On Kant’s Janus-Faced Transcendental and Empirical Conception of the Human Being

ABSTRACT

There has been increased attention to the empirical and naturalistic dimensions of Kant’s philosophy in recent decades, across both his theoretical and practical philosophy. Anik Waldow’s impressively wide-ranging and carefully argued book, Experience Embodied: Early Modern Accounts of the Human Place in Nature (Waldow 2020), clearly demonstrates the fruits of this reoriented focus, not only in the case of Kant, but also in all the embodied agency-oriented conceptions of experience that she brings to light across the early modern period in the thought of Descartes, Locke, Hume, Rousseau, Herder, and Kant. Here, I focus on Waldow’s analysis of the a priori and empirical dimensions of Kant’s conception of the human agent. In particular, I set that analysis within the wider context of Kant’s multi-levelled and presuppositional conception of the sciences, both theoretical and practical, and consider some of the difficult questions that arise concerning Kant’s transcendental idealist conception of our freedom within nature.

KEYWORDS:

• Kant
• Waldow
• transcendental/empirical
• freedom
• agency
• embodiment

Recent decades have seen increased attention to the empirical and naturalistic dimensions of Kant’s philosophy, across both his theoretical and practical philosophy. Anik Waldow’s impressively wide-ranging and carefully argued book, Experience Embodied: Early Modern Accounts of the Human Place in Nature (Waldow 2020), clearly demonstrates the fruits of this reoriented focus, not only in the case of Kant (in her final chapter on ‘Diversifying Method: Kant’s Janus-Faced Conception of the Human Being’), but in all the embodied agency-oriented conceptions of experience that she brings to light across the early modern period in the thought of Descartes, Locke, Hume, Rousseau, Herder, and Kant. Here, I focus on Waldow’s analysis of the a priori and empirical dimensions of Kant’s conception of the human agent, and for this I think some background might be helpful.

Already during his ‘pre-critical’ period and then throughout his ‘critical’ and ‘transcendental’ philosophy, Kant framed the most basic aims of philosophy in terms of the dual achievement of a fundamental metaphysics of nature on the one hand, and a metaphysics of morals on the other. From the Critique of Pure Reason onwards, however, Kant held that both of those a priori metaphysical sciences must themselves be grounded in a transcendental ‘critique’ of the cognitive faculties at issue, one that explains the possibility and limits of the relevant kinds of pure a priori cognition in particular.¹ The resolutely non-empirical principles of Kant’s famous three Critiques (of pure reason, practical reason, and judgment) each in their own way possess an objective validity that is supposed to be demonstrable independently of any particular empirical laws or empirical aspects of human nature, however firmly established by experience the latter may be. As a result, a highly complex, systematic hierarchy of the sciences took shape throughout Kant’s critical period, roughly depending on the degree of methodological abstraction from the relevant empirical aspects of our scientific knowledge and of our practical agency.

Very crudely sketched (all the relationships among these ‘levels’ are subject to divergent interpretations), and proceeding from the maximally abstract and non-empirical to the increasingly empirical, we have: (1) the first and second Critiques, and then ‘under’ them, correspondingly, (2) Kant’s two a priori metaphysics of nature and metaphysics of morals. Each of the latter domains is constituted by empirically enriching or re-embedding certain highly generic empirical principles from which the two Critiques had each sought to abstract. For example, the metaphysics of morals (the doctrine of right and the doctrine of virtue) in some sense presupposes and ‘applies’ the a priori principles of the second Critique to specifically human nature and feelings, generating more specific a priori principles pertaining to the specified empirical domain. In the theoretical domain an example would be that the first Critique’s principle that every alteration has some necessitating cause becomes the more specific principle that every change in matter-as-the-movable has an ‘external’ (i.e. ‘lifeless’) cause (cf. Kant MFNS 4:543;CJ 5:181). In addition, (3) in the first Critique’s ‘Appendix to the Dialectic’ and in the third Critique (of judgment), Kant argues that certain more indeterminate yet a priori regulative principles of reason and reflecting judgment are necessary for the possibility of systematic experience and empirical science. Two examples would be that any possible empirically cognizable nature must exhibit some degree and kind of empirical uniformities or other; and some kinds of experienced beings we ‘find’ must be judged to be ‘organized beings’, i.e. living things. Finally, (4) at the properly empirical level of the theoretical and practical sciences themselves, we have, for example, the sciences of physics and anthropology, respectively. It took the specific empirical and mathematical hypotheses and discoveries of Newton and Kepler to give us physics as we know it, though on Kant’s view it requires an a priori metaphysics of nature to explain the possibility and limits of the a priori principles that physics presupposes. What about anthropology on the practical side?

