1. Introductory comment

I am grateful for all of these thoughtful comments on Kant’s Impact on Moral Philosophy.Footnote ¹ I am heartened by the general support for my project, and the criticisms that have been expressed are well worth reflection and discussion, certainly more than I will be able to provide here. But I will address several issues raised by my commentators.

My book covers a wide range of material, on all of which I could not, of course, be equally expert. I begin with an overall view of my approach to Kant’s moral philosophy and its development and then discuss the issues raised in the first decade or so of the reception of Kant’s work in moral philosophy, from the early reviews of Kant’s Groundwork for the Metaphysics of Morals (published in 1785) to the publication of the Metaphysics of Morals in 1797, Karl Leonhard Reinhold’s review of the Introduction and the Doctrine of Right, and Kant’s response to it. I then turn to major moments of the subsequent reception of and response to Kant’s work, first in the German philosophical tradition from Fichte to Nietzsche and then in the Anglo-American domain from Anglo-American idealism to John Rawls, his students and successors such as Thomas Nagel, Onora O’Neill, Christine Korsgaard and others, as well as to Derek Parfit, whom we might call, although I did not put it this way, a neo-utilitarian. The commentators have been kind enough not to mention obvious omissions: I had originally intended to include a chapter on Neo-Kantianism and at least one on Jürgen Habermas and such important successors of his as Axel Honneth and Rainer Forst (who studied with both Habermas and Rawls), as well as something on Søren Kierkegaard. But I had not read Kierkegaard since college, so I decided to forego a discussion of his work, and while I actually did write a chapter on Hermann Cohen’s version of Kantian moral philosophy and the response to it by Max Scheler (who never named his target), I ended up deleting it in response to the complaint of the referees that the book was too long. As for Habermas and his successors, I do think it would have been important to have included a discussion of them, in particular a discussion of the famous debate between Habermas and Rawls carried out in the pages of the Journal of Philosophy in the 1990s, but I ran out of steam as well as of space. I would be delighted if someone else were to pick up where I left off and cover that valuable material.

I will reply to each of the comments in turn, as far as is possible, in order of the parts of the book that they discuss. But first I will make two points about the scope of the work in addition to what I just said about its omissions, points that will be especially relevant to the comments by David Boucher and Andrew Vincent and by Catherine Wilson. First, as far as possible, I have confined my discussion to Kant’s moral philosophy in general and to his ethics in particular, excluding discussion of his doctrine of right. This is not because I do not regard Kant’s doctrine of right as an entirely proper part of his moral philosophy, for, as I have argued in many papers in the last quarter-century, some already collected in my 2005 collection Kant’s System of Nature and Freedom (Oxford) and a further, larger group now published as Selected Essays on Kant, Volume I: Duties of Freedom (Cambridge), I most certainly do. In Kant’s view, as in the view of every other theorist in the German natural law traditionFootnote ² from Pufendorf and Wolff through Achenwall as well as in the British tradition of Hutcheson, Kames, and Smith, moral philosophy in general has two proper parts: the part that covers moral duties that can and should be coercively enforced, that is, right; and the part that covers moral duties that should not and cannot be coercively enforced but must be fulfilled out of the personal motivation of individual agents, in Kant’s terms, their respect for the moral law, which constitutes ethics – a word that, at least in discussions of Kant (and these others), should never be used as a synonym for moral philosophy, for that is to confuse a part with the whole. Rather, I tried as much as possible to exclude Kant’s doctrine of right from coverage in this volume simply because, in my design for the whole series on the legacy of Kant, I thought that Kant’s doctrine of right and its impact needed a volume of its own, which, indeed, Howard Williams will soon publish. That volume will cover the reception of Kant’s doctrine of right by some of the same figures that I discuss, but also its reception by and influence on figures and movements that I do not discuss at all, such as Marxism and Neo-Kantianism, where the response to Kant’s political philosophy turns out to be much more interesting than the response to Kant’s moral philosophy in general and to his ethics. So I will decline to respond to Catherine Wilson’s request that I discuss ‘Kant’s positions, relevant to well-being and opportunity, in his explicit social and political philosophy’ and his ‘influence on later political theorists’. I do some of that in the collection on Duties of Freedom that I mentioned, and Howard will be doing much more of that in his book, which we should all be awaiting with great anticipation. But it is beyond the scope of the present volume.

