Academia.eduAcademia.edu
Animals Do Have an Interest in Liberty Published in Journal of Animal Ethics, 6 (1), 2016, 20-43. According to Alasdair Cochrane, liberty can have value for most animals only because it allows them to obtain other desirable things, such as well-being. With this he concludes that humans can continue to use other animals as long as they treat them well. In this article, I reject this conclusion by arguing against the positive conception of liberty in favor of its negative or republican conception. I suggest that it is sufficient for a being to be capable of agency in order to have an interest in liberty, which deserves to be protected by the right to be free. Fundamental Rights; ethics, animals; interests; liberty; autonomy In 1976, James Rachels sets out to answer to the following question: do animals have a right to liberty? (Rachels, 1976) Appealing to the Aristotelian principle of equality, he recalls that the decision to deny certain individuals a right that we would grant others must be adequately justified. Now, no morally worthy reasoning, says Rachels, allows us to grant the right to liberty to all human beings, but only to certain human beings. Moreover, for Rachels, certain nonhuman animals should also benefit from the right to liberty. Many important legal documents, including international declarations and conventions, recognize the fundamental right to liberty. According to Rachels’ definition of liberty, being free means for a rational agent to be capable to act as he pleases without interference. He also mentions that it is most usual or intelligible to say of a wild lion that he is free or to say of a oncecaged bird that he was freed. According to Rachels, the colloquial sense of the term does not limit liberty to rational agents, and thus allows for it to be applied to all animals capable of desiring to act in one way rather than another. He concludes that to be free means for all agents to be able to act without running into external constraints. Rachels then tries to understand the reasons behind granting a right to liberty to human beings using a conception of liberty divested of the superfluous requirement of rationality. He first suggests that liberty represents a ‘good in itself’ that we grant to human beings for the sole reason that none should be deprived of a good from which they could benefit. Rachel argues that if this is indeed the case, and liberty does possess an intrinsic value, nonhuman animals should also benefit from the right to be free, since liberty can benefit all those who are able to desire to act a particular way. For all beings capable of intentional actions, liberty constitutes a ‘good’ and all agents thus have an interest in liberty. Rachels, who does not suggest any specific definition of ‘intrinsic liberty’, still appears to think that few philosophers consider liberty to be an intrinsic good for human beings, and generally believe that the value of liberty is instrumental—or, in other words, derived from more fundamental goods. However, even in this case, adds Rachels, nonhuman animals can have an interest in liberty, since the absence of liberty is detrimental to some of their other interests. To further 1 support this statement, Rachels describes how captivity (or restriction to the freedom of movement) often harms the well-being of nonhuman animals, and is thus incompatible with their fundamental interests. By way of example, he asks us to consider animals who were captured in their natural environment and relocated to zoos, where they live a life of boredom and frustration. He also mentions a case where the level of aggressiveness in a colony of baboons was due to their state of captivity; these animals are usually peaceful when born and raised in their natural environment. Rachels nevertheless believes that it would not be necessary for animals such as laying hens to be completely free in order to have their interests fulfilled. It might be the case that the interest in liberty for certain animals, as for the laying hens, be not only instrumental, but also weak. Lions, for their part, should not in any case be kept captive; they always have an interest in being left free in nature. The reason for this, explains Rachels, is that lions suffer much more when confined than do laying hens. To his mind, the interest in liberty involves many different levels, depending on each individual’s specific need to live and thrive in a way that corresponds to his or her natural tendencies, and without interference. While insects might not have any interest in liberty (if they are incapable of intentional action, they could be indifferent to what happen to them), and lions may have an interest in being fully free, hens could simply have an interest to be kept out in a farmyard rather than in small cages; hens’ welfare would then only require ‘relative’ liberty. Rachels’ analysis leaves open the question of direct interest in liberty because, as he is mainly interested in understanding the instrumental interest in liberty, he barely mentions the possibility of liberty having an intrinsic value for humans. According to him, interest in liberty is dependent on other basic interests, which can be impeded by a lack of liberty. Cochrane on the interest in liberty of nonhuman animals For his part, Alasdair Cochrane suggests that human beings said to be ‘normal’ can have an intrinsic interest in liberty because they are autonomous agents—that is, they are capable of ‘framing, revising and pursuing their own conception of the good’. In his article ‘Do Animals Have an Interest in Liberty?’ (Cochrane, 2009), Cochrane wishes to demonstrate that the only kind of interest in liberty attributable to the majority of sentient nonhuman animals is instrumental, and would depend on whether or not the liberty promotes, in each case, the fulfillment of more basic interests. To support this thesis, Cochrane must first argue that human interest in liberty is not purely instrumental, and that the importance of liberty for one’s well-being is not reducible to the pleasures and/or pains which can be achieved or avoided through said liberty. To better illustrate this idea, the author uses the case of Truman Burbank, the main character of the Truman Show. While Truman’s life can be considered pleasant – he appears to enjoy the company of good friends and a loving wife, and to lead a globally comfortable daily life – Truman is not free because he does not know that his life is being filmed for a television show, nor that his friends and wife are actors. Although Truman is happy, it seems clear to Cochrane that Truman’s interests are ill- 2 served, specifically because he is not really free. Cochrane concludes that liberty can be a good which, for human adults, does not solely depend on the hedonistic well-being which it allows. Cochrane then considers the hypothesis that the interest in liberty rests on the pursuit of the satisfaction of preferences and desires, and on the success or failure of this pursuit. It certainly is in the interest of human beings to pursue the goals they set for themselves, and this venture is easier when we are free. However, says Cochrane, we can imagine situations in which an individual’s preferences are satisfied while this person is manifestly not free. Take, for instance, the case of a slave who would have assimilated the belief that his life is less worthy than that of his owner, and who would desire, for this specific reason, to continue serving him. This slave can satisfy his desire to serve his owner while not being free. The individual appears to be in an appalling situation despite the fact that his preferences are met. According to Cochrane, the case of adaptive preferences seems to show that the human interest in liberty is more fundamental than their interest to satisfy their preferences. After showing that the human interest in liberty is not always instrumental and that it can be intrinsic, Cochrane wants to establish a link between liberty and the status of autonomous agent. To do this, he turns to Gerald Dworkin and defines autonomy as the capacity of an individual to reflect upon her desires and to modify them according to her values and her own conception of the good.1 Cochrane explains that the problem with Truman, in the example of the Truman Show, was that the protagonist was not capable of determining for himself the ends and goals he wanted to achieve. His choices and ambitions were limited and largely impeded by other agents the people producing the television show in which he was unknowingly participating. The problem brought about by this lack of liberty, insists Cochrane, was not that it was keeping Truman from being happy or from satisfying his desires, but that it was preventing him from framing, revising and pursuing the goals he would have set for himself. The same explanation could hold in the case of the slave desiring to serve his master out of indoctrination. From this analysis, Cochrane concludes that liberty has an intrinsic importance for those capable of autonomy. Should we take Cochrane’s reasoning to imply that all humans and only humans have an interest in liberty? Cochrane admits that not all humans are autonomous and that some of them do not have an intrinsic interest in liberty. He recognizes that those people who are highly dependent on others, such as children or mentally disabled people, have very little control over their own lives and that it is often in their best interest when those who are autonomous act paternalistically towards them. According to Cochrane, a presumption against paternalistic attitudes (based on the recognition of an intrinsic interest in liberty) plays a role when it comes to autonomous people, but not when it comes to those who are not autonomous. Of course, he says, this presumption is not absolute, and is dismissed when a person’s use of her freedom impedes the freedom of another. With regard to nonhuman animals, Cochrane suggests that