Rawls, Animals and Justice: New Literature, Same Response
- First Online:
- Cite this article as:
- Garner, R. Res Publica (2012) 18: 159. doi:10.1007/s11158-011-9173-z
This article seeks to revisit the relationship between Rawls’s contractarianism and the moral status of animals, paying particular attention to the recent literature. Despite Rawls’s own reluctance to include animals as recipients of justice, and my own initial scepticism, a number of scholars have argued that his theory does provide resources that are useful for the animal advocate. The first type takes Rawls’s exclusion of animals from his theory of justice at face value but argues that animals can still be protected within a moral realm independently of justice, or indirectly through the motivations of human contractors. The second type adapts his theory in a way that enables animals to be included within a contractarian theory of justice. It is argued, though, that none of the responses offered is successful in providing a sphere of protection for animals from within Rawls’s contractarian theory. It is doubtful if Rawls’s intention was for animals to receive a significant degree of protection within a moral realm independently of justice, and equally doubtful if the contractors in the original position would be motivated to act on behalf of animals. In the case of the second, whilst Rawlsian resources can be utilised to justify the attempt to amend the veil of ignorance so as to include animals, these are not dependent on a contractural agreement. Similarly, placing emphasis on social-co-operation as a means of incorporating animals into a theory of justice is flawed, not least because, paradoxically, it works for domesticated animals whilst they are being exploited.
In the two decades or so after the publication of Rawls’s A Theory of Justice (1971) a considerable amount of literature appeared designed to assess the degree to which Rawls’s contractarian political theory could be utilised by those concerned about protecting the interests of animals. Despite Rawls’s own reluctance to include animals as recipients of justice, many argued that his theory was still useful for the animal advocate (Richards 1971; Rowlands 1997, 1998; Vandeveer 1979). My initial contribution to this debate, published almost a decade ago, answered in the negative. Those interested in theorising the protection of animals, I suggested, ought to look beyond contractarianism (Garner 2003). Since then, a number of scholars have sought to renew interest in Rawls as a source of resources for animal protection (Abbey 2007; Coeckelbergh 2009; Filice 2006) and Rowlands (2009, chap. 6) has, in a second edition of a book originally published in the 1990s, repeated the claim that a revised version of Rawls’s contractarianism represents the most promising means of theorising the protection of animals.
The aim of this article is to revisit the relationship between Rawls’s contractarianism and the moral status of animals, paying particular attention to the recent literature, in order to assess the degree to which my original dismissal of Rawls’s position can still be maintained. The article is structured around the two major responses to Rawls’s insistence that animals are not recipients of justice. The first takes him at face value but argues that animals can still be protected, either within a moral realm he identifies independently of justice or as a result of their interests being represented by those who are participants in the contractural arrangement. The second seeks to adapt Rawls’s theory in a way that enables animals to be included as recipients of justice, either through amending the veil of ignorance so as to make species, along with gender, race and social situation, an unknown, or through denying that animals do not contribute to society’s cooperative relationship, thereby challenging Rawls’s assertion that justice is only due to those who add to the collective welfare.
Neither of the responses, I argue, is successful in providing a sphere of protection for animals within Rawls’s contractarianism. In the case of the first, it is doubtful if Rawls intention was for animals to receive a significant degree of protection within a moral realm independently of justice, not least because of the importance he attaches to moral pluralism. Likewise, it is equally doubtful if the contractors in the original position would be motivated to act on behalf of animals, or at least all animals, because they perceive that when they are out from behind the veil of ignorance they might care about them. In the case of the second, Mark Rowlands’s attempt to amend the veil of ignorance so as to include animals is only justified if one imports principles from outside of the contractual agreement. These principles can be found in Rawls’s work but they are not dependent on a contractural arrangement. Similarly, placing emphasis on social-co-operation as a means of incorporating animals into a theory of justice is flawed. It may, counter-intuitively, exclude some humans as beneficiaries of justice, requires that co-operation is voluntary given, in order to exclude inanimate objects, and only works for domesticated animals whilst they are being exploited.
