While many readers are sympathetic to the substance of Kant's moral theory, few are comfortable with his arguments that purport to establish the authority of the moral law. The sticking point—or, for those less favorably disposed to Kantian morality, the breaking point—is the prominent role played by freedom. In Kant's hands, the innocuous and commonsense view that morality has authority only for those capable of acting morally engenders a thorny philosophical problem. For Kant, only those with free will possess such a capacity, and as a result, his justification of the authority of morality requires him to establish, or demonstrate, that there is such a thing as free will. But since Kant conceives of freedom as independence from natural causality and, at the same time and for the same reason, denies the possibility of cognizing freedom, a successful demonstration must clear formidable philosophical hurdles. Not surprisingly, Kant's efforts have been subject to vigorous attack ever since they emerged in print. In what follows, I defend Kant against the most damning of these criticisms, which is that his post-Groundwork demonstration ultimately retreats to a dogmatic assertion of the authority of morality.
Before outlining my strategy, I want to lighten my load by situating Kant's demonstration of freedom within its proper dialectical context. Two points in particular alleviate some of the looming challenges. First, Kant's demonstration is not meant to repel challenges to morality issued by the hard determinist or fatalist (Sch 12–13); Footnote 1 Kant's intended addressee is not someone worried that determinism as a metaphysical doctrine undermines the possibility of morality (A 803–4/B 831–2; Sch 13). Kant advises those troubled by such thoughts to reread the Critique of Pure Reason (KpV 97), which, he contends, proves the metaphysical possibility of freedom. (Allison aptly dubs this the ‘therapeutic’ function of transcendental idealism [Reference Allison and Baiasu2013: 297].) Second, at least in the Critique of Practical Reason, Kant does not try to win over those who claim that an argument establishing the authority of morality must be based on theoretical, nonmoral premises.Footnote 2 The Critique is rather aimed at someone who acknowledges the importance of morality but worries that her undeniable susceptibility to sensible desire indicates a subjection to sensible desire that renders her incapable of acting morally. More specifically, the Critique’s demonstration of freedom responds to worries that our wills are ‘empirically conditioned’ (KpV 15; see also KpV 94, G 406). An empirically conditioned will is limited to means-ends deliberation regarding ends ultimately grounded in sensible interests and to motivation by these ends. Such wills lack the capacity to act on the basis of pure practical principles, hence the worry that provokes Kant's response: if we possess empirically conditioned wills, we cannot act morally; and if we cannot act morally, moral obligation is a mere chimera (G 407).
This characterization of Kant's project has the salutary consequence of relieving him of two onerous argumentative burdens, namely, defeating the fatalist and convincing the skeptic. But his path is still daunting. A non-empirically conditioned will is on some occasions the source of its own causality, and the possibility of cognizing a will that is its own cause is excluded by the structure of human experience (KpV 29, 48): on Kant's account of theoretical cognition, our minds structure the sensible manifold by ordering it into a temporal series, an ordering that requires that all events be cognized as following from antecedent causes. Since an experience or cognition of free will is impossible, it is hard to see how a demonstration of free will could get off the ground.
The seeming impossibility of demonstrating freedom has led even friendly readers to complain that Kant fails to vindicate morality's claim to authority. With respect to the second Critique, the most popular (and lethal) objection is that in lieu of a demonstration of freedom that would fulfill this function, Kant retreats to a dogmatic assertion of morality's authority (Bittner Reference Bittner1989: 88–90; Guyer Reference Guyer2000: 138; Ameriks Reference Ameriks2003: 254–55n; Reference Allison and Baiasu2013: 171–72; Wood Reference Wood2008: 135). On this view, Kant eschews any attempt to provide a knockdown proof of human freedom and opts for an unpalatable alternative, which is to trot out the ‘fact of reason’, hoist its flag, and hope that we salute. In Schopenhauer's memorable words, this strategy is said to rely on ‘an outrageous petitio principii’.
The plausibility of this objection can be seen by examining the orthodox interpretation of Kant's post-Groundwork argument for free will. Although proponents of the orthodox interpretation differ on the details (Henrich Reference Henrich and Velkley1994a: esp. 83; Baron Reference Baron1995: 43; Schönecker Reference Schönecker2013), they agree on the basic steps:
1. Our consciousness of the authority of the moral law includes the awareness that we ought to will in accordance with the moral law.
2. Our consciousness of the moral ‘ought’ establishes that we can will in accordance with the moral law (via ‘ought implies can’ [KpV 47]).
3. Our consciousness of the ability to act in accordance with the moral law is a consciousness of our freedom (via the conceptual equivalence of a moral will and a free will [KpV 27–30; Rel 50n]).
Now for Kant's argument to go through, the consciousness in (1) must not be mistaken or illusory and the ‘ought implies can’ principle invoked in (2) must be sound. But both premises leave Kant deeply vulnerable to charges of dogmatism or circularity. Since a person's ordinary moral consciousness of the authority of the moral law for her will be veridical only if she is free, this consciousness cannot be used to establish that freedom. Second, as I noted above, the Critique is targeted at those plagued by the specter of their inability to act as morality demands. Thus, (2)’s bald insistence that ‘ought implies can’ will not, absent additional supporting argument, vindicate morality against such doubts. Kant's critics can thus be forgiven for suspecting that his argument is propped up by ‘moralistic bluster’ rather than rational support (Wood Reference Wood2008: 135).
