I. Introduction
Constitutivism is the general view that some class of normative standards can be derived from the constitutive features of agency. The constitutivist claims that all persons, as agents, are bound by certain norms simply because they are agents. Christine Korsgaard, for example, holds that the constitutive features of agency require adherence to Kant’s categorical imperative (Korsgaard Reference Korsgaard2009). David Velleman, on the other hand, believes that features of agency require persons to act to increase their self-knowledge (Velleman Reference Velleman2009). Paul Katsafanas develops a Nietzschean version of constitutivism, which holds that we aim to perform actions that we can approve of knowing their causal origins and that also allow us to overcome obstacles. Thus, he argues, we have reason to perform those actions that provide opportunities to overcome obstacles (Katsafanas Reference Katsafanas2013).
While there are several varieties of constitutivism, they all have a similar structure. Each starts with a conception of action and its constitutive features. These features specify what one must aim at if one is performing an action. The accounts then derive norms of behaviour from these features. Constitutivists argue that persons have reason to adhere to these norms because failure to adhere constitutes failure to be an agent. These accounts seek to ground normativity in agency and allay sceptical worries about a person’s reasons to abide by norms, moral or otherwise.
In this article, I argue for a Hegelian constitutivist view. In addition to Kant and Nietzsche, Hegel has interesting things to say about human action and its relationship to normativity. The view developed here is epistemic in nature. Adherence to certain norms of recognition is both necessary to know what one has done in the past and to guide future action. The view is also minimal. I do not believe the constitutive features of agency can tell us much about which actions to perform or what moral code to follow. In this way, I believe that many versions of constitutivism may promise too much. The alchemist cannot turn base metals into gold and it is unlikely that we can derive many compelling norms of action from its constitutive features. But, as I argue, Hegel allows us to see that the foundations of morality can be gleaned from the nature of action.Footnote 1
I will proceed as follows: In the next section, I provide general reasons to care about constitutivism. This section aims to provide some standards by which to judge the success of any version of constitutivism. In section III, I offer criticisms of Kantian and Nietzschean constitutivism. I focus on these two versions of constitutivism because they provide nice contrasts to Hegelian constitutivism that are also grounded in the history of philosophy. Section IV is a defence of Hegelian constitutivism. And, in section V, I reply to possible objections to the view. Section VI serves as a brief conclusion.
II. Why constitutivism?
Admittedly, it is sometimes hard to see the point of more meta-ethical philosophizing. It seems that we have settled into our camps and the various sides are not moving. And, let us be honest, it is not like the world outside of academic philosophy is craving a new defence of constitutivism or any other meta-ethical theory. But, there may be a point to all of this. Here I say a bit more about the potential philosophical benefits of a true constitutivist view and why we should care if constitutivism is true.
Constitutivism aims to ground a certain range of normative claims. For the most part, we believe we should not lie and steal from our friends, believe that 2+2=7, or arrogantly ignore the opinions of others. But we are hard pressed to find a reason for these beliefs that is not circular or question-begging. We appeal to morality to ground moral claims and mathematics to ground mathematical claims. But what grounds the claims we appeal to? We could take the Wittgensteinian route and say ‘nothing’, or ‘our form of life’, but this seems unappealing because we clearly want something to make us right and we believe that certain forms of life are just bad or incorrect. Likewise, appealing to a divine source is problematic for a whole host of reasons. Constitutivism aims to restore our confidence in our beliefs in a way that is satisfying.
Of course, it is easy to just claim that it does not matter or that we do not really care. And this is true, for the most part. I, for one, do not spend my days thinking about meta-ethics. But, sometimes we do care. There are people who believe in wild conspiracy theories, vote for incompetent politicians and are committed racists. We take it that these people are wrong. But, sometimes they make us question ourselves, maybe we are irrational! In these moments, we do care. We want to hold on to something in the face of a world full of wildly irrational and immoral people who are fully confident in themselves.
Further, constitutivism promises to provide an answer to the normative sceptic, the person who believes that there is no reason to abide by norms of a particular type. The constitutivist’s response to the sceptic is simple and straightforward: you have reason to abide by these norms because they are constitutive of the thing you are doing. Take for instance a person who wrestles yet is sceptical about the norms that guide her to score more points than her opponent and try to get a pin if possible. She may ask, ‘why should I try to pin my opponent?’. The constitutivist has an easy answer: ‘You should try to pin your opponent because you are in a wrestling match and pinning your opponent is the ultimate goal of the sport. If you abandon that goal, you are no longer wrestling. You are just out there making yourself tired!’. The norms, then, are built into the activity for the constitutivist.
To be clear, the global normative sceptic is a creature made up by philosophers. She is as much a real being as Descartes’s evil demon. I just do not think it is true that a person exists who is sceptical about all norms. Perhaps she serves some important methodological purpose. I am doubtful. However, I am less doubtful about the existence of local normative sceptics, folks who think there’s no reason to abide by a particular set of norms. Take, for instance, a law professor who believes he has no obligation or reason to vote. He publishes articles about public policy, but does not vote! Now, no existing version of constitutivism can respond to this particular local sceptic, but the point is that local sceptics are real and constitutivism is designed to address some of them.
Moreover, sometimes the sceptic is us. Sometimes I do not feel like fulfilling my promises, or giving to charity, or working with problematic colleagues. I know I need to do these things, but sometimes the reasons escape me. Sometimes I feel there is no reason and I am just going along with the flow. To the extent that we are sceptical about some normative domain, versions of constitutivism aim to give us reassurance. Korsgaard, for instance, will tell us that we have reasons to perform actions that unify us. Admittedly, reflecting on constitutivism is not always motivational, but that does not mean constitutivism is not right.
