1. Introduction
Kant’s appeal to a (or ‘the’) fact of reason is generally regarded as an important feature that distinguishes his argument in the second Critique from that in the Groundwork; rather than trying to establish the moral law (or, more accurately, the various formulae of the Categorical Imperative) on the basis of the idea of freedom, as he argues in the Preface, ‘the Idea of freedom is revealed through the moral law’ (5: 4), and that ‘had not the moral law been distinctly thought in our reason, we would never been justified in assuming anything like freedom’ (5: 4, note).Footnote 1 But there is wide disagreement on issues as fundamental as what the fact of reason is, what role it plays in Kant’s arguments for pure practical reason’s supreme principle and how this ‘fact’ ‘reveals the idea of freedom through the moral law’.
I shall focus on the latter issue, examining the passages in which Kant refers to the fact of reason and attempting to determine what can be said legitimately to follow from his arguments in those passages. I will also entertain the views of some commentators who have weighed in on what the fact of reason establishes about the status of freedom. My thesis is that, despite the claims of some commentators, the most charitable reading of Kant’s arguments supports a relatively deflationary account of what the fact of reason justifies asserting about the status of freedom. I conclude with a brief argument to the effect that this is also what Kant himself thought the fact of reason establishes.
But first I need to discuss briefly the methodology that I shall employ. Although Kant himself never provides a specific, definitive statement of the moral law, I follow Lewis White Beck, who parses the moral law as follows: a perfectly rational being would act only on maxims that are consistently willable as universal laws, and an imperfectly rational being should do so out of pure reverence for the moral law (Beck Reference Beck1960: 121, 173). Assuming that Kant thought that we are, at best, imperfectly rational beings, for us the moral law has two distinct components: one related to the content of morality’s fundamental principle – that our maxims must be consistently willable as universal laws; and one related to moral motivation – that we ought so to will out of pure reverence.
When Kant claims in the second Critique that our apprehension of the moral law is a ‘fact of reason’, it is important to consider exactly what is being ‘apprehended’. Kant employs the term ‘fact of reason’, or one of its close relatives, to refer to two distinct phenomena. One refers to taking morality as supremely authoritative, whereas the other refers to consciousness of the content of the moral law. Even though some interpreters and even Kant himself (at times) run the two together, each is a distinct phenomenon in its own right, and each addresses an important feature of Kant’s moral theory, especially in the second Critique: that reason, as opposed to experience or some other source, determines both what motivates morally worthy action and what morality requires us to do. Kant argues that pure reason can be practical in each of these respects.
While I believe that Kant claims each to be a ‘fact of reason’, my concern here is not to provide a rigorous defence of that claim.Footnote 2 My employment of this distinction is limited to (a) attempting to make sense of the range of things that Kant says about and on the basis of the fact of reason, and (b) organizing the positions defended by commentators who disagree about what the fact of reason is and, most importantly, (c) what those commentators claim that the fact of reason establishes. Ultimately, the main argument I am making here – concerning what the fact of reason reveals about freedom – is independent of the issue of whether Kant’s claim to have discerned the ‘sole fact of pure reason’ (5: 31) can be sustained.
2. The Fact(s) of Reason
Having argued that ‘[e]mpirical grounds of decision’ are unsuitable for universal practical legislation (which requires practical ‘laws’, not ‘mere maxims’), Kant argues that a will that is suitable for universal legislation ‘must be conceived as wholly independent of … the law of causality’, and hence free ‘in the strictest, i.e., transcendental, sense’ (5: 29). But Kant also recognizes that the idea of transcendental freedom is deeply problematic: there can be no experience that confirms it, in addition to the difficulties discussed in the third antinomy in the first Critique. He asserts that ‘it is the moral law that leads directly to the concept of freedom’ (5: 30). Freedom, Kant claims, is revealed to us on the basis of a ‘fact of reason’.
But to exactly what does ‘fact of reason’ refer? Briefly, Kant appears to use the term to denote two distinct referents. At times he claims it to be a ‘fact of reason’ that we regard the moral law as authoritative – that moral concerns provide a counterweight capable of opposing and overriding inclination-based interests. This is illustrated by the false witness example (at 5: 30), in which Kant asks us to imagine an individual who is threatened with death by a magistrate should he not bear false witness against an honourable person whom the magistrate wants destroyed. He also employs the term ‘fact of reason’ in this manner in discussing the ‘knowledge we can have of and through pure practical reason’ (5: 91–3), which concerns how pure reason can be practical in the sense of having the capacity to motivate us to act ‘without any admixture of empirical motives’ (5: 91). In each instance Kant claims it to be a fact of reason that we take moral considerations as authoritative.Footnote 3
However, Kant also identifies our awareness of the content of morality’s fundamental principle as a ‘fact of reason’. He claims to have shown that
it is at least not impossible to conceive of a law that alone serves the purpose of the subjective form of principles and yet is a ground of determination by virtue of the objective form of a law in general. The consciousness of this fundamental law may be called a fact of reason, since one cannot ferret it out from antecedent data of reason, and since it forces itself upon us as a synthetic proposition a priori based on no pure or empirical intuition. (5: 31)
The sense in which pure reason is ‘practical in itself’ and is ‘directly law-giving’ is in providing the content of morality’s fundamental principle – what perfectly rational persons would do and imperfectly rational persons should do. At the conclusion of the ‘exposition of the supreme principle of practical reason’ Kant proclaims to have ‘shown, first, what it contains’ as well as ‘how it differs from all other practical principles’ (5: 46). Here ‘fact of reason’ refers to our apprehension of the content of the moral law.
