Eidos. Revista de Filosofía de la Universidad del Norte

ISSN 2011-7477
n.° XIV, enero-junio de 2011
Fecha de recepción: septiembre 14 de 2010
Fecha de aceptación: diciembre 10 de 2010


The practicality of pure reason.
A normative defence of Kant ' s theory of moral motivation

El carácter práctico de la razón pura.
una defensa normativa de la teoría kantiana de la motivación moral

Triantafyllos Gkouvas
University of Antwerp, Faculteit Rechten, Centre for Law and Cosmopolitan Values


El propósito de este trabajo es defender la opinión que Kant ha propuesto sobre la teoría internalista de la motivación moral. En particular, argumentaré que la adopción de Kant de internalismo se evidencia en su afirmación de que la relación de la razón pura de la voluntad se basa en una práctica una proposición sintética a priori. Lo que se pretende demostrar es que Kant trata la sinteticidad práctica como un concepto fundamental para su relato de lo que significa ser motivados por los principios de la razón pura. En mi conceptualización de la teoría motivacional de Kant, la relación entre las máximas universalizables y el interés moral de actuar sobre ellos es necesaria pero no tautológica, ya que violaciones de los derechos son lógicamente posibles a pesar de que tengamos una razón moral de actuar. Lo que impide que este último argumento colapse en una cuenta casi externalista de la motivación moral es que el impacto de motivación de las máximas como la ley, es en última instancia, la premisa de una concepción normativa de nosotros mismos como agentes libres.

Palabras clave

Motivación moral, juicio moral, razón pura, voluntad, autonomía.


The purpose of this paper is to defend the view that Kant has propounded an internalist theory of moral motivation. In particular, I shall argue that Kant's espousal of internalism is evidenced by his claim that pure reason's relation to the will is premised on a practical synthetic a priori proposition. What I aim to demonstrate is that Kant treated practical syntheticity as a pivotal concept for his account of what it means to be motivated by principles of pure reason. On my construal of Kant's motivational theory, the relation between universalizable maxims and the moral interest to act upon them is necessary but non-tautological, since violations of duty are logically possible despite our having a moral reason to act. What prevents the latter argument from collapsing into a quasi-externalist account of moral motivation is that the motivational impact of law-like maxims is ultimately premised on a normative conception of ourselves as free agents.


Moral motivation, moral judgment, pure reason, will, autonomy.

A note on the text

For convenience I cite Kant's works infratextually in parentheses. Citations from Kant's works will be located by volume and page number of the Academy Edition, Kant's gesammelte Schriften, edited by the Royal Prussian (successively the German and then Berlin-Brandenburg) Academy of Sciences (Berlin: Georg Reimer, later Walter de Gruyter & Co., 1900-). References to the Critique of Pure Reason will be located in the traditional manner by the pagination of its first ("A") and second ("B") editions.

Translations from the Critique of Pure Reason are from Immanuel Kant: Critique of Pure Reason: A revised and expanded translation based on Meiklejohn, ed. by Vasilis Politis, London: J. M. Dent, Everyman, 1997 (4th edn.). Translations from the Critique of Practical Reason are from Immanuel Kant: Critique of Practical Reason, trans. by Thomas Kingsmill Abbott, Dover Publications, Inc., 2004 (unabridged republication of the 1954 reprint by Longmans, Green and Co., London and New York, of the work originally published in 1909); those of Religion are from Immanuel Kant: Religion within the Limits of Reason Alone, trans. by Theodore M. Green and Hoyt H. Hudson, HarperOne 1960; those from the Groundwork of the Metaphysics of Morals from Immanuel Kant: Groundwork of the Metaphysics of Morals, trans. By H. J. Paton, New York: Harper & Row Publishers, Inc., 1964; and finally those of The Metaphysics of Morals from Immanuel Kant: The Metaphysics of Morals, trans. with an introduction and notes by Mary Gregor, Cambridge: Cambridge University Press, 1991. The abbreviations used throughout this essay are as follows:

KrV = Critique of Pure Reason

GMS = Groundwork of the Metaphysics of Morals

KpV = Critique of Practical Reason

R = Religion within the Limits of Reason Alone

MM: The Metaphysics of Morals

Internalism and the Kantian Paradigm

Internalism can accommodate profoundly opposing first-order moral views. Internalists primarily hold that there exists a necessary connection between normativity and motivation. I deliberately frame the internalist argument in very broad terms

in order to bracket certain difficult questions that could not be explored adequately in the space of this interpretive essay, among them, the question of what is the precise form1 in which normativity establishes its relation to motivation. My aim in this paper is to provide a cogent account of how Kant responds to a more fundamental question, that is, in what way normativity is necessarily connected with motivation. To illustrate my point I take the next two exemplary propositions P1 and P2:

P1: If A has a reason to φ, that is because she could be motivated to φ, and P2: If A has a reason to φ, then, because of this, she could be motivated to φ.

Both propositions fall within the internalist class because both accept that in some way to be further specified moral judgments are necessarily connected with a certain motivational process. Nevertheless, they depart in opposite directions with regard to the prioritization of the related terms. In the first example, the moral reasons an agent is presented with are dependent on the psychological capacity to arrive at a desire to act accordingly, whereas, in the second case, a moral demand becomes the ground of the ensuing motivation. In either case there is an obvious asymmetry2 in the relation between normativity and motivation and what actually distinguishes the two examples is the choice to bestow primacy on either term.

As I shall try to demonstrate, Kant has espoused a type of internalism akin to the second proposition. Showing that Kant acknowledges the primacy of pure practical reason over motivation will serve as a basis for explicating his understanding of the precise way in which they are necessarily related. To begin with, I shall provide a necessary textual background for filling in the details of my approach to Kant's motivational theory. The Kantian corpus abounds in references suggestive of an internalist conception of moral agency. More specifically, Kant's grasp of internalism standardly takes the form of the claim that "the objective principle of determination must always and alone [emphasis added] be also the subjectively sufficient determining principle of the action" (KpV 05:71). From this short passage we immediately infer that the objectively binding moral law always, i.e. necessarily (1), suffices alone, i.e. prior to any conative state (2), to motivate the will. Kant's claim about the primacy of pure practical reason over the agent's motivational capacities is even more lucidly depicted in his Groundwork remark that "this much only is certain: the law is not valid for us because it interests us... the law interests us because it is valid for us [emphasis added] men in virtue of having sprung from our will as intelligence and so from our proper self" (GMS 04: 460). In the Metaphysics of Morals, Kant's internalist argument is further refined by means of the distinction between what he calls ethical (ethische) or internal (innere) and rightful (rechtliche), juridical (juridische) or external (ãuftere) lawgiving (Gesetzgebung). Although both kinds of lawgiving comprise a law "which represents an action that is to be done as objectively necessary" (MM 06: 218) and an incentive "which connects a ground for determining choice to this action subjectively" (MM 06: loc. cit.), only in the case of moral lawgiving does the law make duty the incentive, whereas a lawgiving which does not include the incentive of duty itself is juridical. Moral internalism or Moralität consists in "that conformity in which the Idea of duty arising from the law is also the incentive to the action" (MM 06:219) whereas juridical externalism or Legalität amounts to "mere conformity or nonconformity of an action with law, irrespective of the incentive to it [emphasis added]" (MM 06: loc. cit.).

Disambiguation of Some Key Terms

In the rest of this paper, I shall capitalize on the textual evidence to point out that Kant's particular way of referring to the relation of pure reason to the will is expressed through his claim that the motivational grip of the moral law "forces itself on us as a synthetic a priori proposition, which is not based on any intuition, either pure or empirical" (KpV 05:31). I shall term this type of internalism practical syntheticity although I will interchangeably refer to it as autonomous motivation. What I aim is to show is that pure reason's synthetic relation to the will preserves the necessary link between normativity and motivation while safeguarding the categoricity of moral rules. Before embarking on this task, a few additional Kantian jargon needs to be spelled out so as to facilitate the course of the argument.

First, it should be noted that Kant employs a variety of terms signifying the objective necessity of the moral law (moralisches Gesetz). Terms such as mere legislative form of maxims (bloß gesetzgebende Form der Maximen, for similar formulations see KpV 05:27, 29, 31, 34), practical law (praktisches Gesetz) and objective determininggroundof the will(objectiver Bestimmungsgrund des Willens, KpV 05:71, 81 and GMS 04:449) are alternative ways of expressing his central idea that pure reason is capable of determining the will by "the mere Idea that a maxim qualifies for the universality of a practical law" (MM 06:225). For Kant moral oughts are rational requirements that every agent could adopt as universally binding.

Nevertheless, a law-like maxim is not motivationally efficient solely in virtue of our grasping its meaning. What is further required is an ensuing moral interest (moralisches Interesse) which "is found only where there is a dependent will which in itself is not always in accord with reason [emphasis added]" (GMS 04:413n). Kant frequently refers to this moral interest as a subjective determining ground of the will (subjectiver Bestimmungsgrund des Willens). Both terms interchangeably denote the motivational impact of pure reason on the agent as evidenced by the fact that the moral interest and the subjective determining ground alike are analyzed in terms of "a motive to obedience to the law" (KpV 05:79). The term I will be employing in the rest of this paper will be that of moral interest since it fits better to the contemporary talk of moral motivation.