Rawls’s Position on Animals
Contractarian theories have provided one arena for those seeking to theorise animal protection, existing alongside more traditional rights and utilitarian approaches (Regan 1984; Singer 1990). The key characteristic of the contractarian approach is the postulating of an hypothetical arena whereby rational individuals decide which principles of justice or morality they can accept. In other words, contractarian theories are committed to the idea that legitimate principles (of justice and morality) are those that are, or would be, agreed upon by rational individuals in appropriate circumstances. That is, legitimate principles are the outcome of an agreement. The advantage of the social contract approach, as Kymlicka (1993, p. 186) suggests, is that it seems to provide valid answers to two crucial questions that any moral theory ought to be able to answer; what does morality require of us and why should we be obliged to accept those demands? Contractarian theories answer by declaring that morality requires us to accept only those principles that we have agreed to and we should be obliged to accept them precisely because we have agreed to them.
Because of the key characteristics of the contractarian approach, it is widely assumed that animals cannot be protected directly from within it (Carruthers 1992, pp. 98–99; Sandoe and Christiansen 2008, p. 19). Rawls appears to confirm this since he excludes animals from his theory of justice on two related grounds. The first is that, since society is regarded as ‘a cooperative venture for mutual advantage’ (Rawls 1971, p. 4), to benefit from society’s collective endeavours requires an ability to be able to provide something in return. Since animals do not contribute to society, at least in conventional ways such as earning money, they are not entitled to be recipients of justice. Secondly, Rawls argues that only ‘moral persons’ are entitled to be beneficiaries of justice (1971, pp. 504–505). Moral personhood, for Rawls has two features. First, moral persons ‘are capable of having…a conception of their good (as expressed by a rational plan of life); and second they are capable of having…a sense of justice, a normally effective desire to apply and to act upon the principles of justice, at least to a certain minimum degree’ (Rawls 1971, p. 505).
One of the problems with Rawls’s exclusion of animals, as he recognizes, is that if animals are excluded from a theory of justice because they are not moral agents then so too are so-called ‘marginal’ humans such as infants, and the severely mentally disabled (see below). This may well explain the subtle change Rawls makes to his position in A Theory of Justice. In an earlier article, ‘The Sense of Justice’, he argues that ‘it is necessary and sufficient that he be capable, to a certain minimum degree, of a sense of justice’ in order to qualify for membership in the original position (Rawls 1963, p. 284). By the time of A Theory of Justice (1971, p. 505, 512) however, the capacity for moral personality’ is now only a ‘sufficient condition for being entitled to equal justice’, and ‘I have not maintained that the capacity for a sense of justice is necessary in order to be owed the duties of justice’. This latter position, of course, raises the possibility that those not possessing moral personhood, including animals, are still entitled to be beneficiaries of the deliberations in the original position. Rawls, though, does not follow though the logic of his argument since he still wants to maintain that ‘it does seem that we are not required to give strict justice anyway to creatures lacking this capacity’, and again that ‘Our conduct towards animals is not regulated by these principles, or so it is generally believed’, and finally that animals are ‘presumably’ excluded from equal basic rights (Rawls 1971, p. 504, 505, 512).
Despite their exclusion from a theory of justice, Rawls does not rule out protecting animals from cruel treatment but, for him—as for many liberal political theorists—this concern belongs to a broader moral arena which is not part of the realm of justice. ‘It does not follow from a person’s not being owed the duty of justice’, Rawls writes, ‘that he may be treated in any way that one pleases. We do not normally think of ourselves as owing the duty of justice to animals, but it is certainly wrong to be cruel to them’ (Rawls 1963, p. 302). In A Theory of Justice he makes a very similar claim writing that ‘it is wrong to be cruel to animals…The capacity for feelings of pleasure and pain and for the forms of life of which animals are capable clearly impose duties of compassion and humanity in their care’ (1971, p. 512). What these ‘duties of compassion’ are we are not told. What we do know is that, as a result of the structure of Rawls’s contract, there is a disincentive for participants to accord any consideration to the well-being of animals since they know they will be human once the veil of ignorance is lifted.
Accepting Rawls at Face Value
Two general types of arguments have been made in response to Rawls’s reluctance to include animals as beneficiaries of justice. The first response is that we can accept his insistence that those not represented in the original position cannot be beneficiaries of the principles of justice there decided. We can then either explore what Rawls might have meant by including animals as part of a wider moral realm or look elsewhere for a more convivial theory of justice.
