I am grateful to both Tom BeauchampFootnote 1 and Carson StrongFootnote 2 for their thoughtful responses to my article.Footnote 3 I respond to each individually and then discuss an issue that is of concern to both, namely, my pessimism that the existence and scope of common morality, in my sense of the term, could be demonstrated through empirical research.
Strong argues that I am wrong to be worried about the normative significance of the commonness of common morality, because common morality, as he reads it, does not require near-universal acceptance, but only universal applicability (to use my terminology):
In regard to common morality in the normative sense, the term “common” should not be taken to refer to the acceptance of the precepts by almost all moral agents in the group. Rather, I have suggested that we might take it to refer to the idea that the moral rules and principles in question are binding on all moral agents in the group; it is common in the sense that it applies to all.
Strong can use the phrase “common morality” however he likes, of course, but it seems to me that in caring only about universal applicability, he has moved away from the distinctive interest of common morality theories. Almost all major ethical theories—Kant’s, Mill’s, Rawls’s, and so on—are supposed to be universally binding. What animated my interest in common morality theory was a particular sort of naturalistic methodological and metaethical commitment: the proper, indeed the only viable, place to begin ethical reflection and justification is from within and by way of attention to our shared moral life and commitments. If Strong wishes to sever “common morality” theory from this starting point, thereby letting most major ethical theories, no matter how abstract and aprioristic, count as examples of common morality theory, then I am not sure we have enough shared philosophical motivations to agree or disagree productively about the merits of common morality theory.
It seems to me that that Strong’s response to my discussion of the Pirates’ Creed misses what motivated that discussion. I argued that Beauchamp and Childress were wrong to base their argument against the sufficiency of coherentism on the premise that the Pirates’ Creed is a “coherent code.” My point was to use the example of the Pirates’ Creed to point out the multidimensionality of moral life, and to argue that questions of coherence could not be answered by looking at lists of rules alone. I used the richness of pirate moral life and its embeddedness in a larger habitus as the way into building my alternative picture of common morality as shared embodied practice. Strong argues against my appeal to the Pirates’ Creed by claiming that there are other reasons to be suspicious of coherentism, particularly in its most popular Rawlsian forms, and he gives arguments for why neither wide nor narrow reflective equilibrium is a sufficient justificatory method. Because I never claimed to be any kind of Rawlsian, I am not sure why he takes this to target me. I do see an important role for reflective equilibrium in justification, as I think virtually everyone does. But my goal was not to defend (Rawlsian or any other) coherentism against Beauchamp and Childress’s attack, but rather to point out that their conception of what made up a moral life that could be judged coherent or incoherent was far too thin, at least in the context of their discussion of pirates. I defended a kind of naturalistic transcendental foundationalism that I think doesn’t fit neatly into the existing coherentist/foundationalist debate.
Furthermore, I find Strong’s complaint against reflective-equilibrium-based coherentism peculiar, in context. His argument is that we can never be sure, no matter how much reflective equilibrium we engage in, that our practice of adjusting our beliefs isn’t infected with bias. This seems right to me, but it is an equally strong charge against a foundationalist. Strong gives us no tools or standards by which we can check whether the beliefs and principles we take as foundational are themselves bias infected. He seems to be conflating foundationalism and coherentism, taken as theories of the metaphysics of morality, with foundationalist reasoning and reflective equilibrium, taken as methods of moral justification. I certainly propose no metaphysics of morality in my article. In practice, foundationalist reasoning from what seem to be settled judgments seems at least as bias-prone as reflective equilibrium.
Notice that I don’t have to argue that common morality is unbiased—indeed, odds are overwhelming that it is biased in various ways. I am making the transcendental claim that we cannot possibly opt out of common morality and still engage in meaningful reflection and practical interactions. Likewise, justification has to begin in medias res, with stable, embodied normative practices as we find ourselves embedded in them. If this is right, then the worry that biases may linger can’t tell against my account and in favor of another. Luckily, on my account, we have a full and rich roster of critical tools with which we can critique and refine our own moral habitus, although there is no guarantee that in using them we will ever completely eliminate bias.
I turn now to Beauchamp’s response. I think that he is right to accuse me of unfairly overemphasizing the role of thin general principles in his and Childress’s picture of the common morality; they do indeed say that the common morality includes much more than these. In my defense, it still seems hard not to take general principles as the heart of common morality, given that the book is entitled “Principles” of Biomedical Ethics and is organized for the most part around their four key mid-level principles. Their commitment to the centrality of these principles predates their metaethical story about common morality, which was designed to accommodate their role. And it would be odd if principles were central to bioethics but not practical ethics more generally.
