Introduction
While factory farming has been condemned by ethicists all over the philosophical spectrum, humane farming is still considered a moral option by many. By humane farming, I refer to those farming practices in which the animals lead happy lives but are nevertheless killed in order to use their carcasses for human consumption. Additionally, the killing is supposed to be humane as well, that is, to cause no pain or distress to the animal. This practice, arguably, does not exist nowadays. Although the animals raised in extensive, organic or traditional farms are in many respects better off than their counterparts in factory farms, their lives are by no means free of suffering, both physical and psychological. Moreover, it seems to be the case that in some respects they are worse off than those in factory farms,Footnote 1 as, for instance, they do not receive medications to treat their diseases. Furthermore, and importantly, these animals are slaughtered at the same abattoirs, where they experience not only physical pain but fear and distress. However, even if humane farming does not exist now, it could exist in the future, and many argue, for a variety of reasons, that it should exist, so it is imperative that we examine the moral implications of such a practice.
My objective is to evaluate one of the arguments used in the utilitarian debate on humane farming, in particular, how utilitarians deal with the uncertainty associated with this practice. First, I will explain the main tenets of utilitarian theory and what the common utilitarian response to animal suffering is, and why this position does not help utilitarians in devising a consensus answer to humane farming. Then I will address the question of moral risk and which possible principles could guide us in these cases. Following this, I will show how this notion is relevant to our discussion and which risks we need to consider. I will focus on the position of Singer and de Lazari-Radek to argue, against them, that we do have risk-related reasons to abstain from this practice. Finally, I will discuss what I consider to be the more plausible counter-arguments to my position.
Utilitarianism and animal ethics
Utilitarianism is a moral theory that claims that actions (and omissions) are to be evaluated morally depending on their consequences. More concretely, utilitarians focus on how the consequences of the action impact on the well-being of those involvedFootnote 2 (utilitarianism is compatible with different theories of well-being, hedonistic and preference satisfaction views being the most common ones). As well-being is the criterion against which the consequences are appraised, utilitarianism grants moral status to all beings that can experience well-being, that is, the criterion of moral worth is sentience. Furthermore, according to utilitarian theory, the objective of morality is to maximize utility (that is, well-being, determined by whichever standard we are using), and in order to measure this maximization, the interests of all involved are to be considered impartially. This means that equal interests are to be valued equally, regardless of who is the holder of the interest. Thus, the same interest in a human or a non-human animal is to be given the same weight. For these reasons, utilitarians have historically been ahead of society in caring for the interests of animals.
It comes as no surprise, then, that so many utilitarian philosophers have spoken against animal cruelty.Footnote 3 However, in the case of humane farming, there seems to be no such general agreement. This is because the question of (legitimately) humane farming cannot be solved by appealing to suffering, as the animals lead happy lives (and their deaths do not entail suffering). In contrast, the morality of humane farming depends on complex and more controversial topics, such as the badness of death for non-human animals and the value of coming into existence. There are, nevertheless, other questions related to the possibility of humane farming that might lead a utilitarian to reject the morality of humane farming, such as the impact on the environment, sustainability, the effects on human health, or the distribution of resources. These issues, not directly related to the effect of the practice on the animals, lead many utilitarians to deem this practice immoral.Footnote 4
My argument, however, will not focus on these issues, as they have been widely discussed already.Footnote 5 Nor will I directly engage with the discussion of the effects on the animals themselves. As mentioned before, the key elements of the discussion seem to be (1) whether death is bad for animals, and (2) the value of coming into existence, specifically, whether a short and happy life is better than no life at all. Although I find both questions extremely interesting, I recognize that both subjects are highly controversial and nothing close to an agreement has been reached after decades of debate. While academic discussions regarding the value of life and death are valuable in themselves, we nevertheless need to offer a prompt response to a pressing moral issue. As the animal welfare and liberation movements are getting more attention, and factory farming is being deemed as immoral by large parts of the population, we are in need of a new model. Whether this model is to be a vegan one, or one based on humane farming, is a moral question that needs an urgent answer.
