Skip to content

Kant’s Shadow Model of Moral Causality

The question that concerns Kant in his second Critique is this: how can reason be practical? This might look like a question about the role that reaSoning plays in action and the straightforward answer to this question, outside the Kantian context, would be an explanation of the place reasoning has in human action set out in terms of natural causality. Natural causality, as Kant understood it, is a sequence of events in which an earlier event (the cause) is connected to a succeeding event (the effect) according to a rule.~ Thus when we wish to explain human behaviour in terms of natural causality this will involve a description of several events including the agent's reasoning. Typically this description makes reference to a given stimulus arousing a desire for something which in turn sets in motion the process of working out, through reasoning (based on the given information and past experience), the best means to the end in view. This in turn leads directly to appropriate action and to satisfy as the motivating desire is fulfilled.

When there is more than one desire the situation, although more complicated, can still be expressed in terms of a single unbroken causal sequence. In this case, two separate stimuli lead to the more or less simultaneous activation of two distinct desires. Subsequent experience reveals the fact that both desires cannot be achieved at once and one or the other must be given precedence. There follows a felicific calculation2 in which the strongest desire makes itself known. In effect this calculation amounts to a weighing-up of the pros and cons of satisfying one or other

~ See CPR, p. 464, A 532 B 560. "When we are dealing with what happens there are only two kinds of causality conceivable by us; the causality is either according to nature or arises from freedom. The former is the connection in the sensible world of one state with a preceding state on which it follows according to a rule."

The terminology is Bentham's but here it is being used to refer to the fact that left to ourselves, we, like all animals, are able to come up with (as if automatically calculated) the answer to the question: 'What will please me most?' (or 'Which course of action do I prefer?').17 of the competing desires, a process accomplished (so far as the agent is concerned) by simply considering which course of action, on balance, will produce the greatest satisfaction. When this becomes apparent, action follows and the agent achieves satisfaction.

As with any natural explanation, there are no gaps in the causal sequence. Each step in the sequence is naturally determined ('brought about') by the prior steps. Thus there is no place for any free exercise of the will as an intervening factor as we might suppose when, for example, we think of ourselves as choosing between competing desires. Where we are employing a natural explanation schema, the decision to pursue one course of action rather than the other is not to be understood as a distinct element intervening and so governing the course of the causal sequence Instead the decision must be regarded as a function of the relative strengths of the desires in question as these emerge in the consideration of the likely consequences of pursuing alternative courses of action. We call the discovery, through felicific calculation, of the strongest desire a decision only in retrospect in order to call attention to the point at which we became aware that one of our desires outweighed the other. To sum up: as a component in the natural causal sequence which is initiated by those stimuli which prompt desires in us, reasoning does not determine (in the sense of choose) which desire will be supreme.

This natural model of explanation for human behaviour we will dub the original model. We now wish to show how the shadow model of moral behaviour is developed from the original and to see how the shadow model deviates from the original as a kind of explanation.

Kant's question in the Critique of Pracrical Reason ('How can reason be Practical?') concerns the possibility of reason functioning so as to direct human behaviour by itself, as it were, and not simply as the servant of desire. For it is Kant's view that reason is not limited to an advisory role, the role it plays in formulating hypothetical imperatives: 'If you want x (some satisfaction) then it would be rational to do y.' This Kant refers to as the heteronomous function of reason. But in its most important function, Kant was convinced that reason is to be regarded as autonomous, supplying its own motivation and its own satisfaction and thus not serving as a mere tool of external motivations supplied by the faculty of desire. (3)

Now Kant is convinced that the reason must be practical, since if it were not, morality would be impossible and moral ideals empty But since, in his view, morality manifestly exists as a form of (3) See CPrR, pp. 144-145, A 58-59 Theorem IV. Life (4) he concentrates on providing an explanation of how such autonomous rational behavior is brought about rather than whether it ever actually occurs. That is, Kant concentrates on supplying a model which will serve to explain reason s practical capacity to produce moral behaviour. This system of moral causality (what we have called the ~shadow model of causality') is needed because the mechanism of natural causality is inappropriate to explain moral behaviour. It is inappropriate because it has no place for reasoning as an autonomous practical faculty. As we have seen, the original model describes rational human behaviour as determined naturally, i.e. as activated or set into operation by a heteronomous influence, viz. desire. Any explanation of human behaviour on this natural model (the original) leaves no room for the imputation of responsibility to the person who acts, since everything a person does is understood as a direct consequence of some naturally occurring desire. There is simply no room, then in this model for an autonomous, a self-governing, principle of action. But this is precisely what is required if the action in question is to be thought of as Iying within the agent's responsibility. However, an explanation of moral behaviour-if it is to sound like an explanation-must describe a linked series of events that are related as cause to effect and which, altogether, account for how the behaviour that is to be explained has come about. The explanation of moral behaviour must therefore be at least analogous to explanations of behaviour in terms of natural causality. To repeat: the model of moral behaviour must be modelled on the original model of non-moral behaviour if it is to present itself as an explanation.

