Moral supervenience

Another way to put this is to say that moral facts are a function of, depend upon, or are constrained or constituted by, some non-moral facts, and that the latter are a sufficient condition for making the moral facts true. Yet another common way of putting this is that any change in moral facts must be accompanied by a change in the non-moral facts (i.e., those on which it supervenes), but that the reverse is not the case.[2] A moral fact can be constituted by more than one set of non-moral facts-i.e., it is multiply realizable-but any given set of non-moral facts determines the moral facts which supervene upon it.

For instance, its being obligatory for Alice to pay Bob $10 supervenes upon non-moral facts, like that Alice borrowed $10 from Bob and promised to repay him. However the same obligation could have arisen if Alice had bought something from Bob priced at $10, or if she promised to give him a gift in that amount, etc. But if any such fact is sufficient to generate this obligation, then it is not possible for her to lack that obligation unless the relevant facts change. Likewise, if another person, say Cindy, has the same non-moral relationship to David that Alice does (e.g., having borrowed $10 from him), then she must have the same obligation Alice does.

The principle is compatible with a very fine-grained analysis of the supervenience base for moral predicates, and hence is compatible with moral particularism. (R.M. Hare, in the first recorded usage of the term moral particularism, defined these as incompatible, saying they were contradictories, but his definition of particularism is not identical with its contemporary usage.[3]) The non-moral facts on which a moral fact supervenes could be quite complex, with many exceptions and qualifications; for instance, one might only be obliged to pay someone $10 if the recipient has not waived the debt, or does not intend to use the funds to cause immediate harm to someone else, etc.

As the principle of moral supervenience by itself places no limit on what the non-moral supervenience base of any moral fact is, nor on how complex the former might be, it is generally considered a very weak principle. It is even compatible with the idea that obligations may vary for persons of different nationalities, races, genders, ages, etc. It only disallows exceptions for certain persons merely on the grounds that they are different people from other persons. Note that the claim that Alice's obligations supervene on her properties must be understood as properties other than her simply being Alice (as opposed to being, say, Cindy). "Being Alice" is not a property of Alice apart from all the other properties which distinguish Alice from, say, Cindy, such as the fact that she is called "Alice," or has a different residence, birthplace, etc. from Cindy. Hare put this point by saying that the supervenience base of a moral fact could not include "individual constants" (including proper names of persons, countries, etc.)[4][5]

Moral supervenience is a kind of moral universalizability principle, like the golden rule or Immanuel Kant's categorical imperative, so the underlying idea may be as old as the golden rule. Moral supervenience differs from most other universalizability principles in that it adds no specific criterion for the supervenience base of permissible behaviors, so it cannot function as a comprehensive test for moral permissibility, as most other universalizability principles purport to do.[6]

The earliest precise specification of the principle may be found in Samuel Clarke's "Rule of Equity" according to which "Whatever I judge reasonable or unreasonable that another should do for me: that by the same judgment I declare reasonable or unreasonable that I should in the like case do for him."[7] A few years later William Wollaston echoed this claim with the principle that "Whatever is either reasonable or unreasonable in B with respect to C, would be just the same in C with respect to B, if the case was inverted. Because reason is universal and respects cases, not persons."[8] A few years later Richard Cumberland restated this as a requirement of "right reason," which entailed that “It is included in the notion of a true proposition, (a practical one, for instance,) and is consequently a necessary perfection of a man forming a right judgment in that affair; that it should agree with other true propositions framed about a like subject, tho that case should happen at another time, or belong to another man.... Whoever therefore judges truly, must judge the same things, which he thinks truly are lawful to himself, to be lawful in others in a like case. "[9] Later versions are found in Reid,[10] Moore,[11] Sidgwick,[12] and Sharp.[13]

The first usage of the general term "supervenience" and the specific term "moral supervenience" in print was by R.M. Hare,[14] although he later suggested that the former term was used by other philosophers conversationally before he put them both into print.[15] While Hare was primarily interested in the supervenience of moral concepts on non-moral ones, he also argued that other evaluative concepts, e.g., aesthetic ones like beautiful, pleasant, nice, etc., must also supervene upon non-moral facts. For instance, it is senseless to call one room "nice" and another "not nice" unless there is some underlying difference between them describable in non-aesthetic terms (like the arrangement of the furniture, color of the walls, etc.) If the two rooms are identical in all their non-aesthetic properties, then they must also be identical in their aesthetic ones.[16][17]

In a series of later books, Hare made moral supervenience, combined with the criterion that a rational being would prefer the satisfaction of his preferences over their frustration, the basis of his idea of universal prescriptivism. From this he derived a version of utilitarianism, by arguing that to prescribe a particular action in one's circumstances was only rational if you would prescribe anyone's else's doing it, even if you were equally likely to be any agent (including all those affected, for good or ill, by the action). This would only be true if, were you to personally experience all the good and bad effects of the action upon all affected persons (i.e., the satisfaction and frustration of their preferences), you would not prefer some other action to the one in question.[18][19] He often simply called moral supervenience "universalizability" and equated it with Kant's principle of universal law,[20][21] although the two are not the same (see moral universalizability).