In her 1958 paper "Modern Moral Philosophy", G.E.M. Anscombe argued that the expression "morally ought" has no content outside of a "divine law conception of ethics", and that it has become, in modern usage, an expression with mere "mesmeric force" (Anscombe 1958, 8). Is Anscombe right about this? In this blog post, I examine what "morally ought" might mean in secular discourse. I don't reach a conclusion, but at least I manage to narrow down the possibilities by eliminating some.
1. Defining "ought"
Before we try to understand "morally ought", we must first get clear on what we mean by "ought". Not every ought is a moral ought. For example, consider the statement, "If you want to reach the bus stop on time, you ought to leave now." This is a perfectly legitimate use of the word "ought", yet it has no moral content. It's a completely open question whether I morally ought to leave for the bus stop. Likewise, consider the statement, "You know A; A implies B; therefore, you ought to believe B." Again, this is (at least in everyday, non-philosophical discourse) a perfectly legitimate use of "ought", but it has nothing necessarily to do with morality.
Therefore, I propose the following definition of "ought": it is true, within a given normative sphere, that P ought to do X if and only if, within that normative sphere, P has sufficient reason to do X.
This definition relies on the notion of "normative spheres". An example of a normative sphere would be the epistemic sphere. The statement "You know A; A implies B; therefore, you ought to believe B" is a statement within the epistemic sphere. And, if true, it entails that, within the epistemic sphere, I ought to believe B: we might say that I epistemically ought to believe B. Notice that there might be conflicting oughts in other normative spheres. For example, if someone will get tortured to death if I come to believe B, then I might have a moral obligation to do everything I can to avoid believing B. In other words, both of the following may be true: (1) I epistemically ought to believe B; (2) I morally ought not believe B.
To be a bit more precise, I define "ought" as follows: it is true, within a given normative sphere, that P ought to do X if and only if, given only the reasons for and against doing X that are available within that normative sphere, P has sufficient reason to do X, all else being equal. P epistemically ought to believe B if and only if, given only the epistemic reasons for and against believing B, P has sufficient reason to believe B, all else being equal. Of course, all else may not be equal. If someone will be tortured to death if P believes B, then P may not have sufficient reason to believe B, all things considered. This doesn't change the fact that P epistemically ought to believe B, because it doesn't change the fact that, given only the epistemic reasons for and against believing B, P has sufficient reason to believe B, all else being equal.
I admit that I can't easily identify the boundaries of the various normative spheres—with the possible exception of the epistemic sphere. But, for my purposes, I actually don't need to. For the purpose of this blog post is (partly) to investigate what the boundaries of the moral sphere are. If we can figure out what those boundaries are, then we can simply define "P morally ought to do X" as "Given the reasons for and against doing X that are available within the moral sphere, P has sufficient reason to do X, all else being equal".
2. First proposal: moral ought = altruistic ought
Nowadays, many people regard morality as mainly a matter of benefitting and respecting others. For many of us, there would be nothing immoral about ruining your own life if you could do so without hurting or burdening others. This leads us to a first attempt at defining the moral sphere: the moral sphere is the sphere of altruistic action.
Ultimately, this first proposal seems to mischaracterize the moral sphere. There's nothing altruistic about the mere avoidance of premarital sex. Thus, on this first proposal, an obligation to avoid premarital sex wouldn't qualify as a moral obligation. Yet an obligation to avoid premarital sex clearly qualifies as a moral obligation.
I must be clear about what I mean here. Clearly, many people don't think that there's any obligation to avoid premarital sex. But that's not the point. Everyone surely agrees that if there were an obligation to avoid premarital sex, then it would be a paradigm case of a specifically "moral" obligation. (Indeed, people sometimes say that they "Don't believe in morality", meaning that they don't believe in traditional cultural norms such as avoidance of premarital sex.) Thus, we shouldn't define the moral sphere solely as the sphere of altruistic action.