most of them are not autonomous, such that they cannot have an intrinsic interest in liberty. For one thing, animals who are not subjectively conscious (which, he states, includes at least all of the invertebrates) are clearly unable to determine what is right or good for them, to revise their judgments or to aim at obtaining or bringing about what they believe to be good. The case becomes more contentious 3 with vertebrate animals. These sentient beings seem to have desires which they seek to satisfy. However, this does not qualify them as autonomous since, for that, they would also have to be able to reflect on those desires and be able to rectify them according to what they judge to be right. Cochrane refers specifically to Peter Carruthers’ terms to suggest that, in order to be autonomous, an individual must be able to have second-order thoughts – to have thoughts about other thoughts. According to him, we do not have significant proof that members of any particular species – other than those belonging to humanity – possess the sophisticated cognitive capacities required in order to perform such an action. In fact, even if we had sufficient reason to allow the possibility that the great apes and some cetaceans are autonomous,2 these animals represent a minority of all sentient beings. Even if a certain number of nonhuman animals are capable of making choices, the vast majority cannot have second-order thoughts and, thus, cannot determine, reconsider or pursue their own conception of the good, or their own goals; they are not autonomous and, therefore, have no intrinsic interest in liberty. Cochrane turns to a thought experiment to better illustrate his reasoning. He suggests we compare human individuals who were forced to train to become excellent gymnasts with horses who were trained to compete in equestrian games. According to him, if we put aside the pain involved in the process, then the harm done to the horses would be less important than that done to the humans. While the people forced to be gymnasts would clearly be harmed, it remains unclear that the prejudice brought upon the horses would be significant. According to Cochrane, these horses are unable to construct or to seek to fulfill their own projects. To reduce their freedom and to force them to take part in activities which they have not chosen themselves will thus not necessarily go against their interests (as long as this is done painlessly). Cochrane’s arguments rest on a definition of autonomy which is quite stringent. Again, it is because they can frame, revise and pursue their own conceptions of the good that humans are presumed to have an intrinsic interest in liberty. It might however be the case, as Tom Regan thinks it is, that just because we have preferences which we seek to satisfy, we have an interest in seeking to satisfy them freely (I will come back to this shortly).3 The problem with such reasoning, according to Cochrane, is that it is possible to have desires and to have the means to try to fulfill them without being free. Such would be the case for a dog who would be allowed by his loving owner to pursue his desire for food, sleep, exercise, companionship and so on, as he wishes. According to Cochrane, the interest of the dog, as opposed to that of a human slave who would be enslaved to a kind master, lies solely in the possibility of satisfying his preferences, something which is not necessarily incompatible with his status as property. While this dog’s life is nothing like the one he would have chosen if he were left to himself completely, it is unclear that this kind of life undermines his interests, since he is able to satisfy his daily desires when he wants. We have good reason to suppose that it is generally in the interests of all living creatures to lead the life for which they were evolutionarily best adapted. It seems reasonable to suppose that an individual has an interest in being able to use her ‘natural functions.’ Martha C. Nussbaum (2004, p. 306) and Paul W. Taylor (1986, p. 108) are among the authors who defend this theory. Cochrane is nevertheless unconvinced of its relevance. While he recognizes that the exercise of natural functions allows one to gain pleasure and/or avoid frustrations, he holds that it is not the exercise of natural functions in itself which serves the interest of nonhumans. For Cochrane, the 4 idea that the life of an animal (or any other living organism) is better when the latter can act in agreement with its natural functions rests on a perfectionist conception of individuals, a conception suggesting that the value of a being is determined by the fact that it represents, or not, a good specimen of its species (Cochrane, 2009, p. 672). Yet he suggests that the well-being of an individual does not necessarily have anything to do with whether or not it is a good specimen of its species; a deer is most likely a better specimen of his species when he manifests a propensity for fighting for territory or mating, but it is certainly not in his interest to be harmed, sometimes fatally, during such fights. Even if the exercise of its natural functions can often turn out to be satisfying for a being, this kind of liberty in itself would not imply that its life is better from its subjective point of view. Naturally, Cochrane does not deny that it can be appropriate to recognize certain rights to sentient nonhuman animals, since they most certainly have interests that should be protected. He even admits that there exists among these interests an instrumental interest in liberty, to which great importance must be given. Nevertheless, he concludes that animals who are not autonomous can sometimes be kept and utilized for certain human ends, without it conflicting with their individual interests. It would not be appropriate to grant these animals the right to liberty because their other rights already suffice to protect the interests which, in reality, a right to liberty would aim at protecting. He even suggests that if our interference in the lives of sentient nonhuman beings does not constitute a violation of their interests in itself, then we perhaps have the moral obligation (given that we do not risk causing more harm than good) to intervene in the lives of those who live in nature to reduce their suffering, rather than to leave it all up to fate. Robert Garner offers a twofold criticism as an answer to Cochrane’s article. The first part of his argument is about the rather counter-intuitive concession which Cochrane is forced to make towards marginalized humans, whom he admits do not have a more intrinsic interest in liberty than nonhuman animals. Garner extrapolates from this concession that, following Cochrane’s reasoning, marginalized humans do not have a more intrinsic interest to live than their nonhuman counterparts. The second part of Garner’s argument discusses the exceptional nature of situations in which it is permissible to restrain the liberty of nonhuman animals without obstructing any other of their interests. Garner’s second point echoes some hesitation which Rachels first experienced as he demonstrated that the interest not to suffer and the interest not to be killed imply, from a practical point of view, an interest not to be exploited. The reason behind this notion is that sentient nonhuman beings who are utilized for human ends – especially on a large scale – generally undergo several major injuries before being, in most cases, slaughtered. Garner insists that recognizing that sentient animals have an interest in not suffering and that granting them the right not to have pain inflicted upon them would involve giving up a large proportion – even the majority – of human activities related to or implying the use of sentient nonhuman animals. All of this is completely independent of the question of whether or not we grant them an interest in liberty, according to Garner. He thinks that Cochrane is left with no other choice than to side with most of the demands made by the animal exploitation abolition movement, (which requires that we abolish all forms of animal exploitation), since Cochrane himself admits that the interest 5 of sentient nonhuman animals not to suffer must be taken as seriously as that of humans (Garner, 2011, pp. 180-85). Having addressed the question of the instrumental interest in liberty of sentient nonhuman animals,4 let us examine Garner’s first point, which concerns the case of marginalized humans. Garner notes that Cochrane’s accepts the following implication: if it is the case that marginalized humans are given as much moral value as humans considered ‘normal’, then nonhuman animals who possess the same cognitive capacities as the former must also be given the same moral value as the latter. However, instead of seeing fit to raise the moral status of nonhuman animals, Cochrane prefers to endorse a rather delicate and controversial position: marginalized humans do not have, not more than most nonhuman animals, an intrinsic interest in liberty. This position is problematic, says Garner, because it is counter-intuitive in itself and because it implies that children, mentally disabled people and senile people do not have an intrinsic interest to live. Garner recognizes that, in certain situations, it is not controversial for moral agents to interfere in a paternalist manner in the lives of marginalized humans. Indeed, when it comes to keeping a severely handicapped person from harming herself, for instance, it seems justified to intervene in order to protect her from herself. On the other hand, says Garner, the way in which we commonly interfere with the liberty of animals greatly differs from the way in which we intervene in the lives of marginalized humans: The denial of an interest in liberty for animals, though, means much more than [interference in their lives so that their interest is upheld]. It means that we regard it as legitimate to keep them in captivity, regarding them as our property and depriving them