DELIMITING JUSTICE: ANIMAL, VEGETABLE, ECOSYSTEM?

This paper attempts to bring some clarity to the debate among sentientists, biocentrists, and ecocentrists on the issue ofwho orwhat can count as a candidate recipient of justice. I begin by examining the concept of justice and argue that the character of duties and entitlements of justice sets constraints on the types of entities that can be recipients of justice. Specifically, I contend that in order to be a recipient of justice, one must be the bearer of enforceable moral claim rights. I then suggest that this has important implications for the dispute among sentientists, biocentrists, and ecocentrists. In brief, I show that sentientists cannot exclude nonsentient entities from the domain of justice merely by denying that they have “the right kind of interests,” and biocentrists and ecocentrists cannot move seamlessly from some feature of living things or ecosystems to entitlements of justice. I further argue that ultimately this disagreement on the bounds of justice bottoms out in a normative disagreement about which entities possess moral claim rights, and that the case for biotic or ecosystem rights has yet to be convincingly established.


Article abstract
This paper attempts to bring some clarity to the debate among sentientists, biocentrists, and ecocentrists on the issue of who or what can count as a candidate recipient of justice. I begin by examining the concept of justice and argue that the character of duties and entitlements of justice sets constraints on the types of entities that can be recipients of justice. Specifically, I contend that in order to be a recipient of justice, one must be the bearer of enforceable moral claim rights. I then suggest that this has important implications for the dispute among sentientists, biocentrists, and ecocentrists. In brief, I show that sentientists cannot exclude nonsentient entities from the domain of justice merely by denying that they have "the right kind of interests," and biocentrists and ecocentrists cannot move seamlessly from some feature of living things or ecosystems to entitlements of justice. I further argue that ultimately this disagreement on the bounds of justice bottoms out in a normative disagreement about which entities possess moral claim rights, and that the case for biotic or ecosystem rights has yet to be convincingly established.

ABSTRACT:
This paper attempts to bring some clarity to the debate among sentientists, biocentrists, and ecocentrists on the issue of who or what can count as a candidate recipient of justice. I begin by examining the concept of justice and argue that the character of duties and entitlements of justice sets constraints on the types of entities that can be recipients of justice. Specifically, I contend that in order to be a recipient of justice, one must be the bearer of enforceable moral claim rights. I then suggest that this has important implications for the dispute among sentientists, biocentrists, and ecocentrists. In brief, I show that sentientists cannot exclude nonsentient entities from the domain of justice merely by denying that they have "the right kind of interests," and biocentrists and ecocentrists cannot move seamlessly from some feature of living things or ecosystems to entitlements of justice. I further argue that ultimately this disagreement on the bounds of justice bottoms out in a normative disagreement about which entities possess moral claim rights, and that the case for biotic or ecosystem rights has yet to be convincingly established.

INTRODUCTION
While much theorizing in political philosophy restricts considerations of justice to human animals, several animal-rights theorists have persuasively argued that the exclusion of nonhuman animals from the domain of justice is unjustifiable. 1 Crucially, while the substantive details of these accounts can differ greatly, their advocates often explicitly delimit the scope of justice by appeal to sentience. 2 Consequently, for many defenders of animal rights, all and only sentient beings are to be counted as the proper recipients of justice. Let us call this view sentientism.
The appeal to sentience is compelling. Sentient animals have the capacity for feelings and emotions and so can experience things as better or worse. Put another way, most sentient animals are subjectively aware, which means that what happens to them matters to them. Accordingly, sentient animals have an interest in living lives that go well and derivative interests in the things that make life go well, such as access to adequate nutrition, clean water, suitable habitat, bodily integrity, and health. They also have interests in living lives limited in those things that can make life go badly, such as pain, suffering, and cruelty. It is possession of these interests which makes sentient animals eligible for considerations of justice because we can do them great harm or good in our relations with them. As Martha Nussbaum suggests, "it seems plausible to think that these relationships ought to be regulated by justice, instead of the war for survival and power that now, for the most part, obtains" (2006, p. 326).
However, attempts to delimit the scope of justice by appeal to sentience are contested by some environmental ethicists, who argue that the sentience threshold is theoretically indefensible. This challenge has both a negative component and a positive component. Negatively, critics of sentientism have argued that appeal to sentience is morally arbitrary and relies upon an anthropocentric bias (Fulfer, 2013;Plumwood, 1999). Specifically, it has been suggested that sentience is identified as the morally relevant criterion because it is a quality of humans, and thus the bounds of justice are extended only minimally to those few animals who share something in common with us. As a result, the focus on sentience "shifts the boundary to a new point but still leaves far too much outside, and has the same intense emphasis on the need for a boundary between what counts and what does not" (Plumwood, 1999, p. 199). Positively, these critics have advanced a range of alternative proposals that extend direct considerations of justice to all living things (let us call this view biocentrism) or to ecosystems (let us call this view ecocentrism). Specifically, biocentric and ecocentric theorists have argued that the proper subjects of justice are either all entities with interests (Baxter, 2000(Baxter, , 2005 or all entities capable of a dignified existence (Fulfer, 2013) or all entities that possess integrity (Schlosberg, 2007(Schlosberg, , 2012Crescenzo, 2013).
Unsurprisingly, defenders of the sentience threshold for justice reject biocentric and ecocentric attempts to extend the bounds of justice. For instance, they often deny that nonsentient entities have interests (e.g., Cochrane, 2012, p. 37-38), and then argue that only beings with interests can be harmed. If sentientists are right, and nonsentient entities cannot suffer harm, then it would be inappropriate to regard nonsentient entities as subjects of justice since what we do with regard to them makes no difference to them. On a slightly different tack, others have suggested that even if we grant nonsentient entities interests, only sentient beings are vulnerable in a way that makes direct considerations of justice appropriate (Donaldson and Kymlicka, 2011, p. 33). Of course, biocentrists and ecocentrists do not accept the sentientist critique, and we are left at an impasse, with defenders of each position asserting that their view is the correct one.
