A Carnivorous Rejoinder to Bruers and Erdös

Abstract

In an earlier paper, I defended the moral permissibility of eating meat against sentience-based arguments for moral vegetarianism. The crux of my argument was that sentience is not an intrinsically morally salient property, and that animals lack moral status because they lack a root (basic) capacity for rational agency. Accordingly, it is morally permissible to consume meat even if doing so is not strictly necessary for our nutrition. This paper responds to critiques of my argument by Bruers (J Agric Environ Ethics 28(4):705–717, 2015) and Erdös (J Agric Environ Ethics, 2015). I then show that their criticisms are easily dispatched and therefore fail to undermine my defense of meat consumption.

A number of philosophers have argued that moral vegetarianism follows from the seemingly uncontroversial principle that it is wrong to cause pain without a morally good reason.¹ Since meat consumption isn’t necessary for nutrition, it follows that most cases of commercial meat production result in unnecessary animal suffering. Therefore, it is wrong for us to support these practices by eating meat. Rachels (2004) has called this the “basic argument” for vegetarianism, since it relies on one modest premise that is accepted by nearly everyone.

This is an intuitively powerful argument. However in Hsiao (2015), I criticized the basic argument on the grounds that sentience is not, in fact, an intrinsically morally salient property, and that animals lack moral status because they lack a root capacity for rational agency. Accordingly, it is morally permissible to consume meat even if doing so is not strictly necessary for our nutrition. This paper responds to critiques of my argument by Bruers (2015) and Erdös (2015). I begin by briefly summarizing my original argument in Hsiao (2015). I then show that their criticisms are easily dispatched and therefore fail to undermine my defense of meat consumption.

Against Sentience

The central premise of the basic argument for vegetarianism is the claim that sentience is sufficient for moral status. That is to say, if a being is capable of feeling pain, then it has at least some degree of moral status. The justification for this thesis is simple: Pain is a moral evil because it harms the being who suffers. This is seemingly evident when we reflect upon our own experiences. As Beauchamp and Childress (2013: 73) summarize:

In its most basic form, the central line of argument… is the following: Pain is an evil, pleasure a good. To cause pain to any entity is to harm it. Many beings can experience pain and suffering. To harm these individuals is to wrong them. These harm-causing actions are morally prohibited unless one has moral reasons sufficient to justify them… We need look no further than ourselves to find this point convincing: Pain is an evil to each of us, and the intentional infliction of pain is a moral-bearing action, from the perspective of anyone so afflicted… What matters, with respect to pain, is not species membership or the complexity of intellectual or moral capacities, but the actual pain.

This argument has been disputed.² While pain experiences are no doubt harmful to the being who experiences them, this fact does not by itself show that sentience is harmful in a morally salient sense. A harm in its most broadest form is simply a setback to a being’s welfare, and here a distinction needs to be drawn between moral and non-moral harms. Not every violation of a welfare condition counts as a moral evil. If I peel and cut up a potato, I harm the plant by damaging its physical structure. If I pulverize my television with a baseball bat, I harm the television by destroying its ability to function properly. In each case I have caused harm by way of injuring or damaging the well-being of some entity, yet presumably neither harm is harm in a moral sense. Just as there are moral and non-moral senses of goodness and badness, there are likewise moral and non-moral senses of harm.

Hence, the mere fact that pain is harmful cannot be what makes the capacity to feel pain a morally salient property, otherwise this would renders all harms as moral harms. There must be some further fact about pain experiences that explains why their harm is morally significant. The reason why our own pain experiences are morally salient might have to do in part with their harming us, but this fact alone cannot provide a complete explanation. What we need, then, is a set of criteria by which we can distinguish between those welfare conditions that are relevant to moral standing and those that are not.

However, we cannot talk about status, standing, or membership in the moral community (or any community, for that matter) without some understanding of what this community is centered around. Suppose that I asked you to determine who ought to be invited into a newly-formed professional association of otolaryngologists. Unless you were to know something about what otolaryngology is, you wouldn’t be able to fulfill my request. It would be futile to assign members to a group, organization, or community without first knowing what the group is about, since without this knowledge one could not discern the relevant criteria for membership. The same is true when determining who count as members of the moral community. We cannot talk about who does and who doesn’t belong in the moral community without some idea of what morality is.

