Pets, Power, and Legitimacy

1. Pets, Directive Power, and the Question of Legitimacy

2. Moral Independence and the Justification of Power

In this section, we develop a complaint that is rooted not in the effects of relations of power on subjects, but rather, on the way in which relations of power are justified. To foreshadow the central idea, we argue that morally acceptable justifications of asymmetric power must appeal to the ways in which power advances or protects the interests of the subject of power rather than the interests or goals of the ruler. Justifications that appeal to the interests of the would-be ruler fail, we will say, to respect the moral independence of the subject of power.⁸

To develop these claims, begin with the observation that every nonhuman animal – like every human being – has their own life to live, a life to which they bear a unique and special relation. Furthermore, many such animals are, like humans, the “subject of a life.” Consider Tom Regan’s famous elaboration of this idea:

[W]e are each of us the experiencing subject of a life, a conscious creature having an individual welfare that has importance to us whatever our usefulness to others. We want and prefer things, believe and feel things, recall and expect things. And all these dimensions of our life, including our pleasure and pain, our enjoyment and suffering, our satisfaction and frustration, our continued existence or our untimely death – all make a difference to the quality of our life as lived, as experienced, by us as individuals. As the same is true of those animals that concern us…they too must be viewed as the experiencing subjects of a life, with inherent value of their own. (Regan 1985, p. 22)

Observations of this kind – about our separateness or individuality – are commonly relied upon to motivate the claim that there are moral constraints on how we (humans) can permissibly treat one another. For instance, your body is yours and cannot normally be used to advance others’ ends without your consent. We cannot, for example, test our homemade soap on your skin without your permission. Similarly, your agency – your ability to respond to your environment and pursue the desires, projects, and relationships that constitute your life – is yours to exercise as you see fit. For instance, we cannot ordinarily force you to help us paint our house, go to the shops, or chop wood for our fire. While there are circumstances in which others can permissibly demand or force you to use your body or agency (e.g., to help save lives in an emergency), they cannot simply use you as an extension of their own body and agency to pursue their own interests and projects. To do so would neglect your moral independence, the fact that you are an individual with your own life to lead, your own well-being, and your own projects, preferences, and relationships.

You and others thus stand in an asymmetrical relationship regarding the permissible use of your body and agency (and vice versa). Furthermore, as these examples suggest, the normative asymmetry that reflects our moral independence has implications for the ways in which others can justify exercising power over you. The precise contours of this justificatory constraint are complex. For present purposes, two dimensions of this constraint are particularly important. First, to respect another’s moral independence, justifications of power over them must be subject-focused.⁹

Subject-Focused Constraint: To be legitimate, relations of asymmetric power must be justified by appeal to the interests of the subjects of power. The interests or projects of the ruler(s) that would be served by their possession of power cannot enter the justification.

To illustrate, consider that we might value having power over you because of the ways in which it would serve our own interests and projects. For example, it would enable us to complete our decorating projects, source fine wines with which to impress our dinner guests, or avoid the physical effort of chopping logs. However, the fact that we would benefit in these ways cannot contribute to a justification of our having power over you. This is not just the claim that your interest in having control over what you do outweighs our interest in having control over you. Even if serving us in these ways would in fact promote our well-being more than it would detract from yours, attempting to justify power in this way would disregard your moral independence, the fact that your body and agency are not generally available for others to use in the pursuit of their ends.

Notably, we recognise a similar constraint when interacting with humans who – for a variety of reasons – are not deemed competent to make some or all of their own decisions. For instance, young children, or those suffering from dementia, are sometimes unable to reliably decide how they should use their own body or agency. This means that we are permitted, within bounds, to substitute our own decision-making for theirs and to enforce our decisions upon them. Nevertheless, this does not license our use of children or individuals with cognitive impairments to serve our own purposes. The basic normative asymmetry between us, informing how we may permissibly relate to their body and agency, persists.

Thus, a good justification of the distribution and exercise of power must be subject-focused. To be sure, our power to make decisions and exercise control may also serve our own interests, even our interests in exercising power. For example, Harry Brighouse and Adam Swift argue that parents have a significant interest in having a relationship with children which partly consists in the possession of an extensive right to make and enforce decisions for their children (2014, Ch. 4). However, as Brighouse and Swift point out, this interest of parents, however weighty, cannot justify their extensive power over children. Rather, there must be adequate independent grounds, deriving from the interests of the children in standing in such a relationship, to justify the relation of power (ibid, p. 87; Viehoff 2017, p. 289).

