|IMPARTIALITY: KANT AND MILL VERSUS SIDGWICK
There are philosophers who think that practical reason has fundamental principles from which all practical reasons flow. Some of them are pure impartialists about practical reason. These philosophers hold, first, that all the fundamental principles of practical reason are agent-neutral. And second, that the range of these principles, the set of objects they concern, is human, or rational, or sentient beings. On this view the fundamental principles of practical reason mandate some form of impartiality as between the beings in this set.
Some others are pure partialists about practical reason. They hold that the fundamental principles of practical reason are agent-relative. Anyone, for example, who holds that a principle of formal egoism is the sole fundamental practical principle is a pure partialist. So is anyone who holds that the sole principle is instrumentalism. Hobbes fits into one of these groups, on most readings.
Then there is also the possibility that the foundations of practical reason are mixed, containing both partial and impartial principles. A foundation that postulated formal egoism but constrained it with a theory of rights would be a case in point. Sidgwick’s elusive dualism of practical reason is, or at any rate seems to be, another. Sidgwick departed from his two “masters,” on this point:
The rationality of self-regard seemed to me as undeniable as the rationality of self-sacrifice. I could not give up this conviction, though neither of my masters, neither Kant nor Mill, seemed willing to admit it: in different ways, each in his own way, they refused to admit it.1
‘No-one is more important than anyone else’; ‘no-one’s good matters more – “absolutely” – than anyone else’s.’ ‘No human being has greater absolute worth than another.’ Whether or not such dicta have always been taken for granted, they are widely taken for granted today, certainly at the political but also to a large extent at the ethical level. One could, it is true, agree with them negatively, on the basis that there is no such thing as the ‘absolute’ point of view, or Sidgwick’s ‘point of view of the universe’. If there is not, then it is true that no-one is more important ‘absolutely’ than anyone else. That is because no-one is absolutely important at all. By the same token no-one’s good matters more absolutely, and no human being has greater absolute worth. A doctrine of impartiality, in contrast, mandates a positive reading: at least in certain contexts, everyone, or everyone’s interests, or ends, or rights, count positively and count equally. Now the pure partialist may indeed concede that we are committed politically to some such doctrine – though presumably among citizens rather than among all human and still less among all rational or sentient beings. But on the pure partialist view the construction of an impartial political stance from foundations of practical reason must build with partialist, for example Hobbesian, materials. It is a political project. It is not the discovery of a fundamental insight of pure practical reason.
Kant and Mill, in contrast, hold that that is just what impartiality is. Furthermore, they hold that impartiality is the fundamental insight. They are pure impartialists. They must therefore somehow build the ineliminable elements of partiality that exist in any reasonable ethics into that.
I do not believe that can be done. I believe that certain partial principles are as fundamental as any principle of impartiality. So I endorse the mixed view. However the question of mixed or pure is not the one I am going to pursue. My aim here is to examine what epistemological defence Kant and Mill can give of their respective impartial principles, by testing them against the incisive criticisms of Sidgwick. On what epistemic basis can they hold impartiality to be a fundamental requirement of practical reason? The question remains interesting even if one’s position is mixed – it remains interesting so long as one thinks that a requirement of impartiality is at least one of the fundamental practical principles.
Sidgwick’s criticisms are decisive against the arguments he considers. However I am going to argue that in the end they somewhat miss their targets, by virtue of coming from an epistemological stance that both Kant and Mill would find uncongenial. Sidgwick sought to base ethics on self-evident “axioms” or “intuitions”, but neither Kant nor Mill would have been happy with such apparently intuitionistic talk. Of course given the wide range of things that people mean by ‘intuitionism’, that could be a merely apparent disagreement – but I don’t believe it is. I think it marks a real disagreement, and I want to defend the side of Kant and Mill. Yet Sidgwick’s criticisms are powerful enough to require significant restatement (or perhaps just clarification) of both the Kantian and the Millian positions. And it is important to consider both positions, because, I shall argue, the required reformulation must lead both of them to a similar idea as the epistemic basis of pure practical reason, namely, to the spontaneously disinterested dispositions of the will.
