1. The Nature Objection

This section introduces the “Nature Objection” and explains why it’s a (neglected but) serious problem for relational approaches to ethics.

Let us call the following argument the “Nature Objection”:

This argument is valid, so let us focus on why each of its two premises might be true.

Start with P1. We can start to characterize relational approaches to ethics negatively, contrasting them with standard non-relational accounts of morality: utilitarianism, orthodox Kantianism, and orthodox virtue theory. The utilitarian seeks to explain moral obligation in terms of the fundamental, non-relational obligation to produce states of affairs with net positive utility (Mill, Reference Mill and Sher2002; Sidgwick, Reference Sidgwick1874). Orthodox Kantianism explains moral obligation in terms of an intrapersonal requirement of rational consistency. Later, intuitionist versions of deontology take as fundamental an aggregate of self-standing duties, such as the duty of fidelity or non-coercion (Ross, Reference Ross1930). Finally, orthodox virtue theory explains moral obligation in terms of what a good or virtuous agent would do, i.e., primarily in terms of the relation between an agent and her motivations.

Two important qualifications are in order. First, these are merely analytic stick figures. Second, what might qualify these normative theories as non-relational is that they treat relational properties, like claims, obligations to others, or interpersonal relationships, as important but ultimately derivative. By contrast, let us say that for relational approaches to ethics interpersonal relations are fundamental and ineliminable in our understanding of moral obligation.

Philosophers then offer competing accounts of the nexus between moral obligation and interpersonal relations. Some views are strongly influenced by a Strawsonian approach, privileging the reactive attitudes. For instance, Stephen Darwall emphasizes the central importance of the “second-person standpoint,” consisting in second-personal reasons and our competence and authority to make claims of each other, holding each other accountable through blame (Darwall, Reference Darwall2006, Reference Darwall2013b, Reference Darwall2013c, Reference Darwall2024). R. Jay Wallace shares Darwall’s emphasis on Strawsonian reactive attitudes and blame but insists that moral obligation depends on relational claims we have standing to make of each other (Wallace, Reference Wallace, Bakhurst, Hooker and Little2013, Reference Wallace2019). Both Darwall and Wallace seem to share the idea that a focus on second-personal reasons or directed duties abandons the attempt to explain moral obligation in terms of logically independent values, goodness-facts, or reasons. For instance, the fact that my standing on your gouty toe causes you pain is not, in itself, a necessary or sufficient condition of moral obligation, since the pain is a “third-personal” or, as Darwall calls it, an “outcome-based reason” to move your foot (Reference Darwall2006, p. 8). As Janis Schaab nicely puts it: “through such attitudes, [the foot-stomped victim] would implicitly address [you], demanding that [you] comply with her obligation” (Reference Schaab2023, p. 2).

Yet, it seems to me that a relational approach to ethics need not be committed to a Strawsonian picture of accountability and reactive attitudes. For instance, Margaret Gilbert’s work attempts to explain normative properties and their relationality in terms of both the nature of rights as demands and, in turn, such demands as constitutive elements of joint-decision making.Footnote ² And Michael Thompson’s influential and loosely neo-Aristotelian approach attempts to foreground the bipolar structure of relations of justice (Reference Thompson, Jay, Pettit, Scheffler and Smith2004).

What matters for our purposes, then, is that P1 involves the basic thesis that moral obligation is second-personal/relational insofar as a second-personal and/or relational element is fundamental and ineliminable from our understanding of what it is to have an obligation in the first place. As minimalist as this picture is, it already helps to bring into focus the motivation for P2.

P2, Environmental Wrongs, says that some actions involving nature are morally wrong yet do not fit the second-personal/relational model. To flesh out this premise, consider three cases.

For the sake of argument, I’ll assume that each of these cases involves moral wrongdoing: being cruel to animals; chopping ancient trees for social media views; carelessly polluting rivers.

If true, this assumption raises a twofold challenge for relational approaches. On the one hand, these cases appear to have very poor fit with standard relational accounts. On the other hand, these cases appear to have a very good fit with standard non-relational ones.

Start with poor fit. Central to Darwall and Wallace’s account is the Strawsonian picture of blame and accountability. And central to their account is detaching the relational, deontic domain of morality from impersonal evaluative considerations, such as “outcome-based” reasons pertaining to welfare. The challenge is that reasons pertaining to welfare appear essential in helping us understand why we have moral obligations concerning nature. Further, reactive attitudes toward animals, plants, or habitats appear to misfire, since it makes no sense to be resentful when a tree branch falls on your head or a river floods your home. Viewed from the other angle, calves, trees, and rivers lack the capacities typically thought necessary to enter into Strawsonian relations of accountability.