What then does the shadow model which explains moral causality (i.e. how reason can be practical) amount to? Moral behaviour as a form of life arises from a new kind of internal conflict. As we have seen, when we experience conflicting natural desires, a resolution of this conflict is effected through reasonts investigations which allow us to calculate (felicifically) which desire is likely (when we carefully consider the C°nsequences which reason lays before us) to bring the greatest Satisfaction. However, in the case of morality, the internal conflict which marks the appearance of morality as a form of life is a conflict between a ibid., p. 143, A 56 and p. 157, A 81, where the moral law is considered to be a "fact of pure reason." See also p. 152, A 72, in which e argument of the Analytic is summed up: the fact of autonomy in the principle of morality proves that pure reason can be practical. Kant is rightly convinced that we would not have the idea of the moral law, especially as a compelling directive, if it were not possible to effect its unconditional demands. Cf. note 42 to Chapter One natural desire to do something and a sense of duty which is felt as a constraint upon this natural desire. In the shadow model of mora1 causality, this sense of duty or feeling of obligation plays the role that desire plays in the original model. The sense of duty or obligation constitutes a new sort of 'desire', a 'desire' to do what is right. But as an element within the shadow model of moral causality it cannot have any actual motivational power as desires do. If it did, then moral behaviour would result only when our feelings of obligation were, as a matter of fact, stronger than the natural desire that they set themselves against. In such a case, we would not be acting freely, out of a sense of duty, but simply because of the strength of our feelings of obligation. In other words, the moral behaviour of the agent would have a natural explanation, and thus have no moral value.

The sequence of events which constitutes the moral form of life and at the same time illustrates the shadow model of moral causality can be usefully set out through the form of a dialogue for reasons which become apparent as the dialogue progresses and which will be dealt with explicitly later (5)

With any piece of behaviour, natural or moral, the action is instigated by some stimulus. In the example to be presented it is the arrival of an account in the post.

Desire: I am inclined to pay this account for the usual reason: failure to pay will result in the repossession of the item in question which is a source of satisfaction to me. I am therefore inclined to pay because I want to continue to enjoy the item.

Reason: I can appreciate that, but we lack the funds to pay this bill. I suggest we either give up x or borrow the money we need.

Desire: Let's borrow the money.

Reason: But you6 cannot repay it. You will have to lie to convince the lender to give you the money.

Desire: Who cares? I want satisfaction. At this point the natural sequence (following the original model of natural causality) would continue along the usual lines:

Reason: I can appreciate that, but would it be wise to blindly pursue

5 See pp. 35 '., below, especially p. 39.

6 There is a certain amount of confusion as to who is the 'person' in charge here. This confusion reflects an unavoidable problem when the faculties are person)fied. See pp. 39.~. below where this problem is discussed.

20 ~is satisfaction? What happens when the lender is not paid?

Desire: Borrow again

Reasonn: Word will get around, and no one will lend you money in the filture (it's happened to others).

Desire: I wouldn't like that.7

Reason: No. Best perhaps to give up x now and retain your reputation, since it might come in handy later. If you follow your inclination you will eventually lose both x and your reputation.

Desire: I guess I 11 take your advice.

Here the felicifzc calculus is at work weighing the thought of a present (short-term) pleasure against considerations relating to long-term pleasures and pains. The balance could tip either way depending on past exerience which supplies the felicifzc 'weights' which are thrown into the balance when such thoughts occupy our attention.

Let us now consider the moral resolution of this conflict. (The moral sequence follows on Desire's remark: "Who cares? I want satisfaction.")

Conscience: I can appreciate that, but would it be right8 to borrow money knowing that you cannot repay it, that is, lie to get what you want79

Conscience then asks Reason a strange question.

What would things be like if everyone adopted the maxim for action that Desire wants to act upon?

Reason: The practice of Iying would necessarily self-destruct if universally practised. It is a practically invalid maxim.

Here we must break into the dialogue to ask: "Who is this Conscience character?~, Whence comes this voice of Conscience which asks its reproachful question: ~Would it be right?' Within the original model Reason can ask a Mther similar question on the basis of past experience, v~z. 'would it be wise?' and proceed to adumbrate prudential

This is a crucial step in the ~natural' procedure. If Desire does not react emotionally to Reasonts imaginative projection, then Reason's argument will have no power.

This question marks the appearance of the phenomenon of a sense of duty, unique to the ~inner life' of a person behaving morally. See p. 16, above.