It is thus no surprise that some commentators have claimed that the fact of reason refers to our acknowledging the authority of morality’s first principle, whereas others have insisted that the fact of reason refers to our apprehension of that principle’s content. For example, after considering various passages in which Kant alludes to a ‘fact of reason’, John Rawls concludes that it is ‘the fact that, as reasonable beings, we are conscious of [moral considerations] as … supremely authoritative and regulative for us and in our ordinary thought and judgment we recognize it as such’ (Rawls Reference Rawls1989: 101), which is about the authority of morality. As for identifying the fact of reason with our apprehending the content of the moral law, Patrick Kain claims that ‘in this sole fact of reason’ we ‘know’ that there is an ‘undeniable’ moral law to which we are subject and ‘which we know’. ‘Our consciousness of the moral law, in the fact of reason’ ‘is an objective law that governs, indeed the only law that could govern, the will of every rational being’ (Kain Reference Kain2010: 218–19).
Again, my purpose here is not to offer a robust defence of the two facts thesis, although it seems clear that Kant does use the term as I have suggested.Footnote 4 I will further grant that if one (a) apprehends the content of the moral law and (b) conceives of oneself as an imperfectly rational agent (i.e. that one is free and sensibly affected) then, if Kant is right, one is logically committed to holding that she ought to act only on maxims that are consistently willable as universal laws, and that a rational agent should apprehend this. I would also accept that Kant thought that the two facts often do go together – especially that those who apprehend the moral law’s content generally accept morality’s authority (that those who recognize morality’s authority acknowledge Kant’s version of the moral law seems wildly implausible – see note 3). But none of this would show either that the two phenomena are the same, or that an imperfectly rational agent might not refuse to follow the moral law when its dictates conflict with her inclination-based desires.Footnote 5 So not only are the two conceptually distinct, they can indeed ‘come apart’ in both directions: one who apprehends the content of the moral law might not acknowledge its authority, and one who accepts the authority of moral considerations might not apprehend the content of the moral law as Kant understands it.
3. My Interpretation
In this section I develop an interpretation of the passages in which Kant claims it to be a fact of reason that we acknowledge morality’s authority, and then turn to the passages in which he claims it to be our apprehending the content of the moral law. Briefly, I argue that acknowledging the authority of moral considerations is best understood as referring to how we conceive of ourselves, and hence establishes only that we think of ourselves as being able to act contrary to our inclination-based desires, not that we have ever done so or are even capable of doing so. As for apprehending the content of the moral law, I argue that these passages establish that freedom is real only in the sense that it is a coherent idea, one that would – perhaps necessarily – be apprehended by rational beings who have autonomous wills, but from this it does not follow that I have an autonomous will: it does not establish my freedom.
Authority
As I have indicated, Kant claims that our acknowledging the authority of moral considerations is a fact of reason in his analysis of the false witness case and in the elucidation of the Analytic. I will address each of these passages in turn.
Analysis of the false witness case requires a great deal of caution. Kant claims that the man who is threatened unless he bears false witness against an innocent man ‘would admit without hesitation’ that it would be possible for him to refuse to do so. Specifically, he ‘judges, therefore, that he can do something because he ought’ (5: 30).
Kant then proceeds to claim, on this basis, that ‘[er] erkennt in sich die Freiheit’, which he would not have been able to do without the moral law. Exactly what Kant is claiming about the individual in the example turns out to be a tricky, but very important, issue. Something that counts as an Erkenntnis of Freiheit has occurred due to his affirmation that he could resist the order of the magistrate. But exactly what? Following T. K. Abbott (Reference Abbott1989), Beck translates the original German as ‘he recognizes that he is free’ (Kant Reference Kant1993). Mary Gregor (Reference Gregor1997) and Werner Pluhar (Reference Pluhar2002) are importantly different: ‘he cognizes freedom within him’. Subtle nuances of proper translation aside, ‘recognizes’ (as in ‘re-cognize’) might be slightly misleading, since it suggests some previous acquaintance with its object. ‘Cognizes’ seems more faithful to the original. But the more obvious German version of ‘recognizes that he is free’ would include that er ist frei, as opposed to Kant’s actual words. While hardly conclusive, these basic issues of translation support the Gregor/Pluhar version.Footnote 6
There are substantive philosophical issues in play that also favour the Gregor/Pluhar translation. Among the claims that Kant makes regarding the false witness case is that when an agent recognizes morality’s authority within her deliberations (the point of the case), ‘the will is thought of as independent of empirical conditions, and consequently as pure will, determined by the mere form of law, and this ground of determination is regarded as the supreme condition of all maxims’ (5: 31; emphases added). It is noteworthy that Kant does not claim that the will is independent of empirical conditions, or that moral considerations are the supreme condition of all maxims. Instead, they are ‘regarded’ and ‘thought of’ as such – I see myself as being able to act on moral considerations even when each of my inclination-based desires opposes them. This is what emerges from the false witness case: that the agent would judge or believe or affirm that he is free. The larger question is whether any of this establishes the truth of that judgement – whether it establishes his freedom, as the Abbott and Beck translations suggest. To maintain that his ‘recognizing’ that he is free – in the sense of regarding or thinking of himself as free – establishes that he is free would be to saddle Kant with a view that is far too Cartesian; simply believing or judging it to be the case it would not show that such a judgement or belief is correct.
