KANT AND HIS INFLUENCE Edited by George MacDonald Ross and Tony McWalter
This page intentionally left blank
KANT AN...
214 downloads
3450 Views
17MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
KANT AND HIS INFLUENCE Edited by George MacDonald Ross and Tony McWalter
This page intentionally left blank
KANT AND HIS INFLUENCE Edited by George MacDonald Ross and Tony Me Walter
continuum
This edition published by Continuum International Publishing Group, 2005 Continuum International Publishing Group The Tower Building 15 East 26th Street 11 York Road New York, NY 10010 London SE1 7NX © Continuum International Publishing Group 2005 Introduction and editorial selection © George MacDonald Ross and Tony McWalter and contributors, 1990
British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: 0-8264-8853-6 (paperback) Previously published in hardback by Thoemmes Press Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress. Printed and bound in Great Britain by MPG Books Limited, Cornwall
CONTENTS
INTRODUCTION
vii
'OUGHT' IMPLIES 'CAN': KANT AND LU fHER, A CONTRAST Roger M. White CONFUSED PERCEPTIONS, DARKENED CONCEPTS: SOME FEATURES OF KANT'S LEIBNIZ-CRITIQUE Catherine Wilson
1
73
THOUGHT AND SENSIBILITY IN LEIBNIZ, KANT AND BRADLEY Guy Stock
104
'ORIGINAL NONSENSE': ART AND GENIUS IN KANT'S AESTHETIC Peter Lewis
126
FICHTE, BECK AND SCHELLING IN KANT'S OPUS POSTUMUM Eckart Fdrster
146
v
vi
Contents
IMAGINATION AS A CONNECTING MIDDLE IN SCHELLING'S RECONSTRUCTION OF KANT John Llewelyn
170
THE EARLY RECEPTION OF KANT'S THOUGHT IN ENGLAND 1785-1805 Giuseppe Micheli
202
HAMILTON'S READING OF KANT: A CHAPTER IN THE EARLY SCOTTISH RECEPTION OF KANT'S THOUGHT Manfred Kuehn
315
ASPECTS OF KANT'S INFLUENCE ON BRITISH THEOLOGY Donald MacKinnon
348
NOTES ON CONTRIBUTORS
367
INDEX OF NAMES
371
INTRODUCTION George MacDonald Ross and Tony McWalter
The papers in this volume were originally delivered at a conference held at the University of Leeds in April 1990. The conference was organized by the British Society for the History of Philosophy, one of the aims of which is to promote a broader and more scholarly approach to the study of the history of philosophy than has generally been'characteristic of philosophers in the analytic tradition. In particular, it was felt that, at least in England, too little attention was being paid to Kant and the postKantian philosophy of the nineteenth century; and it is no accident that nearly all the contributors are from Scotland or abroad. The first four papers are primarily concerned with the interpretation of various aspects of Kant's philosophy, but setting him firmly in a historical context. The order in which they appear represents a gradual shift in emphasis from his connections with earlier thinkers to comparisons with subsequent developments. Roger M. White's paper, '"Ought" implies "Can": Kant and Luther, a Contrast', is untypical of the collection as a whole, in as much as his stance on this particular issue is unrepentantly anti-Kantian. White argues that although Kant does not explicitly mention Luther, he must have been aware of the opposition between his own position and Luther's a debere ad posse vn
viii
Introduction
non valet consequentia ('ought' does not imply 'can') but to have made this explicit would have brought him into conflict with the ecclesiastical authorities. White provides a detailed and critical analysis of what Kant may have meant by 'Ought' implies 'Can\ and of how he might have argued for his position. He lays particular stress on the paradoxical consequences of Kant's complete divorce between the absolutely free rational will, and the causally determined empirical self. He then compares Kant's approach with that of Luther, and concludes that Kant, like Erasmus before him, made the mistake of putting the question of praise or blame before that of the nature of the good life and how, if at all, it can be attained. Two authors concerned with Kant's metaphysics have focused on his relations with Leibniz. If it was Hume who awoke Kant from his dogmatic slumbers, it was Leibniz who provided the main starting-point for his critical philosophy. Catherine Wilson's paper, 'Confused Perceptions, Darkened Concepts: Some Features of Kant's Leibniz-critique', will prove a useful corrective for those in the Anglo-American tradition who think of the Leibnizian strains in Kant's work as an embarrassment or an unnecessary encumbrance. It helps us to ascertain how ambivalent Kant's position was, and how crucial it is for us to recognize this ambivalence if we are to understand the critical philosophy properly. Wilson focuses on Kant's criticisms of Leibniz's notion of confused perception. She maintains that it is by no means clear that Leibniz saw concepts and perceptions as differing only in degree (the former being 'distinct', the latter 'confused'), and that it is paradoxical to accuse him of failing to distinguish phenomena and noumena - although some of his followers were indeed guilty of such confusions. In his later writings, Kant acknowledged the possibility of a
Introduction
ix
Platonic interpretation of Leibniz, in which phenomena and noumena are sharply distinguished; and Wilson suggests that Kant himself was troubled by the claims of noumena to be over or behind the appearances in some way. In 'Thought and Sensibility in Leibniz, Kant and Bradley', Guy Stock starts out from Whitehead's dictum that 'Kant, in his final metaphysics, must either retreat to Leibniz, or advance to Bradley'. He concentrates on the connected problems of the relation between thought and individual reality, and of the distinction between the actual and the merely possible. Leibniz resolved the former through his privative account of sensibility and his doctrine of the individual as an infima species, but he failed to distinguish adequately between the actual and merely possible worlds. Kant, on the other hand, made a sharp separation between thought and sensibility, and maintained that the actual world is the one which is given in empirical intuition. Bradley rejected both Leibniz's account of sensibility, and Kant's epistemological dualism, together with the consequential doctrine of the thing-in-itself. However, the resultant metaphysical system was closer to Leibniz's than to Kant's. Peter Lewis's paper, '"Original Nonsense": Art and Genius in Kant's Aesthetic', aims to identify what Kant means by 'original nonsense' in the context of his discussion of art and genius in The Critique of Aesthetic Judgment. Lewis argues that Kant's great insight is that works of artistic genius, in virtue of being original and exemplary, are essentially embedded in traditions constituted by works of art. Works of genius provide rules to be followed in the work of non-genius, and set standards of excellence for the work of future genius. In the course of the paper, Lewis draws attention to significant similarities between the views of Kant and Wittgenstein on genius and taste in art.
x
Introduction
The remainder of the papers in the volume are concerned more historically with Kant's direct influence on subsequent thinkers in various disciplines and countries first, with his early influence in Germany. Eckart Forster is currently preparing an English edition of Kant's Opus postumum, a much-neglected work in which Kant deals with the transition from the metaphysical foundations of natural science to physics. In 'Fichte, Beck and Schelling in Kant's Opus postumum\ Forster describes the aim and content of the work, and argues that it constitutes a significant revision of Kant's critical position. In the Opus postumum there are references to Fichte, Beck and Schelling, who were all followers of Kant. Some scholars have claimed that the changes in Kant's position were due to their influence. After examining the evidence, Forster concludes that Kant may have been influenced by Schelling and Beck (though not by Fichte), but that his thoughts were in any case going in much the same direction as theirs. In his paper, 'Imagination as a Connecting Middle in Schelling's Reconstruction of Kant', John Llewelyn concentrates on Schelling's development of the Kantian concept of imagination as the 'connecting middle' between theory and practice. After describing Kant's account of imagination, Llewelyn analyses the difference made to Kant's critical idealism by Schelling's claim that we have non-discursive intellectual constructive intuition as well as empirical intuition. The focus shifts from the relationship between philosophy and mathematics to that between philosophy and art; and Schelling is seen as the connecting middle between Kant and Heidegger. To turn to Britain, it is remarkable how small a role Kant's thought has played in the intellectual life of the English-speaking philosophical world. The last three papers tell part of the story of this neglect.
Introduction
xi
However, any discussion of the history of philosophical thought in eighteenth- and nineteenth-century Britain needs to preserve a sharp distinction between England and Scotland. In England, philosophy was virtually extinct as an academic discipline, returning only gradually during the second half of the nineteenth century. Such philosophical debate as there was existed largely outside the university world, in literary clubs and journals. In Scotland, by contrast, philosophical discussion among the educated laity was underpinned by a university curriculum which had philosophy at its very core (see Davie, The Democratic Intellect). In his exhaustive study, 'The Early Reception of Kant's Thought in England 1785-1805', Giuseppe Micheli has had to rely mainly on reviews and articles in literary journals, and the few translations and commentaries that appeared between those years. The picture he paints is a bleak one indeed: there was little interest in Kant's philosophy, and even less understanding of his central ideas. He was perceived mainly as a political writer, and a subversive one at that - encouraging his followers to reject the established political order, religious belief, and moral values. By the turn of the century increasing repression, and censorship of ideas emanating from the Continent, put a virtual stop to the study of Kant's work. With the exception of Coleridge, Kant remained a closed book to English thinkers until the 1830s. As for the early reception of Kant in Scotland, Manfred Kuehn, in his 'Hamilton's Reading of Kant: A Chapter in the Early Scottish Reception of Kant's Thought', shows that, despite the widespread interest in metaphysics, Kant's ideas were at first understood hardly any better than they were in England. The conventional wisdom is that Sir William Hamilton (1788-1856) was the earliest Scottish philosopher to be substantially
xii
Introduction
influenced by Kant. Kuehn argues that, although Hamilton had an intimate knowledge of Kant's texts, he was in fact quite hostile to his ideas, and much less influenced by him than has generally been supposed. Finally, in his paper 'Aspects of Kant's Influence on British Theology', Donald MacKinnon stresses that the pivot of Kant's influence lies in his doctrine of the primacy of practical reason. He criticizes the view of Kant as the 'philosopher of Protestantism', and he displays his influence on a wide range of British theologians from different denominations. MacKinnon concludes that Kant's influence did not take the form of slavish imitation, but of conversion of his insights to uses he could not have foreseen, and of which he might not have approved. The papers in this volume give only a taste of Kant's range of interests and of his influence. There is ample scope for a series of conferences and accompanying volumes covering his influence at different periods, in different countries, and in different disciplines. In particular, there is the issue of the revival of interest in his philosophy in Britain in the 1830s, and the subsequent decline in any fundamental influence of his thought during the twentieth century. This raises the question of whether Kant's philosophy deserves to be more influential. The Europe of 1989 shared with the Europe of 1789 the distinction of heralding a new political order. The year 1989 was one of progressive and democratically orientated revolution, and the influence of German culture is an issue for us in the latter part of the twentieth century as it was an issue in the latter part of the eighteenth. What Kant offers is the prospect of a worldview which portrays the ethical as fundamental to politics; and he provides an account of human nature
Introduction
xiii
which at least makes possible an ethical commonwealth. These concerns are relevant not merely to the politics of the late twentieth century, but to political theory in general. Kant's thesis that human nature is so constituted that an ethical commonwealth is possible is itself grounded on metaphysics. He sought to produce a Weltanschauung in which 'epistemology is logically prior to ethics in that it must prepare the ground for a philosophically defensible ethics, but [in which] ethics itself is practically prior to epistemology, because the development of an adequate epistemology is a task set by the highest good' (Van der Linden, p. 10). For Kant, epistemology and ethics, or more generally philosophical theory and praxis, are inextricably intertwined. It has long been fashionable for intellectuals in the British tradition to hold what one can call a modest view of philosophy. Integral to this view is the idea that, while once all knowledge was the province of philosophy, the history of its subsequent development is a history of subjects emancipating themselves from their parent disciplines. Ever greater specialization and ever greater expertise is required for work at the frontiers of knowledge. The modest view has it that, as subjects split from philosophy, as they develop their own methods, and as the knowledge-base becomes ever greater, so the field for philosophy contracts and becomes more focused. Some have felt that the philosophical residuum for the twentieth century has been the analysis of linguistic expressions, while others have gone so far as to maintain that linguistic analysis can itself be hived off, so that nothing remains for the philosophers to study other than the thesis that philosophers have nothing to study. If such conceptions of philosophy were to remain ascendant, then there would be no serious future for the subject.
xiv
Introduction
There is, however, a brighter prospect. While it is true that the knowledge industry grows ever more voluminous, it is also true that we have become ever more conscious of the limitations which the fragmentation of knowledge has engendered. Often the most exciting work, the search for solutions to pressing theoretical and practical problems, involves thinking at the frontier between two or more disciplines. What is needed is just the sort of overview of a whole problem area which philosophers have traditionally sought to attain. Reflection on the limitations of the methods used to acquire knowledge in particular disciplines can make us humble about the attainability of absolute truth; but it can also make us realize the importance of approaching problems from a different direction, or of setting up new forms of enquiry. Prominent English-speaking philosophers have maintained that philosophy provides no answers: but to adopt this as a motto for general philosophical practice is a recipe for the long-term decay of philosophy. It is entirely reasonable for the public to ask what sorts of problems a philosophical training enables one to tackle. We should not be seduced by the frequent demand for yes-or-no answers where these are inappropriate; but we must be willing at least to say what kinds of judgement are cultivated by a philosophical training. If we believe that the philosopher's judgement is more widely informed, more objective, based on sounder reasoning, and less bound by the presuppositions of particular disciplines than that of the non-philosopher, then we must emphasize that philosophy is not just a pleasant, abstract pastime, but a practical and useful activity, the diminution or absence of which would impoverish society in multifarious ways. The claim that philosophy must be practical as well as
Introduction
xv
theoretical is a thoroughly Kantian ideal; and it is testimony to the limited influence Kant has had in Britain that it should still be necessary to plead the case. If the case is accepted, it follows that philosophy should play a far more central role in the educational curriculum - a role it used to play in Scotland, and which it still plays in most of the continent of Europe. Kant himself regarded the whole of human knowledge as the province of the philosopher, and in presenting this book we hope to do something to rehabilitate the view that philosophers should be concerned with the full range of intellectual and practical problems facing mankind. His life's work is not simply an episode in the history of philosophy, but a rich resource from which we can derive inspiration for the future development of philosophy as an academic discipline.
Bibliography Davie, George E., The Democratic Intellect: Scotland and her Universities in the Nineteenth Century (Edinburgh University Press: Edinburgh, 1961). Van der Linden, Harry, Kantian Ethics and Socialism (Hackett: Indianapolis, 1988).
This page intentionally left blank
'OUGHT' IMPLIES 'CAN': KANT AND LUTHER, A CONTRAST Roger M. White
The aim of this essay is to examine a central idea of Kant's moral philosophy by setting it in the context of a radically opposed set of ideas - the ideas which inform the thinking of Martin Luther; in particular, the early Luther of The Disputation against Scholastic Theology and The Heidelberg Disputation, together with the positions that he develops in the famous controversy with Erasmus. For, if the idea that 'Ought' implies 'Can' (that we may infer from the fact that we ought to do something that we are able to do it) has central structural significance in Kant's whole moral thought, the directly opposed idea a Debere ad Posse non valet consequentia is equally fundamental in Luther's thought. How far, in doing this, I am setting Kant in a historical context is impossible to determine. For in those places where one would most naturally look for a discussion of Luther - Religion Within the Limits of Reason Alone and The Dispute of the Faculties - Luther is not explicitly referred to, and even if that which Kant calls 'ecclesiastical faith' has distinctly Lutheran features at many points, its named representatives - above all, Spener and others from the Lutheran pietist tradition 1
2
Kant and His Influence
are closer to Kant than to Luther on many of the key issues that confront us here. Given the sheer extent of Kant's writings, I cannot say for certain that he never explicitly discusses Luther, or even that he was conscious of Luther's theological positions other than through the frequently distorting glass of pietism. But, in view of Kant's religious background and upbringing, he must clearly have been familiar, if not with Luther's own writings, at least with texts such as the Heidelberg Catechism with its formative influence on the pietist movement, and therefore, it seems, would have to be conscious of the extent to which he was putting forward positions that were in direct conflict with the theologians of the classical Reformation. For instance, in the Heidelberg Catechism we find: Question 8: Are we then so depraved that we are utterly incapable of performing any good work and are inclined to all that is evil? Yes: unless it be that we are born again by the spirit of God. Question 9: Does not God then act unfairly by man, in that in His law He requires of him what he cannot perform? No: for God has so created man that he is capable of performing the good; but, by the instigation of the devil and through wilful disobedience, man has deprived himself and all his posterity of these gifts. This gives a clear statement of the complex of ideas to which Kant is reacting in works such as The Dispute of the Faculties and Religion Within the Limits of Reason Alone. But when Kant arraigns ecclesiastical faith before the bar of reason, he does not cite the figures of the classical Reformation, not even Ursinus and Olevianus
'Ought' implies 'Can': Kant and Luther
3
let alone Luther himself. The representatives of ecclesiastical faith come from a generation later (Spener, Franck and Count Zinzendorf). But despite their close historical proximity to the Reformation and, indeed, their working within a framework directly given to them by the Reformers, these authors seem to me already to share many assumptions with Kant which are not to be found in the Heidelberg Catechism and which would be explicitly rejected by Luther himself. As a result, in the confrontation between 'ecclesiastical faith' and 'moral religion', Kant has a surprisingly easy victory. If we wish to test Kant's conception, we must leave Kant's chosen battlefield and confront him with a wholly different range of issues from those we find within the confines of pietism. We shall then find many of the questions are far more subtle than they appear when we allow the terms of the debate to be dictated by the confrontation between Kant and Spener. The historical story leading from Luther to Kant is an enormously complicated one, and, as I have said, it is in any case impossible to determine how much of that story was known to Kant. A reading of his writings in the Philosophy of Religion suggests that it might in crucial respects have been quite sketchy. In any case, Kant was well aware of the extent to which his religious writings were sailing close to the wind with the ecclesiastical authorities, and trouble with the censors was a real difficulty he had to contend with: to criticize Luther directly, or to comment on the divergence between what Kant was himself saying and what Luther had said, might well have seemed needlessly provocative. What I propose in this paper is a direct confrontation between Kant's key positions and the antithetical positions argued for by Luther. This is not in a direct sense a historical confrontation, but an attempt at a
4
Kant and His Influence
confrontation between two radically opposed conceptions of the good life. In an indirect sense this is a historical confrontation, since one of the results of such a confrontation is to highlight a series of assumptions which Kant makes, assumptions which appear selfevident to Kant only because of the particular moment in the history of thought within which he is writing, but the particular nature of which only emerges fully if you confront him with a thinker who is not working within the confines that Kant takes for granted. It is, perhaps, worth emphasizing finally that whatever the precise historical relation of Kant to Luther, comparing him with Luther is the right kind of context in which to discuss his ethical thought. That is to say, despite Kant's theoretical agnosticism, it is clear not only that Kant's starting point is within the Christian tradition but also that the question of ethics remained for him a religious question. Kant would have understood what was meant if the question of ethics had been put in the form 'What shall I do to inherit eternal life?', in a way in which that question would be unintelligible to many of his twentieth-century followers.
I Although the slogan, in the form in which it is usually cited, 'Ought' implies 'Can\ does not seem to occur in Kant's writings, the idea is omnipresent in the ethical writings: most fundamentally and famously in Kant's proof of the existence of freedom. Here there is a clear progression, at least in Kant's presentation of his conception. In the first Critique, the concern is the negative one of showing that theoretical reason, with its insistence on complete causal determination within the world as phenomenon, was still incompetent to rule out the
'Ought' implies 'Can': Kant and Luther
5
possibility of a transcendental freedom. Then, in the Foundations, freedom is presented as one of the three Ideas which practical reason must postulate in making sense of Man's moral endeavour. But in the second Critique, Kant marks off freedom from the other two Ideas God and immortality - as the only one which is known to have application, morality having become not merely a reason for adopting an 'as if hypothesis of freedom, but the ratio cognoscendi of freedom (second Critique, p. 119). (All citations, where appropriate, refer to Lewis White Beck's translation of the second Critique and other ethical writings of Kant.) It is, however, in the first Critique that Kant's argument is first adumbrated where, for example, we find that, in commenting on the third Antinomy, he writes: practical freedom presupposes that although something has not happened, it ought to have happened, and that its cause, as found in the field of appearance, is not, therefore, so determining that it excludes a causality of our will . . . (first Critique, A 534). and, most explicitly, in the Canon of Pure Reason: since reason commands that such actions should take place, it must be possible for them to take place. Consequently, a special kind of systematic unity, namely the moral, must likewise be possible, (first Critique, A 807) In the second Critique, the argument is illustrated as follows: Suppose that someone says his lust is irresistible when the desired object and opportunity are present. Ask him whether he would not control his passion if, in front of the house where he has this opportunity, a
6
Kant and His Influence
gallows were erected on which he would be hanged immediately after gratifying his lust. We do not have to guess very long what his answer would be. But ask him whether he thinks it would be possible for him to overcome his love of life, however great it may be, if his sovereign threatened him with the same sudden death unless he made a false deposition against an honourable man whom the ruler wished to destroy under a plausible pretext. Whether he would or not he perhaps will not venture to say; but that it would be possible for him he would certainly admit without hesitation. He judges, therefore, that he can do something because he knows he ought, and he recognizes that he is free - a fact which, without the moral law, would have remained unknown to him (p. 141f). Taken in isolation, this last passage could be misleading, suggesting some kind of argument from moral experience, an argument that depended upon the empirical phenomena of choice, decision-making and conscience. But that would be to mistake Kant's intentions completely, in a number of respects. (1) For Kant, all empirical phenomena are, as such, susceptible to full deterministic explanations, and that will include all the phenomena of willing and deciding. (2) Any argument which took as its starting point the empirical phenomena of willing could only have as its conclusion empirical freedom, whereas Kant wishes to establish the existence of transcendental freedom - a freedom in which I am the ultimate origin of my action. Kant's fundamental position here is, 'it is useless to endeavour to prove transcendental propositions by examples' (first Critique, A 554).
'Ought' implies 'Can': Kant and Luther
7
(3) Kant, at a number of points, explicitly reckons with the case where I indeed find within myself no capacity to obey the categorical imperative, and yet, in that I still say that I ought to do so, I ipso facto acknowledge that I am able to. (See, for instance, the way Kant treats the example of a malicious lie in the first Critique, A 554ff and, in particular, 'When we say that in spite of his whole previous course of life the agent could have refrained from lying, this only means that the act is under the immediate power of reason, and that reason in its causality is not subject to any conditions of appearance or of time', or, most clearly, where Kant discusses Man as fallen Man in Religion 'Within the Limits of Reason Alone, Bk. I, General Observation. How it is possible for a naturally evil man to make himself a good man wholly surpasses our comprehension . . . despite the fall the injunction that we ought to become better men resounds unabatedly in our souls; hence this must be within our power. . . . But does not this restoration through one's own exertions directly contradict the postulate of the innate corruption of man which unfits him for all good? Yes, to be sure, as far as the conceivability, i.e., our insight into the possibility, of such a restoration is concerned. . . . For when the moral law commands that we ought to be better men, it follows inevitably that we must be able to be better men. In this discussion, it becomes clear that Kant holds that 'Ought' implies 'Can' establishes human freedom to obey the moral law, even where empirically we detect no capacity to do so.) (4) Kant intends his proof of freedom to be valid for
8
Kant and His Influence
all rational agents, qua rational agents, not merely for Man. So that, even if Man is the only rational agent we encounter, we are still to regard this proof as one that would be valid for any other rational agent, if there were such. As such, the proof is intended as one which derives its validity purely from an examination of the notion of a rational agent, without reference to particular facts of human nature or experience. The proof must be understood in the following way: We know that we ought to act in certain ways, act on some maxims rather than others. What is more, we know that this is so even when we do not behave in those ways. This premise is to be regarded as established not on the basis of a supposed moral sense or conscience, at least as popularly understood. We know this on the basis of reason alone: our reason itself inexorably proposes to us a course of action as the right one for us to follow. If something ought to happen, then it must be possible for it to happen. Therefore, we must be free to follow the dictates of reason. The general picture Kant is trying to present here is clear, even if the notion of transcendental freedom is fraught with paradox and difficulty: considered as an empirical being, I must be regarded as subject to the conditions of space and time, and, as such, my actions, as phenomenal actions, susceptible to full causal explanation within the phenomenal world. There will be nothing I do, think or feel, no decision I make, that cannot, in principle, be causally explained in terms of my earlier history. From this point of view, I must be regarded purely as the slave of my environment, nature and passions. However, in addition to considering myself as
'Ought' implies 'Can': Kant and Luther
9
an empirical being - that is to say an object of empirical observation, of scientific study - I have to regard myself as an agent, confronted by decisions to be made. I have to decide what to do. As an agent, deciding what to do, I am aware not only of what I am inclined to do, but also of what my reason tells me I ought to do. Since that is so, it is possible for me to follow not the inclinations of my nature, but the dictates of my reason. I am capable of an act of self-transcendence, in which I translate myself into a supersensible realm and become the ultimate author of what I do. A. Although Kant was of course aware of the difficulties that beset this picture, it is not clear whether he fully realized how close to complete incoherence it is. The central difficulty is that the action that morality dictates will be an action within the phenomenal world. If I do something, whether or not I do it because I ought to do it, that will be an empirical event: so that even when I do my duty, on Kant's own terms, what I do will still be susceptible to a full causal explanation. Kant writes: Inasmuch as it (the acting subject) is noumenon, nothing happens in it; there can be no change requiring dynamical determination in time, and therefore no causal dependence upon appearances. And consequently, since natural necessity is to be met with only in the sensible world, this active being must be independent of, and free from all such necessity. No action begins in this active being itself; but yet we may quite correctly say that the active being of itself begins its effects in the sensible world (first Critique, A 541). This serves to bring out the difficulty, rather than to alleviate it. It may well be that this 'positive' use of the concept of the noumenon is the unhappiest aspect of
10
Kant and His Influence
Kant's development of transcendental idealism: it always seems to place an intolerable strain upon the tenability of his contrast between the phenomenal and noumenal, rather than rendering intelligible his concept of man as a free agent. Kant's own attitude to these difficulties was twofold: (1) We are here within the sphere of practical reason, concerned with the question, 'What ought I to do?' To that extent, for Kant, the theoretical question, 'Do I know that I am free?' was less important than the practical question, 'Ought I to regard myself as free?' Here, the crucial idea would be the idea that if I regard myself as unfree, as not ultimately responsible for my actions, this would threaten to expose morality itself as an illusion. Since, therefore, the moral precept is at the same time my maxim (reason prescribing that it should be so) I inevitably believe in the existence of God and in a future life, and I am certain that nothing can shake this belief, since my moral principles would thereby be themselves overthrown, and I cannot disclaim them without becoming abhorrent in my own eyes (first Critique, A 828). What Kant says here about God and immortality will clearly, a fortiori, apply to the third, and most fundamental, postulate of Practical Reason. From this point of view, the important point was not so much proving that we are free, as establishing the possibility of believing ourselves to be so. Knowledge is to be removed so as to create a space for faith. From this standpoint, even if we could not form to ourselves a clear picture of human freedom, that would not matter if we could establish the bare possibility of our being so, without any precise knowledge of how we are so.
'Ought' implies 'Can': Kant and Luther
11
(2) However, as Kant's thought progresses, he does not content himself with thinking of freedom as no more than a postulate of Practical Reason. Morality becomes the ratio cognoscendi of freedom. Whereas we do not know there to be a God, or that we are immortal, Kant claims that we know we are free: the basis for this knowledge being the proof with which we are concerned. Here, even if we find it impossible to form for ourselves an adequate conception of what it is like for us to be free, if the argument for freedom is valid, then we can know that we are free, even without a coherent account of how we are free. Is our knowledge really widened in such a way by pure practical reason, and is that which was transcendent for speculative reason immanent in practical reason? Certainly, but only from a practical point of view. . . . But how freedom is possible, and how we should think theoretically and positively of this type of causality, is not thereby discovered. All that is comprehended is that such a causality is postulated through the moral law and for its sake. It is the same with the remaining ideas, whose possibility cannot be fathomed by human understanding, though no sophistry will ever wrest from the conviction of even the most ordinary man an admission that they are not true (second Critique, p. 216). B. A few comments about the concept of freedom at stake are in order here. First, there is a minor inconsistency in Kant's way of speaking: am I only free when I obey the categorical imperative, or am I always free to obey the categorical imperative? And, if I do not but succumb to my inclinations, do I freely do wrong - so that I can be justly censured for so doing? This seems
12
Kant and His Influence
largely a terminological matter, but Kant's preferred terminology here is undoubtedly to think of the free man as the man who is acting morally. On this use of the concept, freedom is not to be thought of primarily as the ability to do otherwise but as freedom stemming from the fact that I myself am the author of the principles of my action. Moral freedom is a freedom under law, but the moral law is one 'in whose service is perfect liberty'. The main point here is that, for Kant, completely arbitrary, random action is a nonsense notion: all action is explicable, is action for a reason. Where it is correct to describe someone as acting, it is always appropriate to ask why they did what they did. The difference between free action and unfree action is not that the former is lawless and the latter bounded by causal law, but that the laws to which my free action conforms have been proposed by me myself. I am capable of escaping the tyranny of my inclinations, and animal instincts, because my reason itself proposes to me laws to which I may subject myself: I am the author of the laws of my own action. As free, I am still subject to law, but / subject myself to the law which my own reason commends to me. It is, finally, worth mentioning explicitly that it is part of Kant's conception that although I can know, can indeed know a priori, that I am free, I can never know whether I have, on a particular occasion, acted freely, acted not merely in conformity with the moral law but acted in that way because doing so was acting in conformity with the moral law. This is not merely because of the treachery of the human heart, the endless possibilities of self-deception, but precisely because the freedom at stake is transcendental freedom, concerning the 'intelligible character' of our action, which, as such, is not an object of empirical observation:
'Ought' implies 'Can': Kant and Luther
13
The real morality of our actions, their merit or guilt, even that of our own conduct, thus remains entirely hidden from us. Our imputations can refer only to the empirical character. How much of this character is ascribable to the pure effect of freedom, how much to mere nature, that is, to faults of temperament for which there is no responsibility, or to its happy constitution (merito fortunae), can never be determined; and upon it therefore no perfectly just judgments can be passed (first Critique, A 551). Thus I can know that I am a free agent, I can know what constitutes acting freely, but never know whether I am, in fact, acting freely (or, indeed have ever done so). C. In the light of all I have said so far, it is clear that the crucial step in Kant's argument for freedom is the principle ''Ought' implies 'Can\ Despite an initial, intuitive plausibility which this principle undoubtedly has for us, it turns out to be remarkably difficult to see what in the end it amounts to, or why Kant holds it true: (1) It does not seem to be self-evident, and does indeed fly in the face of a great deal that we also wish to say in concrete moral debate: it, in part, gains its initial plausibility from leading us to think of certain aspects of our moral thinking at the expense of others. (2) Despite this, Kant never gives a detailed defence of the principle. He does, indeed, make several scattered remarks that lend support to it, but these invariably turn out to be suggestive rather than a properly worked out defence. Kant at most can be said to sketch out lines along which a defence of the principle is to be sought. (3) The different things that are said in support of the
14
Kant and His Influence
principle by no means amount to the same thing. I will spend the next section of this paper assembling the different defences which may be found at different points in Kant's ethical writings. The main point to stress here is the extreme variety of these defences. What is more, these different defences seem frequently to lead to very different ways of regarding Kant's claim, to very different interpretations of it. The main difficulty here stems from the bewildering variety of ways we may use the key words 'ought' and 'can'. In particular the language of human ability is extremely ill-explored and ill-understood territory. What I will have in mind when I say that someone can or cannot do something, will vary enormously from context to context. At one extreme, we have cases in which we say 'A can \ meaning thereby that the description 'A is -ing^ is logically possible, or that A is the kind of being which it makes sense to describe as -ing, and, at the other, cases in which I say I cannot do something, meaning thereby that I am inhibited in such a way that I find it emotionally impossible to bring myself to do it (shyness, fear of authority, etc.), with a huge range of very different cases between these extremes. It should also be remarked that what may be described as the logic of ability is opaque and not well understood: ordinarily self-evident principles of modal logic turn out to be treacherous here. Thus, it is true to say of me that I cannot throw a dart at a dartboard and hit the bull's eye: I lack the relevant skill. Nevertheless, I may throw the dart and by chance, or luck, it lands in the bull's eye. So that in many uses of 'I can <j>\ it does not follow from the fact that I have -ed, that I could have 4>-ed. Luther is characteristically more careful than Kant in attempting to bring out how his denial of 'Ought' implies
'Ought' implies 'Can': Kant and Luther
15
'Can' is to be understood. It is worth stressing before going any further, something which Luther stresses against Erasmus: even if he wishes to insist on a human inability to obey the Law of God, this must not be taken as though God were commanding things which it would be absurd to imagine a man actually doing: God's Law is addressed to men, and not to stones or geese, and what is commanded is to that extent a genuine human possibility: the idea of someone actually obeying God's Law is a perfectly intelligible idea. It is not as though Man is being commanded to fly or carry a ten-ton weight. Perhaps the best way to bring out the dominant use of 'can' in this controversy is to use a gloss that Kant sometimes adopts - 'is a possible subject of a command': so that I may, or may not, like someone else, but if there is someone I dislike, a command for me to like them would for Kant be pointless and void of significance, Prima facie at least, if I don't like them there may well be nothing I can do about it: certainly, even if there are steps I may take of an indirect nature - trying to view them in a favourable light, for instance - 1 cannot directly obey the command 'Like X\ Liking X is not the kind of thing that I can do, simply at will; and, to that extent it is something which 'cannot be commanded'. This explanation by no means fits everything that Kant says in this connection, but it may serve as an introduction to an understanding of what he is trying to say. There is, however, one interpretative principle that we may clearly adopt in making sense of the apparent uncontrolled variety of Kant's detailed remarks here: any interpretation of what Kant means by 'Ought' implies 'Can' must bear in mind the conclusion that is drawn from this principle: that is to say, it must clearly yield a sense of 'can' that is strong enough to support the conclusion that man is transcendentally free.
16
Kant and His Influence
II How, then, did Kant understand the principle, and how did he seek to justify it? The trouble in interpretation here is twofold: the great variety in the remarks Kant makes in this connection, and the brevity of those remarks. Rather than force them into a single uniform account, which is, I suspect, in any case impossible, I shall examine in turn the different defences which are to be found scattered throughout Kant's writings. But first I shall look at a defence of the principle which cannot in any direct way be Kant's own, but that was crucial in the debate between Luther and Erasmus. Although there are clear reasons why Erasmus' objection to Luther cannot be Kant's thought, it is, I believe, in the end a secularized version of what Erasmus is urging against Luther that may well form the real basis for Kant's insistence that 'Ought' implies 'Can'. If the particular remarks Kant offers in support of the principle turn out to be inconclusive, it may be that the real reason for his passionate insistence on it is that his belief is sustained by ideas derived from Erasmus' argument, even if the context of Kant's thought is one in which this argument strictly has no place. A. The debate between Luther and Erasmus was of course not a debate in moral philosophy, but a theological debate. When the question, whether it was possible for man to obey the law, was raised, the law at stake was the law of God. Because of this, the principle 'Ought' implies 'Can' has a particular purchase that it cannot have within Kant. Luther's opponents, both the scholastic authors like Gabriel Biel, whom he was reacting against, and writers like Erasmus, who were reacting against him, were concerned with the picture of God that
'Ought' implies 'Can': Kant and Luther
17
seemed to emerge if you denied that 'Ought' implies 'Can'. It is in their writings that the Kantian principle may well emerge explicitly for the first time. The issue here is simple: if you deny 'Ought' implies 'Can', can God be just? A Being who enacted laws for his subjects which it was impossible for them to obey, and then punished them for disobedience, would be a monster. The world would then be like a dictatorship in the hands of a capricious tyrant who deliberately frames the laws of the land in such a way that everyone is guilty of breaking the law, so that it is impossible to be a law-abiding citizen. He could then punish whom he chose, and reward whom he chose, purely according to his whim. Such a land would have the form of law, but not the reality: if a master were to free a slave who had merited nothing, he might have reason perhaps to say to the other servants who murmured against him 'You are no worse off if I am kinder to this one; you have your due'. But anyone would deem a master cruel and unjust who flogged his slave to death because his body was too short or his nose too long or because of some other inelegance in his form. Would not the slave rightly call out 'Why am I punished for what I cannot help?' and he would say this with more justice if it were in the master's power to alter the bodily blemish of his slave, as it is in the power of God to change our will, or if the lord had himself given the slave this deformity which had offended, as for example by cutting off his nose or making his face hideous with scars. In this same way God, in the view of some, works even evil in us. Again, as concerns the precepts, if a lord were constantly to order a slave who was bound by the feet in a treadmill, 'Go there, do that, run, come back', with frightful threats if he disobeyed
18
Kant and His Influence
and did not meanwhile release him, and even made ready the lash if he disobeyed, would not the slave rightly call the master either mad or cruel who beat a man to death for not doing what he was unable to do? and: Furthermore, when God burdens man with so many commandments that serve for no other purpose than to make him hate God more and be more terribly damned, do not they make him worse than the tyrant Dionysius of Sicily, who deliberately made many laws that he suspected the majority would not keep in the absence of restraint, and at first took no notice, and then when he saw that everybody was breaking them, he began to summons them to punishment, in this way bringing everyone into his power (Erasmus, Diatribe, Epilogue). The basis for the principle 'Ought' implies 'Can' is easy to understand here: it is a matter of theodicy. If God is conceived as both the lawgiver, and the one who punishes if the law is disobeyed, then he becomes an insane and capricious monster, if it is thought that the laws he enacts are ones which it is impossible for Man to obey. But, at least in any straightforward way, this simple train of thought cannot be Kant's: for Kant, God is not the author of the moral law, nor are we to obey it out of fear of punishment by God. We ourselves are the authors of the moral law; it is our own reason that commands us to obey it. The framework within which Erasmus' arguments have immediate purchase would be, for Kant, a purely heteronomic framework, and as such the antithesis of true morality. Without a divine Lawmaker and Judge, no issue of the
'Ought' implies 'Can': Kant and Luther
19
justice of God can arise, so that the train of thought we find here can at most be considered as a remote ancestor of the ideas that we find in Kant. Yet, this train of thought seems to me to be far from irrelevant to an understanding of Kant. At some level, ways of thinking about morality that only made immediate sense within the context within which Erasmus and Luther were disputing, continue to inform Kant's whole conception of morality. And, at the end of this paper, I shall return to this first defence of 'Ought' implies 'Can', which at least has the merit of being a clear defence. B. In the first Critique, when Kant first adumbrates the principle 'Ought' implies 'Can', he writes: 'Ought' expresses a kind of necessity and of connection with grounds which is found nowhere else in the whole of nature. The understanding can know in nature only what is, what has been, or what will be. We cannot say that anything in nature ought to be other than what in all these time relations it actually is. When we have the course of nature alone in view 'ought' has no meaning whatsoever. It is as absurd to ask what ought to happen in the natural world as to ask what properties a circle ought to have. All that we are justified in asking is: what happens in nature? What are the properties of a circle? (first Critique, A 547). If we regard this preliminary introduction of the principle 'Ought' implies 'Can' as also an attempted justification of it, what it amounts to is obscure. Kant is claiming that where we regard things as necessarily the way they are, it is meaningless to say that they ought to be otherwise. But this is infected with all the ambiguity that I have already noted surrounding the words 'ought' and
20
Kant and His Influence
'necessarily'. On several understandings of the words, this is simply untrue. A doctor may certainly point to a diseased or deformed organ and describe how it ought to be - and so, indeed, with a deformed animal or plant without claiming that there is any real possibility of ever curing or rectifying the deformity. A dog handler may have clear views as to how his dogs ought to behave and train them so to do, without his activity only making sense if he supposes his dogs to be free. Equally, I may have in mind goals that I want to achieve in this paper and, as a result, clear views as to how it ought to be. If I am then guided by these goals in constructing the paper, success is certainly not guaranteed a priori. The fact that I hold that the paper ought to be a certain way - even justly hold it ought to be that way - in no way implies it is necessarily within my power to produce a paper of the required sort, that I can produce such a paper. The most that can be said on the basis of the alleged meaninglessness of asking what the properties of a circle ought to be, is that if we say things ought to be thus and so, we are implying the logical possibility of their being so. But it is a huge step from there to what Kant requires - that there must be a real possibility of bringing the desired goal about. Hence, Kant must be taken to be making a claim that attaches peculiarly to the moral sense of 'ought': but, to that extent, I do not think this preliminary observation can, in its generality, offer any justification for the claim, and, at most, it only serves to introduce it. C. The other support for the principle 'Ought' implies 'Can', to be found in the first Critique rests on a contrast between a categorical imperative and a hypothetical one. This, unlike the first passage, is unequivocally intended
'Ought' implies 'Can': Kant and Luther
21
as a justification of the principle, and one which, moreover, explicitly ties the principle to the moral 'ought': I assume that there really are pure moral laws which determine completely a priori (without regard to empirical motives, that is, to happiness) what is and what is not to be done, that is, which determine the employment of freedom of a rational being in general; and that these laws command in an absolute manner (not merely hypothetically, on the supposition of other empirical ends), and are therefore in every respect necessary (first Critique, A 807). Clearly, the contrast between a categorical imperative and a hypothetical imperative (together with the related notion of a rule of prudence or happiness) is fundamental to Kant's whole presentation of his moral philosophy, and, at a number of points, he will claim that as a consequence of the contrast, whereas it will not always be possible to implement a rule of prudence or to follow the path which leads to happiness, we can always obey the categorical imperative. This claim seems to have two different sources, one of which I will examine later but now I will concentrate on what is perhaps the more natural reading: does the fact that a law is to be regarded as unconditionally binding, of itself imply that it must necessarily be possible to obey it? As it stands, it looks as if Kant is guilty of confusion here, of running together different ways in which a command could be 'absolutely binding'. Suppose I make a promise to you, which I subsequently fail to keep; there are a number of possibilities here, e.g. (1)1 make the promise, because I want your good will,
22
Kant and His Influence
or friendship. Later, under altered circumstances, these, no longer matter to me, so I disregard the promise. (2) I make the promise, but fail to foresee the disastrous consequences of keeping it. When these become clear, I break the promise. (3) I make the promise. Circumstances that I could not have foreseen, and wholly beyond my control, make it impossible to do what I said. We may assume here that, in each case, at the time at which I make the promise, I fully intend to do what I say: it is because of subsequent, unforeseeable changes in circumstances that I do not do what I said. The main point to stress here is that these are three very different cases, and that Kant, by his uses of the notions of absolutely or unconditionally binding, tends to assimilate them, ignoring morally relevant considerations: (1) Here Kant would say I am treating a categorical imperative as if it were hypothetical: this is not quite right, but it is near enough. (I do not do the thing, in order to keep your friendship: I still do it, because I have promised you. It is rather, that in the case envisaged, when your friendship ceased to matter to me, the fact that I have promised ceased to serve as a reason for me. But this is a very fine distinction.) This is the case where Kant is clearly right to maintain that, if I regard the promise as morally binding, I ought to regard breaking the promise as ruled out. (2) I do not wish to dispute this second case with Kant at this stage, beyond commenting on the fact that it is the kind of case where Kant's 'rigorism' continually leads him to maintain quite preposterous positions.