I need hardly add, of course, that Kant and Mill go beyond mere impartiality in their accounts of the foundations of practical reason. Kant incorporates his version of impartiality in the richer doctrine of the Categorical Imperative, Mill incorporates his in the principle of utility. These are rightly thought to point in different directions – in the one case towards equal respect for everyone’s autonomy, in the other to equal concern with everyone’s well-being. However it is the element of impartiality in their respective views, and its basis in the disinterested will, that I am interested in in this this talk.
The version of the Categorical Imperative that most clearly bears impartiality on its face is the Formula of Humanity:
Act in such a way that you always treat humanity, whether in your own person, or in the person of any other, never simply as a means, but always at the same time as an end. IV 429
This requires us to treat all rational beings, impartially, as ends and not merely as means – “man, and in general every rational being, exists as an end in itself ...” (IV 428).
Kant holds that this imperative follows by pure analysis from the requirement of autonomy. But can we even get impartiality, let alone Kant’s other distinctive ethical doctrines, from autonomy?
Practical reason, from which we take ourselves to act under the causality of freedom, is universal in content, as Kant correctly says. So reasons for action must be universalisable. This was the important lesson Sidgwick took himself to have learned from Kant. But then we encounter his critical point.2 We can make it vivid by considering the standpoint Sidgwick calls rational egoism. Rational egoists follow the maxim ‘Always do the action that is best for you’. They readily acknowledge the universality of that principle: everyone should always do the action that is best for them.
Sidgwick’s point can be put in terms of a contrast between agent-relative and agent-neutral ends, as they have since come to be known.3 The rational egoist’s reasons are agent-relative; the relation in question is that the act will advance his, the agent’s, good. In contrast the categorical imperative and the utility principle are agent-neutral. Both posit an agent-neutral end: rational nature, or the general happiness. Yet the rational egoist’s reasons and ends are just as universalisable as are the Kantian’s or the utilitarian’s. Hence no grounds for thinking that reasons must be agent-neutral can be merely formal, since the rational egoist’s reasons are in perfectly good formal shape to be reasons. Rational egoism may be mistaken, but it is not a misunderstanding of the very notion of a reason. If our substantive insight into practical reasons reveals to us that all or some reasons are agent-neutral, what it reveals is a substantive normative truth. This truth is not analytically derivable from the very notion of a reason.
So if we interpret Kant as arguing to impartiality solely from the sound formal point that reasons are universalisable, we have to conclude that his argument is fallacious. We cannot get from universality to impartiality by analysis alone. Impartiality requires not just the universality of reasons but also the existence of agent-neutral reasons.
At this point a promising response is made by Henry Allison.4 Recognising the force of Sidgwick’s objection, Allison acknowledges that no impartial principle can be deduced from the universality of reasons alone. However he denies that that is Kant’s intention. Instead he points out that Kant claims to derive impartiality from autonomy: that is, from the idea of acting solely from recognition of what one should do.
Acting thus means that one accepts no aim for one’s action, and no constraint on it, unless one sees reason to pursue that aim or observe that constraint. For anyone who accepts an aim or a constraint which they see no reason to accept is not acting from reason-responsiveness alone; they are being driven heteronomously by non-rational factors, “alien causes”.5 This is an important feature of Kant’s account. It generates, I agree, a sound objection to an instrumental, means-end conception of rationality, according to which rationality consists in adopting efficient means to one’s ends. Against this conception we can ask: why should we pursue our ends if there is no reason to pursue them? This may be your or my end, but it remains an open question whether it should be. Strictly speaking, in fact, one should deny that an instrumental conception of rationality is a conception of rationality, the capacity to come to a purely reason-responsive conclusion about what one should do, at all. Rationality thus understood – free and unconstrained deliberation about reasons – requires that we should be able to pursue reasons all the way down, never accepting an end or constraint simply as given: there must be no end, no constraint, that practical reason cannot put in question, by asking and answering whether there is reason to accept it.
This is a plausible interpretation of Kant’s notion of autonomy, and of why he thought it so important. Autonomy itself entails the existence of categorical, not merely hypothetical, imperatives. It is the crucial thing an instrumental conception of rationality omits. The instrumentalist maxim is ‘do whatever will most efficiently advance your actual ends’. The objection to it is not that it cannot be universalised, for it can be. The objection is that this principle simply takes ends for granted, without asking whether they should be adopted in the first place, and so cannot be the principle of an autonomous, fully free rational agent.