An example of the first response is provided by Abbey (2007) who argues that we should accept Rawls’s exclusion of animals from a theory of justice, but take at face value his argument that we still have moral duties to them. There are, Abbey argues, more non justice-based normative resources for animals in Rawls’s work than has been recognised previously, particularly in A Theory of Justice (1971) as opposed to Political Liberalism (1993). For Abbey, then, Rawls is suggesting that ‘humans have duties to animals that derive not from the considerations of justice, but from those of morality’ (Abbey 2007, p. 6). Since Rawls refers to our ‘duties’ to animals, we should regard him as serious about the moral claims of animals, despite the fact that he does not provide any detail. Abbey infers that ‘Rawls is seeking to limit the hegemony of rights discourse by recognising that not all issues of ethical concern can be appropriately dealt with via this discourse’ with the implication that ‘morality with its duties, requirements and obligations, is greater than, and different from, justice and its rights’ (Abbey 2007, p. 9).
It is certainly the case, as we have seen, that Rawls recognises that we have duties to animals, and he seems to be saying that we owe these duties directly to them.1 This is consistent with his claim that justice is a much narrower area of inquiry than morality, and the treatment of animals is an issue that is incorporated within the latter and not the former. Rawls touches upon this in A Theory of Justice where he remarks that ‘a conception of justice is but one part of a moral view’ (Rawls 1971, p. 512), but a more detailed account can be found in his book Political Liberalism. Here he suggests that a ‘political’ conception of justice is narrower than a comprehensive view in that it only concerns the basic political structure and not ‘all kinds of subjects ranging from the conduct of individuals and personal relations to the organization of society as a whole’ (1993, pp. 12–13). As a result, ‘the status of the natural world and our proper relation to it is not a constitutional essential or a basic question of justice’, and therefore that ‘our conduct towards animals is not regulated by’ the principles of justice (Rawls 1993, p. 246; 1971, p. 504).
The question of whether there is a viable moral realm independently of justice within which the interests of animals can be protected is an interesting one. The problem with Abbey’s position, however, is that Rawls does not attempt to answer it, so that her interpretation of Rawls represents supposition. We cannot know that Rawls did take the moral worth of animals seriously, and we cannot know that he regarded the duties of morality as ‘greater than, and different from’ those of justice. It seems just as likely that Rawls was putting forward the conventional animal welfare position which holds that the rights of humans trump the interests of animals, or ‘Kantianism for humans and utilitarianism for animals’ as Nozick (1974, pp. 35–42) puts it. The problem is that in the form understood by Rawls, cruelty to animals probably means very little, since, according to one interpretation of the animal welfare position, the treatment of animals only becomes cruel when it ceases to have any useful human benefit. It is difficult to see how gratuitous cruelty, such as inflicting suffering on an animal for the fun of it, could be objected to on the grounds that it is an interference with a conception of the good. Most suffering on animals, by contrast, is inflicted for a reason, whether it be on religious grounds (as in ritual slaughter) or in the pursuit of cheap food or public health.
More significantly, Abbey does not take enough account of what Rawls actually does say, and in particular the primary importance that he attaches to his principles of justice. Crucially, any suggested moral duties to animals are likely to conflict with these principles. In particular, Rawls’s liberty principle, designed to allow individuals to pursue their own conceptions of the good without interference from the state, are always likely to trump attempts to protect the interests of animals, where such attempts conflict with the liberty of humans. Thus, the problem for animals within Rawls’s theory of justice is that central to it is the common liberal assertion that it is no part of the state to intervene in competing conceptions of the good. Since the treatment of animals is regarded as a conception of the good for Rawls, the logical conclusion is that the state should not intervene to regulate the treatment of animals. Whatever is done to animals, therefore, is entirely legitimate, and the state has no justification in intervening to prohibit it. I would argue that this paradigmatic liberal principle, of moral pluralism, has had a considerable practical impact on the way animals are treated in practice, particularly in relation to the issues of hunting and ritual slaughter (Garner 2005, pp. 70–81).
Abbey suggests two possible responses to this negative impact that Rawls’s principle of moral pluralism seems to have for animals. First, she asks whether the cruel treatment of animals might be regarded by Rawls as a type of ‘unreasonable’ pluralism in the moral sphere, in the sense that it would be prohibited precisely on the grounds that it harms animals unjustifiably (Abbey 2007, p. 11). The problem here is that, for Rawls, what counts as ‘unreasonable’ is defined by what would be allowed by principles of justice. That is, Rawls wants to rule out those doctrines and conceptions of the good life which are ‘in direct conflict with the principles of justice’ and, in particular, those ‘requiring the repression or degradation of certain persons on, say, racial, or ethnic, or perfectionist grounds’ (Rawls 1993, pp. 195–196). On these grounds, Rawls rules out human slavery as unacceptable (1993, pp. 151–152, p. 161). By contrast, since animals are not included in his theory of justice, what is done to them cannot be constrained by rules of unreasonable pluralism.