This demonstrates that what underlies the form of moral independence with which we are concerned is not the ability of a fully-fledged rational or autonomous agent to, say, form, revise, and pursue a conception of the good. That capacity will certainly make a difference to what respect for our moral independence requires, but it is not necessary for the underlying constraint on justifications of power to kick in. Given this, we think the claim to moral independence, and the subject-focused constraint on justifications of power it brings in its wake, applies to all nonhuman animals who are the subjects of a life. Thus, a successful justification of power over them must appeal to the ways in which this relation of power advances their own interests.

This has important implications for our ability to justify the relations of power involved in the institution of pet keeping. Centrally, human interests and motivations for living with pets cannot contribute toward a satisfactory justification of those relations of power. For example, our desire for companionship, our intention to test our caring skills before parenting children, or our aim of educating our children about love and death, cannot contribute to a justification for deliberately instituting extensive relations of power over domesticated animals. To appeal to such interests involves wrongly assuming that these animals are available to be used to serve our own interests and projects. This is not to deny that we may have these (and other) interests in living with pets. Furthermore, it remains open that we could permissibly serve these interests by having pets. However, to avoid the complaint that these relations of power are disrespectful of pets’ moral independence, the justification for maintaining and reproducing these relations of power must be grounded in the interests of pets themselves.

At this point, many are likely to argue that it surely is necessary to exercise power over pets for subject-focused reasons, namely, to ensure that their interests are adequately protected. After all, since pets are dependent on humans to meet their needs, we surely have a duty to exercise power over them to ensure that those needs and interests are met. This brings us to the second dimension of the justificatory constraint that derives from moral independence, which we will call the Source Constraint:¹⁰

Source Constraint: Relations of asymmetric power are not legitimate even when they are necessary to solve a subject-focused problem if the would-be ruler is culpably responsible for the problem to which their exercise of power provides the solution.

To illustrate and support the Source Constraint, imagine that Anna pushes a Child into a pond and that Billy, Charlie, and Den are bystanders.¹¹ Anna knows that the only way of saving Child will be for Billy, Charlie, and Den to cooperate in a complex manner that only Anna can orchestrate (given her superior knowledge of the local area). Indeed, Anna pushes Child into the pond because she craves the feeling of power that will come from issuing orders to the bystanders as part of the rescue effort. Assuming this is the only way to save Child, and the costs will not be too high for Billy, Charlie, and Den, they will be duty-bound to follow Anna’s directives to ensure that Child is saved. However, Anna surely cannot claim that the power she wields is legitimate. After all, she would not have needed to exercise such power if she had refrained from pushing Child into the pond. While Billy, Charlie, and Den would be morally required to defer to Anna to save the child, they could simultaneously complain that the power Anna wielded over them violated their moral independence because she manipulated the situation so as to secure this power.

Returning to the case of pets, we agree that humans often are morally required to exercise power over pets who depend upon us to protect and promote their well-being. This can supply a local justification for the exercise of power, just as it can supply a local justification (in the form of a duty) for Anna to exercise power over the bystanders to save Child.¹² However, we cannot appeal to this necessity to argue that the relations of power central to the institution of pet keeping are legitimate. Such an argument assumes that we have pets and then points out that their interests depend on our exercising power over them. But this ignores the fact that we are culpably responsible for the very situation in which pets require that we exercise power over them if they are to live anything resembling a good life.¹³

One might object that there are important disanalogies between the case of Anna and the institution of pet keeping. First, Anna is directly responsible for creating a situation in which she must exercise power to adequately serve the interests of Child. By contrast, we did not initiate the process of domestication which has developed over many thousands of years. Second, while Anna is individually responsible for pushing Child into the pond, individual pet owners generally do not act to create the situation in which an animal requires that they exercise power to adequately satisfy their interests.