'Ought' implies 'Can': Kant and Luther
23
Suppose I promise to meet you at a certain time 'without fail'. As I am setting out someone has a heart attack, necessitating urgent medical attention. The meeting, although of some importance to you, is scarcely a matter of life or death. I consequently skip the meeting, ensuring instead that the appropriate assistance is given to the victim of the heart attack, 'knowing you will understand'. To do otherwise here strikes me as a form of insanity, part of what led Freud to dismiss Kant's conception of the categorical imperative as not genuine, but taboo, morality. But, for Kant, 'knowing you will understand' is neither here nor there - my obligation is not to you, but to the promise. What is of immediate concern to me is that this is already a very different case from (1). In no sense whatsoever am I treating a categorical imperative as if it were hypothetical - in Kant's sense of a hypothetical imperative. The fact that I behave in the way envisaged in no way shows that if the unforeseen accident had not occurred and I had kept the promise, I would only have kept the promise to secure some desirable end. But, even so, in (2) as in (1), rightly or wrongly, I may be said to have bargained the rigour out of the promise. So that, even if in very different ways, I have treated a promise as not unconditionally binding. (3) I promise to intervene on your behalf; before I have a chance to do so, and in a way that I could not have envisaged, I lose the status that would enable me to intervene. This case is completely different from (1) and from (2) above. Here, in no sense whatsoever, need I cease to regard the promise as unconditionally binding: if the promise is overridden by events, I have no part in the overriding. The only blame that could
24
Kant and His Influence conceivably be attached to me would be one of rashness in making the promise, and, clearly, if I knew my position to be precarious, you might justly censure me for promising, thereby entitling you to rely on my promise. But, it need not be like that. I might have every reason to suppose that keeping the promise will be unproblematic, and yet. . . . If it were then said, you ought only to make promises in circumstances in which you can absolutely guarantee delivery, then given our limited control on the future, that is tantamount to abandoning the practice of promising. Every promise to <j> is to be taken, and will naturally be taken, as a commitment, if at all possible, to . Now, does this make the practice of promising conditional, so that there is no absolute obligation to keep a promise? Not at all: it certainly does not, as Kant sometimes seems to think, convert the imperative to keep the promise into a hypothetical one. The difference between the hypothetical and the categorical imperative is that, in the former case, a condition has to be satisfied before the imperative comes into force. Here, the imperative to keep your promise, if at all possible is and remains unconditionally binding. Kant may simply be guilty here of confusing two different logical forms, represented by the same surface grammatical structure:
If P, 0! might mean, either, as in the hypothetical imperative, that a certain condition had to be satisfied, before any imperative came into force, or, that what was (unconditionally) commanded was, 'If p, 4>V.
'Ought' implies 'Can': Kant and Luther
25
But, and this is the point, it is only if one runs together these three very different cases, (l)-(3), that considerations of the unconditional validity of the moral law lend any support to the principle 'Ought' implies 'Can'. Only if the thought, that we should regard the demands our reason makes of us as unconditionally binding, leads to the idea that our reason may not demand of us something that turns out to be impossible to do, may we infer from the fact that there is a categorical imperative, that it must be possible to carry it out. There is perhaps here the further thought that if I am really unable to execute a demand my reason makes of me, then I cannot be blamed for failure: but that thought is powerless to prove the existence of freedom. It is at this point that the principle 'Ought' implies 'Can' threatens to become utterly empty, as follows: Suppose I were to say: when you are (morally) commanded to 0, the most that can be expected of you is that you do all that is in your power to <j>, and if you do
that you will be morally blameless. Hence, from a moral point of view, the commands, '0!', and, 'Do all that is in your power to >!', come to the same thing: so that not only do you necessarily do precisely the same thing in obeying the command, but you will be appraised as having obeyed or disobeyed the command, not according as you did, or did not, 0, but as you did, or did not, do all in your power to <j>. On this interpretation, moral commands can always be obeyed, it being the merest tautology that it is always in your power to do all in your power to do something. But on this, minimalist, interpretation of 'Ought' implies 'Can\ the principle is powerless to prove anything, let alone the idea that Man is transcendentally free. Kant's position is only to be rescued from complete vacuity at this point by his insistence that there is one
26
Kant and His Influence
thing which is and remains necessarily under my control, namely good and bad willing, and that this must be simultaneously regarded as the true locus of human freedom and of moral appraisal. But that development lies ahead of us (see section E, below), and leads on to different aspects of Kant's thought from any which can be derived from a simple reflection on the difference between the categorical and hypothetical imperative: in itself, that difference is powerless to give Kant what he wants. D. There is a strand in Kant that suggests that the principle 'Ought' implies 'Can' follows from the fact that the moral law consists in commands: that what can be commanded, must be possible. This strand has been followed up by Professor Hare in Freedom and Reason, Chapter 4. Hare's discussion turns out to be elusive, and often he seems to be saying: 'Ought' implies 'Can', except when it doesn't. It is 'ought' 'with its full force' which implies 'can'; but it is difficult to see what exactly this 'full force' is, other than that it is the force which attaches to it when it implies 'can'. (1) The crucial point here is that Hare is wishing to maintain that imperatives, as such, in some way imply the possibility of carrying them out: so that his considerations will apply as much to advice, or other non-moral imperatives, as to morality. For this reason alone, the considerations to be adduced here cannot be central to Kant's position, where he clearly wishes us to see his principle as saying something that has purchase for the categorical imperative, as opposed to other imperatives. (2) Hare's key point is the relation between ought and what he calls 'Practical questions' ('What am I to do?'):
'Ought' implies 'Can': Kant and Luther
27
unless the practical question arises, the 'ought' question cannot arise, if 'ought' has its full force. . . . And the reason for this is that, when the word is being used in this way, its function is to offer help and guidance in answering this practical question . . . and so, naturally, there is no point in asking the 'ought' question when the practical question does not arise {Freedom and Reason, p. 36). (3) Of course, if the considerations Hare wishes us to take into account derive from reflections on practical questions as such, they are bound to be impotent to establish Kant's principle 'Ought' implies 'Can': for such questions arise with respect to the phenomena of acting and deciding, and reflections upon such phenomena will from the very outset be incapable of settling the question whether Man is transcendentally free. On Kant's conception, the phenomena of acting, including choosing and deciding, will all be compatible with a thorough determinism: indeed, Kant himself accepts such a determinism with respect to them. The most that Hare's arguments could establish would be the empirical phenomena of choosing and deciding, whose existence Kant would not dispute, but which Kant would not see, as such, to be incompatible with a metaphysical determinism: if we could investigate exhaustively all the appearances of men's wills, there would not be a single human action which we could not predict with certainty, and recognise as proceeding necessarily from its antecedent conditions (first Critique, A 550). (4) It should be noticed, quite independently of the particularities of transcendental psychology, that the sense of 'can' that is at stake in the main thrust of Hare's remarks is very weak: I can only order someone to 0, if
28 Kant and His Influence 'the question of their -ing arises'. If we consider practical questions, we typically have two concerns in mind: we consider what needs doing, and what we are able to do. In considering what needs doing, although we will characteristically not consider what it is obviously absurd to suppose that human beings could do, we may well consider much that we subsequently have to work out how to do, and whether it can be done at all. The only things that could be ruled out from such considerations would be things that we already know to be impossible, but many things will be included for which we will have then to seek to find out whether they are possible. Positing ideals, where we then have to see whether or not it is possible to realize them, is an important part of the business of reflecting on practical questions. Maybe 'the question of our realizing those ideals must arise', but that does not imply that they may not, in the event, prove impossible to realize. The language of 'ought' moves between setting ideals of conduct and presenting the practical steps to realize those ideals; frequently, it will, as a result, present ideals prior to knowing whether or not they may be possible to realize. Indeed, sometimes the use of 'ought' to posit an ideal of behaviour may be put uppermost, so that we explicitly posit an impossible ideal; as, for example, when a patient is told by their doctor 'What you really ought to do is avoid all situations that create stress: I realize that will not be possible; so, what you should do instead is . . . .' I may summarize the result of this examination of Hare's position - and those elements in Kant that suggest it by saying that this train of thought, considered as supporting Kant's principle 'Ought' implies 'Can', would be truly feeble. And it would be truly feeble for one
'Ought' implies 'Can': Kant and Luther
29
simple reason: the most they could possibly establish is that it is inappropriate to say 'You ought to >' if it is known to be impossible that you should 0. So that if it is not known whether or not it is possible to >, we certainly cannot infer from the fact that it is correct to say that we ought to 0, that we can <j>. Although the positions I have considered in the last three sections all appear to be operative at various points in the Kantian texts, it is the remaining sections which give interpretations of the principle 'Ought' implies 'Can' that relate most particularly to Kant's own ethical theory: E. If we consider the following passage from the second Critique, we arrive at another interpretation of 'Ought' implies 'Can', different in kind from any I have looked at so far: It is always in everyone's power to satisfy the commands of the categorical command of morality; this is but seldom possible with respect to the empirically conditioned precept of happiness, and it is far from being possible, even in respect to a single purpose, for everyone. The reason is that in the former it is only a question of the maxim, which must be genuine and pure, but in the latter it is also a question of capacity and physical ability to realize a desired object (second Critique, p. 148). Here, the categorical and the hypothetical imperative are sharply contrasted, in the respect which interests us: for the hypothetical imperative as explained here, 'ought' does not imply 'can'. This is Kant's clearest passage concerning the relation of 'ought' and 'can', but the thought moves in a different direction: now, the reason that 'Ought' implies 'Can'
30
Kant and His Influence
is that the moral 'ought' is seen as restricted in its scope to that which I am seen as necessarily having within my power: issues concerning the maxim that I choose to adopt, and the 'purity' of that maxim 'Do I choose the right maxim?' and 'Do I choose the maxim that I do, because it is the morally acceptable maxim?' But the thought here should not be regarded as a justification of the principle 'Ought' implies 'Can', so much as a consequence of it. The thought is not so much ''because morality only concerns the purity of the will, it is necessarily within the power of everyone to be moral', as 'because being moral is necessarily within the power of everyone, it must be conceived as being restricted in its scope to what must be within their power'. We may perhaps see best what has happened here if we anticipate the next section of the paper by looking at a case I shall reconsider there: one of the first kinds of counter-example to Kant's claims that will occur to anyone is the case in which someone does the wrong thing under extreme duress. Suppose someone is tortured to betray a solemn confidence and reveal the whereabouts of their colleagues, thereby putting their lives in jeopardy. Here is a case in which we may clearly wish to say that, regardless of the severity of the torture and distress of the agent, they ought not to betray their friends. But it seems a monstrous claim to suppose that, quite regardless of the severity of the torture, everyone will necessarily be able to resist it indefinitely. It could well be that, at a certain point, their will is overridden and as an almost physiological reflex they blurt out the required information: that possibility does not seem to be one that could be ruled out a priori. The rejoinder to this example suggested by the passage we are now considering is as follows. Let us suppose someone does indeed blurt out the words. There will still
'Ought' implies 'Can': Kant and Luther
31
be a morally significant difference between two cases: between, on the one hand, the person who at some stage in the proceedings adopts the maxim 'If the pain threatens to become intolerable, do whatever is in your power to stop it', and as a result betrays the whereabouts of the colleagues, whether they adopt this maxim prior to the torture or under the stress it induces; and, on the other hand, the person who never adopts any such maxim, even though, in the end, they too ejaculate the words. Even if the latter case cannot be ruled out a priori^ it is only the former case in which the agent is to be regarded as morally culpable. From a moral point of view, the traitor is not ipso facto the person who betrays the confidence, but the person whose will at the same time consents to the betrayal. Because Kant unfortunately never considers this kind of case, we cannot say for certain that he would have adopted the defence of his position just envisaged, although the passage we are considering means that it is clearly consistent with his thought, and does, indeed, seem the only defence to rescue his position from obvious absurdity. What must be stressed is that here we see the full importance of a transcendental psychology for Kant's ethics: here it emerges clearly that it will depend upon a conception of the will which is exempt from all empirical conditioning; the transcendental will is and remains in Olympian isolation above and unconditioned by the empirical rough and tumble of the situation in which I find myself. So that regardless of the vicissitudes of my life - both my passions and the external constraints to which I find myself subjected - there is a capacity for choice that remains wholly under my control. F. The third chapter of the second Critique can yield another, totally different, reading of 'Ought' implies
32
Kant and His Influence
'Can'. Here it could be read as saying 'Ought' enables. One of the questions that dominates the early parts of the second Critique is the question, 'What makes moral reasons, reasons for me? Kant's empirical determinism, as far as human beings are concerned, is not simply the general argument for strict universal causation of the second Analogy, but a reflection on the notion of human action as such; if an action is to count as an action of mine, I will have made a decision, and have a reason for my decision - in Kant's terminology, I will be acting upon a maxim. This does not imply that I necessarily do the reasonable thing, but it does imply I can give some form of answer to the question, 'Why did you do that?' A purely gratuitous action would not be an action at all. If that is so, there will be some things that serve for me as reasons for action, and other things that do not. It will be a matter of quite contingent empirical fact that I have certain desires, drives and wants which lead certain proposed courses of action to be open to me and not others. We may call the sum total of such drives, 'my nature'. Kant rather naively develops his account of such a nature in terms of a narrow psychological egoism: All inclinations taken together (which can be brought into a fairly tolerable system, whereupon their satisfaction is called happiness) constitute self-regard {solipsismus). This consists either of self-love, which is a predominant benevolence toward oneself (philautia) or of self-satisfaction (arrogantia). The former is called, more particularly, selfishness; the latter, selfconceit (second Critique, p. 181). Given the existence of such a system, it will be intelligible that certain reasons for action are reasons for me: quite simply, to show why I do something will be to show why it appears to me to be in my interest that it
'Ought' implies 'Can': Kant and Luther
33
should be done. One should not be deceived by the naivety of the psychology here. Kant may well have been confused at this point - together with a large number of his contemporaries and predecessors - into thinking that psychological egoism was some kind of necessary theory of human motivation; but this does not affect the picture which ultimately concerns him here. We could substitute for the psychological egoism of this quotation, some other theory of human motivation which was not necessarily self-regarding - such as, for instance, a Freudian theory of the Instincts where the instincts are ultimately to be explained in biological terms, in terms of the value of precisely these instincts to the survival of the human species. Either way, I will have an empirical system that makes it intelligible that certain reasons are reasons for me: I will have a theory of the springs of my action. There need not be anything particularly 'rational' about the way I behave, and certainly nothing morally to be commended in my behaving in this way - it is just the way I am. But, alongside this system - the sum total of my empirical drives - my reason will commend to me a way of life as the way things ought to be, how I ought to behave, where there will be a discontinuity between the ways I am naturally inclined to act and the way my reason advocates. What my reason tells me I ought to do may, or may not, coincide with what I am, in any case, inclined to do. How then do I have any incentive to do what my reason tells me ought to be done? Kant is fully aware of the difficulty he confronts here. He has to show the existence of a moral motive and yet make out that it is not simply one motive alongside other motives; and, above all, not simply find a place for the moral motive within the empirical system of my motives. The moment I make the moral motive simply one among
34
Kant and His Influence
the diverse motives I, as a matter of fact, have, then it ceases to have for him any moral value. There is nothing particularly commendable about having moral reasons for action if it is just as much a brute fact about me that these reasons serve as reasons for me as it is that I have other natural dispositions. Kant's attempted answer may be put in the form: it is the 'ought' which makes it possible. How successful this answer is I leave to the reader to judge, but it is at this point we may find the Achilles' Heel of Kant's ethical theory. He wants the moral motive both to be an inclination and not an inclination. It must be an inclination, since it must be that it should motivate, but not an inclination in such a way that someone might protest 'There's nothing commendable in your acting that way, you are merely acting in accordance with your inclinations: my inclinations lead me in- the direction of my self-interest, yours in the direction of duty. It's just a difference of temperament.' So we seem close to having to talk of an 'inclining noninclination'. The difficulty here is not with the idea that human beings are filled with respect for the moral law, or that there are objective features of the law which engender that respect, but that the respect thus engendered is different in kind from the other feelings that may motivate me. Kant will reply here: 'Although there obviously has to be a psychological story to tell here, and we must be concerned with genuine human incentives, there is a big difference here in that it is my reason that commends this course of action to me, tells me that it is the right path to follow, that I ought to do it, quite independently of whether I otherwise want to do or it is in my interest to do it.' My reason does not commend my duty to me by making the way of life that it presents to me as my duty attractive to me, but by telling me: That is how things ought to be.
'Ought' implies 'Can': Kant and Luther
35
Perhaps Kant wishes us to see an analogy here with having reasons for believing: if I believe something, then a psychological investigation can seek to find out why I do so. Such an investigation will include an investigation of my temperament and include a description of emotional factors that incline me to believe it. But if my reason tells me it is so - if I have, for instance, followed a mathematical proof that it is so - then a correct explanation of why I believe it need go no further than explaining why it is objectively right for me to do so. So here, if my reason tells me that I am to do something, I escape the subjective realm of my empirical nature into a realm where it is objectively the right thing to do. Just as in the one case, I believe something because my reason shows me that I ought to believe it, so in the other I do something because my reason shows me that I ought to do it. But this appears no more than a somewhat remote, and not wholly convincing, analogy. My reason is seen by Kant here as functioning in a twofold way: it humiliates me, and it fills me with respect for the ideal which it presents to me. Sensuous feeling, which is the basis of all our inclinations, is the condition of the particular feeling we call respect, but the cause that determines this feeling lies in the pure practical reason; because of its origin, therefore, this reason cannot be said to be pathologically effected; rather, it is practically effected. Since the idea of the moral law deprives self-love of its influence and self-conceit of its delusion, it lessens the obstacle to pure practical reason and produces the idea of the superiority of its objective law to the impulses of sensibility; it increases the weight of the moral law by removing, in the judgment of reason, the counterweight to the moral law which bears on a will
36
Kant and His Influence
affected by the sensibility. Thus respect for the law is not the incentive to morality; it is morality itself, regarded subjectively as an incentive, inasmuch as pure practical reason, by rejecting all the rival claims of self-love, gives authority and absolute sovereignty to the law (second Critique, p. 183f). It is this aspect of Kant's thought that may be summarized by the slogan 'Ought' enables: to the question, 'How is it possible for me to be moral?', that is to say, 'What makes a moral reason for action, a reason for me?', Kant wishes to say that it is because my reason not only presents me with an ideal way of life, it confronts me with that ideal as the one I ought to adopt: humiliating me for not doing so, and filling me with respect for the ideal it presents. Hence, it is because my reason itself shows a course of action as the one I ought to adopt, that that course of action becomes a live option for me - even if that course of action contradicts my natural impulses: it is because I can see that I ought to do it, that I can do it. G. I have left to last what I suspect may have been for Kant the most basic reason for his principle, and the one that also explains the emotional pathos surrounding it: and perhaps, also, his frequent claim that to deny 'Ought' implies 'Can' is to threaten to expose morality as an illusion. This is the quite fundamental significance which the notions of merit and the meritorious have in Kant's whole ethical outlook. As everyone knows, Kant begins the Foundations by claiming that the only thing that is good without qualification is the good will, but his initial explanations of this claim are unconvincing, even self-refuting. Kant begins by saying that if you take other humanly desirable
'Ought' implies 'Can': Kant and Luther
37
traits, such as intelligence or resoluteness, if they are dissociated from a good will, they can become harmful in their consequences. But (a) this is irrelevant if the claim is that the good will is good in itself and not to be commended for something else, such as its good consequences; and (b) if it is true that intelligence only serves to make the villain more dangerous, it can be equally true that a good will detached from other qualities can be destructive in its effects. A nincompoop with a good will may behave like a bull in a china shop. If we are concerned with truly creative human action, 'being wellmeaning' may be a necessary condition, but it is certainly far from a sufficient condition. Without other qualities such as empathy, an awareness of your own limitations and a capacity for an imaginative understanding of the situation that confronts you, the person who has 'a good will', but no more, may well be a menace. But Kant's real thought is not captured by his opening remarks; he is not really thinking of the consequence of having certain traits at all. What he is thinking is more closely captured by what he goes on to say: Moderation in emotions and passions, self-control, and calm deliberation not only are good in many respects, but even seem to constitute a part of the inner worth of the person. But however unconditionally they were esteemed by the ancients, they are far from being good without qualification. For, without the principles of a good will, they can become extremely bad, and the coolness of a villain makes him not only far more dangerous but also more directly abominable in our eyes than he would have seemed without it (Foundations, p. 56). Now Kant is saying: 'Forget about the consequences: imagine someone censuring someone else's wickedness. If
38
Kant and His Influence
they receive the reply "But, don't you admire their ingenuity, their resourcefulness, daring . . . " then the appropriate thing to say is, "That only makes it worse!" - even if, in some curious quasi-aesthetic way, we may continue to "admire the ingenuity"'. Given any human trait, with the sole exception of the good will, it can be seen that in the wrong context the presence of that trait will detract from, rather than enhance, our estimate of the person. Is this true, and why does Kant insist it must be true? Pritna facie, Kant makes things easy for himself by his choice of examples. Suppose that instead of choosing a skill or an ability such as ingenuity, we had considered a disposition such as being naturally compassionate would his remarks hold? Consider a Falstaff - much that he does is monstrous, and it is undoubtedly right to censure much of what he does. But if we, nevertheless, are made by Shakespeare not simply to condemn him but even to feel affectionately disposed towards him, it is by virtue of his other personality traits - his spontaneous warmth, his generosity towards friends, his capacity for self-mockery. Here, we do not simply say 'That only makes it worse!', we regard these as mitigating, even 'redeeming', traits; despite the fact that these qualities have no obvious connection with the Kantian good will, with reverence for the moral law. Can we actually imagine circumstances in which someone, having censured someone else's villainy, could justly say - when we had retorted 'But still, they are capable of great spontaneous and natural compassion' - 'That only makes it worse!' The most that would be appropriate to say would be 'That still doesn't excuse it!' Kant seems to be wanting to convict us of a kind of incoherence if we insist on regarding a trait such as being
'Ought' implies 'Can': Kant and Luther
39
naturally and spontaneously compassionate as of any ultimate value, as 'good without qualification'. What he has said so far has not brought out why. To see why he held this attitude we have to consider the distinction he draws between admiration {Bewunderung) and respect {Achtung). We may clearly be moved to admire (or have affection, love or even fear of) a wide variety of different human beings, for a wide variety of different reasons, including some quite trivial ones. We may admire someone's intelligence, fortitude, strength and even their good looks or special talents - and they may feel proud of these traits. We may even, as Kant remarks, admire someone for their social status, including factors such as their aristocratic blood over which they have no control whatsoever. The existence of such admiration is not disputed by Kant, but he sees it as not different in kind from the feelings we may have for certain purely natural phenomena such as the strength, agility or beauty of an animal, or even an inanimate phenomenon such as a lofty mountain. 'All of this however is not respect' (second Critique, p. 185). Despite the ludicrousness of some of these phenomena (admiration for rank), Kant is not wishing to say that these feelings are simply inappropriate but that they are different in kind from the respect that is evoked in us by morally good action. Kant says that in the latter case the respect is really respect for the moral law itself, as exemplified in the action: I am not sure that I fully understand his thought here, but this does not seem quite right. The point is that a crucial element in respect is that I myself feel humbled by the action, and that is not simply being humbled by the difference between me and another human being but by the consciousness this prompts of the disparity between my own behaviour and that demanded by the
40
Kant and His Influence
moral law: so that I am not really humbled by them, but really by the moral law itself. However, it still seems important to say here that I do respect the person as well. But, even if I did feel humbled by the prospect of a person with talents far greater than my own, that would be an inappropriate and irrational response. I ought, rather, to say simply that it takes all sorts to make a world. To the extent that the admiration is directed towards what must be regarded as no more than a brute fact about its object, it would be absurd, even if the admiration is in place, to look on what is admired as though it were meritorious. To take an extreme case, we talk of 'venerable age', and it may well be appropriate to show deference to those whose age has given them a wider and deeper experience of life than our own. But it would be a kind of insanity to reproach myself for not being as old as they are. It is only to the extent that I can represent to myself the difference between myself and someone else as other than a brute fact, that 'reproaching myself for the difference would make sense; but that means representing the difference as an achievement, something for which they are responsible, something under their control. Certainly great talents and activity proportionate to them can occasion respect or an analogous feeling, and it is proper to accord it to them; then it seems that admiration is the same as this feeling. But if one looks more closely it is noticed that it is always uncertain how great a part of the ability must be ascribed to innate talent and how much to cultivation through one's own diligence. Presumably reason represents it to us as a fruit of cultivation, and therefore as merit which perceptibly diminishes our self-conceit and
'Ought' implies 'Can': Kant and Luther
41
therefore either reproaches us or else imposes it upon us as an example to be followed in a suitable manner. This respect which we have for a person (really for the law, which his example holds before us) is, therefore, not mere admiration (second Critique, p. 185). Kant draws from this train of thought the startling conclusion that the sum total of what may be called natural philanthropic feelings are without moral value. It is a very beautiful thing to do good to men because of love and a sympathetic good will, or to do justice because of a love of order. But this is not the genuine moral maxim of our conduct . . . (second Critique, p. 189). At this point, Kant develops the second element in his thought which he presents by the striking metaphor that we do not confront morality as 'volunteers'. If I am confronted by, let us say, a great piece of philosophical writing that fills me with admiration, it is inappropriate for me to reproach myself if I lack the natural abilities necessary to emulate the achievement. I may also refuse to reproach myself simply on the grounds that I have chosen not to dedicate myself to philosophy - and I am none the worse for that. If I do reproach myself, it is for failure to achieve a goal that I have chosen to set myself, whereas moral self-reproach is for failure to achieve a goal that my reason presents to me as one that I ought necessarily to aim for. The picture to emerge from this is of morality as a task that is necessarily set to every rational agent, and one where it is appropriate to reproach oneself for failure. To the extent that I am bound to represent a feature of myself, or my behaviour, as a brute fact about myself that is not under my ultimate control, the notions of
42
Kant and His Influence
ultimate praise and blame, and of merit lose coherence. Even if a dog-handler does train dogs using in part not only rewards and punishments but also praise and blame, there is a kind of absurdity in taking this praise and blame with acny ultimacy: its force is exhausted by the extent to which it can profitably change the dog's behaviour. But to represent a feature of myself as not simply a brute fact about myself, as under my ultimate control, is to represent myself as free. This seems to me the real core of Kant's insistence that 'Ought' implies 'Can'. If the denial of this principle is said to threaten to expose morality as an illusion, it is not so much because it undermines the possibility of serious reflection on what is the good life, which is the life worth living; what it threatens to undermine is the ultimate significance of the practices of praise and blame. It is these notions which had fundamental importance for Kant's whole ethical outlook and which were put in jeopardy if 'ought' did not imply 'can'. The whole pathos of Kant's position is captured by the following passage: if nature has put little sympathy in the heart of a man and if he, though an honest man, is by temperament cold and indifferent to the sufferings of others perhaps because he is provided with special gifts of patience and fortitude, and expects or even requires that others should have the same - and such a man would certainly not be the meanest product of nature would not he find in himself a source from which to give himself a far higher worth than he could have got by having a good-natured temperament? This is unquestionably true even though nature did not make him philanthropic, for it is just here that the worth of
'Ought' implies 'Can': Kant and Luther
43
the character is brought out, which is morally and incomparably the highest of all: he is beneficent not from inclination but from duty [Foundations, p. 60). Here we have a picture of a man of high moral integrity, even if not a particularly likeable man, and Kant is wishing to commend precisely such a man as this as worthy of the highest possible respect, quite on a par with the respect we, perhaps more automatically and ungrudgingly, give to someone of great personal warmth and compassionate self-giving to others (and, moreover, by whom most of us would prefer to be helped). The thought here is the same as Christ's on 'the widow's mite': Of a truth I say unto you, that this poor widow hath cast more in than they all: For all these have of their abundance cast in unto the offerings of God: but she of her penury hath cast in all the living that she had (Luke 22.3-4). The moral 'ought' demands 'all we have', and no more. But that it does demand, and if it is not vacuous, there must be something that we do have: we must be free.
Ill Periculosa est haec oratio: lex praecepit, quod actus praecepti fiat in gratia dei. Contra card, et Gab. (DST, 57, Weimar Ausgabe, I, p. 227) If 'Ought' implies 'Can' could be taken as at the centre of Kant's whole moral vision, this antithetical thesis could equally be taken as close to the centre of the moral vision
44
Kant and His Influence
of Martin Luther. The terseness of the Latin creates difficulties for translation, and, indeed, the two standard translations make Luther say something very different. In The American Edition, Vol. 31, H.J. Grimm translates this: It is dangerous to say that the law teaches that its performance takes place in the grace of God. This in opposition to the Cardinal and Gabriel; whereas in The Library of Christian Classics XVI, James Atkinson renders it: It is dangerous to believe that the existence of a law implies that it can be obeyed, for the law is fulfilled by the grace of God (against Peter d'Ailly and Gabriel Biel). Of these two, undoubtedly the somewhat free rendering by Atkinson is the one that comes close to Luther's meaning. Grimm's, indeed, is close to gobbledegook, and whatever it is meant to mean does not fit the sequence of Luther's thoughts in this Disputation. I would render it, again somewhat freely: It is misleading to say that the Law is prescriptive (action-guiding) [and hence, with Hare, that 'Ought' implies 'Can'], since the action it prescribes only occurs by the grace of God . . . . Despite the fact that the opposition between the two visions is immediate, it is difficult to organize an appropriate confrontation. For, not only is Luther writing exclusively as a theologian and Kant as a philosopher, but that difference is to be taken strictly. Whereas Luther presents his thought to be judged only by its fidelity to the scriptural witness to divine revelation, Kant equally wishes his thought to be judged exclusively at the bar of
'Ought' implies 'Can': Kant and Luther
45
pure reason. What I propose then, in this - which is intended in the first instance as a philosophical essay - is to set Kant's position in relief by asking whether his concentration on the principle 'Ought' implies 'Can' does not by implication carry with it a severe truncation or distortion of much that we also wish to say in ethics; illustrating this, as appropriate, by the elements in Luther's thought that, by doing justice to those elements within a Christian framework, constantly led him to the apparently paradoxical denial of the principle that Kant treats as axiomatic. Although my sympathies here undoubtedly lie in the end with Luther, and I shall make no attempt to disguise them, my purpose is not the absurdly ambitious one of adjudicating the confrontation. It is to see more clearly the precise nature of the two fundamentally opposed visions of the good life, and what is involved in choosing one side or other of the confrontation. A. Kant's picture of Man is like a Pre-Raphaelite morality painting - we always stand like an Arthurian knight at a crossroads, with our inclinations as a houri beckoning seductively down one road while our reason is a grim matriarch sternly pointing along the other. This picture, as we have seen, is not based on any empirical study of Man, but is a product of 'transcendental psychology', a piece of a priori reasoning. Essential to this picture is that (a) even if I were always to be seduced down the wrong path, the other one is always open to me, and that (b) the crossroads is permanently manned: my reason always tells me what I ought to do. Luther's objection to this whole picture is that it failed to take sin radically enough. The complexities of Luther's conception of the Fall, and Man as sinful, are such that I can at most sketch what is at stake in what is already
46
Kant and His Influence
a long paper. Kant, indeed, is far from an absurd optimist about human nature and talks of a Radical Evil in Man. However, he writes: It must indeed be presupposed throughout that a seed of goodness still remains in its entire purity, incapable of being extirpated or corrupted; and this seed certainly cannot be self-love which, when taken as the principle of all our maxims, is the very source of evil {Religion Within, p. 41). The crossroads is permanently doubly manned. There is 'Radical Evil' in me, in that there is a perpetual seduction down the wrong path to which I perpetually succumb, but still I remain in a state of permanent internal conflict between reason and inclination. If Luther contests this account, it is not because he has a more 'pessimistic' or 'cynical' assessment of Man than Kant: anyone reading Religion Within the Limits of Reason Alone will be struck by the extent to which, in contrast with the overwhelming majority of his contemporaries, there is a pervasively pessimistic view of the phenomena of the human. Despite the obvious differences, part of the substance of Luther's opposition to 'Scholastic Theology' translates well into the Kantian debate, and the picture that emerges from transcendental psychology. Luther was opposed to any picture of the Fall and Sin in which Man, as fallen Man, was seen in terms of Man having a 'higher' nature (his rationality) and a 'lower' nature (his sensuality), and the Fall was seen to consist in some form of displacement of the higher nature by the lower - with the higher nature remaining intact as 'an image of God in Man'. Sin involved the whole man, his reason and capacity for decision making every bit as much as his animal passions. Man was in rebellion against God, and this involved a disorientation of the whole person so
'Ought' implies 'Can': Kant and Luther
47
that you could not isolate within Man a faculty or part that was and remained exempt from corruption. This is, of course, precisely what Kant does, and indeed insists on a priori grounds must be possible. It is not my concern here to become involved in the theological debate but to ask how credible is the conception of Man on which Kant is insisting here. The point is that to maintain the connection between 'ought' and 'can', Kant is led to assume a power of reason, of decision making, that must remain exempt from all empirical conditioning, intact throughout the emotional vicissitudes of life. In section II of this paper I looked at a possible Kantian defence against the counter-example of someone breaking under torture: the core of the defence was the idea that even if under duress the person snapped, their reason might nonetheless refuse at every point to cooperate with the betrayal. It was always possible, no matter how severe the torture, to refuse to adopt a morally corrupt maxim. It is clear that the defence sketched here would have to apply, mutatis mutandis, to a wide variety of situations in which someone, under emotional or physical stress, does the wrong thing; if we say they ought not to have done what they did, it remained within their control whether they did or did not do it, at least to the extent that they need never decide to adopt the wrong maxim. This at least seems grotesquely implausible; only to be held as a dogma of transcendental psychology in splendid isolation from Man as we actually encounter him. Let us consider Othello here. If anything is clear, it is that a series of moral judgements is in order. He ought not to have killed Desdemona; he ought not to have judged and sentenced her without confronting her with his suspicion and allowing her to speak in her own defence. What is
48
Kant and His Influence
more, these judgements stand with full moral rigour. One will only hesitate over saying this if one insists on hearing it as saying the same as 'he was fully to be blamed for what he did'. But 'being wrought, perplexed in the extreme', however adequate or inadequate to the situation the remainder of what Othello says at the end of the play may be, this can clearly be allowed to stand. He has been subjected to a complex emotional manipulation directed at his most basic insecurities and vulnerabilities, and thereby plunged into a horrible confusion of thought and feeling. To suppose here that Othello retained throughout the process an ability for clear thought about what was right and wrong, and that the decisions he made were deliberate flouting of what his reason was telling him he should do, is clearly absurd. Confronted by an extreme case like this, Kant is surely forced into giving a perverse description of what occurs. In fact, when we look at the actual complexity of the situations within which people make moral decisions, Kant's account seems only preserved from instant falsification by being transposed to a noumenal realm. In the first Critique, he writes: The action to which the 'ought' applies must indeed be possible under natural conditions. These conditions, however, do not play any part in determining the will itself, but only in determining the effect and its consequences in the field of appearance. No matter how many natural grounds or how many sensuous impulses impel me to will, they can never give rise to an 'ought', but only to a willing which, while being very far from necessary, is always conditioned; and the 'ought' pronounced by reason confronts such willing with a limit and an end - nay more, forbids it or authorizes
'Ought' implies 'Can': Kant and Luther
49
it. Whether what is willed be an object of mere sensibility (the pleasant) or of pure reason (the good), reason will not give way to any ground which is empirically given (first Critique, A 548). Kant is saying here that, in human action, there may indeed be a number of constraints, both physical and psychological, which influence my capacity for doing what I set out to do, and also a number of emotional factors that influence the choice I make, but that there is one thing which must remain immune to all external influence, namely my reason, that will continue to confront my action with an account of what it ought to be. It is difficult to make sense of this, save on the assumption that Kant is guilty of a confusion here: a confusion which perhaps stands out more clearly if we transpose his remarks to the sphere of theoretical reason. If I follow a valid piece of reasoning, then this will provide an objective check on by belief, so that whatever I may feel, for whatever reason, 'inclined to believe' must give way to what here confronts me as what must be the case. But to infer from this that my reasoning processes are not subject to empirical conditioning, that I may not simply fail to follow the proof because I am tired, or that I may not become confused, or that my judgement of the strength of the argument I am considering may not be warped by prejudice, self-deception or other emotional factors, would be absurd. There is a sort of equivocation here involved in the phrase 'the "ought" pronounced by reason'. If Othello is persuaded that he ought to kill Desdemona, we might well say that it is not his reason which thus persuades him. We would then mean no more than that this is an irrational conviction. But we are certainly not
50
Kant and His Influence
entitled to say that no process of reasoning has led him to this conviction: it is just a bad one, which has been distorted and corrupted by his emotional confusion. Is the 'ought' of 'I ought to kill Desdemona' 'pronounced by reason' or not? Not by 'pure reason', but by Othello's process of reasoning. When Luther inveighs against reason as 'the Devil's whore', he is not indulging in irrationalism, or the kind of enthusiastic celebration of feeling at the expense of reason before which Kant had such evident horror and of which he was justly suspicious. If by 'reason' you mean embodied reason, the actual thinking processes of people in concrete situations, these are every bit as subject to distortion and corruption as any other aspect of the mental life, and someone can be every bit as much persuaded by 'their reason' in this sense to embark on a wicked project as they can be seduced by their animal passions into doing what they know to be wrong. What both Luther and Kant saw clearly was that to talk in this way had profound repercussions for the principle 'Ought' implies 'Can'. If our reason can persuade us to do great evil, so that it is with 'righteous indignation' that we do it, we are far more radically enmeshed in evil than Kant could ever permit. When Othello kills Desdemona, with pathetic sincerity he is portraying himself to himself as wielding 'the sword of justice'. If he had simply been tempted to do something which he could see all along was wrong, then the Kantian picture, that he could nevertheless have turned aside from his evil project, has purchase. But if what it is right for him to do has here become inaccessible to his mind, then it becomes almost meaningless to insist that, since he ought not to have killed Desdemona, it must have been possible for him not to have done so. For Luther, the Crucifixion would clearly have central
'Ought' implies 'Can': Kant and Luther
51
significance: '. . . the wisdom of God . . . which none of the princes of this world knew; for had they known it, they would not have crucified the Lord of glory' (1 Corinthians 2.8). If, on any understanding, the Crucifixion was an act of great human wickedness, the men who conspired to bring it about would nonetheless have included those who were not just firmly persuaded that they were doing what was morally permissible, but, indeed, were acting out of moral indignation, conceiving themselves to be doing their religious duty. Now, if that is how it can be with Man, then Man is clearly not in a situation from which he can extricate himself by his own efforts, by 'doing what in him lies', but only by the intervention of divine grace. B. Closely connected with the train of thought we have just developed is the question of our knowledge of what we ought and ought not to do. Kant makes the astonishing claim that we always know what we ought to do. This is not just a naive assumption about the powers of human reason, but is presented as a consequence of the claim that 'Ought' implies 'Can'-. But the moral law commands the most unhesitating obedience from everyone; consequently the decision as to what is to be done in accordance with it must not be so difficult that even the commonest and most unpractised understanding without any worldly prudence should go wrong in making it (second Critique, p. 148). This must be wrong. It is perfectly clear that if there are objective answers to questions of right and wrong, these answers are not automatically available to all people, at all times and at all places. This follows quite simply from the fact that there are considerable divergences in the moral practices and beliefs of people.