Given that Kant attributes such an interest only to the will of finite agents defining it as "a motive [Triebfeder] of the will in so far as it is conceived by the reason [emphasis added]" (KpV 05:79), the moral interest cannot refer to a pre-existing disposition to behave morally, but presupposes the agent's engagement in a

Given that Kant attributes such an interest only to the will of finite agents defining it as "a motive [Triebfeder] of the will in so far as it is conceived by the reason [emphasis added]" (KpV 05:79), the moral interest cannot refer to a pre-existing disposition to behave morally, but presupposes the agent's engagement in a rational appraisal of her maxim's potential as a universal norm of conduct. Moreover, it is important to point out that the notion of moral interest does not pertain to an unexceptionally law-abiding will for the reason that, as I shall try to show in detail below, Kant treats the relation of motivation to pure reason as synthetic and hence as between two semantically non-identical terms. Should the law refer to an agent possessed of a perfectly good will (vollkommen guter Wille, GMS 04:414), internalism would turn out to be an analytic claim since in such a case the very concept of a perfect (vollkommen) or holy (heilig) will would contain in itself the notion of the capacity to always act out of reverence for the moral law, or as Kant puts it, " 'I ought' is here out of place, because 'I will' is already of itself necessarily in harmony with the law" (GMS 04:414, see also 04:455). As a result, the necessity of the relation between a perfect will and the moral law would be conceptual, not synthetic. The internalist requirement pertaining to the motivation of finite agents presupposes that the moral law is vested with its imperatival 'ought' (Sollen) formulation which marks "the relation of an objective law of reason to a will which is not necessarily determined by this law in virtue of its subjective constitution [subjective Beschaffenheit, emphasis added] (the relation of necessitation)" (GMS 04:413). It is precisely this relation of necessitation (Nöthigung) that Kant characterizes as practical synthetic a priori and it is precisely this claim that will serve as the core of my analysis.

The Motivational Relevance
of Respect for the Moral Law

I have deliberately excluded from the preceding analysis the ardently debated concept of respect for the moral law (Achtung fürs moralische Gesetz). What I aim to demonstrate by examining this concept separately is that the distinctness of its motivational role has been mistakenly overemphasized. Indeed, there has been a lot of controversy as to whether respect for the moral law denotes a conative state or, conversely, signifies an intellectual appraisal of the action-guidingness of the moral law. As I will try to show, both assumptions are fallacious insofar as they insist on interpreting the concept of respect independently of its relation to Kant's internalist thesis. What I propose instead is a normative reading that treats respect as synonymous with the moral interest to act in accordance with pure practical laws.

It is indeed perplexing how Kant oscillates between respect as "consciousness of immediate obligation of the will by the law" (KpV 05:117) and respect as a moral feeling (moralisches Gefühl) cognizable a priori. Nonetheless, the mutual exclusiveness of these two employments can be shown to be a nonexistent dilemma. Kant gives a clearer view of his thought when he mentions that human beings are often under "the illusion that regards the subjective role of this capacity of intellectual determination as something sensible and the effect of a special sensible feeling (for an intellectual feeling would be a contradiction)" KpV 05: 117, cf. GMS 04:460). Kant terms this fallacious perception vitium subreptionis, namely "as it were [of] an optical illusion, in the self-consciousness of what one does as distinguished from what one feels -an illusion which even the most experienced cannot altogether avoid" (KpV 05:116). The importance of this remark lies in the distinction it draws between the ground of the subjective determination of the will and the way the latter is perceived by the agent. For Kant rational action is not an embodiment of feelings, but of practical laws which simultaneously describe and prescribe a type of action. The moral feeling is always an effect of an already consummated determination of the will and thus it falls beyond the scope of Kant's internalism. The positive feeling resulting from a morally motivated will retains a descriptive function explicating "what effect it [the moral law], as such, produces (or, more correctly speaking, must produce) on the mind" (KpV 05:72). On the contrary, respect as the interest arising from an autonomous will is a normative concept denoting the "assessment of a worth [emphasis added] which far outweighs all the worth of what is commended by inclination" (GMS 04:403, see also 04:402 n2).

My argument is not meant to compromise the motivational function of respect. All I aim to show is that the account of respect as the crux of Kant's theory of moral motivation needs to be cautiously qualified. Put more succinctly, the motivational relevance of respect should not be thought of as distinct from the concept of moral interest as explicated in the previous section. Kant treats these concepts co-extensively as evidenced by his claim that "all moral interest, so-called, consists solely in reverence for the law" (GMS 04:401n2, see also KpV 05:81). Consequently, what I propose is the embedding of the concept of respect qua moral interest in the internalist structure. This is the only way to account for the normative character of this concept which denotes a practical stance of the agent towards her freedom or, as Kant puts it, it is "purely practical and free [emphasis added]... not suggested by any inclination, but [is] commanded and actually brought about by reason through the practical law" (KpV 05:81). The idea of autonomous motivation entails the deliberative choice to act on universalizable maxims. It is implausible to assume that a moral feeling or a benign disposition could accommodate Kant's strong claim about the free will. For Kant, choosing to act upon a universally adoptable maxim just is an exercise of one's autonomy and hence of one's positive freedom. Moral feeling does not mediate between pure reason and the will's determination and thus cannot be the ground of autonomous acts. This is precisely the point that Kant wants to make when he claims that "pure reason must of itself be practical; that, is, it must be able to determine the will by the mere form of the practical rule without supposing any feeling [emphasis added]" (KpV 05:24).

The Formulations of Practical Syntheticity

The preceding remarks on the primacy of pure reason over motivation and the role of respect are aimed at bringing to the fore the conceptual tools Kant has used to expound his motivational theory. This might facilitate the understanding of his particular way of using the concept of practical syntheticity as a means to defend his internalist thesis3. An early formulation of his proposition is found in the second chapter of the Groundwork where the will's necessitation by the moral law is said to be a "practical proposition in which the willing of an action is not derived analytically from some other willing already presupposed (for we do not possess any such perfect will), but is on the contrary connected immediately with the concept of the will of a rational being as something which is not contained in this concept" (GMS 04:420 n1, see also 04:440). According to this formulation, the willing of an action (das Wollen einer Handlung) refers to the agent's interest in fulfilling her moral duty. That interest is not contained in the concept of autonomy of the will which is the source of practical laws enjoining action. In the Religion within the Limits of Reason Alone, the same argument is refashioned in terms of the relation between the predisposition to humanity (Anlage für die Menschheit) and the predisposition to personality (Anlage für seine Persõnlichkeit). Kant claims that the predisposition to humanity which is equated with man's capacity to employ practical reason instrumentally does not contain in itself the predisposition to personality, namely "the capacity for respect for the moral law as in itself a sufficient incentive of the will' (R 06:27). The argument adduced in support of this non-analytical relation is based on the idea that "from the fact that a being has reason it by no means follows that this reason, by the mere representing of the fitness of its maxims to be laid down as universal laws, is thereby rendered capable of determining the will unconditionally, so as to be practical of itself; at least, not so far as we can see" (R 06:26n). The predisposition to humanity "is based on practical reason, but a reason thereby subservient to other incentives" (R 06:28) while the predisposition to personality "alone is rooted in reason which is practical of itself, that is, reason which dictates laws unconditionally" (R 06:loc. cit.). It is precisely the possibility of desire-based actions that necessitates the semantic difference between pure reason and the will's determination. Hence, our having a universally adoptable maxim to φ does not mean by itself that we are motivated to φ Inclination might still impede action or assume the role of the will's motive. As I shall argue in the last two sections of this essay, the argument from practical syntheticity is aimed at demonstrating that the lack of a conceptually necessary link between pure reason and motivation cannot tell against the possibility of an a priori necessity4 grounded in the normative demand to regard ourselves as free agents.

The Fact of Reason as Consciousness
of the Possibility of Autonomous Motivation

Kant does not content himself simply with stipulating the synthetic relation between pure practical reason and the ensuing moral interest. The mere assertion of their semantic difference does not entail anything at all about the possibility of autonomous motivation. This task is mainly carried out in the Second Critique where Kant highlights the importance of the distinction between theoretical and practical synthetic a priori necessity. It is only the latter kind of necessity that is deemed capable of attaining a vindication of the practicality of pure reason. Although Kant does not explicitly reveal his intention to vindicate practical a priori syntheticity -as he does in the First Critique with regard to theoretical a priori syntheticity-, he couches it under the notion of the Fact of Reason (Factum der Vernunft). Indeed, we can legitimately infer that the Fact of Reason just is consciousness (not knowledge) of the possibility of practical syntheticity or equivalently of the possibility of autonomous motivation. This connection is perspicuously reflected in Kant's remark that "it is at least not impossible to conceive that a law, which only applies to the subjective form of principles, yet serves as a principle of determination by means of the objective form of law in general. We may call the consciousness of this fundamental law a fact of reason" (KpV 05:31). As clearly stated in this passage, the synthetic concurrence of the subjective and objective determination of the will is considered to be the content of every rational being's consciousness.

