1. The Neo-Aristotelian Teleological Account of Practical Identities

Although one does not need to be an Aristotelian to endorse the idea that practical identities are a source of reasons for action, some passages in Aristotle’s works unambiguously tie the aim of our actions to such identities. Consider, for instance, Book III of the Nicomachean Ethics, where Aristotle underlines that the bearers of specific practical identities do not deliberate about the ends they ought to pursue. Such ends do not depend on our individual preferences, Aristotle contends, but are internal to our identity:

The point here is that it would be absurd for a physician to wonder if she really ought to cure her patients, as the very point of having physicians is to make sure that patients will be cured. In Aristotle’s perspective, the end ‘to cure’ teleologically belongs to the practical identity ‘physician’: the very function of physicians is to promote health, and this fact allows us to conclude that physicians have reasons to cure their patients. What physicians may deliberate about is not whether they ought to pursue the end that belongs to their practical identity (independently of their preferences), but how they should pursue this end given the circumstances of the case at hand.

Let us note that one can accept this teleological conception of practical identities without endorsing Aristotle’s wider and philosophically controversial natural teleology. More specifically, I can accept that it is a physician’s function to cure her patients without simultaneously believing that human beings as such also have a natural function (ergon). I may also believe that only some practical identities have internal ends, but that the identity ‘human being’ itself does not. In many contemporary Aristotelian accounts of the normativity of practical identities, indeed, Aristotle’s natural teleology plays little to no role. According to his own words, Alasdair MacIntyre’s most influential work—After Virtue—does not rely on what he defines as Aristotle’s “metaphysical biology.”^{Footnote 5} Quite the contrary, its philosophical basis is a teleological understanding of identities and practices that is “rooted in the forms of social life.”^{Footnote 6}

To see this, consider MacIntyre’s Aristotelian account of the practical syllogism, that is, the reasoning through which an agent determines how she ought to act in a concrete situation. By constructing the major premise of this syllogism, the agent first identifies an end she ought to pursue. Let us imagine, for instance, a soldier who considers that she ought to defend her city-state. Then, by constructing the minor premise of her practical syllogism, the agent defines the circumstances of the case at hand. In our hypothetical scenario, let us say that the city-state to which our soldier belongs is under attack and that she has the means to join the fight against the invaders. Finally, MacIntyre follows Elizabeth Anscombe by claiming that the conclusion of the practical syllogism must be an action: our soldier now runs to the battlefield with her comrades and starts combating the attackers.

What interests us here is the major premise of the practical syllogism: how does an agent manage to identify the ends that she must pursue? MacIntyre’s answer is that she must first determine which social practices she is engaged in, and then identify which goods are internal to these practices. The end internal to the practice of playing chess is to win chess games and, as Aristotle suggested, one end that is internal to the practice of medicine is to cure patients. In MacIntyre’s view, practices involve “standards of excellence and obedience to rules as well as achievement of goods” that teleologically belong to them, and entering those practices is to accept the authority of those standards and goods.^{Footnote 7} Like the physician who wonders if she really ought to cure her patients, an Olympic runner who questions the idea that the end of Olympic running is to win races is asking one question too many.

A second non-naturalistic teleological account of the normativity of practical identities can be found in the works of French Aristotelian philosopher Vincent Descombes, who was also largely influenced by Anscombe’s reflection on practical reasoning. Descombes more specifically relies on such an account to provide us with an answer to Anscombe’s influential ‘rational Nazis’ challenge.^{Footnote 8} In Anscombe’s thought experiment, a group of Nazis are caught in a trap where they are sure to be killed. They nonetheless can exterminate Jews before they die. In fact, they are firmly committed to doing so under the pretext that “it befits a Nazi, if he must die, to spend his last hour exterminating Jews.”^{Footnote 9} What is disturbing in Anscombe’s hypothetical scenario is that the imagined individuals use a teleological understanding of their practical identity qua Nazis to justify the pursuit of a morally reprehensible end. Such an end is internal to their identity: it befits a Nazi to pursue it. This raises the worry that the neo-Aristotelian teleological account of practical identities is a poor guide to morality. It may be true that exterminating Jews is an end internal to the practical identity ‘Nazi,’ but this surely does not constitute a moral reason to pursue this end. In fact, if we found that such an account authorizes individuals to engage in genocidal violence, we may be tempted to reject it as a starting point for the philosophical reflection about what agents ought to do.