The latter is, in effect, the central question that Waldow sheds light on in her aptly titled examination of Kant’s Janus-faced a priori and empirical conception of the human being. Waldow’s analysis, understandably and probably wisely given the limits of a single chapter (or two: chapter 6 is on Herder and Kant on history), tends to move back and forth directly from Kant’s most abstract, pure a priori ‘critical’ level (1), straight to the properly empirical level of anthropology as a science (4). Consequently, we do not find many comments on the complex systematic relationships between the transcendental and empirical levels of experience that are involved in Kant’s moving from the pure a priori transcendental Critiques (1) to the ‘mixed’ a priori and empirical metaphysics of nature and morals (2), or in the relationships between either of those and (3) the various regulative maxims of reason and reflection that govern the formation of empirical concepts as well as important aspects of our ideas of history and religion, or the relation between (3) and the properly empirical sciences (4). But I will suggest that the important structural relationships between the transcendental and empirical dimensions of Kant’s philosophy on which Waldow does not focus can integrate smoothly with, and perhaps provide additional support for, her interpretation of the pivotal relationships and distinctions that she does seek to illuminate.

Waldow provides a helpful brief discussion of how Kant’s anthropology lectures developed over his career, culminating in his 1798 Anthropology from a Pragmatic Point of View (hereafter, the Anthropology). In the theoretical domain, as briefly sketched above, there is a sense in which for Kant physics proper presupposes an a priori metaphysics of nature that grounds the intelligibility of its mixed a priori and empirical subject matters (for example, demonstrating certain a priori principles concerning our concepts of relative and absolute motion in space in general); and the latter metaphysics of nature is itself a specification (to the domain of motion) of the pure a priori transcendental principles of the first Critique concerning any cognizable nature for us. If the parallel holds in the practical domain, anthropology would be an empirical science of the human being that presupposes the a priori principles of Kant’s Metaphysics of Morals (Kant 1995), which are themselves ‘mixed’ specifications, to empirical human nature in general, of the transcendental principles of pure practical philosophy that derive from the pure a priori categorical imperative. However, just as it took the specific mathematical discoveries and empirical hypotheses of Kepler and Newton et al. to discover the laws of physics proper, so anthropology would have its own distinctive empirical methods and conceptions, revealing aspects of human experience and agency that are not derivable from a metaphysics of morals, but which nonetheless presuppose the validity of the latter a priori principles of rational human practical agency in general.

I think that essentially the above is what we find supported by a different route in Waldow’s account of the much-debated relationships between Kant’s Anthropology and the a priori principles of his Critiques, in which she shows how multiple interpretive hurdles must be cleared if we are to understand what is involved in Kant’s overall conception. At the outset she brings out the well-known passages in both the Critique of Pure Reason (B833) and in his Jäsche logic lectures (Logik) in which Kant claims that ‘philosophy in the cosmopolitan sense can be brought down to the following questions: 1. What can I know? [metaphysics], 2. What ought I to do? [morals], 3. What may I hope? [religion], and 4. What is man? [anthropology]’ (Logik, 9:25). Kant continues: ‘Fundamentally, however, we could reckon all of this as anthropology, because the first three questions relate to the last one’ (Logik, 9:25). But Waldow brings out convincingly that Kant regarded his newly founded science of anthropology as an observational science of the empirical nature of the human being. The latter had hitherto been neglected, Kant suggests in his lectures on anthropology (hereafter Anthr.Lec.), because ‘one considered the science of human beings as a dependent part of metaphysics,’ and thus ‘applied only as much attention to it as the larger parts of metaphysics permitted’ (Anthr.Lec. 25:8). As Waldow notes, Kant describes his observation-based anthropology as akin to the ‘empirical psychology’ of the ‘English Authors,’ and ‘empirical psychology belongs to metaphysics just as little as empirical physics does’ (Anthr.Lec. 25:8).