Kant's justificatory strategy, however, is not fully deployed merely on the assumption that moral agents are conscious of the motivational efficacy of pure reason. Such an argument would trivially entrust pure practical reason's vindication in a metaphysically inscrutable power of moral insight. Therefore, the notion of the Fact of Reason should not be construed as a deductive proof of pure reason's motivational impact and hence it should not be thought of as playing a distinctively justificatory role. All that is claimed through this Fact is that the possibility of autonomous motivation is actually (consciously) recognizable by every rational being. Everyone possessed of a rational will can attest that autonomous motivation manifests itself in fact and deed, being traceable in our everyday moral reasoning, feeling and judgment. Scholarly attempts to confer upon this Fact a distinct vindicatory role reasonably attract scathing comments such as that Kant had obliquely introduced a dogmatic metaphysics into his ethical theory. There is nothing, however, metaphysically factual behind this concept. On the contrary, what is suggested is that rational agents are actually conscious of their capacity to make (possunt facere) practical laws out of their subjective maxims or, in Kant's own words, we become directly conscious of the moral law's motivational grip "as soon as we trace [entwerfen] for ourselves maxims of the will" (KpV 05:29). This immediate consciousness which is synonymous with the Fact of Reason results in the construction6 of universally adoptable maxims which can motivate our actions. There is no deductive allusion whatsoever with regard to the function of the Fact of Reason. In the following section, I shall further expand on the semantic (not metaphysical) contribution of the Fact of Reason with respect to the understanding of freedom as the ground of autonomous motivation.

Freedom as the Ground of Autonomous Motivation

Granted that the notion of Fact of Reason is not assigned with the task of vindicating pure reason's motivational capacity, I shall examine Kant's further claim that the possibility of autonomous motivation does not require to be deduced from metaphysically secured premises7. All that is needed to legitimately affirm this possibility is to ensure whether we are entitled to acknowledge the real possibility of freedom for practical purposes.

As early as in the Transcendental Doctrine of Method, Kant correlates the practical use of pure reason to freedom, when he declares "I term all that is possible through free will, practical" (KrV A798/B826). In the Foreword of the Second Critique, the motivational capacity (praktisches Vermögen) of pure reason and freedom are presented as deeply interwoven concepts. Kant firmly asserts that "inasmuch as [emphasis added] the reality of the concept of freedom is proved by an apodeictic law of practical reason, it is the keystone of the whole system of pure reason"(KpV 05:03).

In order to gain further insight into Kant's understanding of freedom as the ground of autonomous motivation, I will begin by explicating the distinctive mark of practical as opposed to theoretical syntheticity. With respect to the theoretical employment of reason, says Kant, "if we are to form a synthetical judgment regarding a concept, we must go beyond it, to the intuition in which it is given" (A720/B748). By sharp contrast with theoretical syntheticity, the practical synthetic a priori relation of pure reason to the will is characterized as totally intuition-independent (see KpV 05:31, 05:99). This intuition-independence is twofold since it refers both to sensible (sinnliche) and intellectual intuitions (intellectuelle Anschauungen). These two kinds of intuition are to be strictly distinguished as Kant permits use of the former as a means of knowing objects qua appearances whereas he explicitly denies any possible use of the latter by finite cognizers either in theoretical or in practical endeavors.

Before explicating the importance of intuition-independence in relation to the grounding of autonomous motivation in freedom, I shall briefly sketch the semantic content of sensible as well as of intellectual intuitions. Kant generally defines intuition as the representational means through which cognition immediately refers to objects (see KrV A19/B33). Sensible intuitions, in particular, either pure or empirical, are necessarily related to the sensory experiences of a finite cognizer (see KrV A50/B74) in the sense that the intuiting subject cannot attain knowledge of an object merely by thinking of it, but also needs to be affected in some way by its existence (Dasein). In that sense, sensible intuitions provide knowledge of things only as appearances (Erscheinungen) palpable by human sensibility but not as things in themselves (Dinge in sich). On the contrary, Kant conceives the idea of an intellectual intuition as belonging only to a divine cognizer who "gives in itself the existence of the object of the intuition" (B70). A supernatural being possessed of the faculty of intellectual intuition would be capable of knowing noumenal objects by her very act of conceptualizing them. Kant refers to a non-sensible yet intellectually graspable object as noumenon in the positive sense (Noumenon in positiver Bedeutung) and contrasts it with a noumenon in the negative sense (Noumenon im negativen Verstande) which is exactly the same thing viewed from the perspective of a finite cognizer (see KrV B306).

Keeping track of the above definitional scheme, I shall try to clarify the bearing of intuition-independence on the possibility of autonomous motivation. To highlight my argument, I shall utilize a distinction Kant makes to illustrate his point about what the two a priori syntheticities refer to. As he remarks, whereas theoretical syntheticity pertains to the relation of pure understanding (reiner Verstand) to objects, practical syntheticity refers to the relation of pure reason (reine Vernunft) to the will (see KpV 05:55). In the former case it is sensible intuitions that allow us to advance beyond the intension of a subject-concept and establish a necessary yet non-tautological relation to another predicate-concept. In the latter case, it is not intuition but freedom that is required so as to ground the connection of pure reason to the will of a finite agent. This idea is clearly portrayed in Kant's statement that "instead of intuition [emphasis added] it [the critique of practical reason] takes as their [of pure practical laws] foundation the conception of their existence in the intelligible world, namely, the concept of freedom [emphasis added]" (KpV 05:46).

In this dense passage Kant glides smoothly from intuition-independence to freedom despite the fact that his background assumptions are much more intricate. Therefore, I shall try to recompose his argument in a chain of premises followed by an analysis of how intuition-independence is specifically applied to moral motivation. I have already shown that Kant's internalist thesis consists in the claim that pure reason can immediately determine the will, the latter being "a kind of causality belonging to living beings so far as they are rational" (GMS 04:446). The immediateness of this determination further implies that a will qua causality can function independently of determination by alien causes, such as desires or extra-mental properties. It is in that sense that Kant refers to the free will as causa noumenon. Consequently, the very concept of a will unimpeded by alien influences "contains that of a causality accompanied with freedom, that is, one which is not determined by physical laws, and consequently is not capable of any empirical intuition in proof of its reality" (KpV 05:55, see also GMS 04:446).

Given the above premises we are in a better position to estimate the weight of Kant's claim that the relation of pure reason to the will is intuition-independent and hence freedom-dependent. As I have already stressed, intuitions can be thought of as either sensible (pure or empirical) or intellectual. As a result, the very concept of a free will immediately determinable by pure reason alone precludes both kinds of intuitional mediation. First, with respect to sensible intuitions, it precludes instrumental conceptions of moral agency which treat reason as capable of motivating the will only by means of the motivating influence of subjective ends such as individual preferences or desires. Second, in relation to intellectual intuitions, it precludes a Platonist model of moral motivation according to which merely thinking of a supersensible, extra-mental property of goodness is considered to be sufficient to move an agent to act. In the case of autonomous motivation this supersensible property or, as Kant would term it, this noumenon in a positive sense would be the order of things ruled by the law of freedom 8 . For Kant, “the possibility of such a supersensible system of nature, the conception of which can also be the ground of its reality through our own free will, does not require any a priori intuition (of an intelligible world) which, being in this case supersensible, would be impossible for us [emphasis added]” (KpV 05:45). If a finite agent could gain insight into her noumenal character, she could have access to the noumenal mechanics of autonomous motivation, that is, she “should perceive that this whole chain of appearances in regard to all that concerns the moral laws depends on the spontaneity of the subject as a thing in itself, of the determination of which no physical explanation can be given” (KpV 05:99).

Nevertheless, the vindication of autonomous motivation is not attained merely by seeking recourse to the intuition-independence of pure reason's relation to the will. Our farthest progress has not yet reached the point of finding the ground beneath the conceptual claim that a free will motivated by pure reason purports to preclude the mediation of intuition. It remains unclear how we can legitimately assume that instead of intuition it is the concept of freedom that really is the ground of autonomous motivation. Kant terms freedom an idea or concept of pure reason (Idee, Begriff der reinen Vernunft) to which no corresponding object can be discovered in sense-experience. As such it can only assume the role of a regulative principle of speculative reason thus limiting its employment to "containing ideas for the guidance of the empirical exercise of reason" (A663/B691). All that theoretical reason is legitimized to do is to refer to freedom as merely logically possible but it totally fails with regard to ascertaining its objective reality. Given Kant's firm position that finite cognizers cannot acquire positive knowledge of what freedom is really about and hence of how moral motivation actually works, on what grounds are we entitled to insist on treating it as the founding idea of pure reason's motivational capacity? I believe that, in grappling with this issue, three distinct yet closely connected questions need to be answered, namely what exactly is to be justified by the concept of freedom, how are we to understand freedom in relation to what it grounds and what entitles us to use freedom as a justificatory concept.

First, it is imperative to clarify that it is pure reason's relation to the will (practical syntheticity or autonomous motivation) not the moral law per se that calls for the employment of freedom as a means to establish its validity. The principle of morality viewed independently of its motivating effect is an idea of reason that, as Kant strongly supports, is by its nature not amenable to any kind of deduction (see especially KpV 05:47 and 05:105). Put more succinctly, trying to ground one concept of reason (the moral law) on another concept of reason (freedom) is a circular endeavour that yields nothing but conceptual oversophistication. Freedom assumes a vindicatory role as soon as we accept that pure reason can guide the will independently of any intuition. This point is lucidly depicted in Kant's remark that the moral law per se "is indeed quite independent of these suppositions [i.e. the ideas of God, freedom and immortality] and is of itself apodeictically certain...and so far it needs no further support by theoretical views as to the inner constitution of things" (KpV 05:142 in fine, 143). Conversely, freedom is called upon to attest the reality of "the subjective effect of this law, namely, the mental disposition conformed to it and made necessary by it [the moral law per se]" (KpV 05:143).