In Descombes’s view, however, we can resist the conclusion that the neo-Aristotelian’s discourse on practical identities legitimizes such violent practices. In order to prove Anscombe’s Nazis wrong, Descombes contends, Kantians are likely to argue that such individuals violate moral principles that all rational agents must respect to act rightly. They will then face the challenging task of providing us with a philosophical justification of these universal moral principles, one that Kant himself attempted to accomplish in the Groundwork. According to Descombes, however, there is no need to resort to Kantian universal principles to demonstrate that Nazis act wrongly as a less metaphysically demanding solution is available. To philosophically defeat Nazis—that is, to rationally demonstrate that they are wrong to pursue a grossly immoral end—what one should do is use the ‘normativity of practical identities’ thesis against them. How so?

Let us start from the fact that our practical identities are complex. Many individuals have a profession, but are also part of a family, have friends, and may be part of associations such as political parties or religious groups. Moreover, many moral dilemmas come from the fact that the ends internal to our practical identities are often difficult to pursue simultaneously. It may be my role qua friend to help my colleague move and my duty qua father to attend my daughter’s martial arts exam, but it is always possible that I will not have time to do both and be forced to choose between the two.

In Descombes’s view, the same goes for Nazis. More specifically, real Nazis are not like the ones in Anscombe’s thought experiment, that is, individuals about whom we know nothing else than the morally reprehensible end they are firmly committed to pursuing. They also bear complex identities from which conflicting reasons for action and obligations can spring. To use one of Descombes’s own examples, imagine a Nazi who bears two practical identities that create significantly different obligations for him: he is a member of the NSDAP, but he is also the rector of a renowned university. Now, the rector judges that it befits him to do anything he can to support the NSDAP. In his view, he has only one objective qua Nazi and qua rector: to help the party’s leader achieve his political objectives. The Nazi rector is right to think that it befits him to act in such manner qua Nazi, argues Descombes, but he is mistaken in believing that it also befits him to act in such a way qua rector. The end internally tied to the practical identity ‘rector’ is to promote higher education and research, not to support a political party by agreeing with everything its ‘enlightened’ leader desires to accomplish, even when this hinders the pursuit of knowledge. When our rector behaves in this ideological manner, he may be an excellent Nazi, but he is nonetheless a mediocre rector.

Descombes’s example is intended to show that agents can make rational mistakes by misidentifying the ends and standards of correctness that internally belong to their practical identities. The Nazi rector’s mistake is not to believe that his identity qua Nazi is to promote the political objectives of the NSDAP, but to consider that promoting these objectives is also his duty qua rector. Indeed, the idea that a good rector is one who subordinates knowledge and truth-seeking to ideological ends is just as absurd as claiming that the main end an Olympic runner ought to pursue is to bake good cakes. Once again, the underlying Aristotelian idea that unites these two cases is that agents cannot choose to assign external ends and standards of correctness to the practices in which they engage. Instead, practical identities and their associated practices have internal ends and standards of correctness that agents ought to respect:

In the end, Descombes believes that Nazis will be rationally defeated when we demonstrate that their ideological way of thinking rests on their philosophical inability to identify the ends that internally belong to the practices in which they engage. It is true that the end pursued by the Nazi rector is internal to his identity qua Nazi, but when he attributes this end to all dimensions of his identity, he makes the rational mistake of replacing the end internal to such dimensions by an external one. It is in this very manner, argues Descombes, that monomaniacal reasoners who pursue one single objective are irrational: by concentrating on one sole end, they assign such an end to practices to which it does not teleologically belong, and fail to recognize that their practical identities may create conflicting duties. Defending the ideology of the NSDAP at all costs is incompatible with promoting truth-seeking, but the Nazi rector remains oblivious to this fact.

