Article body


Zoos and aquariums everywhere are facing financial collapse. The lockdowns implemented in response to the COVID-19 pandemic have forced zoos to close their doors to the public, thereby ending their primary source of income.[1] Even in contexts where zoos are allowed to reopen, ongoing physical distancing requirements typically reduce visitor numbers.[2] Unlike tourist attractions that have been able to reduce costs by suspending their operations, zoos continue to have high maintenance costs because the animals in their care need to be fed, housed, and looked after. With ongoing maintenance costs and no income from ticket sales, zoos are left particularly vulnerable to bankruptcy during the COVID-19 crisis, and the precarity of their financial situation threatens the lives and well-being of the animals who live in them.

Most defenders of animal rights see zoos as institutions that unjustly incarcerate animals for human gain (Francione 2000, pp. 23-25; Malamud 2017). This might make one think that it would be wrong to give financial support to zoos, since doing so will only perpetuate injustice. Indeed, European affiliates of the animal protection organization People for the Ethical Treatment of Animals (PETA) have launched petitions to urge governments to close zoos rather than to subsidize them in their current form.[3] While we share the view that zoos ought ultimately to be abolished, we argue that the current situation is complicated by the fact that there are individual and collective duties to assist animals who are threatened by zoos’ financial situation. This means that doing nothing is not an option, because allowing animals to starve or to be killed in struggling zoos is not morally permissible. At the same time, we argue that the crisis has also created an unexpected opportunity to address the unjust treatment of animals held captive in zoos—an opportunity that it would be remiss to overlook. This paper calls for us to take advantage of this opportunity and offers a positive proposal for beginning the transition from current conditions to a world without zoos.

The paper is structured as follows. We begin by outlining three rights that sentient animals possess and which are crucial to thinking about the ethics of zoos (section 2). Zoos, we argue, are unjust institutions since they routinely violate these rights (section 3).[4] Having established that zoos are unjust, we argue that individuals and collectives have a duty to prevent the injustice done to zoo animals (section 4). The paper then sets out how we can best satisfy these duties and makes a series of positive proposals for how states and individuals ought to proceed (sections 5 and 6).


The argument advanced in this paper assumes that sentient animals have at least some basic rights. We can add very little to the general case for animal rights—which has been persuasively argued at length elsewhere[5]—but here we briefly identify three rights that inform the subsequent discussion.

First, and perhaps least controversially, is the right that sentient creatures have against humans not to be made to suffer unnecessarily. The grounds of this right are straightforward. Sentient animals are those with preferences, desires, and the capacity to experience the world as subjectively aware creatures. This means that they have a significant interest in things going well for them, which in turn gives rise to a range of positive rights, such as rights to access unpolluted water and a suitable habitat. Conversely, they have a significant interest in avoiding negative experiences, such as pain and suffering. The right not to be made to suffer protects sentient nonhuman animals against humans’ treating them in ways that cause them to experience harm (Cochrane 2012, pp. 54-57).

Second, we maintain that sentient animals have a right not to be killedunnecessarily by humans. This right protects sentient creatures’ fundamental interest in continued life. Life is valuable for all sentient creatures because “they have an interest in living to experience the goods that lie in prospect for them” (McMahan 2008, p. 67). Moreover, continued life is a necessary condition for everything else that an animal might have an interest in and thus it is the most important interest they have. Here we agree with Christine Korsgaard, who argues that “even if the rabbit’s life is not as important to her as yours is to you, nevertheless, for her it contains absolutely everything of value, all that can ever be good or bad for her…. The end of her life is the end of all value and goodness for her” (Korsgaard 2018, p. 65, emphasis in original). Once we acknowledge how fundamental continued life is for other sentient animals, the idea that they have a right not to be killed by us is compelling.