While these are important differences, they do not mitigate the normative point of the example. Consider another example. Individual men born in the UK or US in the early 19th century were not directly responsible for establishing the social institutions in which there were significant asymmetries in power between men and women. Nonetheless, these men faced a choice about whether to uphold those institutions. To be sure, different individuals played a more or less active role in reproducing those relations of power and will have had more or less opportunity to meaningfully challenge the status quo. Nevertheless, insofar as they (individually and collectively) failed to take steps to dismantle these institutions, they were responsible for recreating this unjust distribution of social and political power.

Similarly, we have inherited a set of social arrangements in which it is assumed that humans can have pets and that we can collectively rule over their lives. We did not initiate this pattern of relations, but – whatever we choose to do – we are responsible for whether and in what form it persists. Again, different individuals will be differently responsible depending on their participation in the institution and broader contextual factors. For instance, the roles of breeders, owners, legislators, and non-pet-owning citizens in the maintenance of the institution are quite different, and justice will demand different things from these people. However, since most adult citizens plausibly bear political responsibility for the continuation of unjust systems and institutions of society, we each share responsibility for upholding and recreating the relations of power central to the institution of pet keeping.¹⁴

A second objection to the Source Constraint is that it is overinclusive because it extends to birth parents’ relationships with their (pre-adult) children. Parenting children that you have bought into existence involves deliberately creating asymmetrical relations of power within which children are severely dependent. Given that most people do not regard the creation of such relationships as morally problematic or impermissible, it is a mistake to condemn all deliberately created relationships that involve significant asymmetries in power. Thus, it remains an open possibility that keeping pets involves the legitimate exercise of power.

A full exploration of the differences between children and pets would take us far afield so here we restrict ourselves to two responses. Note, to begin, that it is widely recognised that the power exercised by adults over children (at both the domestic and centralised political levels) can be illegitimate (e.g., Callan 1997, Brighouse and Swift 2014). Thus, one cannot simply assume that the power we exercise over children is permissible, and then leverage this assumption to argue that it must also be permissible in the case of pets.

Furthermore, many philosophers argue that the dependencies and vulnerabilities associated with the asymmetric power of parent-child relationships are morally problematic when they are avoidable. For example, Anca Gheaus argues that we “ought to minimise [children’s] dependency as far as it is compatible with ensuring that children’s needs are met” (2018, p. 61). To this end, Gheaus argues that all children should have access to long-term caring relationships with adults who are not their parents. This would avoid parents having a “monopoly of care” and thereby decrease children’s dependency, in part by limiting the power of parents.

However, there is a sense in which all the dependencies and vulnerabilities associated with childhood are unnecessary insofar as we are not morally obliged to have children. But of course, arguments like Gheaus’ assume that it is morally permissible to have children. Given this assumption, the reasoning goes, since it is necessary to exercise power over children to ensure that their interests are met, it is permissible to exercise this power. As such, the permissibility of having children sets the baseline against which the necessity (and so the moral character) of asymmetric power relations are evaluated in parent-child relationships.

The assumption that we are morally permitted to procreate is a substantive moral claim which we are not concerned to defend or deny. What matters for our purposes – and this is the first main response – is that we are surely not entitled to a similar assumption in the case of nonhuman animals. Whatever argument(s) vindicate the permissibility of human procreation, it is very far from obvious that such arguments extend to cover our reproductive control over nonhuman animals. Pets are not, for instance, the result of the exercise of human procreative rights or rights to bodily autonomy. Given this, we cannot assume the permissibility of having or breeding pets and rely on this as the baseline against which the “necessity” of our exercise of power is evaluated. The appropriate baseline of comparison is one in which we do not keep pets. Against this baseline, the apparent need to exercise power over pets dissipates.

A second response emerges from consideration of the fact that a very large part of children’s lives is organised around enabling them to become (increasingly) autonomous and independent. Indeed, we would regard alternative child-rearing practices or relationships – that aimed at maintaining relations of authority and dependence with children – as seriously objectionable (e.g., Callan 1997, pp. 152–55; Brighouse and Swift 2014 pp. 62, 164). This marks a fundamental difference with the institution of pet keeping, in which it is assumed that pets are and always will be subject to our will. Indeed, as Harry Brighouse and Adam Swift remark, “Adults are under a duty to raise children so that, eventually, they no longer need the adults’ fiduciary attention…This makes raising a child very different from having a pet” (2014, p. 62).