Nevertheless, the neo-Aristotelian teleological understanding of practical identities faces an important problem. One implicit premise of Descombes’s reasoning is that the Nazi rector is irrational because he is unaware that his duty qua rector is not to promote the political objectives of the NSDAP, but to encourage the pursuit of knowledge. Yet, we can easily imagine a Nazi rector who understands very well that he is failing at his duties qua rector by pursuing ideological ends, but nonetheless chooses to do so. In other words, this Nazi rector makes the conscious decision to give priority to his practical identity qua Nazi over his practical identity qua rector. ‘Good Nazis are bad rectors,’ he thinks, ‘but being a good rector is not as important as being a good Nazi.’ Would such an individual make any rational mistake? Ultimately, Descombes does not provide us with reasons to believe so, and philosophical questions therefore remain unanswered. Even if we admit that practical identities are a source of reasons, indeed, do we also have reasons to endorse specific practical identities and reject others? If so, where do these reasons come from?

In the following two sections, I will argue that Kantian constructivism should be understood as a philosophical attempt to provide a satisfying answer to these questions. If practical identities are a source of reasons for action and create binding obligations, we must understand how they acquired this normative power as well as which identities it is morally appropriate for individuals to bear. As I suggested earlier, Korsgaard argues that the normativity of practical identities ultimately derives from the value of humanity. Let us then examine the argument that supports this conclusion, beginning with the Kantian theory of action on which it rests.

2. Acting According to One’s Own Law

The first step of Korsgaard’s argument regarding the value of humanity is a defence of the Kantian claim that free actions can only be performed by an agent who acts according to her own law. In her view, an individual lacking the power to act according to a law she gave to herself would necessarily lack free will. To see this, imagine an individual who is solely moved by her strongest current desires. Such an individual is also deeply unreflective: she consistently follows the beckoning of desire without ever noticing that she does. Korsgaard argues that this individual would not count as an agent capable of free will. She would rather resemble an automaton who can respond to stimuli, but who lacks the cognitive apparatus and decisional power required to be free. Note, however, that this conclusion would not be true if our individual reflectively chose to only act on her strongest current desires, as she would then act according to a law she gave to herself. In this case, this law would be the following: ‘Act only to fulfil your strongest current desires.’^{Footnote 11}

Korsgaard’s claims relate to Kant’s considerations on free will in the Critique of Pure Reason, where he explains that the human will is an arbitrium liberum, not brutum, “because sensibility does not render its action necessary.”^{Footnote 12} In the human being, “there is a faculty of determining oneself from oneself, independently of necessitation by sensible impulses,” and the presence of this faculty in ourselves makes us free agents.^{Footnote 13} More specifically, Korsgaard is committed to the Kantian view that free will necessarily presupposes reflectivity. To be free, agents must be able to stand back from their desires and choose whether they represent appropriate motives on which to act:

In Kantian terms, the principle of choice on which you act is the law you give to yourself. At this stage of the argument, however, it remains unclear why such law must be one that you give to yourself. After all, why could you not act freely according to a law that someone else gave you? In which sense, exactly, is this claim contradictory (if it is at all)?