Furthermore, in the relational ethics literature, when philosophers bring up the topic of our obligations involving nature (if they do at all), it’s usually to illustrate the very contrast between relational and non-relational models of obligation and wrongdoing.

Here are some representative examples (emphasis added in bold):Footnote ⁵

These diverse philosophical views share the idea that there’s a fundamental contrast between wrong action and wronging someone. Wrong action involving nature typically falls under the “non-relational” part. Therefore, P2 is true: we may well have moral obligations with regard to nature, but these obligations are not relational.

While relational ethics (however conceived) appears to have a poor fit with environmental ethics, the standard normative theories appear to have a good fit with it—lending further support to P2.

Orthodox utilitarians analyze at least the first case (involving the calf) in terms of the intrinsic disvalue of suffering. For example, Peter Singer argues forcefully in defense of animal liberation (Reference Singer1979, Reference Singer2009). Instrumental treatment of calves is wrong in virtue of the massive suffering such practices produce.

Orthodox Kantian accounts, in turn, analyze all three types of cases indirectly, i.e., in terms of how we have moral duties with regard to (but not owed directly to) beings in nature (plants, animals, ecosystems). We have such duties because being cruel to animals or indifferent to the integrity of nature is a hindrance to our own development as rational, moral agents (Ripstein and Tenenbaum, Reference Ripstein, Tenenbaum, Callanan and Allais2020). As in Owens’ example, paving over the Grand Canyon with concrete is wrong not directly because of what it does to nature, but indirectly because of what it does to us, say, our failure to appreciate aesthetic reasons.

Finally, going beyond utilitarianism and orthodox Kantianism, consider ecocentric holism. Holism appears to be implicit in the founding principle of the holistic, ecocentric account found in Aldo Leopold’s seminal “Land Ethic”:

Such an ecocentric account could explain the moral wrongness of being cruel to the calf, indifferent to the redwood, or reckless with one’s pollution by saying that any such acts hinder rather than preserve the “integrity, stability and beauty” of the biotic community. Yet, these are not relational wrongs. Just as the orthodox utilitarian sees past individuals to the aggregate amount of pleasure, so too the ecocentric view sees past individuals to properties attaching to the biotic community as a whole. Individuals appear merely as derivative accidents.

In sum, what supports the truth of P2 is both that standard relational accounts struggle to explain our moral obligations concerning nature and that more familiar, non-relational accounts appear to explain them better. Yet, when we couple P1 with P2, we get our conclusion by modus tollens: a relational account of ethics is not the correct account of our moral obligations.

There is an obvious rejoinder to the Nature Objection. We could deny P2 by seeking to explain Environmental Wrongs in terms of a special type of agency relation, namely, our acting on behalf of beings in nature. Stephen Darwall embraces (but, to my knowledge, does not fully explore) this strategy:

Similarly, Scanlon suggests: since animals (and plants) cannot be owed contractualist justification, they are not part of the contractualist framework. Still, we can contract on their behalf, as their trustees.Footnote ⁶

However, the agency strategy faces three difficulties. First, it is too indirect: if there are directed duties in the offing, it seems most intuitive to say that they are owed to the agent, i.e., the person making a demand on behalf of the plant or animal. But if so, when I kill a dog for fun, I’m not wronging the dog, but, say, the dog’s “owner” who can make a claim on the dog’s behalf. Such indirectness makes a relational account indiscernible from more familiar, non-relational accounts. Second, the explanation appears redundant. Why exactly do you get to act on behalf, say, of the veal or the redwood? If that’s because we have pre-existing duties owed to the veal or the redwood, then the agency strategy presupposes the explanandum. But third, suppose there is no pre-existing directed duty in the offing. Then it’s not clear what rational warrant there is for making a claim on behalf of animals or plants. Why not make claims on behalf of lumps of mud? In sum, the agency relation strategy does not appear promising. On the contrary, it appears to land us back into some familiar non-relational utilitarian or Kantian accounts of why it’s wrong, say, to harm calves or redwoods.

How to move forward? One option is to grant that the Nature Objection is sound. This may seem like a good outcome for relational ethics, insofar as relational ethics would be the proper account of human ethics—relegating familiar non-relational accounts to the domain of non-human ethics.