of self-ownership. Crucially, too, we do this not in their interests, but for a variety of human benefits, some more detrimental to the interests of animals than others (Garner, 2011, p. 178). Then Garner asks the question which interests us: [I]f it is unproblematic morally to regard animals as our property and to treat them as our slaves (on the grounds that they do not possess an intrinsic interest in liberty) then why is it any different for marginal humans? (Garner, 2011, p. 178) According to him, we would never accept that marginalized humans be confined or treated as mere merchandise or property. In fact, Cochrane himself rejects this possibility by insisting on the idea that, although marginalized humans do not have an intrinsic interest in liberty any more than most nonhuman animals do, they have other interests which we must respect. He even adds that those who care for marginalized humans are opposed to abusively exploiting them. Cochrane notes that using children for behavioral studies, for example, or hiring them for, say, theater plays will not lead to child slavery. Nevertheless, thinks Garner, when Cochrane raises the question of confining or utilizing nonautonomous animals in zoos, circuses, research laboratories or in the home for the sake of company, he seems to forget that we still would not be comfortable treating non-autonomous humans thus, quite regardless of the fact that we would refrain from inflicting pain upon them and from killing them. Garner thinks that Cochrane underestimates the gap between the 6 treatment we consider acceptable for human beings and that which he wants to justify in the case of nonhuman animals. Indeed, if he admits that both human beings and nonhuman animals have similar fundamental interests (both groups have an interest in avoiding suffering and an interest in life, but neither has the desire to be free per se), then they should be treated the same way. As it was just discussed, Cochrane does not deny that nonhuman animals have other interests which we must respect. As such, he is clear that the only acceptable ways to use animals for human ends, in his eyes, are those which cause no pain and which do not lead to death. Garner, however, thinks that Cochrane does not realize that these situations are quite improbable and that, contrary to the kind of interference which seems justified when it comes to marginalized humans, those which we allow ourselves when it comes to sentient nonhuman animals are practically always violent and fatal. After all, both sides of Garner’s criticism aim to show that the practical consequence of Cochrane’s argument is of negligible importance. If, on the one hand, we have to rigorously respect sentient nonhuman animals’ interest in avoiding suffering and death and if, on the other hand, the only kind of acceptable interference in these animals’ lives would be that which we judge appropriate for marginalized humans, then the problems of animal ethics would seem infinitely less pressing and serious than they are now. Garner also adds that arguments similar to those presented by Cochrane have been often used to negate the interest of nonhuman animals in continuing to live. Authors such as Peter Singer, he says, think that autonomy is a characteristic essential for the existence of such an interest. Yet, if we accept the argument from marginal cases (as does Cochrane), then we must either recognize the fundamental right to live of all sentient beings, or take away the fundamental right to live of marginalized humans. This is, according to Garner, the dilemma encountered when we adopt the criterion of autonomy as defined by Dworkin. Since we are not in any case willing to deprive marginalized humans of their fundamental right to live and to not be enslaved, then it is better, according to Garner, to abandon the criterion of autonomy entirely. Garner’s criticism leads us to think that Cochrane undermines the instrumental basis for not exploiting sentient nonhuman animals and not infringing on their liberty. Garner thinks that even if Cochrane is right and that only certain humans have an intrinsic interest in liberty, the demands of abolitionists would remain highly relevant. Cases of morally acceptable use of nonhumans for our own ends would, indeed, be very rare. That said, Garner, just like Rachels, does not succeed in overthrowing Cochrane's main thesis, which is that most sentient nonhuman animals do not have an intrinsic interest in liberty. One might think that we should accept the claim that only certain humans have an intrinsic interest in liberty. However, this conclusion appears to be premature. Indeed, before we conclude that marginalized humans and most nonhuman sentient animals do not have an intrinsic interest in liberty, we should see whether the concepts of autonomy and liberty lend themselves to other interpretations, simpler than that supplied by Cochrane. Non-human animals and the concept of autonomy 7 When discussing of the concept of autonomy, we should first remember that some authors have tried to distance themselves from traditional requirements and have shifted their interest to other paradigms, such as preference autonomy, practical autonomy and natural autonomy. Contrary to Kantian autonomy, which is associated with rational thought and deliberation, preference autonomy, according to Regan, consists in having preferences and having the capacity to initiate actions which aim to satisfy them (Regan, 1984, pp. 84-5). According to this conception of autonomy, it is not necessary to be able to take a step back from our own desires or goals to evaluate with impartiality what others should do if they were themselves placed in the same situation. In order to be autonomous, says Regan, it suffices to have goals or desires, to believe that certain actions will help satisfy them and to act on the motivation given by these objectives and beliefs. While most nonhuman animals are not autonomous in the Kantian sense, many of them, Regan reckons, exhibit preference autonomy.5 To support this, Regan gives the example of Fido, a dog who is hungry and who has not exercised in a long time. When given the opportunity to eat and the opportunity to run outside, Fido eats before he goes out. Regan thinks that it is reasonable to interpret this hierarchy of actions as proof that, in such circumstances, Fido prefers eating to running and that he acts according to this preference. Regan’s second example describes Fido’s reaction in a new situation in which we offer him a bowl of his usual food and a bowl of boiled eggplant. There it seems even more reasonable to interpret Fido’s choice (to eat his usual food over the eggplant) as proof that he prefers his food to the eggplants, and that it is this preference which motivated him to go for the first bowl rather than the second one. For Regan, this can be assumed even if this is the first time that Fido is offered such a choice. To determine which animals meet the criteria for possessing preference autonomy, Regan thinks that it is sufficient to observe their behavior and identify those who behave in a manner which can be intelligibly described and succinctly explained by their preferences and/or by the choices they make according to these preferences (Regan, 1984, p. 86). Regan concedes that the Kantian version of autonomy is adequate when it comes to describing the behavior of moral agents, who, indeed, must be able to rationally and impartially judge their own preferences in order to identify their moral obligations.6 However, he adds that being capable of moral reasoning is not a necessary condition of all types of autonomy and that it is reasonable to assume that at least all mammals above one year of age meet the criteria for possession of preference autonomy. Regan concludes that these nonhuman animals have an interest in liberty because they are autonomous – that is, they can have satisfaction in seeking to achieve the goals they set for themselves ‘in their own way’.7 For his part, Steven M. Wise prefers to speak of practical autonomy, which he defines as follows: [A] being has practical autonomy and is entitled to personhood and basic liberty rights if she: 1. can desire; 2. can intentionally try to fulfill her desires; and 3. possesses a sense of self-sufficiency to allow her to understand, even dimly, that it is she who wants something and it is she who is trying to get it (Wise, 2002, p. 32). 8 Wise explains that this definition implies consciousness (but not necessarily self-consciousness) and sentience. Contrary to Kantian autonomy – which Wise considers so stringent as to be out of reach even for most humans, especially for children and adults who are not rational, – practical autonomy corresponds to a less complex interpretation of autonomy which is privileged by many philosophers as well as by all judges of common law.8 For Wise, all beings capable of this kind of autonomy should benefit from the right to liberty (Wise, 2002, p. 33). Lastly, Bruce N. Waller defends an evolutionist interpretation of autonomy built on the natural propensity of many animals to find equilibrium between, on the one hand, situations in which they can explore and choose between different options and, on the other hand, situations in which they can use their persevering at a thing even when it does not yield immediate benefits. According to Waller, the superior intelligence of ‘normal’ humans allows them to sharpen their research skills to create a more complete inventory of the different options which are open to them. However, human intelligence is not a sine qua non condition of autonomy. Simply put, it is to humans what the senses of sight or smell are to nonhumans: a tool which allows them to have a range of options. As opposed to Kantian autonomy, natural autonomy is not specific to humans. For Waller, this kind of autonomy indeed benefits multiple nonhuman animals: We (along with many other animals) want and need the freedom to do otherwise, to