I actually see the mercury rising in the thermometer-tube, it would be strictly meaningless to ask why I can see this. The characteristic peculiarity of the Aristotelian account of knowledge which, I suggest, is necessitated by the impossibility of making the "unproved principles" mere matters of belief, is that it frankly recognises the same directness which marks our apprehension of sensible fact, at one end of the scale, in the apprehension of ultimate principles at the other end; both are strictly vision. As in the case of vision with the bodily eye, there are a bewildering multitude of links intermediate in the causal order between the occurrence of the event which is the stimulus and the event which is the sensation, but nothing whatever intermediate in the order of apprehension between percipient and perceptum, so also in the case of what we see with the mind’s eye when we apprehend an undemonstrable principle, there is nothing intermediate in the order of apprehension between the knowing mind and the principle it knows. And it is this kind of direct and immediate apprehension of truth which we should regard as the type of true knowing. All that we commonly call our scientific knowledge is an endeavour, never fully successful, to recapture for our mental vision of facts this immediacy and obviousness from which we begin by passing away, the moment judgment supervenes on sense-perception.

Judgment has sometimes been spoken of as the |387| characteristic form of the apprehension of truth. It seems to me that the very fact that most of what we know about the world has to be couched in the form of judgment is the characteristic mark of the inevitable imperfection of our apprehension. This was well understood by the philosophers of the Middle Ages, and it is because they understood it so well that there is a curious point, not often remarked upon, in which their language about God differs from that of philosophers of a later time. When a modern philosopher allows himself to talk of God, he makes no scruple to speak of God as "thinking," or even "cogitating." Locke, for example, in his attempted demonstration of the being of God proceeds at once, after establishing the point that "something has existed from eternity" to argue that this something must necessarily be a "cogitative" being. So in Green’s Prolegomena to Ethics the cardinal point for which it is contended against the naturalist is that both the existence of nature and our knowledge of it are dependent upon the reality of a "self-distinguishing principle" which thinks. St. Thomas, on the other hand, expressly denies that cogitatio is found in God. God knows all truths, but God does not "think," just as God wills the whole creation but does not "deliberate," or "make up His mind." It is just because judgment is the characteristic form of thinking that, to the mind of St. Thomas, it is an unjustifiable anthropomorphism to say that God thinks. God does not think, He knows with a knowledge which is vision.

The thought is that in a knowledge which is always completely "in act" there is a complete and intimate interpenetration and possession by the knower of the object known which is absent from judging. Judging is not "being in possession of" or "being possessed by" the object; it is a preliminary step on the way to such |388| possession, inevitable in us who begin without knowledge, and have to obtain it in successive fragments and parcels; we make our way in the direction of knowledge by thinking "about it and about." But if we could completely achieve the ideal of knowledge which guides us all through the process, the perfect adaequatio intellectus cum re, as we cannot, our knowledge would no longer be thinking, or wear the form of judging; it would have recovered the directness characteristic of the perception from which it began by setting out. It is this conception of complete knowledge as a direct vision, to which the form of judgment is inadequate, which, as it seems to me, is already implied in the admission of a real distinction between knowledge and true belief.

I hope I shall not be understood as suggesting either that to know anything completely I must be that which I know, or that in so far as we know objects it is not properly we who know, but the objects which "know themselves through us." What I mean is not this. I mean rather that the type of perfect knowledge is an apprehension which is at once direct and, as I may say, "self-luminous" or "transparent." The absence of either directness or transparency — what I might call the colour of "of-courseness" — is an indication that our apprehension is not all that complete knowledge should be. In this respect, sense-perception, though it gives us our most familiar illustration of one of the characteristics of complete knowledge, the directness, falls short in respect of the other, because it is apprehension of "brute" fact, about which there is no "of course," but which, for all we can see, might perfectly well be quite otherwise. At the other end of the scale, in the "highest universals" which are indemonstrable, we approximate again to a knowledge which may fairly be called "vision," and is called insight. These principles are no |389| mere tautologies; they are all, in that sense, "synthetic," and yet there is no "because" behind them; they carry their own evidence with them, they are so "of course." The trouble is that this "of-courseness" cannot be communicated to the whole system of what we know. It is communicated to the conclusions which follow by demonstration from the principles, but these conclusions remain incurably abstract; there is always a gulf between them and full concrete actual fact.

And yet it is undeniable that there is a kernel of something which is not merely believed but known in our apprehension of actual fact. Perception, I mean, is knowledge, and perceiving is knowing, and yet perception is not judgment, and nothing but confusion comes of the attempt to talk of it, as some modern logicians have done, as though it were judgment. There is a difference difficult to express clearly by any form of words, and yet unmistakably real, between seeing a green leaf and judging "that the leaf is green." I cannot attempt to communicate the content of a perception to anyone without converting perception into judgment, and hence the possibility of confusing knowledge with mere belief comes in at once, since there is always the possibility of error in the description, or analysis, of the perceived which is involved in the simplest of judgments. The ideal type of knowledge would only be realised if we could substitute for our crude piecemeal perceptions an apprehension of the actual in which we should see it articulated by the ultimate principles, as a system of geometry is articulated by its postulates, and see the articulation not progressively but in one complete act of insight, carried through to the whole of the actual detail, not stopping short, as our scientific insight always does, at the "proximate" universals. This would be nothing other than the |390| scientia visionis ascribed by the mediaeval thinkers in its completeness to God and only to God. We, of course, never reach its completeness just because of the curious position of the mind which is the standing crux at once of naturalism and of a really "absolute" idealism, as at once standing outside and above the complex "event which is nature" and also implicated in it τρόπον τινὰ δύσφραστον καὶ θαυμαστόν. In actual perception, in a very real sense, we may say that the whole of the actual is given to us at every moment, but it is given in a form which is opaque and non-communicable; in our most exact scientific knowledge, when we know, as Aristotle would have us know, our conclusions through, and as guaranteed by their ultimate οἰκεῖαι ἀρχαί, our insight is direct and there is transparency and the "of course" quality, but the whole transparent structure still stops short with "separated" universals, and, if we are to speak strictly, we must say that our application of the system to the actual "passage of nature" is, as Plato maintained, at best "true belief" rather than knowledge.