Second, I want to clarify that by the impact of Kant’s moral philosophy I really meant the impact of Kant’s work on other philosophers, not, as Wilson also asked me to discuss, ‘on government policies relevant to well-being and opportunity’ or on intellectual history or actual mores outside of what we recognise as the academic philosophical tradition (even if some of the figures discussed were not, or for much of their lives were not, academics, such as Schopenhauer and Nietzsche). The criterion of selection, at least once I got past the first generation of reviewers who were largely other philosophy professors, now mostly forgotten (although a great deal of work is now being done on them, even since I wrote this part of the book), but obviously and explicitly responding to Kant, was whether the figures I discussed are now recognised as figures significant to a sufficient degree in the canon of academic philosophy since Kant. So on the one hand, I did not intend the book to be a history of scholarship on Kant’s moral philosophy, which would have been of interest to a very small readership, but on the other hand, I did not intend to write a history of the impact of Kant’s moral philosophy, for better or worse, on culture, politics, and history in general since Kant’s day. I had no intention of writing, as Wilson puts it, a history of ‘the impact of [Kant’s] texts on practices’ or of their effect ‘on the institutions and practices of the world’. That might well be a project of great interest but would be a project for a very ambitious intellectual historian, not an historian of philosophy.

Having made these comments, I now turn to my responses to the several essays.

2. Morgan on Kant and Reinhold

Seiriol Morgan’s main point is that my treatment of Kant’s response to Reinhold’s 1792 response to Carl Christian Erhard Schmid’s ‘intelligible fatalism’ (which I think is more properly Johann August Heinrich Ulrich’s position, which Schmid himself meant to criticise) is a ‘comedy of errors’ since they both held roughly the same position, misrepresenting a significant difference between Reinhold and Kant. Reinhold and Kant are both responding to the charge that Kant’s apparent position in the Groundwork of the Metaphysics of Morals and the Critique of Practical Reason that the moral law is the causal law of the free will makes it impossible to treat immoral action as the product of a free will and thus what is ordinarily taken to be the necessary condition for the assignment of responsibility, in Kant’s term ‘imputability’, for immoral action. In the Eighth Letter of his second series of Letters on the Kantian Philosophy published in 1792 and in the essay on the ‘radical evil in human nature’ first published in the Berlinische Monatsschrft in that year and then as Part One of Religion within the Boundaries of Mere Reason the next year, both Reinhold and Kant respond to this charge by distinguishing the freedom to choose between adhering to the moral law or violating it and our knowledge of the moral law itself as two different things. Reinhold calls this freedom of choice ‘absolute freedom’ and claims that our possession of such freedom is a ‘fact of consciousness’ without appealing either to Kant’s transcendental idealism to open up the possibility of such freedom or to his claim in the second Critique that our awareness of the moral law is a fact of reason but that our freedom to choose to act with or against it is deduced from that fact rather than directly experienced, thus it is anything but a ‘fact of consciousness’. Kant then goes on in the Religion to repeatedly assert our freedom to act with or against the moral law in spite of the knowledge of the moral law that we have; regularly uses the word Willkühr to designate our ability to choose for or against subordinating self-interest to the moral law; and then canonises that usage in the Introduction to the Metaphysics of Morals by explicitly asserting that it is Wille that gives us the moral law and Willkühr that freely chooses whether to act in accordance with or contrary to it, while Wille cannot properly be said to be either free or unfree at all, because it does not choose or act but only presents us with the moral law. Kant does not say so in either the Religion or the Metaphysics of Morals, but of course he takes it that the room for a Willkühr can be and has been opened up only by transcendental idealism. Apart from that, Reinhold and Kant seem to be pretty much on the same page: the moral law and our knowledge of it are one thing and our freedom to act with or against it another, which is then why Reinhold subsequently manifests astonishment in his review of the Introduction to the Metaphysics of Morals and its first part, the Doctrine of Right, at Kant’s claim that ‘freedom cannot be defined – as some have tried to define it – as the ability to make a choice [zu wählen] for or against the law (libertas indifferentiae), even though choice [die Willkühr] provides frequent examples of this in experience’ (MM, 6: 226). That is how Reinhold had defined freedom of choice and how he thought that Kant had at least tacitly defined Willkühr in the Religion – and, I would venture to say, how most readers think that Kant had defined Willkühr in that work.

Kant’s reasons for rejecting Reinhold’s definition seem strained and unconvincing: that choosing against the moral law is something that we are familiar with in our experience of ‘the human being as a sensible being’ means that it is something that we are familiar with from the phenomenal world and therefore that it cannot be used to define anything, a fortiori, the freedom of our own will ‘as an intelligible being’; and that ‘Only freedom in relation to the internal lawgiving of reason is really an ability[, but] the possibility of deviating from it is an inability’. But it is not clear why an inability should not be part of a definition or, perhaps even more fundamentally, why the choice to act against the moral law is not at the same time an ability of its own, the ability to act in accordance with a principle of self-conceit (in the terms of the second Critique) or to subordinate morality to self-love (in the terms of the Religion). And as to the claim that we cannot have a direct experience of the freedom of the intelligible self, well, that had clearly been Kant’s position all along, and his confidence in the Religion that we – or our Willkühr – are always free to choose to act for or against the moral law, to subordinate self-love to morality or vice versa, was clearly based on an assumption about the necessary condition for imputability, the condition for our being imputable for our evil as well as for our good. So why should the fact that we do not have direct experience of the noumenal Willkühr freely choosing evil suddenly become a problem for defining that freedom as including the ability to choose evil, that is, to choose actions or maxims contrary to the moral law?