Third, we believe that sentient animals have a right to self-determination. Since they are subjectively aware agents, sentient animals have preferences and desires that they try to satisfy through self-determined action. Animals with these agential capacities have a significant interest in determining what happens in their lives, and in being able to exert control over the contents of their lived experience (Healey and Pepper 2020). The right to self-determination, then, protects competent animals’ interest in having their wilful agency respected in certain domains of activity. For instance, many wild animals competently negotiate the world without human assistance and so have a right to be self-determining—that is, to live free of human interference.[6] Importantly, the right to self-determination is not all or nothing: one does not need to have a right to be self-determining in all spheres of activity in order to have a right to be self-determining in some. This means that even when domesticated and captive animals cannot survive without human assistance, they may nonetheless be competent judges of whether they want to be touched or seen, what and when they want to eat, who they want to spend time with, what activities they want to engage in, and so on, and so they have a right to have their will respected in these domains (Donaldson and Kymlicka 2016; Healey and Pepper 2020).

While this is not an exhaustive list of animals’ rights, it is our view that whatever other entitlements of justice nonhuman animals may have, they at the very least have these three rights.[7] When these rights are violated, the animals are directly wronged and injustice occurs. Thus, a just interspecies society is one in which nonhuman animals are not made to needlessly suffer at the hands of humans, where animals are not unnecessarily killed by humans, and where their capacity for self-determination is not unjustifiably limited by humans.


Zoos (and aquariums—we use “zoos” to include both) are facilities that display animals in enclosures to the paying public. Though zoos remain popular attractions in most societies, we argue in this section that they routinely violate the rights of the animals in their care. Zoos, then, are unjust institutions.[8]

3.1. The Right Not to Suffer

Advances in knowledge about zoo animals and their welfare have allowed modern zoos to improve the survival and reproduction rates of the animals in their care. However, even for comparatively good zoos, it is difficult to provide the material conditions necessary for animals’ flourishing. Of course, animals in zoos are not a homogenous group, and individuals respond to their captivity in different ways. For some animals, such as small rodents and small reptiles, life in captivity may not have a significant impact on their welfare at all.[9] Importantly, however, these are not the animals that most visitors go to see. Studies suggest that visitors’ interests gravitate towards large mammals (see Carr 2016). And it is large, active mammals for whom captivity is most difficult. Bears, elephants, primates, big cats, whales, and dolphins all fare less well in captivity because confinement deprives them of the resources and opportunities that they need to flourish as the kinds of creatures that they are (Marino 2018, pp. 101-107). Confined to small enclosures, unable to roam freely, unable to avoid the gaze of humans and their fellow inmates, unable to gather food or stake claims to territory, and lacking opportunities for companionship and diverse forms of social interaction, these animals struggle to adjust to the artificial and restricted conditions of their confinement. Their lives are monotonous and limited in ways that have serious impacts on their physical and psychological health. For instance, as critics have long maintained, confining animals in zoos produces anxiety, sadness, neurotic behaviour, and other negative experiences. One source of psychological suffering are visitors, whose presence and behaviour are very stressful for some animals (Fernandez et al. 2009; Hosey 2008). This stress sometimes results in increased intraspecies and interspecies aggression and can cause a range of stereotypies—constant and repetitive behaviours that have no obvious goal or function (Mason 1991)—such as pacing, rocking, swimming in circles, mouthing at cage bars, and excessive grooming, to name but a few. These abnormal behaviours are symptomatic of the psychological suffering caused by confining animals in zoos.

Incidentally, zoo-related news reports during the COVID-19 pandemic frequently make the claim that animals in zoos “miss” their visitors. The general thrust of this claim is that animals in zoos are suffering because they miss the stimulation that interaction with human zoo-goers provides. This is evidenced, or so it is claimed, by the fact that animals who might otherwise be entirely indifferent to visitors suddenly show increased interest in staff or journalists close to their enclosures.[10]

We think that this narrative of “lonely animals” must be resisted. Such reports create a romanticized idea about zoos and the impact of visitors. We do not deny that some animals may have found the recent changes to their daily routine distressing.[11] However, it does not follow that the presence of visitors is a welcome aspect of these animals’ lives, or that once visitors return, these animals can resume living good lives. The animals’ response may simply be an indicator that their lives are so lacking in stimulation and enrichment that even noisy and disruptive visitors are an improvement. This should be seen as an indictment of zoos, not a reason for zoo visitors to think that they are improving animals’ welfare by visiting and supporting zoos.