This would be a Pyrrhic victory. Once a large domain of the moral is excluded from relational ethics, the very unity and elegance of the model is at stake. If, say, utilitarianism or orthodox Kantianism are correct for thinking about non-human animals, why not hold that these models, after all, also work for other subdomains of the moral? How about our relations to human beings who do not or cannot fully develop rationality? How about considerations of well-being and welfare? How about larger issues of social justice that appear to not concern specific individuals, such as voting or paying taxes (Scanlon’s examples above)? The worry, in short, is that far from giving us the promised insight into the very nature of moral obligation, relational ethics can at best illuminate a very limited and narrow jurisdiction of the moral.

Fortunately, that’s not our only option. In what remains, I explore an alternative: I’ll argue that P2 is false. But to get there, we’ll have to rethink the overall framework of relational ethics.

2. Relational Perfectionism

My suggestion will be that we can get a sound account of our duties to non-human living beings by developing a distinctive form of perfectionism. Perfectionism is often associated with egoistic eudaimonism or consequentialism. However, drawing from Aristotle, Leibniz, and Kant, I’ll retrieve a non-consequentialist form of perfectionism in three steps. First, we need to retrieve the notion of immanent purposiveness as the structure of living beings. Then identify immanent purposiveness with a form of self-organization or self-determination. Finally, make explicit the essential nexus between self-determination and deontic properties: a self-determining being is necessarily a bearer of a fundamental claim to respect from others. Yet, such claim to respect is itself a type of moral power, in the metaphysical sense that its actualization requires relations of recognition.

Begin with purposiveness or teleology. A trope in the history of Western metaphysics is that with the advent of modern science, modern philosophers transformed the concept of causality, eliminating final causes from the natural world.Footnote ⁷

Very roughly, this Aristotelian background required that a complete explanation of a being or its activity appeal to each of the four causes (material, efficient, formal, and final).Footnote ⁸ Aristotle defined the final cause as “the end or that for the sake of which a thing is done” (Physics, 194a33). But Aristotle distinguished two basic forms of purposiveness (Physics 192b9-194b15).Footnote ⁹ One is endogenous or immanent: a being’s activity is oriented to the actualization of its own form.Footnote ¹⁰ The other is exogenous or imposed: a being’s activity is oriented to the actualization of a form imposed from the outside by the deliberate goals of a conscious agent. While Aristotle’s paradigmatic example of immanent purposiveness is the nature of organisms, his paradigmatic example of imposed purposiveness is the nature of artifacts, whose purpose is exogenously imposed by a craftsman. Indeed, in its elementary organic form, immanent purposiveness does not require intentional design, opening a path for understanding, say, plants as immanently purposive.

For some time, the standard story was that modern philosophers altogether eliminated final causality from the natural world. To see this, consider the illustrative example of E. A. Burtt.

It’s controversial among scholars whether modern philosophy left “no place for final causality whatsoever” in nature. Some scholars, like Margaret Osler, argue that modern philosophers eliminated immanent purposiveness from nature but left space for imposed purposiveness, making nature “the product of the divine artificer”(Osler, Reference Osler1996, p. 390). For our purposes, it does not matter which side is correct. What matters is their agreement that for many founding modern philosophers, immanent purposiveness was banished from nature. The only form of purposiveness left was exogenous or imposed, the intelligent purposiveness characteristic of the craftsman.

One of the clearest formulations of the thought can be found in Samuel Pufendorf’s highly influential distinction between “physical” and “moral” qualities—a distinction which Pufendorf popularized but which he probably owed to Erhard Weigel. A moral entity, Pufendorf claims, “depends on imposition, that is, on the decision of free agents as such, and these free agents … together with a tacit or expressed mutual agreement which has been entered into, have imposed morality upon things and persons” (Book 1, Def. 2, p. 24). While the material, natural world consists in the non-normative physical properties of bodies in motion, moral qualities emerge from the imposition of value by a rational intelligence.Footnote ¹¹ Pufendorf reintroduces purposiveness into the world by collapsing all purposiveness into Aristotle’s second form: imposed purposiveness.

Abstracting from Pufendorf’s strong voluntarism, we get the following model of normativity:

The basic idea is that normative properties are essentially extrinsic, the product of imposition by some form of intelligent activity.

Exogenous Purposiveness, it seems to me, is a general structure informing a variety of views about the good and the right. Nowadays, the view is often associated with constructivist forms of neo-Kantianism.

