1. Diagnosing the Problem: the ideology of humane use

Current public policy is premised on the assumption that humans have the right to use animals for our benefit, but that we should avoid ‘cruelty’ or ‘unnecessary suffering’ in this use. This proviso against unnecessary suffering is sometimes described as a commitment to ‘animal protection,’ but the basic logic of such laws inherently subordinates animal protection to human use. We protect animals only if and insofar as this protection is consistent with our use of them. For example, it is consistent with our interest in eating meat to require that cows be insensate when killed, but it is not consistent with our interest in eating wild fish to require that fish be insensate when killed. There is no way to kill most wild fish without inflicting intense pain, and to insist on a principle that fish be insensate when killed would require us to give up using fish as food. Since we assert the right to use fish in this way, we abandon the principle of ‘humane’ slaughter. The fish's interest in a painless death counts for nothing since it is incompatible with our using fish for food.

Ani Satz (Reference Satz2009) calls this the ‘interest-convergence’ principle: we recognize the interests of animals only if and insofar as it is consistent with our prior claim to use them in a particular way.Footnote ¹ It operates through what Jessica Eisen (Reference Eisen2010) refers to as ‘human-use typologies’: we categorize animals based on how we intend to use them, and then we define standards of protection that do not interfere with that use. This is quite explicit in animal law and policy. To take one example, Switzerland's Federal Animal Welfare Act (2005) says that animals have ‘intrinsic value’ and so must be treated with respect, and that this requires that anyone who handles animals must ensure their wellbeing insofar as their ‘designated use’ (Verwendungszweck) permits. We see versions of this in every Western country: we protect animals’ interests insofar as is this is compatible with their ‘designated use’. For this reason, it is misleading to talk about ‘animal protection’ policies: the fundamental purpose of these policies is not to protect animals, but on the contrary to assert the right to use animals. We have animal use laws, not animal protection laws. At its core, animal law authorizes the harming of animals.

Robert Nozick famously characterized the philosophical foundations of these policies as ‘utilitarianism for animals, Kantianism for persons’ (Nozick, Reference Nozick1974, p. 39). In relation to humans, we endorse ‘Kantianism’Footnote ² – that is, we recognize that humans have rights which set limits on the extent to which we can be sacrificed for the good of others. Killing one human to harvest organs to save five other humans would fail to treat us as ends in ourselves: it would instrumentalize us, when we should instead be seen as bearers of inviolable respect and dignity. All Western democracies – and international human rights law – accept this basic premise of ‘Kantianism for humans’, reflected in the prohibition on experimentation on humans.

By contrast, in the case of animals, we endorse utilitarianism – that is, we accept that their interests should be counted when engaging in utilitarian calculations about the greater good, but there is no limit on the extent to which those interests can be sacrificed for the good of others. Killing one baboon to save five humans is permissible, and perhaps even morally required. As McMahan puts it, animals are ‘freely violable in the service of the greater good’, whereas human persons are ‘fully inviolable’ (McMahan, Reference McMahan2002, p. 265). All Western democracies accept this basic premise of ‘utilitarianism for animals’, reflected in the humane use provisions of animal law.

This then is the current state of animal policy: an ideology of humane use, tied to human use typologies, grounded in inviolable Kantianism for humans, violable utilitarianism for animals. Virtually everyone who has studied animal law agrees that this framework is a moral failure, permitting and legitimizing the wholesale instrumentalization of animals, sacrificing even the most basic interests of animals to the most trivial interests of humans.

2. Philosophical critiques

As I noted earlier, moral philosophers have played a pivotal role in challenging the ideology of humane use and its philosophical underpinnings, and these philosophical critiques have inspired both individual change and social movements. These critiques have come from a wide range of positions, including feminist care theory, Wittgensteinian ethics, Aristotelian virtue ethics, pragmatist ethics, Christian and other religious ethics, and Indigenous ethics. In this section, however, I will focus on two lines of critique. If we want to challenge ‘utilitarianism for animals, Kantianism for humans’, then two options readily come to mind: we might defend utilitarianism for both humans and animals or we might defend Kantianism for both humans and animals. These are the positions defended by Singer and Regan respectively, and they continue to shape much of the debate. I will suggest, however, that neither is easily translatable into policy reform.

The Utilitarian critique: Singer has argued that it is speciesist to give more weight to the interests of humans than to the comparable interests of animals: all interests should be counted equally in determining the greater good. He thus objects to the status quo of ‘utilitarianism for animals, Kantianism for humans’. But as a utilitarian, the part of the status quo he objects to is not ‘utilitarianism for animals’. On the contrary, he fully supports utilitarianism for animals (so long as the interests of animals are in fact duly counted). What he objects to, philosophically, is Kantianism for humans. Whether someone has ‘rights’ is itself a question to be determined by assessing what promotes the greater good: there are no such things as inviolable natural rights that pre-empt the pursuit of the greater good. What he proposes, therefore, is not to extend inviolability to animals, but rather to insist that humans too should be seen as violable in the service of the greater good. He argues for a more even-handed calculus when deciding who is violable to serve the greater good.

For example, in a famous 2006 BBC documentary, Singer said that if infecting 100 monkeys with Parkinson's and then killing them was necessary to develop a therapy that could cure 40,000 humans, then it was ‘justified’.Footnote ³ Some commentators were surprised by Singer's endorsing of lethal animal experimentation, and suggested that it must reflect some change of mind in ‘the father of animal rights’, and some retreat from his stance against speciesism (Walsh, Reference Walsh2006). In fact, as he clarified afterwards in interviews, he has never denied that animal experimentation is justified if it promotes the greater good, and he explains that this is entirely consistent with his critique of speciesism, since he would apply the same test to experimenting on ‘human beings at a similar mental level’.Footnote ⁴ What is wrong is not experimenting on monkeys to promote the greater good, but the fact that humans have exempted themselves from the utilitarian calculus and accorded themselves inviolable protection from utilitarian violability. His recommendation, in short, is utilitarianism for both humans and animals.