Thus, there are reasons to care about constitutivism. These reasons stem from our lives. A person who truly does not care about morality, normativity, self-reflection or her ideas about herself and others probably does not have a reason to read or consider a paper about constitutivism. And that is fine. There are some philosophical topics I cannot get myself to care about. And that is fine too. But, as I hope to show, Hegelian constitutivism should be interesting and thought-provoking to some. And, perhaps, can deliver on some of the promises offered by constitutivism in general.
III. The shortcomings of other forms of constitutivism
Before diving into Hegelian constitutivism, I will say something about the shortcomings of other versions of constitutivism. I will not linger here long, but the discussion is necessary. Korsgaard has developed a Kantian constitutivism and Katsafanas argues for a Nietzschean constitutivism. I focus on these two views in this section, noting that this is far from an exhaustive list of constitutivist theories. They do, however, provide a nice contrast to Hegelian constitutivism in that they are based in the writings of two historically influential philosophers.
Korsgaard’s Kantian constitutivism entails that we are bound to govern ourselves according to the categorical imperative. That is, her constitutivism entails that we ought to act on principles that we can at the same time will to be universal laws. Indeed, any formulation of the categorical imperative can serve as an entailment of Kantian constitutivism. That is, the view entails that we ought not treat others as a mere means to an end and that we should act such that we are legislating for the kingdom of ends.
Korsgaard’s argument for Kantian constitutivism is relatively straightforward and is a faithful take on Kant’s own arguments. Her argument starts with a distinction between actions and non-actions. Actions, for Korsgaard, are attributable to a whole person and not just part(s) of the person (Reference Korsgaard2009: 18). For instance, hiccups are not attributable to a person, but to involuntary contractions of a person’s diaphragm. Likewise, heartbeats and Freudian slips are not attributable to the whole person, but to the activity of the autonomic nervous system and unconscious thoughts, respectively.
Actions, on the other hand, are attributable to the person as a unified whole because they stem from a person’s voluntary, self-conscious decisions. Korsgaard, following Kant, assumes that a person is identical to their rationality or that part of them that chooses courses of action. Korsgaard tells us that because we are rational, reflective creatures we can view our desires and inclinations from a distance, so to speak (Reference Korsgaard2009: 125–26). We can decide which desires to act on and which to ignore. Further, we make decisions for reasons. We decide to sleep versus going to work for a reason: perhaps, the reason is that we needed a rest day and nothing important is going on at work. This reason is the principle upon which our action is based.
Kant, as we know, believed that principles, or imperatives, come in two flavours: hypothetical and categorical (Kant Reference Kant2012: 28). The hypothetical imperative is binding to the extent that a person has a particular desire. The categorical imperative is binding regardless of a person’s desires. The hypothetical imperative directs me to satisfy my desires, while the categorical imperative tells me how to regulate my desire satisfaction. The hypothetical imperative may tell me to lie to my employees, while the categorical imperative will tell me that lying is wrong. If all my principles of action are hypothetical, then I am governed solely by my desires. But this will mean that my actions are attributable to my desires, not to my rationality, my principle-based decisions.
Korsgaard claims that persons who follow only the hypothetical imperative are not necessarily unified over time. They are unified only to the extent that their desires do not change (Reference Korsgaard2009: 162–74). The person who is governed only by the hypothetical imperative is unable to make commitments. A commitment requires a person to ignore certain desires to stick to the commitment, but this is just what a person who only abides by the hypothetical imperative cannot do. As soon as a strong desire that conflicts with the commitment arises, the commitment is put on the back burner. This person will never save money, for instance, because her desire to spend rules the day.
‘Actions’ governed solely by the hypothetical imperative are not actions in Korsgaard’s sense because they are not attributable to the person as a unified whole, but to a part of her: her desires. Thus, Korsgaard believes that the constitutive principle of action is the categorical imperative, which unifies the agent. Persons who do not aim to abide by the categorical imperative are not acting, but engaging in some other form of human activity. Thus, Korsgaard has a response to anyone who worries about why they should abide by the categorical imperative: if you are not at least aiming to abide by the categorical imperative, then you are not acting. Your ‘actions’ will not be attributable to you as a unit, but instead to one or several of your desires (Reference Korsgaard2009: 78–81).
There are many potential problems with Kantian constitutivism.Footnote 2 First, as Hegel and many others have pointed out, the categorical imperative does not provide much guidance. I can easily craft my principle of action such that it can be willed as a universal law. Further, it is not clear what it means to treat someone as a mere means to an end. Does the capitalist who radically exploits his employees, yet pays them the legal minimum wage, treat them as mere means to an end? Thus, it is not easy to determine when the categorical imperative has been violated. As I explain below, Hegel can respond to this question.
Second, there is the problem of bad action. This is a problem that plagues many versions of constitutivism. If action is constituted by an aim or governing principle and bodily movements that are not so constituted are not actions, then it seems that there is no room for the constitutivist to hold that there can be bad actions. In this case, it seems that Korsgaard is unable to critique the person who is governed solely by the hypothetical imperative. This person is just engaged in a different game; not action, but something else. One aim of constitutivism, however, is to help us distinguish good actions from bad. So, this is a serious problem.
Korsgaard has a solution. She claims that attributability and unity come in degrees. Drawing from Plato, Korsgaard demonstrates that principles can vary in their ability to unify an agent (Reference Korsgaard2009: 175). The timocratic principle unifies more than the oligarchic principle, which unifies more than the democratic principle (which is most like the hypothetical imperative). The categorical imperative offers the most unity. For Korsgaard, there can be bad actions, those actions based on principles that do not fully unify the agent and thus undermine attributability.