3.2. The Right Not to Be Killed

When animals have an interest in continued life, this means that, other things being equal, more life is good for them. While studies show that zoos mostly enhance the longevity of mammals, there are some animals, such as female Asian and African elephants, who live longer in the wild than they do in zoos (Tidière et al. 2016, p. 4). Hence, one obvious way in which zoos harm some animals is by reducing their lifespans. Of course, this is not the same as actively killing animals, but it nonetheless shows how zoos can set back the fundamental interest that individual animals have in living.

More troubling, however, is the intentional killing of zoo animals by those who are supposed to care for them. One recent news story that garnered much public attention involved Neumünster Zoo, whose director explained that the zoo was drawing up a “kill list” of animals that could be killed to feed others if the lack of income made it difficult to pay for the zoo’s ongoing costs during the COVID-19 crisis.[12] This statement may suggest that killing for non-euthanasia purposes is a rare occurrence in zoos, limited to crises such as the one we are currently facing. However, the zoo director herself emphasized that euthanasia is not uncommon in zoos and is sometimes legally required. While the term “euthanasia” is meant to apply to cases where animals are killed because they have a serious, terminal illness and/or are enduring unbearable suffering with little prospect of alleviation or recovery, in the zoo context (and other contexts involving nonhuman animals), the term is frequently applied to the painless killing of healthy animals for reasons other than their own well-being.[13] A common practice in zoos is so-called “population management euthanasia” (e.g., Powell et al. 2018), in which “surplus” animals are killed—because, for example, they can’t contribute to the zoo’s breeding programme. Estimates suggest that European zoos within the European Association of Zoos and Aquaria (EAZA) kill between 3,000 and 5,000 animals each year (Browning 2018).

The case of Marius—a young, healthy giraffe at Copenhagen Zoo—brought attention to this issue in 2014. Marius was shot to death[14] because his genes were not useful to the zoo’s breeding programme. His body was dissected in front of an audience of visitors and then fed to the zoo’s carnivores. Copenhagen Zoo’s treatment of Marius is consistent with EAZA’s “Culling Statement,”[15] and EAZA publicly supported the decision, noting that “the young animal in question could not contribute to the future of its species further, and given the restraints of space and resources to hold an unlimited number of animals within our network and programme, should therefore be humanely euthanised.”[16] Copenhagen Zoo also killed two young leopards in 2012 for similar reasons; it is reported to kill between twenty and thirty healthy “exotic” animals per year.[17]

In addition to the killing of “surplus” animals, captive animals are often killed if they pose a threat to human life. For instance, Margaash, a snow leopard who escaped his enclosure at Dudley Zoo in the UK, was shot dead after he couldn’t be coaxed back.[18] Other animals have been sacrificed in defence of humans, not because they have escaped their enclosures, but because humans have found their way inside. In recent years, perhaps the most infamous case of this involved Harambe, a 17-year-old western lowland gorilla. A three-year-old boy climbed into the gorilla enclosure at the Cincinnati Zoo and Botanical Garden and was grabbed and dragged by Harambe. Fearing for the boy’s life, a zoo worker shot and killed Harambe.[19] To be clear, we are not saying that the zoos were wrong to prioritize the lives of the humans in these cases (though much depends on the circumstances). However, the apparent conflict between the rights of the animals, on the one hand, and the rights of the humans, on the other, is entirely human in creation. If zoos did not exist, neither would these threats to human safety, and animals would not have to pay for these conflicts with their lives.

It’s important to understand that killing in zoos is not a rare occurrence. Nor are decisions about killing individual animals made in light of those animals’ interests or the interests of other animals in the zoo. Instead, such decisions are made in light of the zoos’ goals, particularly its commercial interests. Frequently, animals who require expensive medical treatments, who do not play well with others, or who are surplus to the zoo’s breeding program are killed.

3.3. The Right to Self-Determination

Zoos must be categorized as what Erving Goffman famously called “total institutions” (1961). According to Goffman, a total institution is a closed social system in which life is organized by strict norms, rules, and schedules. Daily life within that system is governed by a single authority who determines the rules and issues orders to the residents and staff (1961, p. 6). Moreover, the “encompassing or total character [of total institutions] is symbolized by the barrier to social intercourse with the outside and to departure that is often built right into the physical plant, such as locked doors, high walls, barbed wire, cliffs, water, forests, or moors” (Goffman 1961, p. 4). This description of a total institution clearly reflects the organization and structure of zoos. Animals are confined in a closed system that they are unable to leave, and all aspects of their lives are heavily regulated and monitored by those in charge: social interaction, movement, diet, reproduction, activity, sleep, observation, and so on. Moreover, the zoo is separated from the rest of society by physical barriers designed to block escape attempts and keep non-fee-paying members of the public out.

Recognizing that zoos are total institutions makes it clear how they limit the degree of control animals can have over the shape of their own lives, thus violating their rights to be self-determining. Animals captive in zoos cannot choose to leave, and few have the freedom to leave their enclosures. Animals captive in zoos cannot choose where to live and with whom to live. Animals captive in zoos cannot choose what to eat and when to eat it. Animals captive in zoos cannot choose their relationships since the communities that animals live in are determined by zoo management. Moreover, animals may be moved in or out of specific communities when this suits zoo breeding programmes or the commercial interests of zoos, which affects not only individual animals’ interests but also undermining communal life. Animals captive in zoos often cannot choose to be unobserved since zoos have an interest in ensuring that visitors can actually see the animals they pay to visit.[20] Zoos interfere with animals’ reproduction—for example, by preventing them from reproducing or by subjecting them to artificial insemination. For all of these reasons, and countless others, the right of animals to be self-determining is incompatible with keeping them captive in zoos.

Since zoos clearly violate the rights of many animals in their care, they are unjust institutions. However, some try to defend the existence and practices of zoos by appealing to the contribution that zoos make to education and species conservation. Might these kinds of benefits justify infringing on animals’ rights in the ways we outline in this section? Note, first, that claims about the educational and conservational benefits of zoos should not be taken at face value: studies assessing zoos’ actual effects typically reveal more modest benefits than is often claimed (Born Free Foundation 2007; Jamieson 2002, p. 187; Moss and Esson 2013). More importantly, educational and conservational benefits, even if large, cannot justify violating the rights of individual animals. While rights are generally not considered absolute, they protect rights-holders by blocking simplistic utility calculations: it does not become permissible to violate rights simply because this would maximize the good. However, the pursuit of other goods can justify infringements on individual rights if, first, the good in question is of similar importance as the right to be infringed upon, and, second, if the benefit achieved is sufficiently large. Education and species conservation are not the kinds of goods that could, even in principle, be weighed against basic rights such as the right to continued existence. Thus, whatever benefits zoos might bring about with regard to education and species conservation, those goods cannot be used to justify the rights violations associated with keeping animals in zoos.


The financial problems that zoos are facing create additional risks for captive animals—as is starkly demonstrated by some zoos’ proposals to kill animals in their care to reduce costs. Animals’ rights not to be harmed and killed clearly impose duties on those who own and manage zoos not to take this approach. However, the pandemic has made it difficult for the individuals and organizations primarily responsible for the care of zoo animals to avoid such desperate measures. Zoos have called on both governments and individuals to donate funds so that these measures can be avoided.

Such pleas made by zoos, especially when tied to accounts of “kill lists,” have the flavour of moral blackmail: “If you don’t help us, then we will be forced to kill our animals.” These kinds of appeals trade on our desire to help cute animals in need, but they obfuscate the fact that those who make the appeals are primarily responsible for the vulnerability of the animals in their care. Zoos deliberately choose to display animals to a paying public and are responsible for capturing them in the wild, acquiring them from other zoos, or breeding them in captivity. That they now ask for donations to “prevent” them from killing animals in their care and allow them to continue their rights-violating practices is deeply problematic.[21]

At the same time, of course, this does not release us from our duties to the animals who are affected by zoos’ response to the current crisis. If we trust that zoos have no other choice but to kill their animals, who else may bear a duty to prevent that outcome?[22]

4.1. State Governments

We argue that just as the state must protect vulnerable humans in circumstances where their caregivers are unable to satisfy their duties of care, the same is true for other vulnerable animals. This means that governments are under a duty to rescue animals in zoos facing financial collapse. However, government subsidies, which many zoos have requested and some governments have begun to provide,[23] are problematic if offered unconditionally: zoos can use such resources in any way they see fit, and it is unclear whether such funds will prevent wrongful killing. Moreover, by subsidizing zoos, governments legitimize what is, as we argued above, a fundamentally unjust institution. Government subsidies to zoos, then, are likely to perpetuate rather than alleviate the unjust treatment of captive animals, both directly (by supporting its operations) and indirectly (by providing symbolic support to the institution).

This leaves current governments in a difficult situation: they have a duty to prevent the killing of captive animals in zoos that are struggling to cope with the crisis, but they must do so in a way that does not perpetuate or compound the injustices that animals in zoos suffer. If unconditional subsidies are not appropriate, how should governments respond? We sketch our preferred solution in section 5.

4.2. Individual Citizens

Individuals too have a duty to assist others in need, even when those others are nonhuman animals. As Peter Singer famously argued, “if it is in our power to prevent something bad from happening, without thereby sacrificing anything of comparable moral importance, we ought, morally, to do it” (Singer 1972, p. 231). Insofar as individuals can do something to prevent animals in zoos from being harmed and wronged as a result of the COVID-19 crisis, we have a duty to do so. Moreover, by acquiescing and participating in a system that allows animals to be kept captive for our entertainment, we citizens are complicit in the institutional structures that allow zoos to exist. This makes us partially responsible for the threat that zoo animals currently face and further places us under a duty to mitigate that threat.

What precisely is the content of these duties? While zoos have called on citizens to donate money to keep them afloat as the pandemic continues, we argue that citizens must not make donations to zoos. While such donations could in principle help alleviate the current financial strain and thus prevent the killing of animals in response to the crisis, they offer no scope for influencing how zoos are run or even how donations are used. For example, as we noted earlier, killing animals for the purposes of “population management” is a routine element of how zoos operate, so there is little reason to be confident that zoos would use these funds to prevent killings; they may simply decide that these funds are better spent elsewhere. In the absence of oversight mechanisms that ensure that such donations would be used for the intended purpose, the risk of supporting the ongoing operation of an unjust institution rather than preventing harm to individual animals is significant.

What if donations are made not in the form of money, but are instead directly targeted to the needs of the animals—for example, by providing food?[24] While this reduces the risk of funds being misdirected, it does not fully address the concern: to the extent that food donations free up resources, they indirectly fund the zoo and thus support its (unjust) operations. In addition, donations—monetary or nonmonetary—legitimize the institution: no matter how critical donors might be of zoos, any donation to a zoo might be interpreted as an expression of support for the institution, and zoos can easily claim that the donations they receive are an endorsement of their work.

Finally, there are plenty of organizations affected by the current crisis that are committed to helping (nonhuman or human) animals, such as animal shelters and rescue organizations.[25] We clearly also have duties to the animals cared for by these organisations. As long as individuals have more duties than they can realistically meet, they should direct their donations towards organizations that are not unjust and where they can be confident that funds will be used for appropriate purposes.

The concerns raised in this section speak against individuals’ and states’ donating to zoos. We turn now to our positive proposals that states should seek to realize, and that citizens should endorse and campaign for.


Calls for boycotts of zoos and aquaria have been successful in some contexts. Consider, for example, the increasingly critical perception of orcas in captivity and the way in which this has affected companies such as Sea World. However, such successes remain focused on animals with higher cognitive capacities, and, for most people, the permissibility of zoos tout court is not in question.[26] It is thus hard to be optimistic that anti-zoo campaigns will make substantial progress, at least in the short term. The present situation, however, presents an opportunity for more radical change.