In order to overcome the deadlock, I propose that we look more closely at the concept of justice and what it means to be a subject of justice. With this as my focus, the central question of this paper is, which entities are candidates for entitlements of justice? 3 Sadly, I do not offer a comprehensive answer to this question; my aims are far more modest. Here I endeavour only to bring clarity to the debate among sentientists, biocentrists, and ecocentrists, by thinking through what is required to count as a candidate recipient of justice and, in so doing, to determine precisely what these theorists must show in order to establish their conclusions about the scope of justice. Though my aims are modest, the conclusions of this paper are nonetheless instructive for our current thinking about what we owe to nonhumans. By doing some conceptual housekeeping, I show that sentientists cannot exclude nonsentient entities merely by denying that they have "the right kind of interests," and biocentrists and ecocentrists cannot move seamlessly from some feature of living things or ecosystems to entitlements of justice. I further suggest that ultimately this disagreement about the bounds of justice bottoms out in a normative disagreement over which entities possess moral claim rights, and that the case for biotic and ecosystem rights has yet to be convincingly established.
The paper is divided into four sections. I begin, in section 1, by outlining those features of justice that distinguish it from other moral concepts. I contend that the character of duties and entitlements of justice sets constraints on the types of entities that can be recipients of justice. Specifically, in order to be a candidate recipient of justice, one must be the bearer of enforceable claim rights. In section 2, I discuss the benefits of adopting the discourse of justice, which explains why defenders of nonhuman life and ecosystems have been keen to extend considerations of justice beyond human beings. In section 3, I conjecture that sentientists have been able to build a persuasive case for the inclusion of sentient nonhuman animals into the domain of justice because, on most plausible accounts of moral rights, sentience is a necessary condition for being a rights holder. I then consider two strategies available to biocentrists and ecocentrists who wish to assign entitlements of justice to all living beings and ecosystems: (1) they might continue to maintain that nonsentient entities possess claim rights; and (2) they might reject the view that claim rights are necessary for justice. Ultimately, I conclude that neither strategy looks promising. Lastly, in section 4, I suggest that biocentrists and ecocentrists need not be too dismayed by this conclusion. Justice claims do not exhaust the moral domain, and limiting the scope of justice to sentient beings does not entail devaluing nonsentient life and ecosystems or rule out the possibility of legal protections for the environment.

THE CONCEPT OF JUSTICE
I want to begin by noting that all parties to the dispute are interested in what I will call "political justice." That is, defenders of justice for sentient animals, nonsentient life, and ecosystems are concerned with institutionally enforceable rightful entitlements. They might, for instance, be concerned with securing legally enforceable protections against harmful human action, fair access to important resources such as water and adequate nutrition, inclusion in climate change adaptation policy, or representation in political decision-making procedures. Political justice involves the direct institutional protection of recipients of justice, which means both that institutions are to be structured in ways that secure justice for all members and that members able to bear political responsibility can be coerced to fulfil their duties of justice to others. While there is significant disagreement about the substantive content of justice, the following four ideas about the concept of justice are widely shared. 4 First, the domain of justice is the domain of rightful entitlement, which means that justice concerns securing for each entitlement bearer that to which they are due. Entitlement bearers are owed duties of justice and they are treated unjustly when those duties are violated. 5 States of affairs might then be described as just when entitlement bearers have received that to which they are justly entitled and duty bearers have fulfilled their duties of justice. By contrast, states of affairs in which entitlement bearers have been denied their due and duty bearers have failed to fulfil their obligations can appropriately be characterized as unjust. This indicates that the idea of justice is thinly relational: justice and injustice are always done by someone or something to some other. 6 Thus, we evaluate the justness of people's actions, institutions, laws, conventions, or, more broadly, states of affairs, by considering how those who fall within their reach are treated or affected.
Second, though the precise content of what we are due as a matter of justice is disputed, entitlements of political justice are usually understood, even if not explicitly, to have the structure of Hohfeldian claim rights. For Hohfeld, a right-to be distinguished from a power, privilege, or immunity-is a claim that one has against some other that they act in some way. Moreover, the right that one possesses is always correlated with a duty possessed by some other to act in accordance with that right (Hohfeld, 1913). For example, I have a right against you that you do not assault me, which entails that you are under a duty not to assault me. More formally, X has a claim right to φ against Y if and only if Y is under a duty toward X to ensure φ. Accordingly, for every holder of a claim right, there is at least one correlative duty bearer.
Importantly, the duties that are correlative with claim rights are directed duties, which is to say, "duties that an agent owes to some party -a party who would be wronged if the duty were violated" (May, 2015, p. 523;emphasis added). Not all moral duties are directed. Some duties are owed to no party in particular, such as duties of beneficence, and others to no party at all, such as duties to protect great works of art. In these cases, X has a duty to φ, but X's duty is not directed toward a right holder. I may, for example, have a moral duty to volunteer at my local community centre, but that duty is not a duty owed to any particular individual. As stated above, unless someone has a claim right to my assistance, I am under no directed duty to provide it.
With this picture of rights and directed duties in the background, we can now return to the specific case of justice relations. If entitlements of justice are conceived of as claim rights, then there are correlative duties of justice. Moreover, the duties of justice that correlate with claim rights are directed-that is, they are owed to someone or some others. 7 This means that, when we violate our directed duties of justice, we do not merely act wrongly with regard to someone or some others, but we wrong those to whom the duty was directed. The satisfaction of our duties of justice thus depends on us acting in ways that respond directly to correlative claim rights-rights others have against us to perform or abstain from certain actions.