Moral risk
Putting things simply, we need to make a decision when we do not have all the information required for that decision to be fully informed. Thus, we are in a situation of what philosophers call moral uncertainty (this uncertainty springing from either moral or non-moral questions). Making decisions in these situations puts us in a position of moral risk, that is, a situation in which we are risking something of moral value. For this reason, I will focus on an approach to this question from a perspective of moral risk. In particular, my objective is to elucidate whether we have enough reasons to abstain from the practice of humane farming from a perspective of moral risk when it comes to its effects on the animals themselves. In doing so, I will engage with the arguments of Singer and de Lazari-Radek,Footnote 6 to claim, against their position, that we do have reasons to do so. I will accept, for the sake of the argument, two of their premises, namely, that animals have moral status, and that it is in fact possible to farm animals without making them suffer. Furthermore, although other strategies of dealing with moral risk have been devised,Footnote 7 I engage with the argument on its own terms, using a strategy that utilizes principles that belong to the family of precautionary principles. There are two reasons for this decision. First, I wanted to show that, even using this kind of strategy, there are reasons to reject humane farming. Second, my main focus of interest is on the values of coming into existence and death, and the discussion of the different theories of moral risk management would distract us from these topics.
Two main arguments are used in this debate. First, those arguing that humane farming is morally acceptable need to claim that death does not harm animals, or if it does, that this harm can be compensated. After all, if animals have an interest in continued existence, we would need to take this interest seriously. And taking such as basic interest seriously would imply not thwarting it for trivial reasons, such as taste preference. Indeed, many authors claim that death does not harm animals, alluding to a variety of reasons, which generally converge on animals not being self-aware.Footnote 8 However, note that even if the animal itself is not being harmed by its death, there is nevertheless a reduction of overall happiness in the universe. As the animal was happy and its happiness has been eliminated, the universe is now worse than it was prior to the animal's death. Yet, this reduction can be compensated by bringing another being into existence, as long as the new one is as happy as the previous one. This argument, as devised by Peter Singer,Footnote 9 is known as the Replaceability Argument (RA). To be fair to Singer, he excludes farm animals from being replaceable, as they show some signs of self-awareness. But if this argument is applicable to farm animals, and they are indeed replaceable, that implies that killing them is permissible, and then, humane farming could be permissible too.
Some go further, and declare, not only that breeding and killing animals is not bad for them, but that is beneficial for them.Footnote 10 If this position is right, humane farming may not only be permissible, but obligatory, other things being equal (although it appears that other things are not equal, for the reasons mentioned above). This position is known as the Logic of the Larder Argument (LLA), and it incorporates the RA together with the idea that coming into existence is a benefit, and thus, a short life is better than no life at all.Footnote 11 This benefit could be for the animal itself, as it benefits from a happy life, or for the universe, meaning that the total amount of happiness in the universe is increased, or for both the universe and the individual animal.
It is fair to say that the foundations of the RA and the LLA are based on highly controversial issues, on which we have not reached a general agreement. This, however, does not mean that they are wrong. For instance, the moral status of non-human animals is still considered highly controversial, and countless people believe that almost any human interest has precedence over fundamental non-human interests. In fact, this is a mainstream view: although no one likes to be presented with images taken in factory farming facilities, most people continue to support this industry through their daily choices. Yet, this support is difficult to justify morally: other things being equal, fundamental interests (such as the interest in avoiding extreme suffering) should not be relegated to the same status as frivolous interests (such as the preference for one flavour over another). So, how controversial a position is cannot be understood as a proxy for how morally right it is. In fact, utilitarianism itself accepts ideas that are considered extremely controversial by other moral theories, but most utilitarian thinkers do not shy away from these conclusions merely because they are not widely accepted.