52
Kant and His Influence
To take an extremely simple case: in many societies, the practice of buying, selling and owning slaves was regarded as morally unproblematic. We would now say, and Kant would unhesitatingly agree, that they had simply missed something of fundamental moral significance. But what are we to say of the slave-owners in such societies? There is no reason to suppose that these would not include some who regarded themselves as decent morally upright people. We certainly do not wish to say anything other than that they ought not to have owned slaves, but, is there any force here to saying that they could have done the morally right thing? It is only if we are in the grip of a dogma that we insist here 'They must have known they were doing wrong. They must have been insincere if they claimed to see nothing wrong with what they were doing.' There are clearly extensive areas of moral disagreement in which, however much the protagonists may insist on seeing their opponents as insincere, there is no reason whatever to dispute the sincerity of the two opposed parties. Thus, Kant is perfectly clear in the Metaphysic of Morals that I cannot have duties to irrational animals, even if I can have duties in respect of them (that is to say, an apparent duty to an animal must always be regarded as an indirect duty to myself). Any duty I have to another always stems from their rationality. Many today would vehemently dispute this, seeing the fact that the animals are sentient as generating a duty towards them that is the same in kind as the duty I have towards human beings. This is not a merely verbal disagreement; there are many situations in which the two views would lead to very different accounts of what was morally permissible or obligatory. Consider, for instance, the following:
'Ought' implies 'Can': Kant and Luther
53
Vivisectionists, who use living animals for their experiments, certainly act cruelly, although their aim is praiseworthy, and they can justify their cruelty, since animals must be regarded as man's instruments {Lectures on Ethics, p. 240). There is no reason whatever to doubt the sincerity of either party in such a disagreement. But if, say, Kant were wrong here, does it have any purchase whatever to say he could do the right thing, when he is firmly persuaded it is no such thing? It is also clear that a disagreement such as this is a radical disagreement: that is it will not be resolved by one side pointing out to the other some minor oversight, or clear inconsistency in their position. In the case of slavery, we might think that it would be possible to convict the slave-owners of an ethical inconsistency - that what they would say on other ethical issues would conflict with what they had to say about the practice of keeping slaves. But, in the case before us, if either side were to adopt the ethical perspective of the other, it would seem to involve a conversion, a change of fundamental moral beliefs, since both sides in the debate have prima facie internally consistent and yet incompatible positions. This may serve to introduce the second major disagreement between Kant and Luther. In The Dispute of the Faculties, Kant takes issue with Spener and Zinzendorf over conversion (op. cit., p. 95ff) and seems to have a surprisingly easy victory. What obviously fills him with horror in their accounts is what he sees as a kind of irrationalism, of 'a mysticism which kills reason' (ibid., p. 107). There is for him only one tolerable answer to the question: 'How is it possible for someone to be awakened again to a new life? The moral predisposition in
54
Kant and His Influence
us'. The aim of his discussion is to reject the 'supernatural' accounts given by Spener, Franck and Zinzendorf. We do not wonder at the fact that we are beings subject to moral laws and destined by our reason to obey them, even if this means sacrificing whatever pleasures may conflict with them. , . . But we do wonder at our ability so to sacrifice our sensuous nature to morality that we can do what we quite readily and clearly conceive we ought to do. This ascendancy of the supersensible man in us over the sensible, such that (when it comes to a conflict between them) the sensible is nothing, though in its eyes it is everything, is an object of the greatest wonder; and our wonder at this moral predisposition in us, inseparable from our humanity, only increases the longer we contemplate this true (not fabricated) ideal. Since the supersensible in us is inconceivable and yet practical, we can well excuse those who are led to call it supernatural - that is, to regard it as the influence of another and higher spirit, something not within our power and not belonging to us as our own {The Dispute, p. 106f). That is to say, the only legitimate answer for Kant to the religious question of conversion is the idea that I have put in the form 'It is the 'Ought' that enables1: Even the Bible seems to have nothing else in view: it seems to refer, not to supernatural experiences and fantastic feelings which should take reason's place in bringing about this revolution, but to the spirit of Christ, which he manifested in teaching and examples so that we might make it our own - or rather, since it is already present in us by our moral predisposition,
'Ought' implies 'Can': Kant and Luther
55
so that we might simply make room for it (ibid., p. 107). What are we to say to this? The main point to stress, which is indeed stressed by Kant, is that both sides in the dispute - both Kant's opponents with their supernatural agency in conversion, and Kant with his insistence that our own reason is the only legitimate agent in any morally tolerable process of renewal - agree on one central point: we know what the good life is: This radical change [according to the pietists], therefore begins with a miracle and ends with what we would ordinarily consider natural, since reason prescribes it: namely morally good conduct (ibid., p. 99). and According to the Moravian view, as man becomes aware of his sinful state he takes the first step toward his improvement quite naturally, by his reason; for as his reason holds before him, in the moral law, the mirror in which he sees his guilt, it leads him, using his moral disposition to the good, to decide that from now on he will make the law his maxim. But his carrying out of this resolution is a miracle (ibid., p. 100). Once the matter is conceived in either of these ways the whole process of conversion is necessarily seen in non-cognitive terms: we learn nothing new. Instead, the only problem that could remain would be one of weakness of will, a will which already knows what it ought to do. In this context, Kant recoils from the prospect of a purely emotional manipulation of the will, leaving room, at most, for the example of Christ to encourage our reason to do what it already knows it
56
Kant and His Influence
ought. Although not strictly relevant to my purposes, there clearly also runs through these pages in The Dispute, 2L considerable scepticism about the actual moral efficacy of such emotional experiences (cf. the delightful footnote on p. 103, where he scarcely disguises his hostility towards 'the sects' whose teaching he is here attacking). What we find here is part of Kant's general attack on the possibility of a revealed theology, which if correct would be far more subversive of traditional Christianity than his more famous refutation of natural theology in the first Critique. The core of the critique of the possibility of revealed theology is to be found in the famous sentence from The Foundations of the Metaphysic of Morals: Even the Holy One of the Gospel must be compared with our ideal of moral perfection before He is recognized as such; even He says of Himself, 'Why do you call Me (Whom you see) good? None is good (the archetype of the good) except God only (Whom you do not see)'. But whence do we have the concept of God as the highest good? Solely from the idea of moral perfection which reason formulates a priori and which it inseparably connects with the concept of a free will. Imitation has no place in moral matters, and examples serve only for encouragement (p. 68). That is to say, let us suppose that we are confronted by an alleged revelation of God. If we are not to sacrifice our moral integrity and rationality, we must test the claim by seeing how far it accords with what our reason tells us is good. If it does not accord, we must reject the alleged revelation as spurious, but if it does accord, then we learn nothing that we could not have already known on the basis of our reason alone. The argument here is,
'Ought' implies 'Can': Kant and Luther
57
of course, not peculiar to divine revelation but to anyone confronting us with a 'new' moral view. What is said here is just as applicable to protagonists of 'animal rights' as to 'the Holy One of the Gospels'. Within the sphere of morals, we are and ought to be locked within the confines of our own reason. The suppositions that arise here are: Suppose we do not know what is right. Suppose we have warped or corrupted views of the good life. These suppositions are ones which, as we have seen, Kant simply rules out of court, but they are the questions with which Luther begins. Through sin comes ignorance of the law {Against Latomus, p. 195); and
35. It is not true that an invincible ignorance excuses one completely (all scholastics notwithstanding). 36. For ignorance of God and oneself and good works is by nature always invincible {DST, p. 11); and
Notice how simple the words are: 'Through the law comes knowledge of sin': yet they alone are powerful enough to confound and overthrow free choice. For if it is true that when left to its own devices it does not know what sin and evil are - as Paul says both here and in Romans 7: 'I should not have known that covetousness is sin if the law had not said "You shall not covet"' - how can it ever know what righteousness and goodness are? And if it does not know what righteousness is, how can it strive toward it? If we are unaware of the sin in which we were born, in which we live, move and have our being, or rather, which
58
Kant and His Influence
lives, moves and reigns in us, how should we be aware of the righteousness that reigns outside of us in heaven? These statements make complete and utter nonsense of that wretched thing, free choice {Bondage, p. 306f). Now the questions raised above have to be posed in more radical terms than they were in the Kant-Spener debate: conversion cannot be construed purely noncognitively: 'Except a man be born again, he cannot see the kingdom of God' (John 3.3). One obvious fact about the Bible which Kant signally omitted in his account of what it has to say in the debate with Spener, is the extent to which it confronts us with natural man as 'blinded'. Such an idea will necessarily be ruled out of court by him, if nothing else because it instantly puts in jeopardy the idea that 'Ought' implies 'Can'. If the idea is once allowed that fallen men might not know what the good life was, that their reason might have become incompetent, that their '. . . foolish heart was darkened. Professing to be wise, they became fools . . .' (Romans 1.21f), the possibility of a radical incapacity to do good emerges. Now we have to reckon with the possibility that we cannot do the right thing, simply because whether through our own fault or not, we no longer know what it is. Now our reason can no longer pull itself up by its own bootstraps. Luther is as aware as Kant of the paradoxical situation that emerges at this point. Confronted by a putative divine revelation all we can and ought to do is, of course, to judge it by our own moral standards: to allow our judgement to be submerged here would be to open the floodgates to the most vicious fanaticism. But if our own moral standards are themselves far from infallible, are indeed corrupt - so that there is the possibility of a gulf
'Ought' implies 'Can': Kant and Luther
59
between the dictates of our reason and the truth - then, prima facie, such subjection of the alleged revelation to our own reason would seem to imply that we were for ever irredeemably locked in error. What we have now to reckon with is the possibility that in the process of judging Christ (necessarily by our own standards), we find that we ourselves, and those standards themselves, are judged: that we find 'our wisdom' convicted of folly by a higher wisdom. However difficult it may be to work this out, it is clear that we are not reckoning with a possibility that we can ourselves engineer. If it occurs at all, it can only be by divine initiative, subject to the rule 'The wind bloweth where it listeth, and thou nearest the sound thereof, and canst not tell where it comes from or whither it goeth: so is everyone that is born of the Spirit' (John 3.8). Here we find a radical limit to the principle 'Ought' implies 'Can'. If we have to reckon with the possibility of an ignorance of the good life on the part of natural man, then for Luther the possibility of leading the good life only comes about, if at all, as a result of a revelation that Man could not himself bring about but which occurred only by the grace of God. Obviously, this paper is not the context to argue through the question of the conceivability of revelation seen in these terms. But what clearly emerges from the confrontation between Kant and Luther is that even leaving the issue of the possibility of divine revelation on one side, Kant's picture cannot be sustained without the, to me, incredible assumption that we are infallible in moral matters, that we always know what is the right thing to be done. Without such an assumption, we must always reckon with the possibility that I am simply unable to do what is right because I have radical misconceptions of what my duty is. But if anything is a moral of a study
60
Kant and His Influence
of the history of mankind, it must surely be that men have done some of their most evil deeds under the firm conviction that they were doing their duty. C. But I have not yet touched on what I believe to be the real core of the disagreement between Luther and Kant, which issues in their equally passionate, but violently opposed, attitude to the principle 'Ought' implies 'Can'. This concerns the central question, 'What is a possible subject of a moral command?' Luther's theology undoubtedly takes its roots in his own experience as a young monk, filled with a zealous concern to dedicate his life to the service of God and who, by all accounts, was punctilious in the observation of the monastic rule. But what confronted him in his most scrupulous endeavours to observe the law as the Law of God, was that the Law seemed to command the impossible. That is to say, that if it were a matter of the observation of the rules of a monastic order, he could observe them to the letter, but the Christian commandments talked in quite other terms. Above all, at their centre there was a twofold command to love. Here, prima facie, was a straightforward impossibility. How can there be a command to love} The terms in which the Bible talked about the good life involves at every turn notions which are by no stretch of the imagination under our control. When he expressed his worries, he was given the popular Aristotelian psychology of the day: that a virtue was a sort of skill, and you acquired a virtue by practising it. Just as you become a good chess player by playing chess, so you become a compassionate man by performing acts of compassion. All his experience convinced him that this was simply bad psychology, that, indeed, all the religious acts of service to God, far from leading towards
'Ought' implies 'Can': Kant and Luther
61
loving God, led him to hate God - precisely for setting him such a fantastical task. His wrestling with the problem that emerged here, and his ultimate resolution of it, was to form the basis of all Reformation theology. I must try here to pick out the salient points which lead into direct conflict with Kant, and, above all for our purposes, two points are crucial: (1) Luther never compromised with the idea that the Law included at its very centre much that simply could not be construed as lying within the power of man to do at will: it included much that, in Kantian terms, 'could not be commanded'. (2) As a result, if there were such a thing as a man being acceptable to God, it could not be as a result of his having merited such acceptance by obedience to the Law. And, equally, if there were actual acts of obedience to God, this could only be by the grace of God, and not the product of Man's efforts. I shall return to this in the final section of this paper. As is well known, Luther and Kant take diametrically opposed positions on (1): Luther: There is also 1 Corinthians 13.2: 'If I have not love, I am nothing.' . . . For it is strictly true that a man is nothing in the sight of God if he is without love. And that is precisely what we teach about free choice {Bondage, p. 286). Kant:
The possibility of such a command as, 'Love God above all and thy neighbour as thyself agrees very well with this. For as a command, it requires respect for a law which orders love and does not leave it to arbitrary choice to make love the principle. But love to God as an inclination
62
Kant and His Influence
(pathological love) is indeed impossible, for He is not an object of the senses. The latter is indeed possible toward men, but it cannot be commanded, for it is not possible for man to love someone merely on command. It is therefore practical love which can be understood in that kernel of all laws. To love God means in this sense to like to do His commandments, and to love one's neighbour means to like to practise all duties toward him. The command which makes this a rule cannot require that we have this disposition, but only that we endeavour after it (second Critique, p. 190). Now, leaving aside the oddity of Kant's actual formulation of what practical love is here (I simply do not know what is meant to be involved in 'endeavouring to like to practise all duties toward' someone), the spirit of his proposal here is clear - in so far as there is a command to love, it must be interpreted in such a way that what is thereby commanded is something I can simply decide to do. The question is, is this reupholstered love, love at all? Part of the trouble is that Kant not only gives an odd account of practical love, but his remarks seem to betray a faulty conception of the 'pathological' love with which it is contrasted. He seems to confound it with the different notion of 'liking someone'. Let us consider here an indisputable case of an act of love: And he said unto them, Come ye yourselves apart into a desert place, and rest a while: for there were many coming and going, and they had no leisure so much as to eat. And they departed into a desert place by ship privately. And the people saw them departing, and many knew him, and ran afoot thither out of all cities, and outwent them, and came together unto
'Ought' implies 'Can': Kant and Luther
63
him. And Jesus, when he came out, saw much people, and was moved with compassion toward them, because they were as sheep not having a shepherd: and he began to teach them many things (Mark 6.31ff). It is clear that we have here a description of a simple act of goodness, a manifestation of love. It is equally clear that this description contains much that would necessarily lie outside the scope of Kantian 'practical love'. The situation here is clear. Jesus has through his ministry to the people reached the point where he needs rest and to be alone; but they are insatiable and refuse to allow him to have that time alone, (i) On any understanding, he would clearly have been justified simply to try to avoid them: whatever he does here is 'beyond the call of duty'. Christ is a 'volunteer', (ii) What is shown here clearly involves the emotions: it is essential to the account that 'he is moved with compassion', (iii) What we have is an act of pure spontaneity, clearly not part of a pre-planned policy of action: indeed, he has gone into the desert precisely to be away from the people, (iv) The act of love is an act directed towards this specific group of people, not the carrying out of some universal maxim of action: he acts 'because they are as sheep not having a shepherd'. It is clear that it is precisely features such as those four which lead us to think of this as a simple act of goodness. The question is: How far can Kant accommodate this into his account of the good life? To answer that we must clearly take into account not merely the earlier formulations of the Foundations and the second Critique, but also the far more nuanced, if somewhat convoluted, discussion of love which we find in the Metaphysic of Morals. Here, at least, we have a more
64
Kant and His Influence
sensible version of what he means by 'practical love' than the frankly silly account in the second Critique: 'The duty of love for one's neighbour can also be expressed as the duty of making others' ends my own (in so far as these ends are only not immoral)' (M., p. 117). In that later discussion, Kant is far from wishing to deny the value of what he calls there 'sympathetic feelings', culminating in saying: Hence we have an indirect duty to cultivate the sympathetic natural (aesthetic) feelings in us and to use them as so many means to participating from moral principles and from the feeling appropriate to those principles. . . . For this is one of the impulses which nature has implanted in us so that we may do what the thought of duty alone would not accomplish (M., p. 126). There are two questions here: Is what Kant says consistent with his basic ethical position? And, secondly: Is what Kant says adequate in itself? He is clearly wanting to acknowledge a value to love, to gratitude, to generosity, to compassion, even in so far as these notions go beyond the notions of 'performing acts of generosity' and the like: 'there is a value in the generous disposition as such. The basic ethical position confronting us at every point in the earlier writings was that the morally good act was not only the act in which I did the right thing, but the act in which I did the right thing 'because it was my duty'. The thought of duty was seen there as the only morally acceptable motive, the only motive that conferred moral value on the act: To be kind where one can is a duty, and there are, jnoreover, many persons so sympathetically constituted
'Ought' implies 'Can': Kant and Luther
65
that without any motive of vanity or selfishness they find an inner satisfaction in spreading joy and rejoice in the contentment of others which they have made possible. But, I say that, however dutiful and amiable it may be that kind of action has no true moral worth {Foundations, p. 59). Here he is vehemently maintaining a position which is at least on the surface in considerable tension with the later discussion. However, a detailed reading of the later passages may, for the most part, be reconciled with the earlier one: most of the time he may be read as saying in the later passages that the feelings are themselves of no true moral worth. It is the cultivation of those feelings as an aid to doing our duty that is morally commendable. However, he culminates the discussion in The Metaphysic of Morals: Would it not be better for the welfare of the world in general if human morality were limited to juridical duties and these were fulfilled with the ultimate conscientiousness, while benevolence were considered morally indifferent? It is not so easy to see what effect this would have on man's happiness. But at least a great moral ornament, love of man, would then be missing from the world. Accordingly benevolence is required for its own sake, in order to present the world in its full perfection as a beautiful moral whole, even if we do not take into account what advantage it brings (in the way of happiness) (M., p. 126f). 'Moral ornament'? 'A beautiful moral whole'? Here he finally uses the word 'moral' in a way that is not only wholly unprepared by the earlier discussion but, if taken seriously, would threaten to explode its most fundamental presuppositions and introduce a use of the word
66
Kant and His Influence
'moral' that was no longer governed by the principle 'Ought implies 'Can'. Perhaps the best way to confront Luther and Kant at this point is to compare the famous passage about Love in 1 Corinthians 13, with the opening paragraphs of the Foundations. Here Paul and Kant are in different ways concerned to single out one element in human life and subordinate in importance every other aspect of human life to it. But Paul, throughout the passage, is clearly describing love in such terms that it does not lie in Man's power whether he loves or not, and yet seeing the performing of those acts which are within one's power as worthless without love ('And though I give all my goods to feed the poor, and though I give my body to be burned, and have not charity, it profiteth me nothing'). Kant, by contrast, is explicitly limiting his celebration to 'the good will' in a way in which that notion is deliberately restricted to what is within Man's power: and it is thus restricted precisely because 'Ought' implies 'Can'. Luther, confronted by such passages as 1 Corinthians 13, refused to compromise in his interpretation of them. Here we are confronted by a picture of the good life, the life acceptable to God, and demanded of Man by God. If that is so, then any adequate basis for Christian ethics would have to begin precisely by repudiating 'Ought' implies 'Can', refusing to restrict his account of what the good life was if it was restricted to the idea of 'doing what in one lies'. But then a fundamental contrast emerges between 'doing the works of the Law' and 'fulfilling the Law' (see, for example, Bondage, p. 302f). One could, and should, 'do all the works of the Law': become a model citizen, obey the categorical imperative . . . and yet even when one had done all that one would not have 'fulfilled
'Ought' implies 'Can': Kant and Luther
67
the Law'. To fulfil the Law one would have not merely to perform acts of generosity but be generous. A text to which Luther recurs is Matthew 7.16ff: Do men gather grapes of thorns, or figs of thistles? Even so every good tree bringeth forth good fruit; but a corrupt tree bringeth forth evil fruit. A good tree cannot bring forth evil fruit, neither can a corrupt tree bring forth good fruit. The problem was not just to do good but to be good: So a man must first be good or wicked before he does a good or wicked work, and his works do not make him good or wicked, but he himself makes his works either good or wicked [Freedom of a Christian, p. 361). and: Why therefore do we yield to an invincible evil lust? Do what is in you and do not lust. But you cannot do that. Therefore you also do not by nature fulfil the law. But if you do not fulfil it, much less will you fulfil the law of love. Likewise, do what is within you and do not become angry with him who offends you. Do what is in you and do not fear danger [Heidelberg Disputation, p. 69). It is clear by now that the difference between Luther and Kant runs deep, but its roots seem to lie in a basic difference of approach. Luther starts by considering two things - the nature of the good life, the life for which God created Man, which made Man acceptable to God, and the nature of sin. Only then, working out from that point, he argues in a number of ways that the nature of sin was such that man had rendered himself incapable of fulfilling the Law, and
68
Kant and His Influence
hence that it could only be by divine grace liberating Man that such a thing as obeying the Law of God could ever occur. Kant, by contrast, always proceeds the other way round: starting by considering what Man can do, he allows that to define the scope of what could count as a legitimate account of the good life. It is because 'Ought' implies 'Can' that 'nothing is good without qualification, save only a good will'. The question is: if we adhere rigidly to Kant's programme, will we not be led, from the very outset, to give a truncated account of the good life? Is there not a great deal that we consider to be of fundamental value in human life, which, if duty is construed in Kant's terms as governed by 'Ought' implies 'Can', simply has to be regarded as lying beyond the scope of duty? We have a simple intuitive contrast between a good man and a moral man, and to my ear at least there is no contradiction in saying 'He's a highly moral man, but scarcely a good man'. Of course, being a moral man matters, but someone may do all that in him lies, and yet not be a good man. But to say this is to attach a great deal of fundamental significance to the good life that simply cannot be regarded as a human achievement.
IV Let us try to draw together the somewhat disparate threads of this paper. When I examined the principle 'Ought' implies 'Can\ I began with an argument from Erasmus and ended with an argument from Kant which I believe to be Kant's real argument for the principle - in any case it seems the only one of the arguments to be found in Kant which does not seem badly flawed when you try to think it through. The others seem at best
'Ought' implies 'Can': Kant and Luther
69
suggestive, and at worst muddled. I described the argument in Erasmus as a distant ancestor of Kant's, and now we have looked at Luther's antithetical position. I wish briefly to comment on what I had in mind. Erasmus' argument rested on the notion of a God who commanded, and who rewarded or punished Man for obedience or disobedience. He then wanted us to contemplate the morally outrageous picture that emerged if the commands were impossible for Man to fulfil: as though life became an insane obstacle race in which the obstacles were deliberately set to trap every competitor. When Luther replies to Erasmus, one feels that despite the fact that both men are arguing well and making good points, the debate really gets nowhere; that in the end they are simply talking past one another. The reason for this is that Luther and Erasmus brought such different presuppositions to the debate, that they simply were unable to hear what each other was saying, and that Luther had in particular questioned a whole range of assumptions which were for Erasmus simply axiomatic to his whole way of thinking. What Luther was breaking completely with was thinking in terms of merit. It is not by chance that the first really public act of the Reformation was his attack on the system of indulgences, which is the sick outworking of a theology conceived in terms of merit. What makes the Erasmus argument seem at first so attractive is that it is presented as though the purpose of God, in issuing his commands, is to set Man a task and reward or punish Man for his success and failure in that task; so that the whole history of the Universe was a history of crime and punishment. If crime then was inevitable, the history became a black comedy. The task had at least to be a possible one, and, however difficult it may be, Man could merit the reward. But this picture is certainly not the only possible one
70
Kant and His Influence
of God and his relation to Man. There is, indeed, something already perverse in viewing life as no more than this kind of obstacle race - whether or not the race is a possible one. One may, for instance, think instead of God's commandments as simply the Law of life: that living according to the Law of love will be the life for which God has created Man, and according to which man will find life and fulfilment. If Man through his own perversity rejects that life, turns his heart in upon itself, he may radically cut himself off from the possibility of the good life so that, thereafter, if freedom, love and even happiness are to be available to him it will only be by an act of divine grace, which is based entirely on a divine, initiative. This is a crude sketch of Luther's picture. But seen in this way, Erasmus' argument has no purchase at all. It is only when we think of the whole purpose of the Law of God to be a complex system of tasks set to us purely for us to be rewarded or punished that it can appear so strong. Now of course, as I stressed, this is not the context of Kant's thought: he is not thinking of God as the author of the moral law, but of the law as dictated by reason alone. But the notions which are implicit in Erasmus' argument are the notions of reward and punishment, and consequently merit. Kant is taking the notions of that which is praiseworthy and that which is blameworthy as the crucial notions in approaching the fundamental questions of ethics. These are, as it were, the secular analogues of the notions with which Erasmus is working, Now, of course, notions of moral praise and blame, of guilt and worthiness, are important moral notions, but they are not the only fundamental moral concepts. It is, I believe, only because Kant insists in giving to these notions absolutely decisive weight from the very outset of his enquiry that he is led to claim that the only thing that
'Ought' implies 'Can': Kant and Luther
71
is good without qualification is a good will; a simple act of true compassion is worth every bit as much, even if not everyone can at will be compassionate. It is only because we at some level all agree with Kant in assigning the fundamental role that he does to the notions of the praiseworthy and the blameworthy, that we find his insistence that 'Ought' implies 'Can' so plausible. An alternative approach to ethics would be to begin by enquiring into the question: 'What is the life which is truly valuable, which is truly worth living?' Only then to raise the question: 'How, if at all, can we attain that life?' We shall then, I suspect, be led to see a large number of the questions of ethics in very different terms from Kant. Questions of praise, blame and guilt will still have their place, but only to be raised after other questions have been raised: and there are whole areas of ethical enquiry in which these questions can become almost irrelevant. Shakespeare, in a play like Othello is engaged in a profound ethical exploration of Othello and his relationships, but it is an enquiry in which the question whether we blame Othello or not is of hardly any interest whatever, even if it remains an exploration of the question: 'How has Othello got human life and human relationships wrong?' The fact that he may not have been able to help doing so, does not in any way mean that he has not gone radically astray. But to begin to explore that question as an ethical question, we have to do what Luther did and dethrone the concept of merit from the centre of our moral thinking.
Bibliography The following translations of Kant and Luther have been used in this paper. Page references are to the editions cited, and abbreviations are as given.
72
Kant and His Influence
Kant Critique of Pure Reason, trans N. Kemp Smith (London, 1929) [first Critique]. Critique of Practical Reason, and other ethical writings, trans L.W. Beck (Chicago, 1949) [second Critique and Foundations]. Metaphysic of Morals, part II, trans M.J. Gregor (New York, 1964) [M.]. Lectures on Ethics, trans L. Infield (New York, 1963). The Dispute of the Faculties, trans M.J. Gregor (New York, 1979) [The Dispute]. Religion Within the Limits of Reason Alone, trans T.M. Greene and H.H. Hudson (New York, 1960) {Religion Within].
Luther and Erasmus For Luther, I have generally cited from The American Edition of Luther's Works, Vols 31-3 (Philadelphia, 1957), referring as appropriate to The Library of Christian Classics, Vol. XVI (London, 1962): Disputation against Scholastic Theology [DST]. For the Luther-Erasmus debate, I have used The Library of Christian Classics, Vol. XVII (London, 1969): Erasmus [Diatribe], Luther [Bondage].
CONFUSED PERCEPTIONS, DARKENED CONCEPTS: SOME FEATURES OF KANT'S LEIBNIZ-CRITIQUE Catherine Wilson
This paper describes a route to Kant's rejection of the metaphysics of the monadology, one which proceeds not by way of the antinomies and the associated claims about the limits of knowledge, but by way of Kant's often repeated claim that Leibniz treated perception and cognition as higher and lower degrees of a single faculty, perception being a mode of confused cognition. Although this criticism, which appears in the Amphiboly section of the first Critique and in many other locations, has certainly received its share of attention, there is a well-founded uneasiness about it. While a few scholars just state their all around endorsement, as though there were obviously a mistake here and it was good of Kant to have seen it, Leibniz scholars have long recognized that it is very difficult to pin this single faculty theory on to the historical Leibniz. This seems to reduce the value of Kant's criticism, for if there is no evidence that it ever related to any real position, its historical significance cannot be very great. Here I would like to try to show that Kant's accusation really is at the centre of his Leibniz-critique. It does, in 73
74
Kant and His Influence
fact, touch its intended object with one arm, as it were, while reaching with the other into the interior of Kant's own scheme of things. On this interpretation, Kant's reassessment begins with a point of what he calls 'logic' - what we would call a point of epistemology - and becomes a point of metaphysics; the attack on Leibniz's epistemological notion of a confused intellectual apprehension becomes an attack on the system of monads. We have here a definite shift of levels: for what begins as an internal, theoretical criticism ends as a metatheoretical criticism of the ability of any philosophical subject to apprehend sensibly the noumenal world. If a subject cannot apprehend the noumenal world cognitively either - there being no such thing as a 'cognitive apprehension' - the end result is that the philosophical subject Leibniz is effectively cut off from his doctrines; he has no right to them. By way of discussing the relation between Kant's internal and external criticisms, I will try to provide a more accurate account of Leibniz's own view; and to show to what extent the Leibnizian doctrine was mediated for Kant by the Leibnizians. I begin, however, with the problem of extracting the confusion doctrine from Leibniz himself.
I 'Leibniz intellectualised appearances', Kant says, just as Locke . . . sensualised all concepts of the understanding. . . . Instead of seeking in understanding and sensibility two sources of representations, which, while quite different, can supply objectively valid judgements of things only in conjunction with each other, each of these two great men holds to one
Confused Perceptions, Darkened Concepts
75
only, viewing it as in an immediate relation to things in themselves. The other faculty is then regarded as serving only to confuse or order the representations which this selected faculty yields.1 This criticism is often taken as bearing wholly on the problem of space, and as reaching out only through the problem of the identity of indiscernibles to more general features of Leibniz's metaphysics. This interpretation is supported by the close proximity of the famous passage talking about comparing and distinguishing objects: Leibniz, Kant says, 'compared all things with each other by means of concepts alone, and naturally found no differences save those only through which the understanding distinguishes its pure concepts from one another'.2 So he is supposed to have arrived as a result at the view that spatial relations and orientations are intellectually known properties of things-in-themselves, confusedly perceived, for 'appearance was, on his view, the representation of the thing in itself. Now, although it is right to see in these passages echoes of the NewtonLeibniz controversy and to read Kant as endorsing a nonrelational theory of space, this does not settle the question of the justice or appropriateness of Kant's accusation; in fact, it is a somewhat strange accusation, for Leibniz assuredly did not hold, could not have held, that righthandedness and left-handedness were properties of 1 Critique of Pure Reason, A 271/B 327, trans N. Kemp Smith (St Martin's: New York, 1965). 2 Ibid., A 270/B 326. This criticism was already in finished form in the 'pre-critical' essay 'On the Form and Principles of the Sensible and Intelligible World' (1770), 'The sensitive should not be described as what is more confusedly cognized and the intellectual as that of which there is a distinct cognition. For these things are only logical distinctions, which do not touch at all the things given', trans and edited by G.B. Kerferd and D.E. Walford, in Selected pre-Critical Writings (Manchester University Press: Manchester, 1968), p. 58.
76
Kant and His Influence
simple substances. (Moreover, Kant seems to have known that Leibniz did not hold this view while he was composing the B version if not the A version of the Critique.)3 What kind of general view of perception as confused cognition is it right, then, to pin on to him? From what basis was Kant actually working? These are the first questions we have to investigate, and by appeal only to works of Leibniz which formed the basis of commentary in the eighteenth century.
1
The 'Authentic' Version of the Confusion Doctrine
Leaving aside, for a moment, the question of whether perception and cognition are two faculties or one, let us note that the claim that normal perceptions are, or may be, thoroughly or somewhat confused enters into Leibniz's philosophy in three different ways. First, it is an internal feature of Leibniz's logic, where it is one of a cluster of terms that qualitatively differentiate 'ideas' or items of 'knowledge'. Second, it occurs as an internal term in Leibniz's descriptive psychology: lower monads or bare monads, resembling people in a swoon or a deep sleep, are said to have more confused perceptions than higher monads. Finally, and here we really do shift levels, confusion is said in some texts to characterize our human perception of sensible things in general; they are really, as intellectual analysis reveals, aggregates of monads, which we perceive confusedly as matter. I will take up each of these broader notions and consider Kant's reaction to it in turn.
1
See p. 100, n. 47.
Confused Perceptions, Darkened Concepts
77
(a) 'Confused cognition' and the mind-body link Leibniz seemed to be motivated by three main sets of problems in his elaboration of the logical-epistemological theory of confused perceptions. The first was the problem posed by the Cartesian account of sensory experience. It is supposed a desideratum in theories of perception that the object of my perception and its cause should coincide. If I am seeing a piece of wax, for instance, my visual experience should be caused by a piece of wax. But the Cartesian theory dispensed with this desideratum in a bold and deliberate way: the cause of my perception of wax - whether we understand here 'real, interactive cause' or 'occasional cause' - is the motion of tiny particles with only the primary qualities of shape and size. My perceived object with its colours, smell, warmth, etc. fails to coincide with it.4 Now, Leibniz has not one but two ways of dealing with this problem. The exoteric solution (of which Leibniz was nevertheless inordinately proud) is the theory of pre-established harmony. On this theory, coincidence is still sacrificed but, by designating the relation between object and cause as 'pre-established', it achieves a kind of legitimacy; and by describing it as 'harmonious', the failure to achieve coincidence is somewhat mitigated. The esoteric solution is the theory of confused perception: object and cause do actually coincide, on this view; the relation is not arbitrary because one is a true mapping of
4
See, for example, the discussion in the 'Principles of Philosophy', Part IV, section 198 in The Philosophical Writings of Descartes, 2 vols, J. Cottingham et al. (eds) (Cambridge University Press: Cambridge, 1985), 1: 284-5.
78
Kant and His Influence
the other, a 'confused' version of it.5 The perception of colour is just the confused perception of colourless moving particles. So Leibniz anticipates the answer to an argument of Berkeley against the corpuscularians which goes something like this: cherries are coloured, so if we can't perceive colours, we can't perceive cherries either. Our text is the 'Meditations on Knowledge, Truth and Ideas'. Here Leibniz says that when we perceive colours or odours 'we are having nothing but a perception of figures and motions, but of figures and motions so complex and minute that our mind in its present state is incapable of observing each distinctly and therefore fails to notice that its perception is compounded'.6 The emergence of a new, apparently 'simple', quality yellow, the scent of violets, the taste of pineapple - is secured to its cause by a striking analogy: unresolved, indistinct, confused figures and motions produce colour, just as unresolved, indistinct, confused blue and yellow particles seem to produce a new simple quality of green. Note that a perfect theory of colour which showed why blue and yellow make green would reduce the arbitrariness to the point where lack of coincidence was no longer a difficulty. This is, by the way, the solution Leibniz indicates to the mind-body problem in general; a perfect theory of mind reduces the arbitrariness of its connection 5 So Leibniz attacks Locke for his resigned acceptance of Cartesian arbitrariness; see the New Essays, trans P. Remnant and J. Bennett (Cambridge University Press: Cambridge, 1981), p. 381f. (In an interesting passage discovered by Parkinson, however, he adopts a more resigned attitude; see G.H.R. Parkinson, 'The Intellectualization of Thought and Appearances' fn. 26 in M. Hooker (ed.), Leibniz: Critical and Interpretive Essays (University of Minnesota Press: Minneapolis, 1982), pp. 3-20.) * 'Meditations', L: 294 ('L' references are to L.E. Loemker, trans and ed. of G.W. Leibniz, Philosophical Papers and Letters, 2nd edn (Reidel: Dordrecht, 1969); G IV: 426 ('G' references are to C.I. Gerhardt (ed.), Die Philosophische Schriften von G. W. Leibniz, 7 vols; reprinted by Olms: Hildesheim, 1962).
Confused Perceptions, Darkened Concepts
79
to the body by considering smaller and smaller units (perhaps individual cells), or more and more refined theoretical constructions (perhaps information networks), to the point where coincidence is achieved. (Sometimes his despair of doing so, or his fear of the materialistic consequences, seem to push him back towards pre-established harmony with its acceptance of 'arbitrariness'.)7 (b)
Degrees of confusion
The second problem concerned the distribution and grades of consciousness. If the soul is essentially res cogitans, as Descartes maintained, what is it when the person is asleep, mad, raving, knocked out, drugged, or, equally important, unborn or dead? Descartes said that the body was the instrument of the soul; when it was diminished, affected, or not all there yet, the soul could not express itself properly even though the soul itself thought as perfectly as ever.8 Add to this somewhat doubtful solution the general dubiousness of the beastmachine hypothesis, and the need for a notion of grades of consciousness is clear. In the 'Monadology' Leibniz divides souls into three main groups: simple monads (regular), souls and spirits. The simple monads have no distinguishable or memorable perceptions; they experience something like our swoons and dreamless sleeps; souls have distinct perceptions because they belong to 7
See, for example, the famous 'mill' pas«?ge at 'Monadology' 17, L: 644; G VI: 609. Solomon Maimon, who saw confused perceptions as 'merely bridges with which to cross from soul to body and back again' registered this tension (though Leibniz had good reason to prohibit this passage). See his letter to Kant of 20 September 1791 in Kant's Philosophical Correspondence 1759-1799, trans and edited by A. Zweig (University of Chicago Press: Chicago, 1967), p. 177. 8 See the 'Reply to Gassendi' in The Philosophical Works of Descartes, 2 vols, trans E.S. Haldane and G.R.T. Ross, 2nd edn (Cambridge University Press: London, 1931), II: 208-9.
80
Kant and His Influence
animals with better developed sensory organs, for example, eyes and ears, which make perceptions more distinct. Spirits are, in addition, self-aware and can grasp necessary truths.9 Observe that our human perceptions are characterized as 'distinct' in this context even though, according to the theory of perception we considered a moment ago, they are also thoroughly confused. The context we are considering is phenomenological and has nothing to do with an analysis of perception. Nor does this notion of distinctness seem to have much to do with consciousness or apperception.10 I can apperceive myself in a swoony state where everything is a blur, or apperceive my hearing when my ears are stuffed with wax and I cannot hear the articulation of words. And animals, which do not apperceive, have distinct sensations in so far as they have well developed sensory organs. Sensory organs concentrate and collect, Leibniz thinks, and that is how they make perceptions distinct. The connection between the physical and the phenomenological is, I believe, afforded by the optical notion of 'resolution': a lens does not just magnify, it resolves a blurry region into two distinct points. The experience of lower beings presumably is indistinct in something of the way that the world is to creatures who do not see different colours, who cannot distinguish speech sounds, or who see only patterns of light and dark, rather than definite contours and edges. (c)
Confused cognition and aesthetic perception
The third subject requiring its own notion of confusion was aesthetics and the problem of the grounding for judgements of beauty. It is commonplace in the history of "Monadology', 19-30; L: 644-6; G VI: 610-12. 10
But cf. R. McRae, Perception, Apperception and Thought (University of
Toronto Press: Toronto 1976), p. 36.