This is a non-starter as a basis for policy reform. There is no public support whatsoever for the idea that we should start experimenting on humans. The vast majority of citizens believe deeply in ‘Kantianism for humans’. They believe – as do I – that respecting others as ends in themselves requires setting limits on the extent to which they can be sacrificed for the good of others. Indeed, the legal institutionalization of such rights is widely viewed as one of the great moral achievements of the twentieth century.

So there is no prospect of replacing ‘utilitarianism for animals, Kantianism for humans’ with ‘utilitarianism for both animals and humans’. And, predictably, in the debate triggered by Singer's 2006 interview, no one took seriously the idea that we might start experimenting on ‘human beings at a similar mental level’ in pursuit of new Parkinson's therapies. Kantianism for humans is simply too entrenched to be worth discussing. The entire debate focused on his justification for animal experimentation: that is, he was widely seen as recommending ‘utilitarianism for animals’. His endorsement of utilitarianism for humans dropped out of the policy debate.

And so the net effect of his intervention was to reinforce the prevailing ideology of humane use: he was seen as endorsing the idea that experimenting on animals to benefit humans is justifiable. To be sure, as Singer and others have demonstrated, Western societies badly misapply the principle of utilitarianism for animals, consistently undercounting animals’ interests, and thereby overestimating the cases where sacrificing animals would in fact advance the greater good. A more rigorous application of utilitarianism for animals would show that the benefits to humans of many of our research practices (or agricultural practices) do not in fact outweigh the harms to animals. However, a more consistent application of ‘utilitarianism for animals’ would still leave untouched the basic premise of Kantianism for humans, utilitarianism for animals. Singer himself may believe in ‘utilitarianism for humans, utilitarianism for animals’, but in a context where human rights are deeply entrenched, the former is immediately dismissed in policy contexts. And this leaves us right back where we started. In a context where Kantianism for humans is beyond question, endorsing utilitarianism for animals is not a challenge to the philosophical foundations of the status quo but a reinforcement of them.

The Deontological critique: If utilitarianism for both humans and animals is a non-starter, an obvious alternative is to defend Kantianism for both humans and animals. This is the crux of Regan's deontological rights-based position: animals too should be seen as having inviolable rights not to be sacrificed for the good of others. Animals have their own lives to lead, and we have no right to sacrifice them for our benefit. They are not a caste group that exists to serve us.

On this view, the central policy task is to ensure that animals are accorded the legal status of a rights-bearing subject. In our legal system, that status is ‘personhood’. Since Roman times, Western legal systems divide the world into ‘persons’ (those who are right-bearing subjects), and ‘property’ (that which is the good of another). And so the central policy goal of deontologists is to secure legal personhood for animals.

From a philosophical perspective, I find the arguments for animal personhood to be compelling. In co-authored work with Sue Donaldson, I have argued that personhood – or some functional equivalent like ‘selfhood’ – is essential to interspecies justice, and hence a crucial long-term goal.Footnote ⁵ However, as a policy proposal for the foreseeable future, it is unrealistic, at least with respect to the animals who are most intensely used in our society, such as farmed animals. It is difficult to know what percentage of the general public would support the idea of personhood for pigs and chickens, but if we take the number of ethical vegans as a quick proxy, the number is perhaps 3% in most Western countries. The idea is not supported by any major political party, and is not even discussed in mainstream political and public debate.

This helps explain why existing legal campaigns in defense of animal personhood have focused on a handful of wild animals who are peripheral to the ‘animal-industrial complex’,Footnote ⁶ such as primates, elephants, whales, and dolphins. This is true, for example, of the Great Apes Project and the Nonhuman Rights Project (NhRP).Footnote ⁷ This is a deliberate part of the legal strategy: pursuing personhood for primates, whales and elephants is politically feasible in large part because their use is not central to contemporary Western economies or ways of life. We can grant them basic rights with only minimal change in everyday practices.

Recall that the interest-convergence principle rests on some idea of the ‘designated use’ of animals – on human use typologies - and then accords weight to animal welfare if and insofar as it is consistent with this designated use. Pursuing personhood for apes, elephants and whales is politically feasible precisely because the average citizen has no clear idea of the ‘designated use’ of such animals. On the contrary, most citizens think that the appropriate life for such animals is to live freely in their own habitat. Of course there are some apes, elephants and whales in labs, zoos and aquariums – and these are the focus of legal campaigns – but these species are not primarily defined by their human use, and many citizens are quite prepared to accept that they have no designated human use, and so should be exempt from the interest-convergence logic.

The deeper challenge is to gain rights for those animals who are central to the animal-industrial complex, and who have been defined socially and legally by their designated use. These are the animals who are most trapped by the current interest-convergence logic, and for whom legal reform is urgent. In particular, we need to reform the status of domesticated animals: the animals brought into human societies to serve us, and whose designated use is a central part of our ways of life.Footnote ⁸ It is a striking fact that NhRP and other campaigns for personhood have been unable to imagine a strategy for achieving personhood for domesticated animals.

3. An Alternative Approach

So we seem to be caught in a bind. The status quo of ‘Kantianism for humans, utilitarianism for animals’ is a moral failure, but the two obvious alternatives – namely, Kantianism for both humans and animals, or Utilitarianism for both humans and animals – are non-starters for policy purposes. Is there a third option?