The instrumentalist conception of practical reason is agent-relative, and is ruled out by the ‘open-question’ challenge that autonomy puts to any merely positive ends. However, it does not follow that all agent-relative principles are ruled out by autonomy. In particular, appeal to autonomy provides no effective argument against the rational egoist. Rational egoists do not hold that you cannot deliberate about ends. They hold that there is reason for everyone to pursue their own good and they deliberate about what constitutes that good; thus about what ends there is reason to pursue. They accept that autonomy requires recognition of categorical imperatives; they think pursuit of one’s own good is a categorical imperative. Each person’s own good is a rational end for that person. For all that mere analysis of the notion of autonomy can apparently tell us it may be the only rational end; it may be that for each person their own good is the only final and unconditional end. In which case a rational egoist who acts from his principle acts autonomously.
So must we must conclude that an argument from the very idea of autonomy does not get us to impartiality, any more than an argument from the universalisability of reasons does? I think we can do a little more that than with the materials that Kant gives us.
Consider the Universal Law Formulation. This says:
Act only on that maxim through which you can at the same time will that it should become a universal law. (IV 421)
There is here, we know, a very suggestive idea: the idea of what you can will. Alas, at this point, one must confess, Kant says various things that those of us who are not committed Kantians find utterly unconvincing. I have in mind his appeals to the idea of a contradiction in the concept or in the will. In particular, it is a depressing lapse into banal moralism to argue, for example, that you cannot will that no-one should come to the aid of another, because you would want others to come to your aid if you were in need. At this point one can’t help seeing, in one’s mind’s eye, the incredulous sneer of Nietzsche. Suppose you encounter another human being, or just another rational being, which has no desire or neet at all for the assistance of others. Must such a being recognise the principle of aid – that one should come to the aid of those in need – as a categorical imperative? On the argument we are considering, it would seem that Kant should say no.
To be sure, I can ask myself whether, given my actual, contingent needs and limitations as a human being, I could ‘will’, that is, whether I could want to live in, a society in which everyone cheats whenever they can get away with it, in which there is no mutual aid, and so on. If I would not want everyone to do that kind of thing, because of the bad consequences such behaviour would have for me, or for people in general, then I am taking unfair advantage of others if I do it myself. I am unreasonably treating myself as special. That is good moral thinking so far as it goes. But the point is that it does not derive impartiality from autonomy. In appealing to fairness it presupposes impartiality. That is what gives this argument from fairness its dignity and strength. But at the same time this argument gives no hint of how pure practical reason itself can be said to bring fairness, impartiality, onto the scene.
Moreover Kant holds that the inference from autonomy to morality is analytic: “if freedom of the will is presupposed, morality, together with its principle, follows by mere analysis of the concept of freedom.”6 But on this argument the principle of aid does not follow by mere analysis of freedom, since it requires a premise about the agent’s wants. So if this is the only argument available, the principle of aid is no part of morality.
Nor, I am afraid, does impartiality follow by “mere analysis of the concept of freedom.” And yet the idea that a connection between autonomy and impartiality can be mediated by the notion of what an autonomous will can will remains intriguing. We can approach it by asking whether there is anything, and if so what, that a being can will inasmuch as it wills autonomously but without knowledge of the content of rational agent-relative ends, its own or those of others, to work on.
The question can be illustrated by a thought-experiment. You are contemplating a world of people that does not include you. All you know about them is that they are pursuing some rational ends or other. In other words you know that whatever these ends may be they are ends they should rationally pursue, and hence ends they would autonomously pursue. You have some green buttons you can press, one per person, to assist them to achieve their ends. You have some red buttons you can press, one per person, to frustrate their pursuit. Pressing any of these buttons has no other effect. You know nothing else; in particular you know nothing about who these people are or any relation they might have to you, or about your own current inclinations. Is there anything you can autonomously will? Is there anything there is still reason to will?
I believe that we are spontaneously disposed to press as many green buttons as we can, chosen at random if we can’t press them all, and likewise disposed to press no green buttons. Furthermore I think it makes sense to say that this is a disinterested disposition of the will, in that the disposition remains even if we abstract from all inclination. And finally, I think that this spontaneous disposition is the epistemic basis of the normative principle of impartiality. All of this, obviously, is controversial; but before we consider it further I want to bring in Mill, where we can find some distinctly similar themes.