Confused Perceptions, Darkened Concepts 81
aesthetics that the subject shifts in Leibniz's period from a concern with the rules for producing aesthetically correct objects, correctly proportioned and appropriately decorated, to a concern with accounting for the experience of the observer, his idiosyncratic response to what may be a unique work. Alfred Baeumler, in his study of the theory of irrationality in the eighteenth century, has pointed to the significance of Father Bouhours, a writer mentioned by Leibniz, who wrote, in opposition to Port Royal, an aesthetics - what he described as a 'logic without thorns'.11 Bouhours thinks that beauty depends on something unique, resistant, and intellectually impenetrable in the work of art: it is not a matter of clarity and distinctness, but of indistinctness. Its appreciation calls for a particular fineness - a delicatesse - in the observer: a sense for half-tones and nuances, a receptiveness for sudden unexpected revelations following long confusing searches, and so on.12 'Confusion' enters into this general conception in several ways. In the 'Meditations' Leibniz draws a parallel between aesthetic perception and ordinary perception; just as seemingly simple perceptual qualities like 'red' are composite and can be resolved into their underlying causes by moving to the physical level of analysis, so aesthetic qualities such as beauty are composite, and judged to be unanalysable. 'We sometimes see painters and other artists correctly judge what has been done well or done badly; yet they are often unable to give a reason for their judgement, but tell the inquirer that the work 11 A. Baeumler, Das Irrationalitaetsproblem in der Aesthetik und Logik des 18. Jahrhunderts (1923) (Wissenschaftliche Buchgesellschaft: Darmstadt, 1967), p. 29. 12 Ibid., p. 32. On Leibniz's interest in D. Bouhours' La maniere de bien penser dans les ouvrages d'esprit (1687), see the editors' note in the New Essays, p. xxix.
82
Kant and His Influence
which displeases them lacks "something, I know not what".'13 A second example is from the 'Principles of Nature and of Grace', and here Leibniz makes the connection between perception and confused cognition stunningly explicit. 'Even the pleasures of the senses are reducible,' he says, to intellectual pleasures, known confusedly. Music charms us, although its beauty consists only in the agreement of numbers and of counting, which we do not perceive but which our soul nevertheless continues to carry out, of the beats and vibrations of sounding bodies which coincide at certain intervals.14 Certain pictures look like confused blurs until they are seen in a spherical mirror, when it is suddenly apparent what they represent, e.g. Julius Caesar.15 So much then, for the sources and main points of the theory of confused perceptions. The reader will note that, just on the basis of what has been said so far, it would be difficult to assign to Leibniz the view that we possess only a single basic faculty of apprehension, or that perception is the lowest grade of cognition. To see whether he has any such systematic view, we need to look at some other theoretical statements; I will not cite the evidence on both sides here, but rather point out the very different constructions that can be placed on the same alleged pieces of evidence. The single most suggestive text comes from the 'Principles of Nature and of Grace': it reads Each soul knows the infinite, knows everything, but "'Meditations', L: 291; G IV: 423. L: 641; G V I : 605-6. 15 New Essays, p. 257. 14
Confused Perceptions, Darkened Concepts
83
confusedly. Thus when I walk along the seashore and hear the great noise of the sea, I hear the separate sounds of each wave but do not distinguish them: our confused perceptions are the result of the impressions made on us by the whole universe. It is the same with each monad. Only God has a distinct knowledge of everything for he is the source of everything.16 This passage, which expounds a doctrine of confused omniscience, seems at first to support the single faculty view. For God's knowledge, our knowledge, and the knowledge of every other creature seem to lie on a continuum. We are all granted a certain omniscience, only our human and animal omniscience is confused while God's is distinct. Assuming that God's knowledge is intellectual knowledge, then, we would have to say that our knowledge is intellectual knowledge too; otherwise it would not be the same kind of omniscience in our case and God's case and the passage would be pointless. So, on this reading, when we hear the waves, we have some confused intellectual knowledge of the waves. However, a thought is cutting across this one: the thought that we know passively, where God knows actively. He knows everything because he is the source of everything - that is what the passage says. If so, there is no continuum. God does not soak up or receive impressions from every part of the universe as I soak up the sound of the ocean; nothing impresses itself on him. He is not there being reached by everything as the creatures are. So the passage could just as well be said to point to a categorical difference between perception and intellection, between the kind of knowledge that "L: 640; G VI: 604.
84
Kant and His Influence
depends upon having sensory organs that are impressed upon, and the kind of knowledge that does not. Here is a second ambiguous piece of evidence for the unity of the apprehending faculty, which depends upon an analogy between the fineness of the natural world and our confused apprehension of it, and the beauty of the moral world and our confused intellectual understanding of it: We also find order and wonders in the smallest whole things when we are capable of distinguishing their parts and at the same time seeing the whole, as we do in looking at insects and other small things in the microscope. . . . This is also true when one looks at the brain, which must undoubtedly be one of the greatest wonders of nature . . . yet one finds there only a confused mass in which nothing unusual appears, but which nevertheless conceals some kind of filaments of a fineness much greater than that of a spider's web. . . . We may say that it is the same in the government of intelligent substances under the kingship of God, in which everything seems confused to our eyes. To try to perceive immediately the beauty of the world is 'like wishing to take a novel by the tail and to claim to have deciphered the plot from the first book; the beauty of a novel, instead, is great in the degree that order emerges from the very great apparent confusion'.17 Note the progression of ideas: the undifferentiated speck or blob of the insect body is resolved by the lens into limbs and organs which give us an intellectual understanding of the insect body; so the undifferentiated mass of the human brain could be resolved into a system of fibres and filaments which would help us understand its 17
'Reflections on the Common Concept of Justice', L: 565.
Confused Perceptions, Darkened Concepts
85
functioning; so the moral confusion of the world could be resolved, enabling us to see that it is a system after all. We have moved from peering intently at something with a microscope to 'looking' at the world from a religious perspective, that is to say, from a particular intellectual perspective. Does this mean that clearing up perceptual indistinctness is a Way of acquiring knowledge? Is an inadequate intellectual grasp much the same thing as an inadequate perceptual grasp? That is what the passage and its internal dynamics might suggest. But one might also say that the three cases are simply related to each other by analogy - that there is a rhetorical, rather than a logical, relation between the three cases. If so, then no assimilation of perception and cognition is necessarily implied. My third piece of ambiguous evidence is Leibniz's seemingly direct statement that there are only two fundamental faculties possessed by 'simple substances': perception and appetition. 'The passing state which enfolds and represents a multitude in unity or in the simple substance is merely what is called perception.'18 No separate description of the state of 'cognition' or 'intellectual apprehension' of an object is given. This suggests that a single faculty of representation must embrace both. There is, however, some discrepant information, which suggests that Leibniz recognizes a difference between perception and intellection; that they are, at any rate, clearly defined subcategories. For in the 'Monadology' we also find the claim advanced that the intellectual apprehension of necessary and eternal truths is a reflexive action in which the soul does not merely mirror or reflect the universe, but turns back to look at 18
'Monadology', 14, L: 644; G V I : 608f.
86
Kant and His Influence
itself. In spirits 'the knowledge of necessary and eternal truths . . . gives us reason and the sciences, lifting us to the knowledge of ourselves and of God . . . we rise to reflective acts'.19 In the New Essays, Leibniz comes close to Locke's formulation, arguing that, besides ideas which are acquired through the senses, there are ideas like those of substance, being, unity, etc., which the mind achieves by reflection on itself. We might as well be talking about a higher faculty here: Kant, if he had taken a mind to it, might as well have accused Leibniz in these contexts of having sensualized cognitions, by supposing that these ideas and truths can be, as it were, seen on the soul, or inscribed on our inner tablets. What of our original text, the 'Meditations'? Here, despite the usual loose seventeenth-century handling of words like 'idea' and 'knowledge', Leibniz implicitly distinguishes between an idea in the sense either of an image or of an immediate impression, later called an Anschauung or a Vorstellung, and an idea in the sense of a concept, a Begriff. Each can be indistinct in its own way, and his whole exposition seems indeed directed precisely toward a separation of experiential and analytical elements. The vividness and simplicity of an impression - an immediate sensory experience - are something other than a discursive or analytical understanding of an object. That is why colours, flavours, odours, and, surprisingly perhaps, beauty are all clear ideas (and, usually, clear impressions), while they are neither distinctly sensed nor distinctly conceptualized. We do not perceive their causes distinctly, nor do we know, discursively, what their causes are. That intellectualized concepts are different from perceived or imagined objects is plain from the fact that, although I "Ibid., 29-30, L: 645-6; G VI: 611.
Confused Perceptions, Darkened Concepts 87
cannot imagine or see a thousand-sided figure except confusedly, and I cannot understand the individual terms of the definition of the word 'chiliagon' distinctly, my concept is nevertheless distinct.20 Still, we should have to say that Leibniz does not always hold ideas in the sense of Vorstellungen or presentations, and ideas in the sense of Begriffe or concepts, apart, and that this opens the door to the perception-asconfused-cognition theory. In his discussion of 'obscure concepts', Leibniz explains that: A concept is obscure, which does not suffice for recognizing the thing represented, as when I merely remember some flower or animal which I have once seen but not well enough to recognize it when it is placed before me and to distinguish it from similar ones, or when I consider some term which the Scholastics had defined poorly, such as Aristotle's entelechy.21 In this case my idea, in the sense of my Vorstellung, is poorly defined and so is my idea in the sense of my Begriff. My mental image is not sufficiently distinct to enable me to give defining criteria. The implication is that if my Vorstellung were sharper, clearer, a better picture, my Begriff would be in better shape too. While happily allowing that to get a better Begriff of a chiliagon I do not need a better Vorstellung, Leibniz fails, perhaps, to be sufficiently interested in the fact that to get a better Begriff of, say, an ash-tree, I do seem to need a better Vorstellung. And sometimes Leibniz does suggest that what we see confusedly is identical with the discursive analysis of the thing we see. What we see confusedly 20 21
'Meditations', L: 292; C IV: 423. Ibid., L: 291; G IV: 422.
88
Kant and His Influence
when we see red is whirling corpuscles of a certain description; 'whirling corpuscles of a certain description' is also part of the discursive analysis of red. There seems to be, in other words, a use-mention confusion.22
2
Kant's Reception of 'Confused Cognition'
While a reading of the Critique might make anyone wonder how much Kant actually knew of Leibniz, and while we have surprisingly little evidence of much firsthand study of Leibniz's works, there can be no doubt that Kant was seriously concerned with Leibnizian themes - from monadology, pre-established harmony and theodicy, to forces, space and time in his early essays, and indeed throughout the course of his life. He did not, at least at first, need to return to the original, as Leibniz was well-furnished with interpreters and systematizes in C.F. Wolff (1679-1754); A.G. Baumgarten (1714-1767); and C.F. Meier (1718-1777). Kant's lectures in metaphysics, for example, were a critical exposition of the Metaphysica of Baumgarten, a monadologist and aesthetician, which he studied and annotated thoroughly. His lectures on logic were a critical exposition of Meier, author of the Vernunftlehre, Baumgarten's pupil, and both Baumgarten and Meier looked back to Wolff, who rightly and wrongly was seen as the explicator and systematizer of Leibniz. In the course of this transformation of Leibniz's writings into a system, the confusion doctrine is clearly articulated. For Wolff and Baumgarten, there is, dogmatically stated, one faculty of representatio with perception 22 Kant seems to have taken Leibniz as saying that there are properties which would be mentioned if we were conceptually distinguishing between a right-handed object and a left-handed object; and that the properties themselves are actually dimly apprehended when we see those objects.
Confused Perceptions, Darkened Concepts
89
and cognition corresponding to its 'lower' and 'higher' parts.23 And, to make matters worse, we see, in Wolff's work especially, the thorough confusion of Begriff and Anschauung, concept and intuition, that so provoked Kant. So, for example, Wolff defines a concept {Begriff) as 'a presentation in our thoughts'. 'I have a concept of the sun,' he says, when I represent it to myself in my thoughts, either through a picture, as though I were to see it before me, or through mere words, with which I would make it clear that I had seen the sun, e.g. that it is the blazing body in the sky by day which blinds the eyes and make it warm and light on the earth; or through another symbol, such as the astrological sign. The same goes for marriage. I have a concept, when I either see in my mind an image of two people getting married or again through mere words: linguistic signs.24 Our concepts, he goes on to say, may be more or less 'dark', darkness being opposed to clarity of perception. For example, he says, drawing on Leibniz, I see a plant in the garden and can't quite remember whether I have seen this plant before, or know its name. Most people, he thinks, have this kind of dark concept of philosophical terms and mathematical ones, and that is why they have 23 See Baumgarten, Metaphysica, 7th edn (1779; reprinted by Olms: Hildesheim, 1963): 'Anima mea quaedam cognoscit obscure quaedam confuse cognoscit. . . . Unde FACULTAS obscure confuseque seu indistincte aliquid cognoscendi COGNOSCITIVA INFERIOR est. . . .' (#520); 'Repraesentatio non distincta sensitiva vocatur' (#521); 'Anima mea cognoscit quaedam distincte . . . facultas distincte quid cognoscendi est FACULTAS COGNOSCITIVA SUPERIOR' (#624). 2A Vernunfftigen Gedancken von den Kraeften des Menschlichen Verstands (1713; reprinted by Olms: Hildesheim, 1965), Chapter 1, section 4. Otherwise known as the Deutsche Logik, this work was directly inspired by Leibniz's 'Meditations on Knowledge, Truth and Ideas' as well as by Locke and Tschirnhaus.
90
Kant and His Influence
difficulties with those subjects. Note that because of the indiscriminate treatment of Begriff in both its pictorial and its verbal sense, darkness of concepts describes any less-than-ideal perceptual situation as well as any lessthan-ideal epistemological situation. I also have a dark concept, Wolff thinks, when what I experience is too far away or too small - or even when it is just too dark outside - to distinguish its features clearly 'as when we see something white in the twilight lying far away in a field' or when a discursive gardener is explaining plants to me, but I am distracted by the person accompanying me through the garden.25 Perceptual indistinctness, indistinct cognition, and darkness of concepts are blended together here in an unfortunate way. And Wolff also associates clear-sightedness with theoretical improvements when he says that microscopes and telescopes help us to refine our concepts. What he means is quite innocent; a microscopical examination of Brennessel shows us that its irritating power really involves tiny points on the leaf; that sparks are really glowing bits of steel or stone; that the milky way is a collection of individual stars and that the moon is mountainous.26 There is nothing 'dark' or 'confused', Kant protests in his own logic lectures, about failing to distinguish the individual parts of an object. For example, if you look at a house, you must take in its parts: doors, windows, etc. 'For if we did not see the parts we should not see the house itself.'27 We are not individually conscious of the plurality of parts, but this does not darken or confuse the perception. And this is his stance in general. The 25
Ibid., section 23. Ibid., section 22. 27 Kant's Introduction to Logic, trans R.S. Hartman and W. Schwartz (Bobbs-Merrill: Indianapolis and New York, 1974), p. 38. 26
Confused Perceptions, Darkened Concepts 91
telescope, he says, changes the white band of the milky way into a host of individual stars: the stars were indistinct, but my perception was not confused.28 In making this substitution he is effectively redefining confusion as a lack of order. It has nothing to do with a failure to recognize, identify or define a perceptual object, or with a failure to note or distinguish its parts. Confusion is not the antonym of 'distinctness' but the antonym of 'order'. Throw some coins on the table in any old way and they are confused; put them in neat piles and you have order. It follows that our impression of red cannot be, as Leibniz stated, 'confused' for it has no parts to become disordered. In the case of the impression or even the idea of beauty, we cannot, Kant admits, state its 'parts' i.e. its analysis or the cause of the impression to ourselves in detail. But our inability to give a conceptual analysis of beauty does not imply that the experience of beauty is something involving confusion. Now, it is worth pointing out that Leibniz, at any rate, did not make the inference he is being accused of here. And it is probable that Kant's own theory of aesthetics does without the notion of confused perception because he is concerned with a different kind of aesthetic object. His focus is, as a result, no longer on the aesthetic experience of the observer, but on his exercise of a faculty of judgement or taste. The typical aesthetic objects of the baroque, the trompe I'oeil painting, the complicatedly plotted novel, the formalistic musical pieces - examples, one might say, of 'court art' - lend themselves to Leibniz's treatment in a way in which Kant's aesthetic objects, with their somewhat stifling air of bourgeois domesticity, do 28
Ibid., p. 3 9 .
92
Kant and His Influence
not.29 The activity of connoisseurship, with its picking, choosing, and comparing, with its judgements, is suited to those objects; there is correspondingly nothing mysterious or intriguing about how they are experienced: certainly no need for the Leibnizian multiplicity of levels. Note as well that Kant is able to redefine confusion as a problem of orderliness because he is not at all concerned with something which preoccupied Leibniz: the coincidence of cause and object in perception. The problem of primary and secondary qualities or visual impressions and their corpuscular causes that so perplexed his predecessors either does not register with him or he has seen beyond it: Transcendental Idealism is perhaps a way of putting the question about the relation between cause and object in perception out of bounds. For what Kant introduces in place of this truthguaranteeing coincidence is the notion that two faculties co-operate to produce our perceptions, each of which guarantees objectivity in a different way. There are two sources of representation, he maintains, 'which, while quite different, supply objectively valid conceptions of things by working in conjunction'.30 Sensibility delivers to us the matter of knowledge, the understanding brings sensibility under rules. This distinction, as he says, is not 'logical' but metaphysical: It has, in other words, nothing to do with concept formation, with introspectively identifiable differences in our thoughts and experiences; it is not part of 'empirical psychology' or epistemology: it has to do with metaphysics, and that is to say with origins. 29 'Porzellandosen und - Figuren, Stockknoepfe, Spitzen, Tapetenmuster und derartige Dinge waren die Gegenstaende, von denen Kant seine Geschmackslehre abstrahierte.' Quoted by Baeumler, Das Irrationalitaetsproblem, p. 254, from Schlapp, Kants Lehre vom Genie, p. 181, fn. 5. 10 Critique, A 271/B 327.
Confused Perceptions, Darkened Concepts 93
And it is evident that this point is specifically meant against Baumgarten.31 The co-operation of two faculties, passive and active, one which delivers, one which organizes, then, replaces the Leibnizian picture which veers to odd extremes. Sometimes Leibniz suggests that all experience in the sense of representatio is wholly passive. Every creature is a mirror of the world: it mirrors some things distinctly, some confusedly. And sometimes he suggests that experience is wholly active, that every mind produces or draws up from within itself all its thoughts, which are clear or confused depending on the type of mind it is. When Kant splits the faculty of representation to derive separate active and passive features in perception, he reminds the reader that it is part of the human character of knowledge that it is so divided. Were I to think an understanding which is itself intuitive (as, for example, a divine understanding which should not represent to itself given objects, but through whose representation the objects should themselves be given or produced) the categories would have no meaning whatsoever in respect of such a mode of knowledge. They are merely rules for an understanding whose whole power consists in thought. An archetypic intelligence would not contain a formal 31
See Kant's 'Reflexion' # 2 2 0 to Baumgarten's 'Metaphysica' # 5 2 0 : 'Der Unterschied der Sinnlichkeit vom Verstande ist 1. formal, da die erste Erkenntnis intuitive, die zweite discursive ist . . . In jeder diesen beyden Formen kann Deutlichkeit oder Undeutlichkeit statt finden, nemlich in der Anschauung oder im Begriff. Deutlichkeit der Anschauung findet statt wo gar kein Begriff ist . . . Man muss auch nicht zur Unvollkommenheit der einen Form als der Anschaung das rechnen, was eigentlich nur die Form des Verstandes angeht. Der Unterschied des Verstandes beruhet also nicht auf Verwirrung und Deutlichkeit, wohl aber der Form nach, auf Anschauung und Begriff. In Kants Werke, 19 vols (Walter de Gruyter: Berlin and Leipzig, 1917-28), 15: 84.
94
Kant and His Influence
component working on the matter as a given; it would produce both.32 There is, then, no continuum of representations from the divine to the dim; instead the archetypal, creating intelligence and the passive receptive intelligence seem to work together.33
3
The Confused Apprehension of the Noumenal World
I now come to the last section of the paper with its question, how did Kant's refusal to treat sensibility and cognition as higher and lower grades of the single faculty of representatio give him a point of departure for criticizing the 'Monadology' qua philosophical document? Kant sometimes presents his criticisms in a single breath, sweeping epistemology and metaphysics together. So he tells us early in the work that his whole teaching with regard to sensibility would be rendered empty and useless if we were to accept the view that our entire sensibility is nothing but a confused representation of things, containing only what belongs to them in themselves, but doing so under an aggregation of characters and partial representations which we do not consciously distinguish. The representation of a body in intuition, he continues,
32
Critique, A 271/B 327. In the logician Meier we get a possible route by which the split into the material and the workman might emerge from the Leibnizian theory of confused perception. In Meier, dark knowledge is the matter, which the craft of the soul shapes into clear knowledge: 'When God made the world, he made something dark and empty: a chaos and out of it he built this splendid worldedifice. Dark knowledge is the chaos in the soul, the raw lumps of matter which creative force works upon and out of which she gradually assembles all her knowledge. Without dark knowledge, we could not have any clear knowledge at all'. Auszug aus der Vernunftlehre (Halle, 1752), p. 131. 33
Confused Perceptions, Darkened Concepts
95
'contains nothing that can belong to an object in itself, but merely the appearance of something'.34 But what we experience is not, in Leibnizian fashion, an appearance of a thing-in-itself: 'With their theory of a single cognitive faculty', Kant thinks, Leibniz and Wolff 'abolished the distinction between phenomena and noumena to the great detriment of philosophy'.35 Now, that Kant should have accused Leibniz of taking appearances for things-in-themselves and so confusing phenomena and noumena seems at first even more incredible than that he should have accused him of treating perception as a variety of cognition. Are not the monads noumenal objects par excellence} Are they not, as incorporeal, non-spatial, spontaneous minds, unaffected and unaffecting, utterly different from the persons, animals, and material things that appear to us, and that act upon us and are acted upon? Kant at some points takes Leibniz as asserting that an ideally clear cognition of a body would reveal it as an ensemble of monads - immaterial simples - while our ordinary, indistinct perceptions show a continuous extension. Phenomena and noumena are thus confused, for the monad is a noumenal object only in so far as it exists below the threshold of distinct perception. And this interpretation is less arbitrary than it might seem, for Leibniz, though distressingly vague on the question of how to derive the continua of experience from the discrete unextendeds of metaphysical reality, did at times suggest that matter is the result of a perceptual blurring and confusion of monads.36 Supernaturally clear perception, for example of a divine being, which would be no different * Critique A 44/B 61. 35 Ibid. 16 See, for example, Leibniz's 'Letter to Sophie', G V I 1 : 564.
96
Kant and His Influence
than that being's cognition, would then resolve the continuum into discrete parts revealing units - monads. Which view was adopted by Leibniz's eighteenthcentury interpreters? Wolff says that space is that thing which we imagine when we imagine some arrangement of things;37 but he also says that many simple substances can be gathered together, whereby they fill up space together 'though indeed each individual one actually fills up no space, but simply occupies its particular place within it'.38 He argues from a logical 'Verknupfung' or 'bond' of the cause-effect type (the earth and the sun, being, for example, verknupft), which implies their spatial distinctness, to the possibility of their spatial Verknupfung and so extension in length, breadth and thickness.39 But it is not simply the case that Wolff leaps from the possibility of logical relations to the possibility of spatial relations: what he seems to have in mind here, surprisingly enough, is something like the dynamical occupation of space by repulsive point-forces.40 After stressing that simple things cannot be stuck together as ordinary objects of experience are, he states that 'Just as they are grasped only through the understanding, so their Verknupfung with each other is also only intelligible'.41 Thus no direct claim is made to the effect that a confused perception of monads yields the appearance of continuous matter. In Baumgarten's work, however, we get a more suggestive statement. Though he, too, seems to have both logical and dynamical ideas of monadic bonding,42 he does state in the context of a 17 'Deutsche Metaphysik', #46. "Ibid., #602. "Ibid., #603. 40 'die Puncte der Natur . . . auch eine in ihnen bestaendig wuerckende Kraft haben, und daher ein jedes ausser dem andere seyn muss', ibid., #604. 41 Ibid., #604. 42 'Metaphysica', #402 ff.
Confused Perceptions, Darkened Concepts
97
discussion on monads, atoms, and corpuscles, that observabilia or phenomena are the things which we can know through the senses, that is, confusedly.43 We explain those phenomena by means of corpuscles, which are not, however, to be identified with monads. Thus, confused perception is at least introduced into the discussion of aggregation, though both Wolff and Baumgarten seem determined to maintain a categorical distinction between simple substances and the seemingly simple objects of experience, one which cannot be effaced by confused perception. We get a much stronger link, however, in the later Leibnizian, J.A. Eberhard: Concrete time, or the time which we feel, is nothing other than the succession of our representations; . . . [it] is therefore something composite; its simple elements are representations. Since all finite things are in a continual flux, these simple elements can never be sensed; inner sense can never sense them separately. . . . Furthermore since the flux of alterations of all finite things is a continual unbroken flux, no sensible part of time is the smallest, or a fully sensible part. The simple elements of concrete time therefore lie completely outside the sphere of sensibility. . . . The understanding therefore raises itself beyond the sphere of sensibility as it discovers the unimageable simples without which the sensible images, even in respect to time, are not possible.44 Thus we have an example of at least one Leibnizian 43 Ibid., #425: and he seems to allow, like Leibniz, that 'material atoms' can be used to explain phenomena, though the notion is actually a 'contradictio in adjecto'. Cf. ibid., #429. 44 J.A. Eberhard, Pkilosophisches Magazin, quoted H.E. Allison (trans and ed.), The Kant-Eberhard Controversy (Johns Hopkins: Baltimore, 1973), p. 23.
98
Kant and His Influence
though the dating of this passage is too late to suppose that it was an initial stimulus to Kant - who believes that intelligibility is a matter of lying below sensory thresholds. And the account being offered here seems distinctly different from the relational theory of Wolff. The experience of time is said to be a blur of tiny insensible intervals between tiny, atomic imperceptible experiences, too brief for us to apprehend singly. Now, for Kant, as we know, time could not be the cumulative experience of successive atomistic subsensible experiences. Experiences are, as such, temporalized; they cannot be built out of temporal atoms. The error of Leibnizians like Eberhard is that they regard something of which we can form no image, something which we cannot recognize or identify in intuition, as a nonsensible object. The objects of understanding, they say, are 'unimageable'. But this unimageability concerns, Kant points out, only the limits of perception, not a metaphysical uniqueness. According to Kant, the result is that Eberhard has reproduced the error of Wolff; if the difference between the sensible and the intelligible is a matter of degree and has to do with human capacities and thresholds, then a nine-sided figure is 'more than half-way from the sensible to the supersensible' and a chiliagon is an entirely intelligible object.45 We have now reached the point where the full weight of Kant's epistemological distinction is thrown against the monad. Mr Eberhard wants the monads to be known as nonsensible objects, either insofar as they are too small for the degree of sharpness of our sense, or the number of them in intuition is too large . . . about which we 45
Ibid., p. 127.
Confused Perceptions, Darkened Concepts
99
should be able to know a great deal through the understanding.46 Kant forces a parting of the ways: if microscopes could enable us to see monads by resolving the confusion of the aggregate into distinct beings, seeing those beings would not give us a distinct understanding of simple substances. If, by contrast, the concept of simple substance can be arrived at by a priori reflection, those substances cannot be apprehended, even confusedly, as objects. To sum up, for a Leibnizian like Eberhard, there are two ways of apprehending the monad-world; directly, though confusedly, through perception, and indirectly, though clearly, by reflection on the necessary properties of simple substances. On the interpretation I have been proposing, Kant's attack on the internal logic of 'confused perception' culminates in his cutting off of the direct route to the monad-world. Other portions of the Critique are of course dedicated to cutting off the other possible mode of access.
4
The 'Platonic' Approach to Noumena
In his later writings, particularly in his reply to Eberhard, Kant tried to reclaim Leibniz from the Leibnizians. In the Metaphysische Anfaengsgruende of 1786, he had already criticized his own understanding of Leibniz, as well as the interpretations of commentators, as insufficiently Platonistic: space was always thought of as adhering to things outside of our power of representation, but the mathematician thought of this quality only according to common concepts, i.e., confusedly (for appearance 46
Ibid., p. 128.
100
Kant and His Influence
is commonly explained in this way). By the same foregoing misinterpretation one attributed the mathematical proposition of the infinite divisibility of matter, . . . to a confused representation of space. . . . In this way it remained open to the metaphysician to compound space of points and matter of simple parts and thus (according to his opinion) to bring clarity into the concept of space. This ground of this aberration lies in a badly understood monadology, which does not at all belong to the explication of natural appearances but is a platonic concept of the world carried out by Leibniz.47 On Kant's view, then, Leibniz, a proto-Kantian, had really wanted to say that space belongs only to the appearance of external things. Reclaiming Leibniz from the textbook writers who had made him a systematist and a dogmatist, especially by overextending his confusion theory, meant 'Platonizing' Leibniz, and showing that he was also a proto-Kantian in regarding phenomena and noumena as belonging to separate realms. And this leaves us with an historical question. As we have seen, there does not appear to be a straightforward claim to the effect that monads are confusedly perceived as extended matter in either Wolff or Baumgarten. The discussion of monadic 'V'erknupfung or '' composition is separate from the discussion of the order of the cognitive faculties. Did persons who are unnamed but whose existence is indicated in the passage from the Metaphysical Foundations really promulgate the 'wrong' Leibnizinterpretation? Or was this doctrine simply extracted from time to time, by Kant among others, from the 47 Metaphysical Foundations of Natural Science (1786), trans J.W. Ellington (Bobbs-Merrill: Indianapolis and New York, 1970), p. 55.
Confused Perceptions, Darkened Concepts
101
known Leibnizian writings? I have tried to show that pertinent passages from Leibniz's more popular writings such as the 'Principles of Nature and of Grace' do not seem to support the doctrine in question very consistently. In looking for a source of the 'wrong' interpretation, we might find it necessary to bypass Wolff and Baumgarten and look to the material published in 1768 in Dutens' edition. The 'Letter to Hansch' might be promising in this respect, as the source of Kant's own initial misreading and the grounds of his self-confessed correction. For it is in this letter that Leibniz, in the space of a few lines, claims that sensible or composite things are in flux and do not really exist, that the mind contains an intelligible world within itself and also represents a sensible world to itself, and that 'God sees all things adequately and at once, while very few things are known distinctly by us; the rest lie hidden confusedly as it were in the chaos of our perceptions'.48 It would be difficult to blame anyone for concluding from this passage that sensible things are composites confusedly perceived where divinely clear perception would be an intelligible representation of substances. In rejecting this conception, Kant assigned priority to the Newtonian world of objects located in space and time and forming together a system of dynamical interactions; the Newtonian world is riot a degraded appearance, a confused representation of a higher, purer world of metaphysically real but non-spatial, non-temporal, noninteracting entities. But validating the Newtonian picture was not Kant's only concern, and it is still worth wondering how Kant's own preoccupation with the supersensible was furthered by his encounter with Leibniz. How far did Kant really approve of the reconstructed Tlatonizing' <8 'Letter to Hansch', 25 July 1707, L: 592-3.
102
Kant and His Influence
Leibniz - the Leibniz who did not mix up phenomena and noumena by conceiving them as passing over into one another? I have suggested that Kant repudiates the scholastic idea of a graded series of cognitions ranging from the ideally dear and comprehensive to the dim and restricted. At the same time he was not prepared to do away with the notion of the 'intelligible world' or even with certain noumenal force-bearing objects, for the purposes of morality if not of physics. Understanding his Leibniz-critique does not itself show what he means by his intelligible world or what his own noumenal objects come to. But it does cast some light: we know both that they are not possible objects of perception and that they are not objects restricted to divine cognition. They are, just as Kant is always stressing, objects of our own cognition, though our own cognition when it assumes, if not the impossible capacity for cognitive affections, something of the dictating power that went with the vacated perspective of Leibniz's God.
Bibliography Baeumler, A., Das Irrationalitaetsproblem in der Aesthetik und Logik des 18. Jahrhunderts (Wissenschaftliche Buchgesellschaft: Darmstadt, 1967). Baumgarten, A.G., Metaphysica, 7th edn (1779) (Olms: Hildesheim, 1963). Eberhard, J.A., Philosophisches Magazin quoted Allison, H.E. (trans and ed.), The Kant-Eberhard Controversy (Johns Hopkins: Baltimore, 1973). Gerhardt, C.I. (ed.), Die Philosophische Schriften von G.W. Leibniz, 7 vols (Olms: Hildesheim, 1962). Haldane, E.S. and G.R.T. Ross (eds), The Philosophical Works of Descartes, 2 vols (Cambridge University Press: London, 1931). Hartman, R.S. and Schwartz, W. (trans), Kant's Introduction to Logic (Bobbs-Merrill: Indianapolis and New York, 1974). Hooker, M. (ed.), Leibniz: Critical and Interpretive Essays (University of Minnesota Press: Minneapolis, 1982).
Confused Perceptions, Darkened Concepts
103
Kant, I., Critique of Pure Reason, trans N. Kemp Smith (St Martin's: New York, 1965). Kant, I., Metaphysical Foundations of Natural Science (1786), trans J.W. Ellington (Bobbs-Merrill. Indianapolis and New York, 1970). Kant, I., 'Reflexionen' in Kants Werke, 19 vols (Walter de Gruyter: Berlin and Leipzig, 1917-28). Leibniz, G.W., 'On the Form and Principles of the Sensible and Intelligible World' (1770), trans and edited by G.B. Kerferd and D.E. Walford in Selected pre-Critical Writings (Manchester University Press: Manchester, 1968). Leibniz, G.W., 'Principles of Philosophy' in The Philosophical Writings of Descartes, J. Cottingham et al. (eds), 2 vols (Cambridge University Press: Cambridge, 1985). Leibniz, G.W., Philosophical Papers and Letters, L.E. Loemker (trans and ed.), 2nd edn (Reidel: Dordrecht, 1969). Leibniz, G.W., New Essays, trans P. Remnant and J. Bennett (Cambridge University Press: Cambridge, 1981). McRae, R., Perception, Apperception and Thought (University of Toronto Press: Toronto, 1976). Maimon, Solomon, 'Letter to Kant', 20 September 1791 in Kant's Philosophical Correspondence 1759-1799, trans and edited by A. Zweig (University of Chicago Press: Chicago, 1967). Meier, C.F., Auszug aus der Vernunftlehre (Halle, 1752). Wolff, C.F., Vernunfftigen Gedancken von den Kraeften des Menschlichen Verstands (Olms: Hildesheim, 1965).
THOUGHT AND SENSIBILITY IN LEIBNIZ, KANT AND BRADLEY Guy Stock
Kant, in his final metaphysics, must either retreat to Leibniz, or advance to Bradley. Alfred North Whitehead1 I do not know why Whitehead made this claim. However, I thought it might be interesting, without further reference to Whitehead, to compare Leibniz, Kant and Bradley with Whitehead's assertion in mind. The very suggestion that one could be forced from a Kantian position back to a Leibnizian one or forwards to a Bradleian one implies the existence of intelligible links between them. Such links are particularly obvious between Leibniz and Kant. A natural way of approaching Kant's account of our knowledge of time and space, and the phenomena that exist in them, is to see it as prompted by the difficulties he found in Leibniz's account of thought and sensibility.2
1 A.N. Whitehead, Process and Reality (Cambridge University Press: Cambridge, 1929), IX p. 269. 1 I. Kant, Critique of Pure Reason, trans N. Kemp Smith (Macmillan: London, 1964); see, for example, A 44/B 61-2; 1 B 316-49/A 260-92.
104
Thought and Sensibility in Leibniz, Kant and Bradley 105
I For Leibniz individual substances, or monads, are orderable along two interrelated continua. They are orderable according to their activity and passivity, and according to the clarity and distinctness, as opposed to the confusion and obscurity, with which they represent or express the Universe. The more active an individual substance and the more clear and distinct its perceptions of the Universe the more perfect it will be: the nearer, so to speak, will it be to God, the purely active, omniscient and all perfect substance. No substance will be utterly passive. As Leibniz puts it: '. . . not only is everything that acts an individual substance but also every individual substance acts without interruption'. 3 Passivity, for Leibniz, is simply a relative lack of activity in a substance. It is merely a privation. Further Leibniz argues that in metaphysical strictness individual substances must be construed as windowless. Given the kind of independence essential to a genuinely individual substance it cannot depend on things outside itself for its nature. As he puts it: From the notion of an individual substance it . . . follows . . . that all operations of substances both actions and passions, are spontaneous, and that with the exception of the dependence of creatures on God, no real influx from one to the other is intelligible.4 The perceptions, or inner states, of a substance must change simply from an inner principle of appetition. 5 1 G.W. Leibniz, Philosophical Papers and Letters, L. Loemker (ed.) (Reidel: Dordrecht, 1976), 'On Nature Itself (1698) section 9, p. 502. 4 G.W. Leibniz, Philosophical Writings, G.H.R. Parkinson (ed.) (Dent: London, 1979), 'A Specimen of Discoveries about Marvellous Secrets', p. 79. ' See, for example, 'Monadology' section 15, Loemker p. 644.
106
Kant and His Influence
Hence, in metaphysical strictness, the influence of any individual substance on another must be ideal only: one individual, A, will act in a certain way on another, B, (which will in consequence be, in a certain respect, passive) in so far as in the ultimately original ordering of things A's state provided the reason for B's being in the state it is.6 Let us now concentrate on Leibniz's account of the experience of a rational monad dominant within a set of substances providing the foundation of a particular living human body: in other words, on the experience of a normal adult human being with, like ourselves, a power of apperception and the mastery of the dating-andplacing vocabulary of a natural language.7 The Universe will be presented in our waking perceptions and thoughts from a unique viewpoint. Physical objects will, at any waking moment, appear as spatially arrayed at various distances around our bodies and we will be able to think, at any moment, of our bodies as occupying a determinate position in a three-dimensional spatial system of unlimited extent. Analogously, we shall be able to locate ourselves in a one-dimensional time series and particular events, which are not present will appear, when at any moment we ponder them, as at determinate amounts of time in the past and future (e.g. the last degree exams will appear as having occurred X months ago and the next ones will appear as occurring in Y months time). This perspectival nature of our experience, and the uniqueness of each individual's viewpoint, which is an irreducible feature of the Universe on Leibniz's account, 6
See, for example, ibid., sections 51-2, Loemker p. 648. For the notion of apperception see 'Principles of Nature and of Grace based on Reason' (1714) sections 4-5, Loemker pp. 637-8; for that of expression or representation see 'Letter to Arnauld 9th Oct 1687', Loemker pp. 33940; also "What is an Idea?' (1678), Loemker pp. 206-7. 7
Thought and Sensibility in Leibniz, Kant and Bradley 107
is explained within his epistemology and metaphysics in terms of the idea of a unique degree of confusion attending the individual's representation of the Universe.8 It is the unique degree of confusion with which a substance expresses the Universe which will individuate it and determine its position in relation to all other individuals or, in other words, will determine its position on the continuum of perfection in relation to God (to the all perfect individual). Leibniz defines this notion of confusion in contrast to the notions of clarity, distinctness and obscurity. An individual's idea or concept is, Leibniz maintains, obscure when it 'does not suffice for recognizing the thing represented'.9 Leibniz gives two different kinds of example to illustrate what he means; the first concerns a perceptible particular, the second an abstract object. If a person remembers seeing a flower but did not 'see it well enough' to know, when it is presented to him again, whether it is the very same flower of just a similar one, then his idea of that flower would be obscure. Also if a person understood an ambiguous general term without being aware of its ambiguities he would have an obscure idea of the thing in question. To adapt Leibniz's example: if a person had a sufficient grasp of the meaning of the word 'cause' to apply it correctly in everyday contexts but was not aware that he was using it on some occasions to refer to efficient causes and on others to refer to final causes he would have an obscure idea of cause.10 The notion of clarity is then introduced by contrast to that of obscurity. 'Knowledge is' Leibniz says 'clear . . . 8
See, for example, 'Monadology' section 60, Loemker pp. 648-9. 'Meditations on Knowledge, Truth and Ideas' (1684), Loemker p. 291; see also 'Discourse on Metaphysics' (1686) section 24, Loemker pp. 318-19; 'An Introduction to a Secret Encyclopaedia', Parkinson pp. 5-9. 111 'Meditations on Knowledge, Truth and Ideas', Loemker p. 291. 9
108
Kant and His Influence
when it makes it possible for me to recognize the thing represented.'11 A person will have a clear and distinct idea if he can not only recognize the thing presented but also 'enumerate one by one the marks sufficient to distinguish the thing from others'.12 So a clear and distinct idea will be, for example, 'the kind of notion which assayers have of gold; one, namely, which enables them to distinguish gold from all other bodies by sufficient marks and observations'.13 It follows that having clear and distinct ideas of things, since this will involve an ability to frame explicit definitions of general terms, will presuppose use of a language. So we can take it that on Leibniz's view the gradual acquisition of language by a child will place it in a position to bring to clarity and distinctness progressively more of the ideas which it previously had obscurely and confusedly. At this juncture we can begin to see how logical and epistemological themes connect in Leibniz's thought. It is a basic premise for Leibniz's view of thought and language that all human thinking of the kind that can be either true or false must involve a basic alphabet or vocabulary of simple ideas. There must, in other words, be a series of attributes which are, as he puts it, simple 'in their own natures', or simple 'to the intellect', and thus can be 'conceived through themselves'.14 Leibniz's vision is that a language is in principle possible having a vocabulary of primitive terms standing for such simple attributes or, as he also calls them, 'ultimate '• Ibid. '- Ibid. u Ibid., Loemker p. 292. " See, for example, 'On the Elements of Natural Science' (1682-4), Loemker pp. 285-7; 'On Universal Synthesis and Analysis' (1679?), Loemker pp. 229-30; 'Of an Organum or Ars Magna of Thinking' (c. 1679), Parkinson pp. 1-4.