Granted that freedom is taken to be the ground of the moral law's 'subjective effect', we still need guidance as to how to understand a concept to which theoretical reason has not furnished a meaning graspable by human intellect. It is precisely with respect to this question that the actuality of autonomous motivation -consciousness of which is called a Fact of Reason-confers a positive meaning on the concept of freedom. Kant analyzes the positive definition (positive Bestimmung) of the concept of a free will as referring to "the notion of a reason that directly determines the will (by imposing on its maxims the condition of a universal legislative form)" (KpV 05:48, see also MM 06:214).

One again I caution against deductive employments of the Fact of Reason. In close connection with my earlier remarks, consciousness of the fact that the world of sense is replete with actual examples of principle-based (as opposed to desire-based) action is not to be treated as the truth-maker either of freedom or of pure reason's motivational impact. On the contrary, the contribution of this Fact consists in disambiguating the employment of the concept of freedom solely with respect to moral motivation. By treating the moral law as the law of free agency, Kant wants to endow freedom with a merely practical meaning (lediglich praktische Bedeutung) "inasmuch as the idea of the law of causality (of the will) has self causality, or is its determining principle" (KpV 05:50). He explicitly denies any further ontological commitment with regard to the employment of freedom noting that pure practical reason "employs the notion of cause, not in order to know objects, but to determine causality in relation to objects in general" (KpV 05:49).

How precisely does the actuality of autonomous motivation suffice to ascribe a positive meaning to the concept of freedom despite the fact that Kant explicitly denies the possibility of establishing its reference to intuitable objects? In other words, how are we to understand Kant's talk of the objective practical reality (objective praktische Realität, KpV 05:56) of freedom as opposed to the unfruitful attempts to prove its objective theoretical reality (objective theoretische Realitát, KpV 05: loc. cit.)? The answer lies in what Kant believes to be the fundamental distinctive mark of the practical vis-à-vis the theoretical use of pure reason. Whereas in theoretical reason concepts must be established by mediation of intuition in order to acquire objective reality and hence "the objects must be the causes of the ideas which determine the will" (KpV 05:44), in practical reason "the will is the cause of the objects; so that its causality has its determining principle solely in the pure faculty of reason" (KpV 05: loc. cit.). Explorations into what we are justified in believing are causally dependent on the representation of things as they are. By sharp contrast, practical deliberation is primarily based on principles, that is, on rational considerations concerning what ought to be done.

Pure practical reason categorically demands actions "which nevertheless have not taken place, and which perhaps never will take place" (KrV A547/B575, see also GMS 04:408). It is precisely Kant's distinction between intuition-based knowledge and principle-based action that underwrites his claim according to which "the practical a priori principles in relation to the supreme principle of freedom are at once cognitions, and have not to wait for intuitions in order to acquire significance, and that for this remarkable reason, because they themselves produce the reality of that to which they refer (the intention of the will) [emphasis added], which is not the case with theoretical concepts" (KpV 05:66, cf. also ibid. 44,55 and 56 in fine). The moral determination of the will is not an instantiation of a pre-existing desire or a Platonist conception of the good. Principle-based actions are good because of the way that they are willed -that is, because the reasons for which they are performed qualify to be a universal law- not because they are responsive to independent good-making properties. Provided that "the concept of good and evil must not be determined before the moral law...but only after it and by means of it" (KpV 05:63), the reality of moral interest cannot be corresponding to pre-existing 'good objects' realized through action but to "the manner of acting, the maxim of the will" (KpV 05:60) and hence to "the form of a pure will, which is given in reason" (KpV 05:66, see also 05:70). But if the moral interest produced by pure practical reason refers to a particular way of willing (not to instantiations of goodness) it follows that the practical meaning ascribed to freedom merely refers to the way or law of directing the causa noumenon in the sensible world. That is to say, autonomous acts as empirical tokens of a good will should not be regarded as an intuitional substitute for the supersensible object of the free will and hence of the concept of the unconditionally good. The free will acquires a practical meaning in the sense that pure practical reason furnishes its supreme law which amounts to nothing more but the way that a free will should be employed even though its noumenal reference remains unknown. By defining "the manner [Art] in which, as such, it can be active [emphasis added]" (KpV 05:106) pure reason allows us to employ the concept of freedom which "enables us to find the unconditioned and intelligible for the conditioned and sensible without going out of ourselves [emphasis added]" (KpV 05:105 in fine). We don't need to stretch our intuitive capacities beyond the limits of empirical knowledge in order to understand how freedom of the will manifests itself in the practical realm because all we need to know -and of that we are actually conscious in virtue of the Fact of Reason- is the law or way of willing an action. We don't need also to become cognizant of what freedom is really about and hence how noumenal causation is really possible. Kant explicitly limits the semantic contribution of the Fact of Reason to providing the law but not the object and the mechanics of the causa noumenon when he remarks that consciousness of the moral law as a sufficient motive of the will "can transfer the determining principle of the will into the intelligible order of things, admitting, at the same time, that we cannot understand how the notion of a cause can determine the knowledge of these things" (KpV 05:49). Freedom becomes practically intelligible insofar as we choose to be motivated by universally adoptable maxims. Hence freedom's practical reality solely consists in the specific way of its employment by moral agents as it "is exhibited in concreto in intentions or maxims"(KpV 05:56).

However, even if we are willing to concede that freedom purports to be the ground of autonomous motivation and that freedom as the ground of autonomous motivation is to be meant as autonomy of the will, we still encounter the question of what legitimizes our acceptance of freedom as a real ground in the first place. Admittedly, intuition-independence which was presented as the distinctive mark of the concept of practical syntheticity purports to require a conception of freedom as its foundation. Furthermore, in virtue of the nature of practical reasoning we could at least initially assume that the actuality of autonomous agency endows the free will with a law governing its practical manifestation although without any further metaphysical implications. Nevertheless, neither intuition-independence nor autonomous motivation as a Fact of Reason can justify Kant's choice to employ freedom as the real ground of pure reason's motivational capacity. In the last part of this lengthy section I shall try to show that what legitimizes Kant's move is his employment of freedom as a postulate of pure practical reason (Postulat der reinen praktischen Vernunft).

Whereas freedom was taken to be the ground of pure reason's relation to the will, the postulate that freedom really is the supreme condition of autonomous motivation is not grounded in the capacity of pure reason to motivate but solely in the supreme principle of pure practical reason which demands the determination of the will "by the mere universal legislative form of which its maxim must be capable" (KpV 05:33). This is a crucial point in Kant's justificatory strategy which calls for more attention. Suppose that Kant contented himself with stipulating that freedom is the ground of autonomous motivation because the actuality of moral interest bears testimony to the fact that pure reason furnishes the law directing the free will in the practical domain. But then, as Kant alleges, the law of a free will just is the moral law. Therefore we conclude that a free will as it manifests itself in experience and a will motivated by law-like maxims are synonymous concepts because the law that governs their practical employment is one and the same. The preceding chain of premises can justifiably raise the objection that Kant is trivially reducing the law of a free causality to the moral law. Convincingly establishing this reduction presupposes that not only the law but also the object it applies to is one and the same. However, the morally or unconditionally good (das sittlich / unbedingt Gute KpV 05:68 and 05:69 respectively) as the object of pure practical reason as well as the Idea of "a natural system not given in experience, and yet possible through freedom" (KpV 05:44) are concepts to which no corresponding intuition can be given (see KpV 05:68-69). Hence, even by Kant's own lights, we cannot help but regard as problematic the transition from the conceptual level of understanding the law of a free will as the moral law to the substantive level of claiming that the moral law is the law of a free will. Insofar as we are unable to epistemically ascertain the identity relation between the free will and the autonomous or moral will, we can never be sure that the way we understand freedom is also the way that makes it true.

Nonetheless I shall suggest that Kant has never attempted this transition in the first place. To be more specific, I believe that there is a promising way to rescue Kant's argument from trivialization as soon as it becomes clear that the reason why freedom is understood as the ground of autonomous motivation and the reason why freedom is really taken to be the ground of autonomous motivation are not identical. The first reason is provided by the Fact of Reason which, as already noted, suggests that actually willing an action for universally adoptable reasons evidentially warrants our understanding of how to employ freedom on the empirical plane even though its noumenal reference is beyond our grasp. The second reason suggests that the only means of attesting freedom's reality is to postulate its existence as a necessary condition of obeying the moral law. Alternatively stated, the Fact of Reason makes freedom meaningful as the ground of autonomous motivation whereas the postulation of freedom in virtue of the supreme moral principle makes freedom practically true, that is, a valid ground of pure reason's motivational capacity. Kant alludes to the distinction between the semantic function of the Fact of Reason and the justificatory or normative function of the postulate of pure practical reason when he remarks that the concept of freedom is rendered meaningful "by means of [its] reference to what is practical [emphasis added]" (KpV 05:132, see also 05:50, 56), whereas the postulate of freedom gives "a right to concepts [emphasis added], the possibility of which it [theoretical reason] could not otherwise venture to affirm" (KpV 05: loc. cit).