Any plausible theory of moral status will therefore need to show how the properties it regards as morally salient are conceptually linked to the concept of morality. In other words, we need to start with the concept of morality, from which we can then determine which welfare conditions are relevant in granting entry into the moral community. References to the moral community, after all, can only intelligible in light of some understanding of what morality is. We must then take a “top-down” approach in thinking about moral status. “Bottom-up” approaches, which construct a theory of moral status around certain properties regarded as intuitively morally salient, are inadequate because the candidate properties may in fact have nothing to do with morality, and hence membership in the moral community.³

Now in order to know who should count as members of any community, we need to know something about the nature or purpose of the community in question. Thus, in order to know who is in the moral community, we need to know what the nature or purpose of morality is. This provides us with a way of distinguishing morally salient properties from those that are not.

On this point, morality is fundamentally about free action in pursuit of the good. It is hard to see what else morality could be if not a code of conduct that exists for the sake of guiding purposeful action in pursuit of one’s flourishing. In order to do this—and thus, in order to be a member of the moral community—one must be rational and free. That is to say, one must be capable of knowing and acting for the sake of the good. Thus, Oderberg (2000a: 1) notes that “knowledge and action are the two essential objects of ethics, what the person who wants to ‘be moral’ or ‘act morally’ has to aim at.” These two factors, which constitute the capacity for rational agency, are what defines the moral community as moral. There may be other properties that are also morally salient, but whatever moral saliency they have will be derived from their contributions toward the flourishing of moral agents. For instance, sentience is morally salient only insofar as it is manifest in moral agents, since it adds another way in which they can do well or badly with respect to their welfare. In itself, however, sentience is not morally salient, since it is not a requirement for the exercise of rational agency. Neither is the mere possession of welfare conditions relevant, since these are possessed by non-rational beings. Not even self-consciousness simpliciter is morally salient, for it is only the capacity for certain type of cognition that is relevant to rational agency.

Animals very clearly lack the capacity for rational agency. This is evident when we consider the extremely limited nature of animal thought. Moreover, if animals really did possess rational agency, then at least some of them would be under moral duties. But our reluctance to attribute duties to animals shows that whatever intelligence animals do have, it is not of the kind relevant to rational agency. Therefore, we can conclude that animals are not members of the moral community. Since their suffering does not matter morally, it is morally permissible for us to consume meat. All humans, however, possess a root or basic capacity to reason, and therefore have moral standing.

This is only a brief summary of my main argument in Hsiao (2015). I now turn to objections raised by Bruers and Erdös.

Bruers on Sentience, Rational Agency, and Moral Communities

In his response paper, Stijn Bruers makes three main objections to my original argument. First, he argues that I have not shown why moral status should be defined in terms of rational agency as opposed to some other property. Second, he points out that there are several senses of rational agency, and that it is not clear why my conception of rational agency should be preferred over other accounts. Third, he argues that the essentialism on which my argument is based has been disproven by evolutionary theory. I shall consider the third objection in another section, as it is also endorsed by Erdös.

As for the first objection, Bruers argues that my definition of moral status is arbitrarily asserted. Although he grants that there is some intuitive force motivating the view I put forth, he asserts that I have not given a reason beyond a mere appeal to intuition as to why we should accept it. But this is simply false. I gave an argument, which I briefly summarized in the previous section, for the claim that moral status must be defined in terms of rational agency and rational agency alone. This argument was based not on a mere intuition, but on the theoretical point that morality is about action in pursuit of the good, and that therefore we cannot call anything “moral” unless it bears some relation to rational agency. This also answers Bruers’s charge that I have given no reason to think that moral patiency should be defined in terms of rational agency: If I am right that the concept of moral status as such is inherently connected to rational agency, then it follows that moral patients must possess in some sense the capacity for rational agency.⁴ Indeed, the whole point of my section titled “How to Think About Moral Status” was to argue that any being with moral status—whether it be a moral patient or a moral agent—must possess the capacity (root or developed) for rational agency. It is surprising that Bruers does not attempt to engage with this key argument anywhere in his paper, even though it formed the basis of my entire paper.

Similarly, Bruers also argues that I have not established why the root capacity for rational agency is relevant over, say, a more developed version of this capacity. But again, I explicitly gave an argument for this. I pointed out that we must appeal to root capacities in order to explain why we preserve moral status through certain temporary changes during which we lose the ability to express certain cognitive capacities. Bruers does not engage with this argument.

What I have just said in response to Bruers’s first objection also provides us an answer to his second objection. Bruers points out that the concept of rational agency can be interpreted in several different ways, and that I did not give a reason to accept the specific account of rational agency that I put forth. He is correct in noting that rational agency can be analyzed in several ways, but his claim that I did not justify my account of rational agency is mistaken. The same argument I gave for defining the moral community in terms of rational agency also tells us which sense of rational agency is morally relevant: namely, the capacity for conceptual thought.