pursue a variety of open alternatives. … [T]here are good natural grounds for favoring open possibilities — and thus our fondness for autonomy — asalternatives run deep: deeper than our higher rational planning powers, though not in conflict with them. … [S]pecies whose evolutionary survival strategy favored contingency-shaped learning above rigid behavioral programming will value autonomy-as-alternatives… (Waller, 1998, pp. 18-9). Thus it seems that the notion of autonomy is not always interpreted as it was by Cochrane – that is, as the capacity of an individual to frame, revise and pursue her own conception of the good. That said, it might not be sufficient to be capable of preference autonomy, practical autonomy or natural autonomy in order to have an intrinsic interest in liberty. As mentioned above, this is precisely what Cochrane means when he gives the example of the dog left ‘free’ to fulfill his desires by his loving owner. Again, for Cochrane, the possibility of having desires and seeking to satisfy them without being constrained to do so in a particular manner by an external agent is not logically incompatible with the state of ‘non-liberty’ which characterizes, say, companion animals.9 Perhaps Cochrane may be right after all to suggest that all beings can have their preference autonomy, their practical autonomy or their natural autonomy respected without really being free – unless, however, defenders of a simplified version of autonomy associate it with a type of liberty which is also simplified. Non-human animals and the concept of liberty Cochrane’s inference that human interest in liberty does not rely solely on the pleasures which liberty yields or on the desires which it allows to satisfy was illustrated above by both the example of Truman from The Truman Show and the example of the slave who desires to 9 continue serving his master. The fulfilled Truman and the satisfied slave have an interest in being able to chose what serves them best for themselves, whatever their level of daily well-being is. If being manipulated goes against their interest in liberty, Cochrane argues, it is because Truman and the slave are both potentially autonomous (in the Kantian sense – that is, able to frame, revise and pursue their conception of what is good for them) and that the circumstances in which they are placed prevents them from actualizing this potential. This is a conception of liberty which rests on self-realization and on self-determination, notions associated with the control which an agent can apply to her own destiny and her own passions. This interpretation of the concept of ‘liberty’ corresponds to that which Isaiah Berlin called ‘positive liberty’. In his famous article ‘Two Concepts of Liberty’ (Berlin, 2008, pp. 166-2017), Berlin distinguishes between positive liberty and negative liberty. Loosely speaking,10 the first kind of liberty considers the agent's internal factors while the second considers the external factors. Positive liberty corresponds to a sophisticated version of autonomy, according to which individuals have an interest in realizing their real nature by controlling their impulses or inclinations such that they can act according to their rational judgment. Although many authors have defended the positive interpretation of the concept of liberty,11 others are opposed to it, especially because they dread a drift into authoritarianism. Indeed, some liberals, including Berlin, fear that the minorities might be considered free when governed by the law of the majority (because they live in a society which manages itself) when in reality they are oppressed. Ian Carter adds that a majority oppressed by a dictator could be similarly considered free when in reality enslaved if we argue that the wisdom and intelligence of the dictator give him the right to decide what is good for the people he rules (Carter, 2003). According to Berlin, the problem with positive liberty is that it rests on an idea of division of the self. This idea consists in elevating the rational self above the instinctive or emotional self. It lends decisive importance to the responsibility of rational moral agents and to the control which they have over their own passions. In this view, it is specifically this control which makes agents free. Yet, we are told, this theory places us in a delicate position because it means that those who are most rational know better than others what is good not only for themselves, but also, perhaps, for less rational individuals. From there it takes just an extra step to be able to justify, with this interpretation of the concept of liberty, illegitimate intervention into the lives of others, based on the supposition that those who are more rational can legitimately force those who are less rational to realize their real nature and liberate them from their irrational desires by deciding, for them, what they can and should do (Berlin, 1998, p. 180). Then, relying on an example similar to the one Cochrane will later use,12 Berlin continues by saying that the defenders of the positive interpretation of liberty are forced to recognize that a slave who succeeded in freeing himself from his desires and passions would be free (Berlin, 2008, pp. 182-91). However, we commonly consider slavery to be the antithesis of liberty. By relying on the internal factors of the agent, the positive conception of liberty cannot address such infringements on liberty as self-denial or abnegation. (Carter, 2003) According to Berlin, history shows the severity of the risks of drifting towards despotism when the positive conception of liberty is involved. We should not forget that the metaphysical foundation of this conception of liberty is the idea that the rational judgment of an individual allows her to determine what others fundamentally want, regardless of whether they are aware of their own desires, and regardless of whether they think they have desires to the contrary. 10 This conception of liberty, grounded on the autonomy of rational agents, also seems to be incompatible with the theory of determinism, suggesting that the behavior and will of humans, like all other natural phenomena, are ruled by causal physical laws.13 For positive liberty to have a strong moral meaning, it needs agents to be able to act as the first initiator of their actions and find themselves actually responsible for their thoughts, their decisions and their movements, since it is grounded on those agents’ control over themselves. The deterministic thesis, if it is true, implies that human action, like all other natural phenomena, is not under the real control of individuals and is, unquestionably, caused by past events. According to this hypothesis, which centuries of discussion have failed to rule out, free will is only an illusion, and will or motivation is under no one's control.14 We should also note that doubts raised by the thesis of causal determinism applied to the human mind are not dispelled by the indeterminism of quantum mechanics, which tells us that two identical causes do not necessarily entail two identical effects. While this scientific theory undermines the universal nature of the classic interpretation of causality, it does not reintroduce any notion of control for the agent of her own thoughts, decisions or actions. Indeterminism allows unpredictability, but not responsibility.15 In fact, whatever unified theory of physics the scientific community might reach over time,16 the question regarding the existence of a perfectly free will (in the incompatibilist sense) remains open. Indeed, the causality thesis, as well as probability theory and chaos theory, is a threat to free will because it explains, with inexorable laws of physics, all decision-making, all volitions and all intentions in terms of natural events of which the agent does not have absolute control. Since it is the basis of positive liberty, the concept of autonomy remains uncertain as long as it is linked to a libertarian interpretation of free will. This might explain why many authors have tried to reconcile liberty and determinism by suggesting a simpler interpretation of the concept of freedom. According to the compatibilist approach, it is possible to be free even if the history of the universe is fixed and the human will is, like any other natural phenomenon, determined by the events which preceded it. Defenders of the compatibilist thesis suggest that we interpret liberty as the mere possibility for an agent to do what she wants, regardless of whether the will of this agent is determined by causes which do not fall under her control (McKenna, 2009). Galen Strawson gives as an example an individual who wants to enjoy her day off, but hesitates between climbing a mountain or reading Lao-tzu, where no one is forcing her and she feels free to choose between either one of these activities (Strawson, 2004). In this view, liberty simply consists in not being physically forced to act in a particular manner. Classic compatibilist liberty generally excludes constraints internal to the agent. Yet, when an agent experiences a neurotic compulsion, an obsession, a phobia, a dependency or other psychological disorder or mental illness, it dos not seem reasonable to consider her actions free (Vallicella, 2009). This might be why certain authors nuance their positions to recognize that internal constraints can be an obstacle for the compatibilist liberty of agents, and sometimes also add internal (psychological) obstacles, such as compulsion or panic, to the list of external obstacles to the liberty of an individual (McKenna, 2009; Strawson, 2004). 