The point I am anxious to make is that just because "vision" is the ideal type of knowledge and vision in its completeness is impossible for us, who enter ourselves, in virtue of our temporality, into the "passage of nature," it is quite impossible for us to construct on a single principle a flawlessly coherent scheme of the omne scibile, after the fashion of philosophers like Hobbes, who take demonstration falsely as the ideal of knowledge and deny the name to all that cannot be demonstrated. It is inevitable that, for us, there should be a rift in the scheme of the scibile, the rift which separates the sciences of demonstration from our knowledge of the historical and individual. And there is one characteristic of this historical knowledge on which I think it may not be superfluous to offer a remark; I |391| mean its inarticulateness, or, as I might say, incommunicability, a quality it shares with actual perception. I suppose it is impossible for any man to immerse himself long and thoroughly in the study of the life and work of a great personality or a great age without feeling that he in the end comes by a direct insight into purpose and meaning which he can call by no name but knowledge, though it is quite impossible to demonstrate the correctness of his insight to anyone or even to communicate it. The point has been admirably put by my lamented friend, Prof. John Burnet, with special reference to his own life-long study of Plato and his circle, and I cannot do better than quote his words.

"A man who tries to spend his life in sympathy with the ancient philosophers will sometimes find a direct conviction forcing itself upon him, the grounds of which can only be represented very imperfectly by a number of references in a footnote. Unless the enumeration of passages is complete — and it can never be complete — and unless each passage tells exactly in the same way, which depends on its being read in the light of innumerable other passages not consciously present to memory, the so-called proofs will not produce the same effect in any two minds. That is the sense in which philological inquiry, like every other inquiry, requires an act of faith. In the long run the positive construction must be left to the individual student, and no two students will see quite alike."1

Much the same experience, as I cannot doubt, would be confessed by every historian of a man or an age. And yet, in spite of the impossibility of demonstration and the disagreements of individual students about this or the other point of detail, it seems undeniable that we do, in favourable cases, succeed in getting an |392| "understanding" of a great historical personage which is no mere affair of the more or less likely, and can be recovered by successive students for themselves, though none of them can directly impart it to another. We can, if we will, succeed in knowing, not merely opining or thinking, what the "historical" Plato, or Cromwell or Shelley was, no matter what uncertainty may hang over many of the alleged incidents of their lives. Where we understand, we may fairly say we know; in fact, in our relations with the members of our circle of personal intimates, it is precisely what we thus understand of their inmost purposes and characters that we may claim with the best right to know; it is details of events in their lives, not related to character and purpose, in connection with which it is most rash to profess more than opinion or belief.

The one great outstanding defect in the Platonic-Aristotelian account of knowledge, inspired as it is, naturally enough, by reflection on the character of the one body of knowledge which had already acquired a high degree of organisation by the opening of the fourth century B.C., the mathematical, is that, by insisting too exclusively on one side of the ideal of a completed knowledge, its thoroughgoing articulation, it really leaves no room for the recognition of historical insight into the individual as genuine knowledge, a one-sidedness which has left its traces on all subsequent philosophical methodology. Against Aristotle’s οὐδὲ δι ͗ αἰσθήσεως ἔστινbἐπίστασθαι, I should like to set as a true and important aperçu Locke’s recognition of the reality of "sensitive" knowledge itself, only with the already given caution that in the very attempt to communicate the content of "sensitive" knowledge we are passing away from perception itself to the "perceptive judgment," as Bosanquet, I think, calls it, and substituting |393| for immediate apprehension the mediated, with the possibility that it may be the incorrectly mediated. And the possibility becomes an overwhelming probability when we consider that, in the case of any "perceptive judgment," the implied premises are immensely numerous, and far the greater number of them, possibly the most important of them, are commonly held only implicitly. The premises which would justify our affirmation have to be sought partly in a vast mass of unanalysed and imperfectly analysed perceptions, partly in an intellectual scheme or pattern embodied in a linguistic structure which we have received as an inheritance, and with a very imperfect comprehension of the complexities of its articulation. It is only when our thought is moving in the realm of "abstractions" which constitutes the domain of the "exact sciences" that we can be confident that the first of these two sources of unrecognised "tacit premises of mediation" is precluded; even there we are at the mercy of the second, as is shown clearly enough by the actual history of the critical search for a complete and accurate list of the postulates of mathematical science.

We readily understand, then, why in the past the tendency among great philosophers should have been to make "exact" science into the one type of all genuine knowledge, and to minimise the importance, or even to deny the existence of "sensitive," or, as I should prefer to say, "historical" knowledge. The fact, as I take it, is that in the rationalisation of the given we are always moving between two extremes, one which we have left behind us when we take the initial step towards science, that of merely immediated "given-ness," not as yet even "recognised," the other, one which we never reach, that in which we should possess and be possessed by vision of the whole system of the true, apprehended in |394| all its articulation and apprehended all together. Here the antithesis between immediate and mediated would once more be done away with, and we should at last have an actual knowledge fully adequate to our ideal of knowledge which is through and through, in all its detail, knowledge and nothing but knowledge. But, in fact, we never achieve the completion of the process. Our human thought is always, as the phrase is, "discursive," advances from point to point, or, as the schoolmen said, proceeds by composition and division, that is judgments, affirmative and negative about, or upon, a material not itself apprehended by judgment. We leave perception for judgment the moment we begin to recognise, to reflect and to communicate the results of recognition and reflection; not being able to complete an "infinite analysis," we never return, at the other end of the process, to the kind of apprehension we are seeking, a vision which has the directness of perception without its inarticulateness. And, as I take it, the ultimate reason why this must be so is the simple one that time and movement are as radically characteristic of our thinking as they are of the processes of our physical life. Thinking and judging are what, so far as I can see, complete knowledge cannot be, temporal processes. Our attempt to say what knowledge is leads us back to a conviction which we may equally reach along very different lines, the conviction that time is at the bottom of our profoundest and most perplexing metaphysical problems.

There is a great passage in Augustine’s Confessions (xii. 18) which puts the point I am very inadequately labouring with admirable clarity. "Will you say," exclaims Augustine to a supposed opponent of his exegesis of the opening words of Genesis, "that what truth rounds so loudly in my inner ear concerning the true |395| eternity of the Creator is false? That it is false that His substance is invariable by succession of times, and that His will does not fall without His substance? Whence He does not now will this and again that, but wills once, all together and always, all that He wills; He wills not subsequently what before He willed not, un wills not what formerly He willed. Such will is mutable, and nothing mutable is eternal, but our God is an eternal God. Again, is that false which truth rounds me in the inner ear, that anticipation of things to come becomes vision when they have come, and vision, again, becomes memory when they have passed; that all activity of mind (intentio) which is thus variable is mutable and nothing immutable eternal, but our God is an eternal God? I combine these positions together and discover that my God, the eternal God, created the world by a will that was not novel and that His knowledge admits of no transitions."

