What’s wrong with eating meat and other animal products, such as dairy or eggs? The usual answer appears to be simple: these products involve a great deal of animal suffering, particularly as most of them are produced on ‘factory farms’, where animals are raised in terribly cramped conditions that exacerbate their suffering. Suffering is the problem. If animal suffering were eliminated or significantly minimised, killing animals would not be such a big deal.
That’s the conventional moral thinking on animal ethics, and it applies not only to the use of animals for food but to all animal use: we can use and kill animals for our purposes, as long as we treat them ‘humanely’ and do not inflict ‘unnecessary’ suffering on them. This position is so widely accepted and uncontroversial that it is contained in laws that allow us to use and kill animals but that prohibit cruelty to animals.
The problem is that conventional moral thinking about animal ethics is unsound.
Because animals are chattel property, the concepts of ‘humane’ treatment and ‘necessary’ suffering are largely meaningless as moral concepts. They are primarily economic concepts that, in reality, translate into very little protection for animals. Moreover, the idea that killing animals is not a serious issue as long as animals are not made to suffer rests explicitly on the widely accepted idea that animals do not have a morally significant interest in continuing to live. And that is nothing more than an anthropocentric stipulation.
Before the 19th century, at least in the West, animals were largely excluded from the moral and legal community. They were considered as things. This is contrasted with Eastern thinking, which generally accorded at least some moral value to animals that accounted for the vegetarianism that remains prevalent in the Jain, Hindu and most Buddhist traditions. The Western view was that we could have moral and legal obligations that concerned animals but were not owed to them. To the extent that the cruel treatment of animals was thought to present a moral problem, it was only because it made us more likely to be cruel to other humans. But any obligation to be kind to animals merely concerned animals; the obligation of kind treatment was owed only to other humans. This was the view of Immanuel Kant, Thomas Aquinas and others. We had a legal obligation not to harm our neighbour’s property – whether that property was a cow or a cart. But that was an obligation owed to our neighbour as a property owner, not to the cow or the cart.
Although, in many instances, the status of animals as things was linked to the theological notion that only humans were deemed to have been created in God’s image, its primary focus was on cognition. Animals supposedly were not rational, self-aware or able to use concepts, and this was thought to justify our treating them as having no moral value.
Animals have no sense of what they lose when we take their lives, Bentham argued
This changed as part of a paradigm shift that occurred in the early 19th century and that was brought about by a number of thinkers, one of the more important of whom was the utilitarian philosopher and law reformer Jeremy Bentham. Bentham argued that the only characteristic that mattered for moral significance was the ability to suffer: ‘The question is not, Can they reason? nor, Can they talk? but, Can they suffer?’ As long as an animal was sentient, or subjectively aware and could suffer, that animal’s interests in not suffering mattered morally. Ignoring that suffering because of the species of the being at issue was no more defensible morally than ignoring human suffering based on race.
Although Bentham maintained that the cognitive differences between humans and nonhumans were irrelevant insofar as animal suffering was concerned, he regarded those cognitive differences as very relevant to the issue of killing animals. The fact that animals are not self-aware meant that they live in the present and have no connection with a future self; they have no sense of what they lose when we take their lives, he argued. They do not care that we use and kill them; they care only about how we treat them and kill them. If we kill and eat them, he said, ‘we are the better for it, and they are never the worse. They have none of those long-protracted anticipations of future misery which we have.’ If, as Bentham maintained, animals do not as a factual matter have an interest in continuing to live, and death is not a harm for them, then our killing of animals would not, per se, raise a moral problem as long as we took seriously the interests of animals in not suffering when we killed them.
Bentham’s view – that animals have a morally significant interest in not suffering, but do not have an interest in continuing to live – is the conventional position that most people embrace, and it is reflected in the law. This position rejects the idea that nonhuman animals are persons – that is, beings who can be said to have a morally significant interest in continuing to live. But it also rejects the idea that animals are things that have no moral value. Animals are not persons; they are what I have referred to as quasi-persons.
So how well does quasi-personhood work? How much protection against suffering does it provide?
The short answer: quasi-personhood status does not work; it provides very little protection.
There are two reasons for this.
First, animals are chattel property. They are economic commodities. Although we may think of animals as quasi-persons, their status as property guarantees that they will remain as things with only external or extrinsic value. Because they are property, and because our conventional thinking rejects the idea that killing animals is wrong per se, we do not ask whether particular institutionalised uses are necessary – we assume that they are necessary as exercises of property rights – but only whether the treatment of animals pursuant to those uses is ‘humane’. But if the use itself is not necessary, then all of the suffering inflicted pursuant to it is, by definition, unnecessary.