How does this follow? The answer is simple: Since morality is a code of conduct that guides activity in pursuit of the good life, it follows that members of the moral community must have among other things the cognitive power to understand what is good, why it must be pursued, and that they are doing it. Thus I pointed out that:

The capacity for rational agency is both necessary and sufficient for having moral status, for morality is essentially about action in pursuit of one’s flourishing. Now in order to pursue something, one needs to be capable of knowing what it is he is pursuing and that he is doing it. To pursue something is to aim for it as an end, and one cannot formulate a plan of action if he cannot know what he is aiming at. Moral subjects must therefore be capable of knowing, understanding, deliberating, choosing, and acting for the sake of the good. Furthermore, since morality is about the pursuit of the good, and because the good is species-specific, a moral subject must also have the further ability to know his own good and the good of others in the moral community. This in turn requires that a moral subject have the ability to have at least partial knowledge of his nature and the nature of others like him, which requires that he possess an intellect capable of grasping the essential nature of things and abstracting it as something held in common by many. This intellect must be capable of classifying, generalizing, and recombining concepts in order to reach new insights, for cognition of this sort is necessary to internalize the good as a reason for action.⁵

Perhaps a less ambiguous term for the relevant sense of rational agency is moral agency. Members of the moral community must be rational in the sense of being able to act for moral reasons. I made this point explicitly clear when I argued that if some animals were capable of rational agency, then those animals would be under moral duties. But since we do not attribute moral duties to animals, this shows that in whatever sense animals may be intelligent, it is not the kind of intelligence that is pertinent to morality or moral status.

Finally, at several points in his paper, Bruers points out that my skepticism of appeals to intuitions is inconsistent with some of the arguments that I employ, for they must at some point appeal to intuitions. This is not the case. My comments about the use of intuitions should not be read as rejecting the use of intuitions, which Bruers suggests. Rather, my comments were aimed at criticizing a particular way of theorizing about moral status, one that constructs a theory of moral status around certain properties that are considered intuitively morally salient from the outset.⁶ Rather than theorizing about moral status from a “bottom-up” approach, we need to adopt a “top-down” approach that begins with a theory of morality from which we can then use to delineate the conditions for membership in the moral community.

Sentience, Preferences, and Moral Status: No Connection

One of the key points in my argument against vegetarianism is that there is no conceptual connection between sentience and moral status. Possession of the ability to feel pain certainly adds a new way in which a being’s life may go better or worse, but the same is true for any other welfare condition, and not all welfare conditions matter morally. The challenge for those who think that sentience is sufficient for moral standing is to show why pain matters in a way that other welfare conditions do not.

In response, Bruers proposes that we should understand moral standing and rights-possession in terms of preferences. Inflicting pain upon a being, we might reason, violates its preferences because pain is unwanted. Therefore, sentience is sufficient for moral status because sentience is evidence of preferences, and preferences are what matter morally. As he puts it:

You value what you want. What you want is important to you. If you want it very much, it might become so important that you would want it to be protected by a basic right. Hence, important preferences are the basis for basic rights.

Why does a preference matter in a way that other conditions don’t? Why are preferences important? By definition, a preference means that there is something that matters to someone, that there is something that is important to someone. With a preference, there is someone who wants or likes something. Without subjective preferences, nothing would matter to anyone.⁷

It is not clear, however, how this argument shows that preferences matter morally. Indeed, it seems as if the same argument that I offered against sentience could simply be reiterated here, only replacing sentience with preferences. It is quite true that preferences matter to the preference-holder. But how is this any different any other welfare condition? Having an adequate source of water certainly matters to a plant’s well-being, having a nutrient rich environment certainly matters to a bacterium’s well-being, and having a functional compressor certainly matters to an air conditioner’s well-being. These things all matter, and they all matter because they are required for each being to flourish. But do they matter morally? Surely not. So what is it that distinguishes preferences from these other welfare conditions? Bruers’s answer is that a preference signals something of importance to the preference-holder, but this again is also true of all three examples just mentioned: Water is important to a plant; a compressor is important to an air conditioning unit; nutrients are important to a bacterium. Perhaps they might not be able to consciously want these things, but surely they still matter because they are necessary for their flourishing. Consequently, I see no reason at all for accepting Bruers’s assertion that “[w]ithout subjective preferences, nothing would matter to anyone.” Why should we think that this is the only relevant sense in which something can matter? Bruers needs an argument for this, and he does not supply one. Given the previous examples, this thesis seems clearly false.