I agree with Waldow’s main thesis, which is based on many such passages as the above, that Kant is defending the methodological autonomy of anthropology as an observational empirical science. But does this threaten the picture of Kant’s ‘layered’ or presuppositional structure of the sciences as I sketched it above? The latter was such that in the theoretical domain, ‘empirical physics’ (compare anthropology in the practical domain) presupposes or is dependent for its intelligibility on a mixed a priori/empirical metaphysics of nature, which itself presupposes the purely non-empirical transcendental ‘critique’ of pure reason for its possibility and intelligibility. Despite the interpretive difficulties involved, one can see how Kant’s Metaphysical Foundations of Natural Science (MFNS) was broadly intended by Kant to elucidate empirically enriched instantiations of the pure a priori principles of understanding in the first Critique (Kant says as much at MFNS 4:541–2). But the relationship between both the a priori Critique and the mixed metaphysical science of nature, on the one hand, and the hypothesis- and observation-dependent aspects of Newton’s own ‘empirical physics’, on the other, is arguably a more complex matter. For however one slices it, physics proper, in the respects in which it generates discoveries about nature, is presumably intended on Kant’s picture to be ‘methodologically autonomous’ from the a priori metaphysics of the science of nature in roughly the observation-dependent respects that make such discoveries possible. At the same time, however, as Kant sees it physics also presupposes for its ultimate intelligibility the enriched a priori framework that the more abstract metaphysics of nature elucidates.

It seems to me that Waldow is rightly attempting to display Kant’s anthropology as intended to have the latter sort of complex status in the practical domain. Thus, she is cautious about many interpretations in the literature that suggest that Kant’s ‘anthropology must be seen as Kantian ethics practically applied’ (Waldow 2020, 226).² ‘Kantian ethics applied’ might well seem to be the right sort of interpretation of the Anthropology, especially in light of the undeniable ‘pragmatic’-practical (e.g. pedagogical) dimensions of Kant’s anthropology that pertain, for example, to the means for individual human development and social improvement. These are aspects of Kant’s conception of embodied human experience and practical agency in the Anthropology that Waldow herself is centrally concerned to emphasize (e.g. on pedagogy see Waldow 2020, 235–238). But I think Waldow is right to resist the tempting idea that Kant’s anthropology is basically just applied Kantian ethics, just as one should hesitate before thinking that on Kant’s view physics is applied a priori metaphysics of nature. The challenge lies in how to clarify, nonetheless, that there is a strong dimension of presuppositional dependence of the properly empirical sciences on Kant’s a priori metaphysics of nature and of morals, along the general lines I sketched at the outset.

So, as I would put it: how does Waldow seek to clarify, in the case of Kant’s anthropology, the complex combination of methodological autonomy (as an observational science) yet presuppositional dependence (in relation to Kant’s a priori metaphysics of nature and of morals) that I think she rightly finds in Kant’s view? Her main strategy, I think, is to focus on Kant’s concept of character (and related concepts), first, as it figures empirically in Kant’s anthropology, and in particular as it is simultaneously shaped by the various moral dimensions of human development, culture, and history. One such stress in Kant’s anthropology that Waldow highlights takes the form of observationally based lessons concerning the cultivation and improvement of one’s character. And then, secondly, to examine the concept of character as it figures transcendentally both in his a priori conception of the will as pure practical reason in his moral philosophy, and in the first Critique’s keystone distinction between a person’s (and their actions’) ‘intelligible character’ qua free rational agent when considered from the pure practical point of view of morality, and their ‘empirical character’ as a causally determined denizen of nature (B566–85). Waldow thus nicely frames her general approach to resolving the central interpretive tension at issue in relation to Kant’s anthropology this way:

