Laura Valentini Canine justice: an associative account Article (Accepted version) (Refereed) Original citation: Valentini, Laura (2014) Canine justice: an associative account. Political Studies, 62 (1). pp. 37- 52. ISSN 0032-3217 DOI: 10.1111/j.1467-9248.2012.01006.x © 2013 The Author. Political Studies © 2013 Political Studies Association This version available at: http://eprints.lse.ac.uk/63699/ Available in LSE Research Online: January 2016 LSE has developed LSE Research Online so that users may access research output of the School. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Users may download and/or print one copy of any article(s) in LSE Research Online to facilitate their private study or for non-commercial research. You may not engage in further distribution of the material or use it for any profit-making activities or any commercial gain. You may freely distribute the URL (http://eprints.lse.ac.uk) of the LSE Research Online website. This document is the author’s final accepted version of the journal article. There may be differences between this version and the published version. You are advised to consult the publisher’s version if you wish to cite from it. http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1467-9248 http://dx.doi.org/10.1111/j.1467-9248.2012.01006.x http://www.psa.ac.uk/ http://eprints.lse.ac.uk/63699/ Canine Justice An Associative Account∗ Laura Valentini University College London Abstract: A prominent view in contemporary political theory, the ‘associative view’, says that duties of justice are triggered by particular cooperative relations between morally significant agents, and that ‘therefore’ principles of justice apply only among fellow citizens. This view has been challenged by advocates of global justice, who point to the existence of a world-wide cooperative network to which principles of justice apply. Call this the challenge from geographical extension. In this paper, I pose a structurally similar challenge to the associative view: the challenge from species extension. This says that the existing network of cooperation extends beyond the human species, to encompass some non-human animals, particularly domesticated dogs. In light of this, if one believes that (i) certain non-human animals are morally significant (i.e. objects of moral concern), and that (ii) justice applies to fellow cooperators, one should also hold that domesticated dogs are owed justice in much the same way our human fellow citizens are. I conclude by considering the implications of this argument for the associative view, and animal-rights theory. Introduction A prominent view in contemporary political theory says that duties of social justice apply only among fellow social cooperators. On this ‘relational’ or ‘associative’ view, people who stand in certain types of social relationships acquire a privileged moral status vis-à-vis one another, over and above the basic moral concern owed to ∗ I am grateful to the participants at the Manchester Metropolitan University Workshop on Political Theory and Animal Rights (Manchester, September 2010), and at the Queen’s College Symposium (Oxford, November 2010) for discussion. Special thanks are owed to Alasdair Cochrane (who patiently read multiple versions of the paper), Steve Cooke, Bob Goodin, Will Kymlicka (and his reading group on animal rights), Christian List, and the anonymous reviewers of this paper for their very detailed comments and helpful suggestions. The paper was written while I was a postdoctoral researcher at Princeton University (2009/10) and revised while I was a visiting fellow at the Swedish Collegium for Advanced Study in Uppsala (Fall 2011). I am grateful to both institutions for their support. 2 human beings as such. In particular, proponents of the associative outlook argue, people who cooperatively support the institutions of their own societies are, unlike others, entitled to social justice. This view, most prominently advocated by John Rawls (1971), has often been taken to entail a conservative stance on the scope of principles of justice. For (many) associativists, the reach of such principles is typically confined to domestic political communities, to the exclusion of distant strangers who do not qualify as fellow cooperators (Rawls 1971 and 1999, Sangiovanni 2007). This ‘narrow’ associativist view has been repeatedly challenged by advocates of global justice, who point to the globalized nature of today’s world as evidence for the existence of a world-wide cooperative network to which principles of justice should apply (see e.g. Beitz 1999 and Pogge 1989). We might call this challenge to narrow associativism the challenge from geographical extension. The challenge does not question the normative foundations of the associative view, but simply says that, consistently applied, the view delivers much more radical conclusions than many of its advocates have acknowledged. In the present paper, I pose a similar challenge to the narrow associative view, which we might call the challenge from species extension. This challenge says that taking the associative view seriously implies explicitly acknowledging that the morally relevant network of cooperation extends beyond the human species, to encompass at least some non-human animals. I make this case specifically with respect to domesticated dogs, though the logic of the argument could be extended to other animals too. In a nutshell, I argue that if one believes that (i) certain non- human animals are objects of moral concern and that (ii) justice applies to fellow cooperators (provided they are objects of moral concern), then one should also 3 believe that domesticated dogs are owed justice in much the same way our human fellow citizens are. The contribution of this paper is twofold, to animal ethics and to political theory. First, the paper illustrates how a relational approach to morality may apply to non- human animals, specifically in connection with the duties we owe them on grounds of justice. This is of value insofar as much advocacy of animals’ ethical status focuses on the type of creatures they are (i.e. sentient beings with lives to lead), rather than on the ways they relate to us.1 To be sure, some thought has already been given to the responsibility-generating character of special relations between humans and animals. However, emphasis on contextual relations is still relatively marginal in the literature and has been often accompanied by a rejection of the justice-based approach adopted here, in favour of what might be termed an ‘ethics of care’. Until recently, the key normative concern of ‘relational’ views has been not so much giving animals ‘their fair share’, but responding to their emotional and physical vulnerabilities2 (see, however, Coeckelbergh 2009 and especially Donaldson and 1 For instance, see the mission statements of the UK party Animals Count; The German Animal Protection Party, and the Dutch Party for the Animals. 2 Mary Midgley (1983) grounds animals’ moral status in their emotional abilities and capacity to create social ties with humans, with whom they share a ‘mixed community’. J. Baird Callicott (1992) defends a somewhat similar view, which emphasizes how our moral obligations towards domesticated animals radically differ from our obligations towards wild ones, which do not belong to our communities. For a critique of this perspective see Hadley (2007). Feminist defences of non- human animals’ moral standing have also focused on animals’ ability to feel emotions and their susceptibility to domination by humans (see Donovan and Adams 2007). For another relational view, according to which nonhuman animals are owed duties of help/assistance in virtue of their contextual vulnerability to, and dependence on, human beings, see Palmer (2010). Finally, Anderson (2004) defends a pluralist perspective on the moral status of non-human animals, with a particular emphasis on their social membership and relationships of reciprocity with humans. 4 Kymlicka 2011, for very important exceptions).3 Second, the paper shows that the associative account of justice, so prominent in contemporary political theory, has potentially radical implications, which its proponents have by and large failed to acknowledge, and which they might not want to accept. In sections 1 and 2, I explain the two basic premises of my argument: (i) non- human animals are objects of moral concern, and (ii) social justice is owed to fellow co-operators (provided they are objects of moral concern). In subsequent sections, I illustrate how, if these premises are endorsed, the nature of our relationship with domesticated dogs should lead us to conclude that they are owed justice as much as human fellow citizens are. Specifically, in section 3, I give a brief descriptive account of how relations between humans and dogs have evolved over millennia, and consider dogs’ role in modern society. I argue that most domesticated dogs are an integral part of our existing schemes of cooperation and are therefore entitled to more than the minimal concern all animals have a right to. In section 4, I offer a tentative account of what giving dogs their due requires. In section 5, I consider an important objection, and I then conclude with a few remarks on what the 3 Sue Donaldson and Will Kymlicka’s book, Zoopolis, was published when this article was nearing completion, and I am therefore unable to discuss it in detail. The book argues that the domestication of animals has created relations of membership that generate distinctive claims of justice. Specifically, Donaldson and Kymlicka (2011) apply liberal citizenship theory to the debate on animal rights, bringing notions such as ‘sovereignty, denizenship, migration, territory, membership and citizenship’ to bear on it (p. 14). Different animals, they suggest, relate to our societies in very different ways (e.g. as denizens, citizens, foreigners etc.), and this qualifies them as holders of correspondingly different sets of rights. To the extent that relationships of fellow citizenship are seen as relationships among fellow cooperators, the (much more comprehensive) approach Donaldson and Kymlicka defend is very much in line with the associative account I am exploring here. In both of our cases, there is an attempt to apply conceptual tools developed in another area of political philosophy (i.e. liberal global justice in my case, liberal citizenship theory in Donaldson and Kymlicka’s) to the debate on animal rights. 5 extensibility of duties of justice to non-human animals implies for the associative account itself. Before getting started, let me clarify that I do not aim to defend premises (i) and (ii) – about animals’ moral status and associative justice – but only to show that their conjunction, coupled with some empirical facts, has surprising implications for our thinking about justice and animals’ (specifically dogs’) entitlements. Since such premises are plausible and, in fact, widely endorsed, my lack of commitment to them should not undermine the contribution of the paper. 1. Non-human Animals as Objects of Basic Moral Concern The claim that we owe particular associative obligations to dogs presupposes not only an associative account of justice, but also the view that non-human animals (including dogs) are appropriate objects of moral concern in the first place. To see this, consider the following example. Although I am engaged in a sort of cooperative (perhaps exploitative) relation with my computer, the suggestion that by virtue of this relation I owe justice to it seems absurd. Why? Not because of a flaw in the associative perspective, but rather because my computer is not the sort of entity to which moral concern is owed, or at least so many would think.4 Animals, however, are unlike computers in that they are sentient creatures with lives to lead (Francione 2004). Although they lack the higher-order capacities for moral reasoning that would qualify them as subjects of moral responsibility, this does not automatically disqualify them as appropriate objects of moral concern 4 Some might disagree, however. For instance, Coeckelbergh (2009) thinks that an associative/contractarian perspective can be developed prescinding from intrinsicist claims about creatures’ fitness for having moral status. For this reason, he puts forward the interesting and controversial view that contractarian/associative justice extends even to plants and robots. 6 (Regan 1983). Within the human realm, we routinely distinguish moral responsibility from moral concern. Small children and the severely mentally disabled lack the necessary features to count as subjects of moral responsibility, but we regard them as worthy objects of moral concern. These kinds of observations have led many theorists to conclude that non-human animals count as members of the moral community. For instance, consequentialists have argued that animals’ sentience – their capacity to feel pain and pleasure – makes them appropriate objects of moral concern, whose well-being should be given weight in the moral evaluation of our actions (see e.g. Singer, 1990). Deontologists, on the other hand, have insisted that (at least some kinds of) non-human animals are appropriately regarded as bearers of fundamental rights (see e.g. Regan 1983 and Francione 2000). Since my primary focus here is on the associative approach to justice, and this approach is situated within a broadly liberal-deontological perspective, I more narrowly concentrate on deontological defences of animals’ moral status. I therefore interpret premise (i) of my overall argument – that animals are objects of moral concern – in deontological terms, whereby the relevant concern is expressed in the form of rights, and duties correlative to them. Different ethicists have focused on different characteristics exhibited by animals as the basis for granting them rights. Tom Regan (1983), for instance, has argued that to the extent that animals are conscious ‘subjects-of-a-life’, for whom things can go better or worse, they have inherent value and should therefore be regarded as right-holders. Francione (2000), by contrast, has grounded animal rights in animals’ sentience. Other animal-rights advocates, such as Rachels (1976), have placed 7 emphasis on animals’ having interests.5 For present purposes, I need not worry about these differences. Readers can simply choose the account they find most congenial. All I need to assume is a broad defence of animals’ entitlement to basic, non-relational moral concern. Granted that such a defence is available, what does this basic moral concern amount to? Two universal duties are generally acknowledged within deontological moral theories: a very stringent duty not to harm others, and a less stringent duty to help them when they are in need, provided this is not too costly to oneself (Rawls 1971, p. 114). These are ‘natural’ duties, which fall upon moral agents regardless of their voluntary actions or the particular relationships in which they stand vis-à-vis one another.   The duty not to harm is universal and correlative to a universal right not to be harmed. This means that every moral agent is under a duty not to harm (e.g. physically injure, steal from, harass) every other moral subject. In turn, every moral subject has an entitlement to not being harmed by any other moral agent. The duty to help the needy has a different structure. Although it is universal, i.e. it binds every moral agent, it is not correlative to a universal right. It is not the case that every moral agent is under a duty to help every other needy creature, a demand 5 Note that what follows from animal rights at the level of policy varies from theorist to theorist. Some animal-rights ethicists broadly subscribe to what is known as ‘abolitionism’, namely the view that the use of animals on the part of humans is morally unacceptable. See e.g. Regan (1983) and especially Francione (2000, 2010 with Garner). Other ethicists, however, are less drastic about the implications of conferring rights on non-human animals. See e.g. DeGrazia (1996 and 2002), and Rollin (1981). There are then those, like Garner (2010 with Francione), who adopt a more gradualist approach, and argue for the regulated and humane use of animals either per se, or as a fruitful strategy to achieve abolitionist goals. I thank an anonymous reviewer for suggesting to clarify this. For a general overview of the debate, from which I have learnt, see DeGrazia (1991). See also section 5, below, for further discussion of the implications of conferring rights on animals. 