But then in what sense does the agent in the false witness case erkennen anything about Freiheit? Following the Gregor/Pluhar line, what that agent does is to ‘cognize freedom within himself’, as follows. In judging that he is free to refuse to comply with the magistrate’s ‘request’, the agent conceives of himself as being free to act contrary to his inclination-based desires. Such a person would apprehend or encounter or become aware of or discern freedom ‘in himself’, where the latter refers to his conception of himself as capable of resisting inclination-based desires, which Kant maintains presupposes freedom. It is not the claim that the agent (veridically) cognizes that he is free, but rather that he cognizes the idea of freedom, which is what Kant specifically claims (5: 4) to be revealed through the moral law. And this he does in himself (in sich), since freedom is something he discerns within his conception of himself.
The example shows only that in our ordinary moral consciousness we think of ourselves as having the capacity to refuse to bear false witness even if threatened with death for refusing to do so, i.e. that we see ourselves as being able to resist all our inclination-based desires if they run afoul of what morality demands.Footnote 7 But that one conceives of oneself as having this capacity does not entail that one (or any other actual agent) actually has it. And this is how it should be; given Kant’s opacity thesis, any psychological state that we believe confirms the presence of the moral motive might involve only the dear self, dressed in moral finery.Footnote 8 Even if one did refuse to lie for what seemed to be moral reasons, this would confirm neither (a) that one is capable of acting on reverence for the moral law, nor (b) that the moral motive is capable of opposing and overwhelming the desire for life. The false witness passage establishes only that we think of ourselves as capable of acting in accordance with morality’s demands even to the detriment of all our inclinations.Footnote 9
Subsequent passages are best understood along similar lines. Kant’s claim is not that such deliberations establish that one’s will is as a matter of fact independent of empirical conditions, but rather that when I take moral considerations as authoritative, I conceive of myself that way.
The discussion at 5: 91–3, which addresses the interplay in our deliberations between moral considerations and inclination-based desires, also refers to how we conceive of our agency: we see ourselves as capable of responding to and acting upon moral considerations. For instance, Kant appeals to ‘the commonest practical use of reason by producing evidence that the highest practical principle is a principle recognized by every natural human reason as the supreme law of its will’ (5: 91). He also maintains that ‘no one, not even the commonest mind, can fail in a moment to discover in an example that, though he can be urged by empirical grounds of volition to follow their attractions, he can be expected to obey nothing but the law of pure practical reason’ (5: 92). In each of these passages, Kant is concerned with our recognizing the authority of moral considerations within our deliberations. It is a fact about our deliberative field, so Kant claims, that we see ourselves as not being ruled simply by inclination and ‘empirical grounds of volition’, but also as being able to respond to the dictates of pure practical reason.
Granting that Kant is right about the details of our deliberations, what might this reveal about freedom? While one might agree with Kant’s claims about our being able to recognize the influence of ‘rational grounds of determination’ on our wills via the feeling associated with resisting ‘interfering inclinations’, this would directly conflict with Kant’s opacity thesis. That we see ourselves as capable of responding to and acting upon moral considerations even to the detriment of our inclination-based desires does not show that we are capable of doing so. It would show that we are committed to there being a distinction between empirically conditioned and purely rational considerations, and that the latter are exclusively associated with morally worthy action, but not that we are capable of responding to purely rational considerations. Hence even if in the course of one’s deliberations one encounters what appears to be the kind of feeling associated with a rational (as opposed to empirical) determination of one’s will, that cannot confirm that one’s will is indeed responding to purely rational considerations. In short, it cannot establish that I am free – only that I see myself as free.
Content
As noted above, the discussion from roughly 5: 31 to 5: 47 includes reference to our apprehension of the content of the moral law being a fact of reason. Kant claims that ‘pure reason, practical in itself, is directly law-giving’ via the ‘a priori thought of the possibility of giving universal law’ (5: 31). And further, ‘this principle of morality, on account of the universality of its legislation … is declared by reason to be a law for all rational beings … in so far as they are competent to determine their actions according to principles’ (5: 32). In his summary of the Analytic of Pure Practical Reason, Kant concludes that ‘[t]his Analytic proves that pure reason can be practical, i.e., that of itself and independently of everything empirical it can determine the will. This it does through a fact which is autonomy in the principle of morality by which reason determines the will to action’ (5: 42; emphasis added).
This much is clear: Kant thinks that something about autonomy is revealed to us as a result of our apprehending the content of morality’s fundamental principle. But how? Here are some hints. ‘The moral law expresses nothing else than the autonomy of pure practical reason, i.e., freedom’ (5: 33), and ‘pure reason of itself can be and really is practical, as the consciousness of the moral law shows it to be’ (5: 121). Kant argues that the Analytic, which concerns the content of the moral law, shows ‘this fact [wherein pure reason shows itself actually to be practical] to be inextricably bound up with the consciousness of freedom of the will, and actually to be identical with it’ (5: 42).
