I.

Kant's struggles with theorizing the possibility and intelligibility of moral evil are notorious.Footnote ¹ Paul Guyer encapsulates the source of his difficulties as follows: ‘Kant holds the moral law to be the causal law of the noumenal self. But if the moral law is the causal law of the noumenal self, how could the noumenal self ever will an immoral action (that is, immoral maxim)?’ (Guyer 2009: 189).Footnote ² Amongst contemporary theorists of autonomy, those that understand autonomy in terms of the concept of self-constitution can at least seem to inherit a version of this problem: if the ground and validity of norms derive from their role in an agent's constituting herself as a rational agent, contraventions of those norms have a paradoxical quality. For, as failures to live up to the claims of agency, do they not thereby fail to be expressions of agency?

For a specific example, consider Christine Korsgaard's discussion of the normativity of hypothetical imperatives. A guiding motivation of her discussion is a dissatisfaction with Humean accounts of instrumental reasoning (Korsgaard 2008: 32–46). On such accounts, she alleges, instrumental principles really have the character of mere predictions of behaviour, thus compromising their status as genuine practical constraints: actions simply reveal the true desire of an agent, their effective ends or preferences, and thus trivially accord with the instrumental principle they could not but ‘obey’. Whilst this empiricist view may capture our ways of understanding behaviour from a third-personal viewpoint, it cannot do justice to the first-personal standpoint of agency: if I have truly set myself a certain end, I must recognize an obligation to pursue the means even if I in fact do not; and the stance from which this recognition is effective must be distinguished from a theoretical stance, which is restricted to determining the likelihood of some course of behaviour.

Korsgaard insists that there should be genuine failures to live up to instrumental principles, failures of practical, not merely theoretical, reason; more than just mistakes about which means are appropriate to antecedently given ends. As an example of practical irrationality, she mentions Howard, who is too scared to have a life-saving medical procedure due to his fear of injections (ibid. 39). Elsewhere, various characters’ fear of the drill has them avoiding the dentist's chair and timidity prevents avowed thrill-seekers from riding rollercoasters (ibid. 45; 88). Rejecting Humean explanations of such failures, she states that ‘[i]f we believe that the instrumental principle is a rational requirement, we will say that these people's terror, idleness, shyness, or depression is making them irrational and weak-willed, and so that they are failing to do what is necessary to promote their own ends’ (ibid. 41).

This statement, however, contains an apparent slide between different explanatory loci: it moves from talking about conative states making subjects irrational and weak-willed, to the subjects themselves failing. Such is alleged by Robert Pippin, who points to the difference between unforeseen emotional factors ‘controlling’ the will and ‘overruling’ it, two terms that Korsgaard uses synonymously. A case of some factor controlling the will represents not so much irrational action as a breakdown of agency as such, whilst if the factor is taken to overrule a course of action, an agent—rightly or wrongly—redirects their actions based upon a further reason. Korsgaard speaks of subjects ‘indulging’ ‘given’ conative states, but the nature of such acquiescence in a psychic given is somewhat mysterious. In a passage where acquiescence appears as permission or allowance, Korsgaard explains how:

Yet, Pippin complains, ‘strictly speaking, the idea that someone can “allow” themselves to be controlled, but is not thereby “actively willing” is incoherent’ (Pippin 2008: 83). There is an irony here: Korsgaard accuses the Humean position of implying the disappearance of the agent, since Hume lacks the resources for ‘distinguishing the activity of the person herself from the operation of beliefs, desires, and other forces in her’ (Korsgaard 2008: 45). Pippin's accusation, in effect, is that Korsgaard winds up fudging this very distinction.

Let's return to her example: Howard goes to the doctor, seems committed to remaining healthy and recognizes that he ought to undergo a procedure. Yet on seeing the needle, he becomes scared, has second thoughts, and puts off the operation (for the time being, he perhaps tells himself). One way of responding to a critique such as Pippin's would be to develop an account satisfying the following—admittedly highly abstract—desideratum: Howard's behaviour can be seen as his own, as his doing as a rational agent, by understanding his cowardly character as a function, rather than a disruption, of his self-constitution. Such an understanding evidently requires breaking with the dichotomy of active and passive sides of human nature that structures Korsgaard's theory, whereby active self-determination according to purely self-given principles is opposed to external influences on a passive susceptibility to inclination. In other words, it requires elaborating the Hegelian insight that when properly thought through, self-determination includes an element of other-determination. In what follows, I'll suggest how habit provides one small, if crucial, element in Hegel's attempts to develop just such an alternative.

II.

Korsgaard's examples are (or are at least presented as) one-off events: a subject adopts an end, is afflicted by a given conative state or episode, and is diverted from their end. To be sure, such cases of punctual failure are not unknown. Nevertheless, more often than not, we do not understand such acts in isolation, but as both expressive of and contributing to someone's character. Perhaps Howard's act comes as no surprise to those who know him, since it fits a pattern comprising other acts of cowardice in which he undermines his avowed ends. If so, we might best characterize their understanding of Howard as one on which his action, in being an expression of fear, is thereby an instance of self-expression.