8 that would clearly be impossible to satisfy. Instead, when it comes to duties to help the needy, agents have a fair amount of discretion in determining when and how to discharge them. While breaching such duties clearly implies wrongdoing, it never amounts to a rights violation (O’Neill 1996, chap. 5).6 Since animals are objects of moral concern, they are addressees of the duty not to harm and of the duty to help. Crucially, these duties are baseline, non-relational duties and do not exhaust the realm of moral obligation. There are indeed many so- called special duties, correlative to rights, which we owe to specified others in particular and bear either by virtue of voluntary acts that we have performed – e.g. promises or contracts – or by virtue of the particular relations in which we stand vis- à-vis other people – e.g. our friends, families, colleagues etc. (see Hart 1955). For instance, a food supplier’s duty to deliver fresh produce on a daily basis is not correlative to a universal right, but only owed to those customers with whom he has signed a contract. Similarly, a mother’s duty to take care of, and protect, her children is owed to her children, and only to them. More broadly, family members or groups of friends have duties towards each other that they do not have towards anyone else, precisely by virtue of the special relationships in which they are involved. Notice that these special duties are not a mere specification of the natural duty to help those in need. Instead, they are duties correlative to rights which pertain to specific relations and are structurally similar to the duty not to harm. Like this duty, 6 Cf. the more general distinction between justice and beneficence. 9 special duties are very stringent and owed to particular individuals who have a right to their performance.7 As anticipated, a prominent strand within political philosophy argues that the enforceable duties of justice fellow citizens owe one another are of this weighty, relational, kind. This leads me to expound the second premise of the argument: the associative account of justice. 2. The Associative Account of Justice In contemporary liberal theorizing, the ‘associative’ account of justice is most commonly traced to John Rawls’s work. His principles of justice are said to apply to society understood as a ‘system of cooperation’ for the production and distribution of goods each citizen needs in order to lead a worthwhile life (Rawls 1971). On this associative account, we owe more to our fellow citizens than we owe to distant strangers because citizens are fellow cooperators within societal arrangements. Society provides the background conditions in which individuals can lead flourishing lives, and by working to sustain this background, each citizen is owed a ‘fair return’ for what he/she contributes. Principles of social justice enforced by the state are meant to provide this fair return (Sangiovanni 2007, esp. pp. 26-7). Unlike fellow citizens, distant strangers are not part of a joint collective enterprise, and are therefore owed only basic moral concern. Since we do not contribute to shaping the background conditions in which they live, so the argument goes, we owe them relatively little. (Of course, as I have already mentioned, this assumption is less and less tenable in an increasingly globalized world.) So long as 7 Indeed, we often characterize breaches of special duties in terms of harm. For instance, a mother who (culpably) fails to provide for her daughter’s subsistence may be said to harm her (McGrath 2005). But cf. Donaldson and Kymlicka (2011, Introduction). 10 we refrain from harming them, and help them when they are in need at reasonable costs, our behaviour towards them is beyond moral reproach (cf. Rawls 1999). The conclusions reached by the associative account in the case of international justice may strike some as unpalatable, but we can leave them aside for present purposes. What we need to establish is instead how ‘fair returns’ for social cooperation between members of the same political community are to be determined. To do so, liberals often resort to a ‘universal justifiability’ test. From a liberal point of view, schemes of social cooperation are fair (i.e. just), only to the extent that they are justifiable in the eyes of all those participating in them.8 Although it is difficult precisely to establish what would positively fulfil the requirement of universal justifiability, we can still confidently identify circumstances in which the requirement falls short of being satisfied. For instance, a society that fails to protect some of its citizens’ liberty, political, or economic rights is unlikely to qualify as a fair scheme of cooperation. How could it be justified to everyone participating in it? Agents whose fundamental rights are disregarded have no reason to join the would-be cooperative scheme. From their point of view, the scheme is exploitative and unacceptable. Their contribution to it is not properly acknowledged. That said, there may be multiple societal arrangements meeting the criterion of universal justifiability, especially when it comes to socio-economic justice. Different resource-distribution patterns, for instance, might meet a broad universal 8 Consider, for instance, Rawls’s (1971) original-position thought experiment, or T.M. Scanlon’s (1998) reasonable-rejectability test. For discussion of the original position in the context of animal rights see Rowlands (2002, ch. 3) and Hilden (2007). 11 justifiability test, provided each person has enough to lead a worthwhile life (see Rawls 1999 on ‘reasonably just’ societies, and Valentini 2011, pp. 173-8). For present purposes we need not settle these complex issues. All we need to assume is that a system of cooperation is fair to the extent that it is in principle acceptable in the eyes of all participants and that, even on a parsimonious account of what this requires, the obligations holding between fellow citizens are ‘thicker’ than standard natural duties. In particular, all fellow cooperators are entitled, vis-à-vis one another, to the social conditions necessary to lead decent lives. These will include at least liberty and political rights as well as rights to ‘sufficient’ economic resources.9 If some of the associates lack the objects of these rights, we can immediately infer that they are being treated unjustly. That is, they are not given fair returns for what they contribute.10 But what does the associative account imply for non-human animals? At first sight, one might say, not much. After all, non-human animals have their own habitats and cooperative relationships outside human society, and are therefore not appropriate recipients of associative justice. A moment’s reflection, though, shows that this conclusion would be too quick. At least some non-human animals, namely domesticated ones, are very much part of our lives within society, in some cases even as friends or family members. In light of this, should we think of them as 9 For those who find this last requirement too strong, it could be replaced with a right to meaningful opportunities for subsistence. 