When I say that the RA and the LLA arguments are founded on controversial ideas I do not mean that they are not popular. What I mean is that there is no agreement among competent and well-informed thinkers about a reasonable answer. Furthermore, in some of the cases, there is not even agreement about what an answer to these topics would look like. There is uncertainty, not with regard to the acceptance of principles, but on the status of truth of the principles themselves. And uncertainty in moral decisions matters, as it means that we are making life-changing decisions based on principles of which we are unsure. When we do this, we are taking a moral risk, meaning that our action is risking harm to something that has moral value. Or, in other words, we are risking doing something seriously morally wrong.
Where does the uncertainty lie in these debates? Some of it is confined to the limits of morality, but it extends to other areas as well. Even if one was confident about the status of truth of utilitarianism and its basic tenets, there are a relevant number of other issues to consider, from axiological issues to factual knowledge. For instance, what is the value of life and death for non-human animals? Do they have an interest in continued existence? Is self-awareness a relevant trait when assessing the badness of death? And if so, are these animals self-aware? Do harms and benefits have the same weight? Do non-existing beings have interests? And, if so, do these interests have the same weight as those of already existing beings? Should we care about the amount of happiness in the universe or merely about how happy its inhabitants are? So, there can be uncertainty springing from both moral and non-moral spheres, but they all place us in the realm of moral risk: we may be risking something of moral worth. The question is, therefore, what to do in these cases.
The topic of moral risk has been ignored by philosophers for a long time,Footnote 12 but is getting increasing attention. As mentioned before, this article will focus on a specific answer to these situations: a strategy that belongs to the family of the precautionary principle. Principles like this have been stated in different ways and applied to different spheres of moral inquiry. For instance, MollerFootnote 13 has argued that, when the stakes are high (as in the case of killing), and one of the options is risk free, one should take that option. This applies to the killing of animals, as there is no theory telling us that we should kill animals, but merely that is permissible. In choosing not to kill animals, then, we ensure we take the option where no duty or moral obligation has been forfeited. The same would apply, for instance, to abortion, the actual topic of Moller's article (although not all those advocating for precautionary principles agree with his conclusion). He argues that given the intense disagreement in the abortion debate, we should admit our ignorance about key issues, such as the relevance of personhood, questions about personal identity, and the epistemic value of thought experiments, and acknowledge that we are operating under moral risk. Given this ignorance, the argument goes, we should take the morally safe option and choose not to have abortions. The safety, as in the animal case, resides in the fact that no theory claims that we should kill foetuses (at least not in cases where there is no physical risk for the mother and the future baby will not be severely disabled), but merely that it is permissible. Given this asymmetry, and the possibility of seriously harming something of moral value, we should abstain from abortion. Note that the argument is not that the pro-life side is morally right, but merely that their arguments are plausible, and thus, we should take them seriously. And taking them seriously implies accepting that there is a real possibility that we are seriously harming someone.
Guerrero argues in a similar fashion in an article appropriately titled ‘Don't Know, Don't Kill’.Footnote 14 Guerrero is mainly concerned with the problem of moral responsibility in cases of ignorance, that is, when is one morally culpable in cases where a morally wrongful action is performed from ignorance. Traditionally, it has been considered that one is only culpable in these situations when one is also culpable for the ignorance. If one refuses to find out about a morally charged issue, or is reckless in doing so, then one's ignorance is not an excuse, but an aggravation of the wrongness of the action. Arguing against this idea, Guerrero defends the application of epistemic contextualism (the position that the standards of knowledge vary with the context) to morality. Moral epistemic contextualism argues that how much one should do epistemically depends on the action (or omission): the more morally significant the action, the more stringent the epistemic demands. In other words, whether one is culpable for a given act performed from ignorance would depend on the moral implications of such an act. If the moral stakes of the act are high, the agent is held to higher epistemological standards. In turn, this implies that, when the stakes are sufficiently high, it may be morally impermissible to perform the act. That is, there are moral blockers.Footnote 15
This leads Guerrero to his principle ‘don't know, don't kill’ (there could be other similar principles when the risk of serious harm is real, such as don't know, don't injure, or don't know, don't destroy), claiming that when one is unsure of a being's moral status, it is morally blameworthy to kill it, unless there is substantial moral reason to do so.Footnote 16 Note that, as before, the morality of the action does not depend on the being having moral status, but on our killing it when we are uncertain about its status. Thus, killing a being with no moral status is still morally wrong if we are uncertain about its moral status. Furthermore, Guerrero argues that this principle can be directly applied to the killing of animals, given that there is much discussion about their moral status and that there is nothing of moral significance to justify their killing. Human gustatory preferences are not sufficient to trump the interest in continued existence of a being with moral status. However, his assessment of abortion is different from Moller's, given that he is open to the possibility that the interests of the pregnant woman are of enough moral significance to justify the killing of the foetus.