Since Kant’s reasoning here seems so strained, I suggested that there must be something deeper underlying it and proposed that Kant’s real reason for excluding the freedom to choose contrary to the moral law in the definition of freedom is that he is demanding a ‘real’ definition that includes a ratio cognoscendi of the moral law, and only our ability to choose in accordance with the moral law supplies that; that is the gist of the ‘fact of reason’ argument from consciousness of the moral law to consciousness of our freedom in the Critique of Practical Reason, after all. Morgan does not directly address this part of my treatment of the ‘comedy of errors’ but does suggest that Reinhold ‘fails to realize that any conception of freedom that a Kantian could endorse must be such as not only to ascribe a capacity for moral action to the moral agent, but also to ground that action’s categorical normativity’. However, I think that is what I said, although in different terms. For Kant, a real definition of freedom will include a ratio cognoscendi of the moral law, and only the ability to choose in accordance with the moral law will do that; or conversely, our consciousness of the moral law must include a ratio cognoscendi of our freedom to act in accordance with it. The argument of Groundwork Section III is that our knowledge of our own spontaneity at the noumenal level is the ratio cognoscendi of the moral law; the argument of the ‘fact of reason’ passage in the Critique of Practical Reason is that our knowledge of the moral law is the ratio cognoscendi of our freedom. In either case, freedom has to be assigned to our noumenal self in order to avoid the threat of thoroughgoing determinism at the phenomenal level. But also in either case, freedom itself is, whatever should properly be included in its definition, freedom to choose one way or the other – the freedom to choose either good or evil. Or, having already chosen evil, nevertheless, the freedom now to choose the moral law in a ‘change of heart’. That’s the whole message of the Religion, so no wonder Reinhold was surprised to apparently be told otherwise. In any case, twists and turns aside, I thought that my interpretation of the distinction between Wille and Willkühr does include an account of the normativity of the moral law, or our knowledge of that normativity – that is the role of Wille – as well as a name for the possibility of choosing to act for or against that norm – that is the role of Willkühr. Whether that possibility should best be described as an ability to choose one way or the other or as an ability plus a disability seems like a side issue; Kant has clearly stated in the Religion that imputability for either our evil or our good demands that we be free to choose to act against the moral law or for it.

In connection with this discussion, I would like to warn against citing without contextualisation, as Morgan does, Kant’s remark at Groundwork 4: 412 that ‘the will [Wille] is nothing other than practical reason’. That would be true of us ‘if reason infallibly determines the will [den Willen]’, as Kant continues, but of course in our case reason does not infallibly determine the will, at least as a faculty of choice. That is why Kant is about to explain that the fundamental principle of morality that would present itself to any rational being presents itself to us in the form of a categorical imperative, a constraining ought. For us, the will is not identical to practical reason. That is why in the Introduction to the Metaphysics of Morals, Kant finally has to regularise the distinction between Wille, which is identical to practical reason but which does not act at all and is neither free nor unfree, and Willkühr, which does act, or choose, and is free, for better or worse. To be sure, Kant had not yet regularised that terminological distinction in the Groundwork, and at least at the beginning of Groundwork III he does treat the moral law as the causal law of the free will, where that seems to include the faculty for choice. So one can understand and sympathise with Ulrich’s (and Schmid’s) objection that Kant’s position is ‘intelligible fatalism’, even though even in the paragraph at Groundwork 4: 412–13 Kant had clearly stated that ‘as is actually the case with human beings … actions that are cognized as objectively necessary are subjectively contingent’, and in the latter part of Groundwork III he had described a struggle between pure practical reason and ‘sensible desires’ that can come out either way (G, 4: 454). But we who have the benefit of also having Kant’s Religion should not be confused on this point – even though notable Kantians of our own time have been, such as Christine Korsgaard, who, in Self-Constitution (Korsgaard Reference Korsgaard2009), attributes to Kant the view that moral failure is only a result of cognitive failure, a mistake about what is the right thing to do, but not of a failure to will the right thing, or a will to do the wrong thing.