First, it is important to note that what Kant claims to be ‘inextricably bound up with’ the moral law is ‘freedom of the will’, not ‘freedom of my will’. His central point here concerns our conception of a will, specifically the nature of autonomy and practical reason. Apprehension of the content of the moral law allegedly shows that I can conceive of a will that is free. But that I can conceive of a will determined solely by pure practical reason does not show that pure reason is practical in me – it does not even show that there are any actual entities in whom pure reason is practical.Footnote 10 Bluntly, ‘autonomy in the principle of morality’ is one thing and ‘autonomy in me’ is another.
Moreover, this ‘fact’ cannot entail ‘consciousness of freedom of my will’: since autonomy is not something of which one can have a conscious experience – no ‘fact’ could confirm that pure reason had actually determined my will. That pure reason can be practical (what apprehension of the content of the moral law allegedly shows) does not entail that pure reason is practical in me. Kant is claiming that the content of the moral law includes the idea of an autonomous will; this is ‘autonomy in the principle of morality’. That the moral law is a dictate of pure practical reason shows that, at least for Kant, freedom is a coherent idea. In this way, our apprehending the content of the moral law establishes the ‘objective reality’ of, and hence provides a ‘credential’ for, freedom.Footnote 11
This interpretation also coheres best with other claims that Kant makes about our apprehension of the content of the moral law. For instance, he maintains that, although apprehending the moral law’s content does not ‘expand our knowledge’ of noumena, ‘this fact points to a pure intelligible world – indeed it defines it positively and enables us to know something of it, namely, a law’ (5: 43). Likewise, this appears at the beginning of section II of the Analytic:
In the moral principle as we have presented it there is a law of causality which puts the determining ground of causality above all conditions of the world of sense. We have thought of the will as determinable inasmuch as it belongs to an intelligible world and the subject of this will (man) as belonging to a pure intelligible world, though in this relation man is unknown to us. (5: 50; emphasis added)
Kant appeals to the content of the moral law, arguing that it ‘points to’ an intelligible world (containing wills that are free) and how through the content of the moral law we ‘think of’ a will as belonging to an intelligible world. He concludes that the concept of a being possessing a free will is that of a causa noumenon, one which is ‘for the theoretic use of reason, an empty concept, although a possible and thinkable one’, hence ‘we are assured that it does not contradict itself’ (5: 55–6). But none of this establishes anything about the freedom of my will. Thus I conclude that ‘autonomy in the principle of morality’ should not be taken to establish that I actually possess an autonomous will.Footnote 12
4. The Status of Freedom: (Relatively) Strong Interpretations
Numerous Kant scholars have argued that the fact of reason establishes our freedom, or demonstrates that we possess autonomous wills. Common threads within such arguments include the identification of each of the referents of the ‘fact of reason’ that I have identified as the fact of reason,Footnote 13 and the subsequent argument that, thus understood, the fact of reason establishes our freedom. In this section I examine a sample of such arguments, and conclude that the fact of reason establishes no such thing; the most plausible interpretation of the fact of reason supports only a set of relatively modest claims about what it ‘reveals’. Even those who identify the fact of reason with both acknowledging the authority and apprehending the content of the moral law cannot on that basis assert that they are free in a sense more robust than simply affirming that they see themselves as such.
Pauline Kleingeld maintains that the fact of reason refers to ‘consciousness of [the fundamental law of pure practical reason]’, ‘as the normative principle guiding [one’s] choice of maxims’; ‘agents who regard themselves as reasoning about courses of action’ ‘are conscious of the fundamental law of practical reason as the normative standard that should guide their deliberation’ (Kleingeld Reference Kleingeld2010: 69). That agents ‘acknowledge this principle as the normative standard for the assessment of maxims’ is what Kant calls a fact of reason. Given the reference to ‘the assessment of maxims’, it is plausible to view Kleingeld as referring to apprehension of the moral law’s content, but she also argues that it is ‘the agent’s recognition of the authority of morality’ that ‘back[s] up the belief in freedom’ (2010: 70). And further,
Only when one realizes that one could act against all of one’s inclinations, as in the false testimony example, does one become aware of one’s freedom in the sense required by Kant. In such cases it is not inclination which is conceived as determining the will, but pure practical reason. Only such cases, and not instrumental practical reasoning, disclose freedom of the will. (2010: 71; second emphasis added)
These arguments concern our taking morality as authoritative, which she claims discloses ‘one’s’ freedom.
As we have seen, it is not clear how apprehending the moral law’s content could reveal one’s freedom. For one thing, the moral law is, in the manner of A-propositions, covertly hypothetical in its structure – it claims that perfectly rational beings would act only on maxims that are consistently willable as universal laws, and that imperfectly rational beings should do so, out of pure reverence for the moral law. But it does not assert that there are any such beings, let alone that one who apprehends the content of the moral law is thereby shown to be either perfectly or imperfectly rational. How then could apprehending such a hypothetical establish the existence of any perfectly or imperfectly rational beings, let alone that its apprehender has an autonomous will? Nor does the false witness example ‘disclose’ one’s freedom. Both Kleingeld’s analysis and the false witness example itself appeal to how we conceive of our will, specifically that we conceive of ourselves as being able to resist inclination. Granted, when I conceive of myself as being able to act against all my inclinations I conceive of my will as capable of being determined by pure practical reason. But that I conceive of myself as having an autonomous will does not ‘disclose’ – in the sense of ‘establish’ – that I in fact do have such a will.
