Notes on Predators and Animal Ethics


Dan Hicks


January 13, 2013

In this post, I’ll be discussing one of the common arguments against (moral) vegetarianism. (Moral vegetarianism is the view that vegetarianism is morally required, or that eating meat is morally impermissible. I’ll just use “vegetarianism” here.) I’ve encountered it every single time I’ve taught an argument for vegetarianism; at a few conversations at my co-op; and even at a job interview. I’ll argue that the force of the argument depends a great deal on how the proponent of vegetarianism has argued for her position.

The argument from predators is, I think, a reductio of vegetarianism:

  1. Assume for contradiction: Eating meat is morally impermissible.
  2. Hence, eating meat is morally impermissible for non-human predators, such as lions and wolves.
  3. But eating meat is not morally impermissible for non-human predators.
  4. Hence, eating meat is not morally impermissible.

Step (2) follows from premise (1) by instantiating an implicit universal quantifier. Step (3), I think, is typically taken to be obvious; but the anti-vegetarian might argue from obligate carnivores such as housecats, or wild predators such as wolves and mountain lions. Given the contradiction between (2) and (3), it follows that (1) cannot be true.

(N.B. I’m simply going to ignore science fiction proposals to genetically engineer predators into peaceful herbivores.)

Again, my claim is that the dialectical force of this argument – how compelling an objection it is to vegetarianism – depends a great deal on how the proponent of vegetarianism has argued for premise (1). I’ll discuss arguments for vegetarianism from four different families of normative ethics.

Animal Rights (Deontology)

Rights-based arguments for vegetarianism are exemplified by Tom Regan’s work; Evelyn Pluhar has another, rather different sort of argument for animal rights in her contribution to Sapontzis’ anthology Food for Thought: The Debate over Eating Meat. For proponents of a rights approach, the argument for vegetarianism proceeds from the claims that (a) animals have rights and (b) eating them violates these rights. Since it’s morally impermissible to violate any agent’s rights, it’s impermissible to eat animals.

What appears to be the standard reply among proponents of animal rights to the argument from predators is to distinguish between moral agents and moral patients. Both moral agents and moral patients have rights; but only moral agents have obligations to respect rights. Or, moral patients cannot violate the rights of other moral patients (or of moral agents).

So, human beings are moral agents, and we violate the rights of pigs and cattle when we kill and eat them. However, pigs and cattle and cats and wolves and deer and so on are all moral patients, not moral agents. So the wolves do not violate the rights of the deer when the former kill and eat the latter. Hence, in the argument from predators, (1) does not actually imply (2). Indeed, (1) as given is ambiguous. Stated more precisely, the vegetarian’s claim is that eating meat is morally impermissible for moral agents. Since non-human predators – at least the ones we know about – are not moral agents, this premise implies nothing at all about their eating habits.

I find this response to the argument unsatisfying for three reasons. First, in generally I dislike appeals to rights. Rights approaches – both in animal ethics and more generally – seem to spend far too much time on whether the beings in question have rights (and why this is the case), and almost none on exactly which rights they have. Given that my cat has rights because he is a subjects-of-a-life, do I violate his rights by not letting him out of my apartment? Or, given that, when I let him out, he chases, catches, and kills the mice and chipmunks that leave in the bushes, do I (as the moral agent who’s causally responsible here) violate the rights of the mice and chipmunks? I don’t see how a rights approach can even begin to answer these question.

A rights-based vegetarian might respond that their account does at least clearly support claims (a) and (b), which together with the general notion of rights is enough to imply vegetarianism. So, even if a rights-based approach can’t answer all of our questions about animal ethics, it can at least answer this one.