3. Hassan on Nietzsche and Kant

In my chapter on Nietzsche’s response to Kant, I argue that in spite of Nietzsche’s hostility to the categorical imperative as well as various of his snide remarks about Kant – the ‘old Chinaman’ and all of that – there is more commonality between the two than Nietzsche recognised, namely that for both, human freedom is the ultimate value, for Kant explicitly in the form of the freedom to set and pursue one’s own ends and for Nietzsche under the guise of the morality of nobles rather than slaves, in the form of the Übermensch, etc. At the same time, of course, there is an obvious and tremendous difference between them, namely that for Kant, the freedom of everyone to set and pursue their own ends is of equal value, and that is why the goal of morality can be expressed as the empire of ends, ‘a whole of all ends in systematic connection (a whole both of rational beings as ends in themselves and of the ends of his own that each may set for himself)’ (G, 4: 433), and why, to turn to the Doctrine of Right after all, the ‘General Principle of Right’ is that ‘Any action is right if it can coexist with everyone’s freedom in accordance with a universal law’ (MM, 6: 230), while Nietzsche is concerned only with the freedom of the gifted and creative, even if its exercise should come at the cost of the freedom of others. (That is why Rawls, having taken Nietzsche as his exemplar of perfectionism, rejected perfectionism as a possible basis for a theory of justice.)

Hassan, however, argues, if I may put it this way, that Nietzsche’s exceptionalist perfectionism is actually more realistic than Kant’s insistence upon the freedom of each to set their ends in a way that is consistent with the freedom of all and thereby to realise the goal or ideal of morality. As he puts it, according to Nietzsche, but on this he clearly sympathises with Nietzsche,

Well, of course in his expositions of his doctrine of the postulates of pure practical reason in the first and second Critiques, Kant does defer the realisation of the combination of perfect virtue with perfect happiness to a world that is ‘future for us’ (A811/B839), a fantasy that Nietzsche could rightly criticise and that I have no interest in defending – but also note that Kant did not think that this postulation of an afterlife is necessary ‘for most if not all human beings’; he clearly thought that it was necessary for all human beings, thus that there was no room for a privileged sort of nobles or Übermenschen in this, our ordinary, earthly existence, as Nietzsche thought. But whether we should regard Kant as having undergone a fundamental change of view or rather think of the Groundwork as constructing stick figures, such as the philanthropist who has lost his sympathetic feelings and is now motivated only by his respect for the moral law, for the very specific purpose of clarifying the content of the moral law but by no means as expressing his actual moral psychology, it is not clear that Kant thought that in actuality there is an unavoidable and constant struggle between inclination and moral reason. Rather, he thought that a conflict between the two is always possible for human beings and that, therefore, moral human beings must, as Kant puts it in Religion within the Boundaries of Mere Reason (6: 35–6), make it their fundamental maxim to subordinate self-love to morality rather than vice versa, a maxim, however, which they will need to act upon only should morality and inclination come into conflict, which they by no means always do. Indeed, Kant also writes in the Religion that human predispositions, including both our ‘animality’, that is, what we share with other animals, such as the inclinations towards food and sex, and ‘humanity’, which in that context means our tendency to compare ourselves to others, are actually natural predispositions towards the good, which we can pervert if our fundamental maxim is to subordinate morality to self-love (6: 26–7), or, as Kant puts it in the Critique of Practical Reason, if we choose to elevate our ordinary, permissible self-love into the principle of ‘self-conceit’, but which we, none of us, by any means have to pervert. Indeed, the message of the Religion, which in my view does undercut Kant’s argument for the postulate of personal immortality, is that we are all free, at any time, to undertake the ‘change of heart’ from the maxim of evil to the maxim of good. This confidence in our freedom is of course based on Kant’s transcendental idealism, with which I have no truck and which found few defenders in the history that I trace, but it is certainly not a doctrine that only a few have the freedom to rise above inclination and that they should enjoy any privilege in real life over others who do not have such freedom. Rather, a more realistic view, let’s say between Nietzsche’s exceptionalist perfectionism and Kant’s radical libertarianism, would be that everyone has some potential to rise above their inclinations when those inclinations do conflict with morality; that nobody can simply eliminate inclination, but nobody knows how far their ability to overcome inclination, if necessary, might go; and that no one by their nature enjoys a greater privilege to exercise their freedom, although only in ways compatible with the equal freedom of others. In my view, this position does not ‘sentence most if not all to servility’ or represent a ‘systemic failure to attain an ideal that cannot be empirically realized’. It combines an ideal that may never be fully attained with the postulation of the possibility to approach it. But most importantly, it concedes to no one any right to do whatever they want in the name of their self-perfection because they believe that they are better fitted than others to achieve it. Nietzsche may not have found it to be a problem that ‘his ideal of freedom [cannot] be pursued by just anyone’, but Kant certainly would have found that a problem; in other words, he would have been repulsed by a position that allows a few to rise above the pain of asceticism and slavishness while condemning the many to continue to suffer from it. Kant’s message in the Religion is precisely that no one need be condemned to slavishness, and even if we all have reason to regret that we do not fully realise the moral idea of complete rationality, Kant is emphatic that this does not give any one of us reason to believe that we are or must be lesser than anyone else, let alone give anyone the right to pursue his own self-perfection at the cost of others. I would certainly be more comfortable defending that position than Nietzsche’s exceptionalist perfectionism.