Thought and Sensibility in Leibniz, Kant and Bradley
109
genera'.15 By means of such a vocabulary, the complex general terms of any language could in principle be defined. Such definition would give a complete analysis of the more or less confused ideas possessed in understanding the complex general terms in any language and thus would enable a complete analysis of any proposition, formulated by means of complex general terms, to be given. Such a language would enable its possessor to have what Leibniz calls adequate ideas or knowledge of things: '. . . when every ingredient that enters into a distinct concept is itself known distinctly, or when analysis is carried through to the end, knowledge is adequate'.16 However, Leibniz is doubtful whether any actual example of such knowledge can be given although he thinks 'our concept of numbers approaches it closely'.17 In other areas of knowledge, although we can achieve distinct ideas of things, at some point in the process of analysis we inevitably come to a level where our knowledge of the characteristics constituting the distinguishing marks of the definiendum will be confused. At such a level of analysis a person's thinking or knowledge will be, what Leibniz calls, blind or symbolic, it will of necessity involve the use of terms 'whose meaning appears obscurely and imperfectly in the mind'.18 Such blind or symbolic thinking or knowledge is contrasted with what Leibniz calls intuitive knowledge. Intuitive knowledge is what one must have when one has an absolutely simple idea, i.e. when one has an idea of an attribute conceived through itself. 15
'On Universal Synthesis and Analysis', Loemker p. 229. 'Meditations on Knowledge, Truth and Ideas', Loemker p. 292. 17 Ibid. 18 Ibid. 16
110
Kant and His Influence
The content of our thinking and knowledge in empirical spheres, on Leibniz's view, inevitably remains blind or symbolic because it involves what he calls 'confused attributes'. In other words, it involves attributes which are 'composite in themselves or by intellectual principles but are simple to the senses and whose definitions cannot therefore be explained'.19 Thus general terms standing for such empirically simple attributes can be defined by us only ostensively. The fact that in the area of contingent truths we are in this way confined to symbolic or blind knowledge explains, in the context of Leibniz's system, how error or false judgement is possible for us. Given Leibniz's predicate-in-subject view of truth20 the only way in which false judgement can, in fact, occur is by the meanings of the words we use to formulate our empirical propositions outrunning our understanding of those meanings. That is what inevitably happens when our thinking is symbolic or blind. Our languages give us the power to construct propositions which, unbeknown to us, involve contradictions.21 In any sphere in which we had a notation meeting the criteria of Leibniz's perfect language this would be impossible. We would be able to have intuitive knowledge in such a sphere and thus, given such a language, false judgement would be impossible. This, of course, is precisely the position that an omniscient creator would be in with regard to the Universe as a whole: his knowledge of the Universe, in all its infinite complexity, would be totally adequate and intuitive, i.e. non-symbolic. For finite individuals, the capacity to 1<(
'On the Elements of Natural Science', Loemker p. 285. See 'Meditations on Knowledge, Truth and Ideas', Loemker p. 293; 'Discourse on Method' section 8, Loemker pp. 307-8; 'First Truths' (1680-4), Loemker pp. 267-8; 'On the Nature of Truth' (1686), Parkinson pp. 93-4. :i 'Meditations on Knowledge, Truth and Ideas', Loemker p. 292. 20
Thought and Sensibility in Leibniz, Kant and Bradley
111
master the use of linguistic symbols is a necessary condition of being able to achieve clarity and distinctness in knowledge, but in the limiting case of perfection knowledge would once again become non-symbolic. It would, nevertheless, on Leibniz's view, at the limit of perfection, remain a capacity essentially homogeneous with our symbolic thinking. It is this that marks the crucial point of difference between Leibniz's account of the relation between thought and sensibility and Kant's. To grasp this difference clearly it is necessary to look more closely at Leibniz's notion of a perfect language. In his 'Dialogue' of 167722 Leibniz introduces two ideas which in the history of philosophy have become essential to the notion of a perfect language. Firstly, he introduces the idea that in so far as different languages can be employed to formulate the same propositions, and can be used as vehicles for the same processes of reasoning, they must involve an underlying, non-arbitrary, universal grammar. Secondly, he introduces the idea that some notations can be better constructed, and thus less misleading as to the underlying grammar of the propositions formulated within it, than others. It is the former idea which is the more important for understanding the difference between Leibniz and Kant. On Leibniz's view, there is a single essentially homogeneous grammar (or faculty of synthesis) spontaneously exercised in the generation of any experience capable of yielding knowledge of the Universe (i.e. of any experience capable of being veridical, or non-veridical, of anything whatsoever inner or outer, self or not-self). A finite individual's possession of the grammar, or faculty of synthesis, must involve having ideas (albeit more or less confused) of the very same series of ultimate genera, 11
'Dialogue' (August 1677), Loemker pp. 182-5.
112
Kant and His Influence
or simple attributes, as would be exercised nonsymbolically by an omniscient creator in creating the Universe. We can see Leibniz as arguing that it has got to be like this if our sense experiences and thoughts are to be capable of being true of a Universe which is such that (not to beg the question with respect to the existence of God) it could have been created by a rational, omniscient and benevolent being. In other words, it has got to be like this if epistemological realism is to be preserved.
II Moving now to Kant, a rather different picture emerges. Of course Kant argues, within the framework of his epistemology, that we must give up Leibniz's transcendental realism. But at what point does the argument for that epistemology come? Kant can be seen as taking over from Leibniz the task of giving an account of the nature of a subject, the essence of which is to represent things - things other than itself as well as itself. But Kant's fundamental thesis is that the passive faculty of sensibility cannot be portrayed as a function of a degree of inadequacy or confusion in a single homogeneous and essentially active representative faculty. Sensibility must, Kant maintains, be construed as a radically passive faculty for the mere reception of given cognitive matter (of sensations).13 It must be construed as a faculty for immediate or intuitive cognitions of things that could, as a matter of logical possibility, be possessed by a sentient subject in the complete absence of any active representative or synthetic faculty. 23
Critique of Pure Reason, A 19-21/B 33-6.
Thought and Sensibility in Leibniz, Kant and Bradley 113
Kant agrees with Leibniz that there must be a spontaneous and original active representative faculty and he agrees that the grammar of that faculty must be strictly universal. It must be possessed by any finite subject whatsoever as a condition of the possibility of the subject having experiences of a kind capable of being veridical or non-veridical of anything whatsoever, inner or outer, self or not-self. But, given the independent function of sensibility, on Kant's view, the active synthetic faculty must be construed as purely general. The active faculty is in essence simply the set of pure concepts which we exercise when we understand the propositional formulae constructed by formal logicians, and when we follow through inference patterns between such formulae.24 If we posit the canonical status of formal logic, the interconnected set of concepts in question must be spontaneously exercised in the consciousness of any subject whatsoever - no matter what the character of the sensorily given data - if (i) it is to be able to have experiences the contents of which have truth-values and (ii) it is to be able to make inferences from the contents of such experiences. Hence the active faculty by itself, on Kant's view, is taken as quite unable to relate to any particular, or individual, existent. Thus, even a perfect instantiation of the active faculty would, by itself, be quite incapable of yielding knowledge of any existent thing. The joint operation of both active and passive representative faculties is necessary if any experiences (veridical or non-veridical) of any existent thing whatsoever, inner or outer, self or not-self, are to be generated in a finite subject.25 Moreover it follows, according to Kant's line of 24
Ibid., A 66-83/B 91-109. " Ibid., A 50-2/B 74-6.
114
Kant and His Influence
reasoning, that the passive faculty itself must have a dual function if a subject is to be able to have knowledge of the existence not only of things other than itself but also of itself. Given that a subject could not have knowledge of the existence of things other than itself if it did not have a capacity to receive data from those things (i.e. if it did not have a faculty of outer intuition) by parity of reasoning, it could not have knowledge of the existence of itself if it did not have a capacity to receive data from itself (i.e. if it did not have a faculty of inner intuition).26 This dual function of the faculty of sensibility plays a role in Kant's account of space and time and our knowledge of them. The spatio-temporality of the objects of human beings' sense experiences is to be construed as resulting from the intrinsic nature of human inner and outer sense. The nature of our inner sense is such that any matter received by it will appear as successive states of a self locatable in, and merely in, time; and the nature of our outer sense is such that any matter received by it will appear as states of objects locatable in threedimensional space. Hence time and space are ex hypothesi nothing but subjective conditions but, in consequence, it can be known in advance of experience, or a priori^ (i) that any possible object as it appears to inner sense will be locatable in time and stand in temporal relations of certain general sorts and (ii) that any possible object as it appears to outer sense will be locatable in space and stand in spatial relations of certain general sorts. Objects as they appear to outer sense are not as such 26
Ibid.; see B 152-6; also B 67-70: inner sense or intuition is not to be confused with apperception which is a specific exercise of the active faculty of representation on the matter of inner sense.
Thought and Sensibility in Leibniz, Kant and Bradley 115
subject to the form of inner sense but since any intuition or representation whatsoever, outer or inner, must itself be an inner state of some subject, objects of outer sense will of necessity appear in a subject's experiences not merely as in space but also as subject to temporal determinations.27 Apart from the well-known argument from the synthetic a priori nature of geometrical propositions28 the motivation for Kant's account of space and time, and our knowledge of it, it seems to me, comes primarily from the difficulties he located in Leibniz's account of the relation between thought and sensibility. These difficulties he recounts explicitly in 'The Amphiboly of Concepts of Reflection'.29 But it is very difficult to sort out the real difficulties in this section since the very notion of an amphiboly of a concept of reflection {if I understand it correctly) just presupposes the structure of Kant's epistemological theory. Transcendental reflection is concerned with the relations which objects have to one another in virtue of the relation they have to our dual cognitive faculties, i.e. to sensibility and to understanding construed as radically disparate faculties. A transcendental amphiboly consists in the 'confounding of an object of pure understanding with appearance'.30 So the source of Leibniz's errors is in general his confounding appearances, or objects of sensibility, with objects of pure understanding, with intelligibilia. It seems impossible to evaluate the force of Kant's detailed criticisms of Leibniz when they are formulated -Mbid., A 34/B 50-1. Ibid., A 46-9/B 63-6. It might also be noted that Kant's absolute account of space and time was, in a sense, in conformity with Newtonian science in a way in which Leibniz's relativistic account was not. "Ibid., A 260/B 316ff. 1(1 Ibid., A 270/B 326. 28
116
Kant and His Influence
within the framework of Kant's own epistemology. However, there does appear to be an internal incoherence in Leibniz's system which in a general way Kant's criticisms point toward. It is in essence the well-known problem which Leibniz himself at times was worried about namely that of giving a satisfactory account of contingency and freedom.31 This also relates to the problem of giving an account of the difference between the actual and the merely possible. Leibniz himself concluded that each individual substance, like Aquinas' angels, must be a species infima.11 Moreover, even possible individuals must have complete concepts, i.e. concepts sufficient to determine all their characteristics and their positions in infinitely complex possible Universes. The constituents of these possible Universes, as Leibniz puts it, all tend toward existence according to the degree of their perfection.33 However, merely possible Universes do not depend for their natures on God's will. Ex hypothesi God exercised his will simply in choosing to create, from the infinity of possible Universes, the best of them. But what difference could God's act of creation conceivably have made to that possible Universe? On Leibniz's account there is no room for a sensible, or non-intelligible, difference between the actual and the merely possible (as there is for Kant's actual phenomenal world and his merely possible Universes; the actual is the one given in empirical intuition). Of course, as just pointed out, there is, for Leibniz, an intelligible difference between the actual and the possible - namely its degree of perfection - but, unfortunately, that is not the kind of difference that can conceivably depend on God's will. •" See 'On Freedom' (c. 1679), Loemker pp. 263-6. ! - 'Discourse on Method', Loemker p. 308. •'•' See 'On the Radical Origination of Things' (1697), Loemker p. 487.
Thought and Sensibility in Leibniz, Kant and Bradley
117
III The incoherence of Kant's notion of a thing-in-itself provides Bradley with a sufficient reason for rejecting Transcendental Idealism. It would be much as if we said, 'Since all my faculties are totally confined to my garden, I cannot tell if the roses next door are in flower.' And this seems inconsistent. . . . If the theory were really true, then it must be impossible. There is no reconciling our knowledge of its truth with that general condition which exists if it is true.34 For Bradley it is a fundamental precept that no satisfactory metaphysics can divide appearance - what we experience and know in the activities of our everyday lives - from reality. What we know and experience exists and what exists must belong to reality. As he puts it: 'The assertion of a reality falling outside knowledge, is quite nonsensical'.35 At least Leibniz is not guilty of that error. But, of course, Bradley would not accept Leibniz's account of how appearances belong to reality: namely, by courtesy of a creator God and an infinite plurality of essentially active individual substances which represent the Universe with varying degrees of adequacy and confusion. There are three features of Bradley's account of thought that I wish to draw attention to. Firstly, like Kant but unlike Leibniz, for Bradley sentient experience is not to be seen as merely privative: it plays a quite fundamental role.36 If anything, for Bradley, thought is 14 F.H. Bradley, Appearance and Reality (Clarendon: Oxford, 1968), XII, p. 111. "Ibid., XII, p. 114. 16 F.H. Bradley, Essays on Truth and Reality (Clarendon: Oxford, 1914), VI, 'On Our Knowledge of Immediate Experience', pp. 159-61.
118
Kant and His Influence
the inadequate partner. Secondly, Bradley rejects what might be called 'the determinateness of sense': in other words, for Bradley there are ultimately no judgemental contents, or propositions, which are unconditionally true or unconditionally false.37 Leibniz's realism would have required this possibility, and Kant's empirical realism would have required it to be possible with respect to our judgements about phenomenal objects. This second point, although very important, I will simply mention here. What it can be seen as ruling out is a radically truth-functional account of the relation between thought and reality of the kind posited in logical atomism. Thirdly, Bradley admits that our thought when exercised, for example, in relation to objects in space and time, necessarily proceeds by means of the 'machinery of terms and relations'.38 In other words, it will of necessity involve an application of those concepts which Kant regards as categories. However, Bradley argues that, because these concepts are internally incoherent, the metaphysician cannot allow that reality could be as it necessarily appears in our empirical thought. A grasp of this argument is, of course, essential to understanding Bradley's metaphysics. However, for present purposes, I wish to concentrate on the role of immediate experience and its relation to discursive thought in Bradley's philosophy. As is well known, Bradley, as much as Frege, was an anti-psychologist with respect to logic.39 So far as logic is concerned there is nothing psychological about the ideal contents that we refer to, or (to use Bradley's 17 F.H. Bradley, Principles of Logic (Clarendon: Oxford, 1967), I, chapter II, section 71, pp. 99-100. '* Appearance and Reality, III, p. 28. 19 A. Manser and G. Stock, The Philosophy of F.H. Bradley (Clarendon: Oxford, 1984), VII, 'Bradley's Theory of Judgment' (G. Stock), pp. 135-7.
Thought and Sensibility in Leibniz, Kant and Bradley 119
alternative expression) predicate of, reality in our acts of judgement. The ideal contents that we predicate of reality and which we understand when people communicate their judgements to one another by means of declarative sentences must be what Bradley calls 'universal meanings' or, 'universals'.40 And although Bradley is well aware of the necessity of syntactical, or synthetic, complexity in the ideal contents of our judgements he nevertheless argues that in a sense any judgement is to be regarded as having but one idea.41 Even if we take for example an analytic judgement of sense (i.e. a judgement expressed by a declarative sentence containing indexical demonstratives like 'I have toothache now')42 the ideal content of the judgement must be a universal. The sentence can, in effect, be regarded as a single general term since identically the same ideal content, or meaning, can be used on indefinitely many different occasions, in indefinitely many different judgements, in relation to indefinitely many different facts. However, Bradley argues that the ultimate logical subject of which the ideal content in a judgement is predicated, and in virtue of which, in so far as it is true, the judgement is true (in other words that which he uses the dummy name 'R' to refer to in his formulae for the general form of a judgement, *R is such that S is P') cannot itself be an ideal content.43 Here we might see a version of the problem that Kant saw in Leibniz's system arising again. If the ideal or communicable content of any possible judgement is 40
See Principles of Logic, I, chapter I, section 9, p. 9. Ibid., I, section 11, p. 11. Ibid., II, section 7, p. 49. 43 See ibid., II, section 3, pp. 43-4 and section 5, pp. 45-6; also section 10, p. 51. 41
42
120
Kant and His Influence
universal and yet the ultimate subject of predication of any judgement, i.e. reality, is necessarily individual, how can our thought succeed in relating unambiguously to reality? Bradley gives a version of what seems to be the only kind of answer possible to this question - a version of the kind of answer supplied, in very different ways, by Kant's notion of empirical intuition and (to use an example from a theory of knowledge contemporary with Bradley's) Russell's notion of acquaintance in sensation. If the ideal, or conceptual, content of our judgements is to be able to relate unambiguously to the individual it must somehow be underpinned by an immediate, i.e. non-ideal, non-conceptual, mode of cognition. In other words, it must be underpinned by a mode of immediate experience. Now the question arises: how is this mode of immediate experience to be construed? Clearly, for Bradley, it cannot be interpreted as Kant construed it: namely a relation between a subject (qua thing-in-itself) and objects (qua things-in-themselves); nor as Russell came to construe sensation in the atomism of his Lectures on Logical Atomism, namely as a relation between a momentary subject and momentary sense-data.44 Bradley's positive account of the nature of immediate experience is not easy to characterize briefly. I will do my best. In Bradley's epistemology our waking sense perceptions play a fundamentally important role in our knowledge. For convenience at this point we can again concentrate on the waking experience of normal adult human beings who, like ourselves, have mastery of a natural language like English. Also, for ease of exposition, I will use - as 44 B. Russell, Logic and Knowledge, R.C. Marsh (ed.) (Allen and Unwin: London, 1956), 'On the Nature of Acquaintance', pp. 165-8.
Thought and Sensibility in Leibniz, Kant and Bradley 121
Bradley often does in similar contexts - the first person singular form of expression. But for Bradley this should not be taken to imply privacy of a Cartesian sort in the objects of knowledge. The world that I will regard in a fundamental way as my real world (or the world of actual fact), Bradley argues, will be the world that I can, at each waking moment, think of as a unique spatio-temporal system of unlimited extent containing my body and the objects I can currently perceive around it.45 I will think of this world as, for example, containing not only every presently existing human being but every actual human being who ever has, or ever will, exist. This world, as I make it an object of my thought at any moment, will be, what Bradley calls, an ideal construction. It will, in other words, be an intentional object identified in terms of a complex ideal content predicated of reality in judgement. Or, to put it another way, reality will appear, or be present, in ideal form in the contents of my present thoughts and sense perceptions as an object of a certain sort - namely a spatio-temporally extended system containing (among indefinitely much else) all actual human beings. Now, with what justification do I think of my real world (i.e. this world containing my body and the indexically demonstrable objects I can see and point to around my body now) as being, as it surely must be, unique? Bradley argues that there is nothing in the idea of a spatio-temporally extended system which excludes the logical possibility of the existence of an indefinite series of qualitatively similar spatio-temporally extended series spatio-temporally unrelated to one another.46 The ideal "s Essays on Truth and Reality, III, 'On Floating Ideas and the Imaginary', pp. 30; 44-5; XVI, 'On My Real World', pp. 461-9. 46
See Appearance and Reality, XVIII, pp. 186-9.
122
Kant and His Influence
content that I predicate of reality when I think of objects existing in my real world is like any other ideal content universal and as such cannot guarantee its own uniqueness of application. So Bradley is arguing that we have no right to take our ideas of space and time to have (in Kantian fashion) a special cognitive status - to be pure intuitions - and thus to be guaranteed, by definition, uniqueness in their objects. As Bradley concludes in a very important passage from the Principles of Logic:
It is not by its quality as a temporal event or phenomenon of space, that the given is unique. It is unique, not because it has a certain character, but because it is given. It is by the reference of our series to the real, as it appears directly within this point of contact, or indirectly in the element continuous with this point, that these series become exclusive.47 Bradley's point is as follows. If I employ the framework of ideas in terms of which I think of my real world I will necessarily, and for ordinary purposes quite properly, think of my sense perceptions, thoughts and so on, themselves as dateable events or, as successive states, in my personal psychical history. But I will then be prone to think of the uniqueness of my perceptions, thoughts and so on, as being a function of their occurrence - their being caused - at determinate positions in a unique spatio-temporal series. However, to do this within the context of epistemology and metaphysics, according to Bradley, is a mistake. To think of the uniqueness of my perceptions and thoughts, as being a function of their positions in a unique spatio-temporal series is to get it the wrong way round. On Bradley's view it is the uniqueness 47
Principles of Logic, II section 22, p. 64.
Thought and Sensibility in Leibniz, Kant and Bradley 123
of these perceptions now as immediate experiences (not as items existing in a uniquely actual spatio-temporal and causally interrelated nexus) which enables me to identify in thought my real world and distinguish it from other, non-actual, spatio-temporal series of which I can think.48 Reality is present in ideal form in the contents of these perceptions now, and in my logically interrelated thoughts, as a unique spatio-temporally extended system of objects and events. But these perceptions and thoughts do not merely have communicable ideal contents. They will be items in an indefinitely complex centre of immediate, non-relational experience or feeling. On Bradley's view, such centres of immediate experience are to be taken as epistemologically basic. And it is in virtue of my present perceptions being unique items in such a centre that I am able to distinguish in thought my real world (i.e. the one containing the intentional objects of these immediately felt perceptions and thoughts) from other thinkable non-actual series. According to Bradley's epistemology and metaphysics it is in such finite centres of experience, which cannot be thought of as locatable in any particular spatio-temporal series, that reality (i.e. the ultimate genuinely individual subject of our judgements) is immediately present; moreover, it is in virtue of being items in such a centre of immediate experience that my present perceptions and thoughts themselves are (a) unique and (b) provide, through their ideal contents, my fundamental point of cognitive contact with reality. Thus characterized, the notion of a finite centre is no 48 Bradley can be seen as giving an indexical account of actuality; for such an account cf. D. Lewis, 'Anselm and Actuality', Nous, Vol. 4 (1970), p. 184.
124
Kant and His Influence
doubt a difficult one to grasp.49 But, as T.S. Eliot suggested, finite centres can be thought of as in many ways analogous to Leibniz's monads.50 Like monads they are 'gradable' - from merely sentient to rational and self-conscious - non-relational centres of experience. And at least part of what the latter means is that they are not to be construed as deriving their individuality from being externally related, and in particular from being causally interrelated, within a spatio-temporally extended system of existents. So a finite centre of experience is (i) that within which the ultimate, genuinely individual, subject of all our judgements (true and false) is immediately present and (ii) that within which the genuinely individual can appear more or less adequately in ideal form (and thus be known more or less adequately depending on the coherence and comprehensiveness of the ideal contents exercised)51 as a spatio-temporally extended system of causally interrelated and enduring objects, animate and inanimate. However, finite centres are not themselves to be construed as ultimate subjects of predication. They are that in which the genuinely individual is present both immediately and in ideal form and thus they are not themselves to be construed as genuinely individual. Reality, then, for Bradley, is not to be construed (like Kant's things-in-themselves are) as something other than, or transcendent to, an indefinite multiplicity of finite centres of sentient experience.52 But the finite centres of 49 For the best characterization of the notion of a finite centre see Essays on Truth and Reality, XIV, 'What is the Real Julius Caesar?'. 50 'Leibniz's Monads and Bradley's Finite Centres', The Monist, XXVI (October 1916), pp. 566-76. 51 See Essays on Truth and Reality, VII, 'On Truth and Coherence' and VIII, 'Coherence and Contradiction'. 52 See ibid., IX, 'On Appearance, Error and Contradiction', pp. 245-50.
Thought and Sensibility in Leibniz, Kant and Bradley 125
experience themselves are not to be construed (like Leibniz's monads) as genuine individuals. Their mode of existence can ultimately be merely adjectival. But it is in and through the activities of finite centres of experience alone that the genuinely individual (the Absolute) can exist and realize itself.
Bibliography Bradley, F.H., Essays on Truth and Reality (Clarendon: Oxford, 1914). Bradley, F.H., Principles of Logic (Clarendon: Oxford, 1967) Vols I and II. Bradley, F.H., Appearance and Reality (Clarendon: Oxford, 1968). Kant, I., Critique of Pure Reason, trans N. Kemp Smith (Macmillan: London, 1964). Lewis, D., 'Anselm and Actuality', Nous, Vol. 4 (1970). Loemker, L. (ed.), Philosophical Papers and Letters (Reidel: Dordrecht, 1976). Manser, A. and Stock, G., The Philosophy of F.H. Bradley (Clarendon: Oxford, 1984). Parkinson, G.H.R. (ed.), Philosophical Writings (Dent: London, 1979). Russell, B., Logic and Knowledge, R.C. Marsh (ed.) (Allen and Unwin: London, 1956). Whitehead, A.N., Process and Reality (Cambridge University Press: Cambridge, 1929).
'ORIGINAL NONSENSE': ART AND GENIUS IN KANT'S AESTHETIC Peter Lewis
The words 'original nonsense' are taken from §46 of the Critique of Aesthetic Judgment in which Kant examines the nature of genius in relation to art. Kant offers no explanation of his use of the term 'nonsense'; at least, he does not in the immediate context of its first appearance. He uses it again in §50, almost at the end of the discussion of genius, but again without any definite elucidation. Most commentators, though not all, tend to repeat Kant's usage without comment, while subjecting almost every other concept in Kant's argument to exhaustive scrutiny. This practice suggests that what Kant means strikes most people as obvious, though so far as I can tell 'unsinn' is an unusual term for Kant to use at all, let alone in the third Critique. Of course, it cannot be understood without considering the other terms in Kant's discussion of art and genius: my hope is that it will be enlightening to confront them with questions about nonsense. According to Ernst Cassirer, Kant's account of genius stands 'at the crossroads of all aesthetic discussions in the eighteenth century'.1 Two of the roads in question here 1
Kant's Life and Thought (Yale University Press: New Haven, 1981), p. 320.
126
Art and Genius in Kant's Aesthetic
127
are the neo-classical emphasis on the rules of art and the romantic notion of genius as 'a power of producing excellencies, which are out of the reach of the rules of art'.2 Kant's presentation of the problem in §46 reflects his view of judgements of taste argued for earlier, that the judgement that something is beautiful is an aesthetic rather than a logical judgement (§1); that is, it concerns an individual's response to an object as opposed to an attribution of a property to an object. In judging that something is beautiful, I do not subsume the object under a concept, I do not apprehend the object in accordance with a rule for the application of a predicate. Rather, the determining ground of such a judgement is a feeling of pleasure resulting from the harmonious free play of the cognitive faculties of imagination and understanding. Judgements of taste, then, are not determinable by concepts (§35); or, as Kant sometimes puts it, 'There can . . . be no rule according to which any one is to be compelled to recognize anything as beautiful' (§8).3 This analysis runs into a problem when Kant considers the nature of art. For works of art are artefacts, which means to Kant that they are made in accordance with rules for the achievement of an end or goal specified in terms of a concept, and that in our appreciation of them we must recognize them as such, as works made for some end or goal. This leads to a dilemma: either we estimate works of art as works, in which case we violate the conditions for judging things to be beautiful, or we judge artworks to be beautiful, in which case we violate the conditions for estimating them as artefacts, as works of 1 J. Reynolds in Discourses on Art, Robert R. Wark (ed.) (Yale University Press: New Haven, 1981), VI, p. 96. 3 Unless stated otherwise, translations are from J.C. Meredith's edition of The Critique of Judgment (Clarendon: Oxford, 1969).
128
Kant and His Influence
art. Kant resolves the dilemma by employing the notion of genius. Works of fine art are products of genius, and 'genius is the innate mental aptitude through which nature gives the rule to art' (§46). Having introduced this definition, Kant raises the question whether it is adequate to the concept usually associated with the word 'genius'. I think it is clear that it is not, for it is not the case that everything we are prepared to call a work of art is a work of genius. Ordinarily, we think of genius as producing very special works of art, great works as opposed to merely good or mediocre works. This is how I will employ the expression 'product of genius', taking for granted that Kant has established a necessary connection between the concept of genius and the concept of fine art. Kant proceeds to elucidate his notion of genius in four numbered points, the first two of which are as follows. Genius (1) is a talent for producing that for which no definite rule can be given: and not an aptitude in the way of cleverness for what can be learned according to some rule; and . . . consequently originality must be its primary property. (2) Since there may also be original nonsense, its products must at the same time be models, i.e. be exemplary; and, consequently, though not themselves derived from imitation, they must serve that purpose for others, i.e. as a standard or rule of estimating. The combination of these two sentences seems to me to be a sufficient justification for Schopenhauer's reference in the first sentence of his doctoral thesis to 'the marvellous Kant'. They indicate the way in which Kant resolves the eighteenth century's conflict over genius and
Art and Genius in Kant's Aesthetic
129
rules in art. The products of genius are not the result of the application of rules, no matter how ingenious the rules or their applications: this means that the judgement of artistic beauty is not constrained by rule. However, the product of genius, the work of fine art, is not thereby arbitrary or accidental, for it must serve as a rule in relation to future work; that is, the artefact cannot count as a work of genius, as a work of fine art, if it cannot function as a rule within the arts. The notion of originality is itself a richly complex one, but at least part of what Kant means when he says that it must be the primary property of genius is novelty. 'Everyone is agreed', says Kant, 'on the point of the complete opposition between genius and the spirit of imitation' (§47). A work of genius is not an imitation of any existing work, it is something new, an invention, an innovation. Given this, then Kant's second point, that there can be original nonsense, seems eminently plausible. Being novel or new is certainly not a guarantee of being good or great. But I am intrigued by the fact that Kant does not say just this. He chooses to say that something new or novel may be nonsense. My question is, with what right does Kant employ such a term in this setting? What justifies him in saying this? One commentator who does not ignore this question is Timothy Gould, in his essay 'The Audience of Originality'.4 Gould explores a suggestion made by Ted Cohen that 'a metaphor may be the best available example of what Kant called products of genius'. Thus, a successful metaphor is to be construed as a new way of making sense: it cannot be made sense of according to 4
Timothy Gould, 'The Audience of Originality' in Essays in Kant's Aesthetics, T. Cohen and P. Guyer (eds) (University of Chicago Press: Chicago, 1985).
130
Kant and His Influence
existing rules and yet it does make sense, and in doing so extends our resources for making sense. Gould puts the point as follows: 'The provision of a sense not guaranteed by old rules is either exemplary - and followed, as though it were a new rule - or else it fails' (p. 185). An unsuccessful metaphor is, then, a failure in the attempt to create a new sense, and so is a candidate for the title of original nonsense: though, because of the associations with Lewis Carroll and errant metaphysicians, Gould prefers to talk of original senselessness (p. 186). He goes on: For reasons which I take to be internal to the category, it is hard to give examples. . . . The most conspicuous might be drawn from recent times, from the works or effects of Dada, of some surrealists, or of certain of the so-called minimalists {loc. cit.). Despite his sensitive treatment of the matter, Gould concludes that metaphor is not the best example of the products of genius, for a bad metaphor, in being banal or forced, is a less radical failure in making sense than failure in art (p. 187). I am in agreement with Gould on this point, but I also want to go further in rejecting the example of metaphor. Take a much-used illustration, Romeo's remark, 'Juliet is the sun'. Taken metaphorically, it is pregnant with sense; taken literally, it is a kind of nonsense, it is like a category mistake. On some theories, its being literal nonsense is a condition of its success as a metaphor. But this structure does not fit Kant's analysis of the products of genius. Kant's view is that, since there can be original nonsense, then the products of genius must be exemplary as well as novel. That is, as I read it, being exemplary excludes being nonsense, and vice versa. I see no ground here, at least, for thinking Kant would accept that from one point of
Art and Genius in Kant's Aesthetic
131
view works of fine art are nonsense and that, at the same time, from some other point of view they are exemplary. What is more, I do not believe that the example of metaphor, as elaborated by Gould, actually throws much light on the crucial question of what determines sense and nonsense in this context. Thus, we are told that a successful metaphor does not make sense in terms of existing rules. But this is also true of nonsense. The difference is that metaphor yields a new sense by providing a new rule, whereas nonsense does not. What needs answering here is what it is that enables us to establish this distinction, how we identify sense as opposed to nonsense. And one thing seems clear, which is that it is not by reference to rules of sense that we make this identification. But in that case we have not been given anything in this example of metaphor which advances our understanding of Kant's view of genius. Finally, I want to maintain that the example of metaphor draws attention away from what seems to me to be one of Kant's great insights about the nature of art. Metaphors do not, normally, relate to one another in the way that works of art do. Ordinary, everyday metaphors, as opposed to metaphors in works of art, are embedded in conversational contexts, taking as their point of departure the literal and non-metaphorical conventions of the language. Works of art, however, relate to one another within various artistic traditions. A new work of art takes as its point of departure previous achievements within the art form. It is precisely this which I think Kant expressed in the passage I have quoted about originality and exemplarity. Every work of art which is the product of genius stands in a twofold relation to other works - being original relates a work to its predecessors; being exemplary relates a work to its successors. Works of art, then, are essentially embedded
132
Kant and His Influence
in traditions constituted by works of art. This seems to be very different from the case of metaphor. Indeed, it would surely be extraordinary if Cohen's claim were correct; if, that is, metaphors, and not works of art, provided the best examples of Kant's account of the products of genius. Although in §49 Kant does employ figures of speech in illustration of the nature of aesthetic ideas, the generation of which is distinctive of artistic genius, this, as we shall see, is not to allow that metaphor constitutes a product of genius in the relevant sense. I now want to look more closely at Kant's idea that nature gives the rule to art. He points out in §47 that the artist cannot formulate a rule to serve as a precept for the construction of the work, for this would undermine the autonomy of the work and of the judgement of taste upon the work. 'Rather', says Kant, 'the rule must be abstracted from what the artist has done, i.e. from the product, which others may use to test their own talent, letting it serve them as their model, not to be copied but to be imitated. How that is possible is difficult to explain'.5 That there is a difficulty here can be seen from the clause 'not to be copied but to be imitated' - this appears to conflict with a point mentioned earlier, that genius is completely opposed to the spirit of imitation. In fact, Kant's difficulty here is mirrored in a curious muddle in the manuscript.6 Initially, the manuscript read 'not to be imitated but to be imitated (nicht der Nachahmung, sondern der Nachahmung)', being subsequently corrected to read 'not to be copied (Nachmachung) but to be imitated (Nachahmung)'. This is the reading adopted by the s 6
Critique of Judgment, Pluhar's translation (Hackett: Indianapolis, 1987). Ibid., p. 177, n. 43.
Art and Genius in Kant's Aesthetic
133
translators Bernard7 and Pluhar.8 However, Meredith introduces a conjecture at this point, his translation reading 'not for imitation (Nachahmung), but for following (Nachfolge)'. This brings the passage into line with what Kant says elsewhere in the Critique. At §32, he says, ''Following (Nachfolge) which has reference to a precedent, and not imitation (Nachahmung), is the proper expression for all influence which the products of an exemplary author may exert upon others. . .' At §49, \ . . the product of a genius . . . is an example, not for imitation . . . but to be followed by another genius . . .' I think this textual muddle partly reflects the complexity of the story Kant has to tell about the relationship of works of genius to future works, but also partly a lack of clarity in his thinking about that relationship. The complexity concerns the difference between the way in which works of genius relate to future geniuses and the way in which works of genius relate to future artists who lack genius. The lack of clarity concerns the numerous notions Kant employs to describe these differing relationships. He talks of abstracting (or gathering) a rule or rules from the products of genius, but whereas it makes sense to talk of following rules, it makes no sense to talk of imitating rules: that would explain the clause 'not for imitation but for following'. Kant also talks about the product of genius serving as a model, but whereas it makes sense to talk of imitating a model, it does not make much sense to talk of following a model: that would help to explain the clause, 'not to be copied but to be imitated'. Kant talks, too, about the product of genius being an example for others, and it makes sense to talk of following someone's example as well as 7 8
Critique, trans J.H. Bernard (Hafner Press: New York, 1951). Op. cit.
134
Kant and His Influence
imitating someone's example, though these perhaps do not necessarily amount to the same thing. It is with respect to this latter pair of expressions that we can begin to appreciate the difference between artistic genius and artistic non-genius. Given that genius is opposed to the spirit of imitation, we can say that genius follows the example of another genius whereas the non-genius imitates the example, the product of genius. Of course, neither goes in for slavish copying of previous artistic products, though the nongenius will imitate the style or the plot or the composition, and so on, found in earlier work. In this way, when a particular work or the work of a particular group of artists is taken as the example for imitation, there arises something like a movement or school. 'That is to say', says Kant, 'a methodical instruction according to rules, collected, so far as the circumstances admit, from such products of genius and their peculiarities. And, to that extent, fine art is for such persons a matter of imitation, for which nature, through the medium of a genius, gave the rule' (§49). One might question Kant's emphasis on rules extracted from the works themselves; not that I deny the possibility - the rules of harmony dictated in music schools would be one instance of the phenomenon. Nevertheless, imitation does not require the formulating or fixing of rules in that sense. Here, Kant's other notions - exemplars, models - come into play. Particular works of art can assume a normative function in so far as a teacher, or apprentice, refers to them as a guide in the making of his own work. 'Do it like thisV might be a typical command, accompanied by a pointing gesture. In this way, a work of art can serve as a rule without a rule having to be abstracted or extracted from it. The genius, qua genius, is supposed to be above or beyond all this. His work is brought about neither by
Art and Genius in Kant's Aesthetic
135
imitation nor by following rules. And yet he does follow the example set by previous products of genius. It is just this which, not surprisingly, Kant found difficult to describe. At one point he puts it like this: the product of a genius . . . is an example . . . to be followed by another genius - one whom it arouses to a sense of his own originality in putting freedom from the constraint of rules so into force in his art, that for art itself a new rule is won . . . (§49). The work of genius, then, inspires - to use a word familiar in this context (and one that Kant himself uses in §§48 and 50) - another artist to create another work of genius. In this respect, the work of genius plays a different role from that which it does for the non-genius. It is still exemplary, but what it exemplifies is not a rule or a style but an achievement. In Kant's term, it is a standard, a standard of excellence, a benchmark, a yardstick by reference to which subsequent achievements are measured. This is the position Beethoven's nine symphonies hold in the symphonic tradition. There is an interesting correspondence between the views just outlined and Wittgenstein's views as expressed in his 1938 lectures on aesthetics. Having talked about rules in art and about judgements of correctness, Wittgenstein says: When we talk of a Symphony of Beethoven we don't talk of correctness. Entirely different things enter. One wouldn't talk of appreciating the tremendous things in Art. In certain styles in Architecture a door is correct, and the thing is you appreciate it. But in the case of a Gothic Cathedral what we do is not at all to find it correct - it plays an entirely different role with us. The entire game is different. It is as different as to judge a
136
Kant and His Influence
human being and on the one hand to say 'He behaves well' and, on the other hand, 'He made a great impression on me.'9 It seems to me that these remarks, in themselves somewhat puzzling, are illuminated by Kant's account of works of genius which set standards of correctness but which themselves are not assessed in terms of correctness. Wittgenstein's analogy of the rules of etiquette emerges again in his reflections on his capacities as an artist. 'In my artistic activities,' he writes, 'I really have nothing but good manners'-™ and again, '. . . the house I built for Gretl is the product of a decidedly sensitive ear and good manners. . .' u Wittgenstein's etiquette analogy does suggest a potential difficulty for the Kantian account of genius. For it might well appear that the person who makes a great impression on us need pay no attention to the rules of etiquette - behaving well, conforming to the rules, is a matter for the meek, the faint-hearted. The great man, the genius, creates his own rules. This is a difficulty Kant explicitly confronts: his solution to it brings us closer to an understanding of what he might mean by original nonsense. Kant writes seeing that originality of talent is one (though not the sole) essential factor that goes to make up the character of genius, shallow minds fancy that the best evidence they can give of their being full-blown geniuses is by emancipating themselves from all academic constraint of rules, in the belief that one cuts 9
Lectures and Conversations on Aesthetics, Psychology and Religious Belief (Basil Blackwell: Oxford, 1966). 10 Culture and Value, trans P. Winch (Basil Blackwell: Oxford, 1980), p. 25. 11 Ibid., p. 38.