Second, I find Val Plumwood’s critique of Regan’s view quite compelling:

Regan attempts to meet this objection by claiming that since the wolf is not itself a moral agent (although it is a moral patient), it cannot violate the sheep’s rights not to suffer a painful and violent death (Regan 1986, 285). But the defense is unconvincing, because even if we concede that the wolf is not a moral agent, it still does not follow that on a rights view we are not obliged to intervene. From the fact that the wolf is not a moral agent it only follows that it is not responsible for violating the sheep’s rights, not that they are not violated or that others do not have an obligation (according to the rights view) to intervene. If the wolf were attacking a human baby, it would hardly do as a defense in that case to claim that one did not have a duty to intervene because the wolf was not a moral agent. But on Regan’s view the baby and the sheep do have something like the same rights. So we do have a duty, it seems, (on the rights view) to intervene to protect the sheep-leaving us where with the wolf? (ibid., 8)

That is, the wolf no more violates anyones rights when stalking and eating a human than when stalking and eating a deer. But then a rights approach seems either to conclude that nothing wrong happens in either case – which I take it to be clearly incorrect – or to fail to provide practical guidance – which, after all, is what ethics should do.

Note the similarity between my first objection and Plumwood’s objection. We both consider scenarios – not vegetarian scenarios, though predation scenarios – in which the rights-based approach does not seem to provide practical guidance. So I suppose the proponent of animal rights could respond in the same way as above, namely, that their account doesn’t claim to speak to all of these issues. But then we might begin to suspect that the rights-based approach can only answer very few of our animal ethics questions.

Third, the distinction between moral agents and patients strikes me as ad hoc, and perhaps even self-destructive. Natural rights approaches get much of their appeal from the idea that all persons are as such equal – that distinctions between lords and serfs, men and women, whites and blacks, humans and animals, are morally irrelevant. If animals are (mere) moral patients, while we are (mighty) moral agents, then why shouldn’t we sacrifice the interests of animals when they conflict with our own? This sort of slide might be especially attractive when we consider Plumwood’s child-stalking wolf.


Utilitarian arguments are probably the most familiar family of arguments for vegetarianism, thanks to Peter Singer. In contrast with rights-based approach (human and animal), utilitarianism is context-sensitive, that is, it can support different normative claims for different individuals in different situations. (Different versions of utilitarianism will be differently context-sensitive, of course: act utilitarianism will be much more sensitive to context than rule utilitarianism.)

Context sensitivity gives at least some versions of utilitarianism enough room to separate (contemporary) humans (in industrialized societies) from non-human predators. For contemporary humans in industrialized societies, the meat we might eat would be produced in factory farms, inflicting great suffering on the animals involved, and providing us with only a small amount of pleasure; assuming vegetarianism for us would reduce this suffering (and perhaps be just as pleasurable), it follows that eating meat is morally impermissible. On the other hand, if non-human predators refrained from eating meat, then they would die out and herbivore populations would skyrocket, leading to significantly greater suffering than under what we might call ordinary moderate predation. So then ordinary moderate predation, not vegetarianism, would be morally required for predators.

Note that utilitarianism can separate us from non-human predators without resorting to questionable distinctions, like the moral agents / moral patients distinction. The rights-based approach responded to the argument from predators by claiming that different rules apply to different kinds of beings – this led to my third criticism, that this response undermined the appeal of the approach. Utilitarianism claims that the same rule – maximize net pleasure – applies to all beings, but that the implies of the rule are different in different situations.

Because of its context sensitivity, the utilitarian response depends very much on its empirical, causal claims: that vegetarianism for us and ordinary moderate predation will reduce net suffering, compared to their respective alternatives. First, it’s not clear that decades of moral exhortation encouraging individuals to become vegetarian has been at all effective. Measured in pounds per capita, US meat consumption increased about 26% between 1960 and 2007. It has decreased about 12% since that high point – but this seems to be largely due to the Great Recession. On the other hand, it’s true that the average dressed weight of cattle has increased over that same time, by about 33% (see the third graphic down on that page). And the number of cattle is currently at about 30 million, down from a break of about 45 million in the 1970s; though this is still a net increase compared to the 25 million or so in 1960 (see the second graphic on that same page). And in pounds per capita we eat about three times as much chicken as we did in 1960.