What is here said with special reference to the knowledge of the Creator seems to be simply true of all knowledge which realises the ideal implicitly aimed at in every attempt to know; an apprehension which is not such vision, whatever its excellence, does not wholly amount to what we really mean by knowledge, and it is just for this reason that, without the slightest desire to depreciate science, or to suggest any doubt about the imperative duty of its prosecution, I feel bound to say that such "science" as is possible to beings as temporal as men — I mean, of course, in the conditions familiar to us from our experience as those of man in the life we see — is never quite identical with "knowledge." Even in our exactest of exact science (in the most philosophical of philosophical arithmetics, for example) there is always, I take it, the possibility that we have overlooked some premise really required by the system, |396| or formulated it without due circumspection, and so there is always the possibility that we are claiming to know where we should be content to say that we believe. What is more, we can, in such a case, make no absolutely clear and sharp cut between the propositions which we know and those which, it may be, we should only be said to believe, any more than I can make such an absolute distinction between what I actually perceive and what I remember. The nature of the distinction between knowing and believing, like that of the distinction between seeing and remembering, may be clear and intelligible enough, but it furnishes no infallible criterion for application to the specific instance, just as, in the realm of moral practice, the doctrine of the Categorical Imperative, accepted without reserve, would furnish no infallible criterion of the rightness of the individual act.

Let me, in conclusion, try to put together, in a set of theses, the results of these rather desultory reflections:

(1) Knowing and believing are two quite distinct attitudes of a mind to that which it apprehends. Knowing is not a special way of believing, nor believing a special way of knowing.

(2) The difference between the two attitudes is not one which can be discerned in an individual case by observational scrutiny, and is, in this sense, not psychological, and cannot be justified by an empirical science of psychology.

(3) Nor does it lie in the fact that what is known is true, since this may equally be the case with what is believed, but not known.

(4) The reality of "sensitive" or "historical" knowledge shows that the difference cannot be simply that the object of knowledge is non-temporal.

|397| (5) Yet the difference must lie somehow in the respective characters of that which is believed and that which is known.

(6) What the difference is seems to be indicated by the consideration that perception affords knowledge, though a knowledge which is inarticulate and incommunicable. Belief is a sub-species of judgment and judgment is mediate, in the sense that it presupposes reference to grounds for judging which fall outside the judgment itself. This is not the case with knowledge.

(7) The ideal of knowledge is that it should be at once immediate and articulated. In actual life, the immediacy is most manifest in perceptional knowledge, where articulation is at a minimum. The articulation is clearly manifest in scientific knowledge, as conceived by Aristotle, where conclusions are known along with and through premises which are, in the last resort, themselves immediately apprehended. But, even in the case where such scientific knowledge is best exemplified, that of pure mathematics, the whole of the premises relevant to the conclusion are not consciously envisaged in the act of "knowing the conclusion through its premises." Hence the ideal of knowledge is not actually attained even here, since the apprehension is not full and complete "vision."

(8) It follows that while Aristotle was justified in his demand for "intuitive apprehension of unproved premises of demonstration," his account of knowledge is defective at the other end by not explicitly recognising the reality of the inarticulate and incommunicable knowledge belonging to sense-perception as such.

(9) The impossibility of limiting knowledge to awareness of the universal, thus implicitly denying that historical fact can be known, arises directly from the fundamentally temporal and successive character of |398| human mental activities, not from the nature of knowledge itself.

I would add two corollaries to these theses:

(1) It is unfortunate that some eminent recent philosophers should have spoken habitually of logic as concerned with judgments rather than with propositions. This language inevitably tends to obliterate the very real distinction between knowing and believing by the misleading suggestion that knowing is a kind of "cogitation."

“How does knowing differ from opining and believing? … By being vision.”

(2) Since knowledge, so far as it really is knowledge, is immediate, there can be no "theory of knowledge" in the sense of a theory of the way in which knowing is done. The whole of what can properly be called "theory of knowledge" is contained in an answer to the question "How does knowing differ from opining and believing?" And the true answer to this question can be given in three words, "By being vision."


1^ Greek Philosophy, Part I. 2-3.

Top ↑

XI. Is Goodness a Quality?

This paper was originally presented before the Aristotelian Society in 1932, and subsequently reprinted in its Proceedings.

The observations I am about to make are likely to be both confused and irrelevant, since I am not sure that I know either what goodness and quality mean in this connection, or what the framers of our programme intended us to be discussing. But I will try to indicate what is in my own mind, so far as I know my own mind, in some prefatory remarks about the extent to which I suppose myself to be agreeing or disagreeing with the distinguished speakers who have preceded me.

I do not find the passage in Aristotle’s Ethics, from which the discussion presumably starts, as obscure as Mr. Moore seems to do. The point, I take it (and I presume I am here in accord with Mr. Joseph), is that the word good has a variety of senses, and yet is not merely equivocal. As in the case of the word surgical, the different senses are interconnected by analogy. I say (1) Hannibal was a good strategist, (2) Gladstone was a good Chancellor of the Exchequer, (3) a good epitaph must be brief, (4) Easter Monday is a good day for a public demonstration. In no two of these cases do I mean precisely the same thing, yet it is no mere accident of language that I use the same word in all four cases. The range of values of x, for which the statement x is good is significant, forms a series with an order of before and after between its terms. As Aristotle himself puts it, "good is predicated in all the categories," |400| and categories are an ordered series. (There is a metaphysical order of dependence between the primary category of all, substance, and the rest, and the relation of dependence is a different one in each case.) The point might, indeed, be made without going outside the bounds of a single category. A lead-pencil and the penknife with which I point it may both be good, but a good pencil must have a soft lead, and a good knife must have a hard blade. Good drinking-water must be odourless; good wine, I presume, must have an aroma. It would be possible that all the characters on the strength of which I call A good should be the same as those on the strength of which I pronounce B bad. This of itself seems to show that when I truthfully call different things good, it is not, at any rate, one and the same "simple indefinable quality" that I intend to denote in all cases, though, so far, it might still be true that what is denoted in each case is some simple indefinable quality, and further, that there is some sort of connection between all these simple indefinables. But I cannot myself believe even this statement to be true. I do not think that any such simple quality is denoted when I truly speak of a given poem or novel as a good poem or novel, and I think it a pity to equate simplicity, where it exists, with indefinability. I do not see why the simple need be indefinable, or why in Principia Ethica Mr. Moore should have identified definition with the analysis of a complex whole into its constituents. Given an indefinable A and a second term B, however unanalysable B may be, it seems to me that I can define B if I can point out some relation which B and only B has to A, and has not to any term but A, and this, I should have said, is the procedure regularly adopted in the most precise of our definitions, those of numbers and numerical functions.