Notice how suddenly Conscience makes its entrance. This is also the case in the Kantian original. See FMM, p. 81, BA 54  considerations. However, Conscience does not speak with the voice of experience, but from another quarter altogether. Conscience is, it would seem, the categorical voice of Reason:~° in Kant's terminology, the pure as opposed to the empirical aspect of practical reason.~ Here we have Reason speaking not as the hypothesizing guide to happiness, computing the best means to achieve satisfaction on the basis of past experience, but rather speaking on its own behalf of a completely different goal-a goal which has nothing to do with happiness. Thus the first assumption of the shadow model of moral causality is that we are capable of being motivated by this goal, that we can 'desire' it even though we do not desire it in the natural sense of desiring, e.g. desiring food when we are hungry. Can we make any sense out of this notion of being motivated to do what is right?

We can provide an answer to this question by asking what interest Conscience (the categorical voice of Reason) might have in our doing what is right? Why does it want us to do what is right? It speaks as if it mattered to it what we did, as if it cared. It reproaches us with its question: 'But would it be right?' The tone of its question indicates clearly that it feels we ought to do what is right and in some sense of the word it wants us to heed its reproach. One way of appreciating its motivation here is to say that it wants us to do what it would do if it were able to act.

We can begin to appreciate why Conscience exhibits this attitude by considering what constitutes the character of this action that it wants us to perform and that it labels 'right'. The hallmark of reasoning is consistency and, were Reason (for whom Conscience speaks) a person, an agent, we can suppose that it would always act according to maxims which were consistent, i.e. practically valid. In other words, it would always act on maxims which did not self-destruct when put into practice. But if we ask: 'Why would it so act?', 'What would motivate it?', we

l0 Notice how ~Reason' rather than ~reason, seems appropriate here. This is also the practice followed in the various translations of Kantts works in which we find ~Reasont, ~Sensibilityt, ~Understanding' and ~Judgment', all capitalized. The process of personification which, as we shall see later, the dialogue form entails, shifts attention from active functions of, for example, reasoning, to something which performs this function-Reason The fact that we identify with this ~something' (for it is we who are reasoning, i.e. the something is a someone, a person), has us naturallY treating reason as a person and is thus deserving of a proper name.

1l As in the question at the centre of CPrR: "Is pure reason aufficient of itself to determine the will, or is it only as empirically conditioned that it can do so?" CPrR, p. 129, A 30. must grant that it would not do so through desire, since reason, even personified, is not burdened with desires. It follows, if we allow some Substance to this personification,~2 that it can only have one other sort Of motivation which, in some fashion at least, is intelligible to us: if ReaSon were a person it would always act consistently simply in order to be what it is. Just as we might insist that invalid reasoning is not reasoning, insofar as we cannot 'follow' invalid arguments, so, by analogy, practically invalid acting could not be construed as Reason acting, since such acts self-destruct: they are not rational acts. So the 'interest, that Reason (as person)fied and speaking with the voice of Conscience) has in getting us to obey its injunctions can be thought of as being ontologically based.~3 The trope of personification, which a description of the moral form of life quite naturally falls into, leads to the somewhat curious conclusion that Conscience wants us to do our duty, wants us to be good, so that it can be (behave) in accordance with its own nature, and thus be true to itself (a goal which is characteristic of the moral form of life). Thus a key element of the shadow model of moral causality is the occurence within us of a sense of duty which is felt as a kind of imposition. We feel it as an admonition presented by one of our faculties to behave as it would if it could, viz. in accordance with maxims which are practically valid.

Now, in reality, Reason cannot act without our being willing to actualize its commands in our lives. Therefore, proving that Reason can be practical involves showing "that freedom does in fact belong to the human will":~4 in other words showing that we can will to be reasonable (to do what is right) without reference to natural motives. We understand, then, that Reason, as personified, is only an abstraction. It must not, therefore, be regarded as capable of actually coercing us into doing its bidding by awakening within us a feeling of obligation which then determines us to act after the fashion of a natural motive. If morality is to be possible Reason~s categorical persona, Conscience, can, it seems, only 'coerce' us by setting up a standard of practically valid behaviour which it would pursue if it could act (viz. dutiful behaviour) and then leaving us the choice~5 as to whether its 'desires' are to be - 12 Such an allowance is necessary once the trope of person)fication is emPIOyed and, as we shall see below (pp. 42 ~.), its use is a natural result of a certain aspect of reflexive consciousness which will be seen to ground the possibility of the moral form of life.