4. Boucher and Vincent on Anglophone idealism

In writing on lesser-known figures in a wide-ranging history of philosophy, I try to frame the contributions of such figures to the particular issue at hand within some account of their overall philosophical assumptions and approaches. That was certainly the case for the modest selection of nineteenth- and twentieth-century Anglophone, that is, American as well as British, idealists that I discussed. Other than the early respondents to Kant, who are remembered only by Kant specialists and whose views are known only to a few of those, the Anglophone idealists are surely less well-known to most philosophers today than either the canonical figures of German philosophy from Fichte to Nietzsche or the leading protagonists of ‘analytical’ moral philosophy from the time of Sidgwick and Moore to the present. No doubt many contemporary professors of philosophy have heard of T. H. Green and F. H. Bradley, though they probably could not say much about them. Maybe they have heard of Josiah Royce but surely could say almost nothing about him, and they are unlikely to have heard of Edward Caird at all or to know that H. J. Paton did anything except translate Kant’s Groundwork. But my own knowledge of Anglophone idealism, which was certainly not any part of my education at Harvard in the late 1960s and early 1970s, even though with Royce Harvard had been the epicentre of American idealism a half-century before, pales before the comprehensive knowledge of this rich period of philosophy, especially moral and social philosophy, enjoyed by David Boucher and Andrew Vincent. Both have done invaluable work on the British version of idealism, and their present survey of the school is a marvel of both comprehension and insight. It could stand on its own as a brilliant introduction to the entire movement. In the present context, it does a far better job of situating the diverse moral philosophies of the British Idealists in their larger philosophical context than I could do given my own objectives for the book.

Boucher’s and Vincent’s discussion of Collingwood in particular leaves me regretting that I did not include a discussion of his New Leviathan and the previously unpublished lecture course on ‘Goodness, Rightness, and Utility’ that Boucher added to his revised edition of that book (Collingwood Reference Collingwood1992), although I had done some preparation for such a discussion. I find Collingwood a fascinating philosopher and person, and some of his work profound, such as his account of the relation between artist, work, and audience in Book III of The Principles of Art or the model of re-enactment in his philosophy of history, but some I find hopelessly obscure, such as ‘the scale of forms’ in The Essay on Philosophical Method. However, I cannot say that I find Collingwood’s critique of Kant’s moral philosophy as recounted by Boucher and Vincent very convincing. The message of the New Leviathan, that ‘the development of morality and civility is the gradual elimination of violence from our relations with each other, between bodies politic, and in our relationship with the environment’, is surely just as valid and important now as it was when Collingwood stated it at the outset of World War II. Collingwood’s thesis that ‘the historical consciousness, in its relation to duty, enables the individual to avoid the problem of acting according to rules. The historical consciousness indicates what it is my duty to do, being the sort of person I am, with the experience I have accumulated’, is less convincing. It may well be true for most people much of the time, for true moral dilemmas may be more frequent in the pages of philosophy journals than in most people’s real lives. But the very moment at which Collingwood was writing the New Leviathan was surely a moment at which this was not true, for their historically developed consciousnesses were driving the peoples of the warring nations of Europe and soon the other world powers such as Japan and the U.S. in violently clashing directions, certainly not in the single direction of ‘morality and civility’ with a ‘gradual elimination of violence’; or even within single nations, historically formed consciousnesses were certainly not unified or consistent. The members of the White Rose, or people like Dietrich Bonhöffer, were hardly on the same moral page as the percentage of the German electorate who had voted for Hitler in January 1933 or the larger percentage that went along with him for the next dozen years. The idea that at moments of crisis historical consciousness could do better than moral rules seems implausible; rather, moments of crisis are just when moral rules are most needed.

More technically, Collingwood’s objection to Kant, as reported, I have no doubt, entirely accurately, by Boucher and Vincent, ‘that it resolves all moral problems into “problems concerning rules and their application”’, is unpersuasive; indeed, if I may be a bit rude, it is more appropriate for the boy who stumbled on the Groundwork in his father’s library than for a mature philosopher, or, a little more politely, it is characteristic of a philosopher who has not read past the Groundwork, specifically to the Metaphysics of Morals for which, after all, the Groundwork is only supposed to be the groundwork. That duty consists exclusively in ‘obedience to rules’ overlooks two key points in Kant. First, it overlooks that Kant takes the categorical imperative, in whichever of its formulations, to be a principle for the selection of particular maxims, not particular actions. It might tell us not to make false promises, a proscription that would hold in any world of interacting rational agents, but it would not tell us, by itself, to return books that we have borrowed, let alone how to return such books, whether by post or in person, for such more specific decisions on particular actions will depend on a great deal of information about our particular world and the particular circumstances of particular agents in that world. Further, in his discussion of the so-called ‘Principle of Autonomy’ (G, 4: 432), Kant is insistent that we must will not merely single maxims that are consistent with our own intentions, but a consistent whole of maxims, and this is also implied by his conception of the ‘Empire of Ends’ (4: 433). And in the Metaphysics of Morals, he is insistent that while there can be incompatible ‘grounds of obligations’, there can be no conflict of obligations themselves (MM, 6: 224): in any situation, there is one right thing to do, although determining what that is requires factual information as well as the pure a priori principle. So I find it hard to see what Collingwood’s distinction between ‘right’ and ‘duty’ is supposed to be.