But, as Katsafanas points out, this undermines Korsgaard’s project (Katsafanas Reference Katsafanas2013: 100–101). Korsgaard’s argument would be strong if there was only one principle that could unify persons. But there are several. Given that there are several, she must explain why persons should choose the categorical imperative versus another principle that provides less unity and less attributability. The answer to this question cannot be constitutivist in nature, because the categorical imperative is not itself constitutive of action, but constitutive of the best actions. To extend my wrestling analogy, one does not fail to wrestle just because one does not choose the move that is most likely to lead to points or a pin. One can still count as wrestling and choose a move that is less likely to achieve the constitutive aim because one wants to dazzle the crowd or try something new.
While this objection is worrying, it is not the one I care about most in this context. Some actions are involuntary. Further, some of our actions have consequences that we do not intend. Imagine you and I are at a party. You have a wonderful cocktail in your hand and we are having a conversation. I have an involuntary muscle spasm and knock the cocktail out of your hand, shattering the glass. Take another example. A few friends are camping in a cabin for the weekend. I volunteer to set and manage a fire in the fireplace. I forget to open the vent. It is late, we are all talking and having fun. Eventually, everyone’s eyes are watering, some are coughing. The entire room is filled with smoke! In both scenarios, I owe someone an apology. I may even need to make amends by cleaning up the broken glass or putting out the fire and making sure everyone has water.
Korsgaard is not concerned with these types of actions. I do not violate the categorical imperative by having a muscle spasm or by forgetting to open a fireplace vent. But, these actions and omissions and my responses to them are apt for moral assessment.Footnote 3 Kantian constitutivism cannot account for the moral significance of non-voluntary actions. Sometimes we do not act on principle. Sometimes we do not think at all. Sometimes we act out of habit. Sometimes we just mess up. Kantian constitutivism is unable to tell us which of our mess-ups we are responsible for. The picture of persons assessing their desires and choosing actions based on principles is elegant and beautiful. But, it does not capture most of our lives. We are just trying our best, sometimes doing kind things, sometimes just flat out failing. A moral theory based in the very concept of action should take account of this. I say more about non-intentional actions in the following section.
Paul Katsafanas argues for a Nietzschean constitutivism. He claims that action has two constitutive aims: agential activity and power. Agential activity is achieved when a person can reflectively approve of her action with full knowledge of its causal, psychological origins. For Katsafanas, we act under the idea of freedom and aim to do so (Reference Katsafanas2013: 121–22). For instance, learning that my desire to date a nurturing person stems from childhood neglect might undermine my approval of my current, nurturing partner. Nietzsche, of course, thought that understanding the origins of contemporary morality would undermine our confidence in moral norms.
The second constitutive aim is power, that of encountering and overcoming resistances to one’s ends. This seems counterintuitive. Katsafanas claims that our actions are not motivated by desires for particular ends, but by drives (Reference Katsafanas2013: 166–68). An end can be reached, a desire satisfied, but a drive can only be temporarily satisfied. Katsafanas argues that Nietzsche believes that the satisfaction and happiness we experience upon achieving an end is not lasting. Happiness, on Nietzsche’s account, is achieved in the progress toward the end (Reference Katsafanas2013: 174). Katsafanas defends Nietzsche’s claim that this fact about satisfaction and happiness is best explained by the hypothesis that human actions are motivated by drives. The most pervasive drive, for Nietzsche, is the will to power, the drive to demonstrate power by overcoming resistances.
Katsafanas argues further that normative conclusions follow from this characterization of human action. A drive aims to engage in a process. This is true of the will to power or any other drive. Engaging in a process entails encountering and overcoming resistances. Thus, Katsafanas argues, the aim of a drive is to encounter and overcome resistances (Reference Katsafanas2013: 174). This aim exists for any drive, it seems, not just the will to power. Katsafanas then argues that this constitutive aim entails that persons have reason to pursue actions that are difficult, those that require us to encounter and overcome obstacles.
This is a much too brief outline of Nietzschean constitutivism, but it will serve my purposes here. I will focus only on the second constitutive aim of action: power. A full analysis and critique of Katsafanas’s theory is beyond the scope of this article.Footnote 4 I find a few problems with Nietzschean constitutivism. First, Katsafanas’s view is subject to an objection that I articulated in relation to Kantian constitutivism. Recall that Korsgaard admits that principles other than the categorical imperative can unify persons. This undermines her view because she is unable to give us a constitutivist reason to opt for the most unity. Similarly, Katsafanas takes it that persons have reason to seek out actions that provide opportunities for overcoming obstacles, but also thinks that all courses of action present obstacles. For instance, he believes that sitting on the couch and watching television presents persons with obstacles, even if small (Reference Katsafanas2013: 181). But this leaves him unable to provide a constitutive reason to pursue actions that provide more than minimal resistance. Or, in other words, Nietzschean constitutivism would like to provide reasons for us to compete in a national tennis tournament or write a novel, but will countenance a person who watches television all day!
Further, if all human actions are motivated by drives, and the presence of drives entails encountering obstacles, then it is hard to see how there is any room for normativity on Katsafanas’s account. Drives, I take it, are natural. They simply exert influence over persons. Now, one can recognize and approve of a drive. In the same way, one can embrace and act to magnify the drive. I can recognize my sex drive and work to have the most sex possible. But the drive itself cannot provide room for constitutive normativity. If I cannot fail to act on the drive, then I also cannot succeed. Perhaps Katsafanas will say that there is room for normativity because we can choose to encounter the most formidable obstacles, but this response runs into the previous objection. Korsgaard is right that normativity requires a separation between our natural desires and drives and our principles of action. Indeed, it seems that moral normativity requires resisting our natural drives. (Though, admittedly, Katsafanas does not aim to provide an account of moral normativity.)