I admit, though, that if Beauchamp is serious that there is nothing special about the place of general rules in the common morality, then I understand his and Childress’s discussion of the Pirates’ Creed even less than I thought I did. If general rules play no privileged role in moral reflection, then why would we think that we could draw any conclusions about the coherence or critical resources of a moral system by artificially extracting a set of rules from its lived context and examining it on its own? If Beauchamp and Childress were not at least roughly equating explicit pirate rules with pirate morality, then it doesn’t seem to me that the example has any clear import for their purposes.
I am actually quite pleased if our points of agreement turn out to be more robust than I thought, as I designed the article not as a refutation or critique of Beauchamp and Childress’s picture, but rather as a development of it in a perhaps unexpected direction. But I think there is at least one fairly deep difference between Beauchamp and me that is not easily dismissed. Common morality, for Beauchamp, consists of judgments to which virtually all people (or in some of his formulations, all “morally serious people” or all people “committed to morality”) agree. Whether these are judgments about general principles, ideals, or whatever else, they are the sorts of things that have propositional content and can be true or false, and common morality is distinctively marked by agreement over them. My picture is quite different. I argued that we should understand the common morality as a common habitus—a shared space of embodied practices, habits, reactions, and skills, crucially including interactive negotiations. There is no reason to think that this shared space could be captured in a set of truth-evaluable statements at all, nor that there is any distinctive list of judgments on which we have agreement.
There are two points here. First, the embodied practices and responses that make up the habitus cannot in any obvious way be said to have content. Skills and dispositions of the sort I focus on are not propositional. We may or may not be able to capture all of the habitus of common morality in descriptive or judgmental form, but this description is not, in the first instance, what we share when we share common morality. Second, partly because I don’t focus on truth-valuable judgments, I don’t focus on agreement at all. My claim is that we are all embedded—albeit differently positioned—within a stable web of embodied practices, skills, and reactions (even if we each participate in different parts of this web); this enables us to negotiate and interpret one another successfully, and to give, as it were, living flesh to our otherwise uselessly abstract moral concepts and reflections. This common moral life we share is consistent with plenty of differences and disagreements within it (including country-specific differences of the sort that interest Strong, I should note).
My rejection of a picture of the common morality as consisting of judgments on which we generally agree is central to my take on the testability question, which concerns both Strong and Beauchamp. I claim that the existence of the common morality need not be demonstrated by empirical test, and that it is not clear, even in principle, how one would test this. It need not be tested, because our confidence in its existence is transcendental; we manage to interact with and interpret one another skillfully, generally speaking, and a condition for the possibility of this is that we share a habitus that gives flesh to our moral reflections and transactions.
If we wished to test for the scope of the common morality, to chart out its boundaries and “measure” it, how would we proceed? Beauchamp writes that he and I
seem to differ over whether claims about empirical facts can be examined in scientific research, even if only piece by piece. I support such research, whereas she thinks that “the idea of trying to establish the existence of a shared habitus through some sort of empirical testing seems hopeless.” In the end this matter cannot be decided until empirical research has been attempted and its results displayed.
Strong writes, “the fact that there can be different formulations of a rule does not preclude presenting the various formulations to someone and asking whether she agrees with any of them.” I think both of these responses to my untestability claim miss the point.
The reason I think that testing seems “hopeless” is that I am not sure what sort of research could be relevantly attempted in the first place; we can’t just “attempt it” and see what happens, as Beauchamp suggests, because we don’t know what “it” is. Strong’s suggestion of checking for agreement about “the” various formulations of rules seems doubly off point. First, there are infinite rule formulations that match any dispositional pattern, and an indefinite number of ways of packing specificity and conditionalization into a rule or leaving specification and balancing up to the user. Hence there is no way of listing all the formulations of a rule. Furthermore, I claim that abstract moral rules literally have no practical content apart from the life they are given within a particular habitus. Moral concepts such as respect, dignity, autonomy, and so forth have no meaning except as animated in a lived context. So if we just present people with abstract statements of moral rules, there is no telling whether they are agreeing or disagreeing with the same thing when they respond. Second and more importantly, on my account, again, statements of rules are not in the first instance what make up the common morality. If the common morality does not consist of a set of shared propositional judgments, then giving people statements and asking whether they agree is prima facie unhelpful. Whatever the results, it would be unclear what they would tell us, if anything, about the scope of people’s shared moral life.
As for testing the extent to which we are embedded in a common web of practices and reactions and so forth, this is going to be difficult given that we are each differently embedded in this web; we do not act and react the same way as one another, nor do we all play the same roles or face the same normative demands within our shared moral community. What makes the web “common” is that it is stable and that we can interact with and interpret one another skillfully within it. Its boundaries are surely fuzzy and fluctuating, and its character is always slowly morphing. I can imagine excellent qualitative anthropology that carefully documents interesting cases of breakdown and success when it comes to our moral transactions. This could illuminate the character and limits of bits and pieces of the common morality. But it will never constitute a “comprehensive analysis” of the common morality, and it wouldn’t count as a test of its existence—which, luckily, I think does not stand in need of empirical proof.