Consider our use of animals for food, which represents our numerically most significant animal use. We kill more than 70 billion land animals and at least 1 trillion sea animals every year for food. To put this in perspective, we kill and eat more animals in one year than the estimated number of human beings who have ever lived. Is any of this animal use necessary? Putting aside instances where people are stranded on lifeboats or desert islands, or otherwise facing imminent starvation, it is not necessary for humans to eat animals or animal products. Indeed, for several decades, a growing number of mainstream healthcare professionals have been telling us that animal foods are not only not necessary for, but are actually detrimental to, human health. Moreover, animal agriculture is an ecological disaster as it is responsible for at least as much greenhouse gas emissions as the burning of fossil fuels for transport, and having many other disastrous consequences. We eat animal foods because we think they taste good, and because we’ve been eating them for a long time. So, although the use of animals for food is not even arguably necessary as a general matter, we ignore that, and ask only whether animals used for food have suffered unnecessarily.
It is also not necessary to use animals for clothing, entertainment and sport, etc. Most – substantially all – of our animal use is unnecessary. Indeed, our only use of animals that is not transparently frivolous is the use of animals to cure or ameliorate serious human diseases. Consideration of that use requires a different analysis. So, if we applied the conventional framework to the issue of animal use and not just to the matter of animal treatment, we would find that most animal uses run afoul of the idea that any animal suffering needs to be justified by necessity.
‘Humane’ handling of animals at the slaughterhouse is relevant as an economic matter
Second, because animals are property, and because we do not focus on animal use but only on animal treatment, the level of protection accorded to animal interests in not suffering will be limited, for the most part, to that which is economically efficient. That is, we pay for a level of protection that will provide a greater economic benefit than what we pay to secure it. That is what ‘unnecessary’ suffering means in this context; it is suffering that results in a greater economic cost than the cost of preventing that suffering.
So, for example, legislation requiring ‘humane’ slaughter is common. Why? Because handling at the slaughterhouse is relevant as an economic matter; not protecting at least some animal interests during the killing process can result in incurring significant losses. The celebrated animal behaviourist Temple Grandin, a leading proponent of the ‘humane’ treatment of animals we use for food, and a designer of slaughterhouses, explains the matter clearly: once animals arrive at the slaughterhouse, ‘proper handling procedures are not only important for the animal’s wellbeing, but can also mean the difference between profits and losses due to meat quality or worker safety.’ Humane slaughter ‘will minimise stress levels, improve efficiency and maintain good meat quality. Rough handling or poorly designed equipment is detrimental to both animal welfare and meat quality.’
When we ask whether suffering is unnecessary in the context of animal use for food (and most other uses that cannot plausibly be characterised as necessary), we are really asking whether we are inflicting unnecessary ‘unnecessary’ suffering – whether the suffering we are inflicting is gratuitous. And in order to determine that, we usually look to what is considered ‘normal’ or ‘customary’ by those who use animals for institutionalised purposes, whether for food, entertainment or medical research. We assume that these users are rational and self-interested property owners and will not engage in the gratuitous harm of their animal property.
Standards of animal welfare are not about morality; they are about economics. They are about identifying standards of treatment that avoid the infliction of gratuitous harm. And the resulting level of protection is generally very low.
Could we improve animal welfare and accord a greater weight to animal interests in not suffering – could we build in a moral component? Sure. Some countries have higher welfare standards than others. But the ability to do this is severely constrained by economic and political considerations, including that markets are less and less national and more and more international, and this limits the ability to deviate significantly from the minimal standards. The result is that the difference between the most ‘humane’ standards and the least ‘humane’ standards in terms of what is legally required is almost always the difference between horrible and, at best, moderately less horrible. Some producers provide ‘higher welfare’ products to more affluent consumers who are willing to pay an increased cost for products from animals who have been raised in less intensive conditions. But even the most ‘humanely’ raised animals are still subjected to practices that cause considerable suffering and distress.