To recap, acceptable justifications of power must respect the moral independence of all individuals who are the subject of a life. Thus, any satisfactory justification of human power over pets must satisfy the Subject-Focused and Source Constraints. If no such justification exists, pets have a weighty moral complaint against these relations of power. Furthermore, many familiar extant or implicit justifications for having pets, and the power relations that pet keeping involves, fail to satisfy these constraints.

3. Power, Control, and Self-Determination

An adequate response to the complaint that our power over pets fails to respect their moral independence would require demonstrating that upholding systematic relations of power over a class of animals kept as pets is in the interests of the pets themselves. However, in this section and the next, we identify two ways in which our power over pets is in fact contrary to their interests. In the present section, we argue that our power over pets systematically thwarts their interests in individual control and self-determination.

As we demonstrated in Section 1, pets are subject to extensive directive power at both the domestic and centralised political levels. Humans determine what pets can and cannot do across all dimensions of their lives and enforce these decisions upon them. Generally speaking, the fact that one person or institution has this kind of directive power over another is held to be pro tanto objectionable. For example, citizens are assumed to have a complaint against the relations of rule imposed by the state, a complaint which must be answered if we are to show that the state is justified or legitimate (Kolodny 2023, p. 33). Thus far, however, it is an open question whether any such objection can be made on behalf of nonhuman animals in general, and those animals kept as pets in particular.

Whether this is so depends on the normative underpinnings of the objection to being subject to others’ directive power.¹⁵ A common suggestion is that the objection derives from or at least depends upon the capacity of autonomous agents to frame, revise, and pursue a conception of the good (e.g., Cochrane 2009, 2014). This capacity involves, in turn, the higher-order capacity to critically reflect upon one’s values, preferences, and plans. Because nonhuman animals are assumed to lack this higher-order capacity, it is argued that they therefore lack “an intrinsic interest in having ultimate control over their own lives” (Cochrane 2014, p. 165). In other words, nonhuman animals are thought to have no complaint against the mere fact of being made subject to other (human) wills because they lack the capacity to conceive of and pursue a reflective plan of life.

In our view, however, this characterisation of the objection depends on an unduly narrow account of the reasons that an individual (including an individual who possesses this higher-order capacity) may have to object to subjection to another’s will. One need not have the capacity to enact a reflectively endorsed plan of life to have non-instrumental interests in exercising control over oneself and one’s environment, and thus, in being free from the control and decisions of others.

To develop this claim, begin with the observation that many sentient nonhuman animals are agents. They experience and represent the world to themselves and act in light of these representations to satisfy their preferences, often in complex ways (e.g., Regan 2004; Korsgaard 2018). We should therefore recognise that many sentient nonhuman animals – including most pets – are wilful individuals: embodied and experiencing subjects with the capacity for self-directed action.

This already suggests that animals have significant interests in control and self-determination, for the familiar epistemic reason that individuals are best positioned to know of and satisfy their own preferences. However, we believe there is a deeper connection between agential capacities and interests in self-determination. Indeed, we submit that it is partly constitutive of the well-being of any subject of a life that they can exercise control over their actions, experience, and aspects of their environment. On this view, the good of self-determination does not depend on promoting an independent interest, or the fact that it enables the implementation of a reflective conception of the good. Rather, self-determination is a component of the well-being of all individuals with the capacity to act in pursuit of their own desires and goals.

This claim finds support in an account of autonomy’s value which locates an important dimension of that value directly in the ability to exercise individual control and agency. As Seana Shiffrin has argued, accounts of autonomy’s value that focus solely on the ability to implement a reflective plan of life neglect “the basic value of being in control of one’s experience and in not having experiences forced or imposed upon one when one’s will is to the contrary” (2004, p. 202). Shiffrin continues:

We value and respect autonomy in part because we recognize that one’s life is solely one’s own and that one must bear and endure singly one’s own conscious experience. Given this, the right and ability to select for oneself the contents of this experience is fitting and appropriate…Respecting other people’s decisions is not simply a way to promote their welfare or to facilitate the valuable process of their creating and expressing a distinctive character. In a more basic way, I believe it serves as an acknowledgement of the moral importance of the uniqueness and separateness of persons and the deep, irreducible fact that one’s life is the only life one has. (Shiffrin 2004, p. 203)

Here, Shiffrin rightly refocuses attention on the significance of being an embodied and experiencing subject with the capacity for self-willed action. This is not to claim that the ability to conceive and enact a conception of the good is of no normative significance. Rather, this account aims to recognise that there is independent normative significance attached to one’s position as an experiencing subject with a unique perspective on and ability to act in the world. The intimate connection between an agent’s mind, body, and experience renders it fitting and valuable that the agent herself exerts control over it (Shiffrin 2012, p. 382).