The above remarks provide strong support for the claim that Kant's vindication of pure reason's motivational capacity is normative all the way down. Neither a metaphysical assumption based on the merely logical undeniability of freedom nor the conceptual claim that the law of freedom purports to be the moral law can qualify for giving sufficient credit to the possibility of autonomous motivation. On the contrary, the fact that it is practically necessary that pure reason motivate the will is ultimately grounded in what is taken to be "a need which has the force of law to assume something without which that cannot be which we must inevitably set before us as the aim of our action" (KpV 05:5). Kant believes that insofar as speculative reason has shown that freedom is at least logically possible, pure practical reason has the right to extend itself beyond this logical possibility and demand freedom's reality as the necessary and sufficient condition of autonomous motivation.

However, one could reasonably retort that Kant's final argue is inconclusive. From the fact that, necessarily, pure reason is capable of motivating the will and that, assuming the truth of this premise, the will must be free we cannot immediately infer that we ought to believe that we are free. That is to say, from the fact that autonomous motivation presupposes the reality of freedom no independent epistemic commitment can be detached. Kant himself attempts to avert this misconception by pointing out that the need of pure practical reason (Bedürfnis der reinen praktischen Vernunft) to presuppose the reality of freedom "is subjective, that is, it is a want, and not objective, that is, itself a duty, for there cannot be a duty to suppose the existence of anything (since this concerns only the theoretical employment of reason)" (KpV 05:125). Pure reason wants the will to be free because only on that condition can it determine the will immediately. This requirement is subjective in the sense that it is raised by pure reason itself and it is not individually addressed to us as an additional duty. Therefore, we don't have a self-standing duty to accept that we can be motivated by pure reason and hence that our will is free because, as Kant remarks, "a faith that is commanded is nonsense" (KpV 05:144). Our only duty is to obey the moral law, not to believe that the moral law can immediately motivate our will and hence that we are free. In what sense then is this postulate normative for us? The answer lies, I contend, in the fact that that the principle of autonomy gives us a reason to accept, not simply to believe, that we are free. Kant keenly expresses the voluntariness of the postulation of freedom by terming it 'an acceptance from a moral point of view' (Fürwahrhalten in moralischer Absicht, KpV 05:146). An act of acceptance9, as opposed to that of a theoretical belief, involves a conscious and voluntary choice of a premise which in our case is the real possibility of freedom and as such it "is not commanded, but being a voluntary determination of our judgment, conducive to the moral (commanded) purpose [...] it has itself sprung from the moral disposition of mind" (KpV 05:146). Nonetheless, the acceptance of freedom needs to be evidentially warranted and the ground for this not an irresistible metaphysical cause, so to speak, but the supreme principle of morality itself which "is not a postulate but a law" (KpV 05:132) and hence applies to us as a categorical imperative.

That is to say, we are objectively bound by the moral law and that legitimizes our choice to assume that we are capable to obey it and so that we are free. I have deliberately emphasized the importance of construing the postulate of freedom as an act of will because merely focusing on the peremptory function of the moral law would make us miss an important normative aspect of Kant's argument. To put it more graphically, we don't have freedom merely as a matter of principle, we have the right to demand that it really exist even though there might be an undiscovered mechanism that either foreordains the way we act morally or shows no interest at all in what the principle of morality has to say. The autonomy of pure reason does not simply induce faith in the reality of freedom.

Faith can be lost and this is not a state of mind that can be restored or maintained at will. Kant keenly alludes to the voluntary nature of the practical postulation of freedom in his remark that "it may therefore at times waver even in the well-disposed, but can never be reduced to unbelief [emphasis added]" (KpV 05:146). Therefore our acceptance that we are really free is an act of freedom itself.

A Kantian Rejoinder to Externalism

In the preceding analysis, I have tried to show that freedom is the ground of autonomous motivation in virtue of its postulation by the supreme principle of pure practical reason. In this last section, I shall capitalize on the merits of Kant's normative defence of autonomous motivation in order to advance a rejoinder to the plausibility of motivational externalism10. What I hope to make explicit is that Kant's intention to "prove that morality is no mere phantom of the brain" (GMS 04:445) further implies his interest in guarding autonomy of the will against the possibility of motivational skepticism. My final argument will focus on an analysis of the concept of practical syntheticity followed by an explication of what the determination of the will by pure reason consists in. In that way I hope to provide adequate support for my claim that failing to be motivated by moral principles cannot disprove the necessary bearing pure reason has on conduct.

It is worth mentioning from the outset that necessity is a common property of both analytic and synthetic a priori judgments. In the case of theoretical reason necessity -analytic and synthetic alike-is equated with strict universality (strenge Allgemeinheit) which further implies a proposition's lack of any possible exception (mõgliche Ausnahme) to its validity (see KrV B3). Nevertheless, there is a significant difference between analytic and synthetic necessity in virtue of the fact that the latter does not possess the inexorability of the former. That is to say, the negation of a synthetic a priori judgment does not entail a logical contradiction. For Kant, synthetic truths always remain logically defeasible11 in the sense that synthetic judgments imply a relation "which is consequently never one either of identity or contradiction, and by means of which the truth or error of the judgment cannot be discerned merely from the judgment itself" (KrV A153/B192). With respect to the practical realm, the synthetic necessity by which pure reason relates to the will also entails universality in the sense that the moral law holds "not merely for men, but for all rational beings as such" (GMS 04:408, see also KpV 05:21). Furthermore, insofar as the moral law applies to the imperfect will of finite rational agents practical synthetic necessity takes the form of "an imperative, i.e. a rule characterized by "shall", which expresses the objective necessitation of the action and signifies that, if reason completely determined the will, the action would inevitable take place according to this rule" (KpV 05:20). Hence the very concept of practical syntheticity is inextricably linked with the finitude of human motivational capacities.

In the rest of this section I shall avail myself of what I profess to be a striking resemblance between the logical defeasibility of theoretical synthetic truths and the failure to be motivated by practically synthetic moral imperatives. To illustrate my point, I believe that there is a common rationale behind the facts that a finite rational agent may fail to comply with the necessity of duty and that a theoretical judgment such as "All bodies are heavy" is true if and only if it is true and its denial does not logically entail a contradiction (see KrV A153/B192). By employing this analogy I hope to provide a cogent account of how the logical possibility of akratic or amoral conduct cannot vitiate the normative foundation of pure reason's synthetic relation to the will.

In order to employ constructively the aforementioned analogy, I shall first digress for a moment to examine the concept of an imperfect will determined by pure reason. Kant's ethical corpus abounds in suggestions that the will's necessitation by pure practical reason is related "only with the determination of the will and the determining principles of its maxims as a free will, not at all with the result [emphasis added]" (KpV 05:45, see also 05:66 and GMS 04:399-400). As I have previously noted, the will is immediately determinable by pure reason only insofar as it is free, that is, insofar as it is not motivationally dependent on intuitable objects such as desires or extra-mental properties. Furthermore, I have argued that the real possibility of a free will is premised on a practical postulate grounded in the principle of autonomy of the will. Once we supply these premises it follows that autonomous motivation cannot be conditional on the desirability of objects of any kind.

Provided that Kant divests the free will of its dependence on objects of desire, how are we to understand the will merely as a formal concept? To be more specific, Kant defines such a will as "the capacity for desiring in accordance with concepts, insofar as the ground determining it to action lies within itself and not in its object [emphasis added]" (MM 06:213). As late as in the Metaphysics of Morals he came to term this will the capacity for choice (Willkür) to the extent that this capacity "is joined with one's consciousness of the capacity to bring about its object by one's action" (MM 06: loc. cit.). By sharp contrast, Kant has been frequently referring to pure practical reason itself as Wille, that is, "the capacity for desire considered not so much in relation to action (as the capacity for choice is) but rather in relation to the ground determining choice to action" (MM 06: loc. cit.). Undoubtedly the latter distinction raises a terminological issue which had remained unsettled at least until Kant came to regiment his use of these two terms in the Religion within the Limits of Reason Alone and later in the Metaphysics of Morals. Nonetheless, the distinction between the, so to speak, executive and the legislative aspect12 of the will might prove fruitful for our present analysis as soon as we realize that it squares superbly with the structure of Kant's internalist proposition.

To clarify this point, my proposal is to read the relation of Wille to Willkür as an equivalent reformulation of the relation of pure practical reason to the will. As Kant says, "the will itself (Wille), strictly speaking, has no determining ground; insofar as it can determine the capacity for choice [emphasis added], it is instead practical reason itself "(MM 06: loc. cit.). Hence, Wille, that is, pure practical reason, is the source of a priori practical principles that serve as the determining ground of Willkür understood as the capacity of rational beings to "determine their causality by the conception of rules" (KpV 05:32).

Consequently, the capacity of the will to generate its own principles and be motivated by them irrespective of the desire for an end is indicative of Kant's understanding of moral motivation as a reflexive exercise an agent engages herself in by asking the second-order question of "whether we should [emphasis added] will an action that is directed to the existence of an object, if the object were in our power" (KpV 05:58).This reflexive quality in Kant's conception of internalism sheds new light on his claim that action in accordance with principles presupposes the determination of the will (qua Willkür) by pure reason "even before I ask whether I have power sufficient for a desired effect, or the means necessary to produce it" (KpV 05:20). Both the action per se and its desired effect occupy the position of what Kant calls the maxim's matter (Materie) which is interchangeably defined as the object "the realization of which is desired" (KpV 05:21) or "the object of the will" (KpV 05:27).