Not all those advocating these types of principles have limited themselves to the issue of killing. Vilhauer,Footnote 17 for instance, has developed an argument against retributive punishment based on a precautionary principle too. His discussion is based on the perennial debate about free will between libertarians, compatibilists and sceptics. Some sceptical thinkers argue that, as there is no free will, it is unfair to punish wrongdoers retributively, as they cannot be held morally culpable for their actions. Vilhauer claims that, given that punishment implies, by definition, harm to the punished person, we have a moral reason to take the disagreement seriously, and thus, refrain from retributive punishments (other reasons for punishments, such as deterrence, can still be morally justified). Again, his argument does not depend on the sceptical side of the argument being right, but on the arguments being plausible, and the possibility of serious harm being real.
Moral risk and humane farming
Some kind of precautionary principle is already used in animal ethics. For instance, it is a common idea among those considering sentience as the base of moral status that we need to give the benefit of the doubt to those animals whose sentience is uncertain. For this reason, many moral vegetarians and vegans abstain from eating animals on the fringe of sentience, such as molluscs. The justification is that these animals may be sentient, and thus, we may be doing something morally wrong if we deny them moral status, especially given that we can do this at no high cost to ourselves. Singer and de Lazari-RadekFootnote 18 have also applied this precautionary principle in cases of moral risk to the issue of moral status of animals. Using Moller's argument, they claim that denying moral status to animals might cause enormous unnecessary suffering and death, while the alternative is risk free, given that ‘there are no strong moral arguments that would make it wrong for me to become vegetarian’. Note that here, Singer and de Lazari-Radek are talking about suffering, so this argument does not apply to our discussion about humane animal farming.
So, the question is, can this precautionary principle be applied to the topic of humane animal farming? Not according to Singer and de Lazari-Radek.Footnote 19 When considering the early animal ethics discussion between Lesley StephenFootnote 20 and Henry Salt,Footnote 21 they claim that principles of caution cannot lead our decision in this matter. Stephen had argued that the practice of breeding, killing and eating animals was in fact beneficial for the individual animals, given that it allows them to exist. In contrast, Salt thought that this idea was absurd, as we cannot predicate happiness or unhappiness (or anything else, for that matter) of non-existing beings. Once one exists one can be happy to do so, but discussing the happiness of non-existing beings is nonsensical. A non-existing being cannot benefit from a future existence, as there is no one to benefit from anything until it actually exists.
Today, after decades of debates in population ethics, it is easy to identify these positions as the Total View and the Prior Existence View. These views deal with how utilitarians account for utility in the universe. While the first one is an impersonal view in which the interests of all beings count the same, whether they are existing, future or merely potential; the second one is a personal view in which only the interests of existing and future beings are to be taken into account, excluding those beings that are merely potential, that is, beings that do not currently exist and that might or might not exist in the future. These two views have radically different implications when applied to practical problems, the main one being that in the Total View creating new beings is considered an acceptable way to increase the amount of utility in the universe, while in the Prior Existence View, utility can only be increased by making existing beings happier.