Jens Timmermann identifies the fact of reason as follows: ‘the deed – or Factum – in question is the act of giving the moral law to the agent’, and ‘Kant uses our recognition of moral commands to establish that we possess a will that is free’ (2010: 86, 64). Each claim might be construed as referring to content: ‘giving the moral law to the agent’ and the ‘recognition of moral commands’. But Timmermann explicitly identifies the fact of reason as ‘our awareness of the authority of the moral law’ and claims that ‘[f]reedom is confirmed or can be proved by means of this fact’, which ‘reveals to us that we are absolutely free to do as reason commands’ (2010: 83). According to Timmermann, ‘Kant argues that the very fact that we realize our self-subjection to the moral law proves that we are free’ (2010: 84). Here Timmerman claims that taking morality as authoritative establishes the autonomy of one’s will. But as with Henry Allison’s ‘submission thesis’,Footnote 14 it seems to be impossible to affirm the antecedent – what appears to be a case of self-subjection might not be. Hence it does not follow that ‘acknowledging’ the authority of morality reveals one’s freedom in any robust sense, i.e. beyond seeing oneself as capable of acting on purely moral considerations. Consequently, the conjunction of Timmermann’s claims fares no better than the individual conjuncts. Acknowledging morality’s authority shows that I see myself as free, and apprehending the content of morality’s highest principle reveals that those who are free would be subject to the moral law, but neither shows that I do have an autonomous will and hence am subject to the moral law. At best, it logically commits me to seeing myself as bound by the moral law.
Owen Ware’s argument faces similar difficulties. He states that the fact of reason refers to moral ‘necessity’, not in the sense of morality’s being authoritative within our deliberations – ‘the constraint of the law upon our sensibly affected nature’ – ‘but to our consciousness of its “necessity” – the authority of the law as a law of pure practical reason’, which at least suggests that what is being referred to is its content (2014: 8).Footnote 15 But Ware’s primary focus is on morality’s authority. He refers to the ‘Tatsache of moral consciousness’ in the false witness case, which allegedly shows that the person is ‘deferring authority to a principle other than his own happiness’ (2014: 13).Footnote 16 But as we have seen, that entails nothing about the content of the moral law as Kant understands it – it is about the agent’s endorsing a ‘principle’, whatever that may be, that overrides her self-interest. While Kant thought that only his version of morality’s fundamental principle can adequately explain why an agent should rationally favour morality’s demands over her interests, there is nothing in Kant’s treatment of the false witness case that even hints at such an argument.
Ware argues that through the moral law’s ability to ‘command without empirical motives’ ‘we acquire a positive idea of our freedom’ and that through the moral law’s ‘necessity’ (as a law that commands ‘completely a priori’) we first become aware of the ‘independence of our power of choice’ (2014: 15). I agree – for Kant this is how one (a) becomes aware of the possibility of choice independent of inclination, and (b) acquires the idea of positive freedom. But neither entails that one thereby apprehends – veridically – her freedom. While I agree with Ware that ‘judgments from common human reason illustrate the reality of moral consciousness’ (2014: 16), what ‘moral consciousness’ establishes is quite limited: that those who accept the authority of moral considerations see themselves as beings capable of acting contrary to their inclination based interests; and to those who apprehend the content of the moral law it reveals that beings who have autonomous wills would or should act as the moral law prescribes. The Tatsache establishes only that we see ourselves as subject to the moral law’s requirements and as free to act in accordance with them, not that we are autonomous beings.
According to Kain, what is ‘given’ in the fact of reason ‘is an objective law that governs … the will of every rational being’, specifically ‘a principle, indeed the categorical principle to which we are subject’ (2010: 220). He argues that through the ‘sole’ fact of reason ‘we cognize “with apodictic certainty” – we “know”– that there is an “undeniable” moral law to which we are subject’ (2010: 218), and further ‘Kant contends that we also have practical cognition and knowledge of our freedom – the “condition of the moral law”’ (2010: 220). While Kant does claim that freedom is the condition of the moral law (5: 4), he does not explicitly state that our (whoever that might be) or my (the discerner of the moral law’s content) freedom is the ‘condition of the moral law’. He refers to ‘the fact that there really is freedom for this idea is revealed by the moral law’. This is about freedom itself, or rather the idea of freedom, and establishes nothing about the wills of those who apprehend the moral law’s content. Hence these passages do not support Kain’s claim that knowledge of the moral law ‘provides the basis of our knowledge of our freedom’ (emphasis added). Nor does Kant claim that ‘our consciousness of the moral law, in the fact of reason, is the ratio cognoscendi of our freedom (2010: 218); his explicit claim is that ‘the moral law is the ratio cognoscendi of freedom’ – the claim is about freedom, not my or our freedom.