The matter of a practical maxim is to be strictly distinguished from its determining ground (Bestimmungsgrund), that is, from the reflexive appraisal of the matter's suitability as a universal practical law. If the matter assumes the function of a determining ground without the mediation of this reflexive evaluation, that is, without ascertaining the moral, not simply the physical possibility13 (physische Möglichkeit) of a desired end, the result will be heteronomous motivation. Although both the matter and the determining ground are necessarily related to the act and its desired effect, the former regards only the empirical attainability of an act as a means to an end whereas the latter provides an answer to the question of moral possibility (moralische Möglichkeit), that is, whether a causally efficient desire to § by means of Φ is also a universally adoptable reason for pursuing it.

The preceding remarks provide a sufficient background for gaining a better grasp of Kant's principle-based conception of moral agency. The desirability of an act as a means to an end is not banished rather it is subjected to the limiting condition of its being fit for serving as the content of a universal law. It is precisely the representation of this fitness for universal legislation that serves as the only genuinely moral motive. Objects of desire can never supply the motive of moral actions although they are indispensable14 to them in terms of supplying the material necessary for their empirical manifestation. Kant stresses this point by saying that "it is indeed undeniable that every volition must have an object, and therefore a matter; but it does not follow that this is the determining principle and the condition of the maxim" (KpV 05:34). Therefore autonomous motivation does not refer to "the relation of the will to the action by which the object or its opposite would be realized" (KpV 05:57) but, as it has been previously mentioned, only to the relation of pure practical reason to the will per se. Insofar as a free will is presented with a principle that fulfills the criteria of universal legislation, there is nothing more to be said about moral motivation. Our appraisal of the moral possibility of what we desire to bring about suffices of itself to provide a motive compatible with our moral status as free agents. Kant explicitly endorses this view when he remarks that "provided that the will conforms to the law of pure reason, then let its power in execution be what it may" (KpV 05:45 in fine). According to his internalist conception, a will motivated by principles of pure reason is to be understood merely as a capacity which is ultimately attested by the practical postulation of freedom.

Granted that autonomous motivation is regarded as a capacity to act on unconditional principles of reason, we stand on firmer ground with respect to the question of whether externalism15provides a serious reason for skepticism about the motivational efficacy of practical reason.

If we utilize the aforementioned analogy between the motivational deviations of finite agents and the logical defeasibility of theoretical synthetic judgments, we can convincingly attribute to Kant the view that the actual violations of moral duty are instantiations of a merely logical possibility which remains totally unintelligible16 from the standpoint of pure practical reason. The necessity17 by which pure reason furnishes the capacity to act on its principles is, by its very nature as synthetic, prone to the logical (not practical) possibility of akratic or amoral conduct. The scope of practical synthetic necessity is not as all-encompassing as that of logical or conceptual necessity. It is Kant's firm belief that motivation is not synonymous with having an unconditional reason for action. Still we can rest assured that our capacity to act on universalizable reasons is necessarily attested by our rightful demand for freedom. For Kant, it is the will as a rational capacity that is necessitated by pure reason, not the will's result-producing acts as evidenced by his comment that " ethics does not give laws for actions...but only for maxims of actions" (MM 06:388). Pure practical reason is the source of principles that necessarily can determine a will to action, whereas the extent to which an agent is actually motivated by the law-like character of her maxim falls beyond the purview of practical rationality. Autonomous motivation becomes possible as soon as we have ascertained that we have a universally adoptable reason for attaining a purpose Φ by means of an act Φ. Even though an agent can desire to Φ simultaneously with having a reason to if, it is only the latter consideration that she should cite as the determining ground of her will. The rest is open to an infinite set of logical possibilities among which moral evil in all its ramifications.

My point, therefore, is that what preserves the necessity of the relation between pure reason and the will's determinability is neither a metaphysical truth nor a nomic physical fact but solely our inalienable right to postulate freedom as the property of "a causality of pure reason for determining choice independently of any empirical conditions" (MM 06:221).

Taking as a premise Kant's claim that only "that choice [Willkür] which can be determined by pure reason is called free" (MM 06:213), we can infer that the will is necessarily determined by pure reason only insofar as the deliberating agent exercises her rightful demand that she be free to act on principle. The asymmetric dependence18of the free will on pure reason rests solely on the need of pure reason to regard itself as immediately practical. Once an agent waives her right to postulate her freedom, everything is logically possible including immoral or amoral conduct. Conversely, no logical possibility can vitiate the practically necessary impact of pure reason as long as freedom is postulated as its ground because, as Kant points out, it is conceptually impermissible to locate freedom "in a rational subject's being able to make a choice in opposition to his (lawgiving) reason, even though experience proves often enough that this happens (though we still cannot conceive how this is possible)" (MM 06:226). To clarify this last remark, I shall recompose Kant's overall argument against externalism in the following chain of premises:

a) Freedom purports to be the ground of the practical synthetic necessity by which pure reason immediately determines the will because immediateness implies lack of dependence on the mediation of intuitions.

b) Freedom is really possible in virtue of our right as rational agents to postulate its reality in the practical realm.

c) If freedom is really possible, then freedom really is the ground of practical synthetic necessity.

d) The concept of freedom as a negative regulative principle of speculative reason cannot be understood as referring to particular objects and hence it is void of meaning (see KpV 05:50 and 05:55 in fine, MM 06:221).

e) Only the Fact of Reason, that is, consciousness of pure reason's practical manifestation, can provide freedom with a positive meaning as autonomy of the will (see KpV 05:48, GMS 04:336-447).

f) Granted that freedom can be understood only as the property of the will of being a law to itself independently of the objects of volition, "freedom of choice cannot be defined [emphasis added]...as the capacity to make a choice for or against the law (libertas indifferentiae), even though choice as a phenomenon provides frequent examples of this in experience" (MM 06:226).

g) Therefore, granted that freedom really is the ground of synthetic practical necessity and that it is conceptually true that freedom of the will is the capacity to be motivated by pure reason, pure reason necessarily motivates the free will.

The fact that freedom is meant as autonomy of the will is not a self-standing conceptual truth. Pure reason itself demands that it be conceptually true that freedom as the ground of autonomous motivation is understood only as the capacity to obey the moral law. The ground of this conceptual claim is ultimately normative19, that is, a free will solely understood as a moral capacity is not a conceptual truth simpliciter but a conceptual truth grounded in a practical demand of reason. Therefore, it is not that we cannot, so to speak, escape from acting as free in virtue of a metaphysical, conceptual or nomic necessity. Neither can we claim ex post facto that our failure to be motivated by the moral law bears witness to our lack of freedom and hence to our lack of moral responsibility. Kant proclaims the asymmetric dependence of the free will on the consciousness of moral duty by suggesting that "the concept of the freedom of the will does not precede the consciousness of the moral law in us but is deduced from the determinability of our will by this law as an unconditional command [emphasis added]... Everyone will have to admit that he does not know whether, were such a situation to arise [the temptation to violate duty], he would not be shaken in his resolution. Still, duty commands him unconditionally: he ought to remain true to his resolve; and thence he rightly concludes that he must be able to do so, and that his will is therefore free" (R 06:49n).

Here Kant willingly concedes that failing to act as morality enjoins is always logically possible, yet reason still demands that we understand our freedom as the ability to overcome inclinations and act on principles everyone could adopt as action-guiding. Kant claims that the possibility of motivational failure is an empirical proposition20 (Satz der Erfahrung, MM 06:226 in fine) which cannot constitute the expository principle (Erklãrungsprincip, MM 06:loc. cit.) of the concept of free choice. The latter concept is an Idea of pure reason and hence transcendent (see KrV A296/B352). As such it can only be given a practical (nor theoretical neither logical) definition that serves the need of pure reason to be of itself practical. The attempt to include in this practical definition the empirically logical possibility of motivational failure is nothing but a hybrid definition (definitio hybrida or Bastardeklarung) that adds "to the practical concept the exercise of it, as it is taught by experience" (MM 06:227).


Are practical synthetic a priori judgments possible? This question could aptly epitomize Kant's overall enterprise regarding the possibility of autonomous motivation. My aim was to show that practical syntheticity establishes an asymmetric dependence of the free will on pure practical reason in a way conducive to a normative understanding of moral motivation. Folk psychology can always occupy the matter of our maxims and thus shape the content of change we bring about on the empirical plane. Nevertheless, it totally fails to establish the ground upon which we choose to shape our reality. I believe that only a normative construal can allow us to see the depth in Kant's claim that "only freedom in relation to the internal lawgiving of reason is really a capacity; the possibility of deviating from it is an incapacity" (MM 06:226-227). All the rest follows from this initial premise. The normative gist of autonomous motivation consists in the idea that the change we can effect in the world by acting morally is the only way to affirm that if we are to be really free, this is how our freedom should manifest itself in the world. There is no need to resort to unfruitful investigations of the inscrutable causal capacities of our intelligible nature, since "for us men it is wholly impossible to explain how and why the universality of a maxim as a law -and therefore morality- should interest us" (GMS 04:460).