It is true, as Abbey points out in her second response to the problem for animals of moral pluralism, that Rawls’s commitment to moral pluralism was more pronounced in his later Political Liberalism than in A Theory of Justice. Nevertheless, it is consistent with the structure set up in A Theory of Justice, whereby justice, and therefore the state, are clearly to be kept out of morality. Thus, it is in A Theory of Justice that Rawls writes that ‘each person is free to plan his life as he pleases (so long as his intentions are consistent with the principles of justice)’ (1971, p. 446). It is in A Theory of Justice too, as was pointed out above, that Rawls sets out the lexical priority of liberty, and this was designed precisely so as to allow the pursuit of various conceptions of the good once the veil is lifted. As Rawls points out: ‘To have a complaint against the conduct and belief of others we must show that that their actions injure us’ since ‘it is, in general, a good thing that individuals’ conceptions of the good should differ in significant ways’ (Rawls 1971, p. 447, 450).
As a result, the question here must be that if Rawls thinks animals are significant enough morally, then why did he not include them as recipients of justice? This is not to say that Abbey (2007, pp. 14–15) is necessarily wrong when she emphasises the potential usefulness of non-rights and justice-based animal protection discourses. It is only to say that the Rawlsian version offers nothing much to animals if they cannot be incorporated into his theory of justice.
Another example of the first response to Rawls—that we should accept his conclusion that animals cannot be direct beneficiaries of justice—is to allow individuals within the original position to protect nonhuman animals, not because they calculate they might end up being nonhuman animals, but because of some other motivational device. Such an approach does not dispense with moral agency as the crucial qualification for being considered a beneficiary of justice. Rather, it sees the moral agents in the original position acting on behalf of animals. Clearly, the motivational device cannot be based on altruism, since Rawls’s contract is predicated on the self-interest of the participants in the original position. However, it could be based on a calculation that, once out from behind the veil of ignorance, the participants in the original position might discover that they do care about animals and the ability to do so is important to them.
Interestingly, Rawls himself does use such a motivational device to accord justice to future generations. Thus, he argues that ‘we may think of the parties as heads of families, and therefore as having a desire to further the welfare of their nearest descendants’ (1971, p. 128). Participants in the original position still pursue their own interests but these interests are expanded so as to include the welfare of (at least some) future generations. The mantle for animals is taken by David Richards, a PhD student of Rawls’s, who incorporates animals into his theory in the same way that Rawls incorporates future generations (Richards 1971, p. 207). Participants in the original position are aware ‘that persons generally have certain basic sympathies with animals and animal life’, and, to add an anthropocentric element, they ‘will understand cruelty to animals as an extension of a personality orientation which is prone to cruelty to persons’ (Richards 1971, p. 182).
However, there are problems with this move. In the first place, as Brian Barry (1989, p. 192) points out in the context of future generations, obligations are ‘dependent purely upon the actual goodwill of contemporaries towards their descendants’ and the same would apply to the goodwill of humans towards animals. Therefore, ‘the demands of justice thus depend on the contingent facts about the extent to which people care about the welfare of (at least some) future people’ (Barry 1989, p. 192). This move is necessary because future generations (and animals) ‘would not have any just claims in their own right, but would simply have indirect claims by virtue of the sentiments of the principal parties to the social contract’ (Barry 1989, p. 245). Secondly, we can speculate that humans are likely to care more about what happens to future humans than they are to contemporary animals. There must be a significant doubt whether the contractors would know that people are generally sympathetic to animals since it is far from being a universal human trait, and in so far as it does exist it tends to be reserved for companion animals and not those used to produce food or as scientific subjects.
Rowlands, Rawls and Animals
The second response to Rawls’s exclusion of animals from his theory of justice is to seek to adapt his theory in a way that enables animals to be included. Two main adaptations are possible. The first is to amend the veil of ignorance so as to make species, along with gender, race and social situation, an unknown. Since participants in the original position will then not know whether they will end up being humans or animals they will protect themselves against the latter eventuality by making sure that animals are owed duties of justice. The second is to deny that animals do not contribute to society’s cooperative relationship, thereby challenging Rawls’s assertion that this characteristic excludes them from being owed duties of justice.