There is, however, an important remark of Kant’s that is noted by Kain and appears to support his position: ‘if there were no freedom the moral law would never be encountered in us’ (5: 4). Kain likely views the antecedent of this claim as referring to ‘my freedom’, supporting something along these lines: ‘If I were not free I would never have been able to discern the content of the moral law.’ As he writes: ‘Because each may find the principle within her own reason and that principle reveals something about her own self, at least one referent of this determinate concept of a causa noumenon is unproblematic – what is designated is the subject herself’ (2010: 223).Footnote 17 But the antecedent does not refer to my or anyone else’s freedom – it refers to freedom simpliciter. As such the passage is better understood as ‘if freedom were not a coherent idea then the moral law [i.e. its content] is not something we would be able to apprehend’.Footnote 18
Rawls argues for something similar as an implication of the gallows and false witness examples: ‘This severe passage expresses the depth of Kant’s conviction that those without a conception of the moral law and lacking in moral sensibility could not know they were free’ (1989: 113). Still, it is one thing to say that a being who is transcendentally free would not recognize her freedom without the moral law (what Rawls claims), and quite another to say that a being who discerns the content of the moral law thereby knows that she is transcendentally free (what Kain claims). Might those who are not transcendentally free still (a) have a conception of the moral law, i.e. apprehend its content, (b) appear to possess moral sensibility as well and even (c) be unable to conceive of themselves in any other way? If so, granting that those who are transcendentally free could know that they are free only through their conception of the moral law and their moral sensibility (they might be necessary conditions), those who have a conception of the moral law and/or acknowledge morality’s authority would not thereby establish their freedom (they might not be sufficient conditions).
Moreover, Kain’s interpretation is straightforwardly inconsistent with Kant’s claim late in the second Critique that ‘The three aforementioned Ideas of speculative reason are not themselves cognitions; they are, nevertheless, transcendent thoughts in which there is nothing impossible’ (5: 135). Although Kant does also say that ‘in the entire faculty of reason only the practical can lift us above the world of sense and furnish cognitions of a supersensible order and connection, though these cognitions can be extended only as far as needed for practical purposes’ (5: 106), it is important to note exactly what Kant claims the fact of reason’s revealing the content and authority of the moral law can provide us with a cognition of: ‘a fact absolutely inexplicable from any data of the world of sense or from the whole compass of the theoretical use of reason, and this fact points to a pure intelligible world – indeed it defines it positively and enables us to know something of it, namely, a law’ (5: 43). The claim is not that we can have a cognition of our own freedom; it is that we have a cognition of one feature of the intelligible world, i.e. the moral law. The claim that our apprehension of the content of the moral law allows us to know that we are transcendentally free is simply not supported by the text.
5. The Status of Freedom: A (Somewhat) Weaker Interpretation
Rawls and Allison identify the fact of reason as referring to our acknowledging the authority of moral considerations, and each argues that it ‘reveals our freedom’ to us. As we shall see, their claims are more attenuated and less ambitious than the bolder claims made by Kleingeld, Timmerman, Ware and Kain. But they better cohere with Kant’s position, and are more plausible on the merits as well.
Rawls argues that ‘Our consciousness of the moral law discloses to us that we can stand fast against the totality of our natural desires; this in turn discloses our capacity to act independently of the natural order’, i.e. that we are transcendentally free. Rawls claims further that ‘Knowledge that we can act from a law of that kind – a law that is a principle of autonomy – is what discloses our freedom to us’ (1989: 112).Footnote 19 Granted, if we are able to ‘stand fast against our natural desires’ and if we knew that we could act on a law of autonomy we would know that we are free. But our taking morality as authoritative cannot accomplish this; it can show that we see ourselves as being able to do so, not that we have the capacity to do so.
A similar line of argument appears within Allison’s ‘submission thesis’. Allison (Reference Allison1989: 121) argues that the act of freely submitting to the moral law establishes our freedom: ‘in the very act of submitting to it I take it as valid for me, I cannot coherently doubt that I am subject to it; nor can I doubt what follows immediately from being subject to it: namely, that I have both an incentive to do what the law requires and the capacity to do it’.Footnote 20 The strongest support for Allison’s claim that those who freely submit to the moral law show themselves to be free occurs at 5: 47. There Kant argues that the moral law shows ‘freedom to be not only possible but actual in beings which acknowledge the law as binding upon them’. Agreed, those who freely submit show themselves not just possibly but actually to have autonomous wills. But that it appears to me that I freely submit to the moral law or acknowledge the moral law as binding is not sufficient to establish that I am freely submitting to it, since no experience, even that of seeming to submit freely to the moral law’s binding authority, confirms that I am free. Simply put, I cannot have an ‘experience’ of ‘freely submitting’ – any ‘experience’ is subject to empirical causality, which rules out ‘freely’ anything. Granted, beings who truly acknowledge and freely submit to it are free. But having the experience of acknowledging or submitting to the authority of the moral law does not establish that I have done so.