Now to be sure, having preferences makes a preference-holder’s life more rich and complex by adding more welfare conditions. But this has nothing to do with moral status. What matters is not the number of welfare conditions a being has, but whether those welfare conditions are such that they are relevant to the nature of morality. It is only a certain type of welfare condition—those connected to rational agency, which matter morally. The argument for this was given at length in my original paper, and was summarized above.

Bruers argues that denying that preferences matter morally results in a performative contradiction: “What you want is morally important. If you disagree with this, you shoot [yourself] in your own foot, like saying that what you want is not what you want.”⁸ This is not the case. What I want is morally important not because it is wanted, but because I am a rational agent, and it is my being rational that renders my preferences morally salient (the arguments for this were, again, stated earlier and in the original paper). There is only a performative contradiction if one affirms that preferences are morally important because they are preferences (i.e. preferences are inherently morally salient), but Bruers has given no reason to regard this as true. If the moral importance of preferences depends instead on some further fact about the being who has them, then one can non-arbitrarily affirm that human preferences matter while denying that animal preferences matter.

The fundamental problem with Bruers’s response paper is that he does not engage with my main arguments. His claim that I arbitrarily asserted without justification that rationality is necessary for moral status, for example, is flatly mistaken. I provided at length an argument showing (a) why moral status must be defined in terms of rationality (and therefore why moral patiency requires this capacity), and (b) why a root capacity for rationality is sufficient for meeting this requirement. Bruers does not address these arguments head on. Indeed, many of his objections were already preempted in my original paper.

Erdös and Bruers on Essentialism, Root Capacities, and Darwinian Evolution

Both Bruers and Erdös argue that the essentialism on which my arguments are based is inconsistent with evolutionary theory. Erdös makes this objection the key focus of his response paper. Specifically, he takes aim at the essentialist response to the argument from marginal cases, according to which humans qualify as members of the moral community because they are all members of a rational kind.⁹ Certain humans may not be able to actually express or exercise the ability to reason, but the ability to reason is nevertheless present in the form of a root or basic capacity, one that is rooted in their species-membership. This idea, according to Erdös, is “biological nonsense. It cannot be verified, and it only obscures reality.”

The charge that essentialism has been falsified by evolutionary theory is a common one. Although neither Bruers and Erdös explain in much detail why evolution poses a problem for essentialism, the standard Darwinian critique of essentialism has focused on two points. First is the non-constancy of species: Because organisms can give rise to radically different organisms, this allegedly shows that are no such things as fixed essences or natures. As Hull (1989) puts it:

If species evolve in anything like the way that Darwin thought they did, then they cannot possibly have the sort of natures that traditional philosophers claimed they did. If species in general lack natures, then so does Homo Sapiens as a biological species. If Homo Sapiens lacks a nature, then no reference to biology can be made to support one’s claims about ‘human nature’… Because so many moral, ethical, and political theories depend on some notion or other of human nature, Darwin’s theory brought into question all these theories.

This criticism is easily dispatched. Oderberg (2007: 2004) points out that it is an “elementary mistake to think that fixed essences exclude substantial change.” The essentialist is not committed to the proposition that an organism that possesses a fixed essence cannot undergo substantial change to become a new organism nor that it cannot give rise to another organism with a different essence. It would indeed be impossible for an essence to change into another essence, but this is not what occurs in evolutionary change (or in any kind of change, for that matter). When evolution produces a new species from an existing species, the essence of the original species does not change. Rather, its descendants simply cease to instantiate the essence of its ancestors and take on a new one instead. This is wholly compatible with the essence of some species being fixed and immutable. What changes isn’t the essence, but the material that exemplifies the essence.

Consider chemical change. When glucose is converted into adenosine triphosphate (ATP), the matter that once instantiated the essence of glucose ceased to do so and instead took on the essence of ATP. The essence of glucose remained unchanged throughout the process while the material that once instantiated it took on a new essence. If substantial change of this sort poses no difficulty for chemical essentialism, then neither does evolutionary change pose a difficulty for biological essentialism.¹⁰ There is nothing in essentialism that says that something with an essence cannot be destroyed or changed into something else.¹¹

The second part of the evolutionary criticism of essentialism has to do with vagueness between species. That is to say, the “messiness” of classifying species should lead us to be skeptical of the ideas that there are any species to begin with. This objection too is exaggerated. Worries of vagueness are present not only in biology, but in physics, chemistry, and just about every other discipline. Change between elements is quite common, and it is not immediately clear why this poses a problem for biological essentialism when chemists and physicists seem quite content in recognizing distinct non-arbitrary essences despite the fact that many of them are barely distinguishable and have very similar properties. Given that there are independent reasons to believe in the existence of essences,¹² worries about vagueness need not prompt skepticism over their reality.