The view that animals do not have an interest in continuing to live because they lack humanlike self-awareness, which, as discussed above, reinforces the property status of animals, is so much a part of our thinking that it is accepted even by prominent modern animal ethicists. That is, these thinkers accept this as the unquestioned starting point and then go on to argue that some animals may be more than merely sentient and may be self-aware. Peter Singer, despite being a utilitarian for whom the consequences of actions – and not rights concepts – determine what actions are right or wrong, is often referred to as the ‘father of the animal rights movement’. Singer maintains, like Bentham, that sentient animals have an interest in not suffering but do not enjoy a presumption that protects their continued existence unless they are also rational and self-aware. The primary difference between Singer and Bentham is that Singer believes that at least some animals, such as nonhuman great apes, elephants, dolphins and some birds are, to some degree, rational and self-aware, and have a morally significant interest in continued existence, although that interest can be outweighed by human interests.
Tom Regan, a rights theorist, maintains that ‘subjects of a life’ have a right to respectful treatment that precludes their being treated exclusively as means to our ends. But a subject of a life must have a psychophysical sense of identity over time; that is, Regan’s view of who can qualify as a nonhuman person with a morally significant interest in continuing to live is very similar to Singer’s. The animals that clearly qualify for Regan are normal mammals of at least one year of age. Although Regan qualifies his position by saying that being a subject of a life may not be necessary for nonhuman personhood, and that sentience may be enough, he claims to be ‘radically unclear’ as to how that could be the case.
Singer, Regan and many other animal ethicists fail to appreciate the significance of the fact that sentience is a means to an end – that of continued existence; sentience is what ensures continued survival and the continuation of conscious states. To say that a being who is sentient does not have an interest in continued existence is like saying that a being who has eyes does not have an interest in continuing to see. Seeing is what an eye does; perpetuating consciousness so that life continues is what subjective awareness does. When we kill a sentient being, we take something of value from the being, however that being values it. We take that which the sentience of the being was designed to protect – the continuation of conscious states. That is a harm to the being. If a being is conscious at all, what it is like to be that being involves an organisation toward staying alive. This does not require that we understand the nature of any animal’s consciousness – a matter about which we will likely never agree. It just requires that we understand what consciousness is.
Every being with phenomenal consciousness is connected with their future self if only in the next second of that future. They have an interest in getting to that next second. They are necessarily oriented toward the future at least to that extent. It is their very existence as a sentient being and their connection to at least the next second of their conscious experience that provides a sense of personal identity. Putting aside humans who do not want to continue to be conscious and who kill themselves, that is true of every sentient being; the fact that they are not or may not be connected to the future in the way that normally functioning adult humans are is irrelevant. And every being who has phenomenal consciousness is necessarily self-aware on some level, if only of the self in that moment. We do not need to be able to contemplate our demise or death in order to be self-aware, as Bentham maintained.
Consider Fred, a human with late-stage dementia who lives in an eternal present
We recognise this where humans are involved. For the purpose of deciding whether it is morally justifiable to use a human exclusively as a resource to benefit others, we don’t ask whether a human thinks about having a life in any way that is similar to the way that normally functioning humans do. We assume that humans have a protectable interest in life if it is in their interest to live, which we assume is the case for any human who has conscious experience. This assumption is so strong that, even when we allow assisted suicide, we have multiple safeguards in place to make sure that death is what the person really wants. We do not require that humans have minds that are similar to the minds of normally functioning humans to be a person; any human who has a mind is regarded as a person. We recognise that for a human to be conscious at all is to be a being concerning whom there is something it is like to be that human, even if the quality of that person’s life, or their connection with their future self, is not the same as is the case with a normally functioning human. We recognise that there is no nonarbitrary way to determine when the connection to a future self is sufficient for human personhood.
I am sceptical of the claim that nonhumans, or at least many of those we exploit, live in an eternal present, but I do not know. I do know, however, that there are many humans who live in an eternal present relative to normally functioning humans. For example, there are many humans who have late-stage dementia. They are not just forgetful or very forgetful; they often have no memory of the past, have no idea about who they once were, do not recognise people with whom they have been very close, and have no ability to plan for the future. They have no psychophysical sense of identity other than the one that they have in that very second. Their experiential welfare is limited to the present moment. They are connected to a future self but only in the sense of wanting to get to the next second of conscious experience. I submit that, as an uncontroversial matter, we see, or at least presume, such humans to be persons with a morally significant interest in continuing to live, and we require a great deal to overcome that presumption.
Consider Fred, a human with late-stage dementia who, as much as any human can, lives in an eternal present. Fred’s sense of the future is very limited; indeed, very probably more limited than that of a healthy dog or mouse. But Fred is otherwise a very happy fellow who enjoys every second of his life although he does not anticipate his life beyond the next second. He is self-aware in the way that any being with perceptual awareness is.