Given the financial problems zoos face because of the pandemic, they are unlikely to survive without assistance. Some zoos have explicitly announced that they will go bankrupt unless they receive substantial financial subsidies.[27] This gives governments a rare opportunity to fundamentally transform zoos. To be clear, we think that governments are obligated to do more than to get zoos to reform their practices by, for example, requiring them to enlarge enclosures or by outlawing captivity for specific species. The fundamental problems with zoos that we described above cannot be addressed through such reforms. Zoos routinely violate animals’ rights to not suffer, to not be killed, and to be self-determining, and the only way to avoid such violations is for animals not to be in zoos in the first place.

How, then, should governments respond to the financial crisis facing zoos? Our proposal is that governments subsidize the transformation of zoos into sanctuaries and then facilitate the gradual closure of most of these sanctuaries. As a first step, governments should seek to return to the wild those animals for whom independent life (or semi-independent life—e.g., with additional food provided by humans) is a possibility. This is unlikely to be an option for more than a small fraction of animals currently in zoos: life in captivity often leaves animals without sufficient skills to survive in the wild,[28] and environmental degradation has reduced many animals’ natural habitats.[29]

Most zoo animals are unfit for release into the wild and should be allowed to live out their lives in sanctuaries rather than zoos. While, in the real world, the lines between zoos and organizations that call themselves sanctuaries can be blurry, on our understanding, sanctuaries differ from zoos in two important respects. First, sanctuaries are oriented towards their residents’ flourishing (Donaldson and Kymlicka 2015). Sanctuaries “provide a safe-haven for displaced beings— they are fundamentally about those inside the walls” (Kemmerer and Kirjner 2015, p. 231, emphasis in original). Unlike zoos, sanctuaries are focused on animals as individuals, rather than as representatives of a particular species. Second, sanctuaries are not run as businesses and therefore do not depend on fee-paying visitors. Currently, sanctuaries are typically charities and rely on donations; while this addresses some of the concern about the incentive structures faced by zoos, the reliance on donations may create problems of its own (see Emmerman 2015). On our proposal, funding for sanctuaries would be provided by governments and taxpayers. This is important not only because it frees sanctuaries from catering to donors, but also because it underlines sanctuaries’ role in offering homes for animals on whom we have inflicted severe injustice.

What kind of life would former zoo animals lead in sanctuaries? In their discussion of sanctuaries for farmed animals, Donaldson and Kymlicka (2015) distinguish between two “models” that sanctuaries can adopt. The first, the “refuge + advocacy” model, seeks to provide refuge for rescued animals, but also uses refuges for advocacy and educational purposes—for example, by holding “open days” on which visitors can interact with residents. The second, the “intentional community” model, on the other hand, sees sanctuaries as models for just interspecies communities: sanctuaries are permanent homes for the animals who live in them and seek to provide, as far as possible, opportunities for agency. For example, residents are permitted to explore different activities and different levels of social interaction with other residents, of the same and other species.

How, precisely, sanctuaries for former zoo animals would look depends on the kinds of animals housed in them. For nonpredatorial animals, such as goats, integration into a sanctuary that adopts the intentional community model is a congenial option. However, the intentional community model assumes that sanctuary residents are nonpredators: this makes it possible for residents to roam freely in the sanctuary and to interact safely with other residents. Zoos, however, house many predators who typically need to be kept separate from other species (and sometimes from members of the same species). This means that sanctuaries for formerly captive animals would be limited in how far they can move in the direction of an “intentional community” model.