Korsgaard’s answer is that acting according to one’s own law—that is, acting autonomously—is part of the structure of free agency itself. When I act according to a law you gave me, I am the one who decides to do so, and this law thus becomes the principle of choice I give to myself. In other words, unless you constrain me to comply with a law of your choosing, what I do when I act on it is to make your law my own. As an agent, “nothing is a law to you except what you make a law for yourself.”^{Footnote 15}

Now, if acting freely amounts to acting according to one’s own law, are there any constraints on the content of this law? Korsgaard does not believe that there are; the principle that one must act according to one own’s law leaves the substance of this law undetermined. For Kant, the problem of free will is that such a will must give itself its own law, but nothing determines what this law must be: “all that it has to be is a law.”^{Footnote 16} On Korsgaard’s reading, the Categorical Imperative, and more specifically its first formula (i.e., the Formula of Universal Law), is the philosophical principle that expresses this very idea:

In Korsgaard’s Kantian perspective, asserting that one must act according to one’s own law simply is alleging that one must comply with the Categorical Imperative. Note that, by defending such a view, Korsgaard remains faithful to the methodology of the third section of the Groundwork for the Metaphysics of Morals, where Kant attempts to demonstrate that freedom itself entails the Categorical Imperative or, in his own words, that once “freedom of the will is presupposed, morality together with its principle follows from it by mere analysis of the concept.”^{Footnote 18} Nevertheless, Korsgaard’s version of Kant’s argument is deliberately idiosyncratic insofar as it clearly distinguishes between the Formula of Universal Law and the Moral Law proper. While the Formula of Universal Law is devoid of any moral content, she contends, the Moral Law is not. Yet, the reasoning we examined so far only intends to show that acting freely amounts to respecting the Formula of Universal Law, and an additional argument must be furnished in order to bridge the gap between this formula and the Moral Law.^{Footnote 19}

As I suggested earlier, Korsgaard ultimately defends the idea that we must value humanity as an end in order to rationally value anything at all, and this view is more reminiscent of the second formula of the Categorical Imperative—the Formula of Humanity—than its first. Contrary to the Formula of Universal Law, the Formula of Humanity—“act that you use humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means”—clearly has substantive moral content.^{Footnote 20} How is it then possible to move from the first formula to the second while remaining, as Kant wanted, within the confines of a priori philosophical reflection? It is in her discussion of this philosophical question that Korsgaard encounters the neo-Aristotelian teleological account of practical identities.

3. Valuing Practical Identities: The Kantian Constructivist Perspective

According to the theory of agency I just sketched, a free action is an action committed according to one’s own law the content of which remains indeterminate. Nothing, yet, precludes the agent from providing this law with any content of her choosing. A patriotic individual could consider, for instance, that the moral and social norms of her nation are more valuable than anything else and consequently give herself the following law: ‘act only so that the results of your action contribute to the glory of your nation.’ It befits a patriot, she thinks, to act according to such a law. While it is true that she does not always respect the Formula of Humanity by acting in this manner—she fails to value, for instance, the humanity of agents who are not her compatriots—she nonetheless respects the principle according to which she must act according to her own law. The question we face, then, is to know whether the Kantian constructivist can demonstrate that our patriot fails to act rationally when she violates the second formula of the Categorical Imperative.

It is certainly Korsgaard’s ambition to argue that she does. In her perspective, the patriot reasons like the neo-Aristotelian when she tries to achieve the end that is internal to her identity qua patriot, but doing so is not a sufficient condition of rational action. More precisely, the patriot’s mistake is that she does rationally assess the value of her practical identity qua patriot before attempting to achieve its internal end. In other words, Kantian constructivists accept that our identities have internal ends and that such identities are a source of reasons for action. As Korsgaard notes, “we may begin by accepting something like the communitarian’s point. It is necessary to have some conception of your practical identity, for without it you cannot have reasons to act.”^{Footnote 21} Such a claim, however, is only the beginning of the philosophical reflection on morality, which neo-Aristotelians like MacIntyre and Descombes do not carry to its conclusion. Surely nobody would object to the idea that physicians have reasons to cure their patients, maybe even that it befits a Nazi to commit genocidal actions, but the important enquiry begins when I try to assess the moral value of such practical identities.