10 It might be objected that what counts as a ‘fair return’ should depend on what exactly each has contributed. On this view, the talented, hard-working and skilled deserve more than others. Proponents of the associative account, however, find this line of reasoning flawed. This is because one’s talents and skills, as well as the market rewards they command, are to a large extent products of the background social structure to which everyone contributes. In light of this, the justification for giving the skilled and talented higher returns considerably weakens. See Sangiovanni (2007, pp. 25- 6), cf. Rawls (1971). 12 fellow associates entitled to justice? In what follows, I answer this question specifically in relation to domesticated dogs (cf. Anderson 2004, p. 287). 3. Dogs and Human Beings: A History of Cooperation To substantiate the claim that dogs qualify as fellow cooperators, it is worth briefly sketching the history of their relations with humans (on this, I am indebted to McNeely and Lindquist 2007, pp. 101-4).11 This history is long, constantly evolving, and contested. Dogs are descendants of wolves, wild animals that live and hunt in packs. With the formation of the first human communities, wolves and humans came into contact and started to benefit from this connection. By living in the vicinity of humans, wolves were able to take advantage of food waste and unwanted leftovers as a source of nutrition, while humans could rely on wolves’ superior predatory instincts and benefit from their assistance in hunting (see Groves 1999). Those wolves that happened to be least aggressive and most docile were particularly successful at interacting with humans and gained the greatest selective advantage. They were best adapted to, and most preferred by, the humans they interacted with. Over generations, this led to an evolutionary process in which what used to be wolves became what we now call dogs. With the establishment of societies based mainly on farming and agriculture, humans began to use dogs no longer primarily in hunting but to transport heavy materials, defend their possessions, herd livestock, and so forth. More recently (arguably), humans also started to appreciate dogs simply for their companionship, 11 Like any reconstruction of human-dog relationships, the one I am about to offer is somewhat controversial. For instance, on some accounts, dogs were first adopted as companion animals, and only later used in hunting. Thanks to an anonymous reviewer for pointing this out. 13 as reflected in the epithet of man’s ‘best friends’ (McNeely and Lindquist 2007, pp. 101-4). The history of cooperation between humans and dogs continues today. Dogs herd cattle, help walk the blind, guard our property, participate in rescue operations, work with the military and the police (among other things as sniffer dogs detecting illegal substances), assist in psychological therapies and are used in hospitals to relieve patients' anxiety.12 These appear to be tasks that warrant granting them the status of cooperating members of society. Dogs, it seems, have jobs and contribute to our societies’ functioning and prosperity in often obvious and direct ways. For this reason, shouldn’t they be given at least some of those rights and benefits that human citizens typically enjoy (mutatis mutandis)? Someone might object that talking of cooperation in the case of dogs is misleading.13 Cooperation only exists when different agents voluntarily come together to pursue a common goal. Our relationships to dogs, however, can hardly be said to fit this description. We are responsible for domesticating them, and we now use them in pursuit of our own goals. Dogs, therefore, are not cooperating members of society, instead, they are exploited members of society.14 True, human beings have turned wolves into domesticated dogs, and our objector might be right to suggest that there is something morally problematic about this original act of domestication (but cf. Groves 1999). But even so, dogs as we know them are very different from wolves and can no longer easily survive in the wild. 12 As the Australian Companion Animal Council (2009) states ‘Dogs serve us not only as companions but also as workers [... such as] guide dogs for the blind, hearing and assistance dogs for the disabled, sniffer dogs used by police and customs, and farm dogs used for stock work.’ 13 Thanks to Steve Cooke for raising this objection. The objection is broadly in line with the concerns expressed by so-called ‘abolitionists’. See note 5. 14 A similar objection is briefly raised and answered in Coeckelbergh (2009, p. 77). 14 They have become humanized in so many ways that, arguably, their natural habitat is human society. Abandoning a Chihuahua, a Dalmatian, or even a Great Dane into the woods would in all likelihood result in their death. From the fact that, due to our original act of domestication, dogs cannot but participate in our societies, does it also follow that their participation is intrinsically exploitative? It does not. To see this, consider an analogous case drawn from the human world. Imagine a capitalist society so far removed from other communities that citizens have no realistic opportunity to leave it, and in which free market forces determine the division of benefits and burdens of social cooperation. Lacking the necessary means to lead a decent life, workers in our society are forced to accept any job offers, even extremely unfair ones (see Cohen 1983). A society so organized would not be regarded as just from the perspective of any plausible liberal-deontological approach. Some of its members are obviously exploited, their cooperation is not appropriately rewarded. However, if the division of benefits and burdens were regulated by principles of justice, no exploitation would occur, and our society would appropriately qualify as a cooperative scheme (even though citizens would have no realistic opportunity to leave it). The case of dogs is not dissimilar from that of workers in the society I have just described. Dogs de facto cooperate and, like our workers, have little alternative to living in society. This, however, does not entail that their cooperation must be inevitably unfair or exploitative. In fact, the whole point of applying the associative account to dogs is to make sure that their cooperation is acknowledged and appropriately rewarded, so that our relations to them are not of an exploitative nature. That said, another objector might complain that only a relatively small number of dogs plausibly count as cooperating members of society. Most domesticated dogs 15 (literally millions of them) are mere companion animals, without proper ‘jobs’. Does this undermine the claim that dogs are fellow co-operators? I doubt so, for two reasons. First, from the liberal-deontological perspective assumed here, numbers cannot be given such moral weight. The (alleged) fact that only some dogs are cooperating members of society does not invalidate the claim that justice is owed to them. Second, and most importantly, the claim that only very few dogs cooperate in the required manner may be challenged by noting that having an officially recognized job is not a necessary condition to count as a productive member of society. Stay-at-home mothers and housewives are undoubtedly crucial for society’s prosperity and continued existence, yet they do not hold what were traditionally regarded as conventional ‘jobs’. As most now acknowledge, however, this should not prevent such women from being appropriately rewarded for their efforts in raising children and providing a healthy family environment. Doing otherwise would amount to unfair discrimination, if not full-blown exploitation (see Rawls 2000, pp. 162-8). Similarly, even dogs that are not classified as ‘working dogs’ provide a significant contribution to society’s prosperity and good functioning. As pointed out by the Australian Companion Animal Council (ACAC), companion ‘[d]ogs can help to lower stress, alleviate loneliness, improve health, and encourage people to exercise more and enjoy the outdoors. They improve feelings of safety – both in the home and in public places. Dogs act to break down the barriers in society, and help people to meet and make friends.’ Some of the functions described in the previous passage are often performed by professionals, such as personal trainers and psychotherapists. If they count as cooperating members of society, why should ‘companion’ dogs not also enjoy a similar status? After all, low stress, physical exercise, good social relationships, and 16 feelings of safety are all important components of a well-functioning social system.15 To the extent that dogs contribute to providing these goods, they should be appropriately rewarded. Of course, there may still be ‘difficult’ or ‘uncooperative’ dogs (cf. McNeely and Lindquist 2007). However, just as the existence of some humans who cannot cooperate in society (e.g. psychopaths or the severely disabled) does not disqualify human beings in general from the status of fellow cooperators, similarly the existence of problematic dogs should not make us blind to the fact that many dogs do in fact cooperate with us. In light of this, what enforceable protections and rights are commensurate with dogs’ status as cooperating members of society? 4. Canine Justice: Giving Dogs their Due As one writer puts it, ‘the domestic dog exists precariously in the no-man’s-land between the human and non-human worlds. It is an interstitial creature, neither person nor beast, forever oscillating uncomfortably between the roles of high-status animal and low-status person’ (Serpell 1995, p. 254 quoted in McNeely and Lindquist 2007, p. 104). Although there exist laws against cruelty towards non- human animals, including dogs, these are far from being fully reflective of their status as cooperating members of society. Indeed, as animal-rights advocates have long-been complaining, existing laws fall short of acknowledging animals’ status as right-holders in the first place. In what follows, I offer an illustrative account of what giving dogs their due as (i) right-holders and (ii) fellow cooperators demands.16 In doing so, my aim is to showcase the potential implications of the associative 15 For a survey of the benefits and harms of companion animals in general see Podberscek (2006). 16 This distinction is drawn for analytical purposes, and there is bound to be overlap between these two sets of requirements in the case of dogs (and cooperating animals more generally). 17 approach. I am aware, however, that a comprehensive treatment of the issue would require a much fuller discussion than I can provide here. 4.1 Dogs as Right-Holders in General17 First, dogs ought not to be regarded as mere property. By considering dogs as property, we implicitly treat them as objects, which may be used by their owners (see Hilden 2007, and Francione 2004). To be sure, as Cochrane (2009a), for instance, has pointed out, to the extent that the ownership relation is suitably qualified (i.e. to the extent that strict limits are imposed on how the object of ownership can be used), treating animals (including dogs) as property is morally acceptable. That is, so long as dog owners bear certain obligations towards their pets, the ownership relation cannot be so detrimental to them. There is no conceptual inconsistency in jointly claiming that (i) A owns X and that (ii) A has a duty of care towards X. This seems to be the view underlying the 2006 UK Animal Welfare Act (section 9), which explicitly establishes the responsibility of the pet owner to ‘take such steps as are reasonable in all the circumstances to ensure that the needs of an animal for which he is responsible are met to the extent required by good practice’. Although it is correct to point out that property relations do not per se preclude the possibility of treating animals as objects of moral concern, they do run counter their status as right-holders. Considering dogs a piece of property partly obscures their moral status, making the presumption that they are ‘things’ to be used more plausible (see Francione 2004). If dogs are mere property, then it is rhetorically easier to justify giving them to shelters, abandoning, and neglecting them. 17 The changes I propose under this heading are relatively familiar ones in the literature. In particular, my account here is indebted to Hilden (2007, pp. 19ff.). 18 This is reflected in the comparatively light forms of punishment risked by those who ignore their duties towards non-human animals in general, and dogs in particular. Even in the much-welcomed 2006 Act mentioned earlier, fines for neglect and cruelty towards non-human animals go up to just 20000 GBP, and periods spent in jail up to 51 weeks (BBC 2011). These sanctions appear strict if we conceptualize them as responses to failures to abide by duties towards one’s property. However, they appear much less demanding when we consider them as responses to rights violations. We would be shocked if the maximal penalty for killing or abusing a child amounted to just 51 weeks in jail. If we take seriously the idea that dogs qualify as holders of rights, current legislation seems excessively tolerant. To fully recognize dogs’ status as right-holders in our legal system, we must stop framing our relationship with them in terms of property (Hilden 2007, pp. 19-20). Needless to say, if dogs publically acquired the aforementioned status, many widespread practices currently considered acceptable, such as using them for meat consumption (in those countries where this is an accepted habit) or in scientific experimentation, would be outlawed. It is clear that such practices violate the ‘basic’ right not to be harmed, which, according to premise (i), non-human animals enjoy. As I have argued, however, dogs are also fellow associates and therefore recipients of justice (according to premise (ii)). What, then, do we owe them in this respect? 4.2 Dogs as Fellow Cooperators Entitled to Justice From a liberal perspective, associative justice requires respect for persons’ liberty, political, and economic rights. However, conferring on dogs the very same rights we typically ascribe to humans – such as the right to vote, to freedom of speech, to 19 education, to hold public office etc. – would be bizarre to say the least. In order to understand what giving dogs their due qua fellow cooperators demands, we cannot simply extend principles of justice for humans to the ‘canine’ realm. Instead, we need to consider how those principles are arrived at in the human case, and then ask whether a similar method could be adopted in the case of dogs.18 As we have seen, in contemporary liberal theory, principles of justice typically emerge from hypothetical consent tests. We look at society and ask: What principles could all citizens accept as giving them fair returns for their contributions? In other words, we ask: What principles could be justified in the eyes of all? As Jeremy Waldron says ‘...[liberal] commitments generate a requirement that all aspects of the social [order] should either be made acceptable or be capable of being made acceptable to every last individual’ (1987, p. 128). Although there may be reasonable disagreement about the exact output of some such thought experiment (e.g. on whether a society can only be just if it adopts Rawls’s difference principle), certain core liberty, political, and economic rights must be guaranteed for a political system to count as just. Indeterminacy, in this context, does not make us completely unable to say what justice demands. A sceptic might suggest that indeterminacy does become fatal when we move to hypothetical-consent tests in the case of dogs. Indeed, while we can imagine what is acceptable or not from the viewpoint of human beings in general, resorting to hypothetical consent in the case of non-human animals might seem pointless, ‘suggesting no clear scenario that would assist our thinking’ (Nussbaum 2004, p. 301). It is sometimes hard to put oneself into other people’s shoes, let alone into the 18 This is a ‘logic of extension’ mirroring the one often used, mutatis mutandis, in the debate on global justice. 