Let us start from the so-called ‘proof’ of the principle of utility. This, as Mill points out in the introduction, “cannot be proof in the ordinary and popular meaning of the term” – “Questions of ultimate ends are not amenable to direct proof.” He adds, however, that
We are not … to infer that its acceptance or rejection must depend on blind impulse, or arbitrary choice. There is a larger meaning of the word proof, in which this question is amenable to it as any other of the disputed questions of philosophy. The subject is within the cognizance of the rational faculty; and neither does that faculty deal with it solely in the way of intuition. Considerations may be presented capable of determining the intellect either to give or withhold its assent to the doctrine; and this is equivalent to proof. (U I 5)
Turning to chapter 4, the ‘proof’ chapter itself, we find Mill asserting that
the sole evidence it is possible to produce that anything is desirable is that people do actually desire it. If the end which the utilitarian doctrine proposes to itself were not, in theory and in practice, acknowledged to be an end, nothing could ever convince any person that it was so. U IV 3
To understand this famous remark it is important to place it in the context of Mill’s epistemology of the normative. This epistemology makes no appeal whatsoever to a special receptive faculty of intuition which tells us of normative truths, in the way that our normal sensory faculties tell us of perceptible facts. Mill does not think that there is any such faculty. And on this point at least, Kant would agree. Mill’s positive epistemological claim is that the “evidence”, that is the epistemic criterion, of what is desirable is nothing but what is desired – “in theory and practice”. In general, the “evidence” for what there is reason to feel – to desire, to admire, to disdain, and so on – consists in our actual affective dispositions – the ones, one might add, that are both spontaneous and resilient, and cannot be reduced to others. Likewise, the evidence for what there is reason to believe consists in our spontaneous, resilient, irreducible dispositions to believe. It is in this way that Mill defends the normativity of induction in the System of Logic – by appeal to what he takes to be our spontaneous, resilient and irreducible inferential dispositions. In general, on Mill’s showing the epistemic basis for normative claims is nothing other than such dispositions. The work that epistemology can do consists solely in criticism of our dispositions, with a view to identifying which ones really are spontaneous, resilient and irreducible.
There is clearly a much longer story to be told about this anti-metaphysical outlook. Taking it as background, however, let us turn to Mill’s next notorious move, in which he infers that if “each person’s happiness is a good to that person” then “the general happiness” must be “a good to the aggregate of all persons” (CWX 234; U IV 3). In a letter in which he explains this unclear remark, he amplifies: “I merely meant in this particular sentence to argue that since A’s happiness is a good, B’s a good, C’s a good, etc, the sum of all these goods must be a good” (CWXVI 1414). Evidently he simply takes it for granted that each person’s happiness is a good – not just good for him, desirable for him, but an agent-neutral good.
We should also take into account two other passages. At the end of the last chapter of Utilitarianism (‘On the Connexion between Justice and Utility’), Mill explains that he takes “perfect impartiality between persons” to be part of the very meaning of the Greatest Happiness Principle; the
principle is a mere form of words without rational signification, unless one person’s happiness, supposed equal in degree (with the proper allowance made for kind), is counted for exactly as much as another’s. (CWX 257; U 5 36)
And in the last chapter of the System of Logic, where he comes to consider the “logic of practice”, he announces that
Every art is ... a joint result of laws of nature disclosed by science, and of the general principles of what has been called Teleology, or the Doctrine of Ends; which, borrowing the language of the German metaphysicians, may also be termed, not improperly, the principles of Practical Reason.
...These general premises, together with the principal conclusions which may be deduced from them, form (or rather might form) a body of doctrine, which is properly the Art of Life, in its three departments, Morality, Prudence or Policy, and Aesthetics; the Right, the Expedient, and the Beautiful or Noble, in human conduct and works.