In our view, the ability to exercise control over one’s own bodily experiences, and environment – and thus in being free from the control and decisions of others – is a non-derivative component of well-being. This account gains support from its ability to illuminate and explain familiar judgments concerning the value of autonomy and the contours of autonomy rights. Consider, for example, cornerstone autonomy rights to bodily integrity that preclude others from interfering with touching or using one’s body without one’s consent. While such rights do play an important role in enabling us to implement a reflective conception of the good, they are also plausibly motivated by a much more immediate concern not to have other people impose decisions or preferences on us that are contrary to our will.

Another example concerns relatively trivial everyday decisions such as whether to eat a third chocolate bar or have another beer (Shiffrin 2012, p. 381). As Shiffrin argues, these decisions are unlikely to always be well supported by appeal to an individual’s privileged epistemic position regarding their own interests, or the value of their enacting a reflective conception of the good. Nevertheless, these are choices that we confidently assume all adults have a right to make without interference from others. However, we can straightforwardly explain these judgments if our account of autonomy’s value “gives central prominence to the ability to exert control over the contents of one’s experience as well as some sphere of one’s environment” (2012, pp. 381–382).¹⁶

These observations support the claim that we have important interests in exercising control over ourselves that do not derive from our privileged epistemic position or our ability to conceive of, revise, and pursue a conception of the good. Unsurprisingly, we think that many sentient nonhuman animals possess similar interests because they too are the subjects of a life.¹⁷ As such, they have their own perspective on and experience of the world, their own preferences, and the ability to engage in self-directed action to satisfy their preferences and avoid what they regard as harmful or unpleasant. This renders it fitting that they, and not others, have control over their experiences and surroundings. Their inability to “step back” and critically reflect does nothing to alter the fact that they are the subject of a unique relation between mind and world. Thus, when human decisions are imposed upon animals against their will, or we structure their lives so that they have limited opportunities to exert control over their environment and activities, we set back their interests in individual control and self-determination.

This is perhaps most obvious when reflecting on the most egregious treatment of domesticated and captive wild animals such as forced breeding, confinement, and scientific experimentation. However, the underlying observation extends to many of the more pedestrian and seemingly benign forms of control that are central to the institution of pet keeping. Consider, for example, the brightly coloured cages shaped like castles and spaceships, which are attractive to human children but offer little opportunity for control to their rodent inhabitants. Alternatively, consider the familiar sight of a human forcefully yanking a leash attached to a collar around a dog’s throat which violates the basic value of having control over their own body and actions. To the extent that restraint is necessary to protect a dog from danger (e.g., cars on a busy road), it is only because we have subjected them to life in such an environment that this is necessary at all.

These are just some examples of the ways in which pets are prevented from exercising control over their own bodies and agency. Yet as we have been at pains to point out, all dimensions of pets’ lives are controlled and imposed upon them by humans. We control whether they are confined alone, given space, or forced to spend time in close physical proximity with others. We have complete control over their sexual and reproductive lives, including whether they are neutered, have opportunities for sexual interactions, or are forced to reproduce. We determine whether they have opportunities for exploration, social contact, and exercise. And no matter the length of a relationship or residence, we can always decide to abandon a pet (to a shelter, if they are lucky) or have them put down.

In all these ways, our possession and exercise of directive power systematically set back the interests of pets in exercising control over their own bodies, experiences, and environment. None of this is a contingent feature of the practice, but a necessary assumption that underlies it. To live with pets, we must assume and exercise a right to exert total control over their lives. In our view, this underwrites a significant moral complaint against those relations of power.