Allison argues that this difficulty can be addressed, since ‘the capacity to be motivated by respect for the law entails the capacity to act out of respect for it’ and ‘the mere fact that we can be motivated by, and act out of respect for, the moral law suffices to establish our transcendental freedom. This replaces the initial assertion that the capacity to impose the law upon ourselves establishes such freedom, since it spells out what is really involved in self-imposition’ (1989: 125). Here it is not the ‘fact’ that we impose the moral law on ourselves – we could never be aware of such a ‘fact’ – but the capacity to be motivated by respect for the moral law, which entails that we can act out of respect for the moral law. This, Allison claims, is what is actually involved in self-imposition: ‘what the fact of reason established is not the existence of actions motivated by respect for the law, but merely the capacity to act on the basis of such motivation’, which ‘suffices to establish our transcendental freedom’ (1989: 126). More recently, Allison argues that ‘what the [fact of reason] supposedly shows is not simply the necessity of presupposing autonomy, but its actuality as a capacity to be motivated by purely moral considerations, a capacity of which we are conscious even if it is never exercised’ (Reference Allison2013: 293–4). But our taking morality as authoritative, even in the sense of having an ‘experience’ of taking morality as authoritative, or of having the capacity for being motivated by reverence, shows only that we conceive of ourselves as having the capacity to be motivated by reverence for the moral law; it does not show that we actually have that capacity. Such a capacity is, for Kant, simply not something of which we can be conscious.Footnote 21
Still, there is an important feature of the claims made by Rawls and Allison about how the fact of reason discloses our freedom that renders their positions eminently plausible. Both Rawls and Allison repeatedly stress that claims about how the fact of reason reveals our freedom must be understood as being limited to ‘the practical point of view’.Footnote 22 When discussing what seem to be contentious claims, especially those that might invite charges of metaphysical overspeculation, Kant often employs ‘from a practical point of view’ and similar phrases, in order to restrict the domain of discourse in question to what is necessary for the employment of practical reason.Footnote 23 For Kant, a claim that would be illegitimate if made within the context of theoretical reason is permissible if its content is necessary for the employment of practical reason itself. And he repeatedly warns that freedom’s legitimate use is limited to what pure practical reason requires of it, i.e. mainly what is required for the purposes of imputation and moral responsibility, the primary ‘practical purposes’ for which practical reason requires freedom, and that this restriction is meant specifically to rule out any theoretical claims.Footnote 24
While much more would need to be said to provide a complete account of Kant’s understanding of the practical point of view, I will note that Kant himself saw the practical point of view as very limited in scope and highly restrictive in terms of what can be said about what is acknowledged or affirmed from that point of view – only what is necessary for practical reason itself. Claims such as those made by Allison and Rawls to the effect that the fact of reason ‘reveals my freedom to me’ involve just the type of claim about freedom that Kant would declare illegitimate, unless restricted to the practical point of view.Footnote 25
Agreed, from the practical point of view, i.e. from the point of view of seeing ourselves as agents who are the authors of their own actions and morally responsible for them, subject to the moral law, we must see ourselves as free. But that is because the ‘practical point of view’ refers to ‘the standpoint of seeing ourselves as deliberating agents’, precisely the standpoint from which we – perhaps inescapably – see ourselves as free. As Allison has argued, although a sceptic may well raise the epistemic possibility that the fact of reason and the freedom that it involves is an illusion, ‘what is denied [by Kant] is … the possibility of considering it as illusory from the practical point of view’ (Reference Allison2013: 298). From the first-person standpoint – in the moment, as deliberating agents facing a choice – we view ourselves as capable of making autonomous choices. As I (as well as Allison and, apparently, Rawls) understand it, adopting the practical point of view is sufficient for seeing ourselves as autonomous. If so, then to question one’s freedom is to abandon the practical point of view.Footnote 26
6. Conclusion and Reply to an Objection
Although some have argued in response to ancestors of this article that once through the Analytic, Kant has no hesitation in asserting that, thanks to the fact of reason, we know that we are free, and hence undoubtedly subject to the moral law, there is an undeniable tension between that claim and much of the text of the second Critique, perhaps most glaringly the status of a ‘postulate of pure practical reason’ that Kant confers on not only God and immortality, but also autonomy of the will. Each of the following appears in the Dialectic:
Is our knowledge really widened in such a way by pure practical reason … ? Certainly, but only from a practical point of view. For we thereby know neither the nature of our soul nor the intelligible world, nor the Supreme Being as they are in themselves, but have only united the concepts … and have done so entirely a priori through pure reason … only by means of the moral law and merely in relation to it. … But how freedom is possible, and how we should think theoretically and positively of this type of causality is not thereby discovered. All that is comprehended is that such a causality is postulated through the moral law and for its sake. (5: 133)
[The three postulates] are nothing else than understanding and will, in their contrasting relation to one another, as they must be thought in the moral law, i.e., as they must be thought only insofar as a pure practical use is made of them. … their use being restricted solely to the practice of the moral law. (5: 137)
[T]his reality [attributed to God, immortality and freedom] is still given only with reference to the practice of the moral law and not for any speculative use. (5: 138)
In each of these passages, which occur late in the Critique, Kant reconfirms the status of freedom as a postulate whose legitimacy is limited to the practice of the moral law, not something of which we have knowledge or a cognition of any theoretical sort. If the earlier arguments have shown that we have a robust cognition of our freedom, that we know that we are free, why subsequently treat freedom as (merely) a postulate of pure practical reason? This provides strong, if short of conclusive, evidence that Kant himself did not believe that the fact of reason warrants our asserting that we are free in any robust sense.