Second, agroecological practices clearly involve much less animal suffering than industrial meat production; by sustainably managing agroecosystems and their required resources (in some sense of “sustainable”) would probably maintain higher levels of net pleasure over time; and arguably require eating some of the animals involved. So a utopian utilitarian argument – see my explanation of “utopian” in the previous link – might conclude that eating meat is morally required.

Environmentalist Ethics

Roughly, ecocentric environmentalism is the view that functionally integrated biological systems have intrinsic ethical significance, and so there are at least some defeasible reasons to promote and protect such systems. There are numerous terms in that definition that really need further explication, but I’m going to take it to be adequate for my purposes here. The standard-bearers for ecocentric environmentalism, as I understand it, are Aldo Leopold and J. Baird Callicott; one might also include Arne Næss’s Deep Ecology, though I do prefer Leopold’s (and Baird’s) concrete, rather American, Land Ethic to Næss’ mysticism.

A somewhat different approach to environmental ethics is agrarianism, as exemplified by Wendell Berry and Paul Thompson (and actually Leopold, notably). Agrarianism is closely tied to community- and place-based agricultural practices and a conception of stewardship of the land, and may not place much value on “wilderness.” Specifically, agrarians are much more likely than ecocentrics to condone human management of/coproduction with/meddling in the extra-human parts of the world.

Both ecocentrics and agrarians are deeply opposed to CAFOs and other methods of intensive, “factory” farming. Ecocentrics are probably more likely than agrarians to emphasize the pollution and environmental destruction produced by CAFOs; agrarians are probably more likely to emphasize the damage intensive farming does to agricultural communities, both cultural (as economic values replace stewardship and community) and environmental (the harm suffered by humans and animals from pollution and overexploitation of the land).

Furthermore, both ecocentrics and agrarians will have complex attitudes towards more traditional, less intensive forms of animal agriculture. In ecological terms, both tend to think of humans as keystone species in our food webs, managing the populations of other species. Agrarians will see traditional forms of animal agriculture as paradigmatic of a well-functioning, human-managed food web, and note that animal husbandry is traditionally one aspect of stewardship. So, insofar as stewardship should be promoted; and stewardship involves raising, working with, and then consuming animals; it seems to follow that consuming animals should be promoted. However, ecocentrics might argue that even traditional animal agriculture is harmful and disruptive to functionally integrated biological systems – the difference between CAFOs and peasant agriculture is one of degree, not kind — and we should return to a hunter-gatherer lifestyle. Or maybe just a gatherer, vegetarian lifestyle? I don’t think either ecocentrism or agrarianism points decisively either towards or away from vegetarianism.

Ecocentrics will almost certain condone predation as the way in which a functionally integrated ecosystem maintains itself – keeps the deer from reproduction out of control and crashing the system, in other words. While agrarians in the past may have condoned killing predators, either systematically or opportunistically, the grounds would have been the threat predators posed to humans. That is, we need to kill the wolves to protect our flocks and herds. However, I think contemporary agrarians are more inclined to see human agroecosystems as integrated parts of broader systems, such as watershed systems. Such systems will incorporate, for example, agricultural fields, woodlands, and wetlands, and predators will be essential to keep both the whole system and its various parts functioning. For example, predation by owls and other birds of prey might be necessary to keep mice populations under control in the bushes on the border between the forest and fields; if mice populations get out of control, they may eat the seeds in the fields.

Care Ethics

Care ethics emphasizes particular, personal relationships and dependency; it has its roots in work on the relationship between parents, especially mothers, and children, but has been extended to animals by Josephine Donovan, Carol Adams, and Lori Gruen. A simple (oversimplified) care argument for vegetarianism might run as follows: since (a) compassion in general and within dependent-caregiver relations specifically is a crucial virtue, (b) the animals that we raise for food are dependent on us, and (c) eating someone is incompatible with having compassion for them, care ethics would recommend vegetarianism as virtuous.

Now we can bring in the argument from predators. Predators depend on their prey; so then they should show compassion to their prey; and so they should be vegetarians, just like us.

