Tuesday, July 12, 2011

Internal and external reasons

Chapter 5 of Joyce's The Myth of Morality continues his discussion of reasons. He argues that all normative reasons are internal, following Bernard Williams' argument that, in Joyce's words:
An external reason claim is one that is applied to the subject of the ascription regardless of what are his desires; and Williams argues that although we make such claims, they are all false. (p. 108)
They are false because reasons for action must be able to explain the action in question, and to do this the alleged reason must be able to motivate the agent. Joyce accepts the Humean view (without claiming to be able to prove it correct) that actions are motivated by beliefs and desires working together.

There is clearly a difference between the reason why someone did what they did and the reason why they should have done something else. I mean, there is a difference in the sense that these do not have to be the same thing. Perhaps you did something for the money but could have made more money doing something else, in which case both reasons are "for the money." But if you kill someone for the money then the reason why you should have refused is rather different. And it might not match any of your desires at all. It is still a reason why you should not have done it.

Is it a reason that you had not to kill the person? Maybe not. But it was a reason that, so to speak, existed for you not to kill them all the same. What kind of existence do reasons that are not acted on have? How can there be a reason that is not owned by someone, that is not someone's reason? And what is the reason not to kill people for money (if, say, Qaddafi makes you a generous offer)? There isn't much of a reason, in a sense. That is, there isn't much I can think of to say to anyone who seriously asked such a question. Because they are people, is about as good as I can do. But it sounds crazy, or like a joke, to say that there is no reason for people not to slaughter innocent civilians for money if that is what they want to do.  


  1. One way of looking at all of this might see the claim that all genuine reasons are internal (does this mean roughly the same as hypothetical, rather than categorical?) is connected to something like egoism--that I only have a reason to do something if it connects in some strong sense to my life/self. In a way, you could probably get fairly far, in convincing others of things, by teasing out the connections between what they think are not reasons (at first glