Incentives of the Mind: Kant and Baumgarten on the Impelling Causes of Desire

1 Introduction

At the core of Kant’s practical philosophy is the claim that pure reason is practical, that is, that “[p]ure reason is practical of itself alone and gives (to the human being) a universal law which we call the moral law” (5:31).^[1] Ever since the first publication of the Critique of Practical Reason in particular, wherein it is Kant’s aim to establish this claim (see 5:3), Kant’s commentators have been especially perplexed by his psychology of moral motivation.^[2] More than anything else, the aspect of Kant’s view that has led to extensive debate in the secondary literature is Kant’s claim that acting on the basis of pure practical reason involves a special feeling “self-wrought by a rational concept” (4:401n), namely the feeling of respect (Achtung) for the moral law, which functions as what he calls “the sole and also the undoubted moral incentive (Triebfeder)” (5:78).

For quite some time, many commentators agreed that Kant’s rationalism precluded psychological states like feelings and desires from playing a meaningful role in the motivation of moral action. As a result, the feeling of respect was thought to account for what it was like to experience being motivated by the moral law but not play a positive, causal role in Kant’s explanation of moral action per se. ‘Respect’ on this reading is merely “the epiphenomenon of moral motivation,” as one scholar put it (Sytsma 1993, 121).^[3] This ‘intellectualist’ interpretation became so widespread that Kant is still understood this way by many working in contemporary moral psychology, and with significant consequences. As one group of scholars has put it, “theories of morally worthy motivation that best fit the current scientific picture […] owe much more to Hume or Aristotle than to Kant” (Schroeder, Roskies, and Nichols 2010, 72), primarily because “[t]he neuroscience raises difficulties for the cognitivist’s [i. e., Kant’s] story, since our moral behavior does not appear to be under the control of cognitive states alone, independently of desire” (Schroeder, Roskies, and Nichols 2010, 106).^[4]

Scholars have now by and large come to agree on a different, more ‘affectivist’ reading.^[5] As Jeanine Grenberg has forcefully argued, for instance, although Kant’s rationalism precludes psychological states like feelings from playing a role in the justification or grounding of the moral law, this does not necessarily preclude such states from playing a positive role in Kant’s account of moral motivation.^[6] Due to the earlier dominance of the intellectualist interpretation, however, for quite some time scholars were focused on showing that feeling was involved in moral motivation rather than explaining how this works.^[7] The task for more recent scholarship has been to explain how exactly the feeling of respect might play a positive role in action motivated by the moral law alone.

The advantages of the affectivist interpretation notwithstanding, in this paper I argue that the ways in which scholars have described the causal role of feeling in Kant’s psychology of moral motivation have disadvantages of their own. A major source of confusion for these interpretations has been Kant’s claim that respect is both a feeling and the ‘incentive’ (Triebfeder) of moral action. Utilizing an underappreciated historical source for Kant’s concept of an incentive, namely Alexander Gottlieb Baumgarten’s Metaphysics, I seek to clarify the confusion by arguing that Kant calls the feeling of respect an incentive because he assigns it the role of an impelling cause of desire.

My discussion proceeds in four sections. In section 2, I introduce the topic at hand by illustrating that Kant’s account of moral motivation is not meant to explain why the moral law determines the will, but only what happens in the mind, that is, psychologically, when the law does so determine the will. Kant’s answer is that in this case we experience certain feelings, in particular the feeling of respect. It is the role of this feeling in moral motivation that has caused a significant interpretive debate. I outline several versions of the affectivist interpretation and illustrate their shortcomings. In section 3, I discuss Baumgarten’s understanding of an incentive, according to which a cognition of the good is an incentive when it functions as what he calls an ‘impelling cause’ of desire, which is distinct from efficient causation in that it does not necessarily produce its effect. In section 4, I offer an interpretation of Kant’s psychology of moral action against this background. I argue that, for Kant, it is feeling rather than cognition that functions as the ground of desire and that respect in particular functions as an impelling cause of desire. In section 5, I discuss the advantages of my interpretation and address some potential objections.

2 The Psychology of Moral Motivation

In Kant’s most considered presentation of his account of moral motivation^[8], namely that contained in the third chapter of the ‘Analytic’ of the Critique of Practical Reason entitled ‘On the incentives of pure practical reason’ (5:71–89), Kant clarifies what his account of moral motivation both is and is not meant to accomplish:

[H]ow a law can be of itself and immediately a determining ground of the will (though this is what is essential in all morality) is for human reason an insoluble problem and identical with that of how a free will is possible. What we shall have to show a priori is, therefore, not the ground from which the moral law in itself supplies an incentive (Triebfeder) but rather what it [the moral law] effects (or, to put it better, must effect) in the mind insofar as it is an incentive. (5:72, my emphasis)

Kant clarifies in this passage that his account of moral motivation is not meant to explain how the moral law determines the will in the sense of supplying the ‘ground’ on the basis of which this determination takes place or explaining why it happens. Rather, his aim is merely to explain what the moral law ‘effects in the mind’ insofar as it does determine the will. The location of the incentives chapter in the second Critique makes Kant’s task even more clear: at this point in the text Kant takes himself to have already accomplished what he announced in the Preface as the main goal of the book, namely “to show that there is pure practical reason” (5:3). Kant presents the argument for this conclusion, which involves his famous doctrine of the ‘fact of reason,’ in Chapter 1 of the Analytic. In Chapter 3, wherein Kant turns to the topic of moral motivation, he therefore presupposes that the moral law can be an immediate determining ground of the will, that is, that we can be moved to action by the moral law alone. Indeed, this is just what it means for pure reason to be practical (5:31). The task of Chapter 3 is to offer further details about the way in which this takes place. We might therefore say that Kant’s aim in the incentives chapter is to discuss the “applied” part of “pure morality”, as he puts it in the first Critique, which assesses moral laws “under the hindrances of the feelings, inclinations, and passions to which human beings are more or less subject”, and thereby “requires empirical and psychological principles” (A55/B79). Similarly, in the Groundwork, Kant refers to this part as the “empirical part” of moral philosophy, or “practical anthropology,” which concerns itself with “the human being’s will, in so far as it is affected by nature” as well as the laws “according to which everything ought to happen, while still taking into consideration the conditions under which quite often it does not happen” (4:387–388).^[9] When we regard pure reason as practical, we therefore seek to explain how “the pure representation of duty” has “an influence on the human heart so much more powerful than all other incentives one can summon from the empirical field” (4:410–411).

In the second Critique, Kant explains that when pure reason is practical, certain feelings are brought about.^[10] In the first instance, these feelings are negative: since acting morally involves acting “not only without the cooperation of sensible impulses but even with rejection of all of them” (5:72), doing what the moral law requires involves the negative feeling of “pain” (5:73). In addition, Kant claims that we also feel “humiliation” because the moral law “strikes down” (5:73) our tendency to make our selfish impulses lawgiving, which Kant calls “self-conceit” (5:74). However, Kant’s more important claim is that when pure reason is practical a positive feeling is brought about as well, namely the feeling of “respect” for the moral law, which Kant calls “the sole and also the undoubted moral incentive” (5:78).

It is relatively uncontroversial that Kant believes that our awareness of the moral law causes us to feel pain, humiliation, and respect and, indeed, that we necessarily feel these feelings.^[11] What has been the subject of intense debate is the answer to a much more basic question: if Kant’s goal is to explain what happens in the mind when the moral law determines our will on its own, why does he turn to feelings in particular? Put differently: what is significant about feelings such that they are the appropriate focus at this stage of Kant’s discussion? Answering this question is difficult because, as Stephen Engstrom has noted, “Kant does not say why indicating a priori what the moral law effects in the mind requires an explanation of the law’s effect on feeling – that is, on the capacity to feel pleasure and displeasure” (Engstrom 2010, 98).

The secondary literature has proposed a variety of ways in which to understand the role of feeling in Kant’s account of moral motivation.^[12] As already mentioned, the ‘affectivist’ interpretation sees feeling as playing a positive, causal role in moral action as opposed to a merely epiphenomenal one. However, scholars have offered several interpretations of the precise nature of the role that feeling plays in Kant’s psychology of moral action. According to one widespread version, Kant understands feeling as a ‘psychological force’ that immediately precedes and is the source of action. Thus, Lewis White Beck famously described Kant’s psychology of action in the following way: “In acts of will, there are two distinguishable factors: a want, which I have called the ‘dynamic’ or ‘conative’ factor, and what we recognize as that which we ought to do, the recognition of which I have called the ‘cognitive’ factor” (Beck 1960, 76). Beck identifies the ‘conative’ factor with what “is sometimes called desire proper” (90) and claims that, for Kant, “Triebfeder [incentive] is the generic name for the dynamic or conative factor in willing” (216). Beck thus links the conative factor in willing to “desire proper,” which is in turn linked to the moral incentive, that is, the feeling of respect. Several interpreters agree with Beck that the feeling of respect is, as an incentive, functionally equivalent to a desire. Melissa Zinkin, for instance, has argued that an incentive, and thus the feeling of respect, is “something in a finite rational subject, such as a desire or a particular end that determines, or motivates, it to act” (Zinkin 2006, 40). Similarly, Jens Timmermann has called the incentive of respect a “motivating desire” in the sense that it is “the force that propels an agent forward if he or she so chooses” (Timmerman 2011, 180).^[13]

While the majority of those who subscribe to the psychological force interpretation do not explicitly identify the feeling of respect with desire, they nonetheless suggest that Kant takes the role of feeling in moral action to be functionally equivalent to desire. To cite just a few examples, A. T. Nuyen claims that “the feeling of respect is a mechanism that propels the sensuous self to moral action” (Nuyen 1991, 40), and Ido Geiger argues that, on Kant’s view, “the effective forces driving naturally affected beings – what actually move us to action in the phenomenal world – are feelings” (Geiger 2011 289).^[14]

While most of these interpreters rightfully note the central role that free choice plays in Kant’s psychology of action,^[15] conceiving of the feeling of respect as a psychological force that immediately precedes and is the source of action has considerable disadvantages. Most importantly, it risks suggesting that there is only one psychological force operative in action. Indeed, without explaining the distinct roles that both feeling and desire play in Kant’s psychology of moral motivation, one risks reducing feeling to desire or vice versa. This is a significant problem because feeling and desire are two fundamentally different kinds of mental states for Kant.^[16] As I will discuss in detail in section 4, Kant subscribes to a tri-partite theory of the faculties of the human mind, according to which the faculties of cognition, desire, and the feeling of pleasure and displeasure are fundamental in the sense that they “cannot be further derived from a common ground” (5:177), and thus cannot be reduced to each other either.

A related problem for the psychological force interpretation is that it fails to appreciate some important nuances of Kant’s conception of desire. Kant defines the faculty of desire as “a being’s faculty to be by means of its representations the cause of the reality of the objects of these representations” (5:9n), and desires can take many forms, such as inclinations, instincts, affects, and passions (see, e. g., 7:751). What these types of desire have in common is that they actively cause the reality of an object. Patrick Frierson has aptly described this aspect of Kant’s understanding of desire in the following way:

Because desires simply are representations insofar as those representations are directed towards action, Kant’s notion of desire is more closely connected to choice and action than the customary English sense of desire, whereby one can desire something without actually pursuing it. Once one has a desire in this general sense, one is committed to action, and action follows necessarily in the absence of unforeseen hindrances. […] [O]ne’s representation will not count as a desire unless it prompts one to action. (Frierson 2005, 10)

To put the point in contemporary terminology, to desire on Kant’s view is not to have a ‘standing’ desire, where one has a desire without it being active in our psyche and prompting us to action.^[17] On the contrary, he primarily has in mind desires that are ‘occurrent’ or ‘manifest,’ that is, are “active,” such that desiring “consists in acting” (28:254), as he purportedly explains in the Metaphysik L1 lecture notes from 1777–80.^[18] On Kant’s view, to desire but not actually pursue the reality of an object is to possess an “inactive desire or yearning” (Metaphysik L1 [1777–80] 28:254), which is nothing but a “wish” (see, e. g., 7:251 and 4:394). Thus, it is inaccurate on Kant’s view to say that we can choose to act on a desire, because for him desiring means one has chosen to act.

Perhaps in recognition of these problems, a number of scholars have recently attempted to assign feeling and desire distinct roles in Kant’s psychology of moral action.^[19] One such example is Janelle DeWitt, who appreciates that “the activity of the faculty of desire is the actual movement toward the object to be brought about” and also rightfully notes that something else “must initiate that movement” (DeWitt 2014, 41). Similar to the interpretation that I will later put forward, DeWitt argues that what ‘initiates’ desire is feeling: “any representation capable of motivating action (i. e., determining the will) is, by Kant’s definition, a feeling” (37). Yet I disagree with DeWitt’s claim that feeling is capable of motivating action only insofar as it is cognitive. For DeWitt, feeling is “the faculty that judges the object of a representation of cognition to be good in relation to a subject, and so brings the representation under the active scope of the faculty of desire” (41–42, my emphasis) and she even calls respect an “action-initiating value judgement” (54–55). Thus, although DeWitt’s reading assigns distinct roles to feeling and desire, it accomplishes this at the cost of intellectualizing feeling and thus not respecting the ‘irreducible’ character of Kant’s tri-partite structure of the mind.^[20] Alix Cohen also notes that desiring consists in acting (Cohen 2015, 5–6) and that cognition, feeling, and desire all play unique, irreducible roles in action. Furthermore, she specifies that “[a]ll desire and inclinations are caused by feelings” (Cohen 2015, 6, my emphasis), but no more than DeWitt does Cohen go on to explain how exactly feelings cause or initiate desire.^[21]

I want to suggest that a major source of confusion for many of the interpretations currently on offer is Kant’s concept of an incentive itself. As already mentioned, Kant claims that the feeling of respect is “the sole and also the undoubted moral incentive” (Triebfeder) (5:78), but he also notes in various places that the moral law itself is the incentive (see, e. g., 5:88), which might explain the tendency of some commentators to intellectualize the feeling of respect. Most interpreters, however, understand an incentive to be a psychological force, which is likely why so many conceive of the feeling of respect as a psychological force as well. Andrews Reath, for instance, has recently claimed that “a Triebfeder for Kant is a psychological state of a subject that is a source of action” (Reath 2021, 244). Similarly, Engstrom says that “when Kant speaks of a Triebfeder, he almost always has in mind something in the subject that generates the action” and that, consequently, “we should think of a Triebfeder as an inner spring or source of choice and action” (Engstrom 2010, 92).^[22] Grenberg even translates Triebfeder as “drive” (see Grenberg 2001, 155) and says that “desires and aversions constitute an agent’s drives [i. e., incentives] to action, and are, all else being equal, what determines action” (163). This leads Grenberg to claim that on Kant’s view, “[f]or the purposes of describing action there is, however, little distinction to be made between the possession of practical pleasure [i. e., feeling] and that of a desire” (163). Understanding an incentive as a psychological force akin to desire is likely what leads her and others to identify the roles of feeling and desire in moral action.

The most promising version of the affectivist interpretation is Patrick Frierson’s recent account of Kant’s psychology of moral motivation from the point of view of empirical psychology. Frierson not only assigns unique roles to each of Kant’s three fundamental faculties, but also specifies the precise nature of the causal relationship between feeling and desire. As Frierson notes, however, empirical psychology is a “purely descriptive and explanatory” (Frierson 2014, 146) discipline according to which all human action is in line with “psychological determinism” (Frierson 2014, 149). Thus, when it comes to the role of feelings in Kant’s psychology of moral motivation, the feeling of respect functions as an efficient cause of desire.

As Frierson acknowledges, however, this is not the only perspective we can take on action, nor is it the perspective on action that Kant is explaining in the second Critique (see Frierson 2014, 16). From the perspective of pure practical reason, we necessarily conceive of ourselves as free and thus not strictly determined by antecedent causes. As Kant writes in the Groundwork: “Reason must view herself as the author of her principles, independently of alien influences, and must consequently, as practical reason, or as the will of a rational being, by herself be viewed as free” (4:448). There must therefore be a way of understanding the role of feeling in moral action that meaningfully distinguishes it from desire, specifies the relation feeling has to desire, and preserves a role for freedom.^[23] This is what the account I develop in the remainder of this paper hopes to achieve.

In the following two sections I argue that Kant’s concept of an incentive is rooted in Baumgarten’s account. For Baumgarten, an incentive does not refer to a desire, but to the ground of desire, which functions as an ‘impelling cause,’ that is, one that encourages but does not necessitate an effect.

3 Baumgarten on Incentives as Impelling Causes of Desire

Alexander Gottlieb Baumgarten (1714–1762) was a prominent proponent of Christian Wolff’s philosophy and the author of textbooks that Kant used in the classroom.^[24] In the Empirical Psychology chapter of Baumgarten’s Metaphysics, which formed the basis of Kant’s lectures on anthropology, Baumgarten employs the concept of an incentive. In fact, Baumgarten’s use of the concept of an incentive in a psychological context played a leading role in extending the application of what was originally a mechanical term to the mind (see Schwaiger 1999, 161–162 n574). That the term ‘incentive’ (Triebfeder) can be found in Baumgarten has been noted by others.^[25] What has not yet been explored is the precise way in which Baumgarten understands this concept, which is my focus in the present section.