Placing his behaviour in an expressive dimension requires striking a delicate balance; perceiving his reaction as expressive of his character presupposes a distinction between Howard as a mere site of various reactions, as it were, and the reactions being his reactions. On the one hand, an emphasis on self-expression is born of the intuition that preserving freedom means seeing Howard, rather than the emotional state of fear, as the relevant locus of responsibility or determining control. On the other hand, saying Howard to be responsible for the act in virtue of his character cannot distort the notion of character beyond recognition, especially by casting it as something to be made and remade at will. Further, satisfying both these requirements assumes a complex identity within difference of subject and character: expressions of character are expressions of the subject himself, and yet the subject is responsible for, and must exercise some control over, his character.

Of course, this is only to mention general conditions on a more satisfying form of action explanation. Nevertheless, placing actions within the context of a developing character is a first step in helping to shift the centre of determining control toward the agent and away from an external intervening state. Doing so, however, requires understanding particular expressions of character as genuine expressions of rational agency, without succumbing to familiar regresses and so resorting to positing ‘character’ as somehow radically self-generated.Footnote ³

To begin with, it helps to avoid a certain false binary in understanding free self-expression, which can be formulated in terms of Gary Watson's influential contrast between valuational and motivational systems. The valuational system results in judgments as to the thing to do by ranking different outcomes in light of what the individual values. What seems to mark out avowals of value, on Watson's account, is that they reflect the considered judgments of individuals, rather than their episodic desires, and thus have a systematic relation to the justifications they would offer and to which they would subscribe.Footnote ⁴ In Watson's words, ‘We might say that an agent's values consist in those principles and ends which he—in a cool and non-self-deceptive moment—articulates as definitive of the good, fulfilling, and defensible life’ (Watson 1975: 215). The motivational system, by contrast, is what moves the individual to act. Freedom consists in the translation of valuations into actions; in other words, it is a function of the extent of the ‘grip’ of the valuational on the motivational system. Watson notes that it would be something of a ‘fiction’ to see individuals as having a systematic and coherent conception of the good, and a complete coincidence of the two systems is accordingly equally fantastical. Nevertheless, we thus have a model on which the criterion of rational self-expression is provided by what an agent would do were they to act according to the deliverances of their practical reason, their considered evaluative judgments, where these are distinguished from their occurrent motivational states. Motivation by ‘desire’ is irrational insofar as such conative states are judgment-independent (ibid. 214).

The problem with this criterion of self-determination is not that the relevant ideal is impossible for finite creatures like us, but that a certain interpretation of it—one occasionally suggested by Watson—is questionable. The idea that deviations of the motivational system from one's true valuations represents an impairment of freedom may be plausible enough, but one can still ask what is criterial of one's real valuations, and in this respect Watson's reference to a ‘cool and non-self-deceptive moment’ is revealing. Likewise, he says that a

The implication is that episodic desires can create an appearance of an evaluation, the actual contents of valuation systems being avowed in cool moments.Footnote ⁵ Pippin's comments on Korsgaard's examples of weakness include a warning against such an implication: what misleads us, he says, is a Christian conception of strength and weakness of resolve and the idea that, in certain trying circumstances, weakness debilitates the decision-making powers of the true agent, identified as the ‘calmly reflective agent’. Adapted to our case: the clinical aura of the doctor's surgery, the sheer size of the needle, the necessity of a quick decision etc. all contribute to Howard's inability to decide as he would were he to make a fully informed and reasoned decision in less pressing circumstances. He is ‘irrational’ here, because he's not ‘himself’. For Pippin, this is ‘a bit of fantasy’ just as much as the common image of degrees of volitional strength (Pippin 2008: 83). Howard is no less the real Howard at the doctor's surgery than he is at home, and faced with the needle, we find out something about the real him, what really matters to him. We deceive ourselves if we think we identify the ‘true’ rational agent when we ask what someone would decide were the influences of emotion, desire, feeling and so on at a minimum.

We thus need to avoid a strict dichotomy between the implied model of rational self-expression and reactions to circumstances conceived as ‘irrational’ responses to sheerly external interferences. Thus, just as we should not identify the ‘real’ Howard exclusively with the ‘calmly reflective agent’, nor should we necessarily model his ‘weak’ response to the needle as an involuntary reflex, conceived as an autonomous controlling influence. And consequently, if we don't want to divide behaviour into two basic classes, of active, fully rational self-possession on the one hand and passive exposure to exculpatory forces on the other, we must see instances of sheer active independence and sheer passive dependence as limiting cases and conceive self-determining activity in a scalar rather than binary fashion. Howard need not be a limiting case, in which action gives way to mere behaviour, but may exercize a degree of autonomy.Footnote ⁶

Further, placing Howard's behaviour between these two limit points requires seeing the conditions and circumstances with which he engages as somehow internal to his self-determination, despite their apparent contingency. Avoiding the limit-point of sheer passivity, we want instead to see the range of his possible reactions to novel circumstances as both informing and informed by his capacity for self-expression. Conversely, avoiding the limit point of sheer rational self-possession, we want to see those conative responses as constitutive ingredients in what an agent can count as rational. One upshot of such an integrated picture, of course, would be the familiar Hegelian claim that what can be ‘actual’ reasons for an agent do not exist in a self-enclosed domain, are not considerations available to an agent from the fantastical position of rational self-possession, as though the nature and ends of reason could be fixed or specified in advance of an engagement with the internal and external circumstances of action.