I am particularly interested in what we might call ‘transformative’ policy reform – reform that not only reduces or mitigates a particular harm, but also challenges the underlying premises used to justify those harms in the first place.Footnote ⁹ And in my view, that requires above all repudiating the idea of designated use. The reason why recent legal moves to recognize animal ‘sentience’, ‘intrinsic value’, ‘dignity’ and ‘respect’ are hollow is that they are subordinated to an animal's designated use. A minimum condition of transformative reform, therefore, is that we ‘Break Free from Human-Use Typologies’ (to quote the title of Eisen's article).

I am also particularly interested in transformative reforms in relation to domesticated animals (hereafter DAs), whose very existence has been defined by human use. Unlike wild animals, DAs are bred by humans to serve us. Many people find it difficult to imagine a life for DAs that is not defined by human use: serving us is who they are, what they do, and what they are for.

Given the extent to which DAs have been defined by human use typologies, we need a strong effort of imagination to break free from these typologies and to visualize alternative futures. If we want to identify options for transformative legal change, we need a clearer idea of our ultimate goal. What would a just interspecies society look like, and what kinds of relationships between humans and DAs would it involve? Once we have this ideal in view, we can then work backwards and identify openings for viable reform.

So what would an ideally just interspecies society look like? We know part of the answer: justice requires that we stop harming, exploiting and killing animals. We must recognize that they are individuals with their own lives to lead, not resources for us. But this is only part of the story: it tells us how we must not treat animals, but it does not yet tell us how we should relate to them. What kinds of relationships, activities, and interactions should we have with them?

It seems likely that we would and should have different types of relations with different groups of animals. Think about wolves and dogs. Since dogs are domesticated wolves, they are very similar in their basic biology, with similar forms of sentience. Yet we would surely have different relations with them. In respect to wolves, our obligation would largely be to leave them alone, to live freely on their own habitat. But having bred dogs to be part of our society and to be interdependent with us, we have obligations to care for them and to provide the conditions for their flourishing in community with us.

How can we make sense of this distinction between dogs and wolves? Domestication has made DAs members of a shared society with humans, and as such, they have membership rights, in addition to the basic personhood rights owed to all sentient animals. Recognition of social membership is an essential component of justice for DAs: having taken them out of the wild, and bred them to live and work alongside us, we must accept that they are now members of a shared society, a society that belongs to them as much as to us. Through domestication, they have acquired a birthright in this shared society.

Recognition of social membership would have far-reaching implications. The idea of social membership is central to our social and political life: contemporary liberal-democracies operate on the basis of an ‘ethics of membership’. T.H. Marshall famously argued that the rise of the post-war welfare state, and its associated ideas of social solidarity and democratic citizenship, was predicated on ‘a direct sense of community membership based on loyalty to a civilisation that is a common possession’ (Marshall, 1950, p. 96). The starting premise is that we form a ‘society’, and that the function of the democratic welfare state is to ensure that all members of society feel equally at home in the community, that everyone can equally partake in its social and cultural life and enjoy its civilization, that everyone can feel that they belong to the community and that the community belongs to them.Footnote ¹⁰ Marshall was centrally concerned with how the welfare state enables the working class to claim common possession of society, but if we accept that DAs are also members of society, the same logic applies. If DAs are members of society, then they too are entitled to shape the social norms that govern our shared life, to have their interests included when determining the public good, to benefit from public goods, public services and public spaces, and to a fair distribution of the benefits and burdens of social cooperation. Concretely, this would include a range of rights, including, for example, a right to a system of publicly-funded health care for DAs, and a right that emergency services be trained and equipped to rescue DAs in case of fires or floods, and a right to accessible public spaces.Footnote ¹¹

This suggests that justice for DAs operates at two levels. First, there is a set of basic rights owed to all sentient animals – wild or domesticated – such as the right not to be killed, experimented on or enslaved. Let's call these the universal rights of personhood, many of which are negative rights. Second, we have a set of rights that vary with the relationship that animals have to human society, which in the case of DAs involves membership rights to flourish within a shared society. (For wilderness animals, by contrast, these relational rights might include rights to habitat and to autonomy, to protect them from encroachment by human society).Footnote ¹² Let's call these ‘group-differentiated’ or ‘relational’ rights, since they vary with an animal's relationship to human society. These tend to be positive rights, to particular relationships or resources.

On this two-level view, justice requires both universal basic rights of personhood and relational rights, such as membership rights for DAs (or territorial rights for wilderness animals). To achieve justice for DAs, we need to promote both an ethic of personhood (DAs cannot be endlessly sacrificed for the good of others) and an ethic of membership (DAs have a birthright in society, which belongs to them as much as to us).

While the first level of personhood rights has been much discussed in animal rights theory, less attention has been paid to the second level of membership rights. This partly reflects the dominance of moral over political philosophy in animal rights theory: the idea of social membership is central to political theories of citizenship and social justice, but less central to moral philosophy (in either its utilitarian or deontological versions). However, the inattention to DAs’ membership rights also reflects a perception that talk about membership rights seems utopian. After all, as we've seen, personhood rights are largely negative rights not to be harmed, whereas membership rights are often positive rights. If it's unrealistic to expect public support for negative rights of personhood, how can we hope to get support for positive membership rights?