I believe that it is uncontroversial to say that Fred is a person in that he has a morally significant interest in continuing to live. There is something it is like to be Fred. We regard that something as having moral value. We have an obligation that we owe to Fred to not use him as a resource for others. Fred is not a moral agent; he is, however, a moral patient. He may not have moral obligations to anyone, but we have moral obligations to him. The protection of his life is not a matter of our sufferance or our concern that, if we don’t protect Fred, we’ll be next. Even though we are not obligated to treat Fred in the same way as we would treat a normally functioning human (indeed, we have an obligation to not do so), we are obligated to treat Fred – and all sentient humans – the same as in having a right not to be used exclusively as a resource. That is the minimal content for the possession of moral value. Fred is not a thing. He is not a quasi-person whose life we can take as long as we kill him ‘humanely’. We assume that Fred values his life, however he values it, and we cannot second-guess him; we assume that he has an interest in continuing to live, even if he is unable to muster any sophisticated thinking about the matter.
Once we require that Fred have additional cognitive characteristics in order to be said to be a person, we necessarily end up in an arbitrary mess. Is there a morally relevant difference between Fred, who has no memory and no ability to plan for the future beyond the next second of his consciousness, and Sara, who has late-stage dementia but who is able to remember one minute in the past and plan for one minute into the future? Is Sara a person and Fred not a person? If the answer is that Fred is not a person but Sara is, then personhood apparently supervenes somewhere in the 59 seconds between Fred’s one second and Sara’s one minute. And when is that? After two seconds? Ten seconds? Forty-three seconds? If the answer is that neither is a person and that the connection with a future self requires a greater connection than one minute, then when, exactly, is the connection with a future self sufficient for personhood? Three hours? Twelve hours? One day? Three days?
Where humans are concerned, we recognise that sentience is sufficient for personhood – to have a morally significant interest in continuing to live. There is no reason – other than anthropocentrism – to treat the matter differently where nonhumans are concerned. Singer tells us that the premature death of cows is not a tragedy because ‘there is nothing that they hope to achieve’. That is just another way of saying that the lives of cows do not matter because they don’t aspire to do the things that philosophy professors do. This does not mean that sentient beings are equal for all purposes; it does, mean, however, that all sentient beings are equal for the purpose of not being used exclusively as a means to the ends of others.
If our conventional moral thinking about animals is to be meaningful at all, it needs to maintain that we can use and kill animals for human purposes only when there is a real conflict between human and nonhuman interests – a compulsion – that necessitates it, and, in such cases, we must take care to make sure that we impose the least amount of suffering on animals. This would mean that we would no longer routinely treat animals as chattel property and use them for all sorts of purposes that cannot plausibly be characterised as necessary. That is, even in the absence of recognising that animals are nonhuman persons with a morally significant right in continuing to live, which, as a proponent of animal rights I would urge, it would mean that we could not justify, among other things, eating animals or animal products in all but relatively a few situations.
And it’s not just meat that is a problem; there is no morally significant difference between meat on the one hand, and dairy and eggs on the other. All of these products involve suffering and death. Veganism is not an extreme position; what is extreme is claiming to believe that animals matter morally and then inflicting suffering on them for no reason other than culinary pleasure or convenience. It is also extreme to continue to ignore that, if we adopted a vegan diet, we could substantially reduce, if not end, world hunger, and take the single most significant step we can take as individuals to address the climate crisis.
Although I am sure that many readers will have various objections to what I have said here, I want to anticipate the one that I think will be the most prevalent: where do we draw the line? What about insects? What about plants? Are they all sentient? The answer is that lines in ethics are almost always hard to draw but, in this case, we can say with confidence that just about all of the animals we exploit as a matter of institutionalised practice – the mammals, the birds, the fish, the lobsters, the crabs, the octopi , etc – are sentient. We can start there and worry about refining the line later on.
As for plants, there is absolutely no evidence to date that they have any sort of minds that prefer, want or desire anything. Yes, they have certainly evolved biological processes that seek to assure their flourishing but no, they are not subjectively aware. If it turns out that plants are sentient, given that it takes many more units of plants to produce one unit of an animal product, we would still be obligated to choose to consume the plants directly if we did not conclude that we have an obligation to starve.
Gary Francione’s most recent book is Why Veganism Matters: The Moral Value of Animals (Columbia University Press, 2020).