I will try to put these concerns to rest by showing that Kant offers a rigorous and coherent demonstration of free will. In my view, Kant hews fairly closely to the schematization just offered, but he supplies arguments for each step that have so far gone unrecognized or unappreciated, despite their considerable merit. I will not be so ambitious as to claim that Kant's program is ultimately sound. I am content to show that if we grant the correctness of Kant's analysis of morality—if we grant that the moral law really is the fundamental principle of morality—his demonstration of free will is on much firmer footing than is usually assumed.
The central tenet of my interpretation is that Kant's demonstration is best understood in practical, rather than theoretical, terms.Footnote 3 In Kant's view, a theoretical demonstration of x’s existence presents x in experience (sense perception, scientific experiment, etc.) or shows that x has some necessary connection to experience. If a theoretical demonstration is successfully prosecuted, propositions regarding the existence of x merit assent by any theoretically rational being. It is for this reason that Kant tries on a theoretical demonstration of freedom before a practical one; the Groundwork famously argues that we earn membership in the intelligible world by dint of the fact that our theoretical cognitions are governed by pure rational principles. This membership allegedly entails that we have the capacity to act on the basis of pure rational principles. While the Groundwork does not assert that freedom can be presented in experience, the demonstration is theoretical in the following sense: the truth of the proposition that assures our seat at the noumenal table—viz., theoretical cognition necessarily employs pure rational principles—is secured by the transcendental claim that the employment of these principles is a necessary condition of experience. Since we cannot deny the reality of experience, we cannot deny either our membership in the intelligible world or our freedom.
In the second Critique, Kant pivots to what I call a practical demonstration of freedom. In a strict, first-person sense, a practical demonstration of x’s existence obtains when an agent wills x into existence. In an extended, third-person sense, a practical demonstration works by showing how one could will the object into existence. Kant's texts provide a demonstration in the extended sense while individual agents carry out a demonstration in the strict sense. It is important to emphasize that we do not practically demonstrate freedom by bringing a free will into existence. Kant identifies freedom with the capacity to regulate one's will by pure rational principles and act independently of sensible interests (KpV 47). This capacity is not itself something we can bring about; if we didn't have it, there would be nothing we could do to acquire it. What the demonstration brings into existence is a moral will, a will regulated by the moral law and capable of compliance with it. But since acting morally is acting freely, bringing a moral will into existence actualizes our capacity for freedom. And by actualizing our capacity for freedom, we demonstrate that we possess it. In short, the second Critique offers a demonstration of freedom by showing how agents can bring about the existence of a will regulated by, and capable of acting in accordance with, the moral law.
Kant's decision to abandon the Groundwork strategy (KpV 8, 47–48), along with all pretense to an argument grounded on theoretical premises, often meets with disapproval (see, for example, Wood Reference Wood2008: 135; cf. Henrich Reference Henrich and Velkley1994a). But interpreting Kant's demonstration of freedom and corresponding justification of morality in practical terms has several significant advantages. It provides a way around the Groundwork’s failed theoretical demonstration, while rebutting charges of dogmatism. (To my mind, the Groundwork argument has at least one fatal flaw: it does not show that we can act on pure practical principles, but leaves open the possibility, discussed below, that such principles can move us to act only when they serve a sensible desire.) It also makes better sense of Kant's fact of reason doctrine, with its emphasis on the role of a deed in the demonstration of freedom. Finally, and most importantly, a practical demonstration coheres with Kant's picture of the will, specifically his conception of willing as efficacious representation, in ways that a theoretical demonstration does not. As I argue below, Kant holds that practical laws determine a will only when they are efficaciously and self-consciously represented by that will. As a result, the demonstration of the existence of a will determined by the moral law—a free will—must originally be a practical one. (Once an agent has practically demonstrated her freedom, she can provide derivative, theoretical demonstrations. But these derivative demonstrations have an epistemic dependence on her original, practical demonstration.)
Before proceeding, I want to clarify a crucial point. Pure reason is practical, Kant says, when it ‘determines’ a will, and he defines a free will as a will determined by pure practical reason. So one of the fundamental tasks of the second Critique is to show that pure practical reason determines the will (KpV 45). But Kant is less careful than he should be about distinguishing two different aspects of determination, one upstream from acts of choice and the other downstream. These must be disentangled if we are to get clear on how Kant's demonstration of the practicality of pure practical reason is meant to work.