Baumgarten’s empirical psychology begins by defining the soul in Wolffian terms as “a power for representing the universe according to the position of its body” (§ 513; see Wolff [1720] 2003, § 192).^[26] Baumgarten then devotes the majority of the chapter to the treatment of two main faculties of the single power of the soul: the cognitive faculty and the appetitive faculty. It is during his discussion of the appetitive faculty, that is, the faculty of desire, that Baumgarten introduces the concept of an incentive. Baumgarten defines desire as follows: “If I endeavour to make an effort to produce some perception, i. e., if I determine the power of my soul, or myself, to produce some perception, I desire” (§ 663). In line with his belief in the principle of sufficient reason (see § 22), we do not simply desire indiscriminately but always for a reason. Thus, for Baumgarten, there are grounds of desire, and these grounds are what he calls ‘incentives’: “Whoever desires or averts intends the production of some perception. Hence, the perceptions containing the ground of this sort of intention are the impelling causes of desire and aversion, and thus they are called the incentives of the mind <elateres animi>” (§ 669). According to this basic definition, then, Baumgarten understands incentives as ‘perceptions,’ that is, representations or mental states, which function as the ground of a desire.

An important feature of Baumgarten’s view is that the ‘perceptions’ grounding desire are associated with the other main faculty he discusses in the empirical psychology, namely the faculty of cognition or knowledge. Immediately after offering the above definition of an incentive, Baumgarten states that “knowledge, insofar as it contains the incentives of the mind, is moving […] and insofar as it does not contain these incentives, it is inert” (§ 669). More specifically, it is knowledge of the good, which Baumgarten identifies with “perfection” (see § 100), that is moving. Indeed, Baumgarten’s rationalist conception of agency entails the thesis according to which we only desire that which we cognize as good or perfect (see, e. g., § 665).

Baumgarten distinguishes between two fundamentally different kinds of desire: sensitive desire and rational desire or the will (see § 676 and § 689, respectively). This is because “the faculty of desire follows the cognitive faculty” (§ 676), and the cognitive faculty is itself of two distinct kinds: sensitive and rational (see § 519 and § 624, respectively). Baumgarten follows Wolff’s Latin works in distinguishing between the ‘inferior’ and ‘superior’ faculties of cognition, such that sensitive cognition is obscure and confused and rational cognition is clear and distinct.^[27] The important point for my purposes is that, because the faculty of desire “follows” the cognitive faculty, the way in which we know the good determines the way in which we desire: if we know the good obscurely and confusedly, then we sensitively desire and have an inferior appetitive faculty (see § 676–§ 688); if we clearly and distinctly know the good, we rationally desire or “will,” that is, we have a superior appetitive faculty (see § 689–§ 699). This means, of course, that there are two types of cognition that can ‘ground’ desire and thus two kinds of incentives: “The incentives of the mind are either stimuli or motives” (§ 690); if sensitive cognition grounds desire, then incentives are called “stimuli” (see § 677), whereas if rational cognition grounds desire, an incentive is called a “motive” (§ 690).

An important aspect of Baumgarten’s view is that he understands incentives in terms of a unique kind of causality, namely ‘impelling causation.’ In Part I of the Metaphysics, devoted to ontology, Baumgarten discusses the four traditional, Aristotelian types of causality (namely, efficient § 319, final § 341, formal, and material causation § 345), occasional causation (§ 323), and a sixth type of causality that seems to apply exclusively to action: impelling causation. After clarifying that an end is that “which seems good to the agent,” Baumgarten says that “the representation of the end is called the intention” (§ 341). As we have seen above, however, Baumgarten understands intention as identical to desire: “Whoever desires or averts intends the production of some perception” (§ 669). Accordingly, just as desire has grounds, so does intention: “The grounds of an intention in the one intending are called the impelling causes” (§ 342).

Although Baumgarten does not make this clear until much later in the Metaphysics, what is distinct about impelling causation is that it does not make its effect necessary. Whereas an efficient cause necessarily brings about an effect such that “[w]hen an efficient […] cause is posited, an effect is posited” (§ 326, see also § 328), an impelling cause does not necessitate its effect. In the context of action, this means that cognition (the impelling cause of desire) does not necessitate desire (the effect). On the contrary, for Baumgarten we only desire what we “prefer,” that is, what we judge to be the best of the goods available to us (see § 726 and § 697), and one judges what is best and thereby determines what one prefers by counting or weighing the impelling causes in favor of or against various actions:

By considering how much good or evil is to be hoped for from either side, they [i. e., a person deliberating] count the impelling causes, which they weigh. By judging how great are the goods or evils that are to be hoped for, and by weighing carefully which is better, they prefer one or the other. (§ 697)

Accordingly, Baumgarten believes that choice is “determined according to preference” (§ 726, my emphasis), that is, we necessarily choose and desire what we judge to be best or “prefer.”^[28] The important point for my purposes is that a singular impelling cause or incentive does not necessarily determine an effect (a desire) on its own. What causes us to desire is rather the “state of preponderance” (§ 674), that is, the greatest cumulative degree of strength that one or many impelling causes have on balance with the impelling causes that speak for opposing options. Whatever option has the greatest impelling force, outweighing those of the other available options, is the one I will choose, for this is what I judge to be best or what I prefer. Thus, one single impelling cause or incentive is simply one reason among many that speaks for or grounds desire and thus, on its own, merely impels or encourages a person to act in a particular way.