3. Second proposal: moral ought = non-self-serving ought
So morality (as it's commonly understood) seems to encompass things like sexual activity, not just altruistic activity. Again, I must be clear about what I mean: we regard traditional sexual norms as specifically (even paradigmatically) "moral" norms, even if we don't believe that those norms actually apply. How can we define the moral sphere in light of this fact?
Here's an attempt: perhaps the moral sphere is the sphere of non-self-serving action. If defined in this way, the moral sphere would encompass both altruistic behavior and the observation of sexual norms. In that case, P "morally ought" to do X if and only if, given only P's reasons for acting in a non-self-serving way, P has sufficient reason to do X, all else being equal.
This proposal also seems wanting. Is it really plausible to say that one can never have a moral obligation to benefit oneself in some way? Perhaps one can't, but that seems to be a synthetic truth, not an analytic one. Nothing in the very concept of morality seems to forbid the existence of a moral obligation to benefit oneself.
4. Third proposal: moral ought = ought that doesn't depend on the agent's contingent desires
Here's yet another attempt: perhaps the moral sphere is the sphere of actions whose normativity doesn't depend on the agent's contingent desires. On this view, P "morally ought" to do X if and only if, given only the reasons for and against doing X that don't depend on P's contingent desires, P has sufficient reasons to do X, all else being equal.
This proposal suffers from a fatal flaw. On this proposal, purely epistemic reasons count as moral reasons. Suppose that I have sufficient epistemic reason to believe B, all else being equal. Furthermore, suppose that there are no reasons against believing B. On the current proposal, it would follow that I morally ought to believe B. This seems overly moralistic, to say the least. Granted, I epistemically ought to believe B. But unless something horrible will happen if I fail to believe B, it's odd to say that I morally ought to believe B. We don't think of purely epistemic reasons, by themselves, as moral reasons.
5. Virtues and vices: relevant?
At this point, one might complain that I've been using too narrow a conception of the moral sphere. Surely morality consists of more than moral reasons for and against specific actions. Even if P and Q perform the same action in the same circumstances, P may be morally better than Q in terms of character. So we can't restrict "the moral" to what I have been calling moral reasons; we must expand the moral to include, among other things, virtues and vices of character.
All of the above points are well taken, but I'm not sure that they're relevant to the current discussion. Right now we're discussing what sets moral oughts apart from other kinds of oughts. It's not obvious to me how expanding the moral sphere to accommodate virtue theory brings us any closer to defining moral oughts. One might claim, as Rosalind Hursthouse does (cf. Das 2003, 331), that X is the right action if and only if a virtuous person would do X under the circumstances—that is, P ought to do X if and only if a virtuous person would do X in P's situation. But notice: that's a substantive claim about the content of actual moral obligations: specifically, the claim is that we have a moral obligation to do whatever a virtuous person would do. This claim may be true, but it's irrelevant to my current aim. My current aim is to explain the very meaning of the statement "P morally ought to do X". This statement can't just mean "X is what a virtuous person would do in P's situation"; if it did mean that, then Hursthousian virtue ethics would be trivially true, and there wouldn't even be debate over it.
6. Conclusion (or not)
Does the statement "P morally ought to do X" have any content over and above that of "P ought to do X"? In conclusion, I'm not sure. However, I am sure that the content cannot be any of the following:
- Given P's reasons for acting altruistically, P has sufficient reason to do X, all else being equal.
- Given P's reasons for acting in a non-self-serving way, P has sufficient reason to do X, all else being equal.
- Given the reasons for and against doing X that don't depend on P's contingent desires, P has sufficient reason to do X, all else being equal.
- X is what a virtuous person would do in P's situation.
References:
Anscombe, G.E.M. "Modern Moral Philosophy". Philosophy 33.124 (1958): 1-19.
Das, Ramon. "Virtue Ethics and Right Action". Australian Journal of Philosophy. 81.3 (2003): 324-39.
Pigden, Charles. "Anscombe on 'Ought'". The Philosophical Quarterly 38.150 (1988): 20-41.
No comments:
Post a Comment