13 See Chapter Eight for a discussion of this notion of an Ontological Imperative. See also p. 28, below. CPrR, p. 129, A 30. This is essential given Kant's conception of the nature of the human realized through our being rational. Conscience then is reason personified and, as such, it is imbued with a 'desire' to be (to have its way) which it can only realize with our cooperation. In the dialogue it expresses this 'desire' with its question 'But would it be right (= practically valid) to behave according to your proposed maxim?, However, as we shall see, Desire has no natural capacity to act in accordance with maxims whose goals are unrelated to happiness. Thus if Conscience is to motivate Desire~6 it would have to present dutiful behaviour as the path to greater happiness.

However, in the nature of the case, this strategy is not open to it because the moral motivation to do what is right cannot consist of a desire for happiness. As we shall see in what follows, this stalemate leads to the emergence of the next element in the shadow model of moral causality, viz. a capacity to act freely which allows for the possibility of acting solely out of the thought of duty and thus without the motivation provided by any naturally occurring desire. We now return to the dialogue.

Reason: The practice of Iying would necessarily self-destruct if universally practised. It is a practically invalid maxim.

Conscience: Reason has just indicated why, as I suspected, acting on your maxim would not be right. I charge you, Desire, not to act on your maxim. Do your duty and resist your inclination to lie.

Desire: Resist? I can't resist myself. I am Desire and I want only satisfaction. Reason said that if my maxim were followed universally it would self-desttuct. As I weigh the alternatives for satisfaction presented to me by Reason I feel that, whether I lie or whether I do not, I will eventually end up in a dissatisfied state. But it seems to me that if I lie, I will initially, at least, get some satisfaction, before this Iying of mine becomes common practice and eliminates itself as a means of obtaining will as affected but nor necessitated. It is not an arbitrium brutum, but an arbitrium liberum. See CPR, p. 465, A 534 B 562.

16 Note that this portion of the dialogue is not intended to reflect Kant's own representation of how Reason works to persuade us to act morally This is not his concern in the CPrR (the question there is how this persuasion is possible at all). It is only in the FMM in rhe four examples Kant gives to demonstrate how the Categorical Imperative is to be applied, that we get any idea of Reasonts methods of persuasion. The dialogue here is an attempt to reconstruct this process, to see how Reason operates when it seeks to effect behaviour. The intention is to check the practicality of Kant's claim that Reason can be practical, by going through the procedure of solving a ~real-life, dilemma using Reason~s methods satisfactory for a person who, like myself, is in financial straits. At this stage, Desire, left to its own resources, would weigh up these alternatives and come down on one side or the other, depending on past experience. Conscience: I can see that you really don't appreciate the character of the coercive power of my injunction to do your duty. I am not asking you to weigh the alternatives of either doing your duty or going against it, in terms of the satisfactions to be gained in either case. You have obviously missed the point. I am demanding of you that you do your duty simply because it is your duty to do what is right. It is just stupid to imagine that you could regard yourself as having done your duty if this were the result of your calculating that to do so would give you greater satisfaction' at least so far as you seem to understand satisfaction, that is, in terms of happiness.~7

Desire: Now we seem to be getting somewhere. Is there another kind of satisfaction, other than happiness, which 'satisfies' those who do their duty? Tell me more. What is it like? Is it better than happiness?

Conscience: I am rather sorry I hinted at another source of satisfaction which a person like yourself might take as grist for your felicific mill. I'll admit however that those who do their duty do find themselves blessed with another kind of satisfaction called self-contentment.~8 However I cannot hasten quickly enough to add that this satisfaction cannot be sought as such. It only comes to those who do their duty solely in virtue of the fact that it is their duty. You will find that it cannot be obtained if it is sought as your goal. You cannot desire to obtain self-contentment, though of course you can 'desire' it, that is, consider it worth having.

Desire Please talk sense. In my experience a desire is a desire. You can't 'want' something without wanting to obtain it.

Conscience So you might think, but the very fact that I am talking to you about doing what is right as if you could do it without being motivated by any specific desire is sufficient proof to the contrary. ~9

7 For Kant's arguments on this distinction between acting from duty and merely in accordance with duty, see FMM, p. 58 ff. BA 8 and CPrR, p. 180, A 127 and p. 188 A 144, the latter two drawing the distinction between morality and legality.

See CPrR, p. 221, A 212 for an example of a uniquely moral C°mP°nent of a person's 'inner life'. See p. 16 above.

This appears to be the essence of the argument from the fact of the moral law; from the reality of our awareness of an 'ought'. See CPrR, p. 143, A 56 and p. 157, A 81.

What you have to appreciate is that moral value is something you create ex nihilo. It is not a natural value like the happiness which you claim is the only value you know. It is created whenever a person acts on my injunctions through the recognition that to do so is right. Desire: Hold on now. I can weigh desires to the turning of a hair, but I can't weigh 'desires' in my balance.20 I don't like to be rude, but frankly I think you are wasting your breath on me. You have been addressing the wrong faculty. What you seem to need (if your appeal to duty is to awaken any response) is a faculty which 'seeks' self contentrnent in a way analogous to my constant seeking after happiness. It would have to be a faculty which could somehow bring itself to choose to do what you call 'right' simply in virtue of the fact that it is right.