In the case of the first, it is, of course, the case that amending the veil of ignorance to incorporate animals challenges Rawls’s assertion that only moral agents can be recipients of justice. As we saw earlier, one difficulty for Rawls here (and for contractarianism in general) is that insisting upon moral agency as an entry qualification for justice also has the effect of excluding some humans, such as the very young and the severely mentally disabled. Rawls’s justification for including so-called ‘marginal’ humans is that infants will eventually become persons and marginal adults used to have personhood. Of course, some marginal humans have never been persons and never will be, but Rawls still includes them within his theory of justice on the grounds partly that to fail to do so is ‘a risk to just institutions’, and partly because these individuals would have been parties to the contract but for their unfortunate circumstances (1971, p. 509). He does not, however, explain further what this ‘risk’ might be, although others, such as Carruthers (1992, pp. 114–118), have opined that excluding marginal humans from an entitlement to justice would lead to a slippery slope whereby the boundary line for inclusion and exclusion would become blurred. In addition, Rawls’s second reason for including marginal humans—that those humans born as non-persons were unfortunate and should not be penalised for their bad luck—is open to the response that being born as an animal is similarly a matter of chance.
Recognising, no doubt, the potential inconsistency of his position, Rawls does, as we saw above, cast doubt on the strength of his commitment to the importance of moral agency, arguing that he is unsure whether it is a necessary condition for being a recipient of justice (Rawls 1971, p. 505). Barry (1989, p. 211) is right to suggest that this doubt is because Rawls recognises the inconsistency of not including animals but including those humans who have not yet gained moral agency or who will never possess it. His initial assumption, that animals are not to be included as recipients of justice is, though, confirmed.
The reality is that Rawls’s contractarian political thought finds it difficult to include those entities who are not moral agents, whether they are marginal humans or animals. The claim that it is counter-intuitive to exclude the former as recipients of justice does not by itself, of course, justify amending Rawls’s contract to include both marginal humans and animals. What it might do is to persuade us that we have to look elsewhere for a theory that can do this work, or, alternatively, we could reject arguments based on intuition and argue that neither marginal humans nor animals are the kind of beings who can be recipients of justice. This position might be made more palatable if we expand the duties owed within the moral realm identified by Rawls to marginal humans as well as animals.
In order to justify amending the terms of the contract to include animals (and marginal humans), we need a further argument. The most sustained and original attempt has been provided by Rowlands (1997, 1998, 2009). To get to grips with Rowland’s argument it is necessary to see, initially, why it is that Rawls (and contractarian theories in general) finds it difficult to incorporate animals as well as marginal humans. When one thinks of a contract designed to come up with principles of justice one envisages a situation where self-interested individuals meet to decide upon principles under which they are to live. In this type of contract, ‘consisting of mutually advantageous rules of conduct’ (Rowlands 2009, p. 123), the principles of justice adopted are constituted by the contract, and their authority derives from the fact that those in the contractual situation have agreed to them because they are perceived to be in the self-interest of the participants. In this scenario, the only justification for seeking to agree principles with others is if they can help or hinder us in some way. Since animals do not come into this category, they must be excluded from the contractual situation.
For Rowlands, this type of contract is a Hobbesian version. But there is another version of the contract, a Kantian version, and Rowlands argues not only that this version enables us to include animals, but also that this is the version that Rawls, in the main, adopts. So, for Rowlands, only a Hobbesian type of contract, where the self-interest of the participants is the only concern, is consistent with the exclusion of animals.2 In a Kantian-type of contract, on the other hand, the principles of justice adopted are not merely constituted by the participants. Rather, the principles arrived at through the contract have to be continually checked against pre-existing moral principles that exist independently of the contract and the contractors. Thus: ‘Contained in the idea of Kantian contractarianism…is an at least minimal conception of moral objectivity that is independent of the contract and the agreements reached by contractors’ (Rowlands 2009, p. 126). Rowlands argues that the key ‘moral law’ that Rawls advocates is ‘equal consideration’ or what Rowlands calls the ‘intuitive equality argument’ (IEA). This is the argument that individuals should not benefit from the possession of characteristics, such as ability or rationality, for which they are not responsible. The principles of justice emanating from the contract, then, must, for Rowlands, be consistent with the IEA.