Looking upstream from acts of willing, or choice, we find determining practical principles. For Kant, practical principles state what it would be morally or prudentially good to do, and they can be sorted into three types: specific principles of action, general principles of ends, and second-order principles stating deliberative rules. Practical principles determine a will when they perform their characteristic role in the will's practical activity, that is, when they ground judgments regarding what we are to do, the ends we ought to pursue, or the way we are to deliberate. In the first two cases, practical principles function as premises of a practical syllogism (see McCarty Reference McCarty2006). For example, ‘treating people with respect is good’ serves as a major premise, which along with a minor premise such as ‘telling the truth is respectful’, grounds the judgment ‘telling the truth is good’. In the third case, practical principles, such as the moral law and the principle of happiness, guide deliberation about the premises employed in practical reasoning. Looking downstream, we see that determinations of choice typically result in action (KpV 42), namely, the action specified by the choice or practical judgment (KpV 15, 42; G 446). (I use ‘choice’ and ‘judgment’ interchangeably since judgments are expressions of choices.) It is this feature of determination that enables practical judgments to function as efficient causes of action. As a bit of shorthand, I will designate the latter as the ‘efficacious’ aspect of determination and the former as the ‘normative’ aspect (Ameriks [Reference Ameriks2003: 252–53], Henrich [Reference Henrich and Velkley1994: 94-5], and Morrisson [Reference Morrisson2008: 15–16] describe this distinction in slightly different terms). An agent normatively determines her will by a principle when she takes it to be a normative standard for deliberation, choice, or action. The efficacious aspect of determination refers to that which moves agents from choice to action.
To establish that pure practical reason can normatively determine the will is to establish that the will's set of practical principles includes pure practical principles and that these principles sometimes ground the will's (Willkür) acts of choice (MS 213). To certify that pure practical reason can efficaciously determine the will is to show that choices grounded in pure practical principles can, of themselves and independently of sensibility, move us to act in the way the principles specify. It is to show that pure practical reason, in the guise of pure rational principles, can be an efficient cause of our actions (KpV 48).
Kant first draws this distinction in his 1775 lectures (Kuehn Reference Kuehn, Denis and Sensen2014: 51), though at this point he still believes that efficacious determination, determination via a principium executionis, occurs by means of thoughts of divine reward and punishment (27: 274–75, 278). In Kant's mature moral philosophy, a will determined by pure practical reason is a will both normatively and efficaciously determined by pure practical principles. Insofar as a free will is a will with the capacity to be determined by pure practical reason, we can characterize free will as a will with the capacity to be both normatively and efficaciously determined by pure practical reason. In effect, the two aspects of determination by pure practical reason correspond to two aspects of freedom. What I call the ‘legislative’ aspect of freedom, or ‘freedomL’, is the capacity to be normatively determined by principles of pure practical reason (KpV 29). By contrast, an unfree will is normatively determined solely by empirical principles grounded on conceptions of the pleasurable or agreeable. Freedom in its ‘efficacious’ aspect, or ‘freedomE’, refers to the capacity to be moved to act by pure rational principles and the practical judgments in which they figure (G 445–47). An unfree will is moved to act by a sensible attraction to an object. Both aspects are necessary for freedom, and a will in which just one obtains is not free.
My interpretation of Kant's demonstration of freedom naturally divides into two sections, one focusing on freedomL, the other freedomE. In the first section, which defends an expanded version of the first premise of the orthodox argument, I argue that Kant is committed to the following claim: we actualize our capacity for freedomL by making the moral law the fundamental law of our will. In brief, we make the moral law the law of our will by bringing about the normative determination of the will by the moral law—hence the practical nature of the demonstration. We bring about the normative determination of the will by the moral law by using the moral law to regulate our choice of practical principles.
Unfortunately, Kant's account of the normative determination of the will does not by itself, nor with the assistance of the efficacity condition, explain how choice is efficacious. Rather, Kant's account leaves open the possibility that agents can act in virtue of their practical judgments only by means of a sensible desire that moves them to act so. If this possibility were not foreclosed, it would sabotage Kant's justificatory project. The second section of my paper illuminates how Kant wards off this danger. Eschewing the popular ‘ought implies can’ strategy implicit in the orthodox argument (which renders Kant's justification of morality circular or dogmatic), I show that freedomE is secured by a Kantian flavor of non-Humean motivational judgment internalism, which holds that practical judgments generate an affective force that moves agents to act.Footnote 4 It turns out that the demonstrations of freedomL and freedomE are two sides of the same coin: freedomE follows from freedomL. When one actualizes freedomL, one actualizes freedomE, demonstrating one's freedom in toto and securing the authority of the moral law.
1. Freedom in Its Legislative Aspect
Kant's discussion of the demonstration of freedom in its legislative aspect occurs at the beginning of the second Critique. He starts by uncovering the a priori principle of practical reason and then explains why we have reason to think that we possess pure practical reason. After briefly sketching the first step of his argument, I will discuss the second at length, as it contains the claim crucial to my interpretation: we can bring about the determination of our will by pure rational principles and make the moral law the fundamental law of our will.
Kant arrives at the a priori principle of practical reason through an analysis of the nature of practical principles and practical laws (KpV 31). Kant defines practical principles as propositions that contain a general (normative) determination of the will. Practical principles are subjective when they are valid only for particular agents, objective when they are valid for any rational agent whatsoever (KpV 19). Objective practical principles are practical laws. Since a priori principles of the will must be unconditionally valid, they must share the features of practical laws (KpV 20). Kant then argues that certain kinds of principles are incapable of functioning as practical laws, viz., principles whose content derives from an interest based on pleasure associated with the object contained in the principle (KpV 21, 25–27). After characterizing such principles as material principles, Kant concludes that the objectivity of practical laws must derive from their form, the form of universal lawfulness (KpV 27). To confirm the purely rational pedigree of practical laws, Kant reminds us that a principle with a lawful form necessarily issues from pure practical reason since ‘the mere form of law can be represented only by reason’ (KpV 28).