20 ‘paws’ of creatures belonging to another species (see Hilden 2007, part II and Rowlands 2002 for critiques of this argument). This is a warranted worry, but not one that undermines the project of making sense of what it is to give dogs their due qua cooperating members of society. Even though it is hard, perhaps impossible, to gain profound knowledge of a dog’s psychology, of its deepest beliefs and desires, we can still form a reasonably clear picture of what conditions must be satisfied for dogs to lead decent lives. Dogs cannot ‘directly’ speak their minds, but their behaviour can be interpreted, and their interests reasonably inferred from it, and from their biological makeup more generally (Donaldson and Kymlicka 2011, pp. 108-12, Rowlands 2002, pp. 66-8, Hilden 2007 pp. 17ff.). This is not to deny that, since dogs do not belong to our species, their needs and interests are bound to remain somewhat opaque to us. But opacity is not the same as inscrutability, and so long as we can at least partly understand dogs’ interests, we ought to provide them with the necessary conditions to lead decent lives on grounds of justice. That is, society should be organized in such a way that not only human beings, but also dogs, could consent to it. For example, we can reasonably assume that, like children, dogs have an interest in generating social bonds of affection as members of a pack or family. They have an interest in bodily integrity and freedom from violence, as well as in having access to food, shelter, and medical care. They have an interest in enjoying sexual pleasure, and in living in an environment where they can move reasonably freely (cf. Nussbaum 2004, pp. 313-17). These are all interests we can plausibly assume dogs have, and which would have to be secured for dogs to consent to living in society, if they could. 21 Of course, making our societies more sensitive to dogs’ needs is a hard task to accomplish. In what follows, I only offer some candidate examples of the sorts of reforms that ought arguably to be undertaken to realize ‘canine justice’.19 To begin with, since there is a strong correlation between the satisfaction of dogs’ interests (in bodily integrity, nutrition, health care etc.) and the families of which dogs become members, those who wish to adopt a puppy (or an adult dog) should be adequately checked. In the same way in which young children are only entrusted to families deemed capable of caring for them, so too dogs should only be entrusted to humans deemed capable of providing them with a healthy and safe environment (cf. Hilden, pp. 21-2). However, to make sure that ‘human guardians’ adequately take account of their dogs’ interests, broader social changes are required, setting limits to what is deemed acceptable animal treatment. For instance, seemingly harmless habits, such as having one’s dog compete in a beauty contest, can be revealing of, or conducive to, a problematic ‘commodifying attitude’ towards the animal, with detrimental effects on its health and bodily integrity. For example, there is little doubt that ‘plastic surgery’ performed on dogs, in the from of ear-cropping and tail-docking, is very often motivated by human perceptions of beauty which have little to do with dogs’ own well-being. In recognition of this, the 2006 UK Animal Welfare Act (sections 5 and 6) has precisely banned such practices.20 Similarly, the search for purity of breed, to ensure that dogs meet specified aesthetic standards, is often harmful to the health and well- 19 For a more general and comprehensive discussion of what treating domesticated animals as citizens would require, see Donaldson and Kymlicka (2011, sec. 5.4). 20 Specifically, the Act bans cutting dogs’ tails for cosmetic reasons, but not in the case of ‘working dogs’. 22 being of the animal itself.21 Moreover, a concern with dogs’ bodily integrity and their interest in experiencing sexual pleasure would plausibly lead to banning permanent sterilization – although I venture the hypothesis that non-invasive contraceptive methods might be used for birth-control purposes (but cf. Fusfeld 2007 for a different view). A ‘caninely just’ society should not only ban social practices that are explicitly harmful to dogs, but also make sure that the environment in which they live is sufficiently hospitable to them. This requires attention at the level of urban planning, making sure that dogs have access to designated areas protected from traffic in which to move freely, socialize, and play (ACAC 2009). Finally, dogs’ interests should be given official representation within society, so as to secure their protection over time. Although, as mentioned earlier, dogs cannot be given the right to vote or to hold public office, appointed representatives may act on their behalf at both the political and the legal level, making sure that their status as objects of moral concern is adequately defended (see Garner 2005, Nussbaum 2004 and Hilden 2007). More could be said, but if at least some of these arguments are correct, we can conclude that giving dogs their due would involve some relevant changes in the way society is organized. This, in itself, should come as no surprise. Indeed, perhaps even more significant changes appear to be needed when we think about what a ‘humanly’ just society requires. There is a big gap between most societies in the world and the ideal of a just social system liberal political philosophers advocate. 21 On this, see the work of The Royal Society for the Prevention of Cruelty to Animals (RSPCA), http://www.rspca.org.uk/allaboutanimals/pets/dogs/health/pedigreedogs/actions and http://www.rspca.org.uk/media/news/story/-/article/PuppiesBredForLooksAreBornToSuffer_Dec11 (last accessed 3/10/2012). 23 In connection with this, it is crucial to emphasize that the ‘caninely just’ society envisaged in this paper is an example of an ‘ideal-theoretic’ scenario (Rawls 1971, pp. 8-9, and Simmons 2010 for discussion). In the same way in which change towards a more just society for humans takes time, change towards a more just society for dogs (let alone other animals) is also a matter of long and gradual reform.22 The policies discussed here could not be implemented straightaway, in our non-ideal world. For instance, in an ideal world, no dog would have to be euthanized in shelters because of overpopulation. However, it may well be that, due to the constraints of the non-ideal world in which we live (where too many dogs are abandoned, and where too little is invested in shelters) euthanizing some dogs might be a sadly inevitable solution, at least in the short term. Or else, in an ideal world, dogs would not need to be permanently sterilized. However, as a transitional, non- ideal measure for the world in which we live, permanent sterilization might be the best we can do to safeguard dogs’ interests. Unless alternative contraception methods are introduced as a way of ensuring birth control, the number of undesired and neglected animals will only increase. Even though a society that gives dogs their due is still a somewhat distant ideal, and many non-ideal measures would have to be undertaken to reach it, we might want to ask whether such a ‘caninely just’ society would inevitably result in greater injustice for humans. Although I am in no position to provide a well-grounded, conclusive answer to this question, I think there are some reasons for optimism. Many of the reforms that would make society more just for humans would also make it more just for dogs, and vice versa. 22 Thanks to an anonymous reviewer for asking me to elaborate on this point, and suggesting some of the examples that follow. 