Thus, it is only because he believes that humans perceive themselves as constrained by nature that there can be room for an empirical anthropology that traces the determining influences of nature on human character formation. And conversely, only because for Kant human existence is fundamentally marked by the possibility to actualize freedom, does it make sense for him to offer guidance [for example, in anthropology] on how we can pragmatically realize this task. (Waldow 2020, 227)

The focus on character enables Waldow to bring together these two different aspects of Kant’s view of our naturally embodied yet rational human agency in a way that I found particularly helpful (and which could also form the basis of a more extended treatment that would rightly focus on the central concept of character in Kant’s metaphysics of morals itself):

Kant’s idea is that having character is an achievement that becomes possible through the use of reason. This use renders causally efficient the laws of freedom in a world governed by the laws of cause and effect. Kant’s concept of character thus brings together what his methodological dualism leaves separate, and thereby helps us to understand what it means to study the human being in its empirical situation. (Waldow 2020, 238)

I found Waldow’s core analysis of this complex region of Kant’s thought to be insightful and clear, and I certainly highly recommend it to anyone interested in both the methodological structure and practical import of Kant’s philosophy. I will end, however, by noting some particular aspects of its execution about which I had some questions or suggestions to make.

Waldow faces up to the daunting task of having to explicate Kant’s notoriously difficult conception of free human agency in a very short space. In tackling Kant’s conception of maxims as subjective principles of the will, for example, Waldow writes:

… unless the faculty of choice counts as fully determined by practical reason (Kant speaks of ‘complete control’ at [Groundwork] 4:401), it does not count as a manifestation of the will [as practical reason] … Actions initiated by the faculty of choice, and which do not manifest the will in this particular sense (i.e. they are affected by impulse), therefore seem to involve at least some determinations grounded in our way of sensing – that is, the influences of the many inclinations and affective dispositions that tend to govern our conduct. If they did not, these actions would be fully determined by practical reason and replace our commitment to maxims through a commitment to the will. (Waldow 2020, 240–241)

Waldow here and in other passages seems to use Kant’s term ‘maxims’ to refer only to subjective principles of action that would not pass the test of the categorical imperative. As she frames the matter above, for instance, if our actions are affected by sensible inclination they are thus not ‘fully determined by [the will as pure] practical reason,’ for if they were thus ‘completely’ determined by reason, ‘these actions … would replace our commitment to maxims through a commitment to the will’ (Waldow 2020, 240–241, emphases added). Similarly, Waldow writes that ‘actions that are determined by our maxims rather than practical reason, are not governed by the laws of morality and freedom … ’ (Waldow 2020, 243, emphases added).

This strikes me as a non-standard way of understanding Kant’s conception of a maxim. Kant’s examples are indeed usually of failures of particular maxims to pass the universalization test. But when we act from duty it is again our particular maxim as a subjective principle of the will that in this case passes or would pass the test. Human beings always act on end-directed maxims, as I understand Kant’s view, and always under conditions of affection by inclinations. From the pure practical standpoint of our agency, sometimes we act freely in accordance with the moral law as pure practical reason, and sometimes we act freely but immorally. I recognize that there are well-known interpretive disputes surrounding Kant’s view of freely chosen evil actions (i.e. morally bad actions), but I don’t think Kant need or ought to be interpreted as denying or radically reinterpreting the basic aspects of our agency just mentioned. I would think that they should also be essential to Waldow’s view of our free yet embodied agency, but her discussions of Kant on maxims and choice leave me worried that she might be presupposing a reading of Kant on these matters that threatens to render free immoral action more mysterious on Kant’s view than it really is. Whether this is true in Waldow’s case I am not sure. (In this I am not intending to question or set aside the idea that Kant has an ontology of ‘things in themselves’ and ‘noumenal affection’ that underwrites his account of freedom and morality.)