Art and Genius in Kant's Aesthetic
137
a finer figure on the back of an ill-tempered than of a trained horse' (§47; cf. §49). Kant's image of cutting a fine figure nicely parallels Wittgenstein's analogy of a person making a great impression: neither amounts to mere conformity but equally neither can be achieved by disregarding all rules. As Wittgenstein put it in reply to an objection, '. . . every composer changed the rules, but the variation was very slight; not all the rules were changed. The music was still good by a great many of the old rules . . .'12 (The context of this remark makes clear that Wittgenstein would have been prepared to accept the qualification, 'every great composer'.) Here we need to remember a point I made at the beginning, that works of art are works, hence, in Kant's view, made in accordance with rules. He endorses this in §47: Even though mechanical and fine art are very different from each other, since the first is based merely on diligence and learning but the second on genius, yet there is no fine art that does not have as its essential condition something mechanical, which can be encompassed by rules and complied with, and hence has an element of academic correctness (Pluhar's translation). To accommodate this, Kant makes two crucial distinctions, the first between the material and the form of art. 'Genius can do no more than furnish rich material for products of fine art; its elaboration and its form require a talent academically trained, so that it may be employed in such a way as to stand the test of judgment' (§47). Corresponding to this distinction, there is the further 12
Lectures and Conversations, p. 6.
138
Kant and His Influence
distinction between genius and taste. Genius is a productive faculty or ability, creating new ideas; whereas taste is a critical rather than a productive faculty (§48). Taste . . . is the discipline of genius. It severely clips its wings, and makes it orderly and polished . . .' (§50). We can unite these two distinctions by recognizing that taste is the ability, which may be acquired by learning, to mould the form of the work of art in accordance with the academic constraint of rules. There is a remarkable similarity between Kant and Wittgenstein on this topic of genius and taste. In Culture and Value Wittgenstein thinks of genius in terms of originality, of inventiveness,13 and contrasts this with taste. The faculty of 'taste' [says Wittgenstein] cannot create a new structure, it can only make adjustments to one that already exists. Taste loosens and tightens screws, it does not build a new piece of machinery. . . . Giving birth is not its affair. Taste makes things ACCEPTABLE. . . . Even the most refined taste has nothing to do with creative power.14 And in an image reminiscent of Kant's trained horse, Wittgenstein says that 'within all great art there is a WILD animal: tamed'.15 This seems to imply that for Wittgenstein great art involves both originality and taste, but this is explicitly denied in a later entry, '. . . I believe that a great creator has no need of taste; his child is born into the world fully formed'.16 If Wittgenstein is equivocal on this issue, Kant is not. 13
Op. cit., p. 36; pp. 18-19. Ibid., pp. 59-60. 15 Ibid., p. 37. " Ibid., p. 59. 14
Art and Genius in Kant's Aesthetic
139
[Ilnsofar as art shows genius it does indeed deserve to be called inspired, but it deserves to be called fine art only in so far as it shows taste. Hence what we must look to above all, when we judge art as fine art, is taste, at least as an indispensable condition (§50; Pluhar's translation). This makes clear that when talking of the products of genius as exemplary, Kant has in mind works which possess both genius and taste, genius in the context of this latter contrast being understood more narrowly as the material of the work (or the soul of the work, as Kant says at one point, §49). Figures of speech, such as metaphor, occurring within works of art, would count as products of genius in this narrow sense in so far as they exhibit aesthetic ideas; but it is only to the work of fine art as a whole, to the product of genius in the wider or fuller sense, that Kant attributes the tradition-embedded features of originality and exemplarity. As we have seen, not all works of art are works of genius; thus, Kant allows that 'in order [for a work] to be beautiful, it is not strictly necessary that it be rich and original in ideas' (§50; Pluhar's translation). There may, then, be works which exhibit taste but not genius. Since taste is a necessary condition of fine art, it seems that there cannot be works which exhibit genius but not taste. In §48, Kant appears to confirm that this is so when he says that 'in a would-be work of fine art we may frequently recognize genius without taste and in another taste without genius'. In fact, 'genius without taste' is a possible candidate for what is meant by 'original nonsense' - a soul without a body, as it were (see §43). It appears to complete the categorization which Kant employs - that is, works of
140
Kant and His Influence
genius, works of non-genius, and original nonsense. Broken down into 'faculties', that yields 'genius and taste', 'taste without genius', and 'genius without taste'. But this is too simple: there is another kind of work which Kant refers to and which is also a candidate for what is meant by 'original nonsense', viz. the work of a bungler (Pluhar) or tyro (Meredith) which exhibits neither genius nor taste (see §§47 and 49). Such work is marked by idiosyncrasy and eccentricity in consequence of the maker's attempt to ape the originality of genius. But whereas in the product of genius 'a deviation from the common rule' is a merit, and shows 'courage', since it . is demanded of the artist in making something 'appropriate' or 'adequate' to his idea (§49), in the work of the bungler it is merely ridiculous, empty-headed (§47), attention-seeking (§49). Equally, though an action which in one context may show courage, an outwardly similar action in a different context may be simply foolhardy. There is a strong case for taking the work of a bungler to be what Kant means by 'original nonsense'. Nevertheless, there is also a case to be made for 'genius without taste' in view of the fact that Kant uses the distinctive word 'nonsense' once more, and only once more, at just the point where he should if this is what he meant by 'original nonsense'. In §50, having claimed that taste is indispensable, he goes on to say that 'in lawless freedom imagination, with all its wealth, produces nothing but nonsense; the power of judgement, on the other hand, is the faculty that makes it consonant with the understanding'. If taste, exercised in judgement, is what brings lawfulness into the ideas of imagination and so makes possible a work of fine art, a product of genius in the full sense, then the absence of taste will result in a work potentially rich in ideas but disorganized to the
Art and Genius in Kant's Aesthetic
141
point of unintelligibility. The proposal here is intuitively appealing. It is the notion of an artist whose work is imaginative and full of invention, though too full, exhibiting no evident consistency or coherence, or it is arbitrary and whimsical, or it is one in which we might think of the artist as having lost control of his material, as being overwhelmed by it all. 'If only', we think, 'he had the discipline, the strength of character and will to bring order out of this chaos.' ('One might say', wrote Wittgenstein, '"Genius is talent exercised with courage"?)17 To quote again a sentence I used earlier, now continuing the passage: Taste, like judgment in general, is the discipline (or corrective) of genius. It severely clips its wings, and makes it orderly or polished; but at the same time it gives it guidance, directing and controlling its flight, so that it may preserve its character of purposiveness. It introduces a clearness and order into the plenitude of thought, and in so doing gives stability to the ideas, and qualifies them at once for permanent and universal approval, for being followed by others, and for a continually progressive culture (§50). Without the discipline of taste, the work may be original but nonsensical. If, as I have suggested, there are two plausible candidates for the title of 'original nonsense', I see no need to eliminate one in favour of the other. The distinction between different kinds of nonsense - paradoxical as that sounds - corresponds to a distinction between different kinds of artistic failure. One is irredeemably vacuous, the result of naive incompetence or outrageous effrontery; the other is a matter of unrealized potential, r
Culture and Value, p. 38.
142
Kant and His Influence
a genuine attempt at artistic innovation which can be taken seriously and from which something can be learned - in short, an exemplary failure. Earlier, I quoted Timothy Gould to the effect that it is hard to give examples of original nonsense for reasons which are internal to the category. I am not sure what is meant by this, unless it is just that what we need to look for are examples of failures, albeit perhaps heroic failures, whereas it tends to be successful works which are deposited on, and not submerged by, the sands of time. Gould's own examples of artistic nonsense are controversial, and inevitably so; though it seems to me that the works of Dada, surrealism and minimalism all too often lack the kind of complexity and potential which I have characterized as belonging to genius without taste. As an example of that latter category, no less controversial, I offer Joyce's Finnegan's Wake. If there is a shortage of examples of original nonsense, there is, ironically, no shortage of works which appear, or did once appear, to be original nonsense. It was in these terms that people, critics as well as the general public, reacted to artistic innovations throughout the nineteenth century - to Turner's paintings for instance, to the first impressionist exhibitions, to the post-impressionists, etc. These works often struck their first spectators to be the work of bunglers, patternless, pointless, unruly and chaotic displays of colour. Even if it is not the fate of genius to be misunderstood, on Kant's view it is not surprising that it often is, at least to begin with. The appearance of nonsense is dissipated, if at all, through the exercise of taste, the critical faculty. This involves not merely discerning some kind of order or pattern in the work there is a fine illustration of this in relation to a Jackson Pollock painting in David Bell's recent paper,
Art and Genius in Kant's Aesthetic 143
The Art of Judgment'18 - but also seeing the work as continuous with accepted examples of the relevant kinds of art. Thus, Turner was made sense of by Ruskin's placing him within the landscape tradition running from seventeenth-century Dutch painting to Constable. I have reverted to talking, as I did at the beginning of this paper, of taste as exercised by the audience of art, whereas in the middle of the paper I talked of taste as an aspect of artistic capability. This reflects the variation in Kant's way of talking about taste as between the Analytic of the Beautiful and the sections on genius in the Analytic of the Sublime. If this looks like inconsistency, then it needs to be remembered that in estimating works of art we are involved in making judgements of dependent beauty rather than judgements of free beauty (§§16 and 45; cf. §48). This means that in judging a work to be beautiful, to be a work of fine art, 'A concept of what the thing is intended to be must . . . be laid at its basis' (§48), a concept which guided the production of the work. At its most general, that concept is 'work of fine art', but that in turn will supervene on more specific concepts such as 'painting', 'literary work', 'sculpture', 'piece of music'; and, more specifically still, 'poem', 'novel', 'landscape', 'portrait', and so on. These concepts do not fully determine our judgement of the work, any more than they do its production: '. . . a mode, as it were, of execution, in respect of which one remains to a certain extent free, notwithstanding being otherwise tied down to a definite end' (§48). The explanation of this is that our grasp of such concepts, both for the artist and the audience, is given through examples. . . .[T]he artist, [says Kant] having practised and II
Mind (1987), pp. 236-7.
144
Kant and His Influence
corrected his taste by a variety of examples from nature or art, controls his work and, after many, and often laborious attempts to satisfy taste, finds the form which commends itself to him (§48). Similarly, the exercise of taste in judging art will involve coming to see a work as a further example, congruent with the old, in which the achieved form is 'adequate' or 'appropriate' (§48) to the rich and original play of the artist's ideas. Our inability to see a work in this way, to see how it can fit in, leaves us with the judgement of original nonsense. My emphasis on seeing a work of art in relation to past works might appear to be in conflict with Kant's view of a work art as exemplary, serving as a rule for future following. But there is no real conflict. A rule points forward in virtue of the way it organizes (our sense of) the past. And it is entirely characteristic of great art that it changes the ways in which we appreciate the art of the past. The point is made superbly by F.R. Leavis in his comments on Jane Austen: She not only makes tradition for those coming after, but her achievement has for us a retroactive effect: as we look back beyond her we see in what goes before, and see because of her, potentialities and significances brought out in such a way, that for us, she creates the tradition we see leading down to her.19 It is as if Jane Austen's work provides a rule for a new ordering of 'the existing monuments' (to use Eliot's famous phrase), an ordering which itself will influence, and be modified by, future literary innovations. Now it might seem that Kant's notion of genius giving 19
The Great Tradition (Penguin: Harmondsworth, 1977), p. 14.
Art and Genius in Kant's Aesthetic
145
the rule to art could be undermined by Kripkean rulescepticism. After all, it might be said, any one work can be thought of as following or fitting in with any other; and, if this is so, then the distinction between artistic sense and nonsense would appear to be arbitrary. However, the ground of this distinction in Kant's aesthetic is his account of the human cognitive faculties and of their harmonious interplay. Not just anything can count as a beautiful work of art; although, crucially, what is and what is not cannot be specified in advance of particular human responses. In the last analysis, Kant's 'continually progressive culture' (§50) rests on 'agreement in judgments'.
Bibliography Bell, David, The Art of Judgment', Mind (1987). Cassirer, Ernst, Kant's Life and Thought (Yale University Press: New Haven, 1981). Gould, Timothy, 'The Audience of Originality' in Essays in Kant's Aesthetics, T. Cohen and P. Guyer (eds) (University of Chicago Press: Chicago, 1985). Kant, I., Critique of Judgment, trans J.H. Bernard (Hafner Press: New York, 1951); trans Werner S. Pluhar (Hackett: Indianapolis, 1987). Kant, I., Critique of Aesthetic Judgment, trans J.C. Meredith (Clarendon: Oxford, 1969). Leavis, F.R., The Great Tradition (Penguin: Harmondsworth, 1977). Reynolds, J. in Discourses on Art, Robert R. Wark (ed.) (Yale University Press: New Haven, 1981). Wittgenstein, L., Lectures and Conversations on Aesthetics, Psychology and Religious Belief (Basil Blackwell: Oxford, 1966). Wittgenstein, L., Culture and Value, trans P. Winch (Basil Blackwell: Oxford, 1980).
FICHTE, BECK AND SCHELLING IN KANT'S OPUS POSTUMUM Eckart Forster
I have taken the title of the conference - 'Kant and His Influence' - quite literally. That is to say, I have taken the 'and' in the title as a conjunctive, not as a disjunctive. Consequently, I want to raise the question: how did Kant respond philosophically to the way others reacted to his philosophy? For while others were struggling with the implications of the critical philosophy he had initiated, Kant was too, virtually until his death. When he died, Kant left behind a voluminous manuscript of several hundred pages - the so-called Opus postumum - on which he had worked for more than a decade. This is, of course, the same period during which disciples and critics, such as Reinhold, Fichte, Beck, Schelling, Maimon, and others, subjected the critical philosophy to a careful examination. Interestingly, we encounter references to Fichte and Beck as well as Schelling in Kant's Opus postumum. To what extent was Kant, in his last work, aware of, or even influenced by, these philosophers? I must begin by saying a few things about the Opus postumum itself. Although the manuscript is virtually complete, Kant did not live to edit it. Ignoring Kant's repeated assertion that the text would be his most important work, and that it would fill 'a gap' in his critical 146
Fichte, Beck and Schelling in Kant's 'Opus postumum' 147
philosophy, his literary executor thought the text not fit for publication, with the result that it soon disappeared in the possession of Kant's heirs. When it resurfaced half a century later, influential philosophers like Kuno Fischer thought they could dismiss it without inspection as a product of senility - after all, had Kant not himself completed the critical philosophy with his Critique of Judgment} But more sympathetic thinkers, too, found it difficult to make sense of Kant's text, for the various sheets and fascicles of the manuscript were not preserved in the order of their composition, making it seemingly impossible to determine the chronological (and logical) order of Kant's reasoning. Eventually, in 1936-8, the complete text was published. When the war was over, Kant's last work slowly began to attract the philosophical attention one would expect, with translations of it being published in French, Italian, Spanish, and an English translation forthcoming. What did Kant hope to achieve in his last work, a work which he initially planned to call 'Transition from the Metaphysical Foundations of Natural Science to Physics'? Why did he think, so late in his life, that another critical work was needed? I will offer an answer to these questions and then give an overview of the content of the Opus postumum, before turning to the discussion of Fichte, Beck and Schelling.1 1 In section I, I am summarizing what I have developed in more detail in a number of publications, including 'Is There "A Gap" in Kant's Critical System?', Journal of the History of Philosophy, XXV, 4 (1987), pp. 533-55; 'Kant's Selbstsetzungslehre in Kant's Transcendental Deductions. The Three Critiques and the Opus postumum, Eckart Forster (ed.) (Stanford University Press: Stanford, 1989), pp. 217-38; 'Kant's Notion of Philosophy', The Monist, 72, 2 (1989), pp. 285-304; 'Die Idee des Ubergangs. Uberlegungen zum Elementarsystem der bewegenden Krafte', in Ubergang. Beitrage zur Spdtphilosophie Immanuel Kants, Siegfried Blasche (ed.) (Vittorio Klostermann: Frankfurt, forthcoming); and in the Introduction to my English edition of Kant's Opus postumum (Cambridge University Press: Cambridge, forthcoming).
148
Kant and His Influence
I We may take our clue from a well-known passage in the Preface to the second edition of the Critique of Pure Reason. It describes the conceptual revolution that set the study of nature 'on the secure path of a science': When Galileo caused balls, the weights of which he had himself previously determined, to roll down an inclined plane . . . a light broke upon all students of nature. They learned that reason has insight only into that which it produces after a plan of its own, and that it must not allow itself to be kept, as it were, in nature's leading-strings, but must itself show the way with principles of judgment based upon fixed laws, constraining nature to give answer to questions of reason's own determining (B xii-xiii). The light that broke upon the students of nature was the realization that nature does not by itself reveal its laws to the observer. No unaided observation suggests that all bodies fall with the same constant acceleration. To discover a law like this, it is necessary to isolate in an experiment certain features from their natural environment, as Galileo had done when he let balls roll down an inclined plane. The experiment, the observer's interference with nature's processes, is a condition under which a scientific study of nature is possible. But more is required for a genuine science of nature. [P]hysics . . . owes the beneficent revolution in its point of view entirely to the happy thought, that while reason must seek in nature, not fictitiously ascribe to it, whatever as not being knowable through reason's own resources has to be learnt, if learnt at all, only from nature, it must adopt as its guide, in so seeking, that which it has itself put into nature (B xiii-xiv).
Fichte, Beck and Schelling in Kant's 'Opus postumum'
149
Any science worthy of the name must exhibit, or guarantee, a systematic connection of its various laws and propositions, and it must do so in accordance with determinate principles. Such systematic unity cannot be gained empirically; it is of a priori origin and presupposes a systematic activity of the subject. No aggregate of perceptions, no mere collection of empirical data, can yield the systematicity we seek in a science. Reason 'must itself show the way with principles of judgement based upon fixed laws'. Hence, if physics is to be possible as a science, philosophy must provide principles for the investigation of nature that yield a priori topoi for the systematic classification of those specific forces of matter that can only be given empirically. The subject must 'adopt as its guide' in seeking to learn from nature 'that which it has itself put into nature'. More specifically, the systematicity of physics rests on two requirements: (1) that we bring together in a systematic way the various laws of nature; (2) that nature permit such classification, in other words, that nature can be regarded as systematic. Now, (1) presupposes (2), for, as Kant had put it in the first Introduction to the third Critique: it is clear that the nature of the reflective judgment is such that it cannot undertake to classify the whole of nature by its empirical differentiation unless it assumes that nature itself specifies its transcendental laws by some principle.2 The Critique of Judgment provided the a priori justification for the assumption that nature specifies its universal laws to empirical ones, according to the form of a 2 Academy edition of Kant's Works (Berlin, 1900-), Vol. 20, p. 215; trans J. Haden (The Bobbs-Merrill Company: Indianapolis, 1965), p. 20.
150
Kant and His Influence
system.3 But it remained silent as to (1). What still needed to be elucidated, then, was exactly how the physicist must initially approach nature in order to be instructed by it, or what kind of a priori principles one must 'put into nature' in order to achieve a systematic understanding of its laws. For this task, the categories and Analogies of the first Critique do not suffice; nor is it enough to analyse the concept of matter in accordance with the table of categories, as the Metaphysical Foundations of Natural Science had done. Physics does not deal with nature uberhaupt, nor with the 'separated (although in itself empirical) concept of matter',4 but with the specific corporeal nature of outer sense. Hence a new work was required, a 'Transition from the Metaphysical Foundations of Natural Science to Physics': The transition from the metaphysical foundations of natural science to physics consists in a method of bringing about the systematic knowledge of physics; this cannot be done simply by collecting experiences, for here the outline [Vorriss] of a system is missing, which must be given a priori.5 If the transition project is relatively clear, its solution certainly is not. Kant's manuscript reveals his long and painful struggle with the problem; in a letter of 1798 to Christian Garve he even writes of 'a pain like that of Tantalus' of seeing before him 'the unpaid bill of my uncompleted philosophy'. What, exactly, is the source of the problem? To account for its systematicity (possibility), physics ' Ibid. Vol. 4, p. 472. 'Vol. 21, p. 492.
4
Fichte, Beck and Schelling in Kant's 'Opus postumum'
151
must be preceded by a special a priori science, the 'science of transition'. But this science of transition, in turn, requires an idea or a plan according to which it is to be carried out, for, as Kant had put it in the first Critique, 'No one attempts to establish a science unless he has an idea upon which to base it' (A 834). But what could function as such an idea for the science of transition? This idea cannot be derived from physics itself, as the transition is supposed to account for the very possibility of physics. Nor can it be derived from the Metaphysical Foundations of Natural Science, from which the transition commences: the concepts of attraction and repulsion, as Kant wrote later, furnish no specifically determined, empirical properties, and one can imagine no specific [forces], of which one could know whether they exist in nature, or whether their existence be demonstrable.6 For a while, Kant hoped to achieve the desired systematic result by 'following the clue given by the categories and bringing into play the moving forces of matter according to their quantity, quality, relation and modality'. But, as the early parts of Kant's manuscript show clearly, the system of categories cannot compensate for the lack of a proper idea on which to base the transition; it cannot itself determine which moving forces of matter must be connected a priori in order to advance to physics as a system. After several years, in 1799, a solution eventually suggested itself. It is reflected in the unique status Kant now assigned to the concept of an ether, which had initially been introduced in the Opus postumum to explain a number of physical phenomena. For example, ' Vol. 22, p. 282.
152
Kant and His Influence
Kant had realized that the dynamical theory of matter of the Metaphysical Foundations of Natural Science, which explains matter in terms of an interplay of attractive and repulsive forces, cannot explain the formation of material bodies. It may explain the differences in density in different types of matter, but it cannot account for cohesion and hence for matter of a particular form - a body (hence it cannot account for mechanics). The possibility of cohesion requires the impact of an all-penetrating, internally moving material, which Kant calls the 'caloric', or 'ether'. But eventually Kant also realized that empirical space, in order to be sensible, has to be thought of as being filled with a continuum of forces which 'as it were . . . hypostatizes' space.7 The ether is thus also referred to as 'hypostatized space'. The space known from the first Critiquey by contrast, the mere form of outer intuition, was 'neither positively empty nor positively full, [hence] not an object existing outside me at all'.8 Finally, Kant argued, since perceptions must be thought of as the effects of moving forces on my subject, the unity of experience, which reason demands a priori and unequivocally - there is only 'one experience' (A 110) - is thinkable only on condition of a unitary basis of all moving forces, or an ether. Kant's ether, in the Opus postumum, thus eventually becomes a necessary concept of reason, a transcendental ideal. But if it is a concept of reason, if it is an ideal of reason, it must be possible to determine analytically, and in accordance with the table of categories, those attributes that necessarily pertain to the ether in virtue of its function (cf. A 580/B 610f). To put it briefly, since the ether can only be one, it must be universally distributed (quantity). 7 8
Vol. 2 1 , p. 224. Vol. 2 1 , p. 232.
Fichte, Beck and Schelling in Kant's 'Opus postumum
153
In order to function as the basis for the connection of all the moving forces of matter, it must be all-penetrating (quality). Since its internal vibrations are to be the condition of the formation of material bodies (cohesion), it must be all-moving (relation). Finally, since the 'unity of experience' permits no interruptions, the internal motions of the ether must be permanent (modality).9 Expressed negatively, the ether must be imponderable, incoercible, incohesible, and inexhaustible. And this, Kant realized, can yield a basic principle for the systematic apprehension of the moving forces of matter, for 'ponderability, coercibility, cohesion, and exhaustibility presuppose moving forces which act in opposition to the latter and cancel their effect'.10 For there to be an object of experience in space, then, we must presuppose a priori such limitations of the ether - limitations, that is, which make possible the constitution of an object of outer sense. We thus have an 'idea' or principle for the elementary system of the transition, and an a priori topic for the empirical forces: as regards quantity, all matter must be either ponderable or imponderable; as regards quality, either coercible or incoercible; as regards relation, cohesible or incohesible; as regards modality, exhaustible or inexhaustible. The forces themselves remain 'problematic' at this level, as Kant emphasizes repeatedly: they can only be given empirically. The systematic account of them, if it is to be a priori, must thus be expressed in the form of disjunctions. At virtually the same time at which Kant finds this solution, he realizes its inadequacy. He had indeed found a principle for the elementary system of the moving "Cf. Vol. 21, p. 584. Vol. 22, p. 610.
111
154
Kant and His Influence
forces of matter, but the question, 'How are these forces to be experienced?' had thus far been left unaddressed. Specific empirical forces cannot be known through reflection, nor are they simply given to a passive subject. Empirical forces are recognized only by way of an active interference with them: We would not know through experience the moving forces of matter in bodies, if we were not conscious of our own activity to exercise ourselves the acts of repulsion and attraction through which we apprehend these appearances. The concept of originally moving forces is not derived from experience but must lie a priori in the activity of the mind of which we are conscious when moving.11 To experience the moving forces of matter, then, the subject of experience must itself exist as a corporeal subject in space and in time; it must first constitute itself as an object or, as Kant prefers to say in the Opus postumum, the subject must first posit itself (as an object). How is this to be understood from the point of view of his transcendental idealism? We can only be affected by moving forces of matter if we exist as corporeal beings in space. And only because we ourselves exercise acts of attraction and repulsion do we apprehend the appearances of moving forces upon us. On the other hand, it is only in the process of such apprehension of moving forces that we appear to ourselves as corporeal. Only in so far as I can represent myself as affected do I appear to myself as sensuous and corporeal, that is, as an object of outer sense. Selfpositing and affection through outer objects must therefore really be two sides of the same coin: 11
Vol. 21, p. 490.
Fichte, Beck and Schelling in Kant's 'Opus postumum' 155
Positing and perception, spontaneity and receptivity, the objective and the subjective relation, are simultaneous, because they are identical as to time, as appearances of how the subject is affected - thus [they are] given a priori in the same actus}1 Hence it is from this original actus of the subject that there emerges the duality of empirical self and material world surrounding it, of observer and observed. Only because I apprehend the undetermined given manifold and, in the process of apprehension, insert or put into this manifold certain fundamental forces, can I represent the manifold as the appearance of an external cause of my perception, and at the same time represent myself as being affected, hence as corporeal: The subject affects itself and becomes an object in appearance for itself in the composition of the moving forces. . . . the moving forces of matter are what the subject does with its body to [other] bodies. The reactions corresponding to these forces are contained in the simple acts by which we perceive the bodies themselves.13 In these simple acts, in what it puts into nature, the subject is guided by the principle of the elementary system of forces, as established in the analysis of the idea of the ether as the unitary material ground of all outer experience. The determination of my own existence in space and time, the transition from pure apperception to knowledge of myself as an empirical being, thus takes place for Kant within the context of the ideal of a single, all-embracing experience. This experience itself depends on the collective unity of the moving forces of matter, 12 11
Vol. 22, p. 466. Vol. 22, p. 364; Vol. 22, p. 326.
156
Kant and His Influence
which the subject investigates, progressing to a thoroughgoing determination of all phenomena, a process never completed but inevitably given as a task: The understanding begins with the consciousness of itself {apperceptio) and performs thereby a logical act. To this the manifold of outer and inner intuition attaches itself serially, and the subject makes itself into an object in a limitless sequence . . . I am an object of myself and of my representations. That there is something else outside me is my own product. I make myself . . . . We make everything ourselves.14 Before proceeding to Fichte, Beck and Schelling, I want to mention briefly that the position sketched here is not yet Kant's final one. Rather, the last fascicles of the Opus postumum contain an extension of the theory of selfpositing to the moral-practical realm. For thoroughgoing determination of my existence in space and time is not the only thoroughgoing determination of myself. As a person, I am also endowed with a will which must be determined in accordance with the moral law of reason, a will, that is, which is subject to the categorical imperative. It is thus appropriate to speak of a moralpractical self-positing, analogous to the theoretical, technical-practical one. 'Man himself must make or have made himself into whatever, in a moral sense, he is or is to become'. Thus Kant had already written several years earlier, in the Religion Within the Limits of Reason Alone. He now expands on this. His main concern, however, is with the idea which practical reason generates in the process of moral-practical self-positing: the idea, or ideal, of God. Practical reason inevitably creates this ideal, 14
Vol. 22, p. 82.
Fichte, Beck and Schelling in Kant's 'Opus postumum' 157
Kant argues, in order to constitute itself as a moral person - just as theoretical reason creates for itself the idea of a world (the ether) as a collective, not merely distributive, unity in the light of which it constitutes itself as a corporeal being in space and time. This, finally, allows Kant to unify the two fundamental branches of his philosophy, namely nature and freedom, in the new theory of transcendental philosophy as Selbstsetzungslehre: Transcendental-philosophy is the subjective principle of the united theoretical-speculative and moralpractical reason in one system of ideas . . . through which the subject constitutes itself into an object.15 Not surprisingly, at this stage the original title, 'Transition from the Metaphysical Foundations of Natural Science', no longer seemed appropriate to Kant for the text at hand. Thus, in the last fascicle of the Opus postumum he tries out various new titles for his last work. Here are three such versions: 'System of Transcendental Philosophy in three Divisions: God, the world (universum) and I myself (man) as a moral being' 'The Highest Standpoint of Transcendental Philosophy in the System of the Two Ideas (God, the World, and the thinking Subject that connects both)' 'Philosophy as Wissenschaftslehre in a Complete System, by I. Kant'. Anyone vaguely familiar with the thoughts of Schelling, Beck and Fichte will immediately recognize the allusion, 15
Vol. 2 1 , p. 67.
158
Kant and His Influence
in these titles, to the main works of these three philosophers: to Schelling's System of Transcendental Idealism, to Beck's The Only Possible Standpoint, and to Fichte's Wissenschaftslehre. And it is interesting that we also encounter other references to Fichte, Beck and Schelling in the Opus postumum. To what extent was Kant influenced, in his last work, by these philosophers?
II Fichte Fichte had studied with Kant in 1791, and it was Kant who had recommended Fichte's first manuscript to his own publisher. The resulting book made Fichte famous overnight. Fichte remained loyal to Kant for quite some time, and for years he continued to send his books to Kant. It would be natural to expect Fichte to have had an influence on Kant's thinking especially in connection with his theory of self-positing. E. Adickes, H J . de Vleeschauwer, and others have thought that this very theory on Kant's part was a concession to Fichte, and a desperate attempt to unify and consolidate his disintegrating school. I think this is mistaken. In fact, I think this theory has roots that are quite independent of Fichte, and that in fact pre-date Fichte's own philosophical beginnings.16 When Kant discusses self-positing in the Opus postumum, there is no mention of Fichte. Fichte is, indeed, not mentioned by name at all in this work, but there is a passage in the second fascicle (sheet VI, page 1) that is a draft for Kant's Open Letter on Fichte's Wissenschaftslehre: A Wissenschaftslehre in general, in which one abstracts " Cf. my 'Kant's Selbstsetzungslehre", op. cit., p. 217f.
Fichte, Beck and Schelling in Kant's 'Opus postumum'
159
from its matter (the objects of knowledge), is pure logic; and to imagine beyond it another, higher and more general Wissenschaftslehre (which, however, can itself contain nothing other than the scientific element of knowledge in general - its form) is, conceptually, to chase one's own tail. In the Open Letter, this passage becomes: I hereby declare that I regard Fichte's Wissenschaftslehre as a totally indefensible system. For the pure science of knowledge is nothing more nor less than mere logic, and the principles of logic cannot lead to any material knowledge. Since logic, that is to say, pure logic, abstracts from the content of knowledge, the attempt to cull a real object out of logic is a vain effort and therefore a thing that no one has ever done.17 This declaration against Fichte is often treated as Kant's general dismissal of his idealistic successors, but this, I think, cannot be right. There are too many peculiarities in the Open Letter to permit us to take it at face value. It was initiated by a review, in the Erlanger Literatur Zeitung, of Johann Gottlieb Buhle's Entwurf der Transcendental Philosophie, in which the reviewer states that Kant, in his Critique, had only sketched the plan for a transcendental philosophy, and that Fichte was the first to have carried out this plan systematically, and hence was the first transcendental philosopher. Kant is challenged to express his opinion of Fichte's Wissenschaftslehre. I have always found it extremely puzzling that Kant 17 I. Kant, Philosophical Correspondence 1759-1799, A. Zweig (ed.) (University of Chicago Press: Chicago, 1967), p. 253.
160
Kant and His Influence
should have written this Open Letter at all. Firstly, it took him a long time to respond. The Open Letter was written more than half a year after the review. Secondly, Kant had admitted in a letter to Tieftrunk that he had not read the Wissenschaftslehre\ what he knew of the book he knew from a review of it in the Jena Allgemeine LiteraturZeitung. Thirdly, what are we to make of Kant's claim, in his Open Letter, that i took the completeness of pure philosophy within the Critique of Pure Reason to be the best indication of the truth of my work'? What completeness? For shortly before the reviewer of Buhle's book issued his challenge, Kant had written to Christian Garve that he felt 'a pain like that of Tantalus' of seeing before him 'the unpaid bill of my uncompleted philosophy. . . . The project on which I am now working [the Opus postumum] must be completed . . . or else a gap will remain in the critical philosophy'.18 One can hardly take the Open Letter at face value. Is there another way in which it might be interpreted? It is important to note that Kant had received a similar challenge two years earlier, in May 1797, this time from the German physiocrat Johann August Schlettwein. Schlettwein insisted that Kant take a side in the ongoing dispute as to who among Kant's followers had best understood the spirit of the critical philosophy, 'whether it is Reinhold, or Fichte, or Beck, or whoever'.19 On 29 May 1797, Kant published an open letter to Schlettwein. In response to the question as to who had understood Kant's texts the way he wanted them to be understood, Kant here wrote: I answer without hesitation: it is the local court 18
Vol. 12, p. 257, my italics (Zweig, p. 251). " Vol. 12, p. 364.
Fichte, Beck and Rebelling in Kant's 'Opus postumum'
161
chaplain and professor of mathematics, Herr Schulz, whose writings about the critical system, entitled: Prufung etc., Mr Schlettwein should consult. But I request that he bear the following in mind: that the court chaplain's words are to be taken according to the letter, not according to a spirit allegedly lying in them (since one can read into words whatever one pleases).20 The distinction between the spirit and the letter of the Kantian system, to which both Kant and Schlettwein allude, had been brought into the contemporary philosophical discussion by Fichte. It had quickly gained notoriety as a means of distinguishing Kantians into rival groups. Fichte himself, of course, belonged to those who followed the spirit, rather than the letter, of the Kantian system. To the latter group belonged - allegedly - Kant's colleague Johann Schulz, whom Kant had once described as the best mind in the area, and whose two volumes of Prufung der Kantischen Critik der reinen Vernunft (1789, 1792) had met with Kant's highest approval. Himself too old and too busy to engage in public disputes about his philosophy, Kant was eager to see Schulz take on that role for him, and above all to continue his Prufungen of the Kantian system. Yet, it seems, Schulz was reluctant to do so; especially, the charge of being a Buchstabenphilosoph seems to have dampened his enthusiasm. So, when Kant was challenged again to express his views of those who proclaimed to have continued in the spirit of the critical philosophy, he may have chosen the opportunity to say a stronger word than was warranted by the situation. That this might be the right interpretation of the Open Letter against Fichte, is 20
Vol. 12, p. 367f.
162
Kant and His Influence
suggested by a little-known communication of Kant's colleague Rink to Villers, written a year and a half later, on 18 April 1801: Schulz is now actually working on the continuation of his Priifung, but age, ill health, and various official duties are creating many obstacles for him. For quite some time he was unwilling to proceed with the work, not wanting to be saddled with the label, made fashionable by Fichte, of literalist [Buchstabler], and this circumstance then provided an occasion for Kant's well-known declaration against Fichte. Since that time Schulz,has once again taken pen to hand.21 I conclude, then, that in spite of terminological similarities, there is no noticeable influence of Fichte on Kant's late philosophy. On the other hand, however, the declaration against Fichte cannot be regarded as the definitive expression of Kant's assessment of Fichte's thinking.
Beck The case is quite different with Jacob Sigismund Beck, a mathematician and former pupil of Kant's. He is mentioned by name in the Opus postwnum and, so it seems, approvingly, for Kant writes: '[W]e have insight into nothing except what we can make ourselves. First, however, we must make ourselves. Beck's original representing'.22 First some background information. Early in the 1790s, Kant had suggested to Beck that he write Erlduternde Ausziige - explanatory excerpts or abstracts 21 22
Altpreussische Monatsschrift, 17 (1880), p. 288f. Vol. 22, p. 353.
Fichte, Beck and Schelling in Kant's 'Opus postumum' 163 - of the main critical writings. Beck complied with Kant's wish in a highly successful way; the first volume appeared in 1793, the second in the following year (Erlduternder Auszug aus den kritischen Schriften des Herrn Prof Kant, auf Anrathen desselben). In 1796, however, Beck published a third volume that went beyond what Kant had expected. Its subtitle is: 'The Only Possible Standpoint from which Critical Philosophy Is to Be Judged'. Beck had come to believe that the structure of the first Critique - especially the sharp separation of Transcendental Aesthetic and Transcendental Analytic - promotes a false view of the text. It suggests that sensibility has a content prior to any synthesizing activity of the subject, which is subsequently subjected to readymade concepts or categories. This view, Beck was convinced, is bound to lead to insurmountable difficulties. In order to avoid the common misunderstanding that Kant really meant to speak of 'given concepts' and of 'objects that affect us', Beck insisted that one begins with the productive activity of the understanding, what he calls the 'original representing'. For sensibility and understanding, according to Beck, have a common root in the original act of representing, an act from which space, time, and the categories originally emerge, not as concepts of objects but rather as ways of representing, and as ways of achieving the original synthetic unity of consciousness through which I can apprehend an object in the first place. From this 'standpoint' of original representing, Beck proceeded to give his own presentation of the main results of the first Critique. As we know, Kant took Beck's efforts very seriously. This is reflected not only in his correspondence,23 but also in several 'Reflexionen' 2J
Cf., for instance, Kant's letter to Tieftrunk, 11 December 1797.
164
Kant and His Influence
(R 6353, 6358). And, of course, Kant himself was very much aware of the problems posed by the structure of the Critique. The book had hardly left the press when he wrote to Marcus Herz that he had 'a plan in mind according to which even popularity might be gained for this study',24 a plan Kant immediately carried out in the Prolegomena. Yet in spite of his sympathies for Beck, Kant could not see any advantage in Beck's plan to begin with the categories, and then to introduce the forms of intuition as being required for the categories to be meaningful. For the argument would first have to establish the necessity of certain forms of intuition to give meaning to the categories, and then proceed to a Transcendental Aesthetic that shows which forms we actually have available to us. And Kant believed that this method 'did not have the clarity and facility' of the synthetic method he had himself adopted in the Critique. But Beck's account looks more plausible (and more Kantian) from the point of view of self-positing that Kant developed in the Opus postumum. No doubt Kant himself realized this. It is perhaps best simply to quote a few passages from Beck's The Only Possible Standpoint, and then to compare them with some typical passages from the Opus postumum. Here is what Beck writes: Original representing consists in the categories. These are nothing else but modes of representation. Accordingly, we have no intention of viewing them here as concepts of objects at all. . . . The reader must transpose himself into this original mode of representation by himself, since our principle is a postulate and not a representation through concepts. . . . [T]here 24
Vol. 10, p. 269.
Fichte, Beck and Schelling in Kant's 'Opus postumum'
165
really is no original representing 'of an object', but simply an original representing. For whenever we have the representation of an object, it is already every time a concept, that is, it is already always the attribution of certain determinations by means of which we fix for ourselves a point of reference. . . . Accordingly, space itself is original representing, namely, the original synthesis of the homogeneous. Before this synthesis there is no space; we generate it, rather, in the synthesis. Space or this synthesis is pure intuiting itself.25 Compare this with a typical passage from Kant: The understanding begins with the consciousness of itself and performs thereby a logical act. To this the manifold of outer and inner intuition attaches itself serially, and the subject makes itself into an object in a limitless sequence. This intuition is not empirical. It is not perception, that is, not derived from a sense-object, but determines the object by the subject's a priori act, [through which] it is the owner and originator of its own representations. . . . Space and time are pure intuitions, not perceptions, that is, contained a priori as intuition in representation. . . . The subject makes this manifold of representations, namely its complex as an object in appearance, be it inner or outer, according to the principle of transcendental philosophy.26 Or take the following passage from Beck: [T]he transcendental statement, 'The understanding posits a something originally', is what first of all gives sense and meaning to the empirical statement, 'The 25
Trans George di Giovanni in Between Kant and Hegel (SUNY Press: Albany, 1985), p. 220f. " Vol. 22, pp. 82-7.