At first sight, this problem would be an ideal one to assess on the basis of risk, given that the stakes are high, the discussion is extremely complex, and competent thinkers hold radically different ideas on the matter. So, why do Singer and de Lazari-Radek argue that this topic cannot be resolved by alluding to moral risk? The reason is that if one takes the Total View seriously, there are risks on both sides. On one side, if we implement a practice of humane animal farming, we have the obvious risk of killing creatures with an interest in continued existence. On the other, if we reject this practice, we may be depriving beings of happy and fulfilling lives, and thus preventing the amount of happiness in the universe from increasing. So, according to them, in this case, there is not a risk-free option.
However, that there are risks on both sides does not mean that the risks are equally distributed. I contend that these authors have failed to assess properly the risks on each side of the bargain. Allow me to elaborate: on one side there is a risk concerning death, on the other, a risk concerning existence. Are these risks of equal weight? In the case of death, the relevant sides of the argument claim that death is either bad, or neutral for the animals. No one, to my knowledge, argues that death is beneficial for animals. Even the defenders of humane animal farming through the LLA do not argue that death is beneficial for the animals, but that the whole practice is, as the possibility of existing compensates for the downsides of an extremely short life. As in the abortion debate, the disagreement is whether killing these beings is permissible, not whether it is obligatory. There are several reasons why it may be permissible: the animals may not be made worse off by death, and thus it is a morally neutral enterprise. Or, the animals could be affected by death, but this harm can be compensated, either with benefits enjoyed by themselves, or by others, or even the universe. In any case, the arguments do not claim that these beings, when already leading a happy life, are better-off dead, and this is the reason why we should kill them.
Once we have clarified the value of death in this argument, let us analyse the value of coming into existence. Singer and de Lazari-Radek rightly point out that there are serious arguments claiming that bringing new happy beings into existence is the right thing to do, as it increases the amount of happiness in the universe. Here, I will ignore problematic implications of the Total View, such as that it implies that we ought to have children if their lives are to be happy,Footnote 22 and the Repugnant Conclusion,Footnote 23 the idea that a universe filled up with barely happy beings may be at least equally good as one with a smaller but extremely happy population. Given the risk framework and the fact that the Total View is taken seriously by competent and informed thinkers, I will accept that it is a plausible view. This would imply that there is a risk in not bringing these animals into existence, as we are failing to increase the amount of happiness in the universe. The risk involved in the other alternative, namely, bringing the animals into existence, is not considered by these authors. This omission springs from the assumption that this option is either beneficial for the animals (as it would be if the Total View is true), or merely neutral (as it would be if the Prior Existence View is true).
But are these the only two alternatives? I contend that Singer and de Lazari-Radek fail to account for all the risks involved, and thus, they fail to assess correctly the risks involved in humane animal farming. An act can be good, neutral or bad. In the case of death, there are plausible arguments claiming that death for animals is either bad or neutral, and we dismissed the idea that death can be beneficial for those leading a life worth living. However, is coming into existence the same? We have discussed the Total View and the Prior Existence View, defending the goodness and the neutrality of coming into existence. But can coming into existence be bad, even for those whose life is worth living?