This two-level view of personhood rights and membership rights is indeed unabashedly idealistic, miles away from the status quo. However, I would argue that having this ideal in mind helps us to identify some real-world possibilities for transformative reform. While the path to personhood for DAs is blocked for the near future, there are forces in society pushing for recognition of their membership rights. Moreover, these efforts have had some success. While still fragile and small-scale, there are emerging spaces where DAs are indeed breaking free from human-use typologies and joining an ethic of membership. These developments have largely gone under the radar of animal rights theorists, but I will argue they are potentially transformative, and that philosophy has an important role to play in advancing them. In the next section, I will consider two such efforts.

4. Membership Rights: Domesticated Animals as Family Members and Workers

According to the ideal I've just articulated, one fundamental task is to ensure that DAs are recognized as members of society, and that our relations with them are governed by an ethics of membership. What would this mean, legally? Where might we look for glimpses of the legal recognition of social membership?

One place to start is with the very term ‘domesticated’. The word domesticated refers to the ‘domus’ – the household. We refer to DAs as domesticated because they have been brought into the domus to live and work alongside us in our homes and workplaces.Footnote ¹³ The home and the workplace are the sites where DAs and humans first formed a society together, and so it makes sense to start here. As I will discuss later on, justice requires that DAs be seen, not just as members of (private) families or workplaces, but also as having claims on society more generally, including public spaces and public institutions. Society is a broader category than family and work. Nonetheless, family and work are good places to start our investigation of membership rights.

Historically, DAs have been defined, for legal purposes, not as members of families and of workplaces, but as the property of families and workplaces. Consider family law. According to the law, pets living in a family home are the property of a family member, not themselves a member of the family. It is therefore entirely permissible for the owner to sell the dog, kill the dog, or in cases of family breakup, to demand custody of the dog as a property right. If the dog is owned by the husband, for example – if he is the one who purchased the dog – then he can claim custody of his property in case of divorce, even if it is clearly in the best interests of the dog to stay with the spouse or children or with other family pets to whom the dog has become attached. Family law gives rights to family members (including the right to maintain family ties wherever possible), but companion animals are not recognized as family members (and hence have no right to maintain family relations). Whereas an ethic of membership would say that the family belongs to both its human and animal members – that the family is a ‘common possession’ of its members – the law gives animals no membership rights in the family.

Similarly, according to the law, animals who work for an organization are the property of those organizations, not members of them. For example, dogs in the US Army have until recently been defined as property, and as such, have often been left behind when the military pulled out of foreign countries. Similarly, police dogs have often been killed when they are no longer able to perform their jobs. Labour law gives rights to workers in the workplace, but does not recognize animals as workers. Whereas an ethic of membership would say that the workplace belongs to both human and animal workers as a common possession, labour law gives animals no membership rights in the workplace.

This is a long-standing feature of animal law, and follows naturally from the ideology of humane use. In both the family and work contexts, we count the animals’ interests only when it is compatible with our use of them. However, what is striking for our purposes is that this legal status quo is increasingly out of touch with public attitudes, and we can see glimpses of a move towards a membership-based model. I will start with the family, and then turn to work.Footnote ¹⁴

Animals as Family Members: Surveys show that the vast majority of North Americans with companion animals consider them as ‘one of the family’, to be treated according to an ethic of membership, and they increasingly expect the legal system to respect and honour this membership relationship.Footnote ¹⁵ The result is a growing gulf between the self-understandings of average citizens and the law, and growing calls for legal reforms to acknowledge animals as family members.

Consider the issue of custody in the event of divorce. In the case of children, the established rule is that custody should be based on ‘the best interests of the child’, including a commitment to maintaining family ties wherever possible (e.g., courts try to avoid splitting up children). In the case of companion animals, however, since they are merely property, custody is based on whoever has the best ownership claim. Custody can be determined, for example, on the basis of who originally paid for the animal, without any attention to who has cared for the animal or whom the animal prefers, and without any commitment to preserving family ties (e.g., no effort is made to avoid splitting up companion animals and those they are attached to).

Many citizens are appalled that the legal system treats companion animals this way. And in response, many scholars have advocated legal reforms that recognize companion animals as ‘members of the family’ for the purposes of custody decisions (including ‘petimony’ in cases where the person best equipped to care for the companion animal is poorer), adapting legal rules developed for custody of human members of the family (Rook, Reference Rook2014).

A similar dynamic is at work in relation to liability for harms caused by animal companions, such as dog bites. In some jurisdictions, owners of companion animals are subject to strict liability for any harm their dogs cause. The result is to equate possessing a dog with possessing dangerous property, such as storing dynamite in one's backyard. Anyone who stores dynamite is subject to strict liability for any explosion, and it is no defense to say that one took reasonable precautions to avoid such an explosion. This is a sound legal principle if one wants to discourage people from having dangerous property. But the same principle is now being applied in many jurisdictions to animal companions: anyone who keeps a dog in their backyard is subject to strict liability for any dog bites, no matter how conscientious and careful they are. The (intended) result is to either discourage people from having animal companions or to keep them continually confined or chained, since they are categorized legally as a threat to others.

This equation of animal companions with dangerous property is out of step with social perceptions. For many people, companion animals are valued members of the family, and so should be subject to the same liability rules as, say, harms done by young children. In the case of young children, parents have a duty to exercise reasonable supervision and control, so as to reduce the potential risk that children might harm others, but children are seen as presumptively sociable members of the family and of society who have a right to be part of social life. We could reduce the risk that children will harm others if we kept children locked up all the time, but children are not dangerous property to be confined. They are members of society with a right to engage in social life, subject to parental duties of reasonable supervision. And so here too we see proposals to replace strict liability for dog bites (animals as dangerous property) with a model based on parental duties of supervision (animals as sociable family members) (Epstein, Reference Epstein2006).