In other words, we do not want to see Howard's behaviour as the value of either of two supposedly already determinate functions. One function would be the state or disposition of cowardly fear, into which the various elements of the surgery situation enter as inputs to produce some behaviour. On this picture, the function as such is blind to an independent system of rational evaluations and operates autonomously to produce behaviour when triggered by different contingent circumstances. Another function would be that separate system of evaluative rationality that we are tempted to identify with the ‘true Howard’. This system would adopt ends and give out means as different circumstantial conditions are fed in. Amongst these conditions are any contingent conative states which may arise, and part of the function of agency on this picture is to negotiate these blind states as helps or hindrances in rationally holding itself to its goals.Footnote ⁷ The worry is that on an account such as Korsgaard's, we will keep oscillating between these two functions, and thus between activity and passivity.

III.

A path out of this impasse, I believe, is offered by Christopher Yeomans's interpretation of Hegel's discussions of modality in the Logic of Essence. Yeomans's account is rich, and I can here give only a condensed presentation of certain elements relevant for present purposes—in in particular, of his interpretation of the transition between the categories of real and absolute modality, which he illustrates precisely in terms of the relation between individual character and external conditions as expressed in action.

As I understand Yeomans's account, the key to deciphering the transition in question is the contrast between an expressive versus a merely representational relation between ‘inner’ idea or potentiality and ‘outer’ manifestation or actuality. Essential to this contrast is that the more simple, abstract and under-specified an idea is, the greater the number of representations it can receive as it is straightforwardly transplanted into different contexts; the more the particularities of a given context inform the idea itself, the more the range of contexts in which it can be represented narrows and, conversely, the more the range of its expressive possibilities widens (Yeomans 2012: 155ff). Accordingly, many of Yeomans's examples are processes of aesthetic production and the development of human skills: a simple idea such as ‘the corrupting effects of power’ or ‘the battle of the sexes’ (Yeomans's example) can receive endlessly many representations as applied to different socio-historical contexts, illustrated through different media etc. Given the inevitable thinness of what all such representations have in common, any interpretive understanding of the idea through these representations could at best consist in a more or less arbitrary taxonomy—perhaps in terms of period, genre etc.—with the multiplicity of representations merely revealing the externality and contingency of the various conditions under which the idea has been developed vis-à-vis the idea itself. By contrast, once the idea itself has been informed by, say, some particular institutional setting, an application to particular individuals, or placed within a particular tradition of production or performance, although the possibilities for representational variety drastically decrease, the opportunities for interpretive understanding accordingly increase: the more the very idea is shaped by circumstances, the greater the opportunity for expressions to bring out telling nuances within it and illuminate the circumstantial conditions in turn. Crucially, any strict opposition between ‘internal’ idea and ‘external’ circumstances is thus inappropriate; what counts as external circumstance will have to be judged on a case-by-case basis, depending on the extent to which a given expression manages to enrich its governing idea through integrating elements of its setting.Footnote ⁸

This distinction provides a tool for understanding the different perspectives of real and absolute modality. The former perspective presupposes a strict separation between a produced actuality and the grounds of its real possibility, i.e., the conditions of its production, and understands the concepts of necessity and indeterminacy accordingly. Regarding produced actuality and prior conditions as possessing mutually independent intelligibility conditions, real modality conceives indeterminacy in terms of the relative under-specification of the latter: real possibility is to contrast with mere logical possibility and its criterion of non-contradiction by referring to the actual enabling conditions of some actuality, in which its potentiality thus lies; the extent to which the future is open, and so as yet indeterminate, is a function of which and how many prior conditions are specified, something that will be determined in practice by the interest-laden context in which a given question of possibility is posed.Footnote ⁹ On the conception of real modality, however, the idea of a complete specification of conditions is perfectly intelligible, and necessity is accordingly understood as an additional factor effecting the transition from a complete set of conditions to resultant actuality; the necessity of an outcome is directly proportional to the number of determining factors specified. Consequently, in the case of human action, the libertarian impulse to inscribe indeterminacy into the ontological structure of the situation (rather than ‘reducing’ it to a function of contextually determined under-specification) leads to the insertion of some kind of ontic gap into the process of necessity that otherwise translates potentiality (conditions) into actuality.

The metaphysical picture shaping the perspective of real modality grounds representative rather than expressive structures. Take its application to our case:Footnote ¹⁰ to see what necessitates Howard's act of, say, cancelling his appointment at the last minute, we specify ever more determining factors of the situation, adding to his fear various features of the surgery, say. Here, his fearful nature is something independent of the environmental conditions in which it is placed, and once it interacts with them to produce a result, we thus receive a representation of a cowardly character. An overall taxonomy of character types could be constructed from such examples, in which the conative base, so to speak, of different characters would nonetheless be blind and related externally to the circumstances in which they are contingently placed.