The idea that coming into existence is a harm has been a marginally held but ongoing worry in Western thought since ancient times. This philosophical position is nowadays most commonly encountered in the form of antinatalism, a theory that includes a value judgement about the quality of human (or sentient) life, but also a moral precept that we should abstain from reproduction. That is, antinatalists contend that creating new life is a morally wrong act. Many would be tempted to ignore this position given how few adherents it has, but antinatalism's unpopularity may be overestimated. Although it is true that antinatalist positions as such are uncommon, antinatalist sentiments have been voiced by a number of genuinely significant thinkers and artists across human history. It has been arguedFootnote 24 that rejectionism (that is, the idea that sentient existence is faulty and should be rejected) has had a major presence in human history throughout all three major cultural forms: religion, philosophy and literature.Footnote 25 While rejectionism and antinatalism are frequently treated as one in the contemporary literature, they are distinct positions, an axiological and a moral position, respectively. Although the arguments I am about to examine are antinatalist (due to the contemporary linkage of both positions), note that the risk argument only requires the weaker option, namely, rejectionism. That is, the idea that coming to existence is harmful, either because existence is bad or because it is worse (or not better) than non-existence, is sufficient for the risk argument.Footnote 26
While it is not possible here to delve deep in the history of this philosophical position, I will give a quick outline of a few relevant contemporary antinatalist arguments, as these are perhaps not readily available to all readers. First, Shiffrin, although not declaring herself an antinatalist, presents an argument that amounts to the antinatalist position. She claimsFootnote 27 that there is an asymmetry in the justification of inflicting harm when this harm is inflicted either to avoid a harm or to bestow a benefit. While inflicting harm seems to be uncontroversially justified when the objective is to avoid a bigger harm, the same cannot be said of cases in which we are merely trying to bestow a pure benefit. So, in cases where the victim cannot consent, we are justified in assuming consent in the first case, but not in the second. For instance, one would be justified in breaking someone's arm to save them from death, but not to do it in order to be able to subject them to cosmetic surgery to improve their already good looks. This is relevant to the ethics of reproduction as, clearly, this is a case in which the interested party cannot give consent. Given the impossibility of getting the consent of the future human being and that non-existence is not a harmful state, Shiffrin argues, we are infringing the consent rights of those that we bring into existence. Note, importantly for our discussion, that she does not need to argue that life is bad, or not worth living, merely that non-existence is not harmful. Thus, there is no reason for us to assume consent, even if the future life of the being would be perfectly happy.
Similarly, David BenatarFootnote 28 bases his antinatalist position on an asymmetry. Benatar argues that there is a basic asymmetry that explains other commonly held procreational asymmetries, as for instance, the common idea that we have moral reasons not to bring a miserable child into existence, but not to create happy children. This basic asymmetry deals with existence versus non-existence: while in existence the presence of pain is bad and the presence of pleasure is good, in non-existence the absence of pain is good while the absence of pleasure is merely not badFootnote 29 (the argument is commonly given in hedonistic terms for simplicity's sake, but it can be translated into other value theories). The reason for this asymmetry is that suffering is so bad that its absence is always good, while the absence of pleasure is only bad if someone experiences that absence. This gives non-existence an advantage over existence, as there is no downside to non-existence: there is no one to experience suffering or the absence of happiness. Here, as in the case of Shiffrin, Benatar does not need to argue that existence is bad, as even the best life will have at least a single moment of pain or discomfort, and given his asymmetry, this is enough to render this life worse than non-existence. Thus, according to Benatar, one is already put in a bad situation when brought into existence. Furthermore, Benatar also argues that existence is, generally speaking, not worth living, and that human beings generally are lousy judges of the quality of their own lives.
Another author who discusses the quality of life is Häyry.Footnote 30 His argument, like our own, is an argument based on risk assessment. He convincingly argues that given that life can be horrible, it is irrational to bring new beings into the universe, as this huge risk can be avoided by an alternative completely devoid of risk: if one does not have children no one gets hurt. Again, his position does not rely on every life being wretched, but on the mere possibility that such a life is possible for the being we are creating. Ignoring this risk is irrational, he claims. Furthermore, procreation is not only irrational, but immoral, as it disregards the real possibility of suffering, and the creation of avoidable suffering is morally wrong.
Going back to our discussion of the LLA, we can appreciate that the question is even more complex than it appeared to be at first. While it is true that there are risks on both sides, the risks are not equally distributed. On the one side, if we do not breed animals with the purpose of killing them, we may be depriving them of a happy life, and failing to increase the amount of happiness in the universe. On the other side, if we breed these animals, we may be acting morally wrongly not only when we kill them, but also when we bring them into existence. Note that this argument, like the previous ones regarding risk, does not need to prove that being brought into existence is morally wrong, merely that there is a reasonable doubt that it may be morally wrong. Therefore, the risk on the side of breeding animals is double, as both actions may be morally objectionable (the breeding and the killing), instead of just one of them (the failure to breed).