This structure of just entitlements and their correlative directed duties brings us to the third distinctive feature of justice, and that is that duties of justice are prima facie enforceable. The enforceability of duties and entitlements of justice is often stressed to draw out the difference between duties of justice and duties of charity and is especially prevalent in discussions about global justice. 8 While someone can be compelled to fulfil his or her duties of justice, the same is not true for duties of beneficence; the state can coerce us to treat others in ways to which they are entitled, but it cannot compel us to engage in voluntary acts of beneficence. Volunteering at the local community centre would no longer be an act of beneficence if I were compelled by the state to perform it.
Importantly, duties of justice are only prima facie enforceable because there may be other salient considerations that would ultimately tell against the use of coercion to achieve compliance: Enforcement imposes constraints on the freedom of agents and involves burdens for them, and such costs must be justifiable as feasible and reasonable. Hence, a duty could be a duty of justice even if a specific implementation of it is not justifiably enforceable, all things considered, in certain circumstances (either because in the circumstances such an enforcement is not feasible or because it imposes costs that are unreasonable given other, stronger, conflicting demands of justice). But since a pro tanto ground for action persists, agents may have to find alternative (feasible, reasonable) ways to honor it, or change the circumstances so that some form of honoring them becomes practicable. (Gilabert, 2016, p. 511 Importantly, not all moral duties and arguably not all directed duties are enforceable, which makes duties and entitlements of justice unique in this respect. For example, if I promise to come to your birthday party, then I incur a direct duty to you to come to your birthday party and you have a moral claim against me that I turn up as we agreed. However, we commonly think that it would be undesirable, and that it would be an illegitimate use of power, if the state were to enforce your right by coercively making sure that I attend your birthday party. That is, there are many moral duties that we ought to voluntarily fulfil, free from the coercive power of the state. What makes a claim right and its correlative duty fall within the protective domain of justice is an interesting question, but one that goes beyond the scope of this paper. 9 All that matters here is that the domain of justice covers moral entitlements and duties that agents of justice have a pro tanto reason to enforce. Finally, most agree that justice claims constitute a subset of moral claims, which means that not all moral considerations are considerations of justice (e.g., Rawls, 1999, p. 448). Justice is, for instance, distinct from loyalty, mercy, charity, and care; it is but one moral value among many. What it is to act justly differs from what it is to, say, act charitably or with care. Whereas justice involves giving to others what is rightfully theirs, charity involves giving to others what is rightfully yours. Likewise, even when claims of justice are satisfied, failure to promote the value of care may lead to the erosion and eventual collapse of the "social fabric of trust and concern" (Held, 2006, p. 71). None of this is meant to suggest that the objects of our moral evaluation only ever exhibit one particular virtue or value, or that the boundaries between moral values are always clear. However, it does mean that the domain of morality is not reducible to claims about justice.
This last point is crucial to the subsequent argument of this paper. Justice is distinct from other moral values and virtues. Injustice is a particular type of wrong and, as such, the nature of the wrong is different in kind from the nature of other wrongs. Though many of the moral flaws that people exhibit in their interpersonal relationships-such as disloyalty, dishonesty, selfishness, and indifference-may be judged to be wrong, they are not necessarily injustices. If I cheat at snakes and ladders or deceive a friend about my whereabouts because I'd sooner spend time with someone else, it is intuitive to describe my actions as dishonest, but inapt to class them as injustices. 10 Though there may be some sense in which my opponent has a moral right to win the game and my friend to be told the truth, in neither of these cases do we think that their right is enforceable.
Justice is a central moral concept, but we must be careful not to overstate its importance. The distinctiveness of justice in relation to other moral concepts is what makes it useful, normatively speaking. Any attempt to make the domain of morality coextensive with the domain of justice would essentially eliminate the distinctive wrongness of injustice-the violation of pro tanto enforceable rights-because all wrongs would then have to be counted as wrongs of injustice.
Moreover, by flattening the moral landscape in this way, we would crowd out all other moral considerations and the distinctiveness of the wrongs they capture. Thus, allowing the discourse of justice to monopolize the moral domain is implausible and limits what we can say about our moral lives in ways that are both practically and theoretically unattractive.

THE ALLURE OF JUSTICE TALK
Having elaborated the concept of justice, we are now in a position to see why some animal ethicists and environmental philosophers are keen to extend considerations of justice beyond the bounds of the human community. There are four key benefits to adopting the discourse of justice. First, duties of justice are often regarded as being more stringent than other moral duties. This means that duties of justice tend to have priority over competing moral demands and that failure to fulfil a duty of justice is worse than failure to fulfil some other moral duty. Accordingly, entities that have entitlements of justice occupy a privileged place in the normative landscape, and this position arguably secures them greater moral concern. Thus, including nonhuman animals, life forms, and ecosystems within the domain of justice would confer upon those entities the degree of moral importance that their advocates take them to be worthy of.
Second, talking in terms of injustice captures a distinctive type of wrong that might more adequately capture the nature of the wrongs in question. That is, injustices are wrongs done to entitlement bearers. This idea is most forcefully taken up by those looking to extend the scope of justice to nonhuman animals and is nicely articulated by Nussbaum: "When I say that the mistreatment of animals is unjust, I mean to say not only that it is wrong of us to treat them in that way, but also that they have a right, a moral entitlement, not to be treated in that way. It is unfair to them" (2006, p. 337).
A third advantage of the discourse of justice is that it better captures the systemic nature of some wrongs and moves us beyond interpersonal wrongdoings. Since justice is a virtue of individuals and institutions, it is necessary to evaluate the justness of institutional arrangements as well as the actions of individuals. This is particularly important for sentientists, biocentrists, and ecocentrists looking to capture the systemic nature of the wrongs done to nonhuman beings and entities and to determine institutional remedies for the perceived injustices.
Finally, since duties of justice are prima facie enforceable, institutional measures, including constitutional provisions, can be introduced to protect recipients of justice. This is to say that recipients of justice may have their entitlements safeguarded by law and are therefore not left vulnerable to the whims of personal morality. Again, this is especially important for sentientists, biocentrists, and ecocentrists who are seeking to extend institutional protection and respect beyond human beings. That people can be compelled to recognize the entitlements of nonhuman life and ecosystems, irrespective of whether they in fact do, is essential to protecting those entities in a world in which their value is largely ignored or denied.