4. Power, Risk, and Dependence

The third complaint pets have against the power relations imposed upon them concerns the relationship between power and the risk of harm. The simple fact is that being subordinate to others’ directive power makes one vulnerable to harm. That is because, if Anna has power over Billy, then Anna may direct or constrain Billy in ways that harm (or benefit) Billy. Moreover, there are several reasons why Anna might exercise power in ways that are harmful to Billy. One reason is epistemic: Anna may be unable or unwilling to identify Billy’s pertinent interests. A second reason arises from the possibility of disregard: Anna may fail to adequately consider how her exercising power over Billy will affect Billy’s well-being. Finally, Anna’s power over Billy enables A to use Billy in the pursuit of Anna’s own ends, and, in doing so, Anna may set back Billy’s interests. Risks of these kinds often motivate claims about how best to distribute or restrict power. For example, many argue for democratic modes of governance, at least in part, on epistemic grounds. Alternatively, the risk that powerful state actors will use power in ways that are detrimental to citizens’ interests supports the distribution of power across independent branches of government, which serve as “checks and balances” on one another.

Turning back to pet keeping, pets are subject to extensive risks of harm at the hands of those who have power over them. This is evidenced, in part, by the widespread harms that in fact eventuate (e.g., du Toit 2019). Moreover, these risks are exacerbated by the fact that pets are highly dependent on their guardians, and the broader community, to satisfy their needs. At the political level, pets are dependent on humans to institute laws and enforcement mechanisms that adequately track their interests through institutions and procedures from which they are constitutionally excluded. At the domestic level, the dependence of pets derives from multiple sources. In part, pets’ dependence is a product of the history of domestication. By selectively breeding the “wildness” out of animals so they exhibit desirable traits, such as playfulness and docility, we have created a class of animals who are often unable to meet their own needs. Yet, the dependency of pets is also a product of the ways in which social power is exercised over them from infancy. By controlling what these animals can do, where they can go, who they interact with, and generally requiring that they live in accordance with our ways of life, we stymie their capacity to be more independent (cf. Donaldson 2020 pp. 719–720).

Of course, dependency is an incontrovertible feature of human and animal life, and not all relationships involving asymmetric power and dependency are morally problematic. However, as Robert Goodin observes, dependencies are often created or exacerbated by social and political arrangements, and when this is so, they are liable to critique (Goodin 1985, p. 192). For example, the vulnerabilities associated with homelessness or unemployment are morally problematic because they are largely created by political decisions about how to distribute resources. Similarly, the vulnerabilities that traditionally arise for women in marriage are morally objectionable because they are the product of political and social arrangements (Okin 1989, Ch. 7). Thus, one cannot appeal to the immutable fact of dependency to support the claim that all forms of dependency and vulnerability are natural or morally benign. To the extent they are under our control, they are subject to moral scrutiny.

As Goodin observes, relationships of dependence and vulnerability often generate or reinforce asymmetric power (Goodin 1985, p. 193). And Goodin suggests that it is the risks that this power will be abused, and not the fact of dependence per se, that underwrites our moral objection to these relationships. Specifically, the objection derives from the fact that (i) those who are dependent are subject to risks of harm at the hand of those who have power over them, and (ii) some dependencies are an alterable product of social and political arrangements (Goodin 1985, pp. 193–194). Granting a general moral objection to exposure to risks of harm without good reason, there is an objection to social arrangements that expose one to risks of harm by creating or exacerbating one’s dependence on others, unless there is some good reason why these arrangements cannot be altered.

To be sure, some relationships pose a greater risk than others. Goodin suggests that the risks become acute, and especially morally dangerous, when a relationship displays four characteristics (pp. 195–196, emphasis in original):

1. The overall relationship embodies an asymmetrical balance of power.

2. The subordinate party needs the resources provided by the relationship to protect their vital interests.

3. For the subordinate party, the relationship is the only source of such resources.

4. The superordinate party in the relationship exercises discretionary control over those resources.

As we have seen, these conditions are met to a high degree in our relations with pets. There are significant and systematic asymmetries in power between humans and pets (1). Pets are also highly dependent on humans for the satisfaction of their needs and interests (2). Furthermore, human guardians are usually the only source of the resources that pets need (3) and have discretionary control over these resources (4). To give an obvious example, guardians normally determine whether, what, and when their pets eat and drink.