Predators and obligate carnivores pose a broader difficulty for our proposal. Obligate carnivores currently in zoos depend on the death of other animals for their survival. They will either kill these animals themselves (if released into the wild) or have them provided by the sanctuary. Would this be a reason to end these predators’ lives instead of releasing them into the wild or a sanctuary? Clearly, allowing the killing of animals to feed former captive predators is problematic. The suffering and killing of animals always matter morally, even if the harm is inflicted by other (nonhuman) animals.[30] But preventing these deaths by killing the predators violates the rights of the predators. The fact that this tragic conflict has been caused by humans makes the situation even more complex: we have created these predators and have made them suffer in captivity for their entire lives; killing them now instead of allowing them to live out their lives in the wild or, more likely, in a sanctuary only compounds that injustice. At the same time, two wrongs don’t make a right: the fact that the predators are victims of injustice does not by itself justify killing other animals to keep them alive. One possible response to this dilemma is to explore how we can allow former captive carnivores to have decent “retirements” while seeking to reduce, as far as possible, the harm done to other animals to feed them. This might involve, for example, procuring roadkill to feed sanctuary carnivores or exploring plant-based alternatives.[31] As we argue below, sanctuaries for such animals will need to be phased out over time, so we are looking for an acceptable, if morally imperfect, solution for currently existing carnivores, not a long-term solution.

One of the questions sanctuaries would need to address is on what terms, if at all, they should allow visitors. Part of Donaldson and Kymlicka’s concern about the “refuge + advocacy” model is that, by allowing visitors, such sanctuaries “enable forms of animal viewing that may reinforce implicit assumptions about a human entitlement to confine and display animals” (2015, p. 55). While this speaks against short-term visitors who walk through the sanctuary as one would through a zoo, the “intentional community” model allows for long-term guests who become part of the community for the duration of their stays.

We share the concern that it is problematic to open sanctuaries to visitors. Existing wild animal sanctuaries that do allow visitors often emphasize that the presence of visitors is not stressful for the animals. The Wild Animal Sanctuary in Colorado, for example, explains on its website that “large carnivores (and most other animals) do not consider air or sky to be territory, so if people are on elevated platforms or walkways they will not be considered a threat.”[32] The sanctuary is visited by almost 170,000 visitors per year. They do not address other sources of stress visitors might create, such as noise, flashes from cameras, or visitors’ attempts to get animals’ attention. More importantly, however, irrespective of how stressful the visits are for the animals, this does not alleviate the primary concern: that allowing nonhuman animals to be viewed in this way normalizes problematic assumptions about their status; visits therefore reinforce the human-animal hierarchy that underlies our unjust treatment of other animals (Malamud 2017; Sorenson 2008). This concern, we think, is a compelling reason for sanctuaries to remain closed to the public. It also speaks against other ways of facilitating the “viewing” of animals, such as webcams that allow us to watch animals virtually.[33] To adapt Donaldson and Kymlicka’s language, we argue that the sanctuaries housing such animals should follow a “rescue-only” model. They must operate as rescues rather than as intentional communities and would not be used for advocacy purposes: visitors would not be allowed.

What should happen to sanctuaries in the long term? Sanctuaries that, because of the animals they house, can follow Donaldson and Kymlicka’s “intentional community” model are consistent with interspecies justice and can continue to provide homes for former zoo animals and their offspring. This is not the case for sanctuaries that must follow the “rescue-only” model. While such sanctuaries would allow animals to live significantly better lives than they would otherwise have lived, they are far from ideal.

First, while such sanctuaries can provide improved living spaces, it is not clear that even under those conditions, animals’ lives are as good as they would have been in the wild (Emmerman 2014). Providing more space is an important advantage of sanctuaries over zoos—but, again, for many animals this will still be inadequate. Carole Baskin, the founder of Big Cats Rescue, notes in an interview that while the space that the sanctuary provides for the big cats in its care is not enough, “it’s way more than anywhere else.”[34] The degree to which the space and enrichment that sanctuaries can provide falls short of animals’ requirements is likely to vary between species and between individual animals.

Second, animals in such sanctuaries are still subject to confinement and captivity (Emmerman 2014). Even if, as some argue, captivity is not in itself a violation of animals’ rights (e.g., Cochrane 2009), it frequently infringes on their rights to self-determination. While sanctuaries for domesticated animals can allow animals significant choice over the extent to which they want to interact with other species, concerns about territoriality and predator-prey relationships will make this much harder in sanctuaries that house nondomesticated animals. Such sanctuaries, then, may still share many of the features of the total institutions that we discussed in section 3.3.[35] While they should seek to attenuate these concerns—for example, by experimenting with different ways to extend the scope for their residents’ agency—we anticipate that such efforts are unlikely to achieve more than modest improvements for animals’ agency.