If we adopt this principle, Rowlands suggests, there is nothing to stop us from including non-moral agents, such as animals and marginal humans, as beneficiaries of principles of justice. Rawls’s failure to see this, according to Rowlands, was because of the ‘unexpurgated and unnecessary elements of Hobbesianism’ in his theory (Rowlands 2009, p. 153). For Rowlands, the IEA insists that we do include animals (and marginal humans) because rationality itself is not a characteristic that we have earnt. Rather, we just have it as a matter of luck. Thus, just as specific human abilities, and more general characteristics such as age, gender, race and class, are hidden behind the veil of ignorance, rationality is an equally undeserved natural advantage which also ought to be hidden. Indeed, a failure to include species as a hidden characteristic, it is argued, would be tantamount to speciesism in the same way that to allow knowledge of race or gender in the original position would be racist and sexist (Vandeveer 1979, p. 374).
Once it is allowed that knowledge of one’s species should be one of those things excluded by the veil of ignorance, it would be just as irrational to opt for a system that permitted harmful or injurious treatment of non-humans as it would be to opt for a system that permitted the same sort of treatment for humans (Rowlands 2009, p. 166).
Three major criticisms can be made of Rowlands’s attempt to amend Rawls. I will deal with the two relatively minor ones before outlining the major problem. The first is raised by Baxter (2005), who questions, with some justification, whether it is possible for participants behind the veil of ignorance to be ignorant of their status as human beings. Rawls imputes to the contractors a number of intellectual capacities, and not least the capacity to understand the ‘maximin’ principle (the risk-averse strategy Rawls claim contractors in the original position will adopt). Given this, Baxter argues, ‘there is no room for the idea that they could also be ignorant of their species, for they would know for certain that they would be members of species which possess at least these minimal attributes’ (Baxter 2005, pp. 95–96).
Of course, Rowlands and others would respond by saying that this does not preclude the possibility of the human contractors imagining that they may still turn out to be nonhumans once the veil is lifted. We might say, for instance, that the contractors possess the characteristics of personhood only for the duration of the debates in the original position, and once the veil is lifted these characteristics no longer necessarily apply. Baxter claims this move is ‘impossible to make sense of’ because the resulting change would be too drastic (Baxter 2005, p. 96). Now, whilst Baxter is right to claim that it would be difficult, if not impossible, for the rational contractors in the original position to imagine turning out to be a bacterium, it is not so difficult for a human person to imagine being a sentient mammal. Indeed, there is a case for saying it is no more difficult for a human person to imagine being a mammal than it is for her to imagine being a severely mentally disabled human.
It should be noted that we can accept that the interests of animals can be incorporated into Rawls’s theory of justice and still dispute Rowlands’s assertion that this would produce principles of justice that would sanction abolitionist objectives such as a prohibition on the eating of animals. There is a great deal of literature on Rawls which questions whether his contractors would adopt the maximin strategy (Wolff 1996, pp. 177–186). In the case of animals, it might be the case that contractors would risk turning out to be nonhuman. This likelihood is enhanced if they rationalise that, as animals, they would not be autonomous agents and therefore had less to lose by death. Of course, the exploitation of animals in factory farms involves the infliction of a great deal of suffering. However, prohibiting intensive animal agriculture is not the same as prohibiting the eating of animals. It is least a possibility, then, that the contractors behind Rawls’s veil of ignorance would choose to prohibit the infliction of suffering on animals—on the grounds that they might turn out to be animals—whilst still sanctioning the eating of animals—on the grounds that this pleasure would not be denied to them if they turned out to be humans.
The biggest problem with the approach suggested by Rowlands relates to his Kantian interpretation of the contract. We can readily accept that this is closer to Rawls’s own position than the Hobbesian version. Crucial here is Rawls’s reliance upon a process of ‘reflective equilibrium’ whereby his principles of justice are measured against widely-accepted moral intuitions (Rawls 1971, pp. 48–52). However, the problem with accepting that there are moral principles we value independently of what is decided by the participants in the contractual situation is that it reduces the importance of the contract device. It amounts to saying that the contract must be so organised as to reflect these important pre-existing moral principles, or at least to consider then seriously. Rawls’s exclusion of animals is arguably, therefore, made prior to his use of the contractual device. That is, by concluding that only moral agents should be beneficiaries of justice Rawls has already made the decision that animals ought to be excluded from his theory of justice. Principles emanating from the contract are, therefore, not the result of an objective account of what participants in the original position would choose, but reflect pre-existing normative judgements, one of which is that justice only applies to persons.