With this analysis in hand, Kant can identify the a priori principle of practical reason. Since a priori practical principles are formal, they tell us nothing about what we are to will. They tell us only how we are to will—we are to will in such a way that our practical judgments are ultimately governed by universal lawfulness. (As I discuss below, formal practical principles are second-order principles governing practical deliberation.) We are to will in such a way that our practical judgments have an objective, rather than merely subjective, validity. Accordingly, there is really only one a priori practical principle, the ‘fundamental law of pure practical reason’: ‘so act that the maxim of your will could always hold at the same time as a principle in a giving of universal law’ (KpV 30).
At this stage, Kant has not yet shown that pure reason is practical for anyone. His a priori analysis of the fundamental law of pure practical reason does not secure the synthetic claim that it is the law for us; it will not be the law for us if we lack pure practical reason. Hence the need to verify that we do posses pure practical reason, which can be achieved by showing that principles of pure practical reason normatively determine some of our practical judgments.
I have insisted that this demonstration is necessarily a practical one. That is, the existence of a will normatively determined by pure practical principles can be demonstrated only insofar as one brings about such a will, or normatively determines one's will by pure practical principles. To see why, we need to pay special attention to Kant's understanding of will and to what it means for a will to be normatively determined by practical principles. For Kant, will is a specific kind of desire (Begehrungsvermögen), which is ‘a being's faculty to be by means of its representations the cause of the actuality of the objects of these representations’ (KpV 9n, translation altered). As a species of desire, willing is a self-conscious representing that leads to the actualization of the object represented. This definition enables us to state two individually necessary and jointly sufficient conditions of willing, which I will christen the ‘self-consciousness condition’ and the ‘efficacity condition’. To see what happens when a principle normatively determines the will and why the demonstration of freedom must be practical, these two conditions must be examined in more detail.
The self-consciousness condition states that if a principle is to determine our will, we must self-consciously represent it. (Here and throughout, ‘will’ refers primarily to Wille in the broad sense, which comprises Willkür, the faculty of choice, and Wille in the narrow sense, the faculty of practical principles [MS 213].) Since will is an exercise of practical reason (G 412), the representations by which willing leads to its objects include concepts, judgments that combine concepts, and principles that render judgments valid or invalid (KpV 32; MS 213). Furthermore, as a type of thinking, the representational activity characteristic of willing is self-conscious (B131–32, 142). As a result, practical principles regulate our practical judgments only if the principles are self-consciously represented. The fundamental law of pure practical reason normatively determines the will only if we self-consciously represent it as so doing. By contrast, as Engstrom (Reference Engstrom2010a: 135) remarks, a theoretical law can govern the objects subject to it regardless of whether this relationship is even capable of being known by the objects in question; the laws governing Earth's orbit around the sun exist as laws even though Earth has no representation of them.
The efficacity condition states that a practical principle determines our will only if it is efficaciously represented, that is, represented in such a way that it at least sometimes leads to the actualization of its object. This condition marks the distinction between practical and theoretical reason. Theoretical reason is primarily employed in the service of determining the concepts of objects, but practical reason also aims at making its objects actual (wirklich) (B x; KpV 67).Footnote 5 While theoretical cognition of an object depends on the actuality, or existence, of its object, practical cognition aims at bringing its object into existence. (A second distinction between theoretical and practical reason is that theoretical reason establishes what is the case, and practical reason establishes what ought to be the case [A633/B661].)
It is important to note that the efficacity condition cashes out differently for different kinds of practical principles. Some practical principles have an action as their object, e.g., ‘always tell the truth’. We make these principles efficacious when we act in light of them. Other practical principles, e.g., ‘health is good’, have as their object a purpose or end. These principles are often too abstract to be directly efficacious. I cannot become healthy just like that, so if such a principle is to be efficacious, it must determine more specific judgments. One could, for instance, combine the principle ‘health is good’ with a constitutive judgment about how it is that one will become healthy (say, by exercising regularly) to form a practical syllogism. From the major and minor premises I draw the conclusion that I ought to exercise regularly. If I then act in virtue of this judgment, the principle is efficacious. Coming at this from the ground up, insofar as I aim at exercising-in-order-to-be-healthy, I turn the principle ‘health is good’ into a practical principle by making it operative in my will.Footnote 6
A third type of principle—second-order principles that purport to govern deliberation about our more substantive practical principles—is especially important for the present discussion. For Kant, the moral law in its various formulations and the principle of happiness are the principles of this type. Neither tells us directly how to act. Rather, the moral law and the principle of happiness furnish the basic rules of two models of practical deliberation, rules that provide the criteria for the validity of substantive moral and prudential principles regarding actions and ends. (Substantive principles that fail to meet these criteria will be invalid or irrational.) The moral law tells us that a valid moral principle is one that any rational agent should adopt. The principle of happiness tells us that valid prudential principles are those that enable agents to promote their own happiness.