On a related point, Nietzschean constitutivism lacks intuitive appeal. Korsgaard appeals to a natural thought: actions are attributable to persons as a whole. We distinguish between involuntary bodily movements and actions. But, there is nothing in our ordinary way of thinking and talking that supports Nietzschean constitutivism. We do not say that a person fails to act because she takes the easy route or does not seek out challenges. We may call this person lazy or uncourageous, but we do not say that a person like this fails to act. Additionally, to return to the first constitutive aim in Katsafanas’s theory, we recognize that persons can act yet have no understanding of the psychological origins of their action. Our friend may unconsciously marry a person who looks and acts like their father. But this action is no less an action because it lacks reflective, open-eyed approval. Nietzschean constitutivism, then, is not intuitively compelling.
IV. Hegelian constitutivism
Hegelian constitutivism embraces a simple and intuitive conception of intentional action: a person is acting if and only if she is attempting to give outer expression to her inner intentions. I offer few remarks about this conception. I am too much of a Wittgensteinian to believe that there is a single aim present in every bodily movement that we consider an action. Consider omissions: a lifeguard sits still while a young child drowns, a mother forgets to pick up her daughter from school, a husband forgets his wife’s birthday. Also consider accidents, micro-aggressions and actions that stem from implicit bias. Some examples: Mike accidentally steps on Susan’s baby, George compliments his Asian-American coworker on her command of English, Marsha hires Frank, a white man, instead of Jamal, a black man, though Jamal is more qualified for the job. These happenings are all actions in the sense that persons can be held morally responsible for them. Further, I am sceptical that every intentional action can be captured by a constitutive aim. I take it that my conception of action is good enough for the present purposes, but I also believe that a conception of constitutivism should explain the moral importance of non-intentional actions.
The intuitive appeal of this conception of intentional action should be clear. If I intend to go to the grocery story yet fail to leave the house because I am tired, I am not performing the action of going to the store but instead refraining from acting relative to this intention. The staying home, while also a doing, and one that I may be held responsible for, is not an intentional action. I did not aim at staying home, it just happened. To go to the grocery store, on the other hand, is to give expression to my intention, to act intentionally. I posit, then, that the constitutive aim of most intentional actions is just this—to act in a way that matches one’s intentions. Hegelian constitutivism is powerful partly because it works with such a sparse and intuitive conception of action.
Indeed, Hegel seems to endorse a conception of action like this. He believes that there is a close connection between action and individual self-knowledge. For Hegel, one comes to know what one is, in part, by acting. In the Phenomenology of Spirit, 401, Hegel writes:
action simply translates an initially implicit being into a being that is made explicit. […] Consciousness must act merely in order that what it is in itself may become explicit for it; in other words, action is simply the coming-to-be of Spirit as consciousness. What the latter is in itself, it knows therefore from what it actually is. Accordingly, an individual cannot know what he [really] is until he has made himself a reality through action. (PhG: 401, 239)Footnote 5
The fit between intention and action—between an inner intention and its outward expression—is normative. One can fail in one’s attempt to act based on an intention. That is, there are actions that are correct and incorrect relative to any intention. Hegel’s insight, I believe, is that the standard of correctness for actions is established not by the world or by the actor’s beliefs, but by a community of persons who articulate, negotiate and enforce norms. In the Phenomenology of Spirit, Philosophy of Right and Encyclopaedia of the Philosophical Sciences, Hegel articulates a historical vision of human action, one in which standards of correctness and the scope of responsibility evolve over time. Before diving into the details of the Hegelian view, I will provide an argument for the claim that the normative standards for human action reside in communities. Hegel does not make this argument explicitly, but I believe an argument of this type is in the background of his thinking.
The argument proceeds by process of elimination. I contend that the standards of correctness for actions are either established by the world, the individual actor or the actor’s community. The first two options are implausible, and thus the third option is correct. I first deal with the idea that the standards of correctness for actions are established by the world. This conception has intuitive appeal. If I intend to drink eight glasses of water a day, what makes it the case that I act according to my intention is the simple fact that I drink the water. If I fail to drink water, and instead drink wine, I have failed. And this failure, it seems, is completely determined by mind-independent features of the world. This view is right, in a sense, but its simplicity obscures a much more complicated story.
As Anscombe (Reference Anscombe2000) and Davidson (Reference Davidson1980) point out, any action can be described in a variety of ways. Take the following example. Sheila is shooting hoops at a local basketball court. She decides she wants to make a three pointer before she heads in. She shoots and her shot is off. But a gust of wind guides the ball into the hoop. As the ball falls through the net, it hits a toddler in the face, knocking him out cold. Additionally, in sticking around to hit that last shot, Sheila missed a scheduled call from her best friend who is suffering from depression. Upset, her friend makes a suicide attempt.
What did Sheila do in this instance? What was her action? Did she make a three-point shot? Injure a toddler? Neglect her friend? Cause a suicide attempt? It is implausible to think that the world itself can answer these questions. Persons can agree on the description of this scenario yet disagree about what Sheila did, about whether her action matches her intention. We can reasonably disagree about whether she made the shot and about whether her action extended beyond or negated her intention. The parents of the toddler, for instance, will likely claim that Sheila did not do what she intended, which was to simply make a three-point shot, but instead negligently injured their son. We could say that she did all these things, but this would still count as us saying it. The mind-independent world does not speak.