What I am arguing here, then, is that a ‘pure’ form of the contract does make it difficult to include animals as beneficiaries of a contractarian theory of justice. Rawls does not, however, offer such a ‘pure’ version of contractarianism in which the principles of justice adopted are constituted by the contract, and where their authority derives from the fact that those in the contractual situation have agreed to them. Therefore, there is a strong case for saying that Rawls does offer resources for the protection of animals. That is, Rowlands may well be right to say that what he describes as the ‘intuitive equality argument’ does in fact, despite Rawls’s denial, allow for the inclusion of animals and marginal humans.
What should be recognised, however, is that the principle that allows for the inclusion of animals as beneficiaries of justice, for both Rawls and Rowlands, derives from outside the confines of the contractual arrangement. Indeed, Rowlands admits that the ‘contract does not determine who does and who does not count morally’ (Rowlands 2009, pp. 127–128). The question to ask, therefore, is what is the value-added of persevering with a contractarian approach for those interested in the protection of animals? And if the answer, as it must be, is that it adds very little, then there is a compelling case for saying that we would be better off invoking the intuitive equality argument as a free-standing principle from which the justice claims of animals and marginal humans can be derived independently of the contract. That is, the use of the contract device, which is the most distinctive element of Rawls’s thought, is not necessary to establish the validity of the intuitive equality principle and the inclusion of animals as beneficiaries of justice.
The second way of adapting Rawls to allow for the inclusion of animals as recipients of justice is a challenge to his assertion that only humans are able to contribute towards society. ‘We are not to gain from the cooperative labors of others without doing our fair share’, Rawls (1971, p. 96) writes, and it is clear that he only has humans in mind here. It has been argued by Coeckelbergh (2009) that animals are not excluded by such a principle because they do, in fact, make a substantial contribution to society. He therefore argues that a moral evaluation of animals should move away from a focus on ‘what non-humans are’ (in terms of their intrinsic value or capacity to suffer) ‘towards what “we”…do together’ (Coeckelbergh 2009, p. 69. Italics original). Animals are therefore morally considerable because of their relations with us. Thus, animals are a source of food for human and other animals, they contribute towards medical research and toxicity testing, and they provide entertainment for us. In return, animals (or at least domesticated animals) depend on us for their lives.
Almost all the animals whose interests are addressed by animal welfare legislation exist explicitly because humans chose to bring them into the world. Indeed, from the moment the animal is bred, until he or she is killed, humans manipulate every aspect of the lives of most captive animals’. As a result, animals are ‘part of a larger co-operative scheme (O’Sullivan 2007, p. 8).
There are a number of problems, however, with Coeckelbergh’s analysis. In the first place, deriving principles of justice from the relational fact of social co-operation, as opposed to the characteristics possessed by individuals, would seem to lead to the exclusion of some humans as beneficiaries of justice. Thus, some humans, such as infants and the severely mentally disabled, may contribute little, if anything, to society and therefore ought, according to the logic of Coeckelbergh’s position, be excluded as beneficiaries of justice. Such a position, of course, might be regarded as counter-intuitive.
Another objection to the employment of social cooperation as a means whereby animals can be incorporated into a contractarian theory of justice is that, if it is established that all contributions to society count, there would seem little to prevent the inclusion in a theory of justice of those inanimate objects, such as rivers and mountains, which contribute to the sum of benefits in society. This is clearly an outcome which Rawls would not accept. In order to exclude the contribution of inanimate objects, we could limit those contributions that are voluntarily given, but this would also produce results undesirable to both Rawls and Coeckelbergh. In the case of Rawls, it would have the effect of denying justice to those humans, such as slaves, whose contribution is not voluntarily given (Baxter 2005, p. 79). For Coeckelbergh, it would deny the force of his argument that animals contribute to society, since they do not volunteer to be raised and killed for food or to be the subjects of scientific experiments.
