Moral Error Theory - III
Chapter 3, The Naturalist's Case for Error Theory, Joyce (2024)
18 Nov 2024
Do moral properties exist in the world, the way that a table or a chair does? Moral naturalists would say yes. For instance, the moral naturalist might hold that the right action is the one that maximizes pleasure. There is nothing about this view that clashes with science. Pleasure as a property is open to scientific investigation, and we can use scientific methods to determine what actions maximize pleasure. Joyce, however, thinks that this is a mistaken view. Moral properties do not exist in the world in the same way that natural properties do. He claims this is so because moral properties are normative (action-guiding) in a way that natural properties are not.
First, let's consider some ways of understanding norms. One starting point is to consider the distinction between hypothetical and categorical imperatives:
Both imperatives are associated with a certain kind of normativity. For hypothetical imperatives, the normativity is contingent on the satisfaction of some end. For example, if you want to be healthy, then you should exercise. If you don't want to be healthy, then the imperative can be withdrawn. On the other hand, categorical imperatives are not contingent on anything. They are applied unconditionally. For example, many institutions have rules that are categorical imperatives. In the institution of golf, it is categorical that you must play the ball where it lies, regardless of any other consideration.
Both hypothetical and categorical imperatives are compatible with naturalism. In the case of hypothetical imperatives, we can take as ends some function of the person's desires or preferences. Categorical imperatives are also naturalistically acceptable, as they can be understood in terms of institutional rules that develop naturally in human societies.
Thus, we have two kinds of normativity: hypothetical and categorical. Which one corresponds to the normativity we associate with moral properties?
Joyce argues that moral imperatives are not equivalent to hypothetical imperatives. Hypothetical imperatives can be withdrawn if the relevant desire is no longer present, but we don't think of moral imperatives as something that can be simply withdrawn. What if we could find some universal desire that everyone has, and then base morality on the satisfaction of that desire? This way, moral imperatives would be hypothetical imperatives contingent on the satisfaction of a universal desire that is always present, and thus are never withdrawn. But this is unlikely: humans are psychologically diverse, and desires can change. Plus, even if we could find such a universal desire, hypothetical imperatives contingent on the satisfaction of that desire would be different from moral imperatives in one important way: our emotional reaction when they are violated. If someone commits a moral transgression, our reaction is likely to be one of moral outrage. But anger would not be the appropriate reaction to someone who frustrates the satisfaction of a universal desire. More appropriate would be disappointment, or perhaps pity, but never anger.
What about institutional categorical imperatives? Are they good candidates for the basis of moral normativity? Joyce thinks not. Moral imperatives are inherently reason-providing. We don't just say that "X is morally wrong", we say that "Don't do X because it's morally wrong." Although categorical imperatives appear to be inherently reason-providing the same way that moral imperatives are (since they are unconditionally applied), this appearance is true only from the perspective of someone who has internalized the rules of the institution, or is speaking from within the institution. Consider the following two statements:
Both are institutional categorical imperatives, but the first one makes explicit the reason for obeying the imperative, namely, that one is part of institution I. In the second case, the qualification is dropped, presumably because the speaker and the listener have internalized the rules of institution I. In either case, the reason for obeying the imperative is the same: both are part of institution I. Now, what if we think of morality itself as an institution? Consider the following:
The first case corresponds to the moral imperative "Don't do X because it's morally wrong," and the second case is the same imperative without the qualification. But, again, morality as an institution does not capture the crucial feature of moral normativity: the property of being inherently reason-providing. Moral wrongness must in itself provide a consideration against acting in a certain way, regardless of whether one is part or not part of the institution of morality. The problem is that, for anyone outside of the institution of morality, there is no particular reason for them to care about morality. And it doesn't matter, from the perspective internal to morality, that anyone outside of the institution should care. People outside should care, but they don't really have to care, since they were never part of the institution to begin with.
Therefore, Joyce concludes that moral imperatives are not equivalent to institutional categorical imperatives, nor are they equivalent to hypothetical imperatives. Since moral imperatives contain certain normative properties that are not captured by any naturalistic account, Joyce concludes that moral imperatives cannot be made to fit within the naturalist framework.