Since these second-order practical principles are aimed at regulating the formation and application of our more specific practical principles, they will be efficacious when they do regulate these principles. (Principles regarding ends can also have this function; for example, I can use the principle ‘health is good’ to determine the validity of a more specific principle such as ‘eating well is good’.) Here another distinction must be drawn. To regulate one's will by a principle in a strong sense is to incorporate that principle into one's character and to be disposed to apply the principle whenever it is salient. In a weak sense, to regulate one's will by a principle is simply to take it as a normative standard and to consider oneself accountable for violations of the standard. Good Kantian reasons favor the weaker version. In the case of moral principles, the distinction between the two senses corresponds to the distinction between autonomy and autocracy, and Kant holds that a demonstration of freedom need only establish our autonomy (Engstrom Reference Engstrom and Timmons2002a; Baxley Reference Baxley2003). On the weaker version, a second-order practical principle is efficacious when it functions as a normative standard for the formation and application of more specific practical principles. (The efficacy of principles of ends and actions depends on the possibility of efficacious determination, hence the necessity of a demonstration of freedomE.)
With these preliminaries on the table, I can present the main argument of this section. Kant's demonstration of freedom must show, or explain how we can show, that pure practical reason normatively determines our will. The key to the demonstration lies in the self-consciousness and efficacity conditions that, taken together, entail that the fundamental law of pure practical reason, ‘so act that your maxim could always hold at the same time as a principle in a giving of universal law’, normatively determines a will if and only if it is self-consciously and efficaciously represented by that will. From this it follows that the fundamental law normatively determines our will when we do self-consciously and efficaciously represent the fundamental law. In other words, the second Critique establishes that we can bring about the normative determination of the will by willing in accordance with the fundamental law of pure practical reason—by using it to regulate, in the weak sense just described, our specific practical principles.
Since a will possessing freedomL is one with the capacity to be normatively determined by pure practical reason, when we do normatively determine our will by the fundamental law of pure practical reason, we actualize freedomL. When we bring about the normative determination of the will by the moral law, we thus demonstrate that we possess freedomL. (We can, however, normatively determine our will without demonstrating freedomL. The legislation of a prudential principle is not the legislation of a law, or objective practical principle [KpV 21, 27]. Since we demonstrate freedomL only when we normatively determine our will by laws, prudential reasoning furnishes no such a demonstration.)
This interpretation might seem to portray normative determination in a voluntaristic light at odds with the familiar role assigned to Wille: the Metaphysics states that basic practical principles are not, as it were, up for grabs, but are legislated by Wille in the narrow sense (MS 213). However, in the Religion Kant grants Willkür the power to choose the principle that stands as our fundamental law (Rel 36); here he is utterly at ease with the notion of Willkür deciding the law of our will (Wille in the broad sense). Nor is the latter view in tension with the former. The Religion asserts only that Willkür determines the effective law of our will, and it leaves unscathed the moral law's status as the a priori principle of willing.
Returning to the argument at hand, the fact that the demonstration of freedomL involves regulating one's will by the moral law and bringing into existence a will normatively determined by the moral law indicates that the demonstration is a practical one. Indeed, the foregoing supports a stronger interpretive claim. If a theoretical demonstration of freedom is impossible, and if we can actualize our capacity for freedom in the legislative sense only by efficaciously representing the moral law, then the demonstration of freedom must be a practical one. To be sure, once an agent is in possession of a practical demonstration of freedom, she gains access to derivative, theoretical demonstrations of freedom in the guise of memorial representations of her legislation, acts of self-appraisal, and so on. But these demonstrations are derivative. They must refer in some way to an original practical demonstration, and they possess epistemic weight only by representing that demonstration.
Before proceeding, I want to address what might seem to be a fatal flaw in the argument on offer. One might worry that freedom is a condition of moral self-regulation since this regulation requires some independence from sensible interests. If so, then Kant's demonstration will turn out to beg the question, and my interpretation will exhibit the same defects as the orthodox version. To defuse this objection, I want to characterize more carefully the capacity at the heart of the matter, namely, the capacity that enables us to choose how to regulate our deliberation. Although Kant doesn't name the power in question, I will call it ‘practical spontaneity’. In one sense, the spontaneity of practical reason is similar to that of theoretical reason: practical reason contributes something of its own to cognition (B 158n). Practical reason contributes practical concepts, ideas, and principles that govern practical judgments (A 68/B 93; G 452). But practical spontaneity is notably different from its theoretical sibling. A practically spontaneous being is spontaneous in that she is the source of her own causality (though not to the degree required by freedom, as I clarify below). Kant defines the will as the faculty of desire ‘in accordance with concepts, insofar as the ground determining it to action lies within itself and not in its object’ (MS 213, emphasis added). What Kant means when he says that the ground determining the will to action lies ‘within itself’ is that agents are in some sense self-determining; in willing, the will (Willkür) itself chooses what it is to do. Positively speaking, the will has the power to set its own ends, to choose the goals it will pursue (MS 381, 382, 385). Negatively speaking, acts of choice are never directly caused by sensible interests or inclinations, even those choices that incorporate actions on ends ultimately grounded in sensible interests.Footnote 7 (By contrast, animals are causally necessitated by their inclinations.)