Of course, we could, as Baxter does, make the claim, in the context of a theory of ecological justice, that ‘something making a contribution to the sum of environmental benefits is a necessary, but not sufficient, condition of its being an appropriate recipient of ecological justice’ (Baxter 2005, p. 84). What is missing here, though, is a theory of moral considerability possessed by those due justice and not by those denied it. The problem here, of course, is that Rawls would insist upon personhood as the necessary criterion which rather takes us back to square one because this would exclude most, if not all, nonhuman animals and therefore deny the force of Coeckelbergh’s argument about the importance of co-operation and the contribution animals make to society.
A further objection to Coeckelbergh’s approach, as he himself recognises, is that social-cooperation only obviously applies to domesticated animals. It is true that, as indicated above, wild animals may contribute to human society (in the sense that they provide aesthetic pleasure for humans or income as in the case of eco-tourism). However, many also conflict with humans in a variety of ways. In so far as they do, they are excluded from a theory of justice dependent upon social co-operation. Finally, according to Coeckelberg, animals will only be owed duties of justice whilst they are in a co-operative relationship with humans. The problem with this is that the objective of the abolitionist animal rights movement is to end this relationship. Once ended, according to Coeckelberg’s position, the ‘liberated’ animals are no longer due justice. This amounts to the rather odd position that animals are only due justice whilst they are being exploited. In practical terms, it means, for instance, that it is just for humans to eat animals because in this way they contribute to human society but, since vegetarians are no longer in a co-operative relationship with food animals, desisting from eating animals is neither a just nor an unjust act.
The main conclusion of this article is that attempts to find support in Rawls’s contractarian political theory for the protection of animals generally fail. Firstly, it is far from obvious that Rawls intended to allocate a significant degree of protection for animals within a moral realm independently of justice, and the moral pluralism that is central to his conception of the good would, in any case, militate against such a project. Likewise, it is wishful thinking to suppose that animals would receive indirect protection as a result of contractors in the original position coming to the conclusion that they might care about them once out from behind the veil of ignorance. Secondly, the tactic of ‘thickening’ the veil of ignorance, so as to include species as an unknown characteristic, is only justified if one imports principles from outside of the contractual agreement. These principles can be found in Rawls’s work but they are not dependent on Rawls’s distinctive contract device. Similarly, placing emphasis on social-co-operation as a means of incorporating animals into a theory of justice may, counter-intuitively, exclude some humans as beneficiaries of justice, requires that co-operation is voluntary given, in order to exclude inanimate objects, and only works for domesticated animals whilst they are being exploited.
It is understandable why attempts have been made to utilise Rawls on behalf of animals. Not only is contractarianism in general able, apparently, to answer key questions about moral obligation, but, in addition, Rawls’s version of it has, of course, taken a central place in Western political thought. It is time, though, to look elsewhere for a theory that can provide the elevated moral status for animals that many of the scholars discussed in this article seek.
A direct duty to animals can be contrasted with an indirect duty. According to an indirect duty view, the protection of animals does not come about because they are regarded as having intrinsic value, but because (some) humans regard the protection of animals as being part of a good (human) life. Thus, whereas a direct moral object is ‘something to which moral consideration is paid’ an indirect moral object is ‘something about or concerning which moral consideration is paid’ (Morris 1998, p. 191).
Rowlands takes this distinction from Kymlicka (1993). Hobbes was clear that the social contract could not include animals. ‘To make covenants with brute beasts’, he wrote, ‘is impossible, because not understanding our speech, they understand not, nor accept of any translation of right, nor can translate any right to another; and without mutual acceptation, there is no covenant’ (quoted in Fellenz 2007, p. 107). A contemporary version of the Hobbesian contract is provided by David Gauthier, and he comes to the same conclusion that: ‘Animals, the unborn, the congenitally handicapped and defective, fall beyond the pale of a morality tied to mutuality’ (Gauthier 1986, p. 268). Narveson (1983) excludes animals on the same grounds. He writes: ‘Contractarianism leaves animals out of it…They are, by and large, to be dealt with in terms of our self-interest, unconstrained by the terms of hypothetical agreements with them. Just exactly what our interest in them is may, of course, be matter for debate; but that those are the terms on which we may deal with them is, on this view of morality, overwhelmingly indicated’ (Narveson 1983, p. 58).
I would like to thank the two anonymous referees for their incisive comments on a first draft of this article. The piece was researched and written whilst I was the holder of a Leverhulme Research Fellowship and I would like to thank the Leverhulme Trust for its support.