This suggests that moving animals from zoos into sanctuaries cannot be the final step of the process we are proposing. The lives they can lead within the sanctuary are of course better than those they can lead in a zoo because the institution is driven by the animals’ interests. But for many animals, such as predators, sanctuaries will involve a significant amount of restriction. For such animals, sanctuaries are temporary solutions to be phased out over time.[36] “Rescue-only” sanctuaries provide captive animals with a home in which to “retire” and comfortably live out their lives, but we cannot be satisfied with the lives these animals lead and we should not seek to repopulate these sanctuaries.

In order for these sanctuaries to become obsolete, their residents must be prevented from procreating, which of course implies yet another restriction of their agency. In addition, this means that animals must not be allowed to procreate even when they are members of endangered species. Our proposal therefore runs counter to the conservation objectives that, to many, support the existence of zoos. This, we think, is a price we must accept. Individual animals have interests and rights; species do not. It is not permissible to make individual animals suffer so that the species can survive (Jamieson 2002, p. 173)—and this would be particularly problematic when it is human activity that has driven a species to the brink of extinction.[37]

The reforms we are proposing here would vastly improve the lives of animals currently captive in zoos. States now have an opportunity to acknowledge the injustice captive animals have suffered and the role they (and we) have played in allowing this injustice to continue. For governments to decide to end the captivity of animals in zoos would send a powerful message of interspecies justice and of the importance of making amends and seeking to repair the injustice we have done.


Of course, we understand that our proposal is unlikely to garner mass appeal, and that support for zoos remains depressingly widespread. If our release/sanctuary proposal is not adopted, does this mean that governments (or individuals) should donate funds to struggling zoos after all? We think the answer is no. First, as we argued above, donations of this kind come with risks, and we cannot be certain that donations would actually prevent the killing of animals, given how zoos are structured. Second, some zoos may simply not survive without significant donations or subsidies and may be forced to close down. Governments should of course do what they can to ensure that the animals in these zoos are placed in sanctuaries rather than moved to other zoos or killed. But bailing out zoos is not a reliable way to fulfil our duties towards the animals who live in them. Indeed, bailing out zoos is morally wrong, and states who make such provisions must be held to account.

This is where the actions of individual citizens are most important. As we mentioned above, individual citizens have a duty to pressure their governments into adopting something like the release/sanctuary model proposed here. This could involve, for example, signing petitions such as PETA’s, writing to local government representatives, discussing the issue with others, or leaving comments on online news stories about zoos. Moreover, individual citizens mustn’t lose sight of the fact that unconditional subsidies to zoos are effectively public funds that are being used to support an unjust institution. Citizens have a responsibility to do what they can to ensure that their taxes are not used for unjust ends, and they have a duty to animals in zoos to end their captivity. It is individual citizens who must hold their governments to account for their failure to adequately protect the rights of animals in zoos.


COVID-19 has been disastrous for many, and it threatens to be disastrous for the millions of animals held captive in zoos. Yet this outcome is not inevitable. It is not inevitable that zoo animals will be killed, nor is it permissible. We have a duty to prevent this. The current situation presents us with the opportunity to abolish zoos by radically reforming the institutions that they currently are.

We have argued here for the following proposal: First, release into the wild (or “supported wild”) as many of the animals currently in zoos and aquaria as possible. Second, the remaining animals are to be housed in sanctuaries; this may well include zoo grounds transformed into sanctuaries. Third, we must distinguish between animals who can have rich, flourishing lives in sanctuaries that protect their rights and agency and those for whom life in a sanctuary necessarily remains limited. While the former can form part of a future that meets the requirements of interspecies justice, the latter must be temporary: they allow animals to live out their lives in better conditions than would have been the case in a zoo, but they must be phased out once the animals in them have died.