A Naturalistic Defense of
“Human Only” Moral Subjects

DILEMATA año 4 (2012), nº 9, 69-73

ISSN 1989-7022

Antoni Gomila

Departament de Psicologia
Universitat de les Illes Balears

Received: 04/05/2012
Accepted: 16/05/2012

The target of Rowlands’ argument in this paper is the Kantian notion of a moral subject: as a self-determined, reflexive, rational, being. He proposes, instead, a naturalistic view of moral subjecthood, along with the “moral sense” tradition, which drops the reflexive, deliberative, requirement of the Kantian tradition; according to this naturalistic approach, the limits of the moral community are wider than the set of rational agents: if all that is required to be a moral subject is to exhibit some kind of moral sensitivity, beings motivated by moral emotions qualify. While I agree with this general position, I disagree with the way Rowlands chooses to defend it.

In my view, the best way to deal with the Kantian view of moral subjecthood, is to make clear that it sets up such a high standard that just a few human beings, and just in a few occasions, qualify as moral subjects in the fullest sense; therefore, there is no way to make coincide the class of moral subjects with the class of human beings only. Such a stringent view as the Kantian not only excludes animals, but also many human beings (babies, handicapped, mentally ill,... even normal people when behaving without full deliberation and reflection); while any move to include all human beings in the class of moral subject is doomed to include some other animals as well (Gomila, 1996). In my view, we need rather to acknowledge that the concept of moral subjecthood is fuzzy, so that any attempt to provide its necessary and sufficient conditions is doomed, either to exclude some human beings, or to include some non-human beings.

But Rowlands is still trying to develop an analytical project -even if there is an amalgam of conceptual and empirical arguments in his paper. In this comment, I want to touch briefly on the two main aspects of his argument: the way he tries to undermine the Kantian approach; and the way he tries to justify his view of moral subjecthood.

As regards Rowlands’ critical side, I find his rejection of the reflexive condition both unjustified and unnecessary. First, unjustified, because it is misled by an excessive analogy with perception, which does not carries through, in my opinion, to reflection and deliberation –in fact, which overlooks all the complexities and conflicts of a full moral life. I find deeply problematic to suggest that morality might be the output of some kind of “blind” mechanism (modular, sub-personal, or even at some point, situationist), as it makes sense for perception, because this suggestion is committed to a counterintuive view of our moral psychology, with no place for our sense of duty, the pull of norms, the understanding of value concepts, or the conflict between different reasons: for the subjective dimension of morality. Morality is a personal-level phenomenon, even for the Mishkin case: it is misleading to describe this case as if there is no subjective point of view involved. In the same vein, for animals to be considered moral subjects, they should qualify as subjects in the first place.

Second, Rowlands’ rejection is also unnecessary. In my view, there is no need, in order to go beyond the Kantian view, to cast doubt on deliberation and reflection per se, as a sufficient condition for moral subjecthood, as Rowlands tries to do. It is enough to claim that it is not a necessary condition, that the highest standard of morality is not to be confused with morality per se (just as the class of artworks includes as a, rather small, subset the class of great artworks).

On the other hand, I’m also critical of Rowlands’ way to carry out the programmatic part of his argument: that morality is a matter of being motivated by moral emotions. I disagree with his cognitivist view of emotions and also with his notion of moral emotions.

Let’s start with the latter. In talking of moral emotions, Rowlands is not thinking of moral emotions strictu senso: shame, guilt, embarrassment, indignation,... –what Strawson (1974) called the reactive attitudes, and which he saw as constitutive of the involved attitude towards others to which morality, just as responsibility attribution and normativity belongs. I feel guilty when I realize that I’m responsible of an offence; and I feel indignant about somebody else’s action as long as I assume that he is responsible for what he did to me, and that it was wrong. It is clear to me that a cognitivist view of emotions is adequate of these kind of emotions: that they are a kind of propositional attitude, that they involve an intentional content. However, it is not equally clear that these sort of emotions are available to non-human primates, just as it is clear that they are not to human babies: they appear in development around seven-eight year of age (Harris, 1989).

Rowlands focuses instead on “emotions with moral charge (or content)”, such as compassion, sympathy, and tolerance, and their negative counterparts, such as jealousy, maliciousness and resentment (resentment is generally included in the moral emotions, but I will leave aside its proper characterization, given that it is a collateral question to the discussion). But it is highly dubious that these kind of emotions involved propositional attitudes, as Rowlands contends –that they involve intentional states with propositional content. Sometimes they may, of course; what I challenge is the assumption that they always do –as a cognitivist view of emotions requires and Rowlands explicitly asserts. For it is also possible that emotions just take individuals as objects, rather than propositional ones. Thus, it is possible for the zebra to fear the lion, or for María to love John, or for little Mary to feel sympathy for her sister, or for my dog to be jealous of my cat.

The importance of the latter for our discussion is double. On the one hand, understood this way, it is not so clear whether these emotions involve per se any moral content (any kind of moral predicate, or norm), and therefore they come short of a proper grounding of moral subjecthood: it is only when they properly take a propositional content as intentional object. On the other, from the presence of these emotions, and the conceptual skills that they involve, in animals, it does not follow that these animals are also able to hold propositional contents –necessary to have these emotions with proper intentional content. Thus, the inference Rowlands makes “if animals may hold contents –if they may think– then they may hold propositions”, is a non-sequitur. Put it conversely: I agree with Rowlands that morality requires propositional thinking, but I disagree that the sort of emotions he spots require propositional thinking, in contrast with moral emotions strictu senso. On the contrary, it seems to me that there are plenty of independent reasons to conclude that propositional thinking comes with linguistic competence (Gomila, 2012) –not on the grounds of the direct identification of language and thought, à la Davidson, but on the grounds of empirical evidence that suggests different kinds of psychological contents, propositional ones being just one of them. While emotions have older phylogenetic roots, at least in mammals, as a fast system of appraisal and orientation –which get transformed when higher cognition is in place.

Therefore, I conclude that moral subjecthood is open just to linguistic beings, because they are the ones capable of holding the sort of intentional moral contents in question. Just being able of the kind of emotions Rowlands focuses on is not enough, then, to qualify as a moral subject; it is just when these emotions take propositional contents as their intentional objects, involving the tokening of moral concepts, that morality appears –in a continuum of possibilities, deliberation being just the most sophisticated form.

The moral, in my view, is that it is important, for naturalistic approaches to morality, to pay due attention to the natural differences that matter for moral subjecthood, just as to the complexities of our moral psychology. Otherwise, the hope to overcome the Kantian view is dim. While I agree with Rowlands that morality is present in minds much simpler than those able of reflection and deliberation, I disagree with him in that, for me, it is just humans that are capable of the propositional thinking that “emotions with moral contents” require. Of course, were any other system to satisfy these characterization, it would qualify as a moral subject –no speciesism here.


Gomila, A. (2012): Verbal Minds: Language and the Architecture of the Cognition. Londres, Elsevier.

Gomila, A. (1996): “Personas primates”. En J.Mª Gómez-Heras (ed.): Ética del Medio Ambiente. Madrid, Ed. Tecnos.

Harris, P. (1989): Children and Emotion. Oxford, Basil Blackwell. [“Los niños y las emociones”, Madrid, Alianza, 1992.].

Strawson, P.F. (1974): “Freedom and Resentment”, in Freedom and Resentment and Other Essays. Londres, Methuen. [“Libertad y resentimiento”, Paidós/ICE UAB, 1995.].