“Without attempting,” he continues a little later,
in this place to justify my opinion, or even to define the kind of justification which it admits of, I merely declare my conviction, that the general principle to which all rules of practice ought to conform, and the test by which they should be tried, is that of conduciveness to the happiness of mankind, or rather, of all sentient beings: in other words, that the promotion of happiness is the ultimate principle of Teleology. (System of Logic CWVIII 232)
In all these passages the assumption of agent-neutrality is clearly evident. Furthermore the passage from the System of Logic makes it clear that Mill takes the utility principle to be the fundamental principle not just of morality but of practical reason as such. So the sole, agent-neutral, principle of practical reason itself is the principle of utility.
Now as Sidgwick is quite right to point out, this principle requires a separate epistemic basis. However, given Mill’s epistemology, its epistemic basis cannot be a Sidgwickian “intuition”. It must consist of the kind of dispositions that we have just been talking about. The dispositions in question cannot however be dispositions of feeling or desire. Perfect impartiality cannot be founded on these, since we have no such impartial dispositions of feeling or desire. Nor can they be founded on dispositions to believe, since the norms founded in this way are epistemic not practical norms – principles governing the theoretical intellect. Mill is not an intellectualist who thinks that fundamental practical principles are known by a purely intellectual intuition. In these two specific respects Kant agrees: knowledge of pure practical principles is not founded on dispositions of feeling, but nor is it founded on dispositions of the understanding. It is founded on the spontaneous dispositions of the will. For Kant the spontaneity that epistemically grounds pure practical principles must be the spontaneity of the autonomous will.
Now of course Kant sharply distinguishes inclination and will, and hence he has this epistemic resource to call on. But, it might be asked, does Mill make any such distinction? Given his associationist psychology can he invoke dispositions of the will as against dispositions of desire? He does and he can – or at least he thinks he can. He accepts, to quote Utilitarianism again:
that the will is a different thing from desire; that a person of confirmed virtue, or any other person whose purposes are fixed, carries out his purposes without any thought of the pleasure he has in contemplating them, or expects to derive from their fulfilment; and persists in acting on them, even though these pleasures are much diminished, by changes in his character or decay of his passive sensibilities, or are outweighed by the pains which the pursuit of the purposes may bring upon him ... All this I fully admit, and have stated it elsewhere, as positively and emphatically as anyone. Will, the active phenomenon, is a different thing from desire, the state of passive sensibility, and though originally an offshoot from it, may in time take root and detach itself from the parent stock; so much so, that in the case of an habitual purpose, instead of willing the thing because we desire it, we often desire it only because we will it. (U IV 11, CWX 238)
The other place where he has emphatically stated this doctrine is the System of Logic, in a section entitled “A motive not always the anticipation of a pleasure or pain”, where we find the following interesting remark:
… A habit of willing is commonly called a purpose; and among the causes of our volitions, and of the actions which flow from them, must be reckoned not only likings and aversions, but also purposes. It is only when our purposes have become independent from the feelings of pain or pleasure from which they originally took their rise, that we are said to have a confirmed character. "A character," says Novalis, " is a completely fashioned will:" and the will, once so fashioned, may be steady and constant, when the passive susceptibilities of pleasure and pain are greatly weakened, or materially changed. CWVIII 154
This brings Kant and German romanticism to mind, as I am sure it is meant to do.
Importantly, of course, Kant thinks that that the feelings are phenomenal, that they belong to our phenomenal character, whereas the will does not. Fatefully, he puts his good distinction between feeling and will into the distorting framework of transcendental idealism. But from our present epistemological standpoint this issue about transcendental idealism makes little difference, though it makes some. The important thing is the distinction between feeling and will, which, as we have just seen, Mill also recognises.
Let’s go back to the thought-experiment about green and red buttons. In this situation I think I would be disposed to press as many green buttons as I could, chosen at random if I couldn’t press them all, and disposed not to press any red buttons. I think that would apply to most of us. Furthermore this disposition comes packaged, so to speak, with a disposition to take it, to see it as, an apt, appropriate, reasonable thing to do – the right thing to do. I am at home with it, it doesn’t feel like an alien intrusion, an adventitious psychological phenomenon that puzzles me. It feels spontaneous, something that comes naturally from the way I am. Since it’s a disposition to act in a certain way, it is a disposition of the will. Furthermore, this particular disposition of the will is not conditioned, or dependent on, any disposition of the feelings. It is a spontaneous, unconditioned and disinterested disposition of the will.