All-too briefly stated,Footnote ¹¹ the transition from real to absolute modality proceeds via the insight that, given how the identity conditions of ‘prior’ factors come to be specified in terms of what they produce, the process of exhausting those factors will become a description of the produced actuality itself. Consequently, if divisions between producing and produced factors are not to be as conventional and external to the productive process itself as the choice of which factors to include (and which under-specification is relevant) in making determinations of real possibility, it becomes necessary to think in terms not of mere production, but of self-production: the conditions of something possible are grounded in the possible itself.

Yet the shift of perspective to absolute modality affects how we see this self-production. In particular, it requires different understandings of necessity and indeterminacy, both of which can best be illustrated through our example: necessity, rather than being an additional factor mediating prior, potentiality-determining conditions and a produced actuality, is identified with the ‘inner’ potentiality itself, in our case: a cowardly character. But for the locus of responsibility to be in Howard's character rather than dispersed amongst a set of conditions, that character's self-manifestation in action has to be an instance of self-constitution as such a character, which, moreover, shapes the conditions that form the background of the action. Rather than supposing a set of prior conditions that are constitutively independent of subsequent actualities, absolute modality has it that manifestations of subjectivity posit their own conditions as presupposed.

One way of motivating this perspectival shift is to consider what makes Howard's doing teleologically intelligible: nothing about his situation in itself necessitates any particular action, but for his response to be intelligible as an exercise of agency, identifiable as the action of avoidance that it is, the needle must be constituted through his perception of the situation (i.e., presupposed) as, say, threatening. And his tendency to perceive it in this way, and so to act in a way that counts as an avoidance of the situation, is what constitutes his character as a certain stripe of cowardly.

For this reason, the requirements on his behaviour being intelligible as an expression of agency also require a different conceptualization of indeterminacy. To repeat the contrast used above, viewing Howard's action as a case of his expressing himself as someone with such and such a character, rather than just viewing him as mere site of reactions to external circumstances, means seeing him as making something of the situation in which he finds himself, and thus as rejecting some alternatives in favour of others. In real modality, we have perhaps a perception of the needle, a feeling of apprehension etc., and these determine a reaction that we can classify as a representation of some character type. But now, we are to see his response as one that takes that apprehension and exploits it in expressing one kind of character rather than another, e.g., as brave rather than cowardly or hesitant—his doing makes sense as this kind of action only within such a space of contrasts. And for his response to be thus intelligible, it must itself unify its conditions by articulating and coordinating them in one way rather than another (Yeomans 2012: 167; 176ff). Thus, if he places himself on the continuum of cowardly (rather than brave, equanimous, etc.) options, the approaching needle can be taken up in a range of ways, from unpleasant to terrifying, the response in action can be more or less drastic or restrained, and the character thus manifested could thus be variously timid, anxious or craven. The important point is that the action's being this coordinating activity means that it is constitutive of the character it expresses: it is intelligible as the action it is only insofar as it is a rejection of other options, and that rejection is in turn intelligible thanks to the way it constitutes its conditions, namely in perceiving the features of the situation as having a certain location within a field of possible alternatives and manifesting this rather than that individual character. Alternative possibilities are thus inscribed within the action itself, as a condition of its teleological intelligibility.

Since the inner character that receives an outward manifestation is formed via that very manifestation, we here clearly have the traits of an expressive relation: the expressive possibilities of a certain character are limited insofar as its possessor places herself on one continuum rather than another; but within that option the possibilities for coordinating the conditions of action are correspondingly richer, facilitating the manifestation of a distinctive individual character. On this model, Howard's panicked reactions are necessary, determined by his character, but only given how those reactions, as actualizations of his fear, manifest reciprocally dependent constitutive activities. From the perspective of absolute modality, in short, the process of necessity is one of self-making or self-constitution, rather than determination via externally given conditions. The ingredients of the doctor's surgery do not give us a context in which Howard's fear can be represented, but are appropriated as conditions of his self-expression. The centre of determining control moves from intruding factors into the subject.

The perspectival shift to absolute modality is supposed to help us see how Hegel understands the nature of will-formation in a way that preserves expressive freedom without needing to challenge the picture of real modality by positing an ontic gap between prior conditions and actuality, i.e., filling it with a capacity for contingent choice or Willkür. Instead, actions are comprehensible against the background of the character they express, which generally—though not necessarily in every case (radical transformations are always possible)—restricts the range of someone's expressive possibilities. Yet that character is the product of a series of responses which, by constituting their conditions as they do, render the character more determinate and subtly alter the space of possibilities to be exploited in further responses. The element of self-constitution in this process means that expressions of character do not reduce to reactions determined by intervening conative states, but represent a choice of self. In Yeomans's words: ‘If action makes motivations more determinate, and motivations are the vehicles of self-expression, then action helps to set the direction and scope of possible future expressions of self’ (ibid. 178).

IV.

Lest this theory of action should seem like an artificial imposition of certain categories of the Logic of Essence onto recalcitrant material, it is important to bear in mind that the expressive structure Yeomans seeks to elaborate just is the structure of Geist for Hegel.Footnote ¹² At the start of the Philosophy of Subjective Spirit, he reminds us that the very essence of Geist is freedom (Enc.: §382), which is the self's (the ‘Dasein of the concept's’) maintaining an identity with itself in its particularity; and he claims that this particularization is to be understood as self-expression:

For Hegel, this immediately implies a by now familiar modal claim: ‘Its [mind's] possibility is therefore immediately infinite, absolute actuality.’