JUSTICE, RIGHTS, AND SENTIENCE
Though the benefits of appealing to the discourse of justice are plain, we must determine the fitness of the concept for human relations with nonhuman entities. The discussion above alerts us to the central distinguishing feature of justice, which is that it is the domain of enforceable rights. This means that, in order to be eligible for entitlements of justice, an entity must possess features or standing sufficient to ground claim rights against others. Thus, the question of which entities are eligible for entitlements of justice can be answered only by attending to the prior question of which entities are eligible for rights.
Some of my readers may be frustrated by the laboured nature of the preceding discussion. Surely it is obvious, one might say, that, given the intimate connection between justice and rights, entitlement bearers must be rights holders, mustn't they? Though the point may seem trite, I nonetheless think that it is important to bring this feature of justice into focus. In the most recent disputes between animal and environmental ethicists over the bounds of justice, the idea that recipients of justice must be rights holders is often problematically absentor underexplored.
Environmental ethicists are, for instance, more prone to ignore the concept of rights altogether and move straight from some feature or quality that they take to be morally relevant to theorizing entitlements of justice for nonsentient entities. 11 In so doing, they make appeals to dignity, integrity, and vulnerability to ground entitlements of justice, but no story is given about how these things work to justify claim rights. But, as I hope to have shown, such a story is needed. Moreover, the preceding discussion calls into question the common sentientist strategy of demonstrating that nonsentient entities do not have the right kinds of interests for justice. Arguing that nonsentient entities lack interests of the relevant kind for the possession of rights only gets us to the claim that they do not possess rights grounded in interests. It does not yet show us that nonsentient entities are disqualified from counting as recipients of justice, because the possession of rights might be grounded in something other than interests.
With these considerations in mind, let us explore two avenues open to biocentrists and ecocentrists looking to make justice claims on behalf of nonsentient life and ecosystems. First, it might be the case that sentience is not necessary for an entity to count as a rights holder. Second, it is open to defenders of justice for nonsentient life and ecosystems to maintain that justice need not be concerned exclusively with rights. Specifically, it might be argued that, while possession of rights is sufficient to count as a recipient of justice, it is not necessary. I briefly discuss each of these strategies in turn.

Sentience and Claim Rights
In my discussion of the concept of justice, I talked at length about the structure of claim rights and their correlative directed duties. It is important to note that nothing about that part of the story can tell us who, or indeed what, can or does possess claim rights. In short, the concept of a claim right cannot, by itself, yield a normative conclusion about which entities possess such rights. Conceptually, any entity could stand in the place of X, where X has a claim right to φ against Y if and only if Y is under a duty toward X to ensure φ. There is nothing, then, about the concept of rights that would prevent a tree, an ecosystem, a bacterium, the universe, a chair, a toaster, or a car from possessing claim rights. So, in order to answer the question of which entities can possess rights, we must introduce further normative considerations.
All of this suggests that the dispute among sentientists, biocentrists, and ecocentrists over the scope of justice ultimately bottoms out in a normative dispute over which entities possess claim rights. There is, however, a general observation about rights discourse, which casts doubt on the project of theorizing moral rights (and, hence, entitlements of justice) for nonsentient entities. As noted in the introduction, defenders of sentient-animal rights have made a compelling case for extending moral rights beyond the human species-a case that rests firmly on the idea that justice is owed only to sentient animals. In light of the preceding discussion we are now in a position to say a little more about the connection between justice and sentience. Specifically, sentience is a precondition for entitlements of justice insofar as it is an ineliminable feature of all plausible accounts of the function and justification of rights.
There are two central views about the function of rights: the Will Theory and the Interest Theory. On the Will Theory, the function of rights is to protect the capacity of personal autonomy. Having a right to something means that you have control over others' free will in regard to that thing. Your right to private property, for instance, allows you to control whether people are allowed to enter your house. Accordingly, someone violates your right by acting contrary to your will with regard to the object of your right. Since the Will Theory is concerned with protecting personal autonomy and our capacity to make choices about the objects of our rights, it presupposes sentience. However, while sentience is a necessary condition, it is not sufficient, since one must be not only subjectively aware but also capable of making choices and waiving rights. Thus, the Will Theory excludes all beings (including many humans) who cannot make choices by asserting or pressing claims or waiving their rights. As such, the Will Theory will be of little use to those who want to extend moral rights to nonsentient entities. 12 The second view, the Interest Theory, has, by contrast, been considered as far more amenable to the inclusion of nonhuman entities. On the Interest Theory, the function of rights is to protect fundamental interests-interests sufficiently weighty to ground duties on the part of others. Many sentientists have keenly embraced the Interest Theory, which they take as open to extension: The prima facie case for viewing all sentient creatures as right-holders is extremely simple and draws upon two conventional ideas in moral and political philosophy. The first idea is that interests are the necessary and  (MacCormick 1977, Raz 1988, Kramer 1998. As such, on this view, all and only interest-holders possess rights. The second conventional idea is that sentience is the necessary and sufficient condition for the possession of interests (Feinberg 1974, Singer 1979, Sumner 1996. As such, on this view, all and only sentient creatures possess interests. When these two conventional views are combined then, the prima facie case is complete: all sentient creatures, as possessors of interests, are possessors of rights. (Cochrane, 2013, p. 657) Here again sentience features as a necessary condition of rights. Nevertheless, biocentrists and ecocentrists often deny the "second conventional idea" that only sentient beings possess interests. I do not have the space to review arguments that reject sentience as necessary for the possession of interests, but it is worth noting that claiming that nonhuman life forms and ecosystems have interests of the kind necessary to ground rights is very controversial. Our ordinary language may allow us to make sense of the idea that nonsentient life forms have biological interests or that integrity or good health is in the interest of flourishing ecosystems, but the fact that these interests are not the interests of welfare subjects (subjects who are able to experience what happens to them as good or bad) makes a difference to the normative status of those interests. That is, bare biological interests and ecosystem interests are not the same kind of interests as those possessed by sentient animals, and thus we need a story that explains how these other kinds of interests can ground rights. Importantly for our discussion, even if nonsentient entities fail to qualify as rights holders on the Interest Theory, biocentrists and ecocentrists could appeal instead to an alternative theory of rights. The Interest Theory and the Will Theory provide us with accounts of the function of rights-they explain what rights are for-but the question of what justifies rights can be viewed as separate. So, while rights may protect interests or autonomy, the question of what justifies rights might be appropriately answered by appeal to something other than interests or autonomy. Indeed, this is the position of those who defend status-based theories of rights (Kamm, 2007;Nagel, 2008).