Is it resilient and irreducible (or in Mill’s words “steady and constant”) as well as spontaneous? If it is it remains firm after I have made it clear to myself that I have it – it does not wilt in the daylight of clear self-knowledge, and is not be explained as a means to some other end that I am disposed to seek. If it has this character, then by the epistemology we have found in Mill, it provides the epistemic basis for a pure practical principle: namely, the principle that achievement by anyone of their rational goals is something I have some reason to promote. We may conclude that everyone’s rational ends, considered comprehensively and impartially, have normative standing as ends for the autonomous will.
Once you bring knowledge about your inclinations and relations to other people back into the picture, you may well bring back agent-relative reasons. The claim is not that the autonomous will can only will agent-neutral ends. The claim is that when we focus on whether there is anything that an autonomous will can will before any further material is brought in that might give it other, inclination-based reasons for action, we find that there is indeed still something, and that something is the achievement of any being’s rational ends as such. Furthermore the moment of impartiality that is (on this view) contained in the autonomous will exercises a constraint of some form on our pursuit of our agent-relative ends. Obviously there are difficult further questions about what substantive form that constraint takes, which I certainly don’t think can be worked out from the very notion of impartiality. Even so, these points about the disinterested will give us an approach to understanding the Universal Law Formulation:
Act only on that maxim through which you can at the same time will that it should become a universal law.
That is, reject maxims which you can will to be universal law only at the cost of contradicting the moment of impartiality that is present in your autonomous will. That is the sense in which we should ask whether we can will our maxim to be a universal law: we are asking whether we can will it in the presence of that pure or disinterested disposition, which is the spontaneous disposition of the autonomous will itself.
Now Kant thinks, as we noted, that it follows analytically from the concept of autonomy that an autonomous will is bound by this categorical imperative. To defend this (even in part) we would have to argue that a will that does not contain the disinterested moment of impartiality is not autonomous, not fully free. Since autonomy is full positive freedom, that amounts to arguing that it is analytic, or at least constitutive of the very concept of full positive freedom, that a will that lacks the disinterested moment of impartiality is not full free. The question obviously arises: are we simply dressing up a substantive and attractive ideal of freedom as an analytic truth? Put the question in another way. We have an understanding of what it is to be fully free, ‘morally free’ to use John Stuart Mill’s term.7 In effect Kant wants to add to the negative conception of moral freedom, that of a will that is not determined “by alien causes,” a positive conception: that of will positive capable of disinterested willing. Can we say that that is, indeed, a part of our notion of full freedom, so that to lack that capability is to fall short of full freedom?
But I shall leave that question hanging, because it takes us away from my main point, which is that this epistemological basis for impartiality is equally available – if available at all – to both Kant and Mill. Of course Mill is committed by his associationist psychology to showing how such a disinterested disposition of the will can emerge. We have already seen how he argues that the will and its purposes can emerge from desire. He must further argue that a disinterested disposition of the will is a spontaneous element on which a developed human nature builds. Something like that is implicit in chapter III of Utilitarianism. Here Mill rejects a “transcendental origin of moral obligation” and states his belief that the “moral feelings are not innate, but acquired”. But, he continues,
they are not for that reason the less natural… The moral feelings are not indeed a part of our nature, in the sense of being in any perceptible degree present in all of us; but this, unhappily, is a fact admitted by those who believe the most strenuously in their transcendental origin. … the moral faculty, if not a part of our nature, is a natural outgrowth from it; capable … in a certain small degree, of springing up spontaneously; and susceptible of being brought by cultivation to a high degree of development. (U III 8)
He then goes on to discuss how this outgrowth is natural because consonant with and supported by a powerful class of natural sentiments: “the social feelings of mankind”, “the desire to be in unity with our fellow creatures” – feelings which are not only natural to us but also continually strengthened, he thinks, by the growth of social specialisation and mutual interdependence.