The fourth edition of the Metaphysics, which is the one that Kant heavily annotated and used in the classroom, is the first edition to include Baumgarten’s own German translation of key philosophical terms. Next to elateres animi, Baumgarten notes “Triebfedern des Gemüths” (see § 669), which, if we translate directly from the Latin, might be rendered as ‘springs of the mind.’^[29] In the second Critique, Kant defines an “incentive (elater animi)” as “the subjective determining ground of the will of a being whose reason does not by its nature necessarily conform with the objective law” (5:72). The inclusion of the Latin expression would surely have signaled to Kant’s students and eighteenth-century readers that he is using Baumgarten’s terminology. Beyond this fact, there are further reasons to think that Kant understood the concept of an incentive in a broadly similar way to Baumgarten, namely, as referring to the ‘ground of desire.’ First, as we will see in more detail below, Kant also distinguishes between two kinds of incentives: in the Groundwork, for instance, he distinguishes between incentives “from the empirical field” (4:411) and the motive of duty (4:412), that is, the incentive of respect.^[30] Second, Kant’s definition of an incentive in the Groundwork explicitly describes it as the ground of desire: “The subjective ground of desiring is the incentive” (4:427).^[31] Furthermore, both the early Prize Essay and the second Critique explicitly take the faculty of feeling to supply the “subjective ground of desire” (see 5:90 and 2:300). In the next section I suggest that Kant understands the concept of an incentive in largely the same way as Baumgarten, albeit with some qualifications, and that this reveals a different way of understanding the role that the feeling of respect plays in Kant’s psychology of moral motivation.

4 Kant on Feelings as the Ground of Desire

Before turning to the role that the faculties of feeling and desire play in the case of moral action in particular, it is relevant to discuss both Kant’s faculty psychology more generally and the structure of these faculties in the case of non-moral action.^[32] As briefly mentioned in section 2, the human mind, according to Kant, has three fundamental faculties that are reducible neither to each other nor to a single, more fundamental power.^[33] Kant states this clearly in the Critique of Judgement:

[A]ll faculties or capacities of the soul can be reduced to the three that cannot be further derived from a common ground: the faculty of cognition (das Erkenntnißvermögen), the feeling of pleasure and displeasure (das Gefühl der Lust und Unlust), and the faculty of desire (das Begehrungsvermögen). (5:177, cf. also 20:205–206 and 10:513–516)^[34]

These three faculties are distinct because they allow the subject to relate to objects in distinct ways. Cognition and desire are inverse ways in which the subject relates to objects: whereas cognitions are representations that “are related merely to the object” (20:206), desires are representations that are “the cause of the reality of this object” (20:206), that is, they are representations that bring about objects outside of the subject. Feelings are drastically different from both cognitions and desires insofar as they tell us nothing about objects at all; although always occasioned by a cognition, as I will explain below, feelings relate “merely to the subject” (20:206).

Although Kant disagrees with his rationalist predecessors in a number of respects, he makes use of an important terminological distinction employed by both Wolff and Baumgarten, namely that between the superior and inferior faculties of the mind, and applies it to all three fundamental faculties.^[35] However, whereas Baumgarten distinguished between superior and inferior cognitions, for example, on the basis of their degree of clarity and distinctness, the basis of the distinction for Kant is a difference in kind.

In the Anthropology, for example, Kant says that the mind is either active or passive, such that the representations we passively receive “belong to the sensible (sinnliche) cognitive faculty, whereas ideas that involve a sheer activity (thinking) belong to the intellectual cognitive faculty. The former is also called the inferior; the latter, the superior cognitive faculty” (7:140, translation modified; see also A15/B29). The superior and inferior faculties of feeling and desire are also said to be distinct in kind because they stem from different sources: feelings are either “sensuous” or “intellectual” (7:230) based on whether they have their source in sensibility and the imagination (inferior cognition) or ideas and concepts (superior cognition) (see 7:230). According to the lecture notes, Kant believes that all “desires are, like pleasure and displeasure, intellectual or sensitive” (Metaphysik Mrongrovius [1782–83] 29:894) based on whether their ultimate source is in the understanding or sensibility, respectively (see Metaphysik Arnoldt [1794–95] 29:1014–1015).^[36]

In moral as well as non-moral actions, each of these three faculties performs a specific role. As Frierson has illustrated in great detail (see Frierson 2005 and 2014), Kant is committed to the idea that moral and non-moral actions have the same structure such that, as we are told in the Metaphysik Mrongovius lecture notes, “the cognitive faculty is connected with the faculty of desire by the feeling of pleasure or displeasure” ([1782–83] 29:890). The following passage from the same set of notes offers a clear explanation of this view:

Pleasure precedes the faculty of desire, and the cognitive faculty precedes pleasure. […] [W]e can desire or avert nothing which is not based on pleasure or displeasure. For that which gives me no pleasure, I also do not want. Thus pleasure or displeasure precedes desire or aversion. But still I must first cognize what I desire, likewise what gives me pleasure or displeasure; accordingly, both are based on the cognitive faculty.^[37]

To take just one form of non-moral action as an example, consider drinking a coffee: I might see someone drinking a coffee (a sensible/inferior cognition), which brings to mind the (sensitive/inferior) pleasure I have experienced in the past when drinking coffee, which in turn causes me to have the (sensitive/inferior) desire to drink a coffee.^[38]

Let us now turn to the case of moral action. The role played by cognition in this case is fairly clear: as I mentioned in section 2, Kant’s account of moral motivation presupposes that the moral law does in fact determine the will on its own. As Kant explains via his doctrine of the fact of reason, the moral law “forces itself upon us of itself as a synthetic a priori proposition” (5:31), and as a synthetic a priori proposition, the moral law is a superior cognition. This is important because, as we have seen, Kant believes that inferior and superior feelings and desires are ultimately distinguished on the basis of the kind of cognition that gives rise to them.

The role played by the faculty of desire in moral action is the same as in the case of non-moral action: both cases involve actively causing the reality of an object, that is, active desiring. Again, the desire at issue in the case of moral action is, however, a superior desire because its ultimate source is a rational and therefore superior cognition. Thus, the second Critique maintains that “reason [is] a true superior faculty of desire” only when “reason of itself (not in the service of the inclinations) determines the will” (5:25, translation modified).