Status-based theories of rights hold that rights are valuable because they reflect the inviolability of rights holders. Inviolability is a moral status that entails a description of what can and cannot be done to the individuals who possess that status (Kamm, 2007, p. 253;Nagel, 2008, p. 107). Importantly, the value and justification of rights does not depend fully on the interests that they protect. As Frances Kamm notes, "insofar as respecting a right not to be harmed involves respecting a status of inviolability, whether one has succeeded in respecting the right does not depend on whether the person is in fact violated or his interests set back" (2007, p. 253). The reason for this is that the value of rights is not limited to their instrumental role in protecting fundamental interests. Rather, the value of rights stems from the fact that they are derived from the noninstrumental value of the beings whose worth they express. Furthermore, what matters

V O L U M E 1 3 N U M É R O 1 H I V E R / W I N T E R
is not simply what is actually done to us, but what can permissibly be done to us. Thus, rights are valuable not just because they protect our interests but also because they express the noninstrumental value of beings who are liable to certain treatment (Nagel, 2008, p. 108).
For biocentrists and ecocentrists, the status-based theory offers an alternative picture to the Interest Theory that is also amenable to the inclusion of nonsentient entities. If it can be argued that all living beings or ecosystems have the moral status of inviolability, then that status would entail claims to certain treatment by humans, and thus, those entities would possess rights. 13 However, sentience is commonly viewed as a precondition for rights on the status-based theory. For example, Kamm argues that "a work of art or a tree may count in its own right in the sense that it gives us reason to constrain our behavior toward it (for example, not destroy it) just because that would preserve this entity... But this is still to be distinguished from constraining ourselves for the sake of the work of art or the tree" (2007, p. 229; emphasis added). For Kamm, only beings who can be helped for their own sake-sentient or conscious beings who can "get something out of continuing existence" (2007, p. 29)-have the kind of moral status that makes them eligible for claim rights. 14 If the notion of inviolability is necessarily tied to the notion of sentience, then it would seem that the status-based justification of rights does not offer much hope to biocentrists and ecocentrists. Moreover, we appear to have returned to the question of whether things can be good for nonsentient entities.
Additionally, it might be argued that the status-based theory of rights will be unattractive to at least some biocentrists and ecocentrists, those who would reject the appeal to inviolability. 15 To attribute the status of moral inviolability to all living things or ecosystems would have implications that at least some biocentrists will find disagreeable. It would mean attributing the exact same moral status to nonsentient entities as to humans, and it would mean holding that there are some ways of treating nonsentient entities that are never permissible. These conclusions will be too strong for some who might argue that we can always use nonsentient life so long as we do so with respect and within reason, and that the moral status of entities is gradated depending on the kind of entity that they are.
It looks, then, as though none of the main theories of rights is easily extended to nonsentient entities. So what? one might say. We humans have a long history of viewing ourselves as superior to other living beings and dominating nature, so it should come as no surprise that all traditional accounts of the function and justification of rights presuppose sentience as a necessary condition. Doing justice to nonsentient life and ecosystems will necessarily require a radical reconceptualization of how we think about moral rights and quite possibly altogether new ways of theorizing about rights. Perhaps this is right. Perhaps there is some hitherto undiscovered theory of rights that extends to nonsentient life and/or ecosystems while capturing our considered moral judgments about the rights of humans and sentient nonhuman animals. But, as yet, no such theory has been articulated. At present, we do not have a theory of rights for all living entities or ecosystems that enjoys even moderate support, and this is because such accounts have very counterintuitive implications-implications that few are willing to accept.
For example, attributing rights of equal weight to all living things and ecosystems would likely result in "ecofascism," where humans are sacrificed to protect the rights of other entities. Since human life depends on the killing of living things, and many of the current practices that we depend upon destroy nonhuman life and ecosystems, the just solution (to protect the rights of nonhumans) would be to either cull the human population or at least deprive many humans of the means of survival. Few would accept such extreme conclusions. An alternative approach rejects equal rights across living beings and ecosystems and instead proposes a graded model of rights whereby rights have greater or lesser moral significance according to some criteria (e.g., Nash, 1993). However, there is good reason to doubt the feasibility and efficacy of attributing rights to nonsentient beings (Benton, 1998). What do nonsentient rights protect? How do we determine the good of nonsentient entities? By what criteria should more or less weight be given to an entity's rights? Does the moral weight of an entity's right remain constant across contexts or does it change depending on the other types of beings involved? Even if these questions can be answered, attributing rights to nonsentient entities means that we infringe on those rights whenever we act contrary to them. Thus, if plants have a right to life, we directly wrong plants, even if excusably, every time we kill and eat them. Moreover, if plants have a justice claim against being eaten, then we have a pro tanto obligation to find alternative forms of nourishment.