In these contexts, and others (such as legal reforms allowing the creation of trusts for pets whose human guardians predecease them),Footnote ¹⁶ there is a push to redefine companion animals not as the property of the family, but as members of the family, with membership-based claims (to the preservation of family bonds in the case of divorce; to engage in social life in the case of liability; to be eligible for family benefits in the case of wills). All of these reforms involve rejecting the older use-based principles of animal law and replacing them with principles taken from family law for governing the relationships amongst family members. In my view, this trend will only grow in strength. The social reality is that companion animals are indeed members of the family, and whenever the law is so far out of touch with social realities, it needs to adapt.

Do these changes represent a genuinely new type of animal law reform that replaces the older logic of interest-convergence grounded in human use typologies with a new ethics of membership? Perhaps not. A skeptic might argue that these reforms are less about recognizing the membership-based rights of the animals, and more about advancing the interests of the humans in having secure use of the animal. We have a long history of using animals for companionship, and these legal reforms can be seen as operating to provide greater legal security to humans who use animals for companionship. In that sense, we can see them as strengthening the rights of the owners of companion animals, not as recognizing the rights of animals themselves. Humans assert a right to use animals for companionship, and these reforms simply specify how that right to use animals will be legally protected.

That is undoubtedly part of the story. However, it is not the only, or primary, dynamic at work. When people mobilize to secure these legal reforms, they are not saying ‘please respect my right to use animals for companionship,’ they are saying ‘please recognize that companion animals are members of my family.’ Just as they do not think of their companionable relationships with children or spouses as use-based relationships, so too they do not think of their relationships with dogs or cats as use-based relationships. They are, rather, relations of co-membership in a shared social unit – in this case, the family. People are not saying that ‘we’ (humans) have a right to use ‘them’ (companion animals), but rather are saying that companion animals are members of the ‘we,’ and are demanding legal recognition of this interspecies we-ness, this interspecies familial relationship. Recognition of family membership is precisely a membership relationship, and it carries with it membership claims, rather than claims to ‘humane use.’ Animals have always provided companionship but recognizing them as co-members of the family is new, and entails claims to shared membership of the home.

Animals as co-workers: Many DAs were originally brought into human society to labour, and humans and animals today continue to work alongside each other in a striking array of workplaces, from farms and labs to hospitals and seniors homes to military installations and airports (Coulter, Reference Coulter2016). As we've seen, legally speaking, these animals are not considered workers, but as the property of the workplace. However, as in the case of the family, this is increasingly contested. In some of these workplaces (though by no means all), the human workers have come to think of animals as their co-workers, with similar working hours and working conditions, undergoing similar training, facing similar risks, enjoying shared accomplishments. And once animals are seen as co-members of a shared workplace, there is a natural tendency to ask whether animals should have the membership rights of co-workers. These rights might include rights to safe working conditions, to maximum working hours, to union representation, to a retirement, to workers’ compensation if injured, to a living wage, and rights against forced labour.Footnote ¹⁷

This idea is nonsensical within the old use-based paradigm. However, an increasing number of human workers refuse to accept this framework for thinking about the status of their animal co-workers. One example concerns dogs in the US Army. As noted earlier, until recently these animals were seen as property – literally listed as ‘equipment’ – and as such, were often left behind (or killed) when the military pulled out of foreign countries. However, in response to pressure from their human co-workers, the Army now defines military dogs as ‘personnel’, and as such they have rights to repatriation and to rehoming upon retirement (Cruse, Reference Cruse2014). Similar developments have occurred in relation to police dogs – for example, the Nottinghamshire police force in Britain provides retired police dogs with a pension. And there are glimpses of a similar trend regarding emotional support animals. In all of these cases, it is the human workers who are mobilizing for recognition of DAs as colleagues and co-workers – as members of a shared interspecies workplace – rather than as property or equipment belonging to a human workplace.

Can we view these reforms as examples of transformative policy change that replaces the older logic of interest-convergence with a new ethics of membership? A more sceptical interpretation is certainly possible. Humans have a long history of seeing animals simply as ‘beasts of burden’ who exist to work for us, and these reforms could be seen as reaffirming our right to the ‘humane use’ of animals for their labour. But that does not seem to fully capture these developments. When soldiers in the US Army mobilized to secure these legal reforms, they did not say ‘please respect my right to use animals for work,’ they said ‘please recognize that animals are my co-workers.’ Just as they do not think of their collegial relationships with other human soldiers as use-based relationships, so too they do not think of their collegial relationships with military dogs as use-based relationships. They are, rather, relations of co-membership in a shared social unit – in this case, the workplace. People are not saying that ‘we’ (humans) have a right to use ‘them’ (animals), but rather that animals are members of the ‘we,’ and are demanding legal recognition of this interspecies we-ness, this interspecies working relationship. Recognition of collegiality is precisely a membership relationship, and it carries with it membership claims, rather than claims to ‘humane use’. Animals have always laboured for us, but recognizing them as co-workers is new, and entails claims to shared membership of the workplace.

So we have two examples of potentially transformative membership claims, as family members and co-workers. It is worth noting that philosophers have been largely invisible in these debates regarding family and worker status. These issues have not been central to mainstream animal ethics, and not have not been prioritized by the animal advocacy movements inspired by philosophers. Rather, these membership developments have largely occurred as a result of pressure from members of the general public who have demanded that the state recognize the ties they have developed with their companions and co-workers. Some philosophers are now playing catch-up with these developments, trying to make theoretical sense of them. But in general, philosophy has been ‘caught napping’ with respect to membership claims. Philosophers did not identify this as a potential sphere for transformative legal reform, and have not actively engaged with public debates on these reforms. I think this is regrettable, since these reforms raise a number of puzzles and dilemmas that cry out for philosophical analysis.