For this reason, I contend that an assessment of the risks involved leads us to refrain from breeding and killing animals for human consumption, even if these animals experience a positive balance of well-being during their short lives. While it is true that this case is not as clear cut as some of the ones we have discussed, such as the retributive punishment or the moral status of animals, where one of the options involves no risk, there is nevertheless an uneven distribution of risks. Thus, when we breed and kill animals, we may be doubly harming them and given this risk, we are not justified in breeding and killing animals by appealing to their own interests. Guerrero's proposal can help us to elucidate the issue further. Recall that, according to Guerrero, we need more stringent epistemic standards as the risk of harm increases. In our discussion we have seen that the risk involved in one of the options is higher, so there would be stronger blockers on this option.
Moreover, as Guerrero bases his precautionary principle on the possibility of causing harm to creatures with moral status, it is not clear that the other option involves a comparable risk, given that even if one accepts the Total View idea that these animals have an interest in coming into existence, and this interest is frustrated, they do not experience the frustration. The harm done to them, thus, seems less severe than that done to those that can experience the frustration of their interests. A further and related argument hinges on the difference between personal and impersonal harms. Admittedly, even if we consider the personal harm done to the non-existing beings less severe, we still need to contemplate the risk of failing to increase the amount of happiness in the universe, and I take that risk seriously. However, I cannot but feel that failing to fulfil an impersonal duty is not as serious as harming a living and sentient being for no good reason. After all, the universe itself is not conscious or sentient, and thus, cannot care about the amount of happiness in it. This argument, I am aware, may not work with those utilitarians committed to the Total View, who may interpret this further argument as begging the question. They may be right, but I find it difficult to come to terms with the idea that an impersonal loss of utility is morally equivalent to personal harm done to a sentient being. As my main argument does not hinge on this intuition, I leave it to each reader to decide whether they do or should share this intuition.
As previously mentioned, there is one more possibility when assessing the value of coming into existence: it might be considered neutral. The Prior Existence View is such a position. Proponents of this view argue that we should not count the interests of merely possible beings, which implies that one cannot be benefited (or harmed) by coming into existence, and that happiness in the universe can only be increased by making existing beings happier, not by creating new beings. This commonly leads them to dismiss both the RA and the LLA,Footnote 31 especially when they accept that death can be bad for animals, given that the negative utility created by the death cannot be compensated by any benefits, to either the animal or the universe. Singer and de Lazari-Radek consider this possibility, which leads them to acknowledge the risk in humane farming. However, as we have seen, they argue that there is also risk in abandoning the Total View. They therefore conclude that this issue cannot be resolved by alluding to risk-related reasons. However, now that we have discussed the possibility that coming into existence can be a harm too, we can appreciate the stakes differently. On the one side, we have the Total View, which declares the practice permissible (or even obligatory), but on the other, we have two utilitarian positions that reject the practice, alluding to greater harms. Thus, if we want to minimize the probabilities of harm to something of moral value, we need to reject the LLA and with it, the possibility of humane animal-husbandry.
A further argument is possible, not directly related to the values of starting and ending existence. Although most utilitarians adhere to the position that benefits and harms have symmetrical value, there are also different versions of what are commonly known as asymmetrical views. This strand of utilitarian theory argues that benefits and harms are not of equal value, with benefits having less impact in the utilitarian calculus, or even no impact at all. A well-known instance of these views is Popper'sFootnote 32 negative utilitarianism, which argues that we have a duty to minimize suffering, but not to increase happiness. Theories of this kind have appeared outside utilitarian theory as well.Footnote 33 These views also support the idea that humane animal-husbandry is to be rejected on moral grounds. First, if death is indeed bad for animals, this negative utility could not be compensated by a similar amount of positive utility experienced during the animal's life. Furthermore, as our efforts are to be aimed (totally, or for the most part) at alleviating suffering, the justification of humane farming as a way of increasing the amount of happiness in the universe is not available. Thus, this alternative version of utilitarian theory also points to the immorality of the practice.