Geist exits in bringing itself to expression,Footnote ¹³ a feature Hegel elaborates in terms of the relation between form and content: the content revealed [offenbart] through expression is not simply prior to expressive form: ‘In mind, therefore, form and content are identical with each other’ (Enc.: §383A). Further, it is constitutive of Geist that it not be merely self-contained, but manifest in an external medium, whilst nonetheless having sublated its externality: ‘Mind which is for itself, or mind as such […] is, therefore, that which reveals itself not merely to an Other but to itself, or, what amounts to the same thing, that which accomplishes its revelation in its own element’ (Enc.: §383A). As we have seen, the very point of the appeal to the perspective of absolute modality is that the condition of freedom, of ‘not being dependent on an Other, the relating of itself to itself’, is an appropriative engagement with outer contingencies, without which there is no self for the self to relate to.Footnote ¹⁴

The intelligibility of this paradoxical-sounding structure is owed to the conceptual determinations Hegel thematizes at the beginning of the Logic of Essence under the heading of ‘reflection’. This is a form of self-relation characterized by a certain open-ended, developmental structure within which reflection's self-misapprehension plays a constitutive role.Footnote ¹⁵ It is essential to a reflective self-relation that it determines itself by taking itself, i.e., what is determined, to be determined independently of its determining activity, as though 1) the determining is not a self-determining and so 2) not of something that is self-determined. To be sure, there is more than a slight danger here of replacing one set of paradoxical abstractions with another,Footnote ¹⁶ yet the features of this relation are just those with which the perspective of absolute modality attempts to explicate the teleological intelligibility of action: the manifestation of character in action posits its conditions as presupposed. And immediately after claiming expression to be the structure of Geist, Hegel says that this manifesting just is a positing of its ‘natural’ conditions as independent of that positing: ‘As the revelation of mind, which is free, it is the positing of nature as its world; but because this positing is reflection, it is at the same time the presupposition of the world as independent nature.’ (Enc. §384)

In this regard, Christian Martin's observation that the three forms of reflection—positing, external and determining—together exhibit something like a hermeneutic process could not be more relevant (Martin 2012: 53): positing reflection reflects the truth that it is identical with the posited, but by abstracting from the conditions of that truth, it sees the relation between reflection and reflected-upon as immediate unity or agreement: it is like presupposing that an utterance or piece of text has a fixed sense, there waiting to be grasped as it simply is. External reflection is an immediate recoil from this naivety, which thus distinguishes itself from the presupposed: but by simply flipping from immediate unity to immediate difference, it sees itself as imposing alien determinations on immediate being-in-itself: it supposes that any understanding of a text comes too late, so its determining activity can only ever be its own interpretation of something essentially inaccessible. Finally, determining reflection is the realization that it cannot be as incapacitated as it assumed: the immediacy of the presupposed is a chimeraFootnote ¹⁷, and the presupposed has the determinacy and meaningfulness that it has through its being mediated, through reflection's always already having departed from it. Whereas both positing and external reflection assume a certain ultimate sense, subject to no inherent development or mediation, determining reflection appreciates that the original text is mediated through and through; its component elements cannot just have their meanings immediately in-themselves, but only through their combination and context. In other words, any putatively self-standing element of meaning will bear significance only through its retrospective constitution, through having been placed within a certain context of meaningfulness—which placing always simultaneously co-determines the context. The developmental process of sense-formation has no inherent boundaries, where the residual external interpretations of the Ur-meaning are finally stripped away and the meaning stands revealed.Footnote ¹⁸ Rather, the ongoing process of interpretive reflection is responsible for the meanings there to be sought by and revealed to external and positing reflection.

Evidently, the progression through these reflective forms partly characterizes the perspectival shift from real to absolute modality. Real modality involves external reflection between character and the further conditions of action; the latter are constituted independently of the former, which is simply added to them as an intruding determining factor. In absolute modality, by contrast, the determination of conditions by character exhibits the form of determining reflection, a determination that presupposes its having shaped what it determines; the expressive activity that is the necessity of character is therefore self-expression, and just as the interpretation of signs must abandon the myth of original meaning and always presuppose a developmental process of retrospective constitution, the teleological intelligibility of action presupposes an ongoing process of character-formation.

Hence Hegel claims that what brings about mind's actuality is not ‘only a form externally added to its content’, but rather ‘the revelation of itself to itself is […] the very content of mind’ (Enc. §383A). It is in this sense that what manifestation reveals is manifestation itself, and hence the manifestation of possibility in actuality is self-revelation.Footnote ¹⁹

V.