11 Thus understood, liberty is reducible to freedom of action, and does not imply a freedom of the will linked to personal autonomy (McKenna, 2009), whose existence remains uncertain: Personal (or individual) autonomy should […] be distinguished from freedom, although [...] there are many renderings of these concepts, and certainly some conceptions of positive freedom will be equivalent to what is often meant by autonomy (Berlin 1969, 131-34). Generally, one can distinguish autonomy from freedom in that the latter concerns the ability to act without external or internal constraints and also (on some conceptions) with sufficient resources and power to make one's desires effective (Berlin 1969, Crocker 1980, MacCallum 1967). Autonomy concerns the independence and authenticity of the desires (values, emotions, etc.) that move one to act in the first place.17 (Christman, 2009) Thus, according to this perspective, an agent is free even if her temperament, her personality, her preferences and her motivations are entirely determined by events over which she has no control and for which she is thus not responsible – such events as her genetic makeup, the environment where she was born, her life experiences and so on. (Strawson, 2004) Indeed, these determinants do not rule out the liberty which consists only in being capable to choose or act according to one's preferences, regardless of what causes them. Compatibilism is subject to serious criticisms and might fail to salvage the notion of moral responsibility from the determinist threat (Strawson, 2004; 2010). Nevertheless, this approach, which rests on a distinction between the agent's external constraints (to which some internal constraints are exceptionally assimilated) and internal constraints, represents a conception of liberty which is probably more intuitive and less controversial – the absence of serious physical or psychological obstacles. In this sense, the metaphysical concept of liberty which is adopted by classical compatibilism corresponds to the political concept of negative liberty which is preferred by Berlin. Indeed, Berlin defines negative liberty as the absence of the type of intervention in individuals’ lives on the part of other agents18 that would prevent these individuals from doing what they want (Berlin, 2008, p. 170) Similar to the discussion of whether or not certain psychological or internal constraints should be added to the set of external constraints related to compatibilist liberty, authors interested in negative political liberty work with a distinction between ‘pure’ and ‘impure’ negative conceptions.19 The first conception limits what is considered to be a hindrance to liberty to cases of physical coercion; this conception excludes threats and any constraints based upon (even minimally) the beliefs, desires or values of the agent. The second conception, more encompassing, includes threats, as well as other serious psychological constraints such as ignorance, phobia, illusions or irrational desires, if, for instance, the latter are provoked by others, by manipulation, by brain-washing or by indoctrination.20 While it is mitigated by the impure version, negative liberty, as opposed to positive liberty, requires only that there be no constraints imposed on the agent by others, constraints keeping her from doing what she wants. According to this interpretation of the concept of liberty, it seems that all mobile beings, capable of intentionality and animated by their desires or their preferences, can be wronged when their motivations or their inclinations are stopped by an external agent.21 Now we have good reasons to claim that sentient beings are capable of 12 intentional actions, of beliefs and of desires (Griffin, 2001). Thus, if liberty is understood in its negative sense, it seems that all sentient nonhuman animals can have an interest in being free and do deserve the right to have such liberty. In this case, it seems right to conclude that Cochrane is mistaken when he holds that nonhuman animals, like all humans who are not autonomous, do not have an intrinsic interest in liberty. It seems as though Cochrane has a reductionist concept of liberty and that he takes the sufficient condition for having an intrinsic interest in being free, which is rational autonomy, for a necessary condition. Indeed, when he uses the Truman example, Cochrane specifies that the situation into which this character is placed undermines his liberty, regardless of how liberty is interpreted (Cochrane, 2009, pp. 664-5). Yet, before discovering the truth and realizing that the universe in which he lives was fully created by other agents, Truman felt free to go about his daily activities as he saw fit, to choose what he wanted to do and to move around as he pleased. He did not suffer from any dependency or mental illness which would have made him incapable of fulfilling his personal interests and behaving how he wanted. The purely negative liberty of Truman remained whole, until he discovered the difficult truth and was physically prevented from fleeing. If it seems to us that Truman was not free, this might be because his situation was, after all, impeding his impure negative liberty and his positive liberty. However, at this stage of the discussion, nothing authorizes us to think that an individual must be autonomous in the Kantian sense in order to have an interest in not having her purely negative liberty hindered. Cochrane presents a line of reasoning which disregards the purely negative version of the concept of liberty. Yet, when we accept the concept of negative liberty (including the pure version), as Cochrane explicitly does and as many other authors22 do, we have to recognize that all beings who can suffer from an infringement upon their liberty have a direct interest in enjoying liberty. In this case, sentient animals have an intrinsic interest in liberty, in the purely negative sense of the term, because they have an interest in moving as they desire without intervention from others. As we have seen, negative liberty does not rely on the capacity for self-determination or autonomy as Kant understood it. Contrary to what Cochrane suggests, it is not necessary for individuals to be able to ‘frame, revise and pursue their own conception of the good’ to have an interest in liberty. Indeed, for individuals to have an interest in negative liberty, nothing more is required than for other people’s interference to cause them harm by keeping them from doing what they want. Even if sentient nonhuman animals were moved only by their instincts or their desires and never by their reason, they would be harmed whenever an external agent keeps them from behaving as they are inclined to behave, since – as is the case with humans – this causes them pain, fear, anxiety or, at the very least, frustration. It seems reasonable to conclude that these animals have a certain type of intrinsic interest in liberty, one based on the negative version. However, this does not mean that the negative liberty of all animals used by humans for their own ends is necessarily violated. It is indeed possible to imagine cases in which animals are human property without ever being subjected to any constraints by their owners. For instance, we can think of a dog who was born in a human environment and whom we have taught to respect traffic lights and to relieve herself in the backyard, solely with a positive reinforcement training method (i.e., with rewards and never with punishment). This dog could find herself in a situation in which she walks around as she pleases without any collar or leash, enters and exits the house by her own ‘dog door’, drinks and eats when she wants and rests when she so decides, 13 without ever being disturbed in her initiatives. This dog could keep her status as property without her master interfering in her activities. Her ‘negative’ interest in liberty could be respected despite her legal status as a thing or property. One might conclude from this view that it is possible to exploit nonhuman animals while respecting their interest in liberty. If this were the case, then Cochrane would be right to suggest that it is sufficient to aim merely to improve our treatment of the animals whom we use and possess, and to impede their will as little as possible, thus succeeding in protecting their negative liberty without having to change their status. The view that an animal can be considered free even while remaining the property of a human is very similar to the one denounced by critics of the concept of negative liberty in the human context. These authors point out that the concept of negative liberty invites us to think that human slaves are free on the condition that their owner be lenient and not use his power to do whatever he wants with them. Even if the dog is free in her movements and she is able to satisfy her desires, she nevertheless remains in constant danger of losing this negative liberty, as she is legally considered to be the property of her owner. For instance, the owner could suddenly start treating her more harshly or he could be forced to abandon his animal to new owner who is illdisposed toward her. Similarly, slaves who belong to a charitable master who does not interfere in their daily lives are nevertheless subject to this master’s authority and are at the mercy of his will because he can always decide to start controlling them more rigorously, thus infringing on their negative liberty. Can the well-treated dog and the slave with a kind master really be considered free in such circumstances? Does not their interest in liberty require more than mere non-interference? Many political philosophers think that an interpretation which reduces liberty to the absence of constraint would force us to conclude that slaves can be free, which is, in their view, highly counter-intuitive.23 This is why some of them hold that liberty can be violated even when there is no such interference: The lack of freedom suffered by slaves is not a consequence of their being hindered in the exercise of their desires. Slaves whose choices happen never to conflict with the will of their master may be able to act without the least interference. They nevertheless remain wholly bereft of their liberty (Quentin, 2008, p. xi). These authors want to resuscitate a third version of the concept of liberty to exclude the possibility of slaves being considered free. Such a version which was privileged by the English anti-monarchist philosophers at the beginning of the 17th century, before Hobbes and the utilitarians succeeded in having it wholly forgotten. Quentin Skinner defends this view and reminds us that some of those opposing the Stuart monarchy were denouncing the royal politics which gave the king a discretionary and arbitrary right, and which thus placed his people (who were supposed to be free citizens) in a situation similar to that of the slaves.24 These dissidents held that the subjects of a tolerant king, who lets them do as they please and who interferes very little in their lives, cannot be considered free because they are, nevertheless, submitted to the supreme authority of a monarch. 