If the world cannot establish standards of correctness, it may be that the individual actor can establish standards of correctness. In this case, Sheila determines whether her intention and action match. On this assumption, the match is contingent not on the world but on Sheila’s attitudes. If Sheila thinks she made the shot—as opposed to thinking that she missed the shot because it was off when it left her hand—then she made the shot. Likewise, if Sheila believes that she caused her friend’s suicide attempt, then she did. After all, these are her intentions. Who better to judge their correct expression than Sheila?
But the problem with this subjective standard of correctness is apparent. As Wittgenstein famously observed in his private-language argument, if standards of correctness are to be coherent, there must be a distinction between a person’s thinking they are correct and their actually being correct (2009: §258, 98). Wittgenstein’s private language is unintelligible because the person in Wittgenstein’s example has a merely subjective standard of correctness for her sensation terms. Whatever she feels is the correct use of the term is the correct use. But this does away with the normativity of the terms. In the same way, if the fit between intention and action is determined subjectively, then normativity falls out. If the standards of correctness for language were merely subjective, there would be no such thing as language. In the same way, if the standards of correctness for actions were merely subjective, then there would be no such thing as action. Of course, there would be bodily movements, but the idea of attempting to give outer expression to one’s inner intentions would make no sense.
This leaves us with the third option. The standards of correctness for actions are set by the actor’s community. This, I believe, is Hegel’s view. In §118 of the PR, Hegel writes:
The self-consciousness of heroes (like that of Oedipus and others in Greek tragedy) had not advanced out of its primitive simplicity either to reflection on the distinction between deed [Tat] and action [Handlung], between the external event and the purpose and knowledge of the circumstances, or to the subdivision of consequences. On the contrary, they accepted responsibility for the whole compass of the deed. (PR: §118,118)
For Hegel, the deed (Tat) is what happens because of one’s bodily movements. The action (Handlung) consists of those features of what happens because of one’s bodily movements that one is responsible for. Hegel contends that the notion that there should be a distinction between Tat and Handlung, that one should be held responsible for only some of the effects of one’s bodily movements—usually, those effects one intends to bring about—is a particularly modern innovation. In other words, the distinction, for Hegel, is not natural but developed historically, within human communities.
Oedipus kills his father and marries his mother. This is his deed. He is unaware that the man he kills is his father, or that the woman he marries is his mother. He only intends to kill a quarrelsome stranger he encounters on the road to Thebes and to marry the queen of Thebes. So described, this is his action (Handlung). Oedipus, however, holds himself responsible for the deed. As a pre-modern individual, Hegel contends, Oedipus is not able to distinguish between deed and action. He is unable to contemplate the notion that the scope of his responsibility is related to his knowledge and intentions. Oedipus’s heroism necessitates that the scope of his responsibility not turn on facts about his intentions.
Hegel contends that modern persons have innovated the ‘right to knowledge’. In the Philosophy of Right, he writes:
The will’s right […] is to recognize as its action, as to accept responsibility for, only those presuppositions of the deed of which it was conscious in its aim and those aspects of the deed which were contained in its purpose. The deed can be imputed to me only if my will is responsible for it—this is the right to know. (PR: §117, 116)
The right to knowledge, for Hegel, restricts the scope of one’s responsibility to those aspects of the deed that one intends to bring about in acting or those aspects of the deed that are the result of one’s pursuit of one’s purpose in acting. From the perspective of the right to knowledge, Oedipus should not be held responsible for killing his father and marrying his mother because he lacks knowledge of the character of his action, so described. He neither intended to kill his father and marry his mother, nor did these outcomes, so described, result from his pursuit of any purpose he had in acting.
To further illustrate: Adam offers his houseguest, Brenda, a bottle of water. Unknown to Adam, the bottle of water was contaminated with cyanide at some point during the bottling process. Brenda drinks the water and dies. In this case, the right to knowledge bars Adam from responsibility. Adam had no way of knowing that the bottle of water was contaminated and did not intend to serve Brenda contaminated water. Nor did Brenda’s death result from any action taken by Adam in pursuit of some further purpose Adam had in acting. Adam is barred from responsibility for Brenda’s death by the right to knowledge.
Though Hegel takes the right to knowledge to be an important innovation, he warns his reader to not make the mistake of taking the right to knowledge as a reason to conceive of action in terms of intention alone. To do this would be to think that an agent’s action is complete once she commits herself to realizing some end, and to think that what happens because of her pursuing this end is simply the deed. To think of action in this way is to restrict the scope of the agent’s responsibility to the results she specifically intends, regardless of what actually results from her intentional bodily movements.
At several points in the Philosophy of Right and the Phenomenology of Spirit, Hegel expresses the idea that an action should not be understood in terms of the agent’s intentions alone. He writes:
The consequences, as the shape proper to the action and immanent within it, exhibit nothing but its nature and are simply the action itself; therefore the action can neither disavow or ignore them. […] It is because of this that it is to the advantage of the criminal if his action has comparatively few bad consequences […] and that the fully developed consequences of a crime are counted as part of the crime. (PR: §118, 118)
An addition, inserted by Hegel’s student Eduard Gans, expresses this Hegelian sentiment well:
It happens of course that circumstances may make an action miscarry to a greater or lesser degree. In a case of arson, for instance, the fire may not catch or alternatively it may take hold further than the incendiary intended. In spite of this, however, we must not make this a distinction between good and bad luck, since in acting a man must lay his account with externality. The old proverb is correct: ‘A flung stone is the devil’s’. To act is to expose oneself to bad luck. Thus bad luck has a right over me and is an embodiment of my own willing. (PR: §118A, 118).
In Hegel’s view, the full action, what one is ultimately responsible for, consists of both the agent’s intentions in acting and (some of) the results of agent’s intentional bodily movements. Thus, while intentions matter for Hegel, they do not determine the full scope of responsibility.