In sum, sentience features as a necessary (though not always sufficient) condition in our central theories of rights. This, I think, helps to explain why defenders of animal rights have been able to make a successful case for extending moral rights to at least some nonhuman animals. It also explains why biocentrists and ecocentrists have not enjoyed the same kind of success. I cannot here offer a concrete refutation of the claim that nonsentient entities possess claim rights. However, the preceding remarks cast doubt on the project of extending predominant conceptions of rights to nonsentient entities. One consequence of this discussion is that, if biocentrists and ecocentrists want to attribute claims of justice to nonsentient entities, the burden is now on them to develop a theory of rights that can adequately accommodate sentient and nonsentient entities alike.

Justice: Recognition not Rights
A different strategy might be to accept that nonsentient life and ecosystems cannot be rights holders, but deny that rights are essential to the domain of justice. Something like this line of thought is present in David Schlosberg's appeal to the injustice of misrecognition and the harm of "status injury" (2007, p. 138-142). 16 Importantly, by taking up and adapting Nancy Fraser's account of recognitional justice, Schlosberg hopes to move us beyond the atomistic language of liberal rights and the liberal paradigm of distributive justice (2007, p. 140). Following Fraser, Schlosberg identifies three forms of status misrecognition: (a) a general pattern of cultural domination, (b) a pattern of nonrecognition, which is equivalent to being rendered invisible, and (c) disrespect, or being routinely maligned or disparaged in stereotypic public and cultural representations (Fraser, 1998, p. 7). These are structural, social, and symbolic indicators of misrecognition or lack of respect, and they are directly related to the status of the individual or community being maligned (Schlosberg, 2007, p. 140).
Nonhuman life and nature suffer "status-injurious misrecognition" because the relationship between humans and nonhuman nature is one of institutionalized subordination (Schlosberg, 2007, p. 140, p. 18). Moreover, Schlosberg argues that nature is "maligned and disrespected" in ways that clearly exhibit the three forms of status misrecognition posited by Fraser (2007, p. 140). Crucially for Schlosberg, on Fraser's view, status injuries can occur without being experienced by their victims; there may be no psychological distress or felt psychological need for recognition. This allows for an easy extension of the idea of recognitional injustice to nonhuman entities that may lack the psychological capacities that are commonly understood to give recognition its value. 17 On this view, then, nonhuman nature, including nonsentient life and ecosystems, suffers injustice when it is denied the social recognition that it is due. And it is denigrated by status injury when it is systematically subordinated to human interests or otherwise ignored and rendered invisible in human decision making.
There are two main things to be said about Schlosberg's argument. First, I think his account depends on an incomplete and problematic reading of Fraser. Though Schlosberg convincingly argues that status injuries do not necessitate complex psychological desires-thereby opening the door to nonsentient entities-he does not fully explain why status injuries constitute injustice for Fraser. Looking more closely at why status injuries constitute injustice complicates Schlosberg's appropriation of Fraser's theory of recognition. For Fraser, the wrong of misrecognition through social subordination is that it represents a barrier to "participating as a peer in social life," hence the politics of recognition is "aimed at overcoming subordination by establishing the misrecognized party as a full member of society, capable of participating on a par with other members" (Fraser, 2001, p. 24). So, on Fraser's view, misrecognition constitutes an injustice because it denies those who suffer from status injury the opportunity to participate and interact with others in our shared social life as a full partner (2001, p. 27). I take this to mean that only beings with the capacity to participate and interact with others as full partners in our shared social life can suffer the wrong of recognitional injustice. Indeed, Fraser suggests as much when she says that "justice requires social arrangements that allow all (adult) members of society to interact as peers" (Fraser, 2001, p. 29). Here "adult members" can plausibly be interpreted only as humans and thus Schlosberg's attempt to extend Fraser's theory to nonhuman entities seems to be in trouble.

V O L U M E 1 3 N U M É R O 1 H I V E R / W I N T E R
Now, of course, it is open to Schlosberg to argue that Fraser's view is needlessly narrow, but, given the character of Fraser's recognitional theory, the inclusion of nonsentient nature seems decidedly strange. What would it mean for nonsentient life forms or ecosystems to participate fully in social life? Whatever one's views about the agency of nonsentient entities, it seems implausible that such entities have either the capacity to participate or an interest in participating as equal partners. Moreover, it is difficult to see what justice would require to secure "participatory parity" for nature. On Fraser's account, justice requires two things: (a) that material resources be distributed in a way that ensures each member's inclusion and voice; and (b) that "institutionalized patterns of cultural value express equal respect for all participants and ensure equal opportunity for achieving social esteem" (2001, p. 29). Neither of these conditions can be aptly applied to nature. Lacking the ability to interact as a member of the political community, or as a peer, nature has no voice that might be enabled by giving it more resources or providing it with opportunities for social esteem.
The second problem is that Schlosberg obfuscates the central role of claim rights in Fraser's theory of recognitional justice. Notice that recognitional injustice occurs when individuals are denied that to which they entitled-namely, social relations of "reciprocal recognition and status equality" (Fraser, 2001, p. 24). Since members of a social community are victims of injustice when their moral right to participatory parity is denied, we need to know more about who can bear such a right and who in fact does. That is, before we can declare that nature has suffered a status injury, we need an account of how it can and does, in fact, have a claim against us to recognition and status equality. 18 Thus, the prior question of whether nonsentient entities can be the bearers of moral rights must be settled before we can determine whether or not such entities have suffered a recognitional injustice. 19 At this point, some defenders of justice for nonsentient entities may protest that the concept of justice I am operating with is inherently anthropocentric and thus not fit for purpose. But such a move does nothing to further the cause of protecting nonsentient entities and damages the concept of justice in ways that render it useless to humans and other sentient animals. If justice is no longer about pro tanto enforceable moral rights, then how are we to distinguish the concept of justice from other moral values? Moreover, as we saw, the reason that ecocentrists and biocentrists want to adopt the discourse of justice in the first place is because of the special role that the concept has in our moral and political thinking. When we strip the concept of its distinguishing features, justice is no longer what we believed it to be and thus no longer able to do the job that we wanted it to do.