14 Critics of the royal prerogative began to argue that, to the extent that they were obliged to live in dependence on the power of the king, and obliged in consequence to rely on his goodwill for the continuation of their rights and liberties, they were living in a state of servitude (Skinner, 2002, p. 247). According to the antimonarchists, the subjects live under constant threat of seeing their lives shattered by the sovereign’s decision to limit their actions. The antimonarchists see this precariousness itself as infringing on their political liberty. Following the Roman law – according to which a free person is a person who, contrary to a slave, does not find herself in potestate domini (within the power of a master) (Skinner, 2002, p. 238) – and the republican critics, the neo-republicans think that liberty hinges on the absence of domination more than on the absence of interference. For those who study and defend the republican concept of liberty – Skinner, as well as Philip Pettit (2000; 2001) and Maurizio Viroli (2002) –, to be free means not to be under the domination of others, not to be subordinated to their will and not to be at the mercy of their moods and desires.25 The absence of interference is not sufficient for their definition of liberty.26 To be free, one has to find herself in a state of independence vis-à-vis the arbitrary power of others (Pettit, 1996, p. 576). By ‘arbitrary power’, they mean the power to interfere as one wants and in all impunity (that is, such that the agent who finds himself submitted to this power does not have any recourse for obtaining condemnation or amends in case of interference) (Pettit, 1996, p. 576; Skinner, 2002, p. 248, note 53). We should notice that, interpreted in such a way, liberty allows certain forms of interference. Contrary to Hobbes, for whom all laws or all politics aiming to protect the liberty of citizens constitutes in itself a trespass against their liberty (Larmore, 2004, p. 99), the neo-republicans think that only the risk of arbitrary interference constitutes an attack on liberty. On the one hand, the republican concept of liberty seems intuitively more stringent than the negative concept of liberty because it excludes not only certain actual cases of interference, but also the situations in which interference is only potential. However, on the other hand, the republican concept of liberty is less stringent than the negative concept of liberty because it does not condemn all forms of interference, but only those which are arbitrary and which place the individual or the group of individuals in a position of vulnerability or of domination vis-à-vis one or more other agents (Pettit, 2004, pp. 78-9). The republican concept of liberty then depends on the structure of the relations existing amongst individuals or groups, rather than on the contingent results of this structure: Whether a master chooses to whip his slave on any given day, we might say, is a contingent outcome; what is not contingent (or at least not in the same way) is the broader configuration of laws, institutions, and norms that effectively allows him to do so or not as he pleases (Lovett, 2010). And as it was the case for negative liberty, republican liberty, it seems, can benefit all beings capable of suffering from others’ interference. Indeed, contrary to positive liberty, republican liberty is not at all dependent on Kantian autonomy: 15 Freedom as non-domination does not therefore consist in rational autonomy or democratic self-government, as freedom has often been positively defined. Much less does it signify the activity by which we supposedly realize our true nature or give expression to our higher self, which are the sorts of conclusions to which positive conceptions of freedom typically lead. To be free from the domination of a master does not mean being the master of oneself, since it is a condition which people may enjoy in a variety of ways — as much by letting themselves be carried away by passion (a dangerous thing when at the mercy of the powerful) as by bringing themselves under the rule of their own reason. […] A free person is one whose freedom consists in the nature and the extent of the possibilities which lie open to them. Their freedom extends beyond their participation in democratic self-government, and it need not involve their shaping their lives ‘autonomously’, according to a plan of their own devising (Larmore, 2004, p. 97).27 In this sense, to have an interest in liberty, an individual does not necessarily need to lead a life which she has completely designed herself in a fully autonomous manner. According to the dominant negative conception of liberty, this individual requires only an interest in not undergoing interventions which prevent her from doing what she wants. According to the republican conception of liberty, it is enough that she have an interest in being sure that her status or her situation protects her from the arbitrary and prejudicial interventions of other agents. Since they have an interest in not being prevented by others from what they want to do, sentient nonhuman animals also have interest in being in situations in which this first interest is not threatened by an inferior moral or legal status they might have. If they have an interest in not undergoing constraints or interference, sentient nonhuman animals also have an interest in not risking undergoing them: they have an interest in liberty in the negative sense as well as in the republican sense. Certainly, the well-treated dog and the satisfied slave experience very little or no interference and remain relatively free, in a purely negative sense. However, their status as property places them in a position of inferiority, in which they constantly risk seeing their desires, volitions or inclinations frustrated by the arbitrary intervention of individuals benefiting from a higher moral and legal status. Their republican liberty is thus hindered, which goes against their interests. Contrary to what Cochrane suggests, liberty in itself does not necessarily depend on Kantian autonomy and does not only have value for those who are capable of such autonomy. Consequently, nothing allows us to conclude that sentient nonhuman animals cannot have an intrinsic interest in liberty. Since they most likely have an interest in not risking suffering arbitrary interference, a situation such as that of Truman or that of the satisfied slave would also cause them harm. Of course, this does not mean that certain humans – those who are ‘normal,’ as opposed to those who are marginalized – do not have more to lose in a situation such as Truman’s or in that of the satisfied slave than beings who are not autonomous in the Kantian sense. Indeed, it is possible that humans capable of ‘framing, revising and pursuing their own conception of the good’ 16 undergo an additional loss when they are victim of domination. There might be a positive interest in liberty which is added to their republican one. Furthermore, the act of manipulation or indoctrination might cause them an additional prejudice; an impure negative interest in liberty might be added to the first two kinds of interest. In any case, the pure negative interest and the republican interest in liberty remain in the reach of all those whose well-being can be negatively affected by an intervention from others which keeps them from doing what they want. All sentient animals have interest in liberty because they can suffer from the interference of other agents in their lives, but let us not forget that Cochrane questions not this, but another conclusion. For him, sentient animals can, indeed, have an interest in liberty; however, he thinks that this interest is purely instrumental because he thinks that it rests entirely on their interest not to suffer and their interest not to be killed. As I explained above, a purely negative interest in liberty consists of having an interest in not undergoing interference. Cochrane might be right to think that interference is harmful for animals because it causes displeasure and suffering. That said, we might also suggest that the rationally autonomous person’s interest in not being interfered with is also based on the fact that being kept from doing what one wants is frustrating, upsetting, annoying and painful. And if negative liberty is a good thing because, as Cochrane suggests, it allows autonomous people to frame, revise and pursue their own conception of the good, then is it not an instrumental good? Given the premise that the interest in liberty of autonomous people rests on their interest in bringing about the fundamental good that is autonomy, Cochrane may be incorrect to suppose that liberty can be an intrinsic good; it may always be reducible to obtaining a more elementary good, even for humans. While liberty prevents all sentient beings (human or not) from undergoing the suffering or frustration caused by constraints, it can also allow autonomous humans to realize their autonomy. In both cases, liberty can be considered to be a good which rests on obtaining another good. However, in this case, all the values – except perhaps the absence of suffering – can be considered