Practical spontaneity is put under a spotlight in the Religion, where Kant memorably claims that the power of choice is such that it cannot be determined to act on an inclination ‘except so far as the human being has incorporated it into his maxim’ (Rel 24). Here Kant reinforces the point just noted—neither inclinations nor the sensible interests that are based on them show up on our doorstep as reasons for action (or for adopting ends)—adding that inclinations and interests become operative in willing when an agent takes them to be reasons in virtue of a practical principle that states why the object of the interest is good (see also Allison Reference Allison1990: 97–98, 207–208; Korsgaard Reference Korsgaard1996: 163–36; Reath Reference Reath, Reath and Timmerman2010: 53).Footnote 8
However, spontaneity is not equivalent to freedom. Kant maintains that we are practically spontaneous even when we choose to act on principles that embrace sensible interests (MS 213; A803/B831; Allison Reference Allison1990: 39). When we choose to act on such principles, we are obviously not acting freely. Accordingly, nothing in our ability to bring about the normative determination of the will by the fundamental law depends on robust freedom.Footnote 9 All that is required is an understanding of what it means to reason in accordance with the moral law along with the generic ability to regulate our will by whichever practical principles we decide upon. The former is secured by a priori knowledge of the moral law, the latter by the spontaneity of the will. (Moral self-legislation is made even easier by the fact that Kant allows leeway in formulating the legislated principle; for example, living by the principle ‘what if everyone did that?’ counts as a legislation of the moral law (KpV 69–70).)
2. Freedom in Its Efficacious Aspect
To complete his demonstration of freedom, Kant must show that moral judgments can move us to act independently of sensible desire. This task is frequently overlooked by Kant's commentators, especially those affirming an efficacity condition on willing, and it is often assumed that what I call normative determination entails efficacious determination. On this view, an agent's decision to regulate her will by a moral principle automatically moves her to act on that principle. This dearth of attention is understandable, since Kant ultimately endorses the entailment; the Religion theatrically proclaims that despite our moral failings, ‘the command that we ought to become better human beings still resounds unabated in our souls; consequently, we must also be capable of it’ (Rel 45; see also Rel 50n; KpV 47). But Kant cannot simply assert that the right kind of motivational connection obtains. Further argument is needed due to a gap between normative and efficacious determination: Kant's efficacity condition opens up conceptual space for a will to be normatively and not efficaciously determined by pure practical reason. Although the efficacity condition entails that willing is efficacious, it says nothing about how practical principles regarding actions and ends are efficacious or how we are moved to actualize the objects of our choices. The efficacity condition does not entail that the will can be efficaciously determined or moved to act solely by its cognition of a maxim. Coming at this from the other direction, the efficacity condition is, by itself, compatible with the popular Humean story, according to which one can be moved to act on her principles or maxims only if she has a sensible interest in so doing. On this view, neither pure reason nor the moral law can move us to act of themselves, and this incapacity renders Kantian morality a dead letter. So a second step is needed. If Kant's demonstration of freedomL, and of freedom in toto, is to succeed, he must fill in the gap between normative and efficacious determination in the right way (Henrich [Reference Henrich and Velkley1994b: 94–95] also emphasizes this problem, but he characterizes it in different terms).
The second Critique does not foreground the gap problem, though the Groundwork mentions it at least once (G 460). Kant identifies it most clearly in lectures from the 1770s, asserting that agents are not ‘organized’ in such a way that they are automatically moved to action by ‘objective grounds’ (27: 1429). As a result, ‘[w]hen I judge by understanding that [an] action is morally good, I am still very far from doing this action which I have so judged’ (27: 1428). Here and in other lectures, Kant toys with a theory of motivation reminiscent of Hume and the British sentimentalists (27: 1429; 25: 650; 27: 258).Footnote 10 Although Kant eventually discards this strategy, it is evidence that he appreciates the difficulty and importance of accounting for the efficacity of practical reason.
The next few pages will argue that Kant bridges the gap in the requisite fashion by means of a view that I call, with apologies for the prochronism, non-Humean motivational judgment internalism (I develop this account of Kant's internalism in more detail elsewhere). Kant's internalist claim is that if P judges that she ought to φ, her judgment by itself and independently of any desire associated with the judgment generates an affective force that moves her to φ (KpV 9n, 47; MS 211, 213, 356; Rel 50n).Footnote 11 Accordingly, moral judgments can, of themselves, move an agent to act in the way specified by the judgment. Thus, given internalism, the source of the efficacity of moral judgment turns out to lie within pure practical reason.
There is something surprising about Kant's internalism. Even though he advances a staunchly cognitivist theory of rational agency, he assigns an important and ineliminable role to pleasure. Pleasure, it turns out, provides the link between cognition and action that explains how we are moved to act on both our moral and nonmoral judgments. But since the pleasure involved has its source in cognition, we need not worry that it puts Kant on treacherous Humean footing.
Kant's most important claim in this respect is that ‘every determination of choice proceeds from the representation of a possible action to the deed through the feeling of pleasure or displeasure’ (MS 399; see also MS 211; 28:894). The pleasure in question should not be understood as occurring upstream from choice. Rather, it operates downstream from choice; the determination in question is efficacious determination. Pleasure's essential role, in both the moral and nonmoral domains, is to enable us to carry out our judgments. Once choice is normatively determined, that is, once we have decided on a course of action, we take pleasure in the thought of actualizing the object of choice or performing the action. And it is this pleasure that moves us to act.