Conscience: I take my hat off to you. If through my injunction 'Do your duty' I could bring such a faculty into being, then morality-that kind of human behaviour which creates moral value-would be explicable. I can only conclude- transcendentally (if you will pardon the jargon)-that such a faculty must exist, since the form of life I serve (morality) exists.

If I then postulate such a faculty-call it Freedom-my call to duty would be intelligible.2~ You are quite right in regarding my appeal as unintelligible, so far as you are concerned. You are not free. I now recognize that it is not to you that my appeal is directed, but to that capacity in humankind which can, in virtue of its freedom, act according to the call of duty. The capacity to thus act 'on principle', to act without any ulterior motive, is what creates moral value and gives the person who so acts dignity.22 In so far as people are free they can realize their inherent dignity by acting on principle. Reason is quite capable of testing the practical validity of the maxims which you (Desire) come up with, so long as I am present to put my question, and then Freedom (or perhaps we should say 'Humanity-in-its-freedom') can choose to create moral.

20 The point Desire is making here is that these phenomena-duty, selfcontentment, etc.-are not part of its experience or (person)fication aside) not part of the experience or inner life of a person who is without a sense of moral values.

21 This argument from morality to freedom is explicit and is frequently repeated in the CPrR. See, for example, p. 235, A 238-239 in the discussion of the postulatos. Morality, involving an independence from the world of sense, presupposes freedom. This is also expressed in terms of freedom being the ra~io essendi-the logical ground-of morality. CPrR. p. 119, A 5, also pp. 149-141, A 52-53

22 See FMM, pp. 92-93, BA 77-78. Dignity, another uniquely moral phenomenon, is considered to be an outgrowth of self-respect. For further treatment see pp. 123 t~., below  value by acting only on those maxims which are practically valid. (At this point in the dialogue the shadow model of moral causality has been fully illustrated.) To sum up: the causal 'explanation' of moral behaviour involves 1) a rational stimulus (the recognition of the practical validity of a proposed maxim of action) which produces 2) a sense of obligation (a 'desire' to act in accordance with the proposed maxim in order to be consistent, to do what is right, to be good). However, we lack any natural desire to furfill this obligation and consequently, if moral behaviour is to be regarded as possible we must assume 3) that we possess a capacity to furfill the obligation through a free act of will, an exercise which brings about 4) a feeling of self-contentment (a 'satisfaction') which is identical with that sense of dignity which comes with the knowledge that one has autonomously directed one's behaviour with reference to the rational maxim which initiated this sequence of moral behaviour.

The factor which makes this explanation an 'explanation'-a shadow only-is the notion of freedom and its exercise. Desire has a question or two about this capacity which Conscience will now be allowed to struggle with as its answers raise issues which will play a large part in the following chapters.

Desire: You haven't forgotten that this freedom can cut both ways? A person can freely choose to follow my maxims even though they turn out to be practically invalid.23

Conscience: I haven't forgotten. People can choose to be evil. Negative moral values can be created as well. This is the price that being free extracts from a person.

Desire: And people still have dignity if they choose to be evil?

Conscience: Yes. They are responsible24 for their actions, whether they choose to be good or evil. The measure of their dignity when they choose the evil alternative is that, if they are caught and convicted of wrong-doing, they demand to be punished, although you undoubtedly would want them to escape the punishment. However they cannot give up the birthright their freedom confers upon them. Not to punish people is to deny them their dignity as human beings. It amounts to treating them as things which cannot control their behaviour.

Desire: So, why then should people choose the good rather than the

23 This is the problem that Kant tackles in RWL, especially Book One.

24 But see pp. 112 q. below, where responsible behaviour is discussed in connection with agent-naturalism.

27 evil course of action? Clearly you want (if that term makes sense when anyone but I says it) them to choose to do what is right. You presumably see greater worth resting with the choice of good than with the choice of evil. But if their status as beings capable of morality is conferred upon them by their capacity freely to choose between right and wrong courses of action, do they not retain their dignity, their self-respect, their moral worth, whichever alternative they choose? The greater worth that lies with the good choice is not, it seems, a result of this choice being necessary for human dignity but rather it is a function of your own 'preference' for good over evil. What lies behind this preference? What is good about being good? Why should Humanity-in-its-freedom want to be good?