Kant distinguishes between the desire at play in non-moral versus moral action in another way, namely by distinguishing between two types of interest:

[T]he human will can take an interest in something without therefore acting from interest. The first signifies the practical interest in the action, the second the pathological interest in the object of the action. […] In the first case the action interests me, in the second the object of the action (in so far as it is agreeable to me). (4:413–414n)

When I act from interest, I do not desire the action itself, but “the object of the action.” Thus, in these cases I act in a particular way only in order to attain a particular object. As Kant indicates in the passage above, when we act from interest, this object is pleasure or agreeableness.^[39] In the case of taking an interest in an action directly, however, I do not act in a particular way to secure an object such as pleasure. Rather I desire to act that way as an end in itself. Kant is therefore clear that non-moral and moral action differ both in terms of the type of cognition that is their source and in terms of the object they seek to realize.

What is less clear, and heavily contested in the literature, is the relation between cognition and desire and the role that feeling plays in moral action. As a brief reminder of what was discussed in section 2, some commentators take the roles of feeling and desire in moral action to be functionally equivalent, while others are careful to distinguish feeling and desire but they do not specify the way in which they relate to one another. The first thing to note in this regard is that Kant argues in numerous places that there is no necessary connection between the faculties of cognition and desire and, thus, that our rational awareness of the moral law does not necessarily entail our desiring and, hence, acting accordingly. This is implied by Kant’s definition of an incentive as “the subjective determining ground of the will of a being whose reason does not by its nature necessarily conform with the objective law” (5:72, my emphasis). Kant here signals an important way in which he diverges from both Wolff and Baumgarten: as we saw in the previous section, Baumgarten accepts a version of the thesis according to which human beings necessarily desire what they cognize to be best. For Kant, by contrast, it is only the divine or holy will that necessarily acts in accordance with its cognitive insight into the moral law.^[40] Finite beings like us, on the other hand, are characterized by “sensibility,” which is “an obstacle to practical reason” (5:76) in the sense that it often pushes us in a direction that is opposed to the moral law.

It is for this reason that the moral law needs to supply us with an incentive: for human beings, “the subjective constitution of its choice does not of itself accord with the objective law of practical reason; they presuppose a need to be impelled to activity by something because an internal obstacle is opposed to it” (5:79). Thus, it is in the context of explaining how cognition and desire relate to each other in the case of human willing, where no necessary connection exists between the two, that Kant introduces the concept of an incentive.

As previously mentioned, Kant refers to incentives as grounds of desire in the Groundwork (4:427), the second Critique (5:90), and the Prize Essay (2:300). This view is even more explicit in the Metaphysik Mrongovius notes from 1782–83, where Kant is said to have stated that the concept of an incentive provides the link between cognition and desire:

That which is the cause of desires is the impelling cause or incentive of the mind (elater animi). Now, if they [incentives] arose from sensibility then they are called stimuli and their effect is called desire aroused by stimulus or sensible desires. But if they originate from the understanding, then they are called motives, and their effect is called desire aroused by motives or intellectual desires. (29:895, translation modified)

Kant’s incorporation of Baumgarten’s terminology in this passage is striking. However, Kant is not merely repeating Baumgarten’s position. An important way in which Kant deviates from Baumgarten is by considering a third faculty, feeling, in addition to the two discussed by Baumgarten, namely cognition and desire. Furthermore, Kant assigns the role of an incentive to feeling and not cognition, as is the case for Baumgarten. Consider the following passage where, just a few pages later, Kant purportedly discusses the role of each faculty in the case of moral action specifically:

since we desire merely that which pleases us, pleasure is the cause of our desiring. But the cause of the pleasure is either sensibility or understanding […]. Understanding and reason give laws to the will, to which it must conform if it is to be free. But we cannot be determined by mere representations of reason; it must also give us incentives.^[41]

In this passage, Kant echoes the view from his published works discussed in the previous paragraph, according to which incentives are needed to account for the lack of a necessary connection between cognition and desire in beings like us who possess sensibility. He highlights that cognitions cause feelings, feelings in turn ground desires, and that insofar as feelings ground desires they should be called incentives. We have already seen how, in non-moral action, sensible representations cause feelings of (sensible) pleasure and pain, which then cause (sensible) desire. Although the structure is the same, the case of moral action is special because the feeling of respect “is not one received by influence, but one self-wrought by a rational concept” (4:401n). Thus, even though cognition and desire are connected by means of feeling in moral action as well, this case is unique because it starts with our rational awareness of the moral law (a superior cognition). This cognition is then “inseparably” (5:80) connected, that is, necessarily causes, the feeling of respect in every finite rational being, which then functions as “the sole and also the undoubted moral incentive” (Triebfeder) (5:78).^[42]

I am now in a position to answer the question that has guided my discussion: Kant turns to feeling when his aim is to explain what happens in the mind when pure reason is practical because feeling functions as the ground of desire in his psychology of action. Thus, the feeling of respect, as an incentive, is meant to play the role of the ground of desire in action motivated by the moral law alone. Kant’s psychology of moral motivation might therefore be reconstructed as follows: our a priori consciousness of the moral law (a superior cognition) necessarily causes us to experience the feeling of respect (a superior feeling), which in turn grounds the (superior) desire to act morally even though we do not necessarily act in accordance with this desire.^[43]

Thus far I have suggested that feelings function as incentives in the sense that they are the grounds of desire. What I have not yet clarified is the specific way in which feeling grounds desire in the case of moral action. As I illustrated in section 2, Frierson argues that, seen from the point of view of empirical psychology, feelings function as efficient causes that necessarily bring about desire; this is a view in line with psychological determinism. Yet from the perspective of pure practical reason, we necessarily conceive of ourselves as free, which means that we need a way of conceiving of the causal relation between feeling and desire that preserves a role for freedom.