166
Kant and His Influence
object affects me'. For the first statement is the concept of the original representing itself in which all the meaning of our concepts has to be grounded. Indeed, the concept I have of my understanding as a faculty in me, even the concept of my ego, receives its sense and meaning in the first instance from this original positing.27 Again, compare this with Kant: The representations of sense-objects do not enter the subject; rather, they and the principles of their mutual connection emerge [wirken hinaus] for the purpose of knowledge of the subject, and to think objects as appearances. . . . The function of the categories [is] to constitute oneself (the subject) as an object. These forms of synthesis in appearance are original, not derivative. . . . Existence in space and time, which stems solely from the subject's power of representation . . . is contained in a system according to the principle of transcendental philosophy.28 There is enough terminological difference in these passages to make it unlikely that Kant had Beck's text in front of him when writing these passages. But that he had him in mind, and that he saw their efforts converge, is borne out by the passage quoted earlier. Here it is again: '[W]e have insight into nothing except what we can make ourselves'. Of this much they had been in agreement previously; new is the following - 'First, however, we have to make ourselves. Beck's original representing'. From the point of view of the Opus postumum, Kant seems both to agree with, and to correct or amend Beck. 27
Between Kant and Hegel, op. cit., p. 229. « Vol. 22, p. 86.
2
Fichte, Beck and Schelling in Kant's 'Opus postumum'
167
Schelling Finally, let us move on to Schelling. From him we would perhaps expect the least influence on Kant. More than fifty years younger than Kant, he was a mere twenty-one at the time when Kant began to work systematically on the Opus postumum. Yet, as is so often the case, first appearances may deceive here too. Schelling was a child prodigy; at twenty he had already published two books. The second one, Of the I as the Principle of Philosophy, was in Kant's library. In 1797, Schelling published Ideas for a Philosophy of Nature, which contains a critical exposition and development of Kant's own theory of matter. In it, Schelling tackles the very same problems that were at the forefront of Kant's attention during this period. I do not know whether Kant read the book or not, but he certainly was familiar with Schelling from the glorious reviews Schelling's books received in the Erlanger Literatur Zeitung, one of Kant's favourite literary journals at the time (which also published the challenge to Kant that lead to the Open Letter against Fichte). There, Schelling was celebrated as a new genius, and as the most promising representative of the (Kantian) dynamical theory of matter. On various occasions, Kant and Schelling are named together as the key representatives of the same philosophy. Kant at least knew that Schelling was a co-labourer in the attempt to secure victory for the dynamical theory of matter. But there are passages in the Opus postumum itself that suggest a familiarity with Schelling's writings. Kant's repeated reference to, and criticism of, the concept of a 'world-soul' is only intelligible, I believe, against the background of Schelling's book On the World-Soul, which was published at the same time (1798). And there is the following notorious passage:
168
Kant and His Influence
'System of transcendental idealism by Schelling, Spinoza, Lichtenberg, and, as it were, three dimensions: present, past, and future'. Whether this enigmatic passage shows that Kant finally acknowledged Schelling as another genuine representative of transcendental philosophy (as most commentators believe), or whether Kant is here simply referring to Schelling's book System of Transcendental Idealism (as I believe) is perhaps impossible to confirm. (The problem being that in German, book titles are not italicized, so that there is no direct way of telling whether Kant is referring to the book, or generally to a system of transcendental idealism.) I believe, however, that Kant is here thinking of Schelling's book, since only a few pages later he refers explicitly to a review of Schelling's System of Transcendental Idealism in - again - the Erlanger Literatur Zeitung. This review contains a lengthy discussion of the emergence of the three temporal dimensions in Schelling's theory of selfpositing. A careful examination of this review strongly suggests (to me at least) that Kant's phrase 'and, as it were, three dimensions: present, past, and future' alludes to this discussion, rather than to three stages of transcendental idealism. Be that as it may, it seems clear to me that Schelling was very much on Kant's mind when he wrote the last parts of the Opus postumum - perhaps even more so than Beck. This is not necessarily to say that Kant was directly influenced by either of them; we cannot establish this with any certainty. But there is no need for that either: what we can deduce is that Kant's own thoughts were going very much in the same direction as both Schelling's and Beck's. At the very least, the Opus postumum puts an end to the myth that Kant was too feeble in his old age to take notice of what was going on around him, or that he globally dismissed the efforts of his idealist successors as nonsense.
Fichte, Beck and Schelling in Kant's 'Opus postumum' 169
Bibliography Beck, J.S., Explanatory Abstract of the Critical Writings of Prof. Kant, Vol. 3: The Only Possible Standpoint from which Critical Philosophy Is to Be Judged (1796), trans George di Giovanni in Between Kant and Hegel (SUNY Press: Albany, 1985). Fichte, J.G., The Science of Knowledge (1794), trans P. Heath and J. Lachs (Cambridge University Press: Cambridge, 1982). Forster, Eckart, 'Is There "A Gap" in Kant's Critical System?', Journal of the History of Philosophy, XXV, 4 (1987); 'Kant's Selbstsetzungslehre1 in Kant's Transcendental Deductions. The Three Critiques and the Opus postumum, Eckart Forster (ed.) (Stanford University Press: Stanford, 1989); 'Kant's Notion of Philosophy', The Monist, 72, 2 (1989); 'Die Idee des Ubergangs. Uberlegungen zum Elementarsystem der bewegenden Krafte' in Ubergang. Beitrdge zur Spdtphilosophie Immanuel Kants, Siegfried Blasche (ed.) (Vittorio Klostermann: Frankfurt, forthcoming). Kant, I., Kants gesammelte Schriften, Herausgegeben von der Koniglich Preussischen Akademie der Wissenschaften (Walter de Gruyter: Berlin, 1900-); Critique of Pure Reason, trans N. Kemp Smith (The Macmillan Press: London, 1929); Prolegomena to Any Future Metaphysics, trans Paul Carus, revised Lewis White Beck (Bobbs-Merrill: Indianapolis, 1950); Metaphysical Foundations of Natural Science, trans J.W. Ellington (Bobbs-Merrill: Indianapolis, 1970); The Critique of Judgment, trans J.G. Meredith (Oxford University Press: Oxford, 1952); The First Introduction to the Critique of Judgment, trans J. Haden (Bobbs-Merrill: Indianapolis, 1965); Religion Within the Limits of Reason Alone, trans T.M. Greene and H.H. Hudson (Harper & Row: New York/Evanston, 1960); Opus postumum, Eckart Forster (ed.), trans Eckart Forster and Michael Rosen (Cambridge University Press: Cambridge, forthcoming); Philosophical Correspondence 1759-1799, trans and edited by A. Zweig (University of Chicago Press: Chicago, 1967). Schelling, F.W.J., Of the I as the Principle of Philosophy (1795) in The Unconditional in Human Knowledge: Four Early Essays (17941796), trans F. Marti (Lewisburg, Pa., 1980); Ideas for a Philosophy of Nature (1797), trans E.E. Harris and P. Heath (Cambridge University Press: Cambridge, 1988); Von der Weltseele; eine Hypothese der hoheren Physik zur Erkldrung des allgemeinen Organismus (F. Perthes: Hamburg, 1798), no translation available; System of Transcendental Idealism (1800), trans P. Heath (University Press of Virginia: Charlottesville, 1978). Schulz, J., Prufung der Kantischen Critik der reinen Vernunft, 2 vols (Hartung: Konigsberg, 1789 and Nicolovius: Konigsberg, i792), no translation available.
IMAGINATION AS A CONNECTING MIDDLE IN SCHELLING'S RECONSTRUCTION OF KANT John Llewelyn
In a footnote to the ninth edition of his Philosophical Letters on Dogmatism and Criticism composed in 1795 Schelling writes: 'It is to be hoped that time, the mother of all development, will . . . foster and eventually develop, unto the completion of the whole science, those seeds of great disclosures about this wondrous faculty which Kant has sown in his immortal work.1 The wondrous faculty in question is the imagination, the Einbildungskraft. If this hope for the development of this seed is realized it is because the mothering is supplemented by not a little husbandry at the hand of Schelling himself. No one contributes more than he to explain in what way the imagination is, as he puts it, 'the connecting middle' between theory and practice and, by implication, to explain in what way the third Critique mediates between the other two. 1 F.W.J. Schelling, The Unconditional in Human Knowledge: Four Early Essays (1794-1796), trans Fritz Marti (Bucknell University Press: Lewisburg/Associated University Presses: London, 1980), p. 190; K.F.A Schelling (ed.), Sdmtliche Werke (J.G. Cotta: Stuttgart and Augsburg, 185661), hereafter SW. 170
'Imagination' in Schelling's Reconstruction of Kant
171
Imagination . . . is analogous with theoretical reason inasmuch as this is dependent upon cognition of the object, and analogous with practical reason insofar as this produces its object, itself. Imagination actively produces an object by putting itself in complete dependence on that object, into full passivity. What the creature of imagination lacks in objectivity, imagination itself supplies by the passivity which, through an act of spontaneity, it voluntarily assumes toward the idea of that object. Thus imagination could be defined as the faculty of putting oneself into complete passivity by full self-activity.2 If, following the cue of the words Schelling stresses here, we consider the polarity of the theoretical and the practical within the architectonic of the three Critiques, we are struck by the thought that imagination is already involved in the first Critique qua concerned primarily with theoretical knowledge and that it is arguably involved in the second Critique qua concerned primarily with practice. The argument for this latter claim would turn on whether one understands the role of the Type {Typik) to be primarily practical or indistinguishable from the imagination's role as a connecting middle between theory and practice. But it can be a connecting middle only if it partakes of both the theoretical and the practical through being their common root. One reason for saying that the Type is a product of the practical imagination is that, according to Kant, the imagination is the faculty which produces an example or model for the application of a law. In so far as the law to be applied is a rule of skill or counsel of prudence it reduces to a falsifiable theoretical statement about the empirical 2 Ibid.
172
Kant and His Influence
world. Only the moral law is irreducibly practical. Nevertheless, our understanding of its universal applicability in the intelligible world of ends can be assisted by the thought of a world in which everyone in the world of ends and means behaves according to the principle of prudence or self-love. Here, however, it is only the universal form of the moral law that is modelled by the law of self-love. Since the moral law is a law of pure reason prescribing with regard only to the form of subjective maxims, hence not grounded in sensibility, the law of the system of nature which typifies it does so only in respect of form. 'Consequently the moral law has no faculty but the understanding to aid its application to physical objects (not the imagination)'.3 'Not the imagination'. These words seem to settle once and for all that there is no place in Kant's thinking for pure practical imagination. This may indeed be so, unless in the second Critique a place can be found for the ambiguity to be found in the first, for example at B 162 note b, between imagination contrasted with and imagination identified with understanding or regarded as the understanding performing one of its roles. Whether or not a place can be found in the second Critique for this ambiguity and thereby for the notion of a properly practical imagination, one of the lessons to be drawn from some of the writings of Schelling seems to be that a place for such a notion must be found, and that the way to find it is along a path which reconstructs each Critique in terms of the others with the help of an intellectual intuitus originarius which Kant denies to man. In attempting in this essay to follow Schelling's first steps along this path let us set out from the theory of transcendental imagination of the first Critique. 1
Kant's Critique of Practical Reason and Other Works on the Theory of Ethics, trans T.K. Abbott, 6th edn (Longman: London, 1909), p. 161.
'Imagination' in Schelling's Reconstruction of Kant
173
I On a simple reading of Kant's theory of theory, the transcendental imagination occupies a Zwischenland between sensibility and understanding defined as the faculty of pure a priori concepts and principles. It is the faculty of judgement where judgement is taken to mean the subsumption of particular cases under concepts and principles. Here the simple reading encounters the complication of the ambiguity announced in the footnote to B 162 which we have already mentioned and will have occasion to mention once more: 'It is one and the same spontaneity, which in the one case, under the title of imagination, and in the other case, under the title of understanding, brings combination into the manifold of intuition'. Kant's movement between an account which contrasts and an account which assimilates imagination and understanding is sometimes said to be a symptom of his partial realization that the capacity to apply a concept or principle is part of understanding it. Indeed, there is no reason why Kant should be said to reject this inclusive notion of understanding so long as we recognize what it includes. If we fail to recognize this, we shall fail to recognize that there is a sense in which someone can be in possession of a concept theoretically and be able to define it yet still be hopeless at picking out instances of it. He may have an understanding of the kind that can be learned by rote yet lack the so-called mother wit which at B 172 Kant calls 'so-called' perhaps because however innate this intelligence may be it still requires practice if the fluent exercise of it is to be possible. In that case, Kant's distinction between a more and a less abstractly intellectual understanding of concepts is preserved at any rate for empirical concepts through the concession that the
174
Kant and His Influence
capacity to apply them in practice, which is part of the fuller understanding, is itself acquired through practice. If Kant nevertheless wishes to distinguish also between, on the one hand, the particularity of a dog or of a physical or mental picture of a dog and, on the other hand, the schema intermediate between the abstract universal concept of it and the particular, the empirical schema appears to end up being nothing other than the acquired capacity to pick out dogs and pictures of dogs rather than something that enables us to acquire that capacity. All it will enable is an improvement in that capacity. Nevertheless, where pure non-empirical concepts are concerned, there is a role for the schema to perform which, does not reduce to that of an acquired expertise. Hence, the acquisition of empirical concepts would not be explicable solely in the terms of an empiricist theory of learning. For empirical concepts presuppose pure schemata, namely mathematical ones in the narrower sense, that is to say, geometrical, arithmetical or algebraic ones, and transcendental ones, that is to say, mathematical and dynamic ones in the sense in which Kant employs these terms to separate each of the quadruples of pure categories, pure schemata and pure principles into two pairs. Although in calling these categories, schemata and principles pure, Kant is rejecting genetic empiricism, this does not prevent his sometimes arguing for a variety of logical empiricism along the lines of the logically behaviourist account of understanding we have just been discussing, according to which the understanding of a concept is the ability to cash it by referring to specific instances that do and don't fall under it. Just what variety of logical empiricism Kant subscribes to will depend on what he would count as a specific instance.
'Imagination' in Schelling's Reconstruction of Kant
175
Take, for instance, the concepts of mathematics, considering them first of all in their pure intuitions. Space has three dimensions; between two points there can be only one straight line, etc. Although all these principles, and the representation of the object with which this science occupies itself, are generated in the mind completely a priori, they would mean nothing, were we not always able to present their meaning in appearances, that is, in empirical objects (B 299). The mathematical principles and concepts in question would have no meaning or sense without this reference to empirical instantiation. Furthermore, we cannot define any of the categories and principles, meaning by that not just the specially mathematical ones but the transcendental ones as well, 'without at once descending to the conditions of sensibility'. By sensibility here he means not solely pure sensibility but empirical sensibility. Without that there is no Sinn. This word is translated by Kemp Smith on at least one occasion as 'sensible meaning'. This translation receives support not only from etymology but also from the adverbal phrase in Kant's remark that without 'relation to the object' we cannot define any of the categories 'in any real sense'. But we have seen that in the passage cited from the previous paragraph, mathematical concepts 'mean nothing' {gar nichts bedeuten) unless their meaning can be laid before us (ihre Bedeutung darlegen) in empirical objects. That, we are then told, is how it is with all the categories and principles tabled in the Transcendental Analytic. Without pretending that the paragraphs here in question pose no problems, it must be said that they do not compel us to draw the embarrassing inference that a pure concept can have a purely empirical, so to speak purely 'impure', instance. If there are empirical instances of the
176
Kant and His Influence
pure concept triangle they are instances in the literal sense of the word Bei-spielen: they are instances falling under the concept only in the mode of standing in for, and playing alongside, it. Pure concepts can have real meaning, as Kant says, only if they are able to present their meaning in empirical objects. Their meaning and the empirical presentability of it are two stages. Thus, the pure concept of magnitude 'seeks its support and meaning [Sinn) in number, and this in turn in the fingers, in the beads of the abacus, or in the strokes and points which can be placed before the eyes'. The 'in turn' {diese aber) here makes it clear that Kant is not saying that the science of arithmetic is a counting of fingers, beads or of mental images of any such things. That cannot be right if mental images of beads are as particular as beads, for that would lead to an infinite regress which he seeks to avoid by stating quite explicitly that counting particulars presupposes schemata which are not particular things because they are not things but procedures or methods for identifying things. The four fingers of a particular empirically perceivable hand exemplify magnitude, a universal intellectual concept of the understanding, thanks to the schema of number, a sensuous concept constructed in the imagination. The purely intellectual concept is introduced to the empirical objects of sense, presented to them, not directly, but through the mediation of a construct, the construct of the number four and of number as such. In calling the schema a construct, we seem to be calling it a thing. But the schema as construct is inseparable from the process of constructing. The numbers four and five are unities at which we arrive. They are totalities built up in time by the combination of a plurality of units: the category of quantity temporalized. Number in general is 'the unity of the synthesis of the manifold of a homogeneous intuition in
'Imagination' in Schelling's Reconstruction of Kant
177
general, a unity due to my generating time itself in the apprehension of the intuition' (B 182). Similarly, the undetermined concept of ground becomes determined as the concept of mechanical cause, and the purely logical sequence {Folge) of hypothetical judgement becomes determined as causal sequence through the mediation of the schema of temporal sequence according to a rule. Since sensible intuition is the only intuition available to human beings according to Kant, construction is restricted to time and along with it the real meaning of the categories of the understanding. These pure concepts have only a purely logical meaning. They have no objective reference unless they are given it by the schemata. What difference would it make to the critical idealism of Kant, we may now ask, when it is argued, as it is argued by Schelling, that sensible intuition is not the only intuition available to human beings, and that we have access to non-discursive intellectual constructive intuition as well? II A convenient point of departure from which to move toward an answer to this question is the piece 'On Construction in Philosophy' published anonymously in the Critical Journal of Philosophy in 1802. Although some commentators have attributed this to Hegel and others have suggested that it is a joint production by the co-editors of the journal, most authorities now agree that it is by Schelling. 'On Construction in Philosophy' is a review of Carl H. Hover's Treatise on Philosophical Construction published in the previous year. These two titles indicate that what is at stake is Kant's assertion in the Transcendental Doctrine of Method of the first Critique that philosophical knowledge is discursive or
178
Kant and His Influence
acroamatic, i.e. knowledge gained by reason from the analysis of concepts, whereas mathematical knowledge is knowledge reached by reason through pure temporal or temporo-spatial synthesis, i.e. the construction of concepts in intuition which as such is concrete and individual rather than abstractly or otherwise discursive (B 741, B 758). Schelling maintains against this that there is a place for construction in philosophy. So are we going to find that whereas Kant, because he believes that there is no place for construction in philosophy, rejects Spinoza's and Wolff's assumption that the method of philosophy can begin with definitions, Schelling will allow the possibility of this, even though the construction allowing it will not be one of pure time or space? Schelling makes five main points. (1) The first critical point Schelling makes against Kant in his article turns, if I read his intention correctly, on the fact that mathematics, on Kant's account, employs schemata and these are each a unity of a given pure intuition or form of intuition which is particular and is the product of an act (Handlung) of construction. It is this act which brings the concept of unity to the pure sensible datum. It is the conceptual partner in the constitution of the schema. Now, since the schema is a unity of this with a sensible datum, the intellectuality of the schema as a whole dominates the pure sensibility which is offered as, so to speak, the temporal receptacle in which the constructing takes place. In this interpretation of Schelling's first objection to Kant a distinction is made between the intuitive and the constructive aspects of the schema. Strictly speaking, therefore, we should have to say that it is the constructing which is intellectual. Schelling, however, considers that his objection entitles him to claim to have found an
'Imagination' in Schelling's Reconstruction of Kant
179
opening for intellectual intuition. This locution can be justified on the grounds that it is with distinguishable but inseparable aspects of schematism that we are concerned, as the constructing is distinguished from the construct produced. Unless we insist on this, the notion of schematism will be open to the objection that a second order schema is required to bridge the gap between its two parts and a third and fourth to bridge the gap between their sub-parts, and so on. Kant's insistence that the schema is a process or procedure (Verfahren) puts his reader on guard against this regress. Further, a reason for protesting against Schelling's talk of intellectual intuition or intellectual intuiting would be a reason for protesting against Kant's reference to the schema as a sensible concept. (2) Schelling writes: 'Space as according to Kant it grounds geometry and time as according to him it grounds arithmetic, is wholly intellectual intuition, but expressed there in the finite and here in the infinite'. (Der Rawn, wie er der Geometrie, und die Zeit, wie sie der Arithmetik nach Kant zu Grunde liegt, ist die ganze intellectuale Anschauung; aber dort im Endlichen, bier im Unendlichen ausgedruckt.) Does 'there' mean in the case of space and 'here' in the case of time, as seems likely? Or does 'there' refer to the case of the mathematician and 'here' to the case of the philosopher which are mentioned in that order in the preceding sentence? In the preceding sentence, Schelling says that the absolute difference between mathematics and philosophy cannot lie for Kant in the fact that there is non-empirical intuition in the latter but not in the former, since Kant argues that geometry does call for non-empirical intuition. The difference must therefore lie in the fact that whereas for the mathematician intellectual intuition is reflected in
180
Kant and His Influence
sensibility, for the philosopher it is reflected in itself, in sich selbst. This would be a reason for saying that in philosophy intellectual intuition is infinite, for it is not limited by something other than itself as is the case with intellectual intuition in mathematics, where it is limited by sensibility. This does not exclude the possibility that philosophy's intellectual intuition is temporal. And, indeed, Schelling maintains that it is. His way of maintaining it, furthermore, brings together into one story the two interpretations of the words 'here' and 'there' in the sentence in question. For philosophy's intellectual intuition is an intuition of pure time. Here the purity of time is its independence of spatiality and its closer kinship with arithmetic as against geometry. True, on Kant's account the representation of the arithmetical continuum depends upon the spatial continuum as a (for humans) contingently necessary auxiliary. Even time itself we cannot represent, save in so far as we attend, in the drawing of a straight line (which has to serve as the outer figurative representation of time), merely to the act of the synthesis of the manifold whereby we successively determine inner sense, and in so doing attend to the succession of this determination in inner sense (B 155). So the notion of succession, presupposed in counting, is a notion of pure active motion, Bewegung, regarded as an act, Handlung, presupposed both by the representation of time which, Kant elsewhere tells us, cannot itself be perceived, and by the representation of space: their connecting middle or common root which explains why Kant says that the schema is not a thing but a Verfahren. This of the word Bei-spielen: they are instances falling under which as striving, presupposes the resistance, the objection, of the spatio-temporally real. This opposition is
'Imagination' in Schelling's Reconstruction of Kant
181
the origin of the feeling of self-awareness, that is to say, of the self's awareness of its own temporality. Schelling asks: ''how, then, does the self become an object to itself as inner sense?'; that is to say, how does it become an object for itself as sensation combined with consciousness? He answers: 'Simply and solely through the fact that time arises {entsteht) for it', not of course time as already externally intuited but, he says, in words that relay Kant's reflections on time back to Augustine and forward to Bergson, Husserl and Heidegger, 'time as mere limit, mere point (blosse Grenze, blosse Punkt)\ that is to say, 'as pure intensity, as activity which can expand itself only in one dimension, but is now concentrated at a single point'.4 The intensio of time as limit or point is not yet time itself. It is how time arises for the self. It is time when the unidimensionally extensible activity becomes an object to itself. This is entailed by the first principle that the self is self-consciousness, activity conscious of itself, from which it follows that the self itself is time conceived of in activity, in Thdtigkeit gedacht. The intensio of time as the tension of a spring or as source is inconceivable in isolation from the idea of, respectively, expansibility or flow {distensio). Since this idea of the subject being a temporally extended object to itself implies that of the subject's being conscious of an object in space, we can say that the intensio or intensity of the self implies the interitionality of consciousness, and vice versa. So we could formulate Schelling's second objection against Kant by saying that the latter fails to see that a corollary of his own argument is that without intellectual intuition sensible intuition is blind. The origin 4 F.W.J. Schelling, System of Transcendental Idealism (1800), trans Peter Heath (University Press of Virginia: Charlottesville, 1978), hereafter STI, p. 103; SW III, p. 466.
182
Kant and His Influence
of geometry, and of arithmetic and algebra, the origin of mathematics is the origin of philosophy. (3) Schilling's third objection, or perhaps a strong reformulation of the one we have just considered, is that Kant contradicts himself. For, as we have indeed confirmed, Kant maintains that philosophy is the analysis of concepts. But concepts will be empty unless there is some object to which they can be referred. Since he denies that in philosophy they are independent of intuition of sensible objects, whether empirical or pure, they must on pain of vacuity be given determinacy of meaning by being referred to intellectual intuition. Hence intellectual intuition is presupposed by transcendental imagination and its schematism, by the pure synthesis of transcendental selfconsciousness. Provided this alleged intellectual intuition of the 'I am' is temporal despite not being spatial, a refutation of subjective idealism analogous to Kant's is available to Schelling, as we have just seen. (4) From our discussion of Schelling's first objection against Kant it is clear that in mathematics intuition has two aspects, a universal and a particular. According to Kant mathematics treats of the universal in the particular, whereas philosophy treats the particular only in the universal (B 742). Schelling doubts the validity of both of these characterizations. He casts doubt on the first by citing Kant against Kant. Kant does indeed say that in geometry the concept of a figure, for example a triangle, is expressed in a particular pure or empirical intuition; he also says however: the evident propositions of numerical relation are indeed synthetic, but are not general like those of geometry, and cannot, therefore, be called axioms but only numerical formulas. The assertion that 7 + 5 is
'Imagination' in Schelling's Reconstruction of Kant
183
equal to 12 is not an analytic proposition. . . . But although this proposition is synthetic, it is also only singular. So far as we are here attending merely to the synthesis of the homogeneous (of units), that synthesis can take place only in one way, although the employment of these numbers is general (B 205). Schelling does not cite or refer to these sentences explicitly, but it must be on them that he bases his statement that Kant himself concedes that arithmetic treats of particular numerical formulae stating a relation between individual magnitudes which are expressed in the universal. That would make arithmetic a part of philosophy on Kant's definition of this. Schelling does not, however, share this definition of philosophy. Because the opposition between philosophy and mathematics cannot be based on whether the particular is expressed in the universal or vice versa, and because, in either case, a true identification of the universal with the particular is as such an intuition, the difference between mathematics and philosophy is a difference of intuition or construction. In mathematics, construction is distinguished by having two aspects, the universal and the particular, the former expressing or presenting the latter or vice versa. In philosophical construction what in mathematics is distinguished is united in a point of absolute indifference or identity which is an identity of identity and difference. (5) With this notion of constructive intuition in which the opposition of the two kinds of expression is overcome, a new notion of particularity emerges and with it a new notion of universality, concrete universality, universality concresced from the expression of the metaphysical in the physical and the physical in the metaphysical. This concrescence of visible spirit and invisible nature is, in the language of Spinoza, the identity of Thought and Extension,
184
Kant and His Influence
the point of indifference between the sive of Deus sive Natura and the sive of Deus sive Spiritus. In the language of Hegel and of the Critique of Judgment (section IV), it is a resultant of the joint expression of a determinate and a reflective judgement. This absolute universality is the universality not of a discursive concept but of a concrete Idea that is the object of intellectual intuition. Whereas Kant cannot maintain both the difference between mathematics and philosophy and secure real sense through reference for the concepts of philosophy, Schelling does both by distinguishing two kinds of intuitional object. If Kant had recognized that he is committed to admitting intellectual intuition in both mathematics and philosophy but that the particularity and generality in each case are of two absolutely different kinds, he would have recognized too that to ask a philosopher to work out the sum of the angles of a triangle from the concept of a triangle is like asking a sculptor to make a statue out of musical notes or a musician to perform a sonata with pigments. Ill In the System of Transcendental Idealism Schelling endorses Fichte's partial endorsement of Descartes' Discourse on Method and Meditations. Up to a point this is also another endorsement of Kant's deduction from the 'I think'. The beginning of the system set out philosophically is the end of the method of discovery, and the structure of the end is reflected in that of the beginning. The identity of intuitive construct and constructing, of thought and thinking, in the Absolute Idea is reflected in the 'I am' which is 'the first, absolutely unconditioned principle' which Schelling takes over from Fichte's Science of Knowledge, as illustrated by Fichte's statement that in the act of positing 'I am', the self 'is at
'Imagination' in Schelling's Reconstruction of Kant
185
once the agent (das Handelnde) and the product of action (der Handlung); the active (das Thatige), and what the activity [die Thdtigkeit) brings about'.5 For this identity of the producing and product or constructing and construct, acting (Handlung) and act performed (That) Fichte adopts the word Thathandlung. This word is commonly used of violent deeds and is therefore appropriate to describe what Schelling and Fichte say is for human beings a unique coincidence through imagination of product and producing or, as we might alternatively say, of constative and non-constative performance, such as might be supposed to be reserved for a divine causa sui. Einbildungs-kraft is a craft, a quiet force, and in writings of Schelling later than those with which we are here immediately concerned it becomes a name for the creativity of God. Schelling observes: In becoming an object of myself through self-consciousness, there arises for me the concept of the self, and conversely, the concept of the self is merely the concept of becoming-an-object-to-oneself. . . . Thus we have here that original identity of thought and object, appearance and reality, for which we were searching, and which is nowhere else to be found. . . . That this identity between being-thought and coming-to-be, in the case of the self, remains hidden from so many, is due solely to the fact that they neither perform the act of self-consciousness in freedom, nor are able to reflect in so doing upon what arises therein.6 5 J.G. Fichte, Science of Knowledge (Wissenschaftslehre), with the First and Second Introductions, trans and edited by Peter Heath and John Lachs (Appleton/Century/Crofts: New York, 1970), hereafter SK, p. 97; Gesanttausgabe, I, 2, Reinhard Lauth and Hans Jacob (eds) (Stuttgart-Bad Cannstatt: Friedrich Frommann Verlag, 1965), p. 259 (I, 96). 6 STI, p. 25; cp p. 43, SW III, p. 366, cp p. 389.
186
Kant and His Influence
So pure consciousness, self-consciousness, usually gets hidden by empirical consciousness. Our consciousness is normally directed to things around us or to the feelings to which they give rise. Because such empirical consciousness is not unchallengeably veridical on any specific occasion it cannot supply the supreme principle of knowledge. This is why Fichte and Schelling begin their search for that principle with a principle of logic, the principle of identity A = A which is usually taken to be presupposed by any synthetic proposition. They agree, however, that an identical proposition conveys no knowledge. They agree with Kant that it is purely formal and that for a proposition to convey real knowledge objective content must be provided in some way. The only proposition that fulfils the requirements of absolute priority is one which like A = A is identical, but which is, at the same time, synthetic. That proposition is 'I = I* or 'I am'. From this synthetic identical proposition in which form and content are one, Fichte sets out, in the wake of Augustine, Descartes and Kant, to construct a science of the world. Our main interest here is in the method of construction which Schelling employs. Having now acquainted ourselves with some of the workings of this method first at the end and now at the beginning of his version of what Fichte calls the 'pragmatic history of the human mind or spirit'7 let us return to the end.
IV Marking out the ground for Hegel's history of spirit, Geisfs Geschehen, and Heidegger's history of the concealment and advent of being, Sein's Ereignis, Schelling 7
SK, pp. 198-9; Gesamtausgabe, I, 2, p. 365 (I, 222).
'Imagination' in Schelling's Reconstruction of Kant
187
conceives philosophy as 'the free recapitulation {Wiederholung) of the series of acts into which the one act of self-consciousness evolves'.8 As the science of science, the knowledge of knowledge, philosophy is the one science with a double seriality, that of the reflective and specifically philosophical series and that of the reflected series of ordinary consciousness which in the end coincide. Seven years before the publication of the Phenomenology of Spirit a pre-echo can be heard in Schelling's System when we read: 'subject and object which are absolutely united in the absolute act of selfconsciousness - must be constantly kept distinct for purposes of philosophizing, that is, in order to allow this unification to take place before our eyes'.9 So, as Hegel says in the Introduction of the Phenomenology 'all that is left for us to do is simply to look on'.10 Philosophy is a pragmatic history because it is a repetition of deeds analogous to the procedure in a law-court through which the necessity of those deeds is brought out. As the words pragma, Tatsache and 'fact' remind us, an act may be the doing and what is brought about, whether the doing be praxis or poiesis. This two-in-one reflected in the language of ordinary consciousness is matched by the philosophical term of art 'point of indifference' which Schelling introduces for the identity of constructing and the intuited object whose transfigurations evolve from original self-consciousness and whose so-called indifference therefore is far from the indeterminacy of a night in which all cows are black or a 'formless whiteness' like that of the blank page purporting to portray the Absolute in a certain burlesque of the philosophical journal Mind 8
STI, p. 49; SW III, p. 397. ' STI, p. 42, cp p. 73, SW HI, p. 389, cp p. 427. 10 Hegel, Phenomenology of Spirit, trans A.V. Miller (Clarendon: Oxford, 1977), p. 54, cp pp. 36, 57.
188
Kant and His Influence
entitled Mind!.11 Schelling complains that when he refers to his 'real idealism' as a system of absolute identity the word identity has been misinterpreted when it has been taken to imply that the system suppresses the distinctions between matter and spirit, good and evil, truth and error.12 Likewise, indifference is not sheer emptiness, nothing or night.13 More like the polemos of darkness and light to which Heraclitus and Holderlin allude, it is a 'conflict of absolutely opposed activities . . . one that originally reaches out into infinity . . . the real, objective, limitable activity; the other, the tendency to intuit oneself in that infinity, . . . the ideal, subjective, illimitable activity'. In the System of Transcendental Idealism the genesis of the figures of consciousness follows approximately the order of the Transcendental Doctrine of Elements of the Critique of Pure Reason and the order of the three Critiques. Apart from a disagreement over the doctrine of the things-in-themselves, the chief difference is that from the deduction of the absolute synthesis of self-consciousness, the movement of the deductions takes the form of graduated resolutions of contradictions by transition to some third thing common to each of the opposites. The main body of the book is divided into the three major transitions which Schelling calls epochs: from original sensation to productive intuition, from productive intuition to reflection and from reflection to the absolute act of will, this last being the transition from theory to practice of which, as we noted, Schelling undertakes to trace the common root. Given the already evident similarities in many respects of Schelling's System of Transcendental " Hegel, Phenomenology, p. 31. 12 SW X, p. 107. " SW IV, p. 403; cp Martin Heidegger, Beitrdge zur Philosophie, Gesamtausgabe, 65 (Klostermann: Frankfurt am Main, 1989), p. 199.
'Imagination' in Schelling's Reconstruction of Kant
189
Idealism to the transcendental idealism of Kant, we are primed to expect agreement between them as to this common root of theory and practice. Given the ubiquity in Schelling's System of constructive intuition, how could this common root fail to be the imagination where schematism performs its 'art concealed in the depths of the human soul' and 'whose real modes of operation {Handgriffe) nature is hardly likely ever to allow us to discover, and to have open to our gaze' (B 180)? How can this fail to be in one guise or another the imagination which as productive imagination it is not unreasonable to identify with the 'common, but to us unknown, root' of sensibility and understanding' (B 29 )?14 Yet in the discussion of what he calls productive intuition, that is to say in the section dealing with the construction of objects, Schelling is coy about using the word 'imagination'. The word is very rarely used there. The simplest explanation for this is one that sheds some light also on the fact that in classical lists of the powers of the mind sometimes both memory and imagination are mentioned, but sometimes the former is subsumed under the latter. The importance Schelling attributes to construction leads him to think of imagination primarily as constructive and his decision to keep more or less in step with the first Critique in his ordering of the treatment of topics means that imagination will not receive the full treatment until his System reaches the stage equivalent to that of Kant's pages on schematism. The System's section on productive intuition is roughly equivalent to Kant's transcendental deduction of the conditions of objectivity. So any mention it makes of 14
Cp Heidegger, Kant and the Problem of Metaphysics, trans James S. Churchill (Indiana University Press: Bloomington, 1962), p. 41; Kant und das Problem der Metaphysik (Klostermann: Frankfurt am Main, 1951), p. 41.
190
Kant and His Influence
imagination will be such as is licensed by the fact that in the Critique of Pure Reason imagination has been considered at this stage only in its capacity as the reproductive component of the threefold synthesis where it is in synergy with recognition in concepts and apprehension in intuition. Kant's phrase 'apprehension in intuition' may well have suggested the phrase 'productive intuition' to Schelling. A productive intuition is a productive image, where the reference to image must be read as a reference to Kant's teaching that 'The pure image (Bild) of all magnitudes {quantorum) for outer sense is space; that of all objects of the senses in general is time' (B 182). This statement is made in the section on Schematism where it is argued, as we have seen, that the application of magnitude (quantitas), a general concept of the understanding, to these pure particular objects of intuition is managed by the schema of number. That productive intuition is another name for the imagination is suggested by the fact that we have found Schelling referring to the imagination as 'this wondrous faculty' in 1795 and now find him speaking in 1800 of 'the marvel [das Wunder) of productive intuition'.15 Its marvellous art, notwithstanding its being an art that conceals art, is discovered by Schelling, with Fichte's aid, to be that of effecting a suspension of the oscillation (Schweben) between, on the one hand, the restriction imposed upon the self by, for example, sensation - when the self senses itself opposed to itself - and, on the other hand, the self's infinite striving to pass beyond this real opposition to its ideal activity, and to overcome this difference, which can neither persist nor be simply done away with {aufgehoben), in the identity of a third activity. The fact that this 'third thing' is still an activity and 15
STI, p. 70, SW HI, p. 423.
'Imagination' in Schelling's Reconstruction of Kant 191
destined to give rise to a higher level oscillation or wavering may explain why the imagination is here described as the resolution of a conflict whereas in the Science of Knowledge Fichte says imagination is itself an oscillation. Phantasia in the De Anima is a kinesis (429al). This idea of movement is implied in the etymology of pha which conveys the idea of emerging or coming to light, phainesthai. It is what Merleau-Ponty would call the 'good ambiguity' of the imagination that moves the mind from each of its epochs to the next. This is why it crops up not only in Schelling's first and second epochs, that is to say, not only in the transition from feeling to productive intuition and from its unconscious operation there to reflection. It emerges also in the transition from reflection on objects to the conscious willing of objects as ends, those produced when the art concealed in the depths of the soul emerges in objets d'art. The methodological way-making {Bewegung) from Schelling's second epoch to his third is a shift from reflection to schematism. In other words, it is a transition from what in the third Critique Kant calls reflective judgement to what he there calls determinative judgement. If, in ascending from one epoch to another the structure of the less complex is refigured but retained in the next, then the highest flight of imagination will be symbolic in the special sense defined by Schelling in the lectures on the Philosophy of Art first delivered in 1802. There schematism is defined as a presentation {Darstellung) in which the universal signifies (bedeutet) the particular or in which the particular is intuited through {angeschaut durch) the universal. The allegorical is defined as presentation in which the particular signifies the universal or in which the universal is intuited through the particular. The symbolic is the logical product of
192
Kant and His Influence
these, or, as he says, their synthesis, in which neither does the general signify the particular nor does the particular signify the general, but in which these signifyings are absolutely one.16 Although in making these distinctions Schelling must have had the third Critique open before him at §59 - for his modes of Darstellung are what Kant there calls kinds of hypotyposes and exhibitiones - his definition of the symbolic is slightly but significantly more articulated than Kant's. Although, like Schelling, Kant contrasts the symbolic explicitly with the schematic (the determinative), he does not, unlike Schelling, contrast the symbolic explicitly with the allegorical (the reflective). The symbolic, on Kant's definition, is a variety of what Aristotle calls analogy. It is an analogy with a double function. It functions as a schema, applying a concept to an intuited object; but the rule of procedure it applies to that intuited object is applied also to another object, as when a living body is taken to represent a monarchical state governed by constitutional laws. Here symbolism is similation by simile made possible by a structural rather than an intuitional resemblance between a constitutional monarchy and a living body. A symbol thus defined is a metaphor and corresponds to what Schelling rates as a lesser form of art. For Schelling, symbolism is imagination at its most developed. And although there are some forms of art which are schematic and others which are allegorical, the highest form of art is symbolic. It may be that in symbolism, as defined by Kant, both schematism and allegorism are at work, but they have not fused as they are in symbolism as defined by Schelling. We have just seen that although Schelling says that symbolism is a synthesis of the other two modes of presentation, he also says that neither of them 16
SW V, p. 407.