What I am calling the ‘full action’, the unity of intention and result, Hegel labels the ‘Sache selbst’, which has been translated as ‘the very heart of the matter’ (PhG: 237) and ‘the crux of the matter’ (Hegel Reference Hegel2018: xlv). Hegel believes that there are no clear criteria that will determine which consequences of an action should be considered part of the full action. He writes:
[t]o determine which results are accidental and which necessary is impossible…among the consequences [of an action] there is also comprised something interposed from without and introduced by chance, and this is quite unrelated to the nature of the action itself. (PR: §118, 117)
For Hegel, then, the full action consists of both intention and result, but there exists no formula for balancing the contribution of each of these elements in the final judgement about the scope of an agent’s responsibility.
The final judgement, for Hegel, is determined by persons in the actor’s community. In the Phenomenology of Spirit, Hegel writes, ‘[s]elf-consciousness exists in and for itself when, and by the fact that, it so exists for another; that is, it exists only in being acknowledged’(PhG: 111). This cryptic and pregnant claim is essential to Hegel’s understanding of action and, indeed, his understanding of social reality. Fully explicating Hegel’s conception of recognition is beyond the scope of this article, but I will explain its implications for his theory of action.
In simple terms, Hegel believes that others exist as a mirror. We can only know what we are by looking into the mirror, by consulting the judgements of others. But, unlike actual mirrors, others both create and reflect reality. While an actual mirror reflects an independent reality, other persons both create and inform us about social reality, what we are, what we are responsible for and what we have done. There are no facts of this nature, for Hegel, that exist beyond the judgements of others. Additionally, unlike actual mirrors, other persons do not automatically provide this information. A prior relationship must be in place for individuals to perform this epistemic function. This prior relationship is one of recognition. Others must recognize one as a person, as rational and responsible, to recognize one as acting and to judge one’s individual actions.
Hegel believes that achieving self-awareness (self-consciousness) is partly a matter of having knowledge of the evolving rules and standards that govern one’s recognitive community. With respect to action, a parallel point holds. For Hegel, knowing what one has done—what one is ultimately responsible for—is a matter of knowing what members of one’s recognitive community will take one to have done, and what those persons will hold one responsible for. One cannot unilaterally determine the nature of one’s action or the scope of one’s responsibility. (And, in this way, Hegel can give content to Kant’s categorical imperative. What it means to treat someone as a mere means is determined by a recognitive community.)
Hegel expands on this theme in a poignant section of his Encyclopaedia. He writes:
From what has now been said, we may learn what to think of a man who, when blamed for his shortcomings […] appeals to the (professedly) excellent intentions and sentiments of the inner self he distinguishes therefrom. There certainly may be individual cases where the malice of outward circumstances frustrates well-meant designs. […] But in general even here the essential unity between inward and outward is maintained. We are thus justified in saying that a man is what he does; and the lying vanity which consoles itself with the feeling of inward excellence may be confronted with the words of the Gospel: ‘By their fruits ye shall know them’. […] if a daub of a painter, or a poetaster, soothe themselves by the conceit that their head is full of high ideals, their consolation is a poor one; and if they insist on being judged not by their actual works but by their projects, we may safely reject their pretensions as unfounded and unmeaning. (EPS: §140, 140)
This way of thinking about action and responsibility prevents individuals from unilaterally expanding or contracting the scope of their responsibility. The arsonist cannot escape responsibility for burning down a house by claiming that he only intended to burn a small section of wood attached to the house. Likewise, a would-be hero cannot take on more responsibility than his community is willing to attribute. If Alice negligently kills Byron, she cannot unilaterally make herself legally or morally responsible for the psychological harm caused to Byron’s grieving friend.Footnote 6
Now that we have the basic picture of Hegel’s theory of action on the table, we can turn to the normative implications of the view. Ideal communities, for Hegel, are formed by relations of reciprocal recognition. Each person in the community recognizes and is recognized by the other. This means that community members view each other as rational and authoritative, as able to have a defeasible take on things and having views that are worthy of consideration and respect. Clearly, actual human communities fall short of this ideal. Persons are often marginalized, disrespected and not treated as epistemic peers. But, Hegel takes it that relationships of this type are necessary for full, self-conscious personhood. Just as a nation-state cannot be sovereign independent of the recognition of other nation-states, Hegel believes that persons cannot be persons independent of the recognition of other persons.
Normative standards follow almost immediately from this reading of Hegel. Recall that for Korsgaard, persons have reason to perform actions that unify them as agents. For Katsafanas, persons have reason to perform actions that entail overcoming obstacles. Hegelian constitutivism is different in that it does not tell us which actions to perform or which types of actions to perform, but instead how we must regard and treat others if we are to know ourselves as actors and perform actions that match our intentions. First, self-knowledge, on the Hegelian picture, involves knowledge of others. Specifically, it requires knowledge of what others think about us and our actions. It is incoherent, then, to take oneself to be a person of a certain type—for instance, nice, funny, adventurous, creative—yet disregard the thoughts, opinions and criticisms of all others. Self-knowledge requires trusting the opinions of at least some. This trusting, other-regarding attitude, I contend, is the foundation of morality.
Hegel’s view is not that we must take on the opinions of anyone and everyone. Nor is his view that we must regard any and every community as authoritative. One of the major themes in Hegel’s writings is that of the invention of the modern person, the person who is not automatically beholden to the norms of their community. The ancient Greeks, Hegel believes, did not have a conception of the individual. Their society was one in which persons fulfilled social roles and accepted communal judgements without question. Modern persons, on the other hand, have a conception of themselves as free and separate from the social roles and communal and religious norms they were raised to accept. Hegel believes that the ideal community allows a reconciliation between the ancient Greek self, that is totally dependent on the community and the modern self that is totally independent of the community.