On the topic of essentialism, Bruers and Erdös seem unaware of the “new” essentialists, who have defended metaphysically robust versions of essentialism (see for example Ellis 2002; Elder 2004; Fine 2005; Hacker 2007; Oderberg 2007; Feser 2014).¹³ Likewise, Thompson (1995), Foot (2001), and Alexander (2012) have all argued that teleology and natural kinds are crucial to our understanding of morality.¹⁴

Erdös’s charge that the idea of a root capacity for rationality is something that I “invented” reflects his unfamiliarity with the relevant literature. The term “root capacity” may have been mine, but the concept expressed by this term is one that is well-established in the literature on moral status (see for example Schwartz 1990; Moreland and Rae 2000; Oderberg 2000a; Beckwith 2007; George and Tollefsen 2008; Lee and George 2008; DiSilvestro 2010; Kaczor 2011).

Finally, Erdös’s complaint that the root capacity for rationality is “non-verifiable” is misguided. Capacities, powers, dispositions, and essences are not material things, and so of course they cannot be empirically or scientifically verified. But why exactly is this supposed to be a problem? If the claim is that we should only believe in those entities whose existence can be scientifically verified, then it is patently self-refuting. The principle that we should only believe in what can be scientifically verified is not itself scientifically verifiable!¹⁵ This is no different from the verificationist principle espoused by logical positivism, which philosophers have long since abandoned. Erdös may be writing as a biologist and not a philosopher, but this does not excuse him for making such a philosophically sloppy argument.

But perhaps there is a further epistemic objection lurking behind this: Even if we grant the existence of a root capacity for rationality, how could we know who has it? Since we can’t directly observe our essence, how could we ever know its features?

The answer to this worry was already suggested in my original paper: We come to know what something’s essence is through observing its actions.¹⁶ That is, we know what kind of being/entity something is by seeing how its powers, capacities, and dispositions are manifested. After all, this how we know that many varieties of nonhuman animals are sentient beings, even if we haven’t observed every single instance. So Bruers and Erdös must themselves be committed to the knowability of essences, lest they shoot themselves in the foot. And indeed, unless essences are knowable by us, the discipline of medicine would make no sense. Medicine is concerned with restoring bodily functions to what they should be, a task that is only possible if we can reliably understand our human nature and how it should be expressed. Accordingly, we know that all humans have a root capacity to reason because we understand that this capacity is essential to being human. This explains why we recognize that “marginal” humans are in some sense immature or disabled with respect to the developed capacity to reason.

But let’s suppose that Erdös is right that the root capacity to reason is “biological nonsense.” What follows? Not much, actually. My core argument, which is that moral status as such is based on the ability to reason, would remain untouched. All that Erdös would have shown is that moral status isn’t based on the root capacity for reason, not that moral status isn’t based on the capacity to reason as such, for it could still be the case that moral standing is based on some other version of the capacity to reason (such as the developed capacity). In other words, Erdös would have only shown that a particular version of the rational agency requirement is mistaken. So even if Erdös is right, nothing of significance follows. He does not respond to my critique of sentiocentrism, nor does he show why the rational agency requirement as such is mistaken.

Conclusion

According to Erdös, the “greatest error” of my paper is that I defend “anthropocentrism, the view that has presumably been the very cause of the spoiled non-human–human relationships and today’s serious environmental destructions.” He condescendingly suggests that I wrote my paper in order to “quiet” my conscience and to rationalize my practice of eating meat.

If anything, the opposite is true. I regard the view that animals possess rights as sheer madness in deep need of correction. As much as I do enjoy the taste of a savory medium-rare filet mignon, my goal in writing was not, as Erdös would like to think, to desperately justify my habit of meat consumption against the pangs of a guilty conscience, but to combat a profoundly mistaken way of thinking that has ruinous consequences for human well-being if taken to its logical conclusions.

I am therefore in full agreement with Carruthers (2011: 401), who has said that “the increasing moral importance accorded to animals in our culture can be seen as a form of creeping moral corruption and should be resisted.” I am proud to call myself an anthropocentrist.