Utilizing the background of Kant’s concept of an incentive in Baumgarten’s Metaphysics, I would like to propose an alternative way in which feeling grounds desire in Kant’s psychology of action, namely in terms of ‘impelling causation.’ As we have seen, an impelling cause, for Baumgarten, differs from an efficient cause in that it does not necessarily posit its effect. Kant’s language in a number of places suggests that he shared Baumgarten’s view. Consider the picture provided in the Metaphysik L1 lectures from the late 1770s, where Kant reportedly says that “[e]very act of choice <actus arbitrii> has an impelling cause <causam impulsivam>” and impelling causes can be either sensible or intellectual (Metaphysik L1 [1777–80] 28:254). This is where he distinguishes, like Baumgarten, between sensible and intellectual impelling causes in terms of “stimuli <stimuli>“ and “motives or motive grounds” (Metaphysik L1 [1777–80] 28:254). Furthermore, when Kant allegedly talks about impelling causes in this text, he specifies that they do not necessitate their effects; he states, for instance, that “[t]his driving power of the power of choice can either necessitate, or by itself it can also only impel (impelliren)” (Metaphysik L1 [1777–80] 28:255). In the Metaphysics of Morals this is the basis on which Kant distinguishes between animal choice and free, human choice:

The faculty of choice (Die Willkür) that can be determined by pure reason is called the free faculty of choice (die freie Willkür). The faculty of choice that can be determined only by inclination (sensible impulse, stimulus) would be the animal faculty of choice (thierische Willkür) (arbitrium brutum). The human faculty of choice (Die menschliche Willkür), however, is one that can indeed be affected (afficirt) but not determined (bestimmt) by impulses (Antriebe) and is therefore of itself […] not pure but can still be determined to actions by pure will. Freedom of the faculty of choice (Die Freiheit der Willkür) is that independence from being determined by sensible impulses.^[44]

Again, aside from the fact that Kant utilizes much of Baumgarten’s terminology here, the point is that animals have a “brute” power of choice because they are “determined by stimuli,” whereas in the case of human beings the “stimuli can never determine it, but rather merely affect (afficiren) it sensibly” (Metaphysik Arnoldt [1794–95] 29:1015).

Kant uses the language of being impelled to activity in his published writings as well. For example, in a passage from the second Critique already mentioned, Kant argues that for human beings “the subjective constitution of its choice does not of itself accord with the objective law of practical reason; they presuppose a need to be impelled (angetrieben) to activity by something because an internal obstacle is opposed to it” (5:79, my emphasis; see also 4:422). Thus, although Kant does not make this feature of his psychology of moral motivation explicit, when he conceives of feelings as incentives in the sense of grounds of desire there is reason to believe that he thinks of the type of causality at work here in largely the same way as Baumgarten, namely as functioning in terms of impelling rather than efficient causation.

Whether Kant makes this aspect of his view explicit or not, there are also good interpretive reasons for thinking he conceived of the relation between feeling and desire in terms of impelling causation. Doing so makes room for the kind of freedom Kant sought to safeguard. In fact, the only place for free choice in Kant’s account of moral motivation seems to be between feeling and desire. Since our consciousness of the moral law is a ‘fact of reason’ in the sense that “we become immediately conscious” of the moral law “as soon as we draw up maxims of the will for ourselves” (5:29), and since the feeling of respect follows necessarily from our awareness of the moral law (see, e. g., 5:77), if the feeling of respect were to necessarily cause desire, then cognition of the moral law would determine us to act accordingly.

As we know, however, on Kant’s view “even the most hardened scoundrel” (4:454) is aware of the moral law but nonetheless can act otherwise. I propose that it is by assigning the role of an incentive to feeling rather than cognition that Kant is able to escape the intellectual determinism of Baumgarten’s view: although for both Baumgarten and Kant individual impelling causes (incentives) do not determine choice, for Baumgarten choice is determined by what we cognize to be best, that is, what has more impelling causes than alternative options. For Kant, however, even the recognition of what is unconditionally good in the moral law cannot determine us to act in accordance with such a cognition unless we choose to do so; the most such a cognition can do on its own is impel us to act accordingly, that is, provide us with an incentive to act morally, which it does by causing us to feel respect.

It is therefore no accident that Kant’s famous ‘incorporation thesis’ locates free choice precisely at the level of incentives. As Kant says in the Religion:

freedom of the power of choice has the characteristic, entirely peculiar to it, that it cannot be determined to action through any incentive except insofar as the human being has incorporated it into his maxim (has made it into a universal rule, according to which he wills to conduct himself). (6:24)

This means that in order for incentives, that is, feelings, to bring about their effects, which are desires, we must choose to act on them. On their own, feelings, of pleasure and pain especially but the feeling of respect as well, are incentives in the sense that they ground desire with impelling rather than efficient causation.

5 Advantages, Objections, and Implications

I consider the interpretation I have offered in the previous sections to have a number of advantages over the affectivist interpretations discussed in section 2. First, it makes a meaningful distinction between feeling and desire according to which they each play an important but distinct role in Kant’s psychology of moral action. Second, my interpretation can account for Kant’s active sense of desire, according to which desire only takes place after we have chosen to act based on grounds of desire, namely feelings. Third, unlike Frierson’s deterministic account of moral action, my interpretation assigns a role to freedom at the place where Kant locates it, namely at the level of incentives for action. In this final section, I wish to address a possible objection to the reading I have offered as well as an implication that concerns Kant’s distinction between subjective and objective determining grounds of the will.

First, the possible objection: Kant states explicitly that the feeling of respect is neither a feeling of pleasure, nor a feeling of pain (see 5:77). This might create problems for conceiving of the feeling of respect as the ground of desire: although it is easy to see how a feeling of pleasure might ground a desire, it is not so easy to see how a feeling that is neither pleasure nor pain could do the same. There is, however, textual support for my claim that not only feelings of pleasure and pain function as the ground of desire in the context of action. Kant occasionally calls the faculty of feeling the faculty of ‘satisfaction and dissatisfaction’ rather than the faculty of pleasure and displeasure, seemingly to reflect the fact that not all feelings are about pleasure and pain. Take for example the following passage from Metaphysik Mrongovius [1782–83]:

The cognitive faculty is connected with the faculty of desire by the feeling of pleasure or displeasure. The author [Baumgarten] calls it pleasure <voluptas> and displeasure <taedium>. That is false, for this is true only of sensible satisfaction. – For the understanding can frequently find dissatisfaction with that which best satisfies the senses. This should be named ‘the faculty of satisfaction and dissatisfaction.’^[45]