Second, Collingwood also overlooks Kant’s fundamental distinction between perfect and imperfect duties, which Kant does note in the Groundwork but fully explicates only in the Metaphysics of Morals. Perfect duties, allowing violation of the maxims of which are supposed to contain a ‘contradiction in conception’, are indeed supposed to prescribe inviolable rules, such as the proscription of making false promises, suicide, and homicide – duties of the Ten Commandments’ ‘Thou shalt not’ style, though Kant’s list may be even shorter than ten. But imperfect duties prescribe only general ends which leave much to individual judgement, are sensitive to all sorts of historical and personal circumstances, and are not reducible to rules. On Kant’s analysis in the Metaphysics of Morals, in fact, there are only two ‘ends that are also duties’, the advancement of one’s own perfection and of the happiness of others, which can by no means be reduced to any set of rules. Following predecessors such as Wolff and Baumgarten, Kant can specify the general areas for self-perfection or improvement, namely the perfection of one’s natural capacities of mind and body and one’s moral capacities of self-knowledge and conscience, and he can certainly say that in general benevolence, gratitude, and sympathy are indispensable in promoting the happiness of others. But he is very clear that one has to exercise one’s own judgement, without any obvious rules, about how best to realise for oneself the moral goals of natural and moral perfection and is as emphatic as can be that in promoting the happiness of others one has to try to do that on the basis of their conception of happiness rather than one’s own, which certainly leaves room for judgement because one can more easily formulate determinate rules for discerning others’ conceptions of their happiness than one can formulate anything more than mere ‘counsels of prudence’ for pursuing one’s own. Further, while Kant may recognise it as a rule that in any particular situation satisfying the rules of perfect duty takes precedence over advancing the ends of imperfect duty – no Robin Hoods; one cannot rob the rich even to help the poor – he by no means thinks that there are determinate rules, or many determinate rules, for resolving conflicts in the pursuit of the latter. Yes, it may be pretty clear that ordinarily one should help support one’s ageing parents or educate one’s own children, but those choices might not seem right in extreme circumstances – or, even in extreme circumstances, they may not be determinate. Perhaps Sartre’s young man trying to choose between joining the Resistance or staying home to care for his ailing mother should not have sweated it so much: no one can do everything, after all, and maybe either of these choices would have been permissible or admirable. In any case, only a caricature of Kant would have thought that there was a rule-governed or algorithmic way to resolve such conflicts; the real Kant, certainly the Kant who finally brought the ethical views he had been presenting in his lectures for decades into print in 1797, was not the ‘regularian’ that Collingwood accused him of being.

In sum, I think that the idealists, above all Green, had much to offer in their view that individual and social development go hand in hand, thus that individual perfection is not to be achieved in some sort of glorious isolation, a view in which they could have been encouraged, by the way, by the healthy British tradition going back to Shaftesbury and Hutcheson, not just the German tradition going back to Hegel. But I do not think that Collingwood’s critique of Kant shows Anglophone idealism at its best. So even if I should have thought to discuss the New Leviathan and its appendices, my discussion of The Good Will by Collingwood’s close contemporary H. J. Paton may not have been a bad choice after all; I think that he made more creative and valuable use of Kant than did the otherwise so inventive Collingwood.

5. Wilson on Kant and Rawls

We jump ahead now to the relation between Kant and Rawls. It is important to keep in mind that A Theory of Justice offered just what its name says, a theory of justice, or a part of political philosophy, founded on a Kantian premise, or, as Rawls preferred to say, on a premise that admits of a ‘Kantian interpretation’. Rawls did not construct his own version of a Kantian moral philosophy in general either in A Theory of Justice or even in the later work entitled ‘Kantian Constructivism’, which still focused on a foundation for his own theory of justice on Kantian assumptions. In neither case did he offer a Kantian moral philosophy in general, as, of course, many of his distinguished students and successors have attempted to do, although he did offer his own interpretation of Kant’s moral philosophy in his posthumously published Lectures on the History of Moral Philosophy. And remarkably, and to my mind at least somewhat unfortunately, Rawls completely ignored Kant’s own doctrine of Right in his political writings as well as in his posthumously published Lectures on the History of Political Philosophy. Samuel Freeman, the editor of the latter and as close a student as Rawls had, told me that Rawls told him that he did not make use of Kant’s doctrine of Right because it was too beholden to the natural law tradition to be of value to contemporary philosophy. I think that this was a mistake, because it is only by understanding precisely where Kant maintained and where he departed from that tradition that we can understand what was truly innovative about his moral philosophy and his political philosophy as a proper part of his moral philosophy, as it was for all the modern natural law theorists, and that understanding Kant through this lens also makes clear how close Rawls was to Kant rather than how far from him he might have been.

