Peter Railton\'s \'Moral Compatibilism\': A Critical Review

A Critical Review of Peter Railton's Moral Compatibilism
In 'Made in the Shade: Moral Compatibilism and the Aims of Moral Theory' (1995), Railton defends a theory of morality that draws upon empirical theories in addition to conceptual analysis—in what he calls the 'new' meta-ethics. What he calls 'moral compatibilism' is just the idea that morality is compatible with the idea that we are 'casual beings in a causal world' (102). I will offer a brief summary on the main points of the article, followed by a brief criticism on Railton's account of moral compatibilism.

Railton begins by describing different kinds of criticisms of the very idea of theorizing about morality, which he sorts three distinct types of criticisms. While deliberative criticisms "concern the appropriate role of theorizing as an activity in a good life" (81), regimentation criticisms focus centrally to "bring to substantive moral judgment certain norms of thought appropriate to other domains of judgments" (81). Integrative criticisms, finally, focus on the requirement of the methods of moral thought to square with epistemological standards that are drawn from outside morality; the criticism, which this article is most intimately concerned. Railton claims that the main worry for moral philosophers, are theories that do not respect the autonomy of morality—and this worry is met through integrative criticisms.

Railton's central task is to show that morality and moral reasoning is inextricably tied to empirical inquiry; and so is not autonomous in that sense. His defense of ethical naturalism is through his distinction between methodological and substantive naturalism in ethics; where methodological naturalism maintains that moral questions can only be resolved by scientific methods; and substantive naturalism—the reducted from of naturalism—where moral concepts and principles can be reduced to empirical concepts and properties (Railton 86). Importantly, Railton's version of neither reductive nor is it 'scientistic' in the methodological sense, since he does not claim either that morality is reducible to empirical facts or that the only way to resolve moral dispute is by applying the methods of the developed empirical disciplines. What Railton does claim, however, is that moral knowledge must not be incompatible with what is or at least what can be known by the best scientific theories—and thus it must be compatible with the naturalistic framework; thereby immediately ruling out non-naturalism and supernaturalism. Again, according to Railton, morality must be compatible with the idea that we are 'causal beings in a causal world' (102).

Although Railton indeed does not want to beg the question against his critics who see morality as having its own distinct norms of justification; and thus who see morality as autonomous in a substantial sense. He argues that from within morality, lay two (what he calls) meta-principles, that actually highlight how morality, even if autonomous in some sense, simply cannot be substantially autonomous; the first, which purports that 'ought' implies 'can'; and the second, which purports to 'treat like cases alike'. Although prima facie these meta-principles seem vacuous, they are indeed not completely empty. Concerning the former meta-principle, if ought implies can, it inevitably leads into empirical considerations—for in order for any moral judgment we first must figure out if the guidelines offered by the moral theory are even possible to employ in the practical sense. Therefore, when one makes the claim that you 'ought' to do something, they are presupposing that you can in fact do it—otherwise it is completely nonsensical. Likewise, the latter principle namely to treat like cases alike as well seems to be nonsensical to deny; as it seems like a formal requirement of moral language or moral judgments.

Clearly then, to deny either of the two meta-principles would thereby deny any reasoning let alone moral reasoning, since these meta-principles simply remain a part of our understanding of what morality is. Railton claims this leads not just to narrow reflective equilibrium, but wide reflective equilibrium; where not only does moral reasoning require a coherent package namely of enjoining our general moral principles with our more particular moral judgments (i.e. narrow reflective equilibrium), but all of that in addition to coherence with our best empirical theories and empirical knowledge (i.e. wide reflective equilibrium)—thereby maintaining a much broader range of coherence that includes, for instance, psychology, cognitive science, etc. Importantly, although with narrow reflective equilibrium, it is not logically inconsistent to defend a conception of morality as being autonomous in some sense, since here we are only in the realm of general moral principles and particular judgments; in wide reflective equilibrium, any remaining substantial sense of autonomy of morality drops out, precisely because coherence is required to empirical facts that are completely independent and independent of any moral realm of inquiry. Railton has thus argued simple meta-principles to a much larger claim that morality simply cannot be autonomous in any significant or substantial sense.

Railton also defends the idea of asymmetric supervenience of the moral on the natural—the basic idea of which is that there is no moral disagreement or difference without an empirical disagreement or difference. In other words, any moral disagreement presupposes some kind of empirical disagreement; which means that if two people were to completely agree on all the relevant empirical facts, it would not be possible for them to disagree morally. The supervenience is 'asymmetrical' simply because only the moral supervenes on the natural, not the other way around. His reason for this is because people often change their moral beliefs because they change their non-moral beliefs, which seems right.

Although Railton attributes a weaker sense of the 'autonomy', namely in the sense that is it still indeed logically possible for someone to be a moral skeptic, it is hard to see how this sense of the autonomy of morality is autonomous in any significant sense. Given Railton's claim about the relationship between moral reasoning and empirical inquiry—namely how they are inextricably tied—and how the two meta-principles, which he regards almost as axioms, that inevitably lead into empirical considerations and wide reflective equilibrium, this seems to deny any substantial sense of autonomy that morality or moral reasoning can have in itself. Similarly, in Railton's distinction between regimentation and integrative criticisms, its hard to note any substantial difference between the two, since the former relies upon conformity to standards of acceptance, while the latter relies upon conformity in a similar sense, to epistemological standards; both of which seem the same upon reflection.

Finally, concerning Railton's deference of asymmetric supervenience of the moral on the natural—where any moral disagreement presupposes some kind of empirical disagreement—which implies that if two people agree completely on all the relevant empirical facts, they would thus necessarily agree morally—seems to be not the case with the issue of abortion. It is indeed conceivable that two individuals that completely agree on the facts of the matter, might still have opposing views on the rightness or wrongness of permitting abortion, and would therefore disagree with each other, without disagreement of the relevant empirical facts. Further, 'complete agreement on all the relevant facts' runs into further problems, as it seems utterly difficult and almost insurmountable of a problem to actually establish when we have a full account of the relevant facts, or how we could know such a thing given our limits as fallible and finite creatures.

References
Peter Railton, 'Made in the Shade: Moral Compatiblism and the Aims of Moral Theory', in Couture and Nielson, eds., 79-106.