|401| But I agree with Mr. Moore when he urges against Mr. Joseph that the goodness of a poet or a poem is not identical with the poet or poem, and again when he adds that the goodness of either is not identical with the "complex of characters" on the strength of which we call poet or poem good of its kind. I do not even understand the first thesis, the one Mr. Joseph appears to defend. If the goodness of the poem is the poem, it must also be true that the poem is its own goodness. What, then, about the badness of the same poem? — and the best poem can be bad in some ways, and sometimes surprisingly bad. Paradise Lost is an astonishingly good poem, but it is amazingly bad too, in respect of its total lack of humour; it would not be the poem it is if it were not perversely unhumorous as well as sublime or rhythmically beautiful. Its badness does not fall outside the character which belongs to Paradise Lost as Paradise Lost; the badness is as peculiarly "Miltonic" as the goodness. The vulgarity of mind in Dickens is quite as inseparably Dickensian as the humour and imagination. In fact, to anticipate for the moment, the schoolmen, I should say, were entirely right in holding that it is only of God that it can be said that He is His own goodness.

Even the more mitigated second thesis seems to me, as to Mr. Moore, untenable. I think it plain that the goodness of a poem is not the same thing as the complex of characters on the strength of which it is judged to be good. If a poem has this complex of characters, and if they really are a sufficient ground for calling it good, the poem always is good. When I go on from saying "the poem has these characters" to say "the poem is good," I am not, indeed, in any way crediting the poem with an additional character, and far less conferring such an additional character upon it; I am |402| recognising that it is good, and good because it has the characters already indicated, and the goodness would none the less be there if I did not recognise it. In this sense we may say of the goodness of the poem and its possession of the characters in question that, to use a familiar Aristotelian cliché, ἔστι τὸ αὐτό; but with Aristotle we have to add to τὸ μέντοι εἶναι οὐ τὸ αὐτό. The added statement, "and, since it has these characters, it is good," has brought in something which was not present in the mere statement that the poem has these characters. What is added is a valuation by reference to a teleological standard, a recognition that the poem is what it ought to be. I shall presently attempt to elucidate some of the implications of this ought; for the moment I would merely note the fact of its presence when we speak of the poem as a good one.

I do not mean, of course, that the standard is an externally imposed one, an extrinsic denomination. When I say "this is a good tragedy" I am not intending a comparison with some imaginary tragedy other than the one before me. What is implied is that in being a tragedy at all, the composition aims at exhibiting a certain typical structure, in the phrase of Plato, it βούλεται εἶναι, "means to be," something which it may conceivably not succeed in being. So far as it does succeed, it is a good tragedy. In fact, the thought I want to defend is the old one that what makes the difference between all created things and their Creator is that this nisus to self-transcendence is present in all of them. They all "mean to be" something which they never quite succeed in being, and this is why none of them is identical with its own goodness.

This, I think, is the real point of the account of the ἰδέα τἀγαθοῦ in Plato’s Republic, to which Mr. Joseph alludes in a passage of his recent book referred to by |403| Mr. Moore. According to the Republic, the source of the goodness of everything other than the Form of good lies outside the thing; it lies in the Form of good. The goodness of other things, in the last resort, is not their own; it is a reflection, and an imperfect reflection, of the goodness of the Form, just as their being itself is a similar reflection of the being of the Form. Hence they are neither their own being nor their own goodness. Their "becoming" is a never-completed process of self-transcendence, and at the same time a nisus towards a goodness which is never fully actualised in them. So far as Mr. Joseph’s language indicates such a view, I am in accord with it. But I should diverge from him if he meant, as I do not think he does, that the meaning of the doctrine which he accepts — viz., that God, and presumably God only, is His own goodness, is completely conveyed by saying that God cannot get or lose His goodness. For this would be equally true if goodness were thought of as an "inseparable accident" of God, but the point of the thinkers who invented the famous formula was precisely that in God the distinction between what is essential and what is accidental has no meaning: there are no accidentia of God, and there is no essentia of God, or "nature of deity," primordial or otherwise, distinguishable from God Himself. Plato is human, but Plato is not humanity, and even the phoenix, the "sole Arabian bird," if it existed, would not be "phoenicity," but God is Deity. The reason why God was said to be His own goodness is not, as Mr. Joseph seems to suggest in one passage of his book, no doubt by an inadvertence, that God is substance, but that the whole scheme of the categories is inapplicable to God. In the scholastic sense of the word substance — and it is the scholastic, not the Cartesian sense which is relevant to the scheme of categories — Plato is a |404| substance, and a good one, but no schoolman could have said of Plato as Mr. Joseph is prepared to say of a good poem, that Plato is identical with his own goodness. Their point was that the composition characteristic of created things (whether thought of as composition of form with matter, or of essentia with existence), is not found in God. God is absolutely simple, so that, so to say, in every one of His acts His whole being receives complete expression. It is not with Him as it is with us. A man may be at once powerful, just, merciful. But some of his acts will exhibit one of these characters at the expense of, or in the absence of others. The justice may get in the way of the mercy, or there will at least be occasions when the man’s act discloses his justice but discloses nothing one way or the other as to his mercifulness. But every act of God, just because God is actus purus, exhibits the whole of God.

There is one other point on which I should like to express entire agreement with Mr. Joseph. I hold, as he does, that moral rightness must not be confused with "tendency to produce good," and that a right act is itself good, not merely productive of something else which is good. I cannot deny that wrong acts may be productive of good, and even of moral good, and right acts of evil, and even of moral evil. An atrocious crime produces good, and moral good, when it brings home to many minds the neglected duty for giving proper public protection to the defenceless; the just punishment of a criminal — e.g. a political assassin — is productive of moral evil when it gives rise to sentimental sympathy with the murderer and condonation of his offence.