I say these things because in my view Wilson ends up demanding from Kant’s doctrine of Right in particular, which was not my subject in this book, what we should expect only from the other half of his metaphysics of morals, namely his Ethics, and because I think that her interpretation of Rawls is also shaded somewhat more to moral philosophy than to Rawls’s own more restrictive theory of justice as the primary virtue of the ‘basic structure of society’. I will explain what I mean by these claims, but I will begin with some more general comments about Wilson’s assumptions. I have to take issue with two claims that she makes at the outset of her remarks. The first is her claim that ‘Kant sees the essence of morality as the sacrifice of first-person interests – prudential or merely self-serving – in favour of a less advantaged party’. No, for Kant the ‘essential ends’ of humankind are ‘the greatest use of freedom … under which it can be self-consistent’, and the ‘principium of all duties’ is consequently ‘the conformity of the use of freedom with’ this essential end (Lectures on Ethics, 27: 346, in Kant Reference Kant, Schneewind, Heath and Heath1997, p. 127); that is, the fundamental principle of morality is that each use of any agent’s freedom, or any maxim in accordance with which an agent proposes to use her own freedom, be consistent with the greatest possible further use of her own freedom and the equally great freedom of all others. In the language of Religion within the Boundaries of Mere Reason again, morality requires the fundamental maxim of submitting one’s self-love, or self-conceit in the language of the Critique of Practical Reason, to this principle of equal greatest freedom in any case in which they might conflict – but that the priority of the freedom of all always and that of one’s own freedom now, to put it that way, might conflict is a contingent fact about the human condition, not a necessary fact about the condition of any and all rational beings and not the ‘essence’ of morality. So it is not part of the essence of morality that anyone has to sacrifice their own interest to anyone or all who are less advantaged; it is contingent that there are less-advantaged folk, and, if there are, someone has to be in that less-advantaged position, and surely morality cannot require that they sacrifice their self-interest to the less advantaged, for there are none. But all can make the freedom of all their overarching end.

Second, Wilson says that ‘Rawls’s scheme is beneficent – it is directly concerned with human well-being. While [for Kant] beneficence is a personal duty, Kant does not see it as a duty of government’, but she thinks Rawls does. That is closer to the truth, but far from entirely true. Rawls’s first principle of justice, that everyone be entitled to enjoy the greatest possible scheme of civil liberties, and the first part of his second principle, that everyone be entitled to equal opportunity to compete for offices and positions of dignity in their society, are, in the first instance, matters of the equal right of all to freedom, not to beneficence. This is why Rawls can place his ‘Kantian interpretation’ on his principles of justice: thus far, they are nothing other than, as he puts it, ways of expressing our nature as free and equal beings. This statement could fit right in with Kant’s exposition of the ‘ground of a possible categorical imperative’ in the Groundwork, namely, the formula of humanity as an end in itself, where humanity is understood as our ability to set our own ends, in other words, as our freedom, where Kant says that ‘our nature’ is what ‘marks us out as persons’, that is, as those due to every possible equality with each other. It is only when Rawls reaches the second part of his second principle, the difference principle, that only those inequities in a society should be permitted that would work to the advantage of the least well-off, that his principles of justice ‘directly’ concern well-being. And even then, we could question how ‘direct’ Rawls’s concern with well-being is: the difference principle is supposed to be that principle concerning resources that people would freely choose for themselves in the thought experiment of the ‘original position’, in which they would accept the ‘veil of ignorance’ about their own positions of advantage or disadvantage as a way of operationalising the moral demand for equality. The original position is a thought experiment for figuring out what commitment to the ideal of ourselves as free and equal persons would require of us, and real people are supposed to accept the principles that come out of that thought experiment because they are in the first instance committed to the moral ideal of themselves as free and equal. Considerations about well-being come in only subject to that prior constraint.