Conscience: From your point of view, where all action is a function of desire, that certainly is a problem. For when you ask why a person should choose good over evil, you are asking why Humanity-in-itsfreedom should want to identify with Reason and so act only upon practically valid (universalizable) principles, and you are naturally curious (in terms of the only motivational structure you know) as to what exactly is to be gained by such a person. You, after all, cannot appreciate what 'good' means for a person if it does not relate to the satisfaction of desire, nor evil if not to unsatisfied desire. Nor can you understand what satisfaction means if it is not cashed in terms of happiness.

Humanity-in-its-freedom, however, is determined neither by desire nor by my pleas viz, that it act as the categorical voice of Reason demands. Humanity-in- its-freedom is not a faculty which can be dictated to. As free, this faculty (person)fied) is subject only to an ontological imperative.25 Humanity-in-its- freedom must exercise this freedom in order to be, and there are only two choices: this persona can either be indulgent (as I would like to put it), seeking prudently or imprudently the satisfaction of its desires as they occur, or it can be rational, acting only on practically valid maxims.26 Now it has no 'preferences' in making this choice, in the sense in which you understand this term. This persona is not predisposed to be indulgent or rational by any feelings which it may experience. It is entirely without dispositions, preferences, indeed, without a character at all. Instead it creates positive or negative moral value through the free choices it makes. Without predispositions,

25 See Chaprer Eight.

26 In the CPrR this choice between two clearcut options is expressed as a choice between succumbing to inclination and therefore realizing only the immediate objects of desire or resisting inclination and following Reason's way of duty, thereby realizing virtue. See pp. 180 95., A 126 ff.  without inclinations to tell it what to do, it cannot avoid acting autonomously and this is the source of its capacity to create moral values. Thus the only question it can ask itself is: 'What ought I to be?', not as you might have thought: 'What do I want to be?' The former question can only be answered by answering the ontologically prior question: 'What am I?' since only in terms of what the persona- Humanity-in-its-freedom-is, can it choose what it ought to be. Its answer to this question is that it is whatever it wills itself to be and that therefore it always retains its freedom to be what it wills. It can create a rational or indulgent character for itself but its character is always being created afresh by each act of will.

Now that you have grasped the ontological circumstances of Humanity-in-its- freedom, I will rephrase your question: 'Why should Humanity-in-its-freedom want to be good?' What sense does it make to say that this persona ought to behave as Reason (categorically) dictates and why does it not make sense to say that it ought to be indulgent?

Desire: Yes, why is Humanity-in-its-freedom obliged to be good, to follow Reason? Why is it not obliged to be happy, to follow my lead? (7lease note that I did not ask why it is not obliged to be evil, as if seeking happiness were evil!)

Conscience: How patient I am. I understand what you are on about. And I simply refuse to take offence at your quite delightful consistency in always equating goodness with happiness.

Once again let me spell out my position: people do not become happy by being good, they become good, period. Whether or not they should also experience happiness or unhappiness as a consequence of their decision to be good is a purely contingent matter and it is quite irrelevant to their choosing to be good. So too with the choice to be evil. Evil people may or may not be happy. Their decision to be evil is not a decision to be happy or unhappy but to be evil. Ontological commitments have only ontological consequences.

Desire: But what about self-contentment? Surely this is experienced as an emotional "plus" and would therefore prompt Humanity-in-its-freedom to choose to be good?

Conscience: There is also an emotional "plus", as you put it, associated with being evil. Consider Milton's Satan. But, to repeat myself, these are not the factors which prompt ontological commitments. They are not (and cannot be) causal factors in what is, by definition, a free act. They are irrelevant because they are contingent side-effects. Good and evil must be understood as ontological values, brought into being by ontological commitments. They don't just happen, they are created through free choices.

Desire: Let me ask my question again in your words: why should Humanity-in- its- freedom choose to be good rather than evil?

Conscience: The answer is complicated. You know that your desires do not arise spontaneously. They are always traceable to some empirical stimulus which aroused them. Reason, by comparison, although it can on occasions be activated by desire, can also engage in spontaneous activity, activity that derives from its own nature.27 Its spontaneity is thus a mark of the autonomous or self- governing aspect of its nature. This self-governing aspect of reasoning is exemplified in the fact that reasoning itself sets the standard for whether or not it is reasoning.28

27 See CPR, p. 473, A 548 B 576, for the perfect spontaneity of reason in creating its own order. See also FMM, p. 106, BA 108.