In this passage, Korsgaard explicitly links the unconditional value of an individual (their humanity; rational nature) with a form of value. Such value is fundamentally extrinsic: a value conferred by valuing or choosing.Footnote ¹²

Yet, Exogenous Purposiveness is not confined to neo-Kantianism. Crucially, one of the founding figures in environmental ethics adopts the same framework, J. Baird Callicott:

Callicott analyzes the normative status and the good of calves and redwoods as a form of Exogenous Purposiveness, one “projected by valuing subjects.” A virtue of thinking of value in this Humean way is that once you remove the anchor of value in the objective, intrinsic properties of things, non-instrumental value reduces to an attitude of valuing things non-instrumentally. And this means, Callicott argues, that we can value not only individuals (like calves and redwoods) but even species and ecosystems for their own sakes. In short, as neo-Kantians and Humeans can agree, calves, redwoods, and even ecosystems matter non-instrumentally because we value them non-instrumentally.

However, the great virtue of Exogenous Purposiveness is also its greatest liability. To bring the difficulty into focus, return to Callicott:

The main worry is a familiar Euthyphronic one: if the non-instrumental mattering of calves or redwoods depends on our valuing them, then it appears as if there is no rational warrant for such valuing. If, as a matter of psychological fact, we happened to not value them, we’d be making no mistake. In other words, if goodness depends on valuing, goodness becomes arbitrary: in principle, it can be just as warranted to think of a calf as deserving respect or as a morally neutral feature of our landscape apt to be served on a dish.

To sharpen the problem, consider the following hypothetical. Imagine a world like ours but devoid of rational animals. Lacking rational, valuing subjects, such a world (call it “W1”) would also lack normative properties. W1 would be a normative desert.

Now, imagine a world qualitatively and quantitatively identical to W1 except for one feature: now introduce a single species with the capacity to desire and value. Call this species “S” and this world “W2.” Suppose that members of S are a species of rational crows. The introduction of crows into W2 marks the momentous introduction of normativity into the world. Rational crows develop; form complex societies; even industrialize.

But some among the rational crows protest: “there are moral constraints on how we use other, non-rational animals – even how we use the rest of nature” some say. As a matter of social fact, this minority is silenced. The majority of rational crows treat other living beings as possessing, at best, instrumental value—their use to rational crow life.

Imagine now a third world, W3, where the distribution of the moral sentiments is exactly the converse of W2: the majority of rational crow society values other non-rational living beings non-instrumentally. And they take such non-instrumental valuing as the basis of their moral duties.

Now ask: granted that the quantity and quality of living beings in W1 is identical to that of W2 and W3 (minus the rational crows, of course). Then what justifies the widespread belief in W3 that non-rational-crow living beings matter non-instrumentally? If Exogenous Purposiveness is true, the answer seems straightforward: nothing. There is no rational warrant for such belief. The only difference is the psychological or social facts explaining the difference between W2 and W3—actual patterns in caring, valuing, or choosing among the rational crows. Yet, such differences appear arbitrary, unable to explain the truth of the normative judgments made by rational crows.

I do not present this as a knock-down argument against Exogenous Purposiveness. Instead, I simply mean to clarify and motivate the alternative, which takes the second horn of the Euthyphronic dilemma. Our duties toward living beings are anchored in features about them, not about the valuing subject. But what could such feature be?

The brief history we have considered offers us the first step of an answer: their immanent purposiveness. Yet, this leaves us with another puzzle: even if we reintroduced immanent purposiveness in nature, what’s the connection between immanent purposiveness and moral duty?

The answer I’d like to explore is a distinctive version of perfectionism.

The core thought derives from the Aristotelian tradition. The tradition connects goodness to being (and being to goodness). Goodness is the perfection or completion of an individual as an instance of its kind—a whole, complete specimen. But to say that a thing is perfect of its kind is to say that it’s fully realized or developed to the extent that the potentialities definitive of its kind are actualized (Stump and Kretzmann, Reference Stump, Kretzmann and MacDonald1991, p. 100). So, for Aristotle and Aquinas, there is an equivalence between saying that a thing has fully actualized its specifying potentialities and that it’s good of its kind.

Following David Brink, then, we may characterize perfectionism as follows:

As I’ll understand it, this version of perfectionism rests on Immanent Purposiveness. Normative properties are grounded in the perfection or full realization of a being’s essential nature precisely because such nature is not normatively neutral. On the contrary, internal to the being’s nature is a structure of purposiveness. I propose to clarify this structure by retrieving from Kant the thought that immanent purposiveness is a structure of self-organization or self-determination. Kant’s notion is explicit only in his discussion of the philosophy of biology, the second part of Kant’s Critique of the Power of Judgment. Philosophers working in contemporary ethics are typically not familiar with it. So let me explain Kant’s two principal conceptual moves briefly.