5. Advancing an Ethic of Membership

If the analysis so far is correct, we are at a crossroads with respect to membership rights for DAs. We can see glimpses of transformative legal reforms that embody an ethic of membership, but also a risk that these reforms will collapse back into ideologies of humane use. The fact that these reforms are ambiguous in this way – wavering between an ethics of membership and an ethics of humane use – is significant, and raises important issues that require philosophical attention.

One way to unpack the ambiguities of a membership approach is to compare it with another, perhaps more familiar, strategy for achieving animal liberation: namely, ‘let them be’. This is the slogan Regan and other theorists have used to summarize their vision of animal liberation: we liberate animals by leaving them alone (Regan, Reference Regan1983, p. 357). Palmer calls this the ‘laissez-faire intuition’, and discusses how central it has been to many theories of animal rights (Palmer, Reference Palmer2010). If the ideology of humane use has involved incorporating animals into our society, culture and economy to serve us, animal liberation involves de-incorporation: disengaging animals from human society, culture, and economy, and leaving them free to live on their own.

This ‘laissez-faire intuition’ is indeed an appropriate route to animal liberation for many wild animals. In many cases, we should leave them (and their habitat) alone.Footnote ¹⁸ However, this is not a viable strategy for most DAs. Having been taken out of the wild, and bred to live and work amongst humans, many DAs would not survive if we just ‘let them be’. In any event, as I argued earlier, DAs have a birthright in the society that they have helped to build. DAs may choose to leave human society, through some process of gradual rewilding, but we have no right to expel them from society.Footnote ¹⁹

The membership strategy is therefore a very different route to liberation from laissez-faire. Rather than severing DAs’ relationship to society, it refounds that relationship by giving DAs a fuller stake in society. On the membership model, society is a ‘common possession’ of its members, and so society belongs to DAs as much as to humans.

This idea raises distinctive risks as a strategy for animal liberation. To be a member of society brings with it various rights – including the family and workplace rights discussed in the previous section – but also various expectations and responsibilities. A shared society is only possible if everyone is able and willing to comply with some basic rules of sociability: sharing space, taking turns, avoiding violence and intimidation. Minimally, therefore, to be a member of society involves expectations of civility and sociability. Membership may also involve expectations of contributing to the common good, of doing one's share for the production and reproduction of the shared social world. And all of this requires some degree of socialization. All members of society – human or DA – need to learn to exercise self-restraint and to regulate their natural instincts so as to comply with social norms and social roles. For DAs to enjoy the family and workplace rights discussed in the previous section, they must first be socialized into the expectations of family and work life, to be able to live and work peaceably alongside others.

The membership approach, therefore, seems to involve some of the same problematic practices as humane use. Whether we adopt a membership model or a humane use model, in either case we seem to be involved in ‘molding’, ‘disciplining’, ‘normalizing’ or ‘policing’ DAs to secure their compliance with social norms and their contribution to public goods.Footnote ²⁰ So how exactly is membership liberating?

This is a serious challenge, and is one reason why many animal rights advocates have embraced the laissez-faire model of animal liberation. A laissez-faire approach denies that we have any right to socialize animals, or that we have any right to impose expectations of compliance or contribution upon them. In that sense, a laissez-faire strategy of animal liberation seems safer and cleaner: it avoids the dilemmas that arise when take seriously the idea of living together in a shared social world, with all the restraints and accommodations that are required to live together socially.

However, as I noted earlier, the laissez-faire strategy is simply not feasible or legitimate for most DAs. So in my view, we have no option but to confront these dilemmas head on. We need to think more deeply about what precisely defines an ethic of membership, and how we can prevent it degenerating into an ethic of humane use.

I suggested earlier that the crux of a membership ethic is the idea of society as a common possession, in which society belongs to all its members. The ideology of humane use rejects this: DAs are seen as belonging in society, but society does not belong to them: it belongs only to humans. Humans are the possessors of society, and as such have the right to determine its social norms and social roles. By contrast, the ethic of membership has a symmetrical view of belonging: DAs not only belong in society, but society belongs to them as a common possession of all members. And this in turn means that, on a membership model, DAs have a right to co-determine its social norms and roles. A full-throated ethic of membership rests on these two ideas: common possession of society and co-authorship of its social norms and roles. These are the ideas that underpinned Marshall's conception of the democratic welfare state, and in my view, they should equally define an ethic of membership for DAs.Footnote ²¹

This is a demanding vision, and as soon as we make it explicit, it becomes clear that existing reforms to family and work fall short of it in at least two key respects. First, they are highly selective in terms of which membership rights are accorded, falling short of rights of common possession and co-authorship of society. These reforms have invoked membership to address specific problems – e.g., custody disputes in the case of companion animals; retirement in the case of working animals – but have not embraced a fuller vision of social membership. Second, they are highly selective in terms of which DAs are recognized as members. In short, they extend a limited conception of membership to a limited number of DAs.

However, justice is not built overnight, and so a crucial question is whether these reforms are just one-off tweaks to the status quo, or are they likely to spur further membership claims. Will they have spillover effects, provoking membership claims for a wider set of DAs, over wider areas of social life?