28 When reasoning takes place properly (and, in a sense, it cannot fail to operate properly), this proper reasoning is marked by the fact that the conclusions arrived at follow from the premises on which they are based. What does this 'following' amount to? Well, what is followed, what is adhered to, are the rules which determine whether an argument is valid or not. And where do these rules come from? They are based on Reason's capacity to recognize instances of class-inclusion. For example, in the famous syllogism-All men are mortal; Socrates is a man; therefore Socrates is mortal-Reason sees that, since men are included within the class of mortals, and Socrates is included within the class of men, then Socrates falls within the class of mortals as well. In diagrammatic terms, Reason can see that the smallest circle is contained in the largest:  

If Reason couldn't see these relationships it couldn't reason. Reasoning is just seeing relationships of class-inclusion and seeing these relationships is reasoning. This ability to see-in diagrammatic terms, that one circle is inside another-is so basic that Reason cannot help seeing such relationships wherever it looks. This explains why its spontaneous tendency to universalize relationships (Is it all right for me to lie? Is it all right for everyone to lie?) is always guided by the very rule which constitutes its activity and its validity. Reasoning is thus always valid, i.e. proceeds in accordance with its own rule, because Reason cannot fail to see (again, using the diagrammatic model) that one circle is inside another. So-called 'invalid' reasoning is always a function of the fact that 'the circles' have not been properly drawn, or in other words, that the terms involved have not been clearly defined in relation to 30

This may sound odd, but all spontaneous self-governing phenomena are odd in having this reflexive feature. You could say that Reason's activity was 'motivated' by its own nature, its tendency to universalize. Now let us return to your question: 'Why should a person choose to do what is right (good) rather than what is wrong (evil)?' or as I would prefer to put it: 'Why should a person choose to live as if they were Reason personified?' Since doing what is right is to be understood in terms of acting according to universalizable maxims, maxims which, when followed, are not practically invalid, my version of the question clarifies the problem which your question raises. For Reason, were it an agent, would always act on practically valid lines. To be specific, were Reason an agent it would always ask (following its tendency to universalize): 'Would this action which I contemplate doing be possible (practically valid) if everyone acted as I intend to act, or would it self-destruct if universally practised?'

Why then should Humanity-in-its-freedom ask this question of itself, and like Reason person)fied, act only on practically valid maxims? The answer is that Humanity-in-its-freedom is-if I may put it this way- pure spontaneity: it must will in order to be. But it is under no compulsion to will to be indulgent or rational, although it must will to be one or the other. Now this pure spontaneity which characterizes Humanity-in-its-freedom would have no value or interest if it made its ontological commitment randomly. Unless it chose deliberately to be rational or indulgent the value created by its choice-moral value-would vanish. Thus the choice it makes must be made deliberately, that is, according to some principle. Furthermore, if it simply made up this principle it would, in effect, bc acting randomly and hence would not be seen as a creator of moral value. The choice iS free, but unless it makes a difference whether the choice is made one way or the other, for good or for evil, then the choice cannot create moral value. What could this each other. So when I point out that an argument is invalid (that it doesn't follow) I always point to the fact that a given term is not in fact related to the other term in the way suggested by the argument. It is the facts contained in the premises of an argument not the mechanics of reasoning, that are incorrect. In other words, I would never suggest to someone that his or her reasoning was invalid on the grounds that, while a given circle was drawn inside a larger circle, he or she somehow managed to see it outside the larger circle. That is assumed to be impossible. What is possible is that the thing under consideration does not actually belong in the circle as believed (e.g. if Socrates were not, in fact, a man.) This explanation accords with Kant's idea in the Analytic of the CPR of how Reason operates in its function as Understanding in the sphere of General Logic.

31 principle of moral choice be? Since it must motivate an ontological choice, it can only be an ontological principle and, for an agent, the only29 ontological motive is to be what you essentially are. But what is the essential nature of a free agent? On the one hand it cannot be the agent's freedom since this aspect of its agency only places it under the onus of choosing. Its essential nature as free does not indicate which alternative to choose. But, on the other hand, the essential nature of a free agent must involve the agent's freedom since such an agent is characterized solely in terms of the freedom of its agency.

Now I have already pointed out that if a free agent expresses its freedom randomly, this exercise of its freedom would not create a moral value since it would not create a characteristic pattern of willing which could be thought of as good or evil. And I also said that the characteristic pattern of willing-the person's character, for short-did not, and could not, determine the continuance of the pattern it displays: the agent is always free to alter its characteristic pattern of willing. The exercise of freedom is only sign)ficant, it only creates moral value, when it is expressed in a series of willings which are consistent enough for a characteristic pattern of either good or evil (rational or indulgent) willing to emerge.30 So it follows that the free activity of willing which creates such patterns will derive its consistency from the free adoption by agents of either the principle of indulgent behaviour (in terms of which each action which they subsequently take will be a foregone conclusion as soon as a given desire makes itself manifest) or a principle based on the categorical advice of Reason (in which each action is tested for its practical validity before being enacted).