First is the distinction between the purposive and the non-purposive. Lumps of mud and stretches of sand are not purposive beings, but a watch washed on the shore is. What explains the difference? Picking up from Aristotle, Kant’s distinctive answer is that purposiveness is not a discrete property possessed by some beings but lacked by others. Rather, purposiveness characterizes the nature and organization of a kind of being. While the organization of a lump of mud or a beach as an aggregate of grains of sand makes no reference to purposiveness, the inner structure of a watch does. That’s because the inner structure and organization of the watch cannot be understood independently of purposiveness.

Second is the distinction between what I’ve been calling immanent and exogenous purposiveness—a conceptual move Hegel once labeled as one of Kant’s “greatest services” to philosophy.Footnote ¹³

Very briefly, Kant’s thought is that artifacts are organized beings in the sense that parts are functional organs of the whole. By contrast, something counts as a natural end in virtue of its self-organization. And self-organization just is the reciprocal relation of dependence of parts and whole for the sake of an end. Yet, such end does not depend on an externally imposed intelligent design. Rather, such organization is constitutive of the being, qualifying it as a natural end. Footnote ¹⁴

In particular, Kant argues that the self-organization of natural ends manifests itself in a threefold classification of organic features (Guyer, Reference Guyer and McDonough2020, pp. 207–208): the growth and self-repair of individual organisms as well as the preservation of the species through reproduction (CPJ, §64, 5:371–372). These functions are modes of self-organization insofar as the whole, represented as the full actuality of the kind, determines and organizes the parts.

Though to my knowledge Paul Taylor’s core biocentric thought never explicitly traces its root to Aristotle and Kant, the thought appears to be identical:

On the perfectionist view I’m articulating, immanent purposiveness deepens Taylor’s account of the living being as a “teleological center of life.” The Kantian account gives us a principled way to distinguish the self-organizing structure characteristic of the living both from things like lumps of mud or beaches (which are not purposive) and from artifacts like watches (which are purposive but only externally so).

At this point we arrive at an important fork in the road in the history of perfectionism. It is a common view that perfectionism is an eudaimonist and consequentialist doctrine. On such a reading, perfectionism is the thesis that the development of certain capabilities is of intrinsic rather than merely instrumental value.Footnote ¹⁵ Perfectionism thus sets a consequentialist standard for right actions: actions are right when they produce the maximal amount of perfection.Footnote ¹⁶

Yet, here I want to explore a neglected, non-consequentialist strand of perfectionism. Although Leibniz has been traditionally read as a consequentialist,Footnote ¹⁷ recent research has emphasized the non-consequentialist character of his thought.Footnote ¹⁸ These readings emphasize how Leibniz builds an essential connection between freedom or spontaneity and right: spontaneity is a sufficient condition for Leibniz’s interpretation of the first Ulpian maxim, neminem laedere. The basic principle is that one has an obligation to exercise one’s own freedom in a way that does not limit the freedom of others or, equivalently, in a way that does not undermine their capacity for self-determining action.

As Johns (Reference Johns2013) and Moggach (Reference Moggach2025) emphasize, Leibniz sets this first Ulpian maxim as a basic deontic constraint—rather than as a goal to be maximized. Though this fundamentally deontic aspect of Leibniz’s thought has only recently been foregrounded, its parallel to Kant’s account is striking. Similarly, Kant presents the original right to freedom as an interpretation of the same Ulpian maxim (MM, 6:236). What both thinkers appear to share, then, is the following thesis:

There is an essential or constitutive nexus between one’s self-organizing nature and a basic claim to respect. Why? Because our immanently purposive, self-organizing, and self-determining nature is an object of respect and makes us wrongable beings. Our self-determining activity can be hindered or destroyed by others, but since our self-determining activity has built-in normativity, in the basic case, to destroy or hinder such activity is not merely wrong, but a wrong to another. This necessary nexus between self-determination and a basic claim to respect, I suspect, is the shared understanding of the Ulpian maxim not to wrong anyone.

The perfectionist dimension of the thought is twofold. The basic claim to respect generated by the self-determining nature of living beings is not merely an instrument for the production of some independently intelligible good—perfection. Instead, what makes Leibniz and Kant’s version non-consequentialist is that the nexus between nature and deontic claim is constitutive rather than instrumental. At the same time, the basic claim to respect is not rule-fetishism: the point of the norm is to enable forms of action and interaction consistent with the perfection of living beings, i.e., with each being acting in a manner that realizes its self-determining nature consistently with the equal claim of other self-determining beings. The perfectionist claim to respect already introduces a form of community.