And here, I think, there are reasons for cautious optimism. Let me start with family reforms. As I just said, these reforms are limited in two key respects as compared to a full ethics of membership. First, these reforms have focused on acknowledging animals’ membership within private families, whereas a full ethic of membership requires that DAs be recognized as members of the broader society and polity as well. To be fully recognized as a member of society would not just involve being treated as a member within private familial settings, but also being treated as a member of the public, with rights to public goods, public services, public spaces and public voice. So the first challenge is whether the recognition of private family membership can be a springboard to recognition of a wider public and political membership.

The second challenge is that legal reforms to recognize family membership have only applied to companion animals, and not farmed animals or lab animals. People rarely view sheep, cows, pigs, mice or rats as members of the family (although some do, of course), and so extending membership rights based on perceptions of familial bonds is less likely to help them. So the second challenge is whether we can extend this recognition beyond dogs and cats to farmed animals and lab animals.

These are both big challenges. There is growing evidence that companion animals are being treated in the law as family members, but we need to extend this membership from the private to the public, and from companion animals to all DAs. I am optimistic about the first challenge. We can already see evidence that the recognition of private family membership is having implications for broader social membership. Consider the legal changes adopted after Hurricane Katrina requiring emergency services to be trained and equipped to rescue companion animals. Or consider the growing movement demanding that cities provide dog parks, and access to public services and spaces. Or consider proposals for public health insurance for companion animals. All of these involve claims to public goods and services in the name of companion animals as family members.

This should not be surprising, since once we acknowledge that companion animals are members of the family, people will want to take their families with them into the broader society. They will want to vacation with their more-than-human family, or to bring their more-than-human family into commercial spaces or onto public transportation, or to ensure the health and safety of their more-than-human family. All of these involve claims on public services and public spaces. More-than-human private families will, sooner or later, demand a more-than-human conception of the public and of social membership.

And this in turn might generate claims to political representation. If companion animals are members of the public, and not just the family, their good should be included when determining the public good, and this requires that their interests be represented. There already are some examples around the world of appointing ‘advocates’ or ‘ombudspersons’ for animals within policy-making processes. In some cases, their mandate is simply to ensure effective implementation of existing laws regulating humane use, but more ambitious versions argue that DAs are owed political participation and representation because of their social membership, as members of the public in whose name the state governs.Footnote ²²

So there are grounds for optimism that the scope of membership norms will expand from private family membership to broader public/political membership, at least for companion animals. The second challenge, however, is more daunting: it is difficult to see how this model of family membership can be extended to most farmed or lab animals, who are not perceived as members of the family. This perception might change over time, due to the rise of sanctuaries for formerly farmed animals, especially urban and suburban micro-sanctuaries, which allow more and more people to experience first-hand the possibilities of living alongside such animals.Footnote ²³ Humans are certainly capable of having companionable relations with most DAs if given the chance, and bylaws prohibiting ‘farm’ animals in urban settings are increasingly being challenged in order to make room for such relationships. As more and more people come to know of friends and neighbours who have formerly farmed animals as family members, perhaps the very distinction between companion animals and farmed animals will gradually break down.

Indeed, some opponents of extending family membership to companion animals criticize it precisely on the grounds that ‘with many communities increasingly permitting farm animals as pets, drawing a line even broadly at companion animals can be exceedingly difficult’ (Goldberg, Reference Goldberg2013, p. 48). Far from being a ‘difficulty’ with the reform, I view this as one of its most potentially transformative features. The political significance of extending family membership to companion animals is much greater if it goes hand-in-hand with efforts to dissolve the line between companion animals and farmed animals – for example, by challenging bylaws that prohibit farmed animals in the city, and by establishing urban and suburban micro-sanctuaries for formerly farmed animals.

In short, while existing reforms to recognize family membership represent a highly limited vision of an ethics of membership, we have good reasons to think they can be a springboard to fuller social membership, certainly for companion animals, and also perhaps – more speculatively – for other DAs, if accompanied by efforts to dissolve the hard division between companion animals and farmed/lab animals.Footnote ²⁴

Let me turn now to the case of workplace reforms. As with family reforms, we see the same double challenge: existing reforms to recognize animals as workers offer only a narrow set of membership rights to a narrow set of DAs. To date at least, the key right gained by those animals who have been legally recognized as workers is the right to a decent retirement. The general public seems uncomfortable with the idea that animals who have worked loyally for us should be killed or abandoned when they are no longer able to work, particularly if this work involved risk or sacrifice, as with military and police dogs. Recognizing animals as workers has largely been motivated by the desire to secure a decent retirement for such working animals.

This is very far from the full membership idea that would entail common possession and co-authorship of workplaces and work life for DAs. However, I suspect that recognizing animals as workers will over time generate a wider set of membership claims. Commentators have called this the ‘labour recognition transformation thesis’.Footnote ²⁵ Both as a matter of formal law and as a matter of public ethos, the category of ‘worker’ carries moral weight. It entails not only rights regarding the workplace (such as maximum working hours and safe working conditions), but also rights in relation to the broader economy and society (such as a right to share in the benefits of society's overall scheme of cooperation). Drawing on international labour law, Blattner argues that recognizing animals as workers would entail both a prohibition on forced labour and effective representation in workplace management (Blattner, Reference Blattner, Coulter and Kymlicka2019), perhaps through some form of ‘microboard’.Footnote ²⁶ Once we recognize animals as workers, they should be seen as having the right to shape the design of jobs and the workplaces – in short, the right to co-author the future of work in our shared society.