Kant thus acknowledges that feelings of satisfaction or dissatisfaction more generally function as the ground of desire. This helps explain why Kant calls the feeling of respect an ‘incentive’ even though it is neither a feeling of pleasure nor displeasure. Indeed, and as mentioned in the preceding section, Kant in various works makes a distinction between two kinds of incentive based on whether the feeling that grounds desire has its source in either a superior or an inferior cognition: in addition to the passage from the Groundwork where Kant refers to incentives “from the empirical field” (4:411) and the incentive of duty (see 4:412), in the second Critique Kant contrasts locating the incentive “pathologically (in sympathy or self-love)” with locating it “morally (in the law)” (5:85). Furthermore, in the Religion he distinguishes between cases where “the law alone” is the “sufficient incentive” (6:30) and cases where “incentives other than the law itself (e. g., ambition, self-love in general, yes, even a kindly instinct such as sympathy)” (6:30–31) are additionally required. In the context of moral action, even those feelings that belong to the superior faculty of feeling, such as respect, act as grounds of desire even though they are not feelings of pleasure or pain.

I now turn to an implication of my interpretation: conceiving of Kant’s concept of an incentive as referring to the ground of desire provides us with a way of reading his definitions of an incentive in both the Groundwork and the second Critique as consistent rather than at odds with one another, as a number of scholars have claimed. When defining the concept of an incentive in the Groundwork, Kant says the following:

The subjective ground of desiring is the incentive, the objective ground of willing the motivating ground (Bewegungsgrund), hence the difference between subjective ends, which rest on incentives, and objective ones, which depend on motivating grounds that hold for every rational being. (4:427)

In the second Critique, Kant also distinguishes between “the objective determining ground of the will” (5:74), which he claims is the moral law (see, e. g., 5:72), and the “subjective determining ground of the will,” which is the feeling of respect (see 5:75). Yet he claims that both the moral law and the feeling of respect are incentives (see, e. g., 5:72). This suggests that whereas the Groundwork reserves the concept of an incentive for subjective determining grounds of the will only, the second Critique treats both subjective and objective grounds of the will as incentives. Accordingly, Henry Allison has argued that Kant defines an incentive “differently in the two works” such that in the Groundwork “Kant contrasts ‘Triebfeder’ and ‘Bewegungsgrund’” (Allison 2011, 96). Similarly, Beck has argued that “Kant carefully distinguishes between Triebfeder and Bewegungsgrund (= ‘motive’) in the [Groundwork],” whereas in the second Critique Kant “is using Triebfeder in the sense in which the [Groundwork] defined Bewegungsgrund” (Beck 1960, 91). Timmermann also suggests that the alleged difference between the two terms, informed by their etymology, is that, in the Groundwork at least, “Triebfeder is a mechanical term. It designates the ‘spring’ of motion, as in a clock or an old-fashioned toy. In psychology, it is by extension a motivating desire, the force that propels an agent forward if he or she so chooses” (Timmermann 2011, 180). A Bewegungsgrund, by contrast, “is something static. It is the object that prompts the mechanism of motivation to action” (Timmermann 2011, 181).

If an incentive is considered the ground of desire, as I have done in this paper, then there is a way to make Kant’s view consistent across these two works. On the one hand, in the case of pure practical reason, our cognitive ground or reason for acting, objectively speaking, that is, the reason that can be shared by all rational beings, is simply the moral law. This is why Kant calls the moral law the “objective determining ground of the will” (5:74). On the other hand, the ground of desire from the perspective of what subjectively happens in the mind, that is, in the sense of the cause of our desire, is the feeling of respect, which is why Kant calls respect the “subjective determining ground” of the will (5:75, my emphasis). Indeed, the feeling of respect is the subjective cause of desire because, as we have seen, all feelings relate only to the subject.^[46] Accordingly, Kant claims in the second Critique that the moral law, although strictly speaking

a formal determining ground of action through practical pure reason […] is also a subjective determining ground – that is, an incentive – to this action inasmuch as it has influence on the sensibility of the subject and effects a feeling conducive to the influence of the law upon the will. (5:75, my emphasis; see also 5:85–86)

On my reading, Kant’s view in the Groundwork is no different: a Bewegungsgrund in that text is an incentive too in the sense of a ground of desire, but it refers more specifically to our objective ground of willing rather than our subjective ground of desire. This is because, in the context of moral action, the Bewegungsgrund is the objective cognition of the moral law at the basis of the subjective feeling of respect as incentive. I therefore claim that it makes sense for Kant to call both the objective and subjective determining grounds of the will ‘incentives’ for two reasons: first, because Kant distinguishes between cognition and feeling whereas Baumgarten does not, and second, because incentives are feelings that always follow from cognitions. Once we have the background of Kant’s view in Baumgarten in hand, and the way in which they differ from one other, it is therefore easier to make sense of what at first glance appears to be a change in opinion across Kant’s works.

6 Conclusion

In this paper, I have outlined an underappreciated historical source of Kant’s concept of an incentive in Baumgarten’s Metaphysics, according to which the term refers to the ground of desire, and more specifically to the impelling cause of desire. Against this background, I offered an interpretation of Kant’s psychology of moral action, according to which the feeling of respect functions as the impelling cause of desire in moral action, which I have suggested has several advantages over the existing options. A fuller picture of how choice works on Kant’s view would have to consider not only how “the removal of the counterweight” of self-love increases “the relative weightiness of the law” (5:76), but also how choice works differently in each of the “three different grades” of our natural propensity to evil that Kant describes in the Religion (5:29–30). Moreover, if my interpretation of Kant’s concept of an incentive is correct, then I take it that feelings are the impelling causes of desire in the case of free, non-moral action as well. However, since these topics involve controversial issues such as whether free non-moral and immoral action are even possible on Kant’s account, I leave them to another occasion.