We could allude to further reasons that would direct us to reject the possibility of humane animal-husbandry coming from theories other than utilitarianism, such as theories of rights,Footnote 34 or even virtue ethics.Footnote 35 However, as our objective was to assess how utilitarians should address the risks in this practice, I will not venture into different theories, and their criticisms of utilitarianism. I hope I have shown that, even staying within the boundaries of utilitarian theory, we have enough risk-related reasons to reject the practice of breeding and killing animals for human consumption, even if these animals lead happy lives and are killed painlessly.
Possible criticisms
Those who reject my argument, but who are not dismissive of the relevance of moral risk, could counter my position by arguing that there is a risk in not killing animals. In other words, they might argue that we are morally obliged to kill animals and failing to do so is morally wrong. However, I cannot see how such a statement could be justified. I have come across only one argument of this kind. In Meat, BelshawFootnote 36 argues that death can be better than existence for an animal. He argues that the animal may experience agony, either at the moment or in the future, and given that, according to him, these animals experience life only as a series of isolated events, previous or future happiness cannot compensate for it. Although I disagree that this is an accurate representation of animal minds, I will not pursue that line of argument. My main concern here is with the jump from the possibility of suffering to the acceptability of death. We can accept that, even if these farm animals lead happy lives, some of them would eventually be in agony due to statistically inevitable sickness, and that for them death is the better choice. However, this does not support the idea that death is better for all animals. Imagine I have three dogs and one of them, being seriously ill, should be killed to alleviate its suffering. This killing appears to be morally justified. What this does not justify is me killing my other two dogs when they are healthy. Belshaw's argument shows, at best, that euthanasia is more readily justified in animals, not that death is the best option for all animals.
A further criticism of my argument could be based on the relation between antinatalism and the badness of death. It could be argued that, given that the antinatalist position considers coming into existence as a harm, the contemplation of death should be seen as a relief from a miserable state, and not as a negative event. When applied to the LLA, this means that, if we accept the antinatalist position, the animals are harmed when brought into existence, but not when killed. This would imply that there is only one risk on that side of the bargain. This, however, is not the right assessment. First, although some antinatalists accept a pro-death position, this is by no means the standard stance. For instance, none of the authors we have discussed are pro-death: they hold that being brought into existence can violate our consent rights, put us in a bad situation compared to non-existence, or expose us to unnecessary risks, but once we are alive, we have additional interests, and some of these will be frustrated by death. This means that once we exist, continuing to exist may be the best course of action, at least while we have a life worth living. So, the antinatalist position is compatible with the idea that death harms the victim. Furthermore, this criticism seems to misunderstand the very key of the risk argument. As mentioned, the argument is not that the antinatalist position is right, but that we should take its arguments seriously. At the same time, we should also take seriously other arguments, such as those which hold that death is bad for animals. This leads us to the possibility that there is a double risk when breeding animals with the intention of slaughtering them.Footnote 37
I have analysed the risks that humane animal farming involves for the animals themselves. My objective was to counter an argument offered by Singer and de Lazari-Radek where they claim that although concern about moral uncertainty can help us solve issues involving the moral status of animals, it does not offer a solution in the case of humane animal farming. These authors argue that in this case there are risks on both sides and thus, there is not an option that allows us to minimize the risk of harming something of moral value. I contend that they have failed to take into account the antinatalist position and the fact that coming into existence can be a harm too. Therefore, if we are committed to avoiding the risk of harming individuals with moral status, we should take the option that minimizes the risk, namely, not breeding these animals. Furthermore, other concerns not limited to the effects of this practice on animals also point to its rejection. As the human interests involved in the practice are frivolous (limited to preferences in taste) and marginal (limited to the difference in pleasure between enjoying a meat-based or a plant-based meal), there are no relevant human interests that can justify the practice. Moreover, as the consumption of animal products is linked to serious environmental, health, and distribution problems, there seem to be no alternative arguments supporting the practice.Footnote 38