In a related way, Waldow writes that the free intelligible character, like transcendental freedom itself, ‘emerges as an uncaused cause that, in order to show itself in the world of cause and effect, has to appear in such a way that it presents itself in accordance with the laws of nature. When this happens, the intelligible character becomes the empirical character that manifests itself under the conditions of sensibility’ (Waldow 2020, 242, emphasis added). Again I find this an awkward way of conveying what I take to be Kant’s ‘dual aspect’ conception of our practical agency, as Waldow appropriately characterizes it. Waldow’s picture seems in places to suggest a more strongly bifurcated reading of Kant on agency than it arguably need receive, according to which (I at least at some points worried) the properly moral dimension of our agency would not really be both a free and yet fully embodied human agency at its core. The picture suggested is as follows. (a) Our character and our actions are intelligible from the practical standpoint as actions that would be completely determined by pure practical reason (and thus free). (b) However, in the actual human arena of action, these become actions that are ‘rather’ determined by our all too human ‘maxims’ of action. Thus, (c) ‘the intelligible character’ thereby ‘becomes the empirical character that manifests itself under the conditions of sensibility,’ which are (unfortunately) the ineliminable condition of the embodied human being. When this happens, as Waldow puts it (Waldow 2020, 244), ‘the intelligible character, which is given to everyone qua possession of reason, fails to make its appearance in the empirical character.’ This sort of complete character takeover (as it were) again seems an awkward and non-compulsory way of characterizing Kant’s view of the embodied and deliberating human agent as capable in any given situation of having under consideration both the thought of their moral duty and a contrary inclination to do something else, both of them possible maxims of their action (as I understand Kant on ‘maxims’), in what would normally be thought of as the human condition. The practical perspective of our intelligible moral character, in which, for example, we conceive ourselves and others as having been responsibly able to do otherwise than whatever we did do in a given context, can remain just as conceptually irreducible and scientifically inexplicable from the empirical perspective as it indeed is on Kant’s view, without the more complete sundering of the two aspects of our human agency that Waldow’s non-compulsory characterizations often seem to suggest to me.

Waldow does directly address the fact on Kant’s view human beings do ‘have the power to give their actions a reason-guided, willful direction that brings with it the possibility of complying with the laws of nature while enacting the moral laws’ (Waldow 2020, 244). However again the glosses that she gives to Kant’s conception of this fundamental fact of our agency gave me pause (though admittedly this is a deeply difficult topic on anyone’s view). Waldow’s way of characterizing what would be morally good action if reason had complete control is that in such cases ‘humans let their actions accord with the moral laws’ (Waldow 2020, 244). Why use ‘let’ to describe this aspect of our agency, when the relevant thought and motivation are produced by one’s own active reason? Waldow in this context adds in an accompanying footnote (citing Jacobs 2003):

Jacobs stresses that following the demands of morality means distancing oneself from the natural course of events; he takes this to show that the intelligible character is a negative concept that describes what ought to have happened and ‘stipulates what indeed would have happened if the natural order consisted entirely of beings acting solely according rational principles’ (Jacobs 2003, 121). (Waldow 2020, 244)

Again, this passage seems to me to reinforce the thought that on Waldow’s (and Jacob’s) reading of Kant on agency, actual empirically embodied free human actions cannot be positively conceived as in some cases based on morally good universalizable maxims as products of one’s own pure practical reason, and in other cases based on maxims that cannot be universalized. Of course Kant does stress that one can never be certain that one’s own action or anyone else’s action really has been so motivated from duty; but this point simply reinforces the idea that in his view there need be nothing essentially counterfactual or merely negative-in-concept about the possibility of ordinary empirically embodied right actions that have genuine moral worth.

Suppose Tom repays a loan back to Tina as promised. This action like every free human action (morally good or bad) is on Kant’s view subject to an intelligible characterization in terms of his having so acted, in this case, because it was the right thing to do, though he was tempted to and could have done otherwise. On Kant’s view, the same embodied free action of Tom’s is subject to empirical characterizations at various levels of science, including psychology, according to which the very same chain of thought and behavior is conceived empirically as causally determined in terms of the relevant laws of nature. (In this I am not attempting to ‘deflate’ Kant’s ‘characters’ to ‘characterizations’.) On Kant’s view nothing in nature escapes the net of scientific naturalism, but early on it was the idea of the Antinomies that Kant reports as having first awakened him from his dogmatic slumbers to the realization that neither the scientific naturalist’s nor the rational theist’s image of the world can be sustained as totality claims, thus making room (as Kant sees it) for at least the coherent conceivability of the above Janus-faced ‘dual character’ conception of our embodied human agency. Whatever one’s interpretation of Kant’s transcendental idealism might be, whether more deflationary or more metaphysical, in places it seemed to me that Waldow’s characterization of Kant’s conception of agency made matters harder for her own valuable insights into its empirically embodied nature.