Others have claimed that Kant’s worries about the whether the moral law might be an ‘empty fantasy’ and virtue a ‘mere phantom’ are ruled out as a result of the argument in the second Critique. But Kant reiterates each of these concerns, at 5: 153 and 5: 154. If the argument has shown that we know that we are free, hence subject to the moral law, there should be no such concerns. This also favours the interpretation that I have argued for.
However, one might object on the grounds that the deflationary account that I have endorsed conflicts with Kant’s claim that freedom gains ‘objective reality’ thanks to the fact of reason.Footnote 27 But I believe that Kant’s account of how the fact of reason confers objective reality on freedom actually supports my view.
On Kant’s view, freedom is ‘subjectively’ problematic: the reality of freedom cannot be established deductively or by example. But the fact of reason in the sense of our awareness of the content of the moral law provides grounds for positing not only negative freedom as independence from the causal realm, but also positive freedom, whereby reason as an original cause is capable of moving the will to act. How so? The moral law shows the faculty of freedom ‘to be actual in beings which acknowledge the law as binding’ (5: 47). Since, for reasons discussed above, this cannot be understood as referring to any examples of actual beings, Kant must be understood as referring to the perfectly and imperfectly rational beings referenced in the moral law itself: it is the thought of those beings which illustrates the possibility of causality through freedom, i.e. the idea of positive freedom that could not be established in the first Critique. Given that the moral law includes the thought that pure reason determines the will directly when fully rational agents act as they would and imperfectly rational agents act as they should, our consciousness of the content of the moral law is sufficient to ascribe objective reality to not only negative but also positive freedom. Freedom is thus ‘objectively real’ insofar as both negative and positive freedom are required by pure reason itself, i.e. to fulfil needs of both theoretical and practical reason, without reason’s degenerating into transcendence and overspeculation, especially given the strictures associated with the practical point of view. While this is hardly a rigorous analysis of Kant’s position on objective reality, it does provide a sketch of how a minimally plausible account might proceed.
Commentators whose position is most similar to mine include Pawel Lukow, who argues that the fact of reason ‘is neither an empirical fact nor an indubitable, self-evident datum of reason of which we are aware’ and that it ‘cannot play the role of the Cartesian God who guarantees the truth of the clearly and distinctly perceived contents of consciousness, because there is no guarantee that these contents refer to something, we cannot assume that our consciousness of them mirrors a metaphysical order, let alone that it provides a full or adequate representation of such reality’ (Reference Lukow1993: 210). What it can do, then, is assure that ‘Through this fact we can acknowledge (erkennen) the moral law’ (1993: 213), without there being any guarantee either that our apprehension of its content is correct, or that we possess autonomous wills. This seems correct; although Kant thought that the moral law is indeed as he claims we apprehend it, we could be mistaken, and we could be mistaken as well about whether we possess autonomous wills.
According to Ian Proops, given that we take ourselves to be under genuine obligations, and that for them to be ‘genuine categorical obligations’ we must be capable of discharging them: ‘Then, with that much established, we reason that such obligations are only possible if Pure Reason is practical, that is to say, we are free. This act of becoming conscious of one’s own freedom can be thought of as a proof of one’s entitlement to regard oneself as free’ (2003: 229). While I agree with Proops that Kant believes that we are entitled to regard ourselves as free, I would stress that this is limited to the practical point of view. I would also add that, from the practical point of view, we are not just entitled – we are required to see ourselves as free. However, again following Proops, given that we take ourselves as being under genuine obligations, since we would be under genuine obligations only if we are free, what follows is only that we take ourselves as being free. It cannot establish ‘consciousness of our own freedom’ in any further sense. To claim, as Proops does, that we are ‘conscious’ of our ‘own freedom’ suggests something beyond the limits of what Kant would – and should – consider legitimate.
Marcus Willaschek adopts a position that is similar to mine; at one point he actually gestures in the direction of my argument but never pursues it. Regarding the three postulates of pure practical reason, he maintains that Kant’s claim that they have objective reality does not mean that their objects really exist, ‘but rather that the content of these ideas is specific enough to refer to determinate objects’ (2010: 193). Hence, ‘although Kant’s arguments for the postulates are meant to show that we must believe in the existence of God, freedom and immortality, the practical extension of our cognition [earlier described as belief/faith] does not concern the existence of God, freedom and immortality, but only the “objective reality” of our concepts of God, freedom and immortality’ (Reference Willaschek2010: 191).Footnote 28 As far as ‘theoretical cognition’ is concerned, what is gained is that otherwise merely ‘logically possible’ and thus ‘problematical’ concepts turn out to have ‘real possibility’. My main disagreement with Willaschek’s position is that he treats the three postulates as being on a par. But the fact of reason supplies objective reality to freedom, and not to God and immortality. In addition, freedom’s status as a postulate depends only on pure practical reason’s commitment to imputation and responsibility, whereas God and immortality depend on a more contentious, and far less plausible, commitment to the requirement that virtue and happiness be rendered proportional.