With this conceptual structure in place, the basis of an alternative picture of self-constitution can come into view. To be sure, a host of details need filling in. For one thing, in the brief sketch of absolute modality above, it was vital to the self-constitution of character through action that affective responses to the environment not be conceived as triggerings of conative states that are essentially independent of the objects triggering them, but as ways of apprehending the objects themselves; recall that the nature of Howard's fear depends on how exactly his response characterizes features of the environment, e.g., the needle as a threat to-be-avoided. He could not exhibit the range of affective responses he does, and these would not be intelligible in terms of their contrasts with other possible responses, unless he had mastered a rich variety of conceptual distinctions. If his conative register were restricted to finding things, say, pleasant or unpleasant, he could hardly find himself being variously, say, apprehensive, fearful or anxious. He can exhibit these nuances of affective responsiveness only if given sufficient conceptual resources at his disposal for understanding both his environment and himself. But this evidently requires an account of how feelings and desires can be elements or vehicles of an overall worldview, rather than something which that worldview confronts; or, in Hegel's terms, of how they can be determinations of thinking.Footnote ²⁰

Hegel himself insists repeatedly that lower, affective determinacies of Geist can be forms of higher contents derived from cognition and volition. In his discussion of practical feeling (Gefühl), for example, he castigates the metaphysics of the ‘understanding’ for holding fast to separations between the Seelenvermögen, playing off Geist against feeling, and failing to see how ‘feeling is nothing but the form of the immediate, peculiar individuality of the subject, a form into which this content [the ‘ideas’ of ‘God, right ethics’], like any other objective content to which consciousness also ascribes objectivity, can be put’ (§471A). Hence the temptation to locate the source and justification of right and its objectivity in feeling rather than reason.Footnote ^21, Footnote ²²

But even setting such issues to one side, this picture of self-expression as activity that posits the conditions of its own realization may seem to just bring us back to our starting point. Consider the second of the two functions mentioned at the end of part II; to take up Watson's terminology again, the evaluative system fixes fully determinate ends and principles in advance and then negotiates and manipulates contingently encountered inner and outer circumstances so as to act in accordance with the former. Clearly, such a picture would remain indebted to the perspective of real modality, whereby what brings about an actuality is the addition of further conditions to factors already enjoying a constitutive independence. The putative function thus makes the process of necessity too external to inner potentiality and correspondingly places contingency in too stark an opposition to necessity. In a word, from the perspective of absolute modality, contingency is to be aufgehoben, not eliminated, incorporated into the very self to be expressed. And a concomitant of this does justice to our concept of character as something stable and enduring: for, the business of shaping one's character increases expressive potential by decreasing the number of domains to which a person's character is suited whilst making it more resistant to arbitrary transformation.Footnote ²³ The extent to which Howard's cowardice is reformable will depend upon where he has placed himself, over time, on the continuum between full rational self-possession and blind passivity.

Nevertheless, does the account we have extracted from Yeomans not threaten to collapse into the unsatisfying picture we wanted to overcome? On the present view of self-constitution, conative states are to be understood within an interpretive or hermeneutic dimension, as means by which an agent configures the conditions of their action as the background against which it is intelligible. An agent expresses their character, rather than being simply determined by it, thanks to the reciprocal productive relation between character and action. Insofar as we act upon certain desires or passions, we thus exploit occasions to present ourselves as particular kinds of person, particular characters. Our initial move beyond Korsgaard was to locate actions within the ongoing process of character- development, and the picture we find in Yeomans requires precisely that self-constitution be effected through such a process. But placing the emphasis on self-determination even in affectively laden situations easily invites suspicion. Invoking the category of absolute necessity with its Spinozist echoes might suggest a particularly implausible variety of self-generation. Surely, the picture cannot be that a subject self-consciously reflects on her circumstantial conditions, on her already achieved character, and on the effect that possible responses to affective situations would have on that character, and then decides to respond in a way that ‘expresses’ such a modified character, as though weaving herself into an explicitly constructed narrative. As we saw, Yeomans tells us that ‘the choice to identify with a motivation is the choice to identify with a potential self’. And taken out of context, such a formulation invites the suspicion that the metaphor of self-authorship is being taken all-too literally: it conjures the image of a kind of doubling of the self, the one actively constructing the other as one constructs a fictional character.

As well as threating the return of an obvious regress, this picture would clearly just be a new version of the fantasy of full rational self-possession, albeit with a more pronounced imaginative, or even Romantic,Footnote ²⁴ dimension. It emphasizes the difference between subject and character at the expense of their identity and, needless to say, does not so much represent inhabiting or being guided by a worldview, as reflecting upon one as if from nowhere. To be sure, a reflective reorientation of one's character is always possible. But on Hegel's view, any explicit intentional reflection is always from somewhere and itself conatively inflected.Footnote ²⁵ It must therefore take place against the background of and be grounded in an orientation of the self to its objects that already embodies and realizes a self-constituted character. Evidently, this background to reflection must therefore possess three characteristics: firstly, it must be prior to the level of conscious intentionality; secondly, it must have a conative rather than purely conceptual or judgmental articulation; and thirdly, it must be a mode of the subject's relating to its environment that is expressive rather than merely representational.

VI.

In Hegel's Philosophy of Subjective Spirit, the need for a conative, pre-intentional and expressive basis for reflective intentionality takes the form of ‘consciousness's’ genetic grounding in the ‘soul’, of ‘Phenomenology’ in ‘Anthropology’. And crucially, it is habit which provides the transition from the soul to intentionality, a transition Hegel designates as the attainment of ‘actual’ soul, since here the soul is no longer passive, but wirklich, effective in its mastery of the subject's natural side as its expressive medium.