To be fair, neo-Aristotelians do not ignore this idea. On the one hand, communitarian Aristotelians like MacIntyre argue that valuable practical identities help us realize the ends internal to the philosophical tradition to which our community belongs.^{Footnote 22} On the other hand, neo-Aristotelian naturalists contend that acting well amounts to pursuing the ends that are internal to my practical identity qua member of the human species.^{Footnote 23} For the Kantian, such answers remain unsatisfying as they do not put an end to the philosophical query. What reasons do I have, indeed, to consider that my identity qua member of a specific community or qua member of the human species has value and that I can therefore consider it to be a source of moral reasons?

Korsgaard’s transcendental argument about the value of humanity precisely aims to answer this question. Practical identities are a source of reasons for action, but the reasons that stem from a specific practical identity are not themselves reasons to accept or reject this identity. Being a fireman gives me reason to extinguish fires, but it does not give me reasons to become a fireman. How can I therefore rationally ground my choice to be or not to be a fireman? Are there properly moral reasons to decide in one way rather than the other?

The Kantian constructivist’s answer is that the moral value of specific practical identities needs to be assessed from a standpoint external to those very identities, and more specifically from the standpoint of rational human agency. It is qua rational human agents that we must assess the value of being a father, rector, or Nazi. Without this standpoint, our decision to endorse specific dimensions of our practical identity will be unjustified. But what is so special about the standpoint of rational human agency? What makes my decision to determine what I morally ought to do from this standpoint less arbitrary than the neo-Aristotelian choice to do so by considering my identity qua member of this specific community or qua member of the human species?

As Korsgaard notes, most of our practical identities are contingent: “you are a mother of some particular children, a citizen of a particular country, an adherent of a particular religion, because of the way your life has fallen out.”^{Footnote 24} Because they are contingent, we may come to call the importance of such identities into question, and maybe even stop attributing any practical importance to them. If the members of my family never treated me well, I may simply reject claims that it is important for me to act qua son or brother or even consider that I am no longer, in a purely moral and non-biological way, the son or brother of person x. This practical identity will then stop having practical force: I will no longer consider it to be a source of reasons for action. There is, however, one dimension of my identity that is not contingent: my identity as rational human agent. Whatever I do, I cannot stop relying on my rational powers when I try to assess the value of my practical identities. This fundamental identity is necessarily mine: the one with which I will be left if I stop attributing importance to all my contingent practical identities. Rational human agency is the fundamental standpoint from which I assess their value, and Korsgaard believes that this makes it the identity from which the value and normativity of our contingent practical identities spring. It is, in her view, moral identity per se:

Here, Korsgaard is merely summarizing an argument of which I will offer a piecemeal reconstruction and critical discussion in the next section. Before I do so, however, note that such an argument does not ground the Formula of Humanity by itself. Indeed, what it intends to show is that I must value my humanity to also value my contingent practical identities, but the Formula of Humanity requires me to value my humanity and the humanity of others. Korsgaard is well aware of this fact and offers a second argument to support the conclusion that I must value the humanity of others if I value my own.^{Footnote 26} Yet, this second argument is not the one on which I want to focus as it presupposes that the first argument succeeds, but I will argue that it fails. Even if valuing my own humanity entails that I also value the humanity of others, I contend, we have reasons to contest the idea that I must value my humanity to value my contingent practical identities.

4. Humanity, Well-Being, and the Moral Force of Practical Identities

I propose the following reconstruction of Korsgaard’s transcendental argument regarding the value of rational human agency:

1. (1) In order to rationally value anything at all, I must have reasons to do so.

2. (2) In order to have reasons to value anything at all, I must value a specific practical identity from which these reasons will spring.

3. (3) In order to rationally value a specific practical identity, I must have reasons to do so.

4. (4) In order to have reasons to rationally value a specific practical identity, I must value my identity qua rational human agent from which those reasons will spring.

5. (C) In order to value anything at all, I must value my identity qua rational human agent.