And this leads us to a perfectionist understanding of the nature of deontic claims. These are not static properties you carry around like an amulet. They are dynamic and dispositional properties—only irreducibly normative ones. The realization of rights is relational insofar as the perfection of the individual depends on the recognition by others of said claim. Relations of respect are not responses to impersonal norms or values. They are, instead, relations directed to the self-determining status of other living beings. And such relations enable the completion of the being in question. Once again, it’s easy to instrumentalize such relations. But the perfectionist tradition contains a non-instrumentalist potential: the thought that the perfection of a living being is inherently relational insofar as its completion depends on relations of respect and recognition in an interdependent community of right.

In his earliest work, Leibniz defines a moral quality as

The most basic concept of right is a power to action and is logically correlative with an obligation on duty bearer.Footnote ²⁰ So understood, basic rights are essential qualities of self-determining beings, but they are also dynamic and active, authorizing forms of action and reaching completion (as a power, facultas, or potentia) in an act that is recognized by others. From the addressee’s perspective the authorization to self-determining action appears as a moral necessity, which is “called obligation.”

We now have in place the required pieces, I think, to mount a Relational Perfectionist view of the basic claims of living beings. Although in their normative ethics, Leibniz and Kant restricted the status of self-determination to rational beings, we can find a natural extension of their perfectionist thought to the living in general.

Just as Kant grounds our dignity and moral status on our nature as free and self-determining,Footnote ²¹ extending a form of immanent purposiveness and self-determination to all living beings sets up a ground for their dignity and basic claim to respect.

3. Forms of Dignity: Plant, Animal, Human

With a sketch of Relational Perfectionism, we can now return to the Nature Objection. If Relational Perfectionism is true, P2 is false. Our obligations with regard to non-human nature are relational, and Relational Perfectionism explains why.

Let me elaborate the argument in Section 2 further. To start, distinguish with Joseph Raz “core” from “derivative” rights: a right is derivative when its justification and nature depend on another right; a right is core when its justification and nature do not depend on other rights (Raz, Reference Raz1986, p. 168). For instance, you might regard a right to freedom or autonomy as core (if it does not depend on another right), but a right to freedom of religion as derivative (if it derives from a more general right to freedom or autonomy).

Now, let us amend Raz’s formulation to get the structure of Relational Perfectionism in view. As a general metanormative theory, Relational Perfectionism takes the basic claim to respect as core insofar as this claim is not derivative from other normative properties. And let us understand the notion of “derivation” in such a way that x is derivative from y if, and only if, y suffices for justifying x but y does not require x for its nature or justification. As a normative theory, Relational Perfectionism avoids the consequentialism characteristic of other forms of perfectionism because its core norm is not maximizing some value. Rather, Relational Perfectionism (as in Leibniz or Kant) builds a deontic property into the very nature of self-determination or self-organization. The nexus between self-determination and a right to respect is constitutive rather than instrumental.

At the same time, we can derive more substantive rights from this basic claim to self-determination. Just as freedom of religion is derivative from and a specification of a general right to freedom, so too substantive moral rights will emerge for specific contexts of action and interaction as requirements and specifications of the basic claim to self-determination. For instance, if there’s a moral right to life, then that claim will be derivative from the core claim to self-determination.

It’s also crucial to see that the content of derived rights will also depend on the nature of the being in question. As Paul Taylor puts it,

Taylor’s point is intuitive. For example, grapes can be toxic to dogs but not to humans. To know the content of a being’s good, then, we need to articulate the conditions fixed by the specific kind of being in question. The content of the basic claim to self-determination—including the specific, substantive moral claims derived from it—will depend on the specific kind of being in question. The different capacities of various kinds of living beings will generate not only different kinds of rights, but also for the same type of right, rights with different content.

To elaborate this point, consider this sensible thought from Joseph Raz:

Raz’s point is that to respect human dignity involves respecting distinctive claims. Inter alia, it involves respecting the dignity-bearer’s autonomy, a “right to control their future.” Relational Perfectionism integrates Raz’s point in a distinctive way. Respecting rational animals will include respecting their autonomy, their right to control their future. And this is grounded in the specific type of capacities characterizing our kind, capacities to “plan and plot” our future. However, not all living beings possess such capacities. Let us assume only rational forms of life do. That means that respecting the self-determining status of a calf or a redwood will involve a different type of treatment from that required from beings lacking the capacities to rationally plan and plot their future. The mistake comes from inferring that because such forms of living beings lack such rational capacities (assuming they do) it follows that they are not self-determining at all—or are not wrongable beings. That is the core mistake behind P2.