I turn now to exposition of the view I want to put before you. For the moment I propose to put reference to God on one side, and to start with elementary reflections about the use of the word good as a predicate of |405| the things and persons of everyday life. I confine myself to these because though we often speak of the characters or attributes of persons and things as good, I think what we mean in such cases is that it is the person or thing which is good, and that the character or attribute is specified as the ground for the judgment of goodness. We say "truthfulness is good," "cruelty is bad." But we mean, I think, that a truthful man is good in virtue of being truthful, a cruel man bad, in virtue of being cruel. Now it seems to me misleading to say that what such a predicate means is that its subject possesses a certain quality. The only alternative way of expressing all that I mean when I call a thing good, I believe, is to say that the thing in question is what it ought to be, and it is the presence of this implication of an ought which makes it improper to treat the preposition x is good as being of the same type as x is white or x is sweet.

I do not mean by this that the thing I call good is what I like it to be, what I now want it to be, or am resolved that it shall be. I may call A’s music good, and B’s novel bad, though I know at the moment of speaking that I neither like A’s music nor want to hear it, and that I do like B’s novel, or am bursting to read it. I may, for my own purposes, want to make another person do, and take pains to induce him to do, what I myself judge to be very bad acts, or may be anxious to keep him from doing a very good act. Still less need I mean that the good thing is productive of, or instrumental to, something else. I judge great poetry good, though I do not know that it is instrumental to anything, and though honesty or classical scholarship may be instrumental to various valuable results, I am not thinking of these results, or not primarily of them, when I say that honesty or scholarship is good. Also, in using |406| the word ought in this way, I am employing it in a wider than the specifically ethical sense. I mean the word to cover such cases as those in which we say that this epigram, or this finish to a game at chess, is what it ought to be.

Mr. Moore would perhaps say that I am here committing a "circle in definition," since if I were asked what I mean by saying that the epigram is what it ought to be, I should be brought back to the original statement that it has the character x, and that x is such that the proposition if an epigram has the character x it is good, is true. But I think the alleged circle is not real. The change of expression from this is good to this is what it ought to be brings out in high relief a genuine fact, the teleological reference implied in the use of the predicate good, and it further indicates that when the word good is used in its proper sense, and not as a loose equivalent of useful, the teleology is internal, inherent in the very nature of the thing which is pronounced good. The "nature" of the person or thing in question is not merely to have certain characters, but to tend to a certain completion or fulfillment. In this sense what I may call a forward-looking reference is embedded in the very structure of the subject of the predication. The persons or things of which we are speaking in the class of judgments now under consideration are all persons or things which come to be what they are by a process of development. Their characters are not a mere bundle or complex, nor, again, are they interconnected by relations of mere logical implication: they belong together in virtue of an extra-logical structural plan, which controls the development and expresses itself through it. It is in this sense that they all presuppose a τέλος, a culminating stage in which the structural plan embodied in the whole process finds full completion without either |407| excess or defect. If, or when, this finality of expression is reached, the person or thing actually is just what it was all along tending to be, it is perfect in its kind. So far as it has, as it stands, embodied the plan in its attained development, it is good, though not perfect; so far as it fails to do so, and thus falls short of the exigence of its own "nature," it is defective, and if it falls short of what its own nature demands at this stage of the development, it is bad of its kind. Where, if anywhere, we come across anything in the natural world, like the atom as conceived by the popular science of the last century, which has no internal structural law of development, and so neither grows into, nor is fashioned into anything, but is merely inertly there without any fulfillment of tendency, the predicates good and bad seem to be strictly inapplicable.

This is why, when Spinoza is indulging his more anti-teleological mood, he quite consistently treats good and evil as mere illusions. The difficulty about the doctrine is not merely that on his premises, since he denies that there is any "nature of man" distinguishable from the "nature of this man," he has no right to call Nero a bad man, or bad specimen of man, though this is a serious difficulty enough for a philosopher who sets out to write on ethics: he has not even the right to believe that there is a "nature of Nero," since to admit so much is to allow that Nero βούλεται εἶναι, tends to be, something which Nero actually is not, and so there will be a possible other than the actual. If Spinoza’s professed metaphysic is sound, whenever we speak of the unactualised potentialities of Nero’s nature, or of any man’s nature, we are saying what has no real significance. Hence Blyenbergh was perfectly right in telling him that he had cut himself off from the possibility of having an ethic, as a philosopher who treats potentiality as mere |408| not-being is bound to do. To identify the goodness of a thing simply with its "possession of a quality of some kind" seems to involve this fatal attempt to eliminate process, γένεσις εἰς οὐσίαν, from the world.

When we consider any concrete thing which forms part of the actual world of historical fact, we find it always at once a Werdendes or γιγνόμενον, and a Gewordenes or γεγενημένον. Its history is the record of the passing into actual fact, or the failure to pass into fact, of a nature which belongs from the first to the thing and yet is not, in the first instance, actual in it. The passage may be the growth of an organism or a mind, or it may be the conscious and largely deliberate shaping of a work of art by the craftsman or artist. It is, I would urge, in the realisation of such internal plan or structure as no longer mere possibility, but actual fact, that the goodness of the thing lies. Nor do I think it any real objection to retort that we may say of the bad person or thing, as we do of the good, that it has a history of fulfillment of the "law of its nature," so that I am apparently committing myself to the view that the artistic or moral monster is at once monstrous and also beautiful or virtuous. After all, the old Greek view seems to be sound, that definite form is a principle of goodness in its embodiments, and that evil as such is formlessness. The process by which the bad man, or play, or picture, becomes what it is is not really one of advance to relevant and adequate embodiment of coherent form, but rather one of failure to embody form, or of positive resistance to it. The effect, in these cases, really is an "effect defective." The bad work of art is anarchical; either it exhibits no recognisable controlling form to speak of, or it exhibits one which fits the content no better than a bed of Procrustes, and in either case, it is anarchic. Anarchy again, as Plato reminds us in his |409| picture of the "tyrannical man," is distinctive of the "moral monster." A real Nero is a man distracted; a real Iago is a stunted man.