Such a reconstruction helps us understand why Korsgaard is inclined to describe her reasoning as a transcendental argument; each premise establishes a necessary condition of a specific form of valuing and the resulting conclusion is that valuing my identity qua rational human agent is the necessary condition of all forms of valuing.

To assess the validity of Korsgaard’s argument, let us first examine the premises. The least controversial of the four premises are (1) and (3). Surely, in order to Φ rationally, I must have reasons to Φ, at least if we understand rationality as the cognitive power to respond to reasons. More contentious are premises (2) and (4). While (2) expresses the neo-Aristotelian idea that reasons for actions spring from contingent practical identities, (4) more radically asserts that my reasons to endorse those very identities necessarily spring from a more fundamental identity: my identity qua rational human agent. Yet, it remains possible to accept (2) while rejecting (4). In other words, one can admit that contingent practical identities are a source of reasons without accepting that such identities can themselves only be rationally endorsed from the standpoint of a rational human agency.^{Footnote 27}

More specifically, my suggestion is that an individual can rationally assess the value of one of his contingent identities by using the reasons that spring from the other contingent identities he bears. Let me clarify this idea with an example:

In this case, Miguel rationally assessed the value of a specific practical identity he bears—his identity qua Catholic—without relying on his identity qua rational human agent as such. It is not qua rational human being that Miguel cannot accept the priest’s conservative discourse, but rather qua patriot and qua physician. If it is possible for us to behave like Miguel, then this fact will undermine Korsgaard’s claim that I must rely on the standpoint of humanity to rationally assess the value of my contingent practical identities. Of course, one could object that Miguel’s strategy of rational evaluation is not a very good one and that it would be desirable that he also rely on his identity qua rational human agent to perform this task. Miguel could wonder, for instance, whether the priest’s discourse respects the humanity of others as an end.^{Footnote 28} Note, however, that, in order for Korsgaard’s transcendental argument to succeed, Miguel’s behaviour must be impossible, not merely undesirable. As I suggested, Korsgaard’s transcendental argument portrays reliance on the standpoint of humanity as a necessary condition of valuing my contingent identities. Yet, a coherentist reasoning about the value of his practical identities is available to Miguel, who need not rely on the standpoint of humanity.

There is nevertheless a second manner in which one can criticize Korsgaard’s reasoning without, this time, contesting the validity of premise (4). For the sake of the discussion, let us concede that the standpoint of rational human agency really is inescapable, that is, that we cannot rationally assess the value of our contingent practical identities without relying on it. Let us also grant that whoever relies on the standpoint of humanity attributes some value to it. It would be a curious decision, after all, to rely on the standpoint of humanity if we envisioned it as devoid of any value. What we must now determine is how the value of humanity compares with the value I attribute to my other practical identities. Does the fact that my identity qua human is inescapable necessarily entail that it is more valuable than my contingent identities?

Korsgaard seems to assume that it does, but she does not offer a detailed explanation of why this is so. In her view, if I find myself in a situation where the obligations tied to my identity qua rational human being conflict with the obligations that stem from my contingent identities, I will have to prioritize the former over the latter. Let us consider, for instance, G.A. Cohen’s case of an idealized Mafioso who lives by a code of strength and honour and “does not believe in doing unto others as you would have them do unto you.”^{Footnote 29} In fact, he correctly believes that it befits a Mafioso to commit certain actions contrary to the Formula of Humanity so that his identity qua human conflicts with his identity qua Mafioso and that he will often have to choose between the two. Korsgaard believes that if the Mafioso is reflective and actively tries to rationally assess the value of his practical identities, he will unavoidably conclude that complying with his obligation as human is more valuable than fulfilling his Mafioso duties. As Korsgaard states, “his obligation to be a good person is […] deeper than his obligation to stick to his code.”^{Footnote 30}

Yet, even on the assumption that my identity qua human is rationally inescapable, it remains unclear why its inescapability entails that it has more value than my other identities. Why is it so, indeed, that such an identity is the most valuable of all just because I must rely on it to assess the value of my contingent identities? Does inescapability have by itself the power to confer the greatest value on things? My suggestion is that it does not. Even if I admit that humanity has value, I can still believe that my contingent practical identities are more valuable than my identity qua rational human agent. If I am right, it will then be possible for me to rationally prioritize my obligations qua father, professor, physician, best friend, etc. over my duty to comply with the Formula of Humanity.