One last remark on this excellent book – and I should stress again that here I have only touched on its final chapter on Kant, ignoring the wealth of insights in the rest of the book on the early modern conception of experience as centrally concerned with our embodied agency in the world. Recall Kant’s layered, presuppositional conception of the theoretical and practical sciences: starting from the officially purely non-empirical level of ‘transcendental critique’, to the ‘mixed’ a priori/empirical metaphysics of nature (motion) and of morals (human nature), to the various observation-based empirical sciences such as physics and anthropology. A passage from the Metaphysics of Morals to which we saw Waldow briefly refer (Waldow 2020, 226) is worth quoting more fully:

[J]ust as there must be principles in a metaphysics of nature for applying those highest universal principles of a nature in general to objects of experience, a metaphysics of morals cannot dispense with principles of application, and we shall often have to take as our object the particular nature of human beings, which is cognized only by experience, in order to show in it what can be inferred from universal moral principles. But this will in no way detract from the purity of these principles or cast doubt on their a priori source. This is to say, in effect, that a metaphysics of morals cannot be based upon anthropology but can still be applied to it.

The counterpart of a metaphysics of morals, the other member of the division of practical philosophy as a whole, would be moral anthropology, which, however, would deal only with the subjective conditions in human nature that hinder people or help them in fulfilling the laws of a metaphysics of morals. It would deal with the development, spreading, and strengthening of moral principles (in education in schools and in popular instruction), and with other similar teachings and precepts based on experience. (Kant, MM 6:216–17)

What is implied in this passage, and would I think be consistent with Waldow’s analysis of Kant’s anthropology as a methodologically observational science that nonetheless presupposes his metaphysics of moral agency, is the following idea. There are two kinds of ‘application’ of Kant’s pure a priori metaphysics of morals to the empirical dimensions of our agency. The first occurs in his Metaphysics of Morals itself, in application to specific aspects of empirical human nature at a maximally high level of generality and abstraction (our natural capacity for specific kinds of moral feeling, for instance). This is analogous to the systematic ‘instantiation’ of the first Critique’s ‘permanent substance as matter in general’ to ‘permanent matter as the (lifeless) movable’ in particular, as analyzed in Kant’s carefully ‘mixed’ a priori/empirical metaphysics of nature (MFNS). Observational anthropology, for its part, as Waldow explains, makes use of but does not seek to analyze or problematize our free moral agency as being operative in nature to transformative empirical and social effect. But here in anthropology the ‘application’ of the a priori metaphysics of morals to human nature, as Kant’s remarks above suggest, is very different from the highly abstract a priori yet minimally empirically enriched systematization that takes place in the Metaphysics of Morals. It presupposes an independently rich source of empirical observations of human nature in action that anthropology provides. Spelling out just what the differences in method and consequence are between those two different domains of ‘application’, would, I think, only serve to strengthen and support Waldow’s main contentions in relation to the nature and place of anthropology in Kant’s critical philosophy.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. References to and translations of Kant’s work are to the appropriate volumes of the Cambridge Edition of the Works of Immanuel Kant, including the standard A/B references to the Critique of Pure Reason, and the marginal volume and page numbers of the standard Akademie edition for all of Kant’s other works. As indicated in the text, I use the abbreviations MFNS, CJ, Lect.Anthr., Anthropology, Groundwork, and MM to refer respectively to Kant’s Metaphysical Foundations of Natural Science, Critique of the Power of Judgment, Lectures on Anthropology, Anthropology from a Pragmatic Point of View, Groundwork for the Metaphysics of Morals, and the Metaphysics of Morals.

2. Waldow (2020, 226) cites for emphasis in this regard Munzel 1999, Jacobs 2003, Louden 2003 & 2011, and Wood 2003; however Waldow also notes the insightful contributions of these works, referring here also to Cohen’s 2009 ‘middle position’.