This reconciliation comes in a form of communal life that allows individuals to find themselves within the community, such that they do not feel a tension between it and themselves. Hegel’s solution requires the full apparatus of a nation-state. But I do not go that far here. Following Robert Brandom (Reference Brandom2019: 235–61), I contend that this unity can come in the form of persons choosing their community, that is, choosing the persons or groups of persons who they will regard as authoritative. For instance, a philosopher may decide that only other philosophers can judge the quality of their work. In the same way, a hip-hop artist may determine that only persons versed in hip-hop history can make meaningful comments about their music. In both cases, the actor exercises independence by constituting their community, but also becomes dependent on those communal judgements in forming a sense of their actions and their selves. If the hip-hop artist’s chosen community judges that their latest album is not good, they will have to act differently if they are to match their intention of producing a great album. The same goes for the philosopher. Indeed, on this conception, one may view oneself as a member of several communities and reference different communities with respect to different judgements. The philosophical community, for instance, is not in a good position to judge one’s cooking. Hegelian constitutivism holds that persons must trust, esteem and take into consideration the judgements of some others if they are to know themselves as actors and act in ways that match their intentions.
Further, trusting and esteeming the judgements of others requires that one not coerce those judgements. The argument for this claim should be clear. Coerced judgements do not stem from the rationality of the person coerced, but from the will of the coercer. A coerced judgement is normatively equivalent to a solipsistic judgement. For instance, if Jamal threatens to fire Pearl if she does not praise all his decisions, Pearl’s praise cannot serve as a check on Jamal or help to guide his actions. Here Jamal’s threat is normatively superfluous. This point also extends to coercive and threatening environments. If Jamal bosses Pearl around, demeans her at work, and readily fires employees, Pearl may come to believe that she must praise Jamal to keep her job. Thus, threats and coercion need not be direct, but can arise out of a coercive environment. Hegelian constitutivism holds that self-knowledge and self-regulation in action also requires a community of persons who are not directly or indirectly threatened into making judgements. Again, this version of constitutivism provides a baseline for moral relationships.
I say that Hegelian constitutivism provides a baseline or foundation for morality because it does not provide reasons for persons to esteem the judgements of all others or to treat all persons in ways that are not coercive. One’s recognitive community could be very small, limited to one’s profession or social class. This version of constitutivism provides reason to form moral relationships with some, not all. These more encompassing moral relationships develop historically, in particular contexts, through struggle and sympathy. It is not a fault of Hegelian constitutivism that it cannot provide a justification for all of morality. That’s too tall an order for any version of constitutivism. You cannot get blood from a turnip, and you cannot get all of morality from action; and you should not expect to. But you can get something.Footnote 7
I said that Kantian constitutivism cannot account for the moral importance of non-intentional actions. Hegelian constitutivism fares better here. As stated above, Hegel believes that we can be held responsible for some of the unintended and unforeseen consequences of our actions. The person who steps on a baby, or forgets to pick their friend up from the airport, or loses a wedding ring, can be held responsible to the extent that their community allows. We do not typically hold persons responsible for justifiable accidents and unforeseeable consequences, but this is just a fact about our modern moral community. However, on the Hegelian picture, these unintentional actions have moral importance because we can be held morally responsible for them and be called on to apologize and make amends. For Hegel, our moral lives, then, are not restricted to our intentional actions.
Hegelian constitutivism also avoids the problem of bad action. Recall that the problem arises for doings that do not have the constitutive aim of action, but that we would intuitively criticize as bad actions. The constitutivist is unable to critique those doings by appealing to a constitutive standard. For Korsgaard, this amounts to being unable to critique those ‘actions’ that lack the constitutive aim of agential unity, as categorized by the categorical imperative. Hegelian constitutivism avoids this problem by adhering to a minimal conception of intentional action. On this conception, intentional action has the constitutive aim of giving outer expression to one’s inner intentions. It is hard to conceive of any doings that we would count as intentional actions that do not have this aim. This conception is consistent with the idea that there are things that persons should not do, but these bad actions can be addressed from the perspective of Hegelian constitutivism.
V. Objections
David Enoch (Reference Enoch2006) levels a powerful objection to all versions of constitutivism: the shmagency objection. The objection is simple. It seems that constitutivism must be wrong because for any conception of agency the constitutivist posits, persons can escape the normative demands of that conception by claiming to be a non-agent or an agent of a different type—a shmagent. Put another way, constitutivism seems to ground normativity in an escapable hypothetical imperative—if you desire to be an agent, then you should act according to certain norms—but the demands of normativity should be categorical, binding on all rational creatures.
Addressing Korsgaard’s Kantian constitutivism, on which the aim of self-constitution (agential unity) is constitutive of action, Enoch asks why, when presented with this version of constitutivism, one could not reply in the following way:
Classify my bodily movements and indeed me as you like. Perhaps I cannot be classified as an agent without aiming to constitute myself. But why should I be an agent? Perhaps I can’t act without aiming at self-constitution, but why should I act? If your reasoning works, this just shows that I don’t care about agency and action. I am perfectly happy being a shmagent—a nonagent who is very similar to agents but who lacks the aim (constitutive of agency but not of shmagency) of self-constitution. I am perfectly happy performing shmactions—nonaction events that are very similar to actions but that lack the aim (constitutive of actions but not of shmactions) of self-constitution. (Enoch Reference Enoch2006:179)
For Enoch, the constitutivist must provide reasons for persons to be agents. Enoch claims that absent such reasons, constitutivism fails. In this way, Enoch’s position is that of a normative sceptic, at least with respect to conceptions of normativity grounded in agency. The shmagency objection is powerful, and one that any constitutivist must take seriously.