BIOCENTRISM AND ECOCENTRISM: CROSSING THE RIVER TO GET WATER
In section 2, I outlined the main features of the discourse of justice that have made it attractive to biocentrists and ecocentrists. However, to conclude this paper, I think it is worth noting that some of the advantages of justice talk have been exaggerated and that some of the perceived benefits can be achieved by other means. First, while duties of justice are often taken to be more stringent than other moral duties, the key word here is often. Importantly, it is not always the case that duties of justice must be prioritized. Many things beyond sentient animals matter morally and can count in their own right. This does not mean that we can directly wrong nonsentient entities or that we have duties to them, but rather that they can give us moral reasons to choose one course of action over others. Their value, either instrumental or noninstrumental, can count in our considerations and, in some cases, may place constraints on what we can do in the name of justice.
Second, entities can be afforded legal protection without those entities being the bearers of moral rights. 20 Again, this does not mean that the legal protections are in place to protect against wrongs done to the entity in question or to compel us to fulfil our moral duties to that entity. There are, for instance, many instrumental reasons that can ground institutional protection of the environmentprotecting the environment is necessary for protecting human and nonhuman animal well-being, for instance-and it is therefore unnecessary to cook up justice claims on behalf of nature in order to justify enforceable institutional and legal measures.
All of this is to say that dropping justice talk for nonsentient life and ecosystems does not entail denying their value, ignoring them in our moral and political thinking, or leaving them vulnerable to abuse and neglect by humans. Furthermore, the project of "greening" theories of social justice does not require that we grant moral rights and entitlements of justice to nonsentient nature and ecosystems. Indeed, there are many existing accounts of environmental justice that effectively make visible the instrumental and noninstrumental value of the environment and offer credible frameworks for its protection (e.g., Holland, 2014). 21

CONCLUSION
In conclusion, this paper has sought to bring clarity to the recent dispute among animal-rights theorists and environmental ethicists over the proper scope of justice. I have suggested that, by thinking more carefully about the concept of justice, we are in better position to see what kinds of entities are eligible for considerations of justice. Furthermore, I have argued that entitlements of justice are claim rights with correlative directed duties that can be enforced, and that the structure of justice relations tells us something about the entities that can stand in such relations. Specifically, only entities that can be appropriately assigned enforceable claim rights are eligible for just entitlements. Thus, if biocentrists and ecocentrists want to bring nonsentient entities into the domain of justice, then they must establish that those entities are the bearers of claim rights. Cochrane, 2012;Donaldson and Kymlicka, 2011;Horta, 2013;Nussbaum, 2006Nussbaum, , 2011Pepper, 2016;Regan, 1983;Wyckoff, 2014. One standard way of excluding nonhuman animals is to deny that they have the relevant capacities (rationality or moral agency, say) to count as subjects of justice-to maintain they are not sufficiently like us. The most compelling response to this move is to note that all humans lack these capacities at some points in our lives, such as when we are infants, when we are ill, when we are severely injured or disabled, or when we suffer from the neurological diseases associated with old age. Few are prepared to accept that infants have no claim to having their needs met, or that Alzheimer's patients are not entitled to adequate care and assistance. Thus, it is now widely felt that rationality and moral agency set the threshold for entry into the community of justice too high. However, once we grant that all humans must be included under the protective sphere of justice, there is no good reason to deny entry to the many other sentient animals who are similar to us in their basic needs and vulnerabilities. 2 See, for example, Cochrane, 2012, p. 36-38;Donaldson and Kymlicka, 2011, p. 32-39;Horta, 2013, p. 36-37;Nussbaum, 2006, p. 358-362. 3 The basic question of which entities count as candidate recipients of justice should be kept separate from the question of which entities count as actual recipients of justice. The latter question can only be answered by adopting a particular conception of justice, whereas the former is a conceptual question about the scope of the application of the concept of justice (see n. 5 below). Here I am concerned with the first question. 4 Here I am interested in the concept of justice and not any particular conception. For more on the distinction between concepts and conceptions, see Rawls, 1999. Furthermore, much of what I say here is compatible with Pablo Gilabert's elaboration of the concept of duties of justice which he defines as follows:

225
Duties of justice are duties to preserve or promote people's access to important conditions or goods to which they are entitled and whose fulfillment is prima facie enforceable. This enforcement is all things considered justifiable if it is necessary for or strongly contributes to securing the required preservation or promotion and can be feasibly introduced without imposing unreasonable costs. (Gilabert 2016, 509) Specifically, of the four features that I discuss below, three track Gilabert's account: entitlement, rights, and enforceability. I do not have the space here to show why the concept of justice is not strictly concerned with negative, perfect, or institutional duties of justice, but to my mind Gilabert convincingly argues that those distinctions belong to specific conceptions of justice rather than to the concept. For more on this see Gilabert, 2016. 5 Of course, not all objects that are the source of injustice can appropriately be assigned duties of justice. Laws and conventions, for example, may be unjust, but lacking the capacity for political agency cannot bear duties of justice. This means that the set of duty bearers within a scheme of justice does not necessarily coincide with those things that comprise the site of justice. This being such, the question of who bears responsibility for justice arises. Answers to this question vary, but that dispute is not of concern here because we are strictly concerned with the question of who can count as entitlement bearers. 6 Which entities can be aptly described as just or unjust is a subject of much disagreement and is connected to the issue of what constitutes the "site of justice." The site of justice comprises all those things to which principles of justice can appropriately be said to apply, including, for example, personal conduct, social institutions, laws, and conventions. My argument here does not depend on adopting a particular view about the site of justice, but I take it as standard that institutions, including economic arrangements and systems of law, are liable to evaluations of justice. 7 I want to leave open the question of whether there are some duties of justice that do not correlate with the rights of some other or others. While beneficiaries of justice are owed that to which they are justly entitled, it is conceivable that not all of our duties of justice correspond to a correlative right holder. 