We take the foregoing to support the claim that pets have a further complaint against the relations of power that we impose on them. By subjecting pets to life under our total control, we subject pets to significant risks of harm, a fact that is only exacerbated by their dependence on us. These relations of power and the associated risks are not mere happenstance. Rather, they are deliberately created through the institution of pet keeping. Unless there are good reasons for subjecting animals to these risks, then pets have a moral complaint against being subject to these relations of power.

Many will no doubt suggest that there are good reasons to subject pets to the risks that accompany our having power over them. However, as we argued in Section 2, reasons that derive from human interests and projects – e.g., our desire for companionship, the joy of seeing the world from another perspective, etc. – cannot contribute to an acceptable justification. Good reasons to maintain an institution of pet keeping in which pets are subject to these risks would have to be subject-focused reasons.

5. Conclusion

All philosophers concerned with the plight of nonhuman animals recognise that significant reforms to our practices involving animals are morally necessary. The only real question is how we should reform existing practices. In this paper, we have argued that the power relations instantiated by the institution of pet keeping are illegitimate. The upshot is that we have a considerable moral reason to abolish this institution. Moreover, when this reason is added to other compelling reasons – such as the negative impact of pet keeping on climate change and wildlife (e.g., Palmer and Kasperbauer 2022) – there appears to be a strong case in favour of abolition.

Many might ask whether it is not possible to reform the institution of pet keeping so as to legitimise the relations of power that hold between humans and pets. While a full defence of the abolitionist position is beyond the scope of this paper, we think the argument that we have presented gives grounds for scepticism. Most reform proposals will not suffice because they focus on protecting the welfare of pets while leaving the basic relations of power between us and them unchanged. Furthermore, much of what is desirable about keeping pets seems to depend on extensive asymmetries in power. People want to choose their pets, to live with them, to be loved by them, to command obedience from them, and generally, to have these animals conform to their way of life. None of this is possible unless we possess and exercise the kinds of discretionary power over our “companions” that generates the complaints we have outlined. Thus, it is hard to imagine reforms that might legitimise the power exercised over pets without essentially bringing an end to the practice of pet keeping as people know and love it.

Of course, the prospect of abolishing the institution of pet keeping raises numerous theoretical and practical questions that we cannot address here. In conclusion, three brief points must suffice. First, recognising that the power we wield over pets and other domesticated animals is illegitimate does not commit one to “a fantasy of species apartheid” in which humans and nonhumans live entirely separately (Donaldson and Kymlicka 2017, p. 50). We do not object to all forms of interaction or relationship between humans and animals. Rather, we object to relationships in which humans are assigned extensive asymmetrical power over the lives of individual animals within a socio-political institution created and upheld by humans.

Second, several authors argue against ending the practice of pet keeping on the grounds that it will require unjust coercive interference in pets’ reproductive lives (e.g., Donaldson and Kymlicka 2011, p. 80; Nussbaum 2022, pp. 199–200). This objection raises complex issues that we intend to address in future work, and nothing we say in this article should be read as implying a view about the right way to transition away from the institution of pet keeping. However, the force of this objection is greatly limited by the fact there is no world in which we continue to live with companion animals but do not exercise significant control over their reproductive lives.

Finally, the idea that we might abolish the practice of pet keeping is often deemed infeasible (Donaldson and Kymlicka 2011, p. 79; Korsgaard 2018, p. 236). Whether or not this is true, it does nothing to show that the practice of living with pets is just or can be made to be so. Thus, if our argument in this paper is sound, abolishing the institution of pet keeping should be regarded as a live option in need of careful consideration. Questions of strategy can be considered down the line.

Acknowledgements

For helpful comments and feedback on predecessors of this article thanks to Jessica Begon, Alasdair Cochrane, Clemens Driessen, Diego Esposito, Carl Fox, Valéry Giroux, Jan Kandiyali, Faik Kurtulmus, Jonathan Parry, Jessica du Toit, Daniel Viehoff, Juri Viehoff, and Kristin Voigt, as well as to audiences at the Centre de Recherché en Ethique, University of Helsinki, MANCEPT, Queen’s University (Canada), University of Sheffield, University of Southampton, and Wageningen University. Richard Healey’s work on this article was supported by funding from UK Research and Innovation (UKRI) under the UK government’s Horizon Europe funding guarantee scheme [Grant Number: EP/X01598X/1]. UKRI Data Statement: This work contains no data.