As I see it, this objection cannot get a foothold with Hegelian constitutivism. This is because Hegelian constitutivism works with a such a minimal conception of action. Presumably Enoch’s sceptic has some conception of action, of persons doing what they intend to do. And this is all that Hegelian constitutivism needs to get off the ground. There is no philosophically loaded conception of action to object to here. If Enoch’s sceptic claims that they do not even have a desire to do what they intend to do, but something else, it is not clear what is on the table and what they are objecting to. Presumably, the sceptic wants to get around the normative demands that come with agency and action, but this desire certainly follows from a desire to do something.
Perhaps a similar objection could get a foothold in the following way. The sceptic could claim that they do not care whether their actions match their intentions, especially if the match is to be judged by others. This sceptic claims to be happy with simply believing that there is a match. That is, this sceptic is perfectly fine with incoherent solipsism. They just want to be happy. This sceptical response is problematic for two reasons.
First, we may be blind to our own mental states. We may need the input of others to know if we are truly happy and to guide ourselves toward happiness. Thus, trusting and esteeming the judgements of others may be necessary to achieve even this goal. Second, there is a worry about why the sceptic cares about intentions at all. Presumably, most normal persons want to achieve their goals. If we discount the notion of actual achievement, it is not clear what point there is in having an intention, a desired outcome, a goal. The sceptic could retreat in this way, but only at the cost of incoherence. Thus, making this form of scepticism less philosophically worrying and more psychologically puzzling.
Still Enoch claims that a successful response to the shmagency objection cannot rest on the claim that the shmagent cares about this or that (Enoch Reference Enoch and Brady2010). We can posit that the shmagent does care about agency, however construed by the constitutivist, but from this it does not follow that the shmagent should care about agency, that she has a reason to care. Regarding Hegelian constitutivism, Enoch’s sceptic will ask why she should care that she achieves what she intends to achieve. Here the response seems straightforward. One should care that one achieves what one intends to achieve because of the intention itself. The reason is intrinsic to the intention. The sceptic could press on and ask what reason she has to have intentions at all. But again, this question seems philosophically uninteresting. The larger question, ‘What reason do I have to be moral?’, is philosophically interesting and has plagued philosophers since Plato’s time. The question, ‘What reason do I have to have intentions?’, has neither of these features and appears to be a manifestation of philosophical stubbornness rather than insight.
One might also object that Hegelian constitutivism does not give us enough. Kantian and Nietzschean constitutivism tell us what we should do. And Kantian constitutivism, at least, gives us a robust picture of morality. Hegelian constitutivism only tells us that we need to establish minimal moral relationships with some people, but not all. This may seem like a weakness, but I do not believe it is. I said that constitutivism, in general, aims to ground our normative claims and to answer the normative sceptic, even when that sceptic is us. But it is also the case that constitutivism must be plausible to achieve these goals. It seems that the results of Kantian and Nietzschean constitutivism come at the cost of plausibility. Hegelian constitutivism, I claim, gives us modest results that stem from a modest, yet plausible theory.
Hegelian constitutivism gives us a ground, a starting point, for morality. This starting point is the minimal moral relations that must exist between persons if they are to act as they intend. Trusting and esteeming the judgements of others in a way that is free of coercion is a starting point. For others do not just make claims about the content of one’s actions, but also about which actions one should pursue and about how they themselves should be treated, and about how to treat others. Hegelian constitutivism does not hold that persons are obligated to listen to claims of this type. This obligation could not plausibly flow from the nature of action. But the types of moral relationships that Hegelian constitutivism derives from the nature of action set the groundwork for more robust moral relationships; relationships where persons recognize wide-ranging rights and duties. In other words, Hegelian constitutivism obligates persons to take other persons seriously. These relationships allow for thoughtful moral conversations and the expansion of the minimal moral relationships required for action.
Hegelian constitutivism can also help us think differently about the normative sceptic. In particular, Hegelian constitutivism can address the moral sceptic. The sceptic, as traditionally conceived, looks for an objective, a priori response to the question, ‘Why be moral?’. According to Hegelian constitutivism, such an answer is not possible. This version of constitutivism can answer the question ‘Why should I take any opinions of others seriously and not coerce all others?’. Once the types of relationships that Hegelian constitutivism requires are established, we can give the sceptic a plausible response to his big question: ‘Ask the people you trust’. If the sceptic trusts these others to tell them what counts as writing a good poem, or being a good teacher, they can plausibly trust them to answer his sceptical questions, whether global or local. Indeed, the sceptic may locate a totally separate community to answer their questions about morality. As before, the sceptic is not obligated to take these answers seriously, but this is not a problem for Hegelian constitutivism. To ask a version of constitutivism to do more than this, I claim, is unreasonable. There is no intuitive reason to think that we can provide an unassailable answer to a moral sceptic by appealing to the nature of action. But constitutivism can tell the sceptic where to look for an answer and why to look there.
VI. Conclusion
Constitutivism provides a refreshing approach to meta-ethical issues. But its merits should not be exaggerated. Many forms of constitutivism promise too much and rest on controversial conceptions of action. Hegel provides the materials for a conception of constitutivism that grounds morality in action and human relationships. It rests on an intuitive conception of action and promises just enough. Hegelian constitutivism helps to answer our meta-ethical questions to the extent that philosophy can help. As Hegel believed, the rest of the moral work must be done on the ground, in conversations between persons who trust each other. Our moral norms develop historically, within communities, accompanied by fits, starts, resolutions, new controversies, and new solutions.