Now, of course, Kant is concerned about well-being as a proper part of the concern of morality as a whole, but, as I said in the previous discussion of Collingwood, the promotion of the happiness of others is an imperfect duty and for that reason a part of Ethics, not Right, where the distinction is that between the non-coercively enforceable part of morality and the coercively enforceable part of morality. Kant might draw the line between justice and beneficence more sharply than we think reasonable, but his reason for assigning the one to Right, and to the state as the basic structure of society, and the other to Ethics, as ‘personal duty’, is not exactly that ‘freedom of a specific sort, the removal of certain specific forms of coercion, simply ranks higher than well-being’. As I have already said, for Kant the principle of the greatest possible intra- and interpersonally consistent use of freedom is the principle of all duties and the essential end of humankind, but ‘ranks higher’ is vague, and it is also part of Kant’s position that it is in the realm of Ethics, specifically in the fulfilment of the duties of love, namely beneficence, gratitude, and sympathy, that agents can best demonstrate their own virtue or morality, because those ends and attitudes cannot be forced out of them by fear of coercion but only by their own respect for the moral law. In that regard, fulfilment of the Ethical duties of virtue ranks higher than satisfaction of the duties of Right, which can be enforced by external, aversive incentives (except in the case of rulers who need to be ‘moral politicians’, but that is a long story that I cannot go into here). Wilson recognises that coercion is the essential instrument of government, but I am not sure that she fully appreciates Kant’s inference that the use of coercion should therefore be restricted to the hindrance of the hindrance of freedom, as Kant puts it (MM, Doctrine of Right, Introduction, section D, 6: 231), and not used more liberally to make government into an instrument for the distribution of well-being. We can understand why, thinking of government in this way, Kant has to construct a somewhat strained argument to justify the use of taxation even to support desirable institutions such as homes for foundlings or the aged: taxation, after all, is a form of coercion, or at least unfortunately often has to be.

Kant’s second reason for restricting government to the protection of freedom is, as Wilson rightly notes, his violent objection to paternalism. We can view this objection as due not only to his resistance to the use of coercion for anything other than hindering hindrance to freedom but also to his insight that one can only promote the happiness of others, consistently with their freedom, in accordance with their conception of their happiness (MM, Doctrine of Virtue, Introduction, section V.B, 6: 388), combined with the reasonable assumption that government is no better at figuring out what people want than they are themselves or, in any case, has no right to arrogate to itself the privilege of doing so. This is an objection not just to ‘absolutist, kleptocratic, authoritarian war machines’, but to cameralist government in general no matter how benign their intentions – Father does not necessarily know best.

Now, as I said, Kant may have drawn the line between the properly restricted use of governmental coercion and what should be left to personal duty more sharply than some of us now think reasonable; and as Rawls recognised and especially as Onora O’Neill made so clear in Faces of Hunger (Reference O’Neill1986), the formal concession of freedom to less well-advantaged members of society may be meaningless without provision in some way or other of sufficient material resources to allow them to make use of that freedom, thus having sufficient health and education to make the opportunity for civil and economic positions and to exercise the simplest rights, such as voting, more than a farce. But to my mind that does not undermine Kant’s reasonable position that freedom more than well-being is the proper concern of government or make well-being rather than freedom Rawls’s primary concern.

One last comment: I do not think that it is correct to say that in his concern for equal greatest freedom Kant’s ‘focus is on noumenal dignity’. Kant’s concern is that everybody be as free as they can be consistently with the equal freedom of everyone else to set their own ends. Even if Kant is committed to our equal noumenal freedom to choose between good and evil and all of that, setting ends and, in so doing, either interfering or not interfering with the freedom of others to set their own ends, or even with one’s own further freedom to set one’s own ends, are not things that happen in the supposed noumenal world. They happen in space and time, and Kant is concerned that embodied agents, living on the finite surface of a globe in space and moving around it in time, be as free as is possible consistently with all of them being equally free. The whole programme of the Metaphysics of Morals, after all, of both Right and Ethics, is to show what duties follow from ‘universal moral principles’ valid for all rational beings but in view of ‘the particular nature of human beings, which is cognized only by experience’ (MM, Introduction, 6: 217). And while Kant might have thought that inequalities among human beings, like the division of the globe into different polities, were natural, in certain regards at certain times tolerable, and even, on a teleological view of nature, which he regarded as unavoidable for us, in certain ways at certain times beneficial, that does not mean that he thought inequality is ‘appropriate’; at least, inequality in freedom needs to be overcome. After all, he thought that war had certain benefits, promoting inventions and even certain valuable character traits; but in the long run, he also thought that perpetual peace is the final end of the doctrine of Right (MM, §61, 6: 350), the only ultimately acceptable goal of justice. Kant does not actually say that ‘the miserable masses were willed by nature’, let alone willing themselves, ‘to toil for the sake of advancement of culture for their superiors’ – he is not, as per the previous discussion, Nietzsche. His argument is always that while certain natural mechanisms, even natural inequalities, may afford means – developing skills – that can be used to achieve greater choice, it will always depend on the free choice of human beings and moral politicians, but not them alone, to achieve justice, locally or globally.

Once again, I thank my commentators for their penetrating and stimulating comments.