24 As I have mentioned earlier, sensitivity to dogs’ interests would have an impact on urban planning, increasing the number of parks and recreational areas sheltered from traffic available to both animals and humans. Similarly, a more just society for humans, in which everyone has adequate opportunities to lead a decent life, would in all likelihood be ‘friendlier’ towards dogs too. In such a society, the levels of violence, poverty and frustration among the general population would be lower than in the status quo, thereby reducing the risk for dogs themselves to be mistreated, subjected to violence, or simply neglected. In short, it appears reasonable to conjecture that human and canine justice would work together in synergy. After all, on the associative account, a society that is just to dogs and humans is, by definition, mutually beneficial for both.23 5. Objection In this section, I consider an important challenge to the account presented here. The challenge points to a perceived tension between the two main premises of the argument: (i) animals are objects of moral concern and (ii) justice applies to fellow cooperators (provided they are objects of moral concern). The worry is that, once we grant animals a set of rights on deontological grounds – in accordance with premise (i) – we are immediately led to embrace an ‘animal liberation’ perspective according to which cooperative relations between humans and animals violate those very rights. If the relationship of cooperation ends – as recommended by premise (i) under the present demanding interpretation – the associative account implies that 23 Notice that this conjecture is once again stated at the level of ‘ideal theory’. There are of course trade-offs to make in the status quo, where human and animal interests have to be balanced against each other. This is an extremely complex question I cannot address here. For discussion see Cochrane (2007) and Aaltola (2005). 25 animals ‘liberated’ from human dominion fall outside the scope of justice. This, in turn, amounts to the counter-intuitive view that animals are owed justice only when they are being exploited.24 This objection is based on a radical reading of the implications of premise (i) and an equivocation between exploitation and cooperation.25 While it is certainly true that conferring rights on animals prohibits us from treating them merely as means, it does not prevent us from treating them as means, provided we also treat them as ends in themselves.26 This is a familiar thought from the Kantian-deontological tradition and is clearly in line with what we consider morally acceptable treatment of other humans. Consider the relationship between a professor, Martin, and his research assistant, Bob. It is clear that, by giving Bob a set of calculations to perform, Martin uses him as a means to something: the results he needs for the paper he is writing. However, provided Martin treats Bob with respect, does not overburden, threaten, or grossly underpay him, we have reason to regard their relationship as perfectly morally acceptable. Martin treats Bob as a means but, crucially, also as an end in himself. By contrast, the relationship between a (non-benevolent) master and his slaves, or between the owner of an aggressive multinational corporation and his sweatshop workers, is morally objectionable precisely because the former use the latter as means, without also treating them as ends in themselves. What is going on in these 24 I am grateful to an anonymous reviewer for raising this objection. 25 For an argument that animal rights need not imply animal ‘liberation’, see Cochrane (2009b). 26 This is indeed in line with the original formulation of Kant’s categorical imperative. Although, interestingly, when adopted to characterize the animal-rights position, this formulation at times loses the qualification ‘merely’, thus turning into an absolute ban on using others (including animals) as means. See e.g. Sunstein (2004, p. 5). 26 cases is exploitation, rather than cooperation (recall my earlier discussion of a capitalist society). Similarly, conferring rights on dogs does not automatically prevent us from engaging in cooperation with them, provided such cooperation is fair and non- exploitative, i.e. provided we treat them not merely as means, but also as ends in themselves. Much of this paper has been precisely concerned with sketching what this might require. Moreover, the claim that ‘absent relationships of cooperation, animals are not owed justice’ is not at all counter-intuitive if we keep in mind that justice does not exhaust the space of moral obligation. Wild animals, with which we entertain no cooperative relations, still maintain a right not to be harmed, and a right to assistance. Humans should therefore refrain from killing or injuring wild animals, and from destroying their natural habitats.27 Similarly, humans are under a duty to assist wild animals, when they can help them at reasonable costs.28 What wild animals do not have a right to, on the associative account, is a fair share in their cooperative endeavours with us, and this is simply because they are not fellow cooperators. While Swedes certainly have duties not to harm Australians, and to help them if they are in need, we typically do not think that Australians are entitled to Swedish health care or education. Why is that? The answer given by the associative account is: ‘because they are not part of the same society-wide scheme 27 This should be qualified by adding ‘unless they are posing an immediate threat to their own lives and well being’. 28 This does not imply that we are under a duty constantly to search for wild animals in need (as we are not under any such duty in the case of human beings). Moreover, this does not mean that we should try to save preys from their predators. This sort of interference could not count as a way of discharging the duty to help. By helping the prey, we would end up harming the predator. Cf. Cochrane (2009b, pp. 675-6), Callicott (1992). Thanks to an anonymous reviewer for pushing me to clarify this point. 27 of cooperation as the Swedes, and are thus not entitled to a fair share of its benefits’ (Sangiovanni 2007, pp. 31-2). If one thinks that this stance is reasonable enough in the case of human beings (again, suspending complications due to globalization), there is no apparent reason why one should not think of it as reasonable in that of animals. In sum, a belief in animal rights – as per premise (i) – does not make fair cooperation between humans and animals impermissible, and allows us to show appropriate moral concern towards animals even when they exist outside our cooperative schemes. Conclusion I have explored the question of what, if anything, might be owed to non-human animals from the perspective of a popular, associative account of justice. Focusing on dogs in particular, I have suggested that justice requires their interests to be taken into account in the design of laws and policies within our political communities. Beyond this specific conclusion about the moral status of domesticated dogs, the paper has broader implications for both the debate on animal rights, and that on the associative account of justice. Regarding the former, by showing how an associative perspective can be extended to the case of non-human animals, the paper represents one of the few attempts, within justice theory, to consider what we might owe non- human animals for reasons which go beyond their intrinsic properties (for other such attempts, see Coeckelbergh 2009, and most importantly, Donaldson and Kymlicka 2011). Although I have not explicitly addressed the question of what might be owed to non-human animals other than dogs, the paper still offers a general method (hypothetical consent test) that could be applied to the case of other ‘cooperating’ animals as well. Of course, doing so will present a number of challenges and will 28 demand a fairly radical shift in people’s attitudes towards animals. There are in fact pervasive forms of human-animal ‘would-be