It is fundamental to the view I am trying to present that there should be no division of the knowable into two disjunct realms, one of the merely real or actual, and another of ideals, or values, or goods, and it is to exclude this separation that I am anxious to insist upon the distinction between the persons and things which are the true subjects of predication of goodness, and the characters in virtue of which goodness is predicated of them. That which is good is, properly speaking, a person or thing: it is Paris, not some character of Paris, that is handsome, Hector, not some character of Hector, who is brave. A character, divorced from its embodiment in a thing or person, is just a concept, and I can see no sense in saying that concepts are good, or bad, or that one is better or worse than another. So again with things and persons which are purely imaginary, like the characters of a play or novel. I do not think anything more can be meant by saying that Wycherley’s Manly is a scoundrel than that if there were a real man who acted as Manly is made to do in the play, that man would be a scoundrel. So if a man were to say that the imaginary society of Plato’s Republic is better (or worse) than London society today, I could only understand him by taking this for an abbreviated way of saying that if there really were such a society as Plato depicts, it would be better (or worse) than that of contemporary London. A thing, to have anything significantly predicated of it, must at least be, and so far as I can see, there is no such mode of being as the "being merely for thought" of which philosophers have sometimes spoken. Would there be any real meaning in the statement that the imaginary Iphigenia of which |410| Aristotle has sketched the plot in his Poetics is a better or worse play than the Iphigenia of Euripides or of Goethe, if the meaning is not simply that if a poet actually constructed a play on the lines laid down by Aristotle the play, in respect of its plot, would be better, or be worse, than the actual play with which the statement compares it?

I am not, of course, denying the reality or importance of the distinction between existence and essence; on the contrary, it is vital to me to insist that the distinction is both real and important. But it is a distinction within what is real, not a distinction between what is real and an unreal something else. That can hardly have been the true meaning of Plato, or of any other supreme philosopher.

In a word, I want to maintain that we shall never understand either judgments about good and bad, or historical judgments unless we are clear on the point that the subjects of such judgments are genuinely individual, that an individual — or at least a finite individual — is a complex of existence or actuality, and essentia, real possibility, and that the essentia is a factor, and ought to be the controlling factor, in its own actualisation; it is at once an efficient cause, as the Americans say an "urge," towards actualisation along certain lines, a formal cause, or law of the process, and a final cause; the whole process of the development is directed upon it, its actualisation "tends" towards it, and will bring it about, if not thwarted or prevented. This is familiar Platonic and Aristotelian doctrine, and calls less for explanation than for vigorous reaffirmation. For the purposes of ethical and aesthetic, as well as of historical studies, it is vital to understand that tendency, or real possibility, is neither unreal, nor is it the same thing as actualisation in fact. The range of real possibility is not |411| indefinite; non omnia possumus omnes is true to the letter; on the other hand, it is wider than that of actual fact. A possibility which is never converted into actuality may none the less be a real possibility. If Shakespeare had been carried off by one of the visitations of plague in the 1590’s, it would still be true that Shakespeare had the possibility of writing Hamlet in him, and that Francis Bacon had not, or that I have not. Even those philosophers who try to believe that an unactualised possibility is nothing at all have not been able to deny that a sentence like that I have just written is at least a popular way of saying something which may be true and be known to be true. But on their theory of the matter, I do not see what the statement "I have not ’got it in me’ to write a play like Hamlet" means, unless it is a mere guess about the future, "I shall never write a play like Hamlet, and if I went on living forever, I never should." But if unactualised possibility is nothing, how do I know that I never shall write such a play, to say nothing of knowing that, with an unending life before me to do it in, I never should? Am I entitled to say more than that I have not yet written a Hamlet? Or to leave self-judgment which is proverbially untrustworthy, out of the question, consider a case of a kind with which we are all familiar. There is a literary work of which we are certain that it must have been produced by one or other of several persons, A, B, C, others being excluded by external evidence of some kind. Is a judgment of the type "This work must have been written by A, because it transcends the powers of B and C, and we may therefore eliminate them" ever more than an arbitrary guess which it is impossible either to make probable or improbable by reasons? Swinburne, for example, ascribed the anonymous play Arden of Faversham to Shakespeare on grounds of this |412| kind; I confess that to myself — if my opinion had any right to count — the ascription seems not only baseless, but preposterous. But will anyone say that a judgment on the point either way can in the nature of the case have no genuine grounds at all?

I hold, then, that the character in a thing in virtue of which we denominate it good is precisely this domination of the process of its "becoming" by an immanent form which is also an end, and that consequently the thing which we can call good, or "the nature" of the thing, is regularly a factor in its own production; in a real sense a good thing is causa sui. Of course, it may be retorted that a work of art, a picture or a poem, does not paint or write itself, it is painted or written by someone. But I do not believe that the objection has any real weight. Even in ordinary language we speak more often of the work of art as "growing" in the artist’s mind, or under the strokes of his chisel or pencil, than we do of its being "made" by him, and we should intend to cast doubts on the artistic quality of the work if we spoke of it as a "manufactured product." Whatever we may think of some of Mr. Alexander’s views about artistic creation — and I own that some of them seem to me strange — he is surely right in saying that the artist has always mixed his mind with his materials, and that the embodied product itself contains the factor of "mind" as well as the factor of "material." The great portrait, for example the Strafford in the National Gallery, is something more than an arrangement of pigments which have been transferred from the tubes that once held them to a certain canvas by the efficient causality of Van Dyck’s mind; it is that, no doubt, but it is also an exterioration or embodiment of Van Dyck’s mind through the medium of pigment on canvas, and this is why we speak of the painter not as manufacturing it but |413| as creating it. If it were only a manufactured thing, we might perhaps call it useful, we should have no reason to call it good. When we judge of the work of art as a work of art, we are thinking of it not merely as having been produced by mind, but as embodying mind; the dominant constituent in it has actually been also the agency in the production.

It follows that for this very reason that the thing which can be called good in such judgments is at once a product and the controlling factor in its own production, the essentia of the thing and its existence tend to fall apart. The thing is not its own goodness, because it is not identical with its own essentia. Hence we can attribute the goodness to it, but we cannot identify the two. The essentia, as I have said, is a real possibility actualising itself. At any stage in this process of self-actualisation there is still possibility not yet actualised, a beyond yet to be attained, and in this sense the goodness of the thing lies outside its own existence. Also we have to remember that a thing in the historical world with which we are concerned is not the only finite factor in its own production; it does not, like a Leibnizian monad, simply unfold its own possibilities from within, subject to the concursus of God. The natures of other things — I should myself say with Mr. Whitehead the natures of all other finite things — are negatively or positively factors in the development. Thus, to express the point in scholastic terminology, the composition characteristic of the good or bad things with which we have actual acquaintance may be characterised more precisely than by calling it composition of essentia with existence; it has the specific character of composition of form with matter. It is an interesting speculation, which I do not wish simply to dismiss, that there may be finite creatures, like the angels of Thomistic |414| philosophy, in whom there is the composition of essentia and existence, but not the more specific composition of form and matter. But I have no actual acquaintance with angels; if I believe in them, it is not on any empirical evidence, and even actual converse with an angel would not of itself be proof that the Thomist theory of the angelic nature is sound. So far as my actual experience of things goes, it seems to be a consequence of the commercium in virtue of which all are factors in the development of each, that in the existentia of each there are elements which are recalcitrant to domination by the form or essentia, and that the completest domination of the first by the second which is ever actually attained is imperfect and impermanent. And thus the same nisus towards complete and relevant embodiment of form in virtue of which a thing has goodness attributed to it appears to be also regularly a nisus towards self-transcendence. In this way, too, the goodness of a thing seems to be outside itself.