Our original problem was how a manifestation of character can be a mode of self-realization, can belong to a subject as an expression of her character rather than as the representation of a contingent disposition. Given its connotations of passivity and mechanism, habit seems an unpromising resource to draw upon here. Hegel anticipates this objection. Although we often think of habit as lifeless, contingent and particular, nevertheless, he says:

To see why Hegel sees habit as the means by which objective contents can belong to a concrete individual subject, it is worth expanding upon certain elements of this passage.

The idea of ‘soulful ideality’ is already in play when Hegel first introduces that of the soul:

Anthropology treats of phenomena that are clearly anchored in nature, features of the biologically grounded lives of living beings, but which can no longer be understood through the kinds of causal explanation that are otherwise at home in our accounts of nature. For, the determinations at issue are themselves constituted by a ‘negative’ stance or bearing of the subject towards them; as a result, they longer count as simply natural, and their effects on the subject cannot be understood in terms of relations between logically independent factors. In Anthropology, we are concerned with ways in which subjects take up, rather than just passively receive, their surroundings, even though this ‘taking up’ is still fairly minimal, falling short of self-conscious perceptual judgment or voluntary action, and thus not yet thematic for the subject.Footnote ²⁶ Nonetheless, as the reference to ‘substance’ already indicates, the determinations of the soul provide the necessary background of the developed intentional achievements treated subsequent to the Anthropology.

As minimal stances towards nature, the determinations of the soul already implicate a degree of self-differentiation, albeit one that is essentially affective; the determinations of the soul are modes of appropriating the environment at the level of sensation and feeling that, by together constituting a kind of conative orientation to the world, are productive of an individual subject. Indeed, the soul is just the structure of self-expression realized at the level of the conative. It is the way in which a subject develops its expressive capacities by contrasting itself with something it is not, namely nature, but whereby this contrasting takes the form of appropriating nature as a condition of its own activity.Footnote ²⁷ Hegel even sees minimal instances of such ‘taking-up’ and self-differentiation in our affective, emotionally tinged responsiveness to the changing seasons and times of the day, which he locates at the level of the ‘natural soul’. These qualities are generally not thematic for the subject; it ‘lives with’ [mitleben] them, meaning they are embodied modes of living with its environment: ‘they are natural objects for consciousness, though the soul as such does not respond to them as external objects. Rather, these determinations are natural qualities which it has in itself’ (§391). Yet the way in which individuals have varying receptivities for these qualities, reflected in different characters and temperaments, already shows an emerging distinction within the ‘Mitleben’ of these qualities between experienced and experiencer, and thus the presence of subjectivity in natural soul.Footnote ²⁸

As he moves through to the ‘feeling’ soul, Hegel charts increasingly complex ways in which the soul achieves heightened forms of identity within difference to its corporeality, each reflected in qualitative shifts in its modes of affective taking-up (e.g., from living-with to living-through [Erleben]), corresponding to different affective states: at first sensation, which can be triggered by both environmental impacts and contents of thought and consciousness, and then feeling, which is the soul's ‘living through’ a manifold of sensations. As the soul bundles together various sensations into common feelings, it ‘posits’ them as its own, sensing itself as their common subject (‘self-feeling’), rather than losing itself in its dispersal across a chaos of discrete sensations. Still, even at the level of feeling, the difference between sensing and sensed is still only latently present, and the differentiation of the soul from particular contents is not their objectification to a self-conscious subject, but the soul's incipient sense of itself throughout and within its sensations. But §409 announces in imperatival form that its positing must become thematic for the soul itself: ‘But the self is implicitly a simple relation of ideality to itself, formal universality, and this is the truth of the particular; in this life of feeling the self is to be posited as this universality’. This requires a more definite self-differentiation of the soul from its natural aspect, between self and the contents of self-feeling, which is to be achieved through ‘breaking’ with its bodiliness—a ‘break’ effected through habit. The soul now takes ‘possession’ of its contents, an idea that seems to denote that whilst we are not yet dealing with a conscious relation to contents, the soul's integration of its natural life is far too advanced for us to see it as merely ‘sunk’ within its contents (Enc. §410).

The central mechanism of habituation is repetition: building on the bundling of sensations into feelings, the recurrence of felt contents now allows the self to grasp itself as a constant factor through variation, through becoming both inured to unpleasant external sensations, which thus become less salient, and indifferent to the satisfaction of desires. These processes are liberating, freeing up the self for more truly minded engagements with the world. But the more important form of liberating, repetition-induced habituation is the inculcation of bodily practices, the cultivation of skills, which involves a reciprocally dependent process of ‘self-incorporation’ [Sicheinbilden]: the training of corporeality, the development of bodily facility, is at once an incorporation of determinations into the ‘being’ [Sein] of the soul.

In this regard, Hegel stresses how the process of habituation effects a gradual transition from the subject's having a mediated relation to its body as a kind of tool to its pursuing its ends through their immediate incorporation into bodily action. His example is learning to write: eventually, one ceases to deliberately manipulate one's own body like an external tool for the sake of an independent goal; one masters the skill, and the initial discrepancy between body and subject (i.e., its ends) is overcome. This full interpenetration of goal and behaviour evidently bears the key hallmarks of an expressive structure in which outer form and inner content are mutually informing: not only are the bodily movements of the subject inherently related to its ends, but those ends themselves are corporeally mediated—indeed, through practice, they come to be so immediately.Footnote ²⁹ Cultivating dexterity and skill is, in short, a process that requires, but cannot be reduced, to mere mechanical repetition. Unlike mechanical routinization, all but the most rudimentary skills exhibit an open, developmental structure that shapes and is shaped in turn by the materials it engages with.