cooperation’, such as meat farming, which fail to acknowledge the very status of animals as right-holders. That said, while the systematic and unnecessary killing of animals for food might be impermissible on the view presented here, humane cattle farming for dairy production (if feasible) might constitute a morally acceptable form of cooperation: cattle provide milk in return for shelter, fresh pastures, and care (see Hilden 2007, p. 24, but cf. Donaldson and Kymlicka 2011, pp. 138-9). Although I do not have the space to explore these arguments in detail, the associative approach presented here points in the direction of a balanced and intuitively appealing position, which despite emphasizing animal rights, avoids the extremism of abolitionist animal ethics. The paper also has implications for the debate on the associative account of justice. As I have mentioned at the outset, this account is often thought to entail a rather conservative stance on the scope of our duties of justice, especially with respect to their extension to the global realm. But what about the extension of justice to other species, rather than to citizens of other states? Interestingly, proponents of the associative account have by and large neglected this question. As this article has shown, however, such a cross-species extension is in principle possible. In fact, if we consider how many species are integrated, and used, within our social systems (arguably cats, dogs, horses, cattle, sheep, and so forth), we can immediately see that the implications of the associative account are potentially very radical indeed (cf. Donaldson and Kymlicka 2011). Whether proponents of this account would accept them is doubtful. However, what this article suggests is that these are implications they must reckon with. 29 REFERENCES Aaltola, E. (2005) ‘Animal Ethics and Interest Conflicts’, Ethics and the Environment, 10 (1), 19-48. ACAC, Australian Companion Animal Council, ‘Dogs in Society Position Paper’, http://www.acac.org.au/pdf/dogs_in_society.pdf (last accessed on 19/10/2011). Anderson, E. (2004) ‘Animal Rights and the Values of Non-human life’, in M. Nussbaum and C. Sunstein (eds), Animal Rights: Current Debates and New Directions. Oxford: Oxford University Press, pp. 277-98. Animal Protection Party (1993) ‘Manifesto’, available online at http://www.tierschutzpartei.de/GSPtierschutz01.htm (last accessed 15/4/2011). Animals Count (2006) ‘Manifesto’, available online at http://www.animalscount.org/manifesto/aims/ (last accessed 15/4/2011). Animal Welfare Act (UK, 2006) available online at http://www.legislation.gov.uk/ukpga/2006/45/contents (last accessed 30/9/2012). BBC (2011) ‘Animal Welfare Law in the UK’, online http://www.bbc.co.uk/ethics/animals/overview/latest.shtml (last accessed 15/9/ 2011). Callicott, J. Baird (1992) ‘Animal Liberation and Environmental Ethics: Back Together Again’, in E.C. Hargrove (ed.), The Animal Rights/Environmental Ethics Debate. New York: State University of New York Press, pp. 249-61. Cochrane, A. (2007) ‘Animal Rights and Animal Experiments: An Interest-Based Approach’, Res Publica, 13 (3), 293-318. Cochrane, A. (2009a) ‘Ownership and Justice for Animals’, Utilitas, 21 (4), 424-42. Cochrane, A. (2009b) ‘Do Animals Have an Interest in Liberty?’, Political Studies, 57 (3), 660-79. 30 Coeckelbergh, M. (2009) ‘Distributive Justice and Co-Operation in a World of Humans and Non-Humans: A Contractarian Argument for Drawing Non-Humans into the Sphere of Justice’, Res Publica, 15 (1), 67-84. Cohen, G.A. (1983) ‘The Structure of Proletarian Unfreedom’, Philosophy and Public Affairs, 12 (1), 3-33. DeGrazia, D. (1991) ‘The Moral Status of Animals and Their Use in Research: A Philosophical Review’, Kennedy Institute of Ethics Journal, 1 (1), 40-70. DeGrazia, D. (1996) Taking Animals Seriously: Mental Life and Moral Status. Cambridge: Cambridge University Press. DeGrazia, D. (2002) Animal Rights: A Very Short Introduction. Oxford: Oxford University Press. Donaldson S., and Kymlicka, W. (2011) Zoopolis: A Political Theory of Animal Rights. Oxford: Oxford University Press. Donovan, J., and Adams, C.J. (2007) The Feminist Care Tradition in Animal Ethics. New York: Columbia University Press. Francione, G. (2000) Introduction to Animals Rights: Your Child or the Dog? Philadelphia: Temple University Press. Francione, G. (2004) ‘Animals – Property or Persons?’, in M. Nussbaum and C. Sunstein (eds) Animal Rights: Current Debates and New Directions. Oxford: Oxford University Press. Francione, G., and Garner, R. (2010) The Animal Rights Debate: Abolition or Regulation? New York: Columbia University Press. Fusfeld, L. (2007) ‘Sterilization in an Animal Rights Paradigm’, Journal of Animal Law and Ethics, 2, 255-62. Garner, R. (2005) The Political Theory of Animal Rights. Manchester: Manchester University Press. 31 Groves, C. (1999) ‘The Advantages and Disadvantages of Being Domesticated’, Perspectives in Human Biology, 4 (1), 1-12. Hadley, J. (2007) ‘Critique of Callicott’s Biosocial Moral Theory’, Ethics & the Environment, 12 (1), 67-78. Hart, H.L.A. (1955) ‘Are There Any Natural Rights?’, Philosophical Review, 64 (2), 175-91. Hilden, J. (2007) ‘A Contractarian View of Animal Rights: Insuring Against the Possibility of Being a Non-human Animal’, Animal Law Review, 14 (1), 5-28. McGrath, S. (2005) ‘Causation by Omission: A Dilemma’, Philosophical Studies, 123 (1-2), 125-48. McNeely, C.A. and Lindquist S.A. (2007) ‘Dangerous Dogs Laws: Failing to Give Man’s Best Friend a Fair Shake at Justice’, Journal of Animal Law, 3, 99-158. Midgley, M. (1983) Animals and Why They Matter. Athens: University of Georgia Press. Nussbaum, M. (2004) ‘Beyond “Compassion and Humanity”: Justice for Nonhuman Animals’, in M. Nussbaum and C. Sunstein (eds) Animal Rights: Current Debates and New Directions. Oxford: Oxford University Press, pp. 299-320. O’Neill, O. (1996) Towards Justice and Virtue: A Constructive Account of Practical Reasoning. Cambridge: Cambridge University Press. Palmer, C. (2010) Animal Ethics in Context. New York: Columbia University Press. Party for the Animals (2002) ‘Aims’, available online at http://www.partyfortheanimals.info/content/view/300 (last accessed 15/9/2011). Podberscek, A.L. (2006) ‘Positive and Negative Aspects of Our Relationship with Companion Animals’, Veterinary Research Communications, 30 (1), 21-7. 32 Rachels, J. (1976/1989), ‘Why Animals Have a Right to Liberty’, in T. Regan and P. Singer (eds), Animal Rights and Human Obligations. Englewood Cliffs: Prentice Hall, pp. 122-31. Rawls, J. (1971) A Theory of Justice, Cambridge, MA: Harvard University Press. Rawls, J. (1999) The Law of Peoples, Cambridge, MA: Harvard University Press. Rawls, J. (2000) Justice as Fairness: A Restatement, Cambridge, MA: Belknap Press. Regan, T. (1983) The Case for Animal Rights. Berkeley, CA: University of California Press. Rollin, B. (1981) Animal Rights and Human Morality. Buffalo: Prometheus Books. Rowlands, M. (2002) Animals Like Us. London: Verso. Sangiovanni, A. (2007) ‘Global Justice, Reciprocity, and the State’, Philosophy and Public Affairs, 35 (1), 3-39. Scanlon, T.M., (1998) What We Owe to Each Other. Cambridge MA: Belknap/Harvard. Serpell, J. (1995) The Domestic Dog: Its Evolution, Behaviour and Interactions with People. Cambridge: Cambridge University Press. Simmons, A. J. (2010) ‘Ideal and Nonideal Theory’, Philosophy and Public Affairs, 38 (1), 5-36. Singer, P. (1990) Animal Liberation. New York: Avon, 2nd ed. Sunstein, C. R., ‘Introduction: What Are Animal Rights?’, in M. Nussbaum and C. Sunstein (eds), Animal Rights: Current Debates and New Directions. Oxford: Oxford University Press, pp. 3-18. Valentini, L. (2011) Justice in a Globalized World: A Normative Framework. Oxford: Oxford University Press. 33 Waldron, J. (1987) ‘Theoretical Foundations of Liberalism’, The Philosophical Quarterly, 37 (147), 127-50. Valentini_Canine justice_2016_cover Valentini_Canine justice_2016_author