1 Motivating force has been attributed to a vast gamut of candidate normative concepts. A standard though not exhaustive division of opinion regarding the way in which normativity manifests itself in the practical realm is between proponents of judgment and existence internalism. Briefly put, judgment internalism holds that motivation is the result of making a sincere judgment or believing that one ought to 9 or has a reason to Φ On the contrary, existence internalists claim that it is in virtue of a consideration's being a reason or right-making that an agent is capable of being motivated to act accordingly. Although I am inclined towards classifying Kant as a proponent of existence internalism, this will not be the primary concern of this paper. What I will try to present instead is Kant's position on the more fundamental question concerning the way in which normativity broadly construed is necessarily related to motivation. For an attribution to Kant of existence internalism, see Darwall, 2006, pp. 296-297. For an informative discussion of the existence versus judgment internalism debate see Darwall, 1983, pp. 51 ff.

2 A binary relation between x and y is asymmetric when it holds from x to y, but not also from y to x. Such a unidirectional implication is logically expressed as ∀ x∀y (Rxy → ¬Ryx). The opposite holds for symmetric relations which hold both ways, i.e ∀x∀y (Rxy→Ryx). The asymmetric qualification of internalism works as a safeguard against the logical consequences of entailment, which allows that impossibility entails everything and necessary truth is entailed by everything. In the case of internalism, that would result in our having an invalid moral reason that does not affect the validity of its ensuing motive. For an in-depth analysis of the asymmetric structure of internalist relations see Hurley, 2001. Michael Smith appears to defend a symmetric-analytic account of internalism based on what he considers as a conceptual truth, namely that "If an agent believes that she has a normative reason to Φ then she rationally should desire to Φ" (Smith, 1994, p. 148). Space limit does not allow me to thoroughly critique Smith's argument; hence I shall confine myself to the remark that symmetricity entails a very strong account of concept possession that cannot disprove without any cost the objections raised by a linguistically competent amoralist.

3 The most direct reference to the connection of practical syntheticity with motivational internalism is made by Nelson Potter, see Potter, 1997.

4Matthew Bedke has recently expounded an internalist theory which is based on the possibility of an a posteriori synthetic identity between moral judgments and motivation (Bedke, 2009). His account (moral judgment purposivism) is premised on a synthetic necessity claim according to which "it is metaphysically impossible for a moral judgment to fail to have this purpose, much as it is metaphysically impossible for water to have a chemical composition other than H2O." (p. 201). Bedke's empiricist account views moral motivation as a social phenomenon causally explicable by means of "empirical information, particularly information about the evolutionary history of our moral practices." (p. 190)

5 For the view that the Fact of Reason does not carry the burden of justifying the autonomy of pure practical reason, see Luków, 1993 and O'Neill, 2002, p. 83.

6 My analysis is consonant with a constructivist interpretation of Kant's ethics. Unfortunately the length of this essay does not permit a thorough defence of Kantian constructivism; however, I firmly support the view that constructivism offers the most fitting account of the objectivity of moral principles in light of Kant's overall Critical enterprise.

7 Kant's conception of the relation of freedom and morality has not been unshakeable. Even his critical writings do not suggest a continuity of thought with regard to this issue. In the Groundwork, Kant explicitly embarks on the endeavor to deduce morality from a metaphysically robust conception of freedom, or in his own words "since it [morality] [...] must be derived solely from the property of freedom, we have got to prove [emphasis added] that freedom too is a property of the will of all rational beings" (GMS 04:447). Although almost all contemporary scholars agree on the fact that the Second Critique marks a shift in Kant's argumentative strategy, there has been an ongoing controversy as to whether this reversion signifies a total abandonment of a metaphysical grounding of freedom. Karl Ameriks is the most ardent proponent of the view that Kant never actually denied the need of an ontological proof of freedom; see especially Ameriks, 2003, pp. 161-192 and 226-248 respectively. On the contrary, Christine Korsgaard and Onora O'Neill (although in a less critical tone) share the view that Kant's final position is purely practical. For a concise presentation of this argument, see Korsgaard, Christine M., Morality as freedom, in Korsgaard, 1996, esp. pp. 162-171, O'Neill, 1989, pp. 51-65 and 66-77 respectively.

8 If the concept of free will was intellectually intuitable the relation of pure reason to the will would be analytic a priori. Autonomy of the will would be analytically contained in the concept of causa noumenon. Kant refers to this hypothesis when he remarks that this synthetic relation "would, indeed, be analytical if the freedom of the will were presupposed, but to presuppose freedom as a positive concept would require an intellectual intuition, which cannot be assumed" (KpV 05:31, see also GMS 04:447).

9 Andrew Chignell ingeniously correlates Kant's notion of Fürwahrhalten to the epistemic concept of acceptance. He further emphasizes the voluntary aspect in Kant's normative justification of freedom noting that the latter is effected by means of a positive epistemic attitude which is not to be confounded with a 'pro-attitude - i.e., gladness or hope or goodwill towards the proposition or its truth", rather it should be understood as "a willingness [emphasis added] to take it on board, to take it to be true." (Chignell, 2007, p. 35) For an excellent presentation of the distinction between belief and acceptance see also Cohen, 1989.

10 One of the few externalist approaches of Kant's motivational theory belongs to Karl Ameriks (Ameriks, 2006, pp. 3-22). Ameriks mistakenly assumes that respect as the motivational component of the allegedly internalist relation just is the moral feeling that moves the agents towards the proper action. His view is manifestly reflected in his claim that "although it is crucial on Kant's view of proper action that the specific feeling of respect for duty be present in some way, it also crucial for any proper human motivation that there be, in addition to feeling [emphasis added], a preceding (logically, if not temporally) founding judgment [...] Fortunately, Kant does not insist that the feeling of duty always has to be clearly explicit to consciousness" (Ibid., p. 19).

11 For an excellent analysis of the logical defeasibility of theoretical synthetic a priori judgements, see Hanna, 2001, pp. 239-264.

12 Lewis White Beck astutely distinguishes the qualitative and functional difference between Wille and Willkür by employing the contrast between freedom as spontaneity, that is, the faculty of initiating a new causal series in time, and freedom as autonomy. For a further analysis of this analogy, see Beck, 1960.

13 Kant's claim that the physical possibility of actions should not constitute the ground of autonomous motivation should not be construed as implying that physical impediments to action are unexceptionally of no moral interest. Kant dispels this error by clarifying that "this ought indicates a possible action, the ground of which is a pure concept. This action must certainly be possible under physical conditions [...] but these physical or natural conditions do not concern the determination of the will itself, they relate to its effect alone, and the consequunces of the effect in the world of appearances

[emphasis added]" (KrV A 547/B575, see also MM 06:404). What is clearly stated in this passage is that physical possibility refers to acts as means to certain purposes, whereas moral possibility refers to the reasons for pursuing an end Φ by means of act n. A physical incapacity must also be a moral incapacity in order to supply a reason against a certain action. To clarify this point, suppose that a person suffering from an incapacitating illness is the only witness of a child's drowning. Her handicap bears testimony to the empirical fact that it is physically impossible for her to save the child. If asked why she has refrained from trying to save the child, her first spontaneous answer might indeed be that a physical obstacle prevented her from doing so. However, once again the same question can be pressed on her in an intelligible manner, but this time the answer expected does not concern the cause of the forbearance but its reason and these two features are far from being trivial alternatives. If asked to provide a moral reason for her omission she could by proper reflection refine her answer and say that a moral norm enjoining salvation despite the savior's physical incapacity cannot be conceivably adopted by everyone, because it is not everyone that suffers from a physical deficiency, but only a minority of persons who thus would unequally carry the burden of its enactment.

14 The fact that Kant subsumes acts as a means to desirable ends under the concept of a maxim's matter provides sufficient reason for refuting the platitude that Kantian ethics marginalizes moral sentiments. While it is true that Kant excludes any feeling from assuming the position of a moral maxim's determining ground, he is totally affirmative with regard to the possibility of a maxim's matter having an affective content. The inclusion of both the act and its end within the matter of a maxim lifts the veil of confusion as to whether an emotional disposition is compatible with the maxim's moral worth. Kant does not show enmity towards the psychological involvement of the agent vis-à-vis her actions; indeed he clearly thinks that such an involvement is a necessary condition of agency in general. What he proposes instead concerns the further step of where to locate the normative weight carried by the matter of a maxim, that is, whether there are any universally adoptable reasons validating the attainment of what we feel as right. His view then is that the moral worth of a maxim's matter (act-end) cannot be attested by the matter itself "but merely from this, that the form of universality which reason requires as the condition of giving to a maxim of self-love the objective validity of a law." (KpV 05:34)

15Derek Parfit erroneously tries to condense all versions of synthetic internalism by means of the general term "non-analytically reductive internalism". I believe that his misconception lies in the assumption he makes that internalism always entails the asymmetric priority of the psychological capacities of the agent see Parfit, 2006, p. 335. With respect to the relation between normativity and motivation he says, "Though these claims do not mean the same, when (R) [we have a reason to do something] is true, that normative fact is the same as, or consists in, the fact reported by (M) [if we deliberated on the facts on a procedurally rational way, we would be motivated to do this thing]" (Perfit, 1997, p. 108). He then immediately remarks that "Such Internalists believe that, though (1) [a normative reason] is true only if (2) [a motivating reason] is true, these claims have different meanings. These Internalists would understand -though they would reject [emphasis added]. the view that, despite this man's motivational state, he has reasons to treat his wife better" (ibid., p. 110). Parfit's description of internalisms of the synthetic genre is inadequate in two ways. Firstly, he excludes without argument the possibility that the asymmetric primacy is held by what he calls a "normative fact". Secondly, he equally fails to consider the possibility of an a posteriori correlation of normative and physical facts without the former being eliminatively reduced to the latter (see also supra note 4).