instrumental; we can even interpret the interest of an individual in not being killed as his indirect interest in obtaining the good things in life. Now, since non-interference is usually considered to be intrinsically good for humans and it is not linked to the capacity of autonomy, it seems more reasonable and simpler to recognize that it can constitute also an intrinsic good for animals liable of being the victims of a prejudice when their will is frustrated by others’ intervention. Following this line of thought, let us assume that, according to the definition of human liberty which many authors have accepted, sentient nonhuman animals have at least an intrinsic interest in liberty in the purely negative sense, as much as in the republican sense. A right to physical integrity and a right to life would not sufficiently protect such an interest – at least not when we consider that these rights seemingly fail to properly protect human interest in liberty (and the proof of this is that we do not stop at granting all human beings – marginalized cases included – a right to physical security and a right to life, but that we also grant them a right to liberty). In light of such a conclusion, how are we to morally interpret or evaluate the current situation of sentient nonhuman animals? Pettit gives many examples of relationships between dominator and dominated where the republican liberty of the latter is violated. He mentions a husband who can physically abuse his wife without risking more than the moderate disapproval of his neighbors; an employer who can shamelessly fire his employees at his discretion at any time; a prison guard who can make the 17 lives of the prisoners miserable without himself facing any reprimand (Pettit, 1996, p. 581). These situations infringe on the republican liberty of individuals. Even when those in dominant positions do not actually take advantage of the lack of reprisals for their abuse of power, they infringe on the republican liberty of those who are subject to them and who find themselves in a position of constant and unacceptable vulnerability. We would hope that these unjust situations – or, at least the worst of them – are relatively rare today. If important issues remain in the application of the laws and rules which prevent such abusive behaviors, nonetheless many societies have come to condemn such practices in order to ensure those who can harm others by interfering in their lives can no longer do it with impunity. At least in Western societies, cases of dictatorship or human slavery are thankfully exceptional. On the other hand, all nonhuman animals are currently victims of human domination because they find themselves in a situation where, when they do not actively experience harmful intervention from humans in their lives, they nevertheless live under a sort of Damocles’ sword, constantly risking being victims of such intervention. Their situation corresponds precisely to that of humans reduced to slavery. Indeed, because of their current moral and legal status as objects which can be appropriated, all animals are victims of oppression and domination by humans. Of course we criminalize purely sadistic actions and give nonhuman animals minimal protection. We also claim to pay attention to their well-being, by opposing ill-treatment and by demanding that they be used under more ‘humane’ methods of exploitation. Some of us even agree to pay a bit more for food originating from organic farms because it is thought that animals are better treated on organic farms than on industrial farms. Yet, this charity is not justice and barely unsettles the relation of domination which remains between human and nonhuman animals (Rollin, 1992, p.122). Indeed, by refusing sentient nonhuman animals equal status, we maintain the right to frame, as we wish, what they can be subjected to. And it is our power over animals which allows us, on the one hand, to claim that we wish to treat them well (when we use them for our own ends) and, on the other hand, to mishandle them as seriously and as systematically as we do. The interests of an individual are not only threatened, but literally violated when he suffers from inferior status and finds himself in a precarious position; there cannot be real justice as long as there is inequality of status.28 A satisfying way to ensure that sentient nonhuman animals are no longer dominated is to offer them a republican type of liberty by granting them the moral and legal status of person, status which must come with at least the protection given by the most fundamental human rights.29 It is under this condition and under this condition alone, that we will be able to say that their fundamental interests and ours are equally protected. It is under this condition, and under this condition alone, that we will be able to say that their fundamental interests were considered equally and that the formal principal of equality, on which Rachels bases his arguments, is being respected. Abolitionist conclusion To recognize the fundamental equality of animals and the right of all sentient beings to be free is not without practical consequences. Indeed, if all sentient animals could benefit from equal 18 moral and legal status as well as the fundamental right to liberty, it would be prohibited to treat any sentient being as an inferior, and paternalism would only be justifiable when it essentially aims at serving the interests of the individual herself. While we have good reason to think that most of our domesticated animals should not be reintegrated into nature, and that as a consequence we have to take care of them and respect their interests to the best of our abilities without giving them back their true freedom, institutions based on animal exploitation, on the other hand, must be abolished. To recognize animal equality collectively implies that we must abandon all forms of institutionalized exploitation of sentient animals, which includes prohibiting not only practices such as hunting, fishing and scientific research involving sentient animals and institutions such as zoos or fur farms, but also all forms of domestication which requires domination and which goes against the fundamental interests of sentient animals.30 On the individual level, these changes imply that we adopt a vegan political stand and31—that is, that we must refuse to support unjustified exploitation of sentient beings and that we work to boycott, as much as possible, all products and services which rely directly or indirectly on their exploitation. 1 Cochrane, A. (2009), Do Animals Have an Interest in Liberty? Political Studies, 57, 660-679, here p. 665, where Cochrane paraphrases Dworkin to define autonomy thus: ‘the capacity to reason and reflect on those desires, and change them in relation to one’s values and conception of the good’. As this definition of autonomy is similar to the one Immanuel Kant privileges, we will call it ‘Kantian autonomy’. 2 Cochrane mentions that it might be better, when uncertain about the cognitive capacities of these animals, to suppose that to confine them or to use them is harmful for them, as it is the case with humans. See ibid, p. 668. 3 On the concept of preference autonomy, see infra, pp. 8-9. 4 Again, Gary L. Francione has thoroughly explained that it is practically impossible to make large-scale exploitation of sentient beings without infringing on their interest not to suffer and their interest not to die (Francione, 2000). 5 Regan specifies that at least all mammals show the capacity to choose according to their preferences because they possess cognitive capacities allowing them to have desires and to pursue goals; to feel and remember; to frame and apply beliefs (Regan, 1984, pp. 84-5). 6 It is interesting to note that Rachels, for his part, thinks that even Kantian moral liberty is not humanity’s privilege, since some nonhuman animals are able to act virtuously, and thinks furthermore that it is just as bad to infringe on their actions as it is for humans. See Rachels (1976), where he gives scientific results which show, among other things, that some primates prefer to starve rather then inflict pain on one of their fellows. 7 Regan gives the example of a captive wolf whose desire for food is satisfied by his owner without fail, but whose usually necessary propensity to use ingenuity to find food in the company of other wolves remains frustrated (Regan, 1984, pp. 91-92). 8 Wise refers not only to Tom Regan, but also to James Rachels, William A. Wright and Christopher Cherniak (Wise, 2002, pp. 30-32 and p. 33, note 38). 9 See supra, p. 4. 10 See the criticism of Quentin Skinner on this topic (Skinner, 2002, p. 239). On the pertinence of assimilating some internal constraints to external constraints, see infra, p.12 and ff. 11 See, for instance, Milne (1968), Gibbs (1976), Taylor (1985) and Christman (1991, 2005). 