Taken in isolation, this Metaphysics passage might not seem to compel my interpretation. But additional support can be found in many of Kant's other remarks about pleasure and satisfaction (Wohlgefallen). In the second Critique he writes that ‘pleasure is the representation of the agreement of an object or of an action with. . . the faculty of the causality of a representation with respect to the actuality of its object (or with respect to the determination of the powers of the subject to action in order to produce the object)’ (KpV 9n, translation altered). When the faculty of causality is the will (Willkür), the object of pleasure is the thought of producing an object or performing an action that we have determined ourselves to produce or perform. The forward-looking thought of actualizing the object is what Kant refers to with the disjunct ‘or with respect to the determination of the powers of the subject to action’. Put more simply, Kant's view is that we take pleasure in representing the actualization of our judgments—we take pleasure in the prospect of the activity of actualization, not just of the actualized object. When I judge that I ought to do something, I take pleasure in the thought of my efficacity, even when my goal is onerous, tedious, or painful. (To my knowledge, this view first appears in a 1782 lecture [24: 891]. The third Critique restates the points made in the second [KU 207, 242; KU, First Introduction 230–31].)
It should be noted that the pleasure involved in efficacious judgment is a special kind of pleasure, which the Anthropology labels intellectual pleasure and distinguishes from sensible pleasure (A 230; see also 29:877–78, 1013; KU 222). Sensible pleasure is produced by sensibility and is the product of contingent cultural and biological forces. The Anthropology curiously lacks a description of intellectual pleasure, but we can construct one by contrast with sensible pleasure: intellectual pleasure arises from the active, spontaneous judgments of practical reason. Confirmation can be found in Kant's empirical psychology lectures (28: 586; 29: 1024), where he dubs intellectual pleasure ‘satisfaction’ (Wohlgefallen) and sensible pleasure ‘pleasure’ or ‘sense-pleasure’ (Lust) (28: 675; 29: 890). Intellectual pleasure is nevertheless a feeling bearing affective force—it is just because of this that it can move us to act.Footnote 12 The difference between intellectual and sensible pleasure lies in the origin of the pleasure. If the pleasure originates in sensibility, it is sensible; if it originates in practical reason, it is intellectual.
Kant's view, then, is that when I judge that I ought to do something, I take pleasure in the thought of bringing into existence the object of my judgment and accomplishing the task I have set for myself. As Kant puts it, ‘the state of mind of a will determined by something, however, is already in itself a feeling of pleasure and identical with it’ (KU 222). More generally, his view is that rational agents possess a disposition to take pleasure in the thought of their rational efficacy. Although he at one point dubs this a ‘peculiar’ fact (28: 586), once the details are on the table, we have what sounds like a fairly intuitive psychological and phenomenological story, so long as we accept Kant's conception of the will as fundamentally geared toward bringing its objects into existence. It is natural to think that a being who, by her very nature, strives for efficacy (KpV 9n), and strives for efficacy through her practical judgments, will feel pleasure at the thought of realizing that efficacy and bringing into existence the object of her judgments. (Kant goes so far as to say that even God acts by means of anticipatory satisfaction [28: 1061–62].Footnote 13) It is this intellectual pleasure that brings an agent to try to execute whatever project she has in mind.Footnote 14 And since this pleasure is generated by cognitive acts of choice, it is just the kind of feeling required by Kant's non-Humean motivational internalism.
With this point in hand, Kant's conception of rational agency can be schematized as follows:
Time1: Choice of maxim M on principle P
Time1 or 2: Cognition of salience of maxim M → Intellectual pleasure → Attempt to act in the way specified by M
The connection between internalism and freedomE should now be clear. Given Kant's internalism, whenever an agent makes a practical judgment, her judgment generates an affective force that enables her to act in the way specified by her judgment. As a result, the normative determination of the will by maxim M entails efficacious determination by M. No sensible interest is needed to explain action. The upshot is that whenever an agent makes a moral judgment, she generates an affective force that, independently of sensibility, moves her to pursue the moral action specified in the judgment. In so doing, she actualizes freedom in its efficacious aspect.
3. Conclusion
It is now time to pull together the various strands of the paper and run through Kant's argument in its entirety. According to Kant, we make the moral law the law of our will and actualize freedomL by regulating our will by the moral law. When this regulation involves making a moral choice, as it sometimes must, we generate a feeling that moves us to act in accordance with our moral judgments. As a result, we can do what the moral law says to do. When we make the moral law the law of our will, we thereby actualize the capacity to be moved to act by moral considerations and to act independently of all of our sensible inclinations and interests. Since actualizing freedomL entails actualizing freedomE, when we make the moral law the law of our will, we demonstrate freedom in toto.