'Imagination' in Schelling's Reconstruction of Kant 193
operates as such. Their becoming one is to be understood in the way that in the triads of categories deduced by Kant the third, though a synthesis of the first and the second, is something entirely new. That schematism and allegory achieve an Indifferenzpunkt in the symbolic is neatly encompassed in the German word for symbol, Sinnbild, which mirrors the unity of meaning and form.17 Schelling evinces his wish to ensure that this mirroring be not confused with a one-sided dependency relation of copying or causing subsumed under the principle of sufficient reason when the phrase 'mere speculation' in the Introduction to the first edition of the Philosophy of Nature (1793) is replaced in the second edition (1803) by the phrase 'mere reflection', thereby enabling 'speculation' to be retained for a nobler role, as it is also by Hegel.18 Perhaps the conjunction of his belief that the imagination is fully developed in art with his belief that it is in the sphere of art that the imagination is usually thought to be most at home explains Schelling's reservations over using the word 'imagination' for the earlier stages of the history of the mind. Coleridge begins Chapter X of Biographia Literaria as follows: 'Esemplastic. The word is not in Johnson, nor have I met with it elsewhere'. Neither have I. I constructed it myself from the Greek words, eis en plattein^ to shape into one; because having to convey a new sense, I thought that a new term would both aid the recollection of my meaning, and prevent its being 17
SW V, p. 412. See K. Diising, 'Spekulation und Reflexion: Zur Zusammenarbeit Schellings und Hegels in Jena', Hegel-Studien, V (1969), pp. 95-127, and JeanFrancois Courtine, 'Le deploiement schellingien de 1'unite: de Yuniversio a Yuniversitas1, Les etudes philosophiques (1978), p. 351, note 12. 18
194
Kant and His Influence
confounded with the usual import of the word, imagination. To 'shape into one' is also a translation of ln-eins-bildung which is the power of Einbildungskraft on Schelling's account of imagination.19 Schelling does not want this to be confounded with the merely associative reproductive imagination of the empiricists. Hence, when he is treating the oscillation between the infinite will and the finite reality which opposes it and makes it possible he acknowledges that he has not yet shown his right to employ the word imagination of this oscillation between and ultimate unification of the finite and infinite. A similar caution is expressed by Fichte when of a certain activity he has found it necessary to postulate in the course of the argument of the Science of Knowledge he writes in parentheses 'its name is imagination, as will appear in due course'.20 If the imagination is most fully developed in art and art is most fully developed in symbolism, symbolism is most fully developed in mythology. Schelling's most favoured cases among extant mythologies are those presented in Homer, where the individual is the universal and the universal is the individual; the Sinn is the Bild and the Bild the Sinn rather than either signifying (bedeuten) the other, which is what happens once mythology gives way to allegory, as Homer's mythologies do for a later age, or when his myths are succeeded by such allegories as that of Eros and Psyche. That Homer's Odyssey should be accorded such a privileged place is in keeping with the pragmatic history of the human spirit being described as spirit's Odyssey.21 It is the history of a " SW V, p. 306. -"SK, p. 150; Gesamtausgabe, I, 2, p. 314 (I, 160). -' STI, p. 232, SW III, p. 629.
'Imagination' in Schelling's Reconstruction of Kant
195
homecoming in that the end of that history is the fulfilment of its beginning. The original judgement, the Ur-teil, of that beginning is the simple twofold root of self-consciousness from which grows the highest form of self-consciousness, the aesthetic imagination which is 'the productive intuition reiterated to its highest degree' {die in der hochsten Potenz sich wiederbolende produktive Anschauung)11 and whose expression in the work of art is the objective organon of philosophy. Let us retrace more closely the path followed by this return.
V Schelling's statement that productive intuition of objects reiterated to its highest degree is the poiesis of aesthetic intuition becomes more comprehensible when it is set alongside certain statements made by Kant. In the section of the Critique of Pure Reason dealing with the transition from the table of forms of judgement to the table of categories, Kant writes: 'The same function which gives unity to the various representations in a judgement also gives unity to the mere synthesis of various representations in an intuition* (B 104). We have already noted more than once that this function which gives unity to the mere synthesis of various representations in an intuition is sometimes called 'imagination' by Kant and sometimes 'understanding'. 'It is one and the same spontaneity, which in the one case, under the title of imagination, and in the other case, under the title of understanding, brings combination into the manifold of intuition' (B 162 note b). Schelling must surely intend his reader to recognize that he is repeating this sentence
11
STI, p. 230, SW V, p. 626.
196
Kant and His Influence
when he writes, with reference to poetic intuition and to productive intuition, that is, intuition productive of the space in which real objects appear, 'it is one and the same capacity that is active in both, the only one whereby we are able to think and to couple together even what is contradictory {das Widersprechende) - and its name is imagination'.23 If Schelling and Kant are agreed that this capacity, spontaneity or function, may sometimes be entitled imagination and sometimes understanding, how does this agreement stand in relation to the sentence cited from the Critique of Practical Reason at the beginning of this essay: 'the moral law has no faculty but the understanding to aid its application to physical objects (not the imagination)'? Kant says this because the moral law is purely formal and the order of a moral realm of free agents can have as an analogy in nature only the formality of a system of natural laws, that is to say, an order of understanding whose laws are not exhibited in concrete spatio-temporal reality by schematism. This does not mean that the moral idea of reason cannot be exhibited in an aesthetic idea, an ideal type or archetype constructed by the poetic imagination. When Schelling writes that an Idea of reason is not a concept of the understanding, he is thinking of the antinomies which Kant has shown to arise if this is forgotten; for example, if it is forgotten that freedom is not a concept of the understanding, but an Idea of reason. Remembering the distinction between the practical employment of Ideas of reason and the theoretical employment of concepts of the understanding makes it possible to couple together, to construe, to construct, 21
STI, p. 230, SW III, p. 625.
Imagination' in Schelling's Reconstruction of Kant
197
what would otherwise be contradictory, for example the infinitude of the will and the finitude of resistant reality without which there can be no willing. Remembering this distinction allows not only for this overcoming of contradiction. Since contradiction is overcome by 'the connecting middle' entitled imagination we can now see that the ambiguity between imagination and the schematized understanding to be found in the first Critique is matched in Schelling's reading of the threefold Critical synthesis by the further ambiguity on the possibility of which we touched in our opening paragraphs. There is an ambiguity between fully fledged imagination and unschematized understanding in that the fully fledged form of the latter, the Idea of freedom presupposed by the categorical imperative, is practical reason whose application is guided by the former, that is to say, by an aesthetic Idea. The constructions of aesthetic intuition are the fulfilment of the constructions of mathematical and intellectual intuition of which philosophy becomes conscious when it reflects on the self-consciousness of the 'I am'. So philosophy's becoming conscious of this is its becoming conscious of its own limitations. In the final section of this essay we shall examine in a little more detail why this is so and indicate in broad terms where the philosophy of philosophy which emerges from Schelling's reconstruction of Kant stands in relation to philosophy's past and future.
VI The aesthetic intuition of the poetic faculty (Dichtungsvermogen) objectifies in a work of art the intellectual intuition which operates unconsciously in the production of objects and in our movement among them in action. It is the document and organon which
198
Kant and His Influence
philosophy needs to make manifest the absolutely simple identity of the conscious and the unconscious which, because it is prior to predication, cannot be fully described and must therefore be otherwise presented, exhibited or performed (dargestellt). However, if philosophy can make this manifest only through art, the idea of philosophy as the science of sciences will have to be revised. If Schelling wishes to say both that art is the ancilla of philosophy, is paramount for philosophy, and that it is paramount over philosophy, as might seem to be indicated if, as he says, philosophy returns to poetry, logos to mythos, philosophical constructing and mythopoietic intuition fuse in what we shall have to think of as the heir to what used to be called philosophy. That this is so is suggested by the following sentences from the last pages of the System where it seems that both the poetic and the philosophical genius become absorbed into the political community: What intellectual intuition is for the philosopher, aesthetic intuition is for his object. The former, since it is necessary purely for purposes of that special direction of mind which it takes in philosophizing, makes no appearance at all in ordinary consciousness; the latter, since it is nothing else but intellectual intuition given universal currency, or become objective, can at least figure in every consciousness. But from this very fact it may also be understood that, and why, philosophy as philosophy can never become generally current. The one field to which absolute objectivity is granted, is art. Take away objectivity from art, one might say, and it ceases to be what it is, and becomes philosophy; grant objectivity to philosophy, and it ceases to be philosophy, and becomes art. Philosophy attains, indeed, to the highest, but it brings
'Imagination' in Schelling's Reconstruction of Kant
199
to this summit only, so to say, the fraction {ein Bruchstiick) of man. Art brings the whole man, as he is, to this point, namely to knowledge of the highest, and this is what underlies the eternal difference and the marvel of art.24 Now for Schelling, as we have seen, the most marvellous marvel of art is mythology. Mythology is accorded this pre-eminence because at the end of its history philosophy, the 'whole science' referred to in the quotation from Schelling with which we began, returns to the mythological source from which it sprang. Philosophy was born and nourished by poetry in the infancy of knowledge, and with it all those sciences it has guided toward perfection; we may thus expect them on completion, to flow back like so many individual streams into the universal ocean of poetry from which they took their source. Nor is it in general difficult to say what the medium for this return of science to poetry will be; for in mythology such a medium existed, before the occurrence of a breach (Trennung) now seemingly beyond repair. But how a new mythology is itself to arise, which shall be the creation, not of some individual author, but of a new race, impersonating, as it were, one single poet - that is a problem whose solution can be looked for only in the future destinies of the world, and in the course of history to come.25 That is a problem we cannot pursue here beyond observing that the future destinies of the world and the course of history will see this idea of philosophy realized 24 STI, p. 233, SW III, p. 629. * STI, pp. 232-3, SW III, p. 629.
200
Kant and His Influence
less in the conceptual thinking of the author of the Science of Logic than in the denkendes Dichten of the third of the three Tubingen friends, Holderlin, and in the dichtendes Denken to which Holderlin provoked Heidegger at a time when talk of mythology and race was once more in the air. Heidegger, following Schelling and Kant, is much given to delving for hidden roots, in particular the root shared by theory and practice. We have been examining why, in the paragraph from the Philosophical Letters on Dogmatism and Criticism which we cited at the outset, Schelling believes this root to be imagination. When in that paragraph Schelling writes with reference to imagination, whose Schweben can mean oscillation but also suspense, that it voluntarily assumes a state of passivity toward an object, it is not difficult to hear his words as an expression of the belief that the common root is less likely to be disclosed by grasping (Begreifen) than by letting be [Seinlassen). In the Further Expositions of my System of Philosophy (1802) Schelling distinguishes ratiocinative Begreifen which explains by subordination, from the non-hierarchic thinking of identity, where the 'of signifies a genitive which is both subjective and objective, a 'speculative' genitive we could perhaps say, where the equal right and value of opposites is acknowledged, and where each thing is allowed to be {seynlassen) in its power the same as the other is in its.26 In learning how imagination is a connecting middle for Schelling and Kant we learn how Schelling is a connecting middle between Kant and Heidegger.
26
SW IV, p. 344.
Imagination' in Schelling's Reconstruction of Kant
201
Bibliography Abbott, T.K. (trans), Kant's Critique of Practical Reason and Other Works on the Theory of Ethics, 6th edn (Longman: London, 1909). Fichte, J.G., Science of Knowledge (Wissenschaftslehre), with the First and Second Introductions, trans and edited by Peter Heath and John Lachs (Appleton/Century/Crofts: New York, 1970). Fichte, J.G., Gesamtausgabe, I, 2, Reinhard Lauth and H. Jacob (eds) (Stuttgart-Bad Cannstatt: Fredrich Frommann Verlag, 1965). Hegel, G.W.F., Phenomenology of Spirit, trans A.V. Miller (Clarendon: Oxford, 1977). Heidegger, Martin, Beitrdge zur Philosophie, Gesamtausgabe, 65 (Klostermann: Frankfurt am Main, 1989). Heidegger, Martin, Kant and the Problem of Metaphysics, trans James S. Churchill (Indiana University Press: Bloomington, 1962) [Kant und das Problem der Metaphysik (Klostermann: Frankfurt am Main, 1951)). Hoyer, Carl H., Treatise on Philosophical Construction (Silverstolpen: Stockholm, 1801; German version, F. Perthes: Hamburg, 1801). Kant, I., Critique of Pure Reason, trans N. Kemp Smith (Macmillan: London, 1964). Schelling, F.W.J., System of Transcendental Idealism (1800), trans Peter Heath (University Press of Virginia: Charlottesville, 1978). Schelling, F.W.J., The Unconditional in Human Knowledge: Four Early Essays (1794-1796), trans Fritz Marti (Bucknell University Press: Lewisburg/Associated University Presses: London, 1980). Schelling, F.W.J., 'On Construction in Philosophy', Critical Journal of Philosophy (1802). Schelling, F.W.J., Sdmtliche Werke, K.F.A. Schelling (ed.) (J.G. Cotta: Stuttgart and Augsburg, 1856-61).
THE EARLY RECEPTION OF KANT'S THOUGHT IN ENGLAND 1785—1805 Giuseppe Micheli
1
Introduction
It is only from around 1830 that one can speak of any substantial influence of Kant's thought in England and of its reception in an adequate way, that is from the time of the change which took place in English culture in connection - if not always directly - with the renewal of society and of the educational institutions and also with the political battles of that period. The only outstanding exception was represented by Coleridge, who was one of the few who had any direct knowledge of the Kantian texts and of the first idealists, in particular Fichte and Schelling; he was to use Kant's thought to work out, especially from 1818 onwards, his own personal, although not systematic, philosophical outlook of a Platonic type (cf. especially Chinol, pp. 37-44); but his writings which can be more strictly described as philosophical remained mostly unpublished, and his influence, while it was important in the field of Anglican theology, was conversely insignificant in that of philosophy, at least for some considerable time (cf. Muirhead, Coleridge, p. 256; Chinol, p. 121). 202
Early Reception in England 1785-1805
203
However, Kant was already being spoken of in England from the last decade of the eighteenth century. Opinions about Kant, about his philosophy and, in general, about German thought and culture, developed and began to circulate, linked to elements of the earlier philosophical tradition, in the context of the very widespread cultural and political debate of the time. The area in which this impression was formed, affecting the reception of Kant's thought in English society at that period, was educated public opinion which had its most important sources in the cultural and political circles of London. The first reception of Kant's ideas in England took place completely outside the sphere of the universities, since at that time and at least until the end of the 1820s, - when a slow process of renewal of the English university system began with the foundation of Imperial College (1826), the first nucleus of the new University of London - the universities of Oxford and Cambridge were, as far as philosophy was concerned, closed environments, especially Oxford, and exerted relatively negligible influence on the cultural debate and on the development of educated public opinion (cf. Sidgwick, pp. 235-43; Pattison, pp. 83-97). There was a different situation in Scotland, where the universities, more closely linked to continental models, had already, throughout the eighteenth century, played a much more direct role in the process of the development of culture (cf. Veitch, especially pp. 83-4, 207-23). At Cambridge Kant's thought was to be introduced in the 1830s by William Whewell whose epistemology, at least in its first formulation (we refer to the first editions of the History of the Inductive Sciences of 1837 and of the Philosophy of the Inductive Sciences of 1840) derived in many aspects from Kant's first Critique. Whewell's
204
Kant and His Influence
attempt to introduce Kant to Cambridge was substantially to fail, in as much as it was too far from the tradition of English thought which was more alive at that time. As far as Oxford was concerned, it would have to wait for the teaching of Henry L. Mansel who, with the aim of combating both utilitarianism and idealism, was to develop, with a theological and religious emphasis, Hamilton's philosophy of the 'conditioned', referring extensively, even if in somewhat extrinsic ways as, indeed, did his master himself, to the teachings of Kant. At Oxford a second factor in the revival of German philosophy, though Hegelian more than Kantian, was signalled by the renewal of classical studies, thanks to the teaching of Benjamin Jowett, which led, by way of the study of ancient thought, to an encounter with Hegel and with the historiographical tradition of German idealism. Among Jowett's pupils were Thomas Hill Green, Edward Caird and William Wallace, who in the second half of the century produced monographs on Kant which had a wide circulation and contributed towards the strengthening of an idealistic interpretation of Kant in England (cf. Robbins, pp. 42-9). Thus, in order to have a picture of the first reception of Kant's thought in England, one must look not to the two universities, of Oxford and Cambridge, but to the cultural life of the capital. It will be found, somewhat surprisingly, that for a brief period at the turn of the century Kant was known, explained, discussed and even translated in England more than in other European countries such as France and Italy. However, this period of Kant's popularity was interrupted quite suddenly and rather abruptly at the beginning of the new century. In truth, the reception of Kantianism in England immediately encountered an obstacle in the dominance of the development of Lockian empiricism and, in the ethical field, in
Early Reception in England 1785-1805
205
the spread of the utilitarianism of Paley and Bentham. But what impeded the reception of Kantianism in England in the first place was the sudden, very strong and violent reaction against ideas originating from the Continent, and in particular from Germany, a reaction which took place at the end of the century. Fear of Jacobin contagion and the long war with France finished by isolating England from the Continent, and from that moment, for more than thirty years, Kant's thought, and more generally German philosophy, no longer had any part in the English cultural debate.
2
The Journals and the Spread of Ideas in the Age of Enlightenment
The origin of the picture of Kant held by English public opinion cannot easily be traced, for the same reasons that make it difficult to reconstruct the origin and spread of many opinions about philosophical doctrines which were widespread during the Age of the Enlightenment. The birth and diffusion of literary journals, in particular of reviews, had led to an impressive growth in printed matter and an increase in the speed of its circulation, which produced a change in the ways in which ideas were transmitted and received. A different public, larger, certainly less qualified, but equally eager for literary curiosities, was formed. As a consequence, this ensured a certain trivialization in the reports, due both to the increasing amount of information available and also to the simplification of its content in order to make it enjoyable and comprehensible to the new public. The public wanted to be informed by means of the literary journal of everything that was being published in every field of knowledge. Thus, a philosophical doctrine would be transmitted through the channel of literary information
206
Kant and His Influence
together with innumerable other items of news culled from the world of the arts, letters and science: not only that, but the literary journals of the period also gave much space to political events of the day, both at home and abroad. Information about philosophy was presented within a context that affected the way in which it was received. The importance of the literary journals as instruments for the diffusion of ideas in the Age of Enlightenment has been fully emphasized. Not infrequently, many works, especially of philosophy and theology, were known, abroad in particular, more through the reviews published in the journals than by direct reading of the texts. The role of the literary journals increased in importance even more from the middle of the century, with the birth of a new sort of literary journal. In London in 1749 The Monthly Review, of Whig leanings, began publication on the initiative of Ralph Griffiths (cf. Sullivan, I, pp. 2317). It was followed in 1756 by The Critical Review, of Tory leanings except for the period 1791-1803, the period we are considering, during which it took a definitely liberal position (Sullivan, I, pp. 72-7). Here we are speaking of journals of a new type compared with their predecessors - review everything and promptly: this was the main distinctive characteristic of the new literary journals compared with the more academic ones of the first half of the century (cf. Roper, especially pp. 19-49). The new journals completed their literary survey with periodical reviews of the state of literary and scientific culture in the other European countries: under appropriate headings notice was given of foreign books which were imported and available from the main booksellers in London. Often the issue concluded with a report of the principal political, cultural and scientific events of the moment, in the country and abroad.
Early Reception in England 1785-180S
207
Nicolai's Allgemeine Deutsche Bibliothek appeared in
Germany, on the model of Griffiths' journal, followed by very many others, both in England and on the Continent. The exchange of information was helped by the links between journals of similar political outlook. For example, the main English radical journal, Joseph Johnson's Analytical Review, between 1785 and the end of the century systematically used for information on Germany the reviews published in the Allgemeine Literatur-Zeitung of Jena, which had first appeared in 1785 and had immediately become the organ of the diffusion of Kantianism or, rather, of what was understood as Kantianism at that time at Jena. For the present research, about fifteen English journals have been examined (together with three Scottish periodicals). Some had a relatively short life while others lasted for much longer periods and allow the researcher to have a comprehensive impression of English cultured life and of the transformations it underwent in the course of time. We possess also the data about the number of copies sold for each of the principal English journals at the time. In 1797 these ranged from 5,000 copies of The Monthly Magazine and The Monthly Review to 1,500 copies of The Analytical Review, giving a total of 31,000 copies sold per month of the main London journals (cf. Timperley, p. 795); the figures relating to 1806 are more or less the same (cf. ScotsMag, LXVIII, 1806, p. 691). The systematic examination of this material makes it possible to assess the characteristics and the extent of the relationship between English and German culture and, within this relationship, to examine in detail the origin of the picture of Kant which people had in England, the connections established at that time between Kant's philosophy and elements of the previous philosophical tradition and, finally, the change in ideas about Kant in
208
Kant and His Influence
relation to the evolution of the cultural and political situation in England.
3
German Culture and the English Public
The educated English public does not seem to have shown a particularly strong interest in Germany until the end of the 1770s. For dynastic reasons, there were privileged connections with Hanover and with the university of Gottingen, but until the end of the 1780s it was thought of almost solely for scientific studies, in particular in the field of medicine and biology, and for Biblical scholarship. Even in 1798 when Coleridge would decide to travel to Germany, it would be in order to pursue medical and Biblical studies (cf. Chinol, pp. 256). Where philosophy was concerned, Gottingen, on the contrary, performed a function in the opposite direction, contributing not a little to the diffusion of British, English, and especially Scottish, philosophy on the Continent (cf. Kuehn, pp. 70-85). The English public began to display greater interest in Germany from the 1770s, but it was only with the new German literature, starting with Goethe, that the English public suddenly discovered Germany. The phenomenon assumed striking dimensions during the 1780s and then, above all, in the last decade of the century. Goethe's The Sorrows of Werther was first translated into English in 1779; twenty years later in 1799, at least twenty-four editions of Werther can be counted including translations and reprints (Morgan, pp. 156-7). In the two years 1798-9 alone, there were about one hundred English editions of the plays of Kotzebue which are today universally considered to be of little merit (Morgan, pp. 2809). The phenomenon can be explained only by the fact that the new German literature, whether good or
Early Reception in England 1785-1805
209
mediocre, succeeded in expressing better than any other at that time the feelings of the English reader, and the expectations, widespread among the middle class which made up the readership of the journals, of radical changes in the political, social and religious field (cf. Morgan-Hohlfeld, pp. 37-42). During the last two decades of the century, English society was in a state of excitement and deep crisis. In 1789 the centenary of the Glorious Revolution was celebrated in England, and not only in radical circles but also in the opposition Whig sector the anniversary was considered an opportunity for developing the political system in a more liberal direction. The events in France served to kindle the fire of debate even further. In public opinion the celebration of the centenary was linked with the memory of the recent American revolution and with the news that was coming from Paris. In his famous speech, delivered on 4 November 1789 to the Society for the Commemoration of the Glorious Revolution, Richard Price advised them 'that though the [English] Revolution was a great work, it was by no means a perfect work' (Price, p. 35). He urged them to consider 'the favourableness of the present times to all exertions of the cause of public liberty'; and concluded: what an eventful period is this! . . . After sharing in the benefits of one Revolution, I have been spared to be a witness to two other Revolutions, both glorious. And now, methinks, I see the ardour for liberty catching and spreading, a general amendment beginning in human affairs, the dominion of kings changed for the dominion of laws, and the dominion of priests giving way to the dominion of reason and conscience (Price, pp. 49-50). The conviction that the time was now ripe for a decisive
210
Kant and His Influence
change of direction in the history of mankind was widespread in the radical and liberal sectors of public opinion. The same situation was noted on the Continent also, especially in Germany; what Kroner (I, p. 2), speaking about Germany at that same period, defines as 'the eschatological state of mind' {die eschatologische Stimmung) was observed also in England. The reception of Kantianism in England and everywhere in Europe was to be profoundly affected by it. On the philosophical level, English culture was not in a very favourable condition for receiving Kant's philosophy. Unlike the situation in Scotland, the dominant tradition in England was that of Locke in the version of Hartleyan associationism. In 1775 Joseph Priestley, with his book Hartley's Theory of Human Mind (2nd edn 1790; 3rd edn 1794) had, with notable success, brought back into circulation the teachings of Hartley whose best known work, Observations on Man, first published without much success in 1749, was reprinted in 1791 and then republished several times. The rediscovery of Hartley brought back into prominence a thinker who had been outside the development of thought in his time; although he was writing in 1749 it seems that he knew only Locke and Newton. Hartley's teaching was received with approval by a large part of English cultural circles. The success of Hartley can, in fact, be recorded as part of a vast European phenomenon; on the Continent, the success of Condillac (and of Bonnet) corresponded with that of Hartley. Hartley himself was translated into German in 1772-3, into French in 1802 and into Italian in 1809. In the field of ethics, utilitarianism was predominant. In 1785 William Paley published his Principles of Moral and Political Philosophy, a work destined to be the textbook on morals at Cambridge for more than half a
Early Reception in England 1785-180S
211
century. Four years later, in 1789, Jeremy Bentham published his Introduction to the Principles of Morals and Legislation. Utilitarianism, which Paley reconciled with religious orthodoxy, was taken in the direction of political radicalism by Bentham and Priestley. In place of the dogma of natural rights, common to other radical thinkers of the time, they postulated the maxim of the greatest happiness of the greatest number as the criterion for choice in political questions. If it had not been for the attitude of expectation of a change of direction in the social, religious and political field, which made educated public opinion watchful towards anything in the area of politics as in that of knowledge that could seem to announce the new world, English philosophy at the end of the century, like much of European culture, would not have provided any chance of a favourable reception of the originality of Kant's thought.
4
The First News of Kant
The first references to Kant appeared in the English periodical press in this more generalized context. In 1787 two periodicals of liberal tendency, The English Magazine (IX, January 1787, pp. 66-7) and The Political Magazine (February 1787, pp. 94-5), published 'a letter from a temporary resident of Weimar' which attempted to provide a firsthand account of German culture at that time. To the English traveller Germany seemed more interesting than France, and full of new things: the German taste is more akin to ours than the French. France can boast of no such poet as Klopstock, who may, in many respects, be put on a level with Milton. Gessner is all sweetness and simplicity. . . . Their dramatic compositions are but moderate. Pretending
212
Kant and His Influence
to imitate Shakespeare, many of their writers have mistaken extravagance for genius, and unintelligible absurdity for the sublime {EnglRev, IX, 1787, p. 67). The letter went on to speak about Goethe and Wieland. Towards the end, the English traveller inserted a brief reference to Kant: the metaphysics of professor Kant, at Konigsberg, gain ground. A parson in Brandeburg has been suspended from his office for endeavouring to prove to his peasants, that, upon Kant's principles, there exists no God {EnglRev, IX, p. 67). The text indicates something more than a simple curiosity: it bears witness to the beginning of a change of attitude towards German culture. Not only does it contain, in all probability, the first reference to Kant to appear in an English journal, but it links the name of Kant with opinions which were often to be repeated later. The new German literature was appreciated but it was accused of, at times, mistaking unconventionality for cleverness; Kant's philosophy was linked, though in very vague terms, with the metaphysical topic of the existence of God, and with the problem of irreligion. A little earlier the writer of the letter had spoken of the obscure terminology of the German philosophers, which was to be another recurrent theme in opinion about Kant: in their philosophical writings, they have adopted an arrangement of phrases so complicated and unwieldy, that, without long practice, a stranger can never understand them {EnglRev, IX, p. 66). From 1787 onwards Kant was referred to with increasing frequency. For the period 1780-1820 more than a hundred references, obviously very varied in their length
Early Reception in England 1785-1805
213
and relevance, can be found to Kant and to his philosophy. Between 1787 and 1795 references were still infrequent and, apart from four reviews in The Monthly Review, of little significance. The period of greatest interest in Kant was in the two years between the end of 1795 and the beginning of 1798, a period which coincided with the publication of the essays on Kant, of very varying worth, by O'Keeffe (November 1795), Nitsch (1796), Richardson (1797) and Willich (January 1798), with the translation of Kant's essay on Perpetual Peace (1796), and with the interest in Kant shown by the journals with a liberal and radical orientation. This period was followed by a second, which went from the middle of 1798 to the early months of 1801 and which was characterized by the violent polemic conducted against German philosophy by the conservative and reactionary journals, in particular The Anti-Jacobin Review. After 1801, references to Kant rapidly decreased and few journals recorded Kant's death in 1804. From 1806 onwards hardly anything was said about Kant nor about German idealistic philosophy. It was only after 1820 that Kant began to be mentioned again in English journals.
5
Kantianism in Holland and its Echoes in England: 1790-1794
The most widely circulated and authoritative English literary journal, The Monthly Review, had already had the means to provide occasional, but nonetheless significant, information about Kant's philosophy to its own readers before the period of maximum interest on the part of public opinion which covered the years between 1796 and 1800. Holland had become a centre of study of Kant's philosophy of some importance, first with Paulus van Hemert
214
Kant and His Influence
and then with Hans Kinker (cf. Land, pp. 98-9; Groenewegen, pp. 304-15; Wielema, pp. 450-61; see also the letter of Johannes Glover to Kant, Ak., XII, pp. 337-40). Difficulties due to language had to some extent reduced the influence of these Dutch centres. Kinker's work would become an important vehicle for spreading knowledge about Kantian philosophy in the whole of Europe only when it was translated into French, in 1801, by Jean Le Fevre, and taken up, on the basis of this translation, by Destutt de Tracy in a paper read in 1802 to the Institut National. However, Holland still had close ties with Great Britain, which were expressed also in frequent contacts and exchanges on the level of culture. On the religious level there were links not only with the Scottish Presbyterians but also with English dissenting groups, whose ministers and most prominent leaders, being excluded from Oxford and Cambridge - although by this time they had their own academies - continued not infrequently to make use of the Dutch universities, in particular that of Ley den, in order to complete their studies. Griffiths' journal, because of its liberal tendency in politics and its openness towards and support for English dissenters and, moreover, because of the editor's vast network of friendships, took advantage of the presence of these English people in Holland so as to have reviews of Dutch literary output in all fields of knowledge. Often these correspondents, because of their good understanding of German, provided reviews of German texts also. The French occupation of Holland in 1795 was to interrupt these contacts. 5.1
Cogan on Allard Hulshoff: Kant and Scottish philosophy
In December 1790 Thomas Cogan, minister of a presbyterian congregation and physician, sent to Griffiths'
Early Reception in England 1785-1805
215
journal a long review of four dissertations on the subject of the immateriality of the soul which had been awarded prizes by Teyler's Society of Haarlem. The author of the first dissertation, the Mennonite pastor Allard Hulshoff (cf. Ak., XI, pp. 188, 196), who was later to contribute to the spreading of Kantianism in Holland (cf. Wielema, pp. 452-3, who however does not mention this dissertation), had approached the subject by taking it, according to Cogan, from the point of view of the philosophy of Thomas Reid and, as a premise, had devoted the whole of the first part of his dissertation to the classification of the various theories concerning the matter in relation to the problem of the origin of knowledge (cf. Hulshoff, pp. 1-11). Among the philosophers quoted by Hulshoff (the reviewer mentioned) was Kant, whom he placed among the 'idealists', that is to say those philosophers who, with Berkeley, rejected the Lockian distinction between primary and secondary qualities, and assert that neither impenetrability nor extension can have place, independently of our conceptions (MonthRev, III, p. 483). Kant's idealism was, however, of a much more moderate kind compared with that of Berkeley: Kant, who derives no sensations from the attention of the soul to real existences, though he acknowledges that our ideas are in some other manner excited, by something existing out of the mind, is placed [by Hulshoff] under the idealists; as, according to him, all our observations and determinations are founded on appearances {MonthRev, III, p. 484). In his classification, the author of the dissertation seemed to be suggesting some kind of analogy between the position of Kant and that of Reid: both rejected the Lockian
216
Kant and His Influence
explanation of the distinction between primary and secondary qualities but they did not go on from there to maintain, with Berkeley, that material objects do not exist outside the mind. According to both of them, there is within us a subjective principle which makes us believe in the existence of something outside ourselves. The author of the dissertation does not seem to have developed the subject any further, and the author of the review restricted himself to noting it in these general terms. Nevertheless, the brief mention is not without significance, and this for two reasons. It represents an early associating of Kant's philosophy with that of Reid, on the part of an author described as a convinced adherent of the Scottish school. Through the review this possible interpretation of Kantian thought was suggested to the English reader, and it was presented to him linked, even though in very vague terms, with elements of the philosophical tradition that must have been familiar to him, that is Reid's polemic against Locke, the reasons underlying Berkeley's idealism and the foundation of belief in the existence of the corporeal world. 5.2
Benjamin Sowden: the first English exposition of Kant's system
In the issue of April 1793 a second and fuller reference to Kant's thought was published (MonthRev, X, pp. 523-31; cf. Wellek, pp. 5-7). The author in this case also was an English dissenter in Holland, the Presbyterian Benjamin Sowden, minister of the English congregation in Amsterdam and a chemist, who provided a review of three essays presented for a competition arranged by the philosophical society of Haarlem on the subject: What is the validity of the moral demonstration of the existence of the Deity, and particularly of that which
Early Reception in England 1785-1805
217
professor Kant has proposed as the only one? {MonthRev, X, p. 524). The three essays which were awarded prizes and published were by, respectively Johann Christian Schwab, collaborator of Eberhard and Feder and an anti-Kantian, Friedrich Daniel Behn of Liibeck, also definitely antiKantian, and Ludwig Heinrich Jakob, who was the only one to defend Kant's position {Verh. Holl. Maatschappye, XXVIII, pp. 1-50 [Schwab], pp. 51-175 [Behn], pp. 177-325 [Jakob]; cf. Adickes, Nos 352 and 938; Wielema, pp. 451-2 and the letter of Hulshoff to Kant, Ak., XI, pp. 259-61). From the beginning of his review the author did not hide his surprise at the fact that the Haarlem Society, which since 1754 had always been concerned with scientific and technical themes, had announced a competition on a metaphysical question. He did not disguise his opinion about research being done in the field of metaphysics: we are sorry to find that this society seems at present to have forsaken the delightful and fertile field of practical philosophy and useful science, for the barren and disgusting desert of speculative metaphysics, in which the mind . . . is perplexed with scholastic subtilties. . . . In no pursuit are men more apt to deceive themselves by mistaking words for things; they invent a new technical language, and, because they express themselves differently from others, imagine that they have found a more certain road to the sanctuary of truth, while, perhaps, they are only bewildering themselves in a labyrinth of their own construction, in which they fancy that they are approaching the object of their search, by paths which really conduct them away from it, and lead them into uncertainty and confusion (MonthRev, X, p. 523).
218
Kant and His Influence
There is no doubt that in Benjamin Sowden's opinion, Kant was the representative of such a type of metaphysics, the rapid spread of which in Germany he noted with some anxiety: these observations may, we think, be applied to the system of metaphysics and moral philosophy which was published, about twelve years ago, by Professor Kant of Konigsberg, which is at present much in vogue among the German literati, and is extolled by its adherents as one of the most sublime efforts of human genius {MonthRev, X, p. 523). The reviewer concentrated his attention on the essays of the two anti-Kantians, Schwab and Behn, and in particular, for reasons of space, on that of the former, which contained comments on the moral proof formulated by Kant which he at once declared he shared entirely. Nevertheless in order to enable our readers to enter more fully into the merits of the question, as well as to gratify the curiosity of the learned concerning a philosophical system, which . . . is very little known in England, it may not be improper to premise a brief view of some of M. Kant's fundamental principles of metaphysics and morals, on which the validity of his argument for the existence of the Deity in a great measure depends. To do this, we are obliged to have recourse to the works of this philosopher, entitled Critik der Reinen und der Praktischen Vernunft, in consulting which, we are much assisted by a very judicious account of his sentiments, published in a kind of Dutch magazine, which, we believe, was drawn up by the ingenious Professor van Hemert of Amsterdam {MonthRev, X, p. 524).
Early Reception in England 1785-1805
219
In a footnote Sowden made it clear that he had at hand both the second edition of the Critique of Pure Reason and the first edition of the Critique of Practical Reason. The text of Paulus van Hemert to which he referred must have been his first article on Kant published in a Dutch journal (cf. Groenewegen, p. 309; Wellek, pp. 6, 266; Ak., XIII, p. 524). The premises for a good exposition of Kantian thought did exist: the reviewer had access to texts in the original language and we know that he understood German well. The exposition of Kant's thought occupies about three pages and is accurate enough even if synthetic, given that it had to serve as an introduction to the subject of the moral proof of the existence of God which was discussed in the texts which he had to review. 5.2.1 Kant's theoretical philosophy In his opinions about the theoretical part of Kant's philosophy, Sowden was probably influenced by suggestions that came from Schwab who, like Eberhard and Feder, interpreted Kant's thought in the Berkeleian idealistic sense, as he denied all objective reality. Nevertheless, when expounding the thesis of the Aesthetic, Sowden seems to have been aware of the Kantian distinction between 'transcendental idealism' and 'empirical realism': for Kant all men have a certain innate faculty, by which they become capable of the knowledge of those things which fall under the cognizance of their senses, and are conceived under the forms of time and space . . ., which, when abstractedly considered, Kant calls pure perceptions, are only forms of perception, and not real existences; they are not essential to the absolute existence of things, nor even to those relations of external things that are independent of our knowledge. As
220
Kant and His Influence
space is nothing more than the form of our knowledge within our own minds, so the objects, which we perceive in space, exist not externally, but only internally; they are mere phenomena, but cannot be said to be only ideal, nor to have no objective reality, because they depend on established laws and real principles. When, therefore, they are said to exist, no more is meant than that they are perceived in space, or in the form of external organization {MonthRev, X, pp. 524-5). Also when he went from the Aesthetic to the Analytic, his explanation of Kant's thought, although extremely synthetic, is not without accuracy: as the nature and mode or form of our perception are determined by the nature of our sensible faculty, so the form of our thoughts, or the manner in which we judge concerning phenomena, or arrange our perceptions, is determined by the nature of our theoretical reason; and . . . that by which we determine the connection of our observations, and form a judgment concerning them, is called a pure notion or category. Those pure notions, which are discoverable by an analysis of the judgment, may be reduced to notions of quantity, quality, relation, and modification [sic]. These categories, considered abstractedly, are not deduced from our perceptions and experience, but exist in the mind prior to these latter, and experience is the result of their combination with our perceptions; - but it is only in connection with our perceptions, that these pure notions can be the source of knowledge; for in themselves they are mere forms, without any independent existence. They serve to direct us in the use of our observations, but they cannot extend our knowledge beyond the limits of perception and experience [MonthRev, X, p. 525).
Early Reception in England 1785-1805
111
Sowden did not discuss the Dialectic: he would refer to the subject of God later when expounding Kant's practical philosophy. The examination of the theoretical part of Kant's philosophy concluded with a brief reference to the relationship between experience, synthetic propositions and science: there are . . . two kinds of propositions, concerning which our minds may be employed, analytical and synthetical. The former are those in which we only explain or illustrate that of which we have already some idea; whereas, in the latter, we increase our knowledge, by adding something new to our former idea of the subject. . . . Without experience, we cannot form any synthetical proposition concerning the objects or matter of our knowledge: but, as the forms of our knowledge are independent of and prior to our experience, we may, with respect to the pure notions already mentioned, conceive synthetical propositions, or acquire pure science; and indeed it is only when we have pure perceptions [space and time! and pure notions [categoriesl for our objects, that we can arrive at universal and necessary certainty; . . . in pure mathematics and philosophy . . . we consider truth, abstracted from matter, with respect only to the forms or laws of knowledge and volition {MonthRev, X, pp. 525-6). Here- the text becomes definitely ambiguous; the conciseness of the exposition leads to misunderstanding. The Kantian distinction between sensibility and understanding, and between the receptivity of the former and the spontaneity of the latter, is not clear; the categories, set beside the pure intuitions, could easily have been interpreted as preconstituted forms rather than, as in Kant, as dispositions of unifying activity.
222
Kant and His Influence
Kantian philosophy, in such a way, lent itself to being interpreted in an innatistic sense. 5.2.2 Kant's practical philosophy The reference to practical philosophy was very brief, serving as a direct introduction to the moral proof of the existence of God which was discussed in the texts which Sowden had to review. Beside theoretical reason, Kant ascribes to man another faculty, which he calls practical reason, endued with power sufficient to impel and direct the will . . .; if this faculty were not granted, . . . practical laws would not be universal moral precepts, but only particular maxims, which individuals might prescribe to themselves as the rule of their conduct. To these universal moral laws, practical reason commands our implicit obedience, without any regard to our inclinations or views of advantage. These are indeed sometimes at variance with the dictates of duty, but, in order to diminish their influence as obstacles to virtue, our practical reason must determine us firmly to believe the existence of the Deity, and of a future state in which our happiness will be proportioned to our internal worth. This is what our philosopher calls rational faith, as it is independent of all knowledge of its object (MonthRev, X, p. 526). Here Sowden, who reduced the exposition of practical philosophy to the theme of the existence of God, as a postulate of practical reason, and to the connected theme of the highest good as a perfect balance between happiness and virtue, in substance a summary of §V of the second chapter of the Dialectic of practical reason (cf. Ak., V, pp. 124-32), seemed to interpret faith in God as an instrument of reason in order to oppose the
Early Reception in England 1785-1805
223
inclinations which are hostile to morality. The theme of God and of the highest good, if expounded separately from the totality of the problems of the second Critique, could easily be linked to the speculative agnosticism of the first Critique, but it also lent itself to an interpretation of a fideistic type and particularly, as in fact was to happen, to misunderstanding of the Kantian concept of practical reason. In Kant's view the principles of religion can be neither demonstrated nor disproved by theoretical reason, but are mere postulates of practical reason; and the only theology, that is really founded on our understanding, is moral theology, which depends on moral principles (MonthRev, X, p. 526). Sowden was not clear about what Kant meant by 'postulate' and, more generally, by practical reason. It seemed to him to be a new term, a distinction to which no