We see this, of course, most notably in the life-history of organisms and species of organisms. The goal of the nisus towards complete embodiment of significant form in the history of the individual organism seems to be its own state of adult maturity, its ἀκμή, and yet, so far as the goal is reached, it is only reached to be lost again, as the organism becomes middle-aged, senile, and finally dies. It is an old story, that while the good to which the organism is unconsciously aspiring is its own ἀθανασία, "the complete adjustment" of its own "inner relations" to "outer relations," the good it achieves is something else — the perpetuation of its species in the next generation. And, of course, it is only on a short-sighted view that even this can be said to be really attained; species are not, as the Greek philosophers once fancied, coeval with time; they do not, so far as |415| we can see, even enjoy the privilege of "Dionysiac recurrence"; when they disappear, they disappear once and for all. Hegel spoke once of the "cunning of the Absolute," which dangles an unattainable good before the individual, like the donkey’s bunch of carrots, which is always at the same irreducible distance from the animal’s nose. No doubt, the thought in his mind was that the "race" no more ever gets the carrot than its individual members do. We have, in fact, as Aristotle remarks, in talking about the good, or end, or οὗ ἕνεκα of a process, commonly to distinguish between the οὗ — the actualisation of the possibility — and the ᾧ — the individual in which it is effected, the beneficiary. Thus, in the case of the physician, the οὗ is the establishment of the organic balance which is health, but the ᾧ is the patient who is restored to health, and it is only exceptionally and accidentally, when a man takes his own prescriptions, that the physician’s patient is himself.

If a process of historical development really is, as I have assumed it to be, what Plato calls a γένεσις εἰς οὐσίαν, a process controlled and sustained by and culminating in the embodiment of significant form, where, in the case of the development of the living organism, does the process really reach its goal? Where is possibility finally translated without remainder into actuality? Not within the individual organism, nor yet in the species, or the "kingdom" to which the species itself belongs, nor in any yet wider whole that we can think of. For they all carry within themselves at their fullest attainment the possibilities which will, sooner or later, lead to the breaking down of the embodiment and the "withdrawal" of the form. The goodness of these various wholes, like that of the individual organism, is still not identical with themselves, but beyond them, and so may be said to be adjectival to them, though, for |416| the reasons I have already given, it would not be an adequate statement to call it a "quality" of them.

How far is the case altered if we turn from the growth of the plant or animal to the shaping of a work of art, or the building up of moral character? We have here, of course, to reckon with a new and important factor in the process, intelligent awareness of the self-constitutive nisus and of the general character, at least, of the embodiment to which it is moving. A work of art, no doubt, is born, as we say, in the throes of inspiration; moral character worth having could hardly be acquired by a methodical skill in acquiring one specific "virtuous habit" after another. So far it is true of the artist and the moral hero, nascitur, non fit. But, on the other hand, we may be sure that no real work of art was ever made in the fashion described untruthfully by Aaron, when he said that he threw the people’s offerings into the melting-pot, and "there came out" the golden calf. And I am probably not alone in suspecting that Mr. Alexander’s vision of the statue somehow emerging from the block as Praxiteles struck it at random with his chisel does less than justice to the brooding intelligence which presumably controlled the blows of the implement. Moral character, again, grows, but it does not grow to any purpose apart from an intelligent self-discipline which involves prevision. Whether it be a work of art or his own soul a man is making, to make with any result he must have some prevision, though, no doubt, his foresight only becomes clear and articulate slowly and gradually enough. As Mark Tapley rightly hinted, public buildings do not "grow spontaneous" in the most fertile of soils, and fine moral character, even with the most favourable of social "environments," does not really come up like a flower; |417| it must be worked for and thought taken for it. In virtue of this presence of conscious controlling pre-vision, the making of a work of art, the building up of a character, are less at the mercy of "haphazard" than the growth of an organism. The essentia of the work of art or the character in fieri can dominate the external factors in the process more thoroughly than the "form" can control the growth of the individual animal or plant; less allowance has to be made for the presence within the "matter" of the development of elements intractable to "form." I suppose that we may add that it is in the case of the formation of character that this tractability is at a maximum, since here the "material" itself is peculiarly intimate to the "nature of the individual."

In this respect the work of art, or the personal character, has its good within itself in a sense in which this would not be true of the animal organism. It is at least possible to understand the view that the animal organism fulfills its function completely in making its contribution to the continuance of its species; we could imagine Nature saying, without palpable absurdity, to the creature whose share in this work has been done: "You have no longer any significance for me, and may go to my scrap-heap: you were made for the whole, not the whole for you."But even were it the fact, as it is not, that the example and influence of every work of art and every fine human character are regularly productive of a series of equally good works of art or personal characters, we could not without absurdity, take such a view in these cases. It is no part of the goodness of the Iliad that it should contribute to inspire Virgil to produce the Aeneid, or of the Aeneid that it should be a chief factor in the making of Paradise Lost. It is only the other side of the same fact that the work of |418| art or the human character is not intrinsically subject to old age, as organisms appear to be. The work of art does not decline from its ἀκμή with the passage of time; the worst that happens to it is that it may grow out of fashion. We may lose our appreciation of its beauty, but the beauty remains there herrlich wie am ersten Tag; it is always young, always at its ἀκμή. And when we come to consider personal character, it becomes positively preposterous to conceive of the process of its formation as culminating in an ἀκμή to be followed by a curve of descent. In the nature of the process itself, there is no reason why it should not go on forever, as Kant held that it does, or why, if it does culminate in an attained perfection, this perfection itself should not simply persist. When both these alternatives are rejected, as they are by so many among our contemporaries, they are rejected, not on the ground of their inherent absurdity, but on the strength of a conviction that the permanence of personality is causally dependent upon the preservation of a physical organism. I am not here to enter on the question whether this is or is not a true conviction; I am only concerned with the point that impermanence is no part of the intrinsic character of the individual moral person, as it seems perhaps to be of the intrinsic character of the individual organism. The nisus of a moral person towards his own "personal good" does not necessarily burst the mould of his own personality. This does appear to be what happens to the natural organism, which has to take leave of its own existence in attaining what Whitehead calls its "objective immortality"; and so has its good "outside itself" in a way in which we may hope that a person has not.