The Critique’s demonstration is a practical one in the sense that it provides instructions for bringing about a will determined by pure practical reason and actualizing freedom. If a reader follows these instructions—that is, if she regulates her will by the moral law—she practically demonstrates freedom, in the strict sense, by actualizing it.Footnote 15 In so doing, her idea of freedom ‘also actually produces the deed that accords with it’ (Sch 13). We can at last appreciate the full significance of Kant's famous gallows example, which concerns a man ordered on pains of execution to give false testimony against his prince's political enemy. The man knows he ought to tell the truth and then judges that he can do so despite the fatal outcome; he judges that pure practical reason can be an efficient cause of his action. Kant declares that the man ‘rightly concludes’ (Rel 50n) that ‘he can do [what morality commands] because he is aware that he ought to do it’ (KpV 320). Clearly, the gallows passage illustrates a first-person, practical demonstration of freedom. More important, it invites readers, especially those who doubt their moral capacities, to undertake the demonstration and make it their own. Kant's contention is that if they do so, they will show themselves to be free.
A more formal rendition of my interpretation reveals its structural parallels to the orthodox approach:
1. Our legislation of the moral law makes the moral law the fundamental law of our will (via the self-consciousness and efficacity conditions).
2. When the moral law is the fundamental law of our will, we sometimes judge that we ought to act morally in some respect (via the efficacity condition applied to second-order principles).
3. Our awareness of the moral ‘ought’ generates an ability to act in accordance with the moral law in that respect (via motivational internalism).
4. When we legislate the moral law, we actualize our capacity for freedom.
However, it must be emphasized that my attention to the self-consciousness and efficacity conditions as well as to Kant's non-Humean motivational internalism supplies Kant with the tools needed to repel the charges of dogmatism and circularity that haunt the first two premises of the orthodox view.Footnote 16 As such, on my interpretation, Kant's demonstration can more readily achieve its fundamental aim: instilling confidence in those who acknowledge the importance of morality but worry that their sensible desires render them incapable of moral action. Kant's prospects in this regard are enhanced by a second consideration: it is not terribly difficult to determine one's will by the moral law normatively and efficaciously. This fact provides rational grounds for believing that our attempts at actualizing freedom will be successful and that we do therefore possess the requisite capacity. To be sure, as cognitively limited beings subject to psychological self-deception, we lack absolute certainty that we have legislated the moral law (G 407; KpV 47). But since Kant's argument establishes that the legislation of the moral law is within reach of any agent who acknowledges the importance of morality, such agents have good reason to believe that they are free and that their consciousness of the authority of the moral law is not illusory.
I will conclude with three points of clarification. First, one might wonder what I make of Korsgaard's rather similar claim that ‘by acting morally, we can make ourselves free’ (Korsgaard Reference Korsgaard1996: 176). I agree with a slightly weakened formulation: we make ourselves free in the sense that we demonstrate our freedom by actualizing it. But I disagree with Korsgaard's underlying argument, which relies heavily on Kantian respect. In Korsgaard's eyes, respect is supposed to explain the normative and efficacious determination of the will by pure practical reason. But respect cannot fill this explanatory role, since respect presupposes the fact that the will can be so determined. Furthermore, Kant says that while respect ‘indirectly’ determines the will (KpV 79), we are free only when the moral law immediately determines the will, i.e., when the moral law regulates a moral judgment, which in turn generates a moral motive. And it must be the case that respect only indirectly determines the will, because if it directly determined the will, or was the ground of the determination of the will by the moral law, Kant's moral theory would collapse into the moral sense theory he so strongly opposes. The same issue affects Franks's (Reference Franks2005: 295ff.) ‘performative’ interpretation of Kant's demonstration of freedom, which locates the demonstration in the production of the feeling of respect.
Second, it might seem surprising that I have said nothing about respect. I demur in part because Kant employs the concept in sundry ways, discussion of which would take us too far afield, and in part because respect, in its core sense as the affective impact of the recognition of the authority of the moral law, plays little role in the demonstration of freedom. As I noted above, respect presupposes the immediate determination of the will by the moral law, and it is the latter that must be established by the demonstration. Put differently, respect is a derivative feature of Kant's vindication of the possibility of moral motivation: the demonstration of freedom establishes our ability to act on moral principles, while respect is what enables us to cultivate virtue and make a habit of acting morally (KPV 75, 79; see also Engstrom [Reference Engstrom, Reath and Timmerman2010c]).
Finally, Kant insists that pure reason can be proved to be ‘actually practical’ through a fact, which he dubs the ‘fact of reason’ (KpV 42). Although Kant is not completely consistent in his characterization of the fact of reason, most readers agree that it refers to the consciousness of the authority of the moral law (Allison Reference Allison1990: 230, 233; Rawls Reference Rawls2000: 260; Sussman Reference Sussman2008: 66; Timmerman Reference Timmerman, Reath and Timmerman2010: 83). My interpretation is on all fours with this view, but I would stress that, connotations of ‘fact’ (Factum) aside, the consciousness of the authority of the moral law is not a theoretical cognition of a property (viz., ‘subject to the authority of the moral law’) ascribable to a rational agent, much less knowledge that an agent has this property. As Kant puts it, ‘in order to avoid misinterpretation in regarding this law as given, it must be noted carefully that it is not an empirical fact’ (KpV 31; cf. KpV 48). From what has been said so far, it should be evident that the consciousness Kant dubs the fact of reason is the self-consciousness that corresponds to the act of making the moral law the law of one's will; it is the self-conscious representational component involved in actualizing one's freedom. It is for this reason that Kant can insist that pure reason proves itself to be practical through a consciousness that is also a deed (KpV 3).