Relational Perfectionism explains our obligations to non-rational, living beings in a straightforward way. If P1* and P2* are true, we get an account not just of our obligations with regard to living beings in general, but, more specifically, we get an account of such obligations as relational obligations. We owe obligations to the calf in virtue of its status as a self-determining animal. And we owe obligations to the redwood in virtue of its status a self-determining plant.

To be sure, the relations of respect we ought to enter with humans, calves, and redwoods are not the same. Relational Perfectionism does not flatten the moral landscape. But, at the same time, it’s also a mistake to suppose that because our duties to humans are different from our duties to non-human forms of life, it follows that the latter are impersonal or non-relational duties. In this way, Relational Perfectionism gives us a powerful framework for understanding not only why we have obligations with regard to living beings (their self-determining, immanently purposive nature) but also why such obligations are fundamentally relational.

Before closing, let me consider some important objections to this line of argument. I’ll respond to these in order to elaborate the account further—not as a demonstration of the truth of Relational Perfectionism. I’ll organize them into three groups: one line of objection targets the relationality of perfectionism; a second targets perfectionism itself and its reliance on immanent purposiveness; and a third deals with more substantive normative issues.

The first objection stems from more familiar “relational” approaches to ethics. It’s important to stress that such approaches have taken the perfectionist tradition as an opponent to relational ethics—not its ally. So my attempt to develop a relational variant of perfectionism is novel. Further, viewed from such an angle, it may seem that Relational Perfectionism is not “relational” enough. After all, does it not explain obligation in terms of some conceptually independent value, the value of perfection?

To elaborate the objection, let me illustrate it with a version due to R. Jay Wallace:

The chief challenge faced by such approach to moral obligation, argues Wallace, is that it fails to accommodate the other-directed nature of moral obligation: “at the level at which normative requirements are explained, the interests of other people enter as occasions for the realization of virtue, rather than direct sources of requirements on the virtuous agent” (Reference Wallace2019, p. 46). The worry is that, at bottom, a perfectionist account cannot be relational, since it forces us to conceive the interests of others as mere occasions for my own perfection. But is that true?

For the sake of argument, let us grant that Wallace is correct about familiar eudaimonistic (or even consequentialist) variants of perfectionism. As Wallace puts it clearly, such views seek to derive our deontic requirements (our moral obligation) from independently intelligible facts about natural goodness. On such a view, my obligations stem not from you, but from requirements I face to perfect myself.

Yet, it’s crucial to appreciate that even if Wallace’s objection goes through for standard eudaimonist versions of perfectionism, it leaves untouched the Relational variant of Perfectionism I’ve been elaborating here. Why? Because Relational Perfectionism treats the self-determining status of other living beings, to borrow Wallace’s terms, as “direct sources of requirements on the virtuous agent.” That was the point of recovering, from Leibniz and Kant, a non-consequentialist variant of perfectionism, one that understands the very nature of moral claims as constitutive for the perfection of individuals. If Relational Perfectionism is true, at the fundamental level, I have an obligation toward you, a fellow living being, not because in observing this obligation I become a good human being or avoid defects of my kind, but rather because my recognition is essential to your development as a self-determining being.

Still, a recalcitrant worry is that Relational Perfectionism is not “relational” enough. The objection is that as a variant of Perfectionism there is nothing fundamentally relational about the account. After all, what distinguishes it from more orthodox, non-relational accounts that tell us that we have impersonal duties to promote or honor value? More to the point, both Steve Darwall and R. Jay Wallace have emphasized the conceptual connection between moral obligation and Strawsonian accountability relations. Yet, it’s not clear that we can take the “participant” stance of reactive attitudes toward non-rational animals or plants. If such reciprocal relations of accountability are missing, it’s not clear what makes the account relational. Footnote ²²

Once again, we may grant, arguendo, that the objection hits its target when addressed to orthodox versions of perfectionism. Yet, it’s crucial to see that Relational Perfectionism introduces an alternative conception of relationality. Two properties make my proposed perfectionism relational. First, as in Leibniz and Kant, a fundamental normative property is the correlation between one being’s claim and the other’s obligation. I’ve suggested that every living being is entitled to such a claim, placing us (rational animals) in a moral relation to any living being. But second, I concede the Strawsonian point. It’s true that it’s not fitting or warranted to take the reactive attitudes toward a calf or redwood. Still, even if that is true, it does not follow that the account is not relational. After all, the Strawsonian model may be too narrow a model of relationality.