16 Kant overtly refers to the practical unintelligibility of moral evil when he remarks that "But the rational origin of this perversion of our will whereby it makes lower incentives supreme among its maxims, that is, of the propensity to evil, remains inscrutable to us [.] there is then for us no conceivable ground [kein begreiflicher Grund] from which the moral evil in us could originally have come." (R 06:43). For an insightful elaboration on the practical unintelligibility of moral evil see Korsgaard, 1996, p. 171, 173.

17 Mark Timmons (Timmons, 1989) tries to analyze the notion of practical necessitation in terms of Kant's counterfactual claim that "an imperative [...] signifies that, if reason completely determined the will, the action would inevitably take place according to this rule" (KpV 05:20). Timmons reads this passage in a way that suggests "that the relation of necessitation expressed in 'ought' statements relates the idea of what rules an agent would act on, or will in accordance with, were she completely rational [.] to the idea of an agent who is not by nature completely rational [...] to claim that some agent, P, is necessitated to some A in some circumstance C is equivalent (via an analysis of the notion of necessitation) [emphasis added] to the claim that if P were deliberating in a completely rational manner, P would necessarily will to do A in C" (Timmons, 1989, p. 206). What seems to be the core mistake in Timmons's account is that practical necessity denotes a practical synthetic a priori relation and thus only pertains to finite, not ideally conceived rational agents who are not always willing to apply the moral rules. On the contrary, the necessity he analytically derives from the latter, is rather a conceptual (not synthetic) one, since it refers to an agent endowed with a divine will, i.e. a will that conceptually entails an always available disposition to act morally. As a result, the Kantian notion of practical necessitation (praktische Nõthigung) cannot be analytically equivalent with the conceptual necessity of a perfectly rational being who is unexceptionally willing to fulfill her duty. Timmons's argument constitutes a refinement of a similar argument presented by McCarthy, 1979, as well as his 1985, pp. 28-42.

18 I believe that Kant's conception of the synthetic relation between pure reason and the will can be analyzed into the proposition that the will is asymmetrically dependent on pure reason. The fact that Kant views the will's synthetic necessitation as an instance of asymmetric dependence is evidenced by his standard use of the terms Abhãngigkeit and Bestimmung as a way of reference to his internalist proposition. This idea is perspicuously expressed in his remark that "the relation of such a will [the imperfect will of a finite rational agent] to this law is dependence [Abhängigkeit] under the name of obligation, which implies a constraint [Nothigung] to an action, though only by reason and its objective law" (KpV 05:32, see also GMS 04:454). Jaegwon Kim locates the close tie between dependence and/or determination and asymmetricity within the framework of his account of supervenience, claiming that "dependence, or determination, is usually understood to be asymmetric whereas entailment (... ) is neither symmetric nor asymmetric" (Kim, 1990, p. 13). If we further attempt to translate practical syntheticity in terms of supervenience, we could endorse Kim's definition of a dependence relation as follows: "A-properties [the moral interest] depend on B-properties [the practical law] just in case A strongly covaries with B, but not conversely; that is, any B-indiscernible things are A-indiscernible but there are A-indiscernible things that are B-discernible" (Ibi., pp.13-4). The latter case refers to the logical (not practical) possibility of violating the moral law, which does not of itself abolish the synthetic necessity established between the asymmetrically supervenient law and the subvenient or dependent will.

19 R. Lanier Anderson provides an excellent theoretical analogue of the normative grounding of a priori syntheticity which corresponds glibly to my moral argument against externalism. As Anderson remarks, "Unlike a descriptive natural law, a prescriptive normative rule does not entail that all the particular cases it covers actually conform to the rule [...] The normative rule thus remains binding, even when it is violated, and thereby has a different 'direction of fit' from descriptive rules! (Anderson, 2001, p. 277).

20 My claim that autonomous motivation is impervious to the logical possibility of moral evil is an indirect rejoinder to Henry Allison's famous Incorporation Thesis (Allison, 1990, pp. 155 ff.). Allison, in an attempt to disqualify the causal accounts of autonomous motivation, has focused on a ever since highly quoted passage from the Religion, according to which "freedom of the will is of a wholly unique nature in that an incentive can determine the will to an action only so far as the individual has incorporated it into his maxim (has made it the general rule in accordance with which he will conduct himself); only thus can an incentive, whatever it may be, co-exist with the absolute spontaneity of the will (i.e., freedom)" (R 06:23 in fine, 24). Allison concludes that moral good as well as moral evil must be an act of a free will. Based on this assumption, he suggests that Kant's frequent comments on the universality of the propensity to evil (Allgemeinheit des Hanges zum Bõsen) imply that the relation of evil to the executive will (Willkür) must also be necessary and thus synthetic a priori, exactly as in the case of the will's relation to pure reason. Allison's argument about the incorporation of good or vicious incentives should not be dissociated from his underlying purpose of positing a parallel syntheticity with regard to the evil's relation to the will. I believe that Kant's explicit comments on the practical unintelligibility of evil clearly exclude the possibility of a practical synthetic a priori relation of the latter to the will. Furthermore, even if we tried to discern in his comments on the universality of evil an allusion to a possibly theoretical synthetic a priori relation, such a conclusion would certainly have no bearing on the practical issue of moral motivation. For a more sympathetic -though no less critical- account of Allison's argument see Frierson, 2003, pp. 34-38. Frierson keenly characterizes the propensity of man to evil as an expression of an empirical abuse of freedom, adding that "the 'necessity' that one with a propensity to evil will perform evil is not a logical necessity, nor a morally relevant necessity, but akin to a natural necessity" (Ibid., pp. 189-90).


Allison, H. R. (1990). Kant's theory of freedom. Cambridge: Cambridge University Press.

Ameriks, K. (2003). Interpreting Kant's Critiques. Oxford: Clarendon.

Ameriks, K. (2006). Kant and Motivational Externalism. H. Klemme, M. Kuehn, D. Schõnecker, Deutsche Forschungsgemeinschaft., & Philipps-Universitãt Marburg. Moralische Motivation: Kant und die Alternativen (pp.3-22). Kant-Forschungen, Bd. 16. Hamburg: F. Meiner.

Anderson, R. L. (2001). Synthesis, Cognitive Normativity, and the Meaning of Kant's Question, 'How are synthetic cognitions a priori possible?' European Journal of Philosophy, 9 (3), 275-305.

Beck, L. W. (1960). A commentary on Kant's critique of practical reason. Chicago: The University of Chicago Press.

Bedke, M. S. (2009). Moral judgment purposivism: saving internalism from amoralism. Philosophical Studies, 144 (2), 189-209.

Chignell, A. (2007). Kant's Concepts of Justification. Nous, 41 (1), 33-63.

Cohen, L. J. (1989). Belief and Acceptance. Mind, 98 (391), 367-389.

Darwall, S. (1983). Impartial Reason. Ithaca, NY: Cornell University Press.

Darwall, S. (2006). Morality and Practical Reason: A Kantian Approach. In D. Copp (Ed.), The Oxford Handbook of Ethical Theory (pp. 282320). New York, NY: Oxford University Press.

Frierson, P. R. (2003). Freedom and anthropology in Kant's moral philosophy. Cambridge, U.K: Cambridge University Press.

Hanna, R. (2001). Kant and the foundations of analytic philosophy. Oxford: Clarendon Press.

Hurley, S. L. (2001). Reason and motivation: the wrong distinction? Analysis, 61 (2), 151-155.

Kim, J. (1990). Supervenience as a Philosophical Concept. Metaphilosophy, 21 (1-2), 1-27.

Korsgaard, C. M. (1996). Creating the kingdom of ends. Cambridge: Cambridge University Press.

Luków, P. A. (1993). The Fact of Reason. Kant's Passage to Ordinary Moral Knowledge. Kant-studien, 84 (2), 204-221.

McCarthy, M. H. (1979). Kant's Application of the Analytic/Synthetic Distinction. Dialogue, 18, 373-391.

McCarthy, M. H. (1985). Paton's Suggestion that Kant's Principle of Morality Might Be Analytic. Kant-Studien, 76, 28-42.

O'Neill, O. (1989). Constructions of Reason: Explorations of Kant's Practical Philosophy. Cambridge: Cambridge University Press.

O'Neill, O. (2002). Autonomy and the Fact of Reason in the Kritik der praktischen Vernunft. In O. Hõffe (Ed.), Immanuel Kant: Kritik der praktischen Vernunft (pp. 81-97), Berlin: Akademie Verlag.

Parfit, D. (1997). Reasons and Motivation-I. Supplementary Volume Aristotelian Society, 71, 99-130.

Parfit, D. (2006). Normativity. In S. L. Russ (Ed.), Oxford Studies in Metaethics, vol. 1 (pp. 323-380). Oxford: Clarendon Press.

Potter, N. (1997). The Synthetic A Priori Proposition of Kant's Ethical Philosophy. Jahrbuch für Recht und Ethik / Annual Review of Law and Ethics. 5, 437-459.

Smith, M. (1994). The Moral Problem. Oxford: Blackwell Publishing.

Timmons, M. (1989). McCarthy on Practical Necessitation in Kant. Kant-Studien, 80 (2), 198-207.

Revista de filosofía de la Universidad del Norte

Universidad del Norte
Barranquilla (Colombia)