In these passages Mill appeals to the natural sentiments, whereas in the other passages I have quoted he distinguishes between desire and will. Mill’s interest in the moral sentiments is entirely apt; it is one of the strengths of his ethics. My point in putting the two groups of remarks together is to show that Mill might well be ready to endorse a psychological account of the growth of a disinterested disposition of the will, and thus, in this respect, approach the Kantian epistemological stance as to the foundation of impartiality. I take it that Mill’s emphatic distinction between desire and will is intended to give its due to Kantianism. There remains the difference, of course, that Mill takes the will be a natural phenomenon, whereas Kant takes it to belong to the transcendental domain of freedom. However the crucial point, as far as the epistemology of impartial reason is concerned, is not whether the will is transcendental or natural, but whether it does have a resilient moment of spontaneous disinterestedness, which can be sometimes overwhelmed, but not destroyed. Transcendentalism may be thought to have an advantage in that it gives that moment a transcendental guarantee. It makes its resilient spontaneity tamper-proof, so to speak – and importantly for Kant, something that is present in all of us. An empirical psychology of the will, and especially an associationist aetiology of the will, cannot do these things. However Mill’s associationist psychology is one thesis, his naturalism is another, and his epistemology is a third. Given his naturalism, his claim must be that a disinterested disposition of the will does exist as part of our nature, whatever its aetiology may be. Or at least, that it exists “as a naral outgrowth from [our nature]; capable … in a certain small degree, of springing up spontaneously.” Much more then has to be said, to explain what it means to say that disinterestedness is a spontaneous and resilient disposition of our will. What do we mean by ‘our’ for example? Do we mean everyone’s? And how does that matter? But on the assumption that this claim can be sensibly explained and established, it suffices to ground a normative principle of impartiality, without requiring any appeal to a receptive faculty of intuition.
It is Sidgwick’s merit to have pointed out, clearly and sharply, that impartiality, considered as a pure principle of practical reason, requires a distinct epistemic basis of its own and cannot be derived by any argument from purely agent-relative starting points. I have argued however that both Kant and Mill have or think they have the materials to provide that distinct epistemic basis, though neither of them exactly comes clean about it. It is clear enough, however, that neither of them is interested in intuitionist metaphysics; nor is that because they have somehow overlooked the intuitionist option.
Sidgwick’s view seems more intuitionistic. However there is an important question of interpretation here. In one sense of ‘intuitionism’, intuitionists hold that there is a world of normative facts that we know by a receptive faculty of intuition, in the way that we know the world of space and time by spatio-temporal perception. On the basis of that knowledge of normative facts, we then choose what to do. In other words, their account of the epistemology of normative knowledge gives no epistemic status to our dispositions of belief, feeling, or will as such. Our normative intuitions may indeed cause such dispositions in us, but the dispositions themselves play no role in the epistemology of normative knowledge. For Kant, given his contrast between spontaneity and receptivity, there can be no such separation. Purely normative judgement is a matter of spontaneity, not receptivity. In particular, pure practical reason is epistemically grounded in spontaneous dispositions of the will, and for just that reason it can itself be practical: i.e. give rise to action via the causality of freedom.
One can therefore distinguish between an intellectualist and a realist aspect of intuitionism. I don’t believe that Sidgwick endorsed the realist aspect. When he talks about rational intuitions, or about the self-evidence of certain propositions, he is not taking intuition to be a receptive faculty which gives us knowledge of a world of normative facts. The criteria he gives for intuitive knowledge (self-evidence, consistency, consensus, clarity and precision) make no mention of such a faculty. So far, so good. On the other hand Sidgwick’s view of ethics does seem to have been distinctly intellectualist. What I mean is that he treats ‘intuitions’ as acts of intellectual insight, divorced from dispositions of the will or the feelings. We somehow simply ‘see’, for example, that something is desirable, or that it is the right thing to do – and that may or may not dispose us to desire it, or to do it. In this respect it seems to me that Mill’s epistemology is preferable. It finds firm ground for our normative judgements – judgements about what there is reason to believe, or feel, or do – in our spontaneous and resilient dispositions to believe, feel and act. In doing so, it stick close to our actual, often messy, normative practice. Sidgwick’s intellectualist talk of self-evident ethical axioms has an impressively exact and scientific air. But as with the marmoreal classical façade of a bank, the exterior misleads as to what is really going on inside: the actual human dispositions that are involved in making the thing work. In the case of impartiality, it is only the presence and power of disinterested will that gives it normative force, and makes it work. Mill’s approach directs our attention in the right way: to what we can say, empirically, about the presence and power of this disinterested will, and how we can fashion such facts into an epistemic basis for some principle of impartiality.