8 This is unsurprising because theorists of global justice are also concerned with the question of who is eligible for considerations of justice, but that conversation is centred on political and geographical boundaries and not on species boundaries. For challenges to the anthropocentric bias in contemporary theorizing about global justice, see Horta, 2013;Pepper, 2016Pepper, , 2017and Steiner, 2011. 9 At the very least, claim rights that protect the most fundamental interests must count as entitlements of justice-e.g., interests in bodily integrity, health, and subsistence. 10 In some cases, acts of disloyalty, dishonesty, selfishness, and indifference may also manifest injustice. For example, when dishonesty is employed to dispossess you of what is rightfully yours, and where your right is enforceable against others, it would be appropriate to recast the wrong as a wrong of injustice. Nonetheless, while such cases exist, they do not show that all cases of dishonesty, selfishness, and indifference are instances of injustice or that it would be desirable to class them as such. 11 For instance, both David Schlosberg (2007) and Daniel Crescenzo (2013) move from the descriptive claim that ecosystems have integrity to the normative claim that ecosystems are entitled, as subjects of justice, to have their integrity respected and promoted. But the normative does not follow from the descriptive. The idea that ecosystems are candidate entitlement bearers in schemes of justice cannot be established by simply making the claim that they have integrity, because there is no apparent connection between integrity and just entitlements. What must be shown is that all entities with integrity can possess claim rights and that integrity can ground claim rights. Similarly, both Katy Fulfer (2013) and Anna Wienhues (2017) argue that an entity is a subject of justice if it has the capacity to flourish. However, in neither case are we provided with an account of how entities with the capacity for flourishing possess claim rights or how the capacity for flourishing can ground claim rights. 12 Incidentally, despite his commitment to biotic egalitarianism, Paul Taylor embraces a version of the Will Theory in his seminal work Respect for Nature (1986, p. 234). Adopting the Will Theory of rights for humans leads Taylor to conclude that, while there is no conceptual barrier to extending moral rights to animals and plants, we "should never … think of them as having rights in the same way we have rights, that is, in what might be called the primary sense of being a rights-holder" (1986, p. 254). According to Taylor, not only is ascribing rights to plants and animals "confusing" and "misleading" (1986, p. 254) but "the reference class of the concept, bearer of moral rights, should [not] be extended to include nonhuman living things" (1981, p. 218). So, on Taylor's account, since neither plants nor animals can be considered rights holders, they must, in line with my analysis above, stand outside of the scope of justice. 13 James Nash advances something like this view in his defence of biotic rights: "All life is sacred or intrinsically valuable and worthy of being treated as the subject of human justice. Indeed, the recognition of intrinsic value in nonhuman creatures implies the recognition of their legitimate claims [i.e., rights] for appropriate treatment from the human community and, therefore, for some level of rights and human responsibilities" (1993, p. 240).
14 Sentience is a necessary, but not sufficient condition for rights on Kamm's view, which leaves open the question of whether any sentient nonhuman animals will count as rights holders. 15 Here I am indebted to an anonymous referee of this journal for bringing this point to my attention. 16 There is a similar line of thought in Val Plumwood's work where questions of justice are taken to include questions of "ethical recognition" (1999, p. 189). For Plumwood, central to the injustice done to nonhuman nature is our failure to grant it the respect it deserves. In the case of plants, Plumwood argues that "we do them an injustice when we treat them as less than they are, destroy them without compunction, see them as nothing more than potential lumber, woodchips or fuel for our needs (a form of incorporation), fail to attend adequately to them, radically dissociate from them and deny their organization as intentional (and perhaps communicative) beings, thus adopting the stance of ethical closure or dismissal" (Plumwood, 1999, p. 201; emphasis added). 17 I.e., capacities bound up with self-worth, self-realization, and the desire to have others regard us appropriately. 18 It is also worth pointing out that relations of recognition with nature would necessarily be asymmetrical insofar as nonhuman entities cannot engage in "reciprocal recognition." 19 I leave open the possibility that we might act wrongly with regard to nonsentient entities when we fail to attend to their value, which seems quite plausible to me. On this view, some nonsentient entities, such as plants, ecosystems, and works of art, may be noninstrumentally valuable and thus give rise to reasons that place constraints on what we can and can't do with regard to them (see Kamm, 2007). Note, though, that this falls far short of attributing to them the status of rights holders, where we owe it to them to act in certain ways. 20 One might wonder what the preceding account makes of some legal persons, such as corporations, that are ascribed rights, responsibilities, and powers by law and yet lack sentience. Shouldn't we see these entities as possessing moral claim rights that underpin their legal rights? This interesting question raises a number of complicated issues to do with social ontology and collective rights. Precisely how we should think about these kinds of entities requires us to take a stand on whether collectives have moral status independent of the individuals who make them up. My sense is that if the legal rights, obligations, and privileges of nonsentient legal persons have a moral foundation, then they are derived from the interests of the sentient individuals who constitute them, but I cannot address this issue in any more detail here. 21 While I think that biocentrists and ecocentrists can achieve much without appeal to justice talk for nonsentient life things and ecosystems, I do not mean to suggest that there is no normative or practical import to counting as a subject of justice. As I stressed in section 2, counting as a subject of justice means that one has a unique moral and political status that others are obligated to recognize, and which they can be compelled to respect. Moreover, the unique status of subjects of justice gives us at least a prima facie reason to prioritize our duties to them over other moral reasons that we have for action, and, where we are unable to satisfy our duties to them, we are under an obligation to find alternative ways and means of honouring those duties. It is also the case that subjects of justice, since they can be directly wronged, might be owed compensation when they are denied their just entitlements. The same is not true of entities that are protected by law but that are not subjects of justice; compensation may be owed to someone, but not to entities that cannot be the victims of direct wronging.