If then Humanity-in-its-freedom must act upon its ontological imperative-be what you essentially are-which principle of action ought it to adopt in order to obey this imperative? Now I say that if it adopts the principle of indulgence it, in a sense, limits its freedom to this single act of will, for by definition the adoption

29 But see Chapter Eight, below, where a second ontological motive is suggested.

30 This is true only when we are considering moral worth "in the judgment of men who can appraise themselves and the strength of their maxims only by the ascendancy which they win over their sensuous nature in time." (RWL, p. 43, A 51-52 B 55) That is, when we are summing up the character of a person, we consider his or her actions (retrospectively) to see whether a pattern has developed. But each individual act, in which freedom is exercised (for the moral option) is good in itself, independently of what sort of choices went before or will come after. Ir is this independence of each choice that makes change possible.

32 of the principle of indulgence precludes any further free choices. But Humanity- in-its-freedom is essentially free. It cannot give up its freedom in this fashion and remain true to its nature. Thus from an ontological point of view (accepting the primacy of the ontological imperative) Humanity-in-its-freedom ought to choose to be rational since such a choice involves giving up neither the freedom to oppose Desire, when Reason warns that the following of a given Desire would result in a practically invalid action, nor the freedom to seek satisfaction if Reason rules that such satisfaction is permitted.3~ Nor indeed does the adoption of the rational principle of behaviour rule out, simply in virtue of its initial adoption, the occasional indulgence. It only rules out consistent indulgence on pain of ceasing to display that characteristic pattern of rational willing that marks a person as essentially good. To be what you are is to be ontologically consistent, to be right in the sense of being in line with one's self (which must mean one's essential nature). This is what being good means.

Desire: You seem to be suggesting that to opt for a life in accordance with Reason's dictates is to opt to be more truly oneself, to live more in line with one's real nature as a free being. When you say to Humanity in-its-freedom: "Do your duty." you are really saying: "Create, by an act of will, the identity for yourself most in keeping with your capacity for freedom." You are saying: "Start acting on principle so that you can be someone you can respect, someone you can ontologically live with. Unless you do, you will be nothing but a part of Nature, as much subject to desire as Desire is to those natural occurrences which stimulate it. However you can-in your freedom-become what you essentially are, what you were ontologically 'meant' to be." Now I can appreciate the strength of this ontological appeal. In appealing to Humanity as free, you appeal to the human capacity to create a new, harmonious (which is to say, practically valid) order of existence. As such you treat this persona of humanity as a god, one who has not as yet realized its creative potential but can do so simply through an act of will. But, with this splendid vision before its eyes, will this persona perhaps be blind to the content of the principles it acts on in order to bring this vision into being? I mention this because my world-the world in which happiness is the be-all and end-all-has suffered terribly at the hands of people acting "on principle". You certainly don't want people to identify with a capacity (Reason) which cannot itself distinguish between happiness producing principles and their opposite, do you? Or do you really mean to

3\ There are difficulties with this argument. See pp. 40 '., below.

33 stick to your guns and insist that a good will (which equals the decision to act solely for the sake of duty-whatever the consequences) is the only good thing, the value in comparison with which happiness is an irrelevant contingency.

Conscience: I'm afraid I must, as you put it, 'stick to my guns'. No one can take any pleasure (moMI pleasure that is, the only sort worth having) in the sight of a person visited by happiness who is not also morally good.32 However, I am not against happiness. There is nothing wrong with it per se but, obviously, if the happiness resulting from an action were to be regarded as the standard of worth by which human behaviour was ultimately to be judged, we would have to know, a priori, what principles of behaviour would maximize our happiness if we were to seek happiness "on principle". And Reason, which supplies people with their principles, cannot say a priori what the consequences in terms of happiness of the application of its principles (which could be no more than the counsels of prudence) might be.33 Reason can only determine a priori whether a principle would be practically valid or invalid. It is thus able to distinguish what actions are permitted, but, of course, it cannot say whether such actions would result in a greater or lesser amount of happiness in the world if they were followed. It can only guarantee that people who confine their behavior to that which is permitted by Reason will experience self- contentment, they will know that they are good people, people whose behaviour is determined by Reason.34

Desire: So, if I understand you, you have no objection to people finding happiness incidentally, so to speak, so long as happiness is not what motivates them?

Conscience: That's right  A moral person accepts his or her lot so far as happiness is concerned.

The dialogue may end at this point. The main features of the shadow model of moral causality are now before us. We have been told what it is to be good and how-according to the shadow model of moral causality-this state can be achieved. From the point of view of the original model (based on natuMI causal relationships) the shadow model does not really explain how moral behaviour is possible and, since the original model sets the standard of intelligibility for an explanation, it would seem to be something of a mystery that the shadow model of moMI causality is intelligible at all. In the following chapter we will try to explain this mystery.

32 Cf. p. 4-5, above.

33 See FMM, pp. 57-58, BA 6-7 and p. 77, BA 45-46.

34 ibid., p. 58, BA 7 34