Indeed, Hegel sees expressiveness as the mark of the ‘actual soul’ achieved through habit, stressing the positive, expressive freedom that is the counterpart to the negative freedom won through inurement to sensation: through habit, he says, the body becomes the ‘sign’ of the soul (§411). But this semantic relation is not to be understood representationally, but expressively: the subject of which the body is a sign is not something simply prior to its semantic manifestation. Hence Hegel's remark that the body becomes the soul's ‘artwork’, a totality that cannot be reduced to an isolable content and whose content cannot be reduced to mere outer form.Footnote ³⁰ In trying to capture the expressive interpenetration of soul and body, Hegel deploys suggestive metaphors, explaining how ‘the embodiments occurring with freedom […] impart to the human body a peculiar mental stamp [geistiges Gepräge]’, and speaks of ‘the mind-pervaded aspect of his body [geistgedrungene Ansehen seines Leibes]’ (Enc. §411). But it is important to bear in mind that the dual nature of habituation processes means that Geist does not pervade an external medium with content that is already fully determinate. As is clear from all but the most rudimentary skills, their cultivation clearly does not consist in the successful manipulation of the body as an external tool, but in an appropriation of ‘bodiliness’ that gradually shapes both the skill and the ends to which it is deployed.

In the Addition to §411, Hegel explicitly tells us that the will [Wille] itself is actualized through habit and makes clear that volition is present even in the most elementary forms of human agency: here, the upright posture of the human being is an achievement of the will, albeit an utterly minimal one of which we are barely aware. The very possession of a will is not something prior to the soul's mastery of its affective life, but something that Hegel clearly thinks evolves through that mastery. This is not to say, of course, that the objective contents of the rational will have their source in feeling; but they become contents of an individual's will, principles of a subject's action, only through habituation. It is through habit that a worldview capable of being articulated and developed at the level of conceptual reflection becomes that of a subject, i.e., through being transmuted into the mode of feeling and sensation. These contents too undergo ‘incorporation’ [Verleiblichung] into the very ‘being’ of the soul. As Halbig puts it,

The ‘affective colouration’ of perceptions of situations was, of course, a vital component of the account of character-expression sketched in section III: subjects both develop and manifest character through a series of actions, through processes that gradually shape the ways in which they render intelligible situations, and thus their own actions in turn. We now see that without the habituation intrinsic to these processes, evaluative contents that have their home in rational reflection could not be actual principles of the individual will. For this reason, Hegel says, ‘[h]abit has rightly been called a second nature’ (Enc. §410R), and his explanation of this notorious phrase simply articulates how habit is the vehicle for cultivating expressive behaviours: ‘nature’, because the contents of habit are ‘an immediate being of the soul’; but a second nature, ‘because it is an immediacy posited by the soul’. These determinacies are ‘found’ by the soul thanks to its own activity, of ‘incorporating and moulding the bodiliness that pertains’ not only ‘to the determinations of feeling’, but also those of ‘representation and of the will in so far as they are embodied.’Footnote ³¹

It is thus through habit that a subject appropriates its body as an expressive means—not as a mere instrument, but as what it is; its behaviour can therefore exhibit rational patterns rather than simply exist as a medium which represents given inner states. Returning to our example, our aim was to understand Howard as neither blindly responsive to novel circumstances, nor as acting out a premeditated self-image. We saw that seeing Howard's action as an expression of character cannot involve modelling that character in terms of either of two functions, an affective disposition blind to a rationally structured worldview or a system of rational evaluation with a merely instrumental relation to the subject's affective life. We wanted to see his action itself as taking up a stance to his environment, without this being just a representation of the conative states or normative beliefs that were fixed prior to action. On the expressivist view outlined above, expressive potential expands the more a subject has internalized external conditions into the practical attitudes and contents it will go on to express; the more elaborated the space of distinctions upon which their actions can draw in correlating prior (presupposed) conditions, the greater their expressive potential. Yet that activity of correlation has a wealth of expressive resources at its disposal only if it can select, so to speak, between a range of embodied behaviours which would already manifest cultivated affective stances towards the world. This would be impossible without the work of habit, which transforms conceptual determinations of thinking into ‘flesh and blood’, by transforming discrete episodes of feeling into emotionally accented ways of bodily engaging with the environment. And the way individuals ‘possess’ those determinations cannot be modelled on either of the two aforementioned functions: they are in the ‘self’ neither as the first would have it, be it as something merely dispositional, ‘merely implicitly (as predisposition [Anlage])’, or as an isolated occurrent state, ‘a transient sensation or representation’, and nor as the second would have it, ‘as abstract inwardness, cut off from action and actuality’.

Given this understanding of habit, we can work our way closer to a Hegelian view of how self-constitution can function in and through, rather than in spite of, the unpredictability of our conative lives.Footnote ³²