12 See supra, p. 3. Note that for Cochrane, this kind of example is used to illustrate the insufficiency of a conception of liberty which would be based on the satisfaction of desires and the preferable aspect of an interpretation 19 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 of liberty based on Kantian autonomy. According to Cochrane’s reasoning, we could admit that the slave robbed of his desires cannot benefit from positive liberty because he, then, cannot realize his nature or his real self. ‘In Laplacian determinism, the instantaneous state of the world (the initial conditions) alone decides, once and for all, the state of the world at all other points in time. …. The will is conditioned by a cause, a reason, a motive; it is impossible for the will to act from nothing. Accordingly, free will is an illusion of the mind, exactly like chance and other forms of indeterminacy—the result of our ignorance about the exact causes’ (Espinoza, 2006, p. 209 ; my translation). On this topic, see O’Connor (2008). Hard incompatibilists notice that, to be morally responsible for her own actions, an agent must have causally determined the latter. See McKenna (2009). Carl Hoefer discusses the ‘Final Theory of Everything’ (Hoefer, 2004). Note that David DeGrazia corroborates by using the vocabulary put forward by Harry Frankfurt. See DeGrazia (1996, pp. 204-205). See Berlin (2008, pp. 169-170). According to Berlin, the other types of obstacles do not obstruct the agent's liberty, but rather places him in a situation where he is incapable of doing what he wants. On this topic see also Carter, 2003, where he recalls this distinction, already illustrated by Hobbes, in the example comparing a person who cannot leave a room because he is chained with a person who cannot leave a room because she is confined to bed. See Hobbes (1651, p. 146). See Carter (2003). On the concept of interference and the different possible interpretations of it, see also Edmundson (2004, pp. 96-98). See Carter (2003). See also Philip Pettit who, while he privileges, as we will soon see, a third conception of liberty, defines the interference associated with negative liberty such as to include threat, manipulation and even, in certain circumstances, omission, in Pettit (1989, pp. 165-6, and 1996, p. 579). Adam Swift dedicates a section of one of his books to the distinction between ‘doing what we want’ and being autonomous (as opposed to the Kantian heteronomy). See Swift (2006, p. 59 and ff). The negative liberty as described by Berlin and associated with Thomas Hobbes, Jeremy Bentham and John S. Mill is the prominent concept in contemporary political philosophy, especially for Anglo-American authors. See Lovett (2010). To this, Frank Lovett adds that the concept of negative liberty can lead to even more paradoxical conclusions. For instance, consider the case of a people colonized by an imperialist power which, as it turns out, uses very little of its authority on them. Strangely, this people could be considered less free were they to revolt and attain their own political power by creating an active government that passes laws and adopts policies that interfere in the lives of the new citizens (Lovett, 2010). See Skinner (2002, p. 252), where he refers to the words of a certain Johnson. See Marmore (2004, p.97), where the author so summarizes the position of neo-republicans. It is interesting to note that, according to Philip Pettit, if authors like Jeremy Bentham and William Paley have preferred the concept of negative liberty over the republican one, it is because they wanted to avoid having to contest the relations of domination that were largely accepted between women and men, on one hand, and between masters and slaves, on the other hand. See Pettit (2001, p. 148). See also Lovett (2010), where we can read: ‘Notice that the republican view of freedom is, at least in the broad sense, a negative conception of political liberty. One need not do or become anything in particular to enjoy political liberty in the republican sense; one need not exercise self-mastery, on any view of what that entails, nor succeed in acting on one’s second-order desires. Republican freedom merely requires the absence of something, namely, the absence of any structural dependence on arbitrary power or domination’. This is an idea on which Gary Francione insists repeatedly in his work. This conclusion is inspired particularly by Paola Cavalieri’s demand that all conscious beings should benefit from fundamental human rights. See Cavalieri (2001, p. 139). ‘The word “domesticate” implies not only capturing and taming a free-living creature, but also altering him morphologically and psychologically by way of breeding or cross-breeding. […] Above all, domestication is 20 31 built on the drive, eugenic in character, to interfere in the destiny of a living being. It requires surveillance and the almost total domination of his actions and motions.’ (Danten (1999, p. 20), my translation). Veganism consists precisely in avoiding supporting or participating in any form of animal exploitation, by choosing not to consume any product or service which implies, directly or indirectly, the use of animals. Vegans go without all food originating from sentient animals, such as meat, milk and eggs. They also try to avoid animal products, such as clothing made from leather, suede, wool and fur. Lastly, they refuse to participate to any activities which involve sentient beings’ exploitation, such as visiting zoos, going hunting or attending aquatic amusement parks. Acknowledgments I would like to thank the Center for Research in Ethics (CRE) for its support, as well as Christine Tappolet, the head director, for her valuable advices. I would also like to thank Maude Ouellette for translating the article as well as Xander Selene for carefully revising it. Finally, I want to express my gratefulness to Carl Saucier-Bouffard and to Renan Larue for their inestimable help throughout the preparation of this paper. References Berlin, I. (2008). Two Concepts of Liberty. In Liberty (pp. 166-217). Oxford: Oxford University Press. Carter, I. (2003). Positive and Negative Liberty. In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. Retrieved from: http://plato.stanford.edu/archives/fall2008/entries/liberty-positive-negative/ Cavalieri, P. (2001). The Animal Question: Why Nonhuman Animals Deserve Human Rights. Oxford: Oxford University Press. Christman, J. (1991). Liberalism and Individual Positive Freedom. Ethics, 101 (2), 343-59. Christman, J. (2005). Saving Positive Freedom. Political Theory, 33 (1) 79-88. Christman, J. (2009). Autonomy in Moral and Political Philosophy. In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. Retrieved from http://plato.stanford.edu/archives/fall2009/entries/autonomy-moral Cochrane, A. (2009). Do Animals Have an Interest in Liberty?. Political Studies, 57, 660-79. Danten, C. (1999). Un vétérinaire en colère. Essai sur la condition animale. Montréal: VLB. DeGrazia, D. (1996). Taking Animals Seriously. Cambridge: Cambridge University Press. Edmundson, W. A. (2004). An Introduction to Rights. Cambridge: Cambridge University Press. Espinoza, M. (2006). Théorie du déterminisme causal. Paris: L’Harmattan. Francione, G. L. (2000). Introduction to Animal Rights. Philadelphia PA: Temple University Press. Garner, R. (2011). In Defense of Animal Sentience: A Critique of Cochrane’s Liberty Thesis’, Political Studies, 59 (1), 175-87. Gibbs, B. (1976). Freedom and Liberation. London: Chatto and Windus. Griffin, D. R. (2001). Animal Minds. Chicago IL: University of Chicago Press. Hobbes, T. (1651). Leviathan, or, The Matter, forme, & power of a common-wealth ecclesiasticall and civill, Printed for Andrew Crooke, at the Green Dragon in St. Pauls church-yard, 1651, Chapter XXI [online]. Electronic Text Center, University of Virginia Library, Chapter 21, p. 146. Retrieved from: http:// http://etext.lib.virginia.edu/etcbin/toccer-new2?id=HobLev2.sgm&images=images/modeng&data= /texts/english/modeng/parsed&tag=public&part=21&division=div2 Hoefer, C. (2004). Causal Determinism. In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy [online]. Retreived from: http://plato.stanford.edu/archives/spr2004/entries/determinism-causal Larmore, C. (2004). Liberal and Republican Conceptions of Freedom. In D. Weinstock and C. Nadeau (eds), Republicanism: History, Theory and Practice, London: Frank Cass Publishers, pp. 96-119. Lovett, F. (2010). Republicanism. In E. N. Zalta (ed.). The Stanford Encyclopedia of Philosophy [online]. Retreived from: http://plato.stanford.edu/archives/sum2010/entries/republicanism McKenna, M. (2009). Compatibilism. In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy [online]. Retrieved from: http://plato.stanford.edu/archives/win2009/entries/compatibilism Milne, A. J. M. (1968). Freedom and Rights. London: George Allen and Unwin. 21 Nussbaum, M. C. (2004). Beyond “Compassion and Humanity”. In M. C. Nussbaum and C. R. Sunstein (eds), Animal Rights. Oxford: Oxford University Press, pp. 299-320. O'Connor, T. (2008). Free Will. In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy [online]. Retreived from: http://plato.stanford.edu/archives/fall2008/entries/freewill/ Pettit, P. (1989). A Definition of Negative Liberty. Ratio, 2 (2), 153-8. Pettit, P. (1996). Freedom as Antipower. Ethics, 106 (3), 576-604. Pettit, P. (2000). Republicanism. Oxford: Oxford University Press. Pettit, P. (2001). A Theory of Freedom. Oxford: Oxford University Press. Pettit, P. (2004). Discourse Theory and Republican Freedom. In D. Weinstock and C. Nadeau (eds), Republicanism: History, Theory and Practice. London: Frank Cass Publishers, pp. 72-95. Rachels, J. (1976). Do Animals Have a Right to Liberty? In T. Regan and P. Singer (eds), Animal Rights and Human Obligations [online]. Englewood Cliffs: Prentice-Hall, pp. 205-223. Retreived from: http://www.animalrights-library.com/texts-m/rachels02.htm Regan, T. (1984). The Case for Animal Rights. Berkeley CA: University of California Press. Rollin, B. E. (1992). Animal Rights & Human Morality, revised edition. Amherst NY: Prometheus Books. Skinner, Q. (2002). A Third Concept of Liberty. Proceedings of the British Academy, 117, 237-268. Skinner, Q. (2008). Hobbes and Republican Liberty. Cambridge: Cambridge University Press. Strawson, G. (2004). Free Will. In E. Craig (ed.), Routledge Encyclopedia of Philosophy [online]. London: Routledge. Retreived from: http://www.rep.routledge.com/article/V014SECT1 Strawson, G. (2010). Freedom and Belief. Oxford: Oxford University Press. Swift, A. (2006). Political Philosophy. Cambridge: Polity Press. Taylor, C. (1985). Philosophy and the Human Sciences, Cambridge: Cambridge University Press. Taylor, P. W. (1986). Respect for Nature: A Theory of Environmental Ethics. Princeton NJ: Princeton University Press. Viroli, M. (2002). Republicanism, tr. A. Shugaar. New York NY: Hill and Wang. Waller, B. N. (1998). The Natural Selection of Autonomy. Albany NY: State University of New York Press. Wise, S. M. (2002). Drawing the Line. Cambridge: Perseus Books. 22