This is speculative: there is a vast distance between current reforms aimed at preventing the killing of animals when they stop working and Blattner's vision of animals as co-authors of workplaces and economies. It is too early to tell whether the ‘labour recognition transformation thesis’ will bridge this gap. However, it's worth noting that many social justice struggles have grounded their claims for recognition and respect on work. Historically, when excluded or stigmatized groups have claimed social membership, it is often through work (Shklar, Reference Shklar1991). Participating in labour provides a basis on which subaltern groups can establish their claim to be participants in and contributors to schemes of social cooperation. Of course, people fight to escape or abolish degrading and stigmatizing forms of work, but there is ample evidence that participating in socially-recognized and socially-valued work is an important source of meaning and identity for humans, and an important basis for claims to social membership, and there is no reason to assume that the same is not true of DAs (Donaldson and Kymlicka, Reference Donaldson, Kymlicka, Blattner, Coulter and Kymlicka2019).

So we have grounds for optimism that recognizing animals as workers can trigger broader claims to social/political membership. However, that leaves us with the second problem: existing reforms only extend this social recognition to a small category of working animals. In principle, compared to the family membership route, the workplace membership path seems more inclusive, since it does not require intimate companionship or family ties. Social recognition comes through relations of collegiality, as co-workers in a shared workplace, not intimate family relations. As a result, this trend could in principle apply to a wider range of DAs engaged in a wider range of labour in diverse workplaces, outside the private home.

In practice, however, only a narrow range of work, done by a narrow range of animals, has so far achieved social recognition. It is above all military and police work that has generated social recognition, primarily for horses and dogs. As a result, the trend to recognize animals as workers, to date at least, has failed to challenge the cleavage between companion animals, whom we socially recognize, and farmed/lab animals, whom we ignore and exploit. Even though social recognition for military and police dogs is tied to work, not family bonds, it seems that the animals we are most inclined to recognize as ‘co-workers’ are the same types of animals that are viewed as ‘members of the family’.

However, I do not see any conceptual obstacle to a gradual extension of the social recognition of animals as co-workers to farmed animals. If sniffer dogs at the airport can be seen as co-workers, why not Bill and Lou, the oxen who worked alongside students plowing fields at Green Mountain College farm?Footnote ²⁷ Or the sheep who graze environmentally-sensitive grasslands at the San Francisco airport, or the goats on Faroe Island who take Google Earth cameras into otherwise inaccessible places, or the pigeons who carry pollution sensors to monitor air quality in the skies above London? And why not the animals who perform what Les Beldo calls ‘metabolic labour’, producing manure, or wool, or eggs (Beldo, Reference Beldo2017)? Indeed, there is growing interest amongst both academics and advocates in the transformative possibilities of labour as a route to recognition and membership for a wide range of DAs.Footnote ²⁸

To be sure, this is an uphill battle. For one thing, social recognition of others as co-workers is easier when we interact with them on an everyday basis, in a setting of trust, cooperation, and sociability, where we greet others at the start of the workday, socialize, and then embark on working together. It is more difficult when large numbers of animals are confined in pens or cages, with little or no human interaction or interspecies cooperation. The sociological conditions that make it possible to see others as co-workers in a shared workplace are simply not present in most modern farms or labs.

Moreover, even if we could make this cognitive leap to reimagine farmed animals as our co-workers, those with vested interests in the animal industrial complex will surely mobilize against any recognition of farmed animals as co-workers, since it would threaten the basis of their commercial exploitation of animals. The recognition of military/police dogs as co-workers has been uncontroversial because it is peripheral to the basic structure of the animal industrial complex, but as soon as ideas of workplace-based membership rights start to challenge inherited agricultural practices, and the lifestyles they make possible, the backlash would be quick. So it is an uphill battle to hope that ideas of animals as co-workers can significantly affect, in the near future, the abject status of farmed or lab animals (Eisen, Reference Eisen, Blattner, Coulter and Kymlicka2019).

In short, in relation to both family and work, we see a similar story. Existing legal reforms have introduced a narrow set of membership rights for a select handful of animals, but in both cases, there is evidence of a multiplier effect, as small initial reforms have triggered further debates and claims for stronger membership rights amongst a wider range of DAs.

Whether this trajectory will ultimately lead to a genuine ethics of membership is far from clear. And even if we could make real progress on the membership front, we would still face unfinished business regarding the other crucial track of interspecies justice – namely, personhood rights. As I've emphasized, justice ultimately requires both membership rights and personhood rights. Membership without personhood is a radically precarious and unjust status. To take just one example, under Ontario law, if companion animals who are lost are not claimed within three days, shelters are allowed and indeed may be required to sell them to research labs to be experimented on. When membership status fails, as in the case of lost dogs, they have no personhood rights to fall back upon, and so are subject to pure instrumentalization. Recognition of personhood makes clear that animals, like humans, have intrinsic moral status, prior to and independent of specific roles or relationships.

Our long-term aim, therefore, should be to achieve both universal rights of personhood and relational rights of membership. I have focused on membership in this paper because it has been relatively neglected within the literature, and because it may be a more politically feasible path for the foreseeable future, but we clearly need both. Membership cannot take the place of personhood (or vice versa): they are both irreducible elements of justice. What is less clear is how the two tracks interact. Someone might think that personhood must come first, that we can only recognize DAs as family members and workers if we first recognize them as persons. One might think, for example, that society will only recognize dogs as ‘personnel’ if it first recognizes them as ‘persons.’ But as we've seen this is not true. And indeed my hunch is that the reverse is often the case: we will come to recognize DAs as persons because we have first come to see them as family members and co-workers. Hannah Arendt famously argued that the first step towards the denial of universal human rights was to strip people of their social membership: once we lose the ability to see humans as social members, we lose the ability to see them as individual persons. So too in the reverse: enabling people to see DAs as social members may facilitate and expedite the recognition of personhood.