Yet, when all this has been said, however nobly we may "think of the soul," and however generous the |419| hopes we allow ourselves to entertain of its destiny, it remains true that the good man’s personal goodness is not identical with him; it is something which is his, but is not himself, and so must be said to be adjectival to him. In the first place, as Mr. Joseph’s language reminds us, it is something he does not possess without having to win it first, and it is also something he conceivably may lose, without ipso facto ceasing to be himself. At least, this must be so if man is really an historical being, if process is the very stuff of his being. And if each of us is really a strictly eternal being, a god mistaking himself for something else, like the persons who compose the Absolute in the philosophy of McTaggart, I confess I do not understand how the "misperception" which infects our experience of ourselves and of one another gets its first hold of us. Indeed, in spite of all the acuteness McTaggart shows in repelling the objector — an acuteness which I can only admire and envy — I think he leaves the really fundamental objection to his construction untouched. The one apparent reality of our human situation, into which no element of illusion due to misperception enters, is, as I understand McTaggart, the love of each person for his personal friends, and McTaggart, at least, seems to be fully aware that the reality of the love demands that the various persons who feel it are constitutive of the being of one another. But how can they be thus related if process itself is the illusion which he pronounces it to be? If he is to be consistent with his treatment of becoming as illusory, must he not convert his persons into self-contained Leibnizian monads, none of whom ought, in strictness, to have any suspicion that there is anything except itself? My standing difficulty with a metaphysic like McTaggart’s, is that it seems to me to demand at once that each of us shall be such a monad, |420| and that he shall not be. If he is to love his friends, he cannot be such a monad; if he is to have the character McTaggart describes as a "fundamental differentiation of the Absolute," he can be nothing else.

Let us look again for a moment at the reasons for denying that I am my own goodness. If we do so, shall we not see that though I am certainly the possessor of my own goodness, the subject to which it is truly attributed, I am, so to say, not a possessor in my own solitary and exclusive right? My goodness, such as it is, is the actualisation of my real possibilities, and I have myself been a causative factor, and the determining causative factor, in the process of actualisation; it has been brought about by my use or neglect of my opportunities, and neither the use nor the neglect has been forced upon me. But my “…”real possibilities“…” themselves are rooted in my various relations with the living and the dead, the animate and the inanimate, in the whole world-wide scheme of beings of whom I am only one. Positively or negatively, all of them have gone to constitute the whole system of the possibilities. My essentia is, indeed, mine and not yours, or that of any other being, but it could have no subsistence if it stood alone, just as the essentia of the circle defined by Euclid is other than that of any other of the figures he defines, but that there should be such an essentia at all implies that there should be a region of space as defined by Euclid’s scheme of postulates, and thus implies the essentia of all the other figures of Euclid’s geometry. So far, then, is my goodness from being me, that while I can say that I have it, and in a real sense that I have made it, I cannot say that I have created it. When I say that I have it, I am to remember also that you can retort on me with equal truth, “…”What has thou that thou hast not received?“…” And the same question may be |421| asked with the same force of any being into whose actuality any element of historical becoming enters, even of a being with as little unrealised potentiality about itself as one of the angels of the Thomist philosophy. None of them all can be identical with its own goodness, because all are characterised by the composition of essentia and existence, potentiality and “…”act.“…”

Whether there can be any good thing which is its own goodness, then, depends on the answer we give to a question of ultimate metaphysic — is it true, or is it not, that “…”real possibility“…” presupposes a ground — a principle of distinction between the possible and the impossible — which is itself an actuality? We are all familiar with this position from the prominence given to it in the metaphysic of Lotze, and before him by Kant in the “…”pre-critical days“…” when he made it the foundation of a repeatedly urged “…”demonstration of the being of God“…” which, so far as I can see, is left untouched by the furious assaults of the Kritik upon the “…”cosmological argument.“…” It is also, of course, the foundation of the Aristotelian theism, and so of the famous “…”five ways“…” of St. Thomas. And I need not remind you of the part it has played in our own day in the Naturphilosophie of Dr. Whitehead. It would take us too far afield from our immediate subject to enter on a discussion of the truth of the principle, and therefore I will only say here that I see no way to escape from it. If it is a sound principle, then there must stand at the source of all historical development a being which is strictly eternal and super-historical, because it, and it alone, is actus parus, complete actuality without any element of the merely “…”possible,“…” and it would not be hard to show that such a being, and only such a being, possesses the full character of God. Such a being could |422| have no “…”nature“…” distinguishable from, and making itself through, the phases of its actual existence. Here, and only here, the distinction between essentia and existence could have no meaning, and consequently the distinction between an attribute and that which is the subject of the attribute would be meaningless too. Such a being would be good and perfectly good, and that in an “…”eminent“…” sense. It would be more than even the complete actualisation of all potentialities, since it would be the eternally actual source of them all. Hence, unlike anything which has a history, it would eternally be its own goodness. To say that it is good would be to say that it is that which it is, whereas to say of anything else “…”this is good“…” is only to say “…”this is becoming, or this has become, what it ought to be.“…” Thus here, at last, we should have found something which has what it has not “…”received,“…” indeed has received nothing, but is what it is, and is good, strictly in its own right. If there is such a being, as I believe there is, we can say of it that it is its own goodness. Its goodness is not adjectival to it because, as I say, in it it is all one to be and to be good. But though this may be true of God, it is not true of me, or of anything which is not God. What I have I have “received” and received from many quarters, and however true it may be that I make my own goodness out of what I have thus received by “free” volitions of my own, I can only exercise free volition “under God”; my own volition is itself part and parcel of the historical process which is only possible in virtue of the actual existence of the strictly eternal and actual source of all possibility. Hence my goodness — even in “heaven,” if I ever reach “heaven” — is communicated and dependent, not inherent and original.

Top ↑

Alfred Edward Taylor
Eminent British Idealist philosopher
Fellow, Merton College, Oxford, Professor of Logic and Metaphysics, McGill University, Professor of Moral Philosophy, University of St. Andrews and University of Edinburgh.



Taylor, A. E. Philosophical Studies. London: MacMillan & Co., Ltd, 1934. This work is in the Public Domain.