Other than the claim-obligation bipolar nexus, what makes the perfectionist account relational is the more substantive thought that the completion or perfection of certain beings depends on certain forms of interaction with other beings. When we mistreat a calf or redwood, it’s not that we are failing to produce some impersonal value in the world. Rather, we are directly hindering a form of self-determining activity of a living being. Duties of respect are directed precisely because their targets are self-determining beings. True, this primitive form of relationality does not embed us in a relation of accountability to the calf or redwood. But it does not follow that because I’m not accountable to the calf or redwood I’m not in a moral relation to it.

This brings us to a second set of worries concerning the form of perfectionism I’ve defended here—in particular, concerning immanent purposiveness. The main worry is that the notion of immanent purposiveness is an outdated mode of dogmatic metaphysics that should be banished. After all, the triumph of evolutionary theory should have banished teleological thinking long ago.

I will not be able to discuss in any detail the connection between evolutionary biology and immanent purposiveness, partly because it exceeds my area of expertise. But I will make an important distinction that may at least alleviate the concern.

When we think of teleology in nature, evolutionary theory appears to target three substantive theses:

At least on a superficial first pass, it seems as if evolutionary theory has lent support for rejecting all three of these theses. And if so, the notion of immanent purposiveness—and its accompanying perfectionist metaphysics—should be rejected.

In response, I’ll make two points.

First, the view of immanent purposiveness I’ve appealed to here requires commitment to none of those three theses. Precisely because the view takes immanent purposiveness to be prior to and more fundamental than extrinsic purposiveness, perfectionism can reject the first two theses. True, Aristotelian versions of immanent purposiveness also came along with the supposition that the essence of a natural kind is perpetual and fixed. But there’s nothing in the concept of immanent purposiveness that commits us to the unchanging nature of kinds. On the contrary, an evolutionary view may bring to the fore the essentially relational nature of kinds: kinds are not permanently fixed natures in a Platonic heaven, but nodes in a series of historical relations. Far from opposing immanent purposiveness, evolutionary theory might in fact support it.

Second, recall: I did not just postulate immanent purposiveness, but I introduced two presumptive warrants for it. The first, in Section 2, stemmed from the Euthyphronic issue of purposiveness in nature. If the model of Extrinsic Purposiveness is fundamental, it seems as if we are stuck with a problem of arbitrariness. We have no principled way to distinguish Worlds W1, W2, and W3. By contrast, immanent purposiveness gives us a principled account of our duties. The second presumptive warrant, in Section 3, showed that, contrary to the Environmental Wrongs argument, a relational account can explain our duties with regard to calves and redwoods when such duties are embedded in Relational Perfectionism. Moreover, we get a robust explanation of why it’s wrong to treat such beings in certain ways precisely in a relational manner: our duties to calves and redwoods are in fact relational duties owed to such beings qua self-determining beings.

To be sure, this paper has secured no more than a presumption for a relational account. My only point is that construing a perfectionist account on the idea of immanent purposiveness does not necessarily commit perfectionism to any of the three pre-evolutionary controversial theses I’ve delineated.

Finally, there are many substantive issues in normative theory that this paper lacks the space to tackle. Let me mention an important one: the omission of polluting rivers. Relational Perfectionism either accepts the implausible view that a river is a living being itself—on a par with calves and redwoods—or is incapable of explaining why polluting rivers is wrong.

But this is a false dilemma. We need not think of a river as a living being to see why we have a duty not to pollute the river. Instead, think of the river as a biotic community. The duty is not owed to any specific individual being, but to a collective of them, a collective constituted by the river as a community of living beings. A river may not be a living individual, but it is the living home of complex interdependencies among various types of living beings. When a factory pollutes a river, the factory violates a public duty owed to the community as a whole. The pollution destroys the conditions for the possibility of a stable, flourishing community with interdependencies between animal and plant life forms. And in so doing, the primary relation is not to a single individual but to the community as a whole.

Similarly, return to the Grand Canyon example quoted above. What explains the moral wrongness of filling the Grand Canyon with cement is that to do so would be to effectively destroy an entire biotic community. To be sure, there may be aesthetic reasons why you should not fill the Grand Canyon with cement. But what explains the wrongness of the act is not the aesthetic reasons; it’s the moral reasons stemming from collective claims of the living beings enmeshed in the Grand Canyon as a biotic community.