Care ethics is a virtue ethics, rather than a rule ethics. The recommendations offered by care ethics are always defeasible and sensitive to particularities, by contrast with the universal and absolute prohibitions of rights-based approaches and utilitarianism. So it seems to me that care ethics can say that compassion can be qualified by nutritional needs: insofar as predators need to eat their prey to survive, carnivorism is not a vice. However, obligatory carnivorism can be tempered by other aspects of compassion. Bringing a chipmunk home to torture for a while, before unnecessarily killing it, is vicious, whether this is done by a human or a housecat or a mountain lion. Thus, it is not the case that eating meat is morally impermissible; but rather that eating meat without compassion or respect for the animal who is eaten is vicious.

But what about other kinds of needs? My cat, for example, has no nutritional need to stalk and torture chipmunks. But it does seem like these activities are required for his flourishing – for him to lead a good life for a cat. He certainly has substitutes: toy mice (and other things that he thinks are good to bat around, such as the pen I’m trying to write with), and watching the chipmunks through the window. But it’s quite clear after more than a few days indoors – whether this is because I haven’t been home or because he refuses to walk on snow – that he’s quite frustrated at being confined, and especially at not being able to hunt.

I suspect that many care ethicists would recognize the existence of moral dilemmas: situations in which we have no morally permissible options, or in which we cannot exemplify all of the virtues at once. In the case of my cat, I must make a choice between showing compassion for him – letting him outside to hunt chipmunks – or showing compassion for the chipmunks – keeping him inside, no matter how frustrating this is for him. I cannot exercise compassion towards both my cat and the chipmunks, and so there is no completely virtuous option.

If that is right, then the best way to proceed would be to look for options that don’t require completely sacrificing both virtues. That is, options where (a) the cat can hunt and (b) the chipmunks don’t suffer too badly. For a while, I considered putting a bell on his collar. But he likes to crawl under the chain-link fence in the next yard, and lost two collars in the first month that he lived with me. (The jingling when he’s playing inside at night would also keep me awake.) Still, I can keep him from bringing chipmunks back inside; put them back outside when he does manage to bring them in (or has caught them inside our apartment); and scare him into dropping a chipmunk when I see through the window that he’s caught one. Also, at least while it’s cold, he’s not interested in being outside for more than about half an hour a day, and this doesn’t seem like enough time for him to do too much to the poor chipmunks. None of this completely resolves the dilemma; but it does reduce some of the tension. (Thanks to several people who helped me think through this section up to this point: Amelia Hicks, Daniel Immerman, Emily Dawson, Samantha Noll, Kathryn Pogin, and a few others whose names I cannot recall.)

As a final aside, care ethics also seems relevant to recent controversies over “happy meat.” Proponents of “happy meat” argue against vegetarianism on the grounds that there is little or nothing wrong with eating animals who led decent lives before being humanely slaughtered. “Happy meat” arguments are perhaps typically responses to utilitarian arguments for vegetarianism: we’re not talking about the egregious, lifelong suffering of animals in CAFOs, but instead a moment of suffering at the end of a very pleasant existence in a natural, agroecological setting. But these arguments can also be cast in terms of compassion, and so in terms of care ethics: “happy meat” comes from animals whom we care for properly during their lives and at slaughter. Indeed, proponents of urban farms sometimes maintain that we should only eat meat if we are willing to look the animal in the eye as we kill her or him.

Vegetarians generally respond to such arguments with horror, as the words of a psychopath. How could you claim to both care about an animal and slaughter it? I suspect, however, that the proponents of “happy meat” and vegetarians are working with rather different conceptions of compassion and care, and so talk past each other in these debates. Vegetarians assume that compassion is incompatible with killing and eating. Proponents of “happy meat” disagree. I think they think that compassion requires treating these animals with respect and dignity, but that it’s possible to do so while killing them. For example, a painless, terror-free, quick death that acknowledges the significance of sacrificing their life for ours – looking into their eyes – might seem respectful and dignified.

I can see how this kind of treatment treats the animal with dignity, and maybe even respect. But I don’t think it’s compassionate. So the disagreement I have with “happy meat” seems to come down to our rival understandings of compassion.