Imagine, for instance, that Teresa, a paramedic, recognizes the value of humanity, but is inclined to prioritize her obligations as paramedic over her duty qua human being. By way of example, she often lies to individuals she assists qua paramedic by telling them that their injuries are not serious. This keeps them calmer, she finds, and facilitates her work. If Kant was right about lying, Teresa does not fully respect the humanity of the individuals she assists, and thereby violates the Formula of Humanity. Note, however, that her actions promote the well-being of the people it is her professional duty to help.

Is Teresa making a moral mistake? Possibly, but there remains a missing link between the outcome of Korsgaard’s transcendental argument—that is, the idea that humanity is necessarily valuable—and the claim that would allow us to establish that she did make such mistake. For the conclusion that I must attribute value to my identity qua rational human agent does not entail that I must also believe that it is the most valuable of all identities, and that the obligations that spring from it systematically trump the obligations tied to my contingent practical identities.

So, I may, like Miguel, assess the value of one of my contingent practical identities by relying solely on my other contingent practical identities or, like Teresa, recognize the value of humanity without considering that the obligations that stem from it override all my other duties. While the first case shows that my identity qua human being is not rationally inescapable, the second demonstrates that it is not necessarily more valuable than all contingent practical identities even if we assume that it is, indeed, inescapable. If these conclusions are plausible, we have grounds to (i) reject Korsgaard’s transcendental argument that aims to establish humanity as the source from which the normativity of all contingent practical identities ultimately derives, and (ii) reject the claim that our identity as rational human agent bears more value than any other conceivable contingent practical identity.

We have come full circle and yet, this is philosophically troubling. More specifically, we are still looking for a satisfying account of the normativity of practical identities. As we have seen, neo-Aristotelian thinkers like MacIntyre or Descombes assume that practical identities are normative, but do not provide us with a detailed explanation of why this is so. Moreover, their view is vulnerable to a problem illustrated by Anscombe’s rational Nazi scenario; if we simply assert that practical identities provide us with reasons to pursue ends that teleologically belong to them, we will be forced to admit that the bearers of problematic identities have reasons to pursue morally reprehensible ends. It befits a Mafioso or a Nazi to commit actions we deem unacceptable, so we need an account of why individuals should not be Mafiosi or Nazis in the first place. In other words, we need a point of view from which we can determine which practical identities it is morally appropriate for us to bear and which ends it is acceptable to pursue. The strength of Korsgaard’s proposal is to acknowledge that we need such a point of view, but I ultimately found her philosophical defence of the standpoint of rational human agency to be lacking. Is there any other point of view from which we can critically assess the value of our contingent practical identities?

My final proposition is to assess the value of practical identities by considering their potential contribution to our well-being and the well-being of others. More specifically, we should endorse a version of Joseph Raz’s humanistic principle and assert that the justification of the goodness or badness of practical identities derives ultimately from their contribution, actual or possible, to human life and its quality.^{Footnote 31} The main advantage of evaluating practical identities from the standpoint of well-being is that it dispels the mystery surrounding their normativity. All human beings must fulfil certain basic physiological and psychological needs in order to meet minimal standards of well-being, and the practical identities we see as truly normative are the ones that help us to do so. Physicians, educators, and psychologists are morally obligated to pursue the end that is internal to their practical identities because the consequences of them not doing so would entail a significant decrease in our well-being. Our life quality depends on them doing so, and it is their contribution to our welfare that gives moral value to the practices in which they engage.

By focusing on well-being, we are also in a position to identify practical identities that, conversely, have no moral value. I already raised doubts about the effectiveness of the philosophical strategy used by Descombes against the monomaniacal Nazi rector: when shown that his end qua Nazi conflicts with the internal ends of his other practical identities, such an individual can always respond that these identities are unimportant compared to the one political identity he truly values. Surely, Descombes is right that we will then be able to argue that he is a mediocre rector, but the Nazi can then countercharge that academic activities have no value in themselves, but only insofar as they serve ideological objectives. In doing so, he would not deny the Aristotelian idea that a good rector must pursue the end that is internal to his identity, but more radically reject the claim that he ought to be a good rector in the first place. In the end, Descombes underestimates the rational resources of the individual who is committed to pursuing a morally reprehensible end. In contrast, arguing from the point of view of well-being will be considerably more difficult for the rational Nazi. Experienced orators may convince us, if they are highly skilled, that fervently supporting the NSDAP will promote their own well-being and the well-being of their fellow countrymen, but it will not be difficult for us to identify whose well-being is likely to be negatively affected as a direct result of the NSDAP’s political activities.

One possible objection to my suggestion is the claim that the standpoint of well-being really is the standpoint of humanity in disguise, that is, it roughly corresponds to the Kantian constructivist’s discourse on rational human agency. It is true that such discourses will likely overlap: to guarantee a certain amount of well-being to each human being, for instance, we may conclude like Kantians that certain individual rights shall never be violated. It is also true that the standpoints of well-being and humanity serve a similar function, that is, they represent a perspective from which the value of all contingent identities derives, and without which it will be arduous—if not impossible—to counter-argue against the bearers of morally problematic practical identities. Yet, the proponents of each of these two standpoints will employ a significantly different methodology in order to determine their content. The Kantian constructivist’s ambition is to demonstrate a priori that the standpoint of humanity is rationally inescapable and that humanity therefore has more value than any other contingent identity. As Korsgaard’s argumentation makes clear, she is also committed to the idea that judging from this standpoint amounts to endorsing the Formula of Humanity. In contrast, the nature of human well-being cannot be determined a priori. Even the proponents of objective list theories of well-being according to which living a good life does not depend—or not only depends—on having one’s preferences satisfied attempt to identify human goods that empirical communities value.^{Footnote 32} In fact, it is unlikely that we will attain a satisfying definition of well-being while neglecting Aristotle’s suggestion to begin by examining endoxa, that is, prominent judgements regarding what it is for a human life to go well.

A fully fledged defence of the normativity of well-being goes beyond the scope of this article, but I can already concede that following Aristotle’s dialectical method will undeniably have drawbacks. Indeed, it will be laborious and time-consuming, as it is likely to lead us to cross-disciplinary boundaries and conceive of moral enquiry as going hand-in-hand with the sociological and anthropological study of people’s considered judgements with regard to the constitutive elements of well-being: pleasure, health, knowledge, friendship, autonomy, etc. Surely, such an approach is less elegant than the Kantian constructivist’s alternative methodology; it does not rely on transcendental moral proofs, nor does it lead us to principled certainties. Quite the contrary, tying the reflection on morality to the appraisal of empirical opinions regarding well-being may yield cross-cultural discrepancies and generate strenuous attempts to find overlaps between them.^{Footnote 33} In fact, if the Kantian constructivist’s metaethical project of grounding moral universalism in a priori reasoning had shown successful, the temptation would be strong to conclude that there is no compelling reason to undertake such a taxing empirical enquiry. Yet, I have provided reasons to believe that such a project ultimately fails. Hopefully, the mishaps of contemporary a priorism will pave the way for thinkers who intend to construct moral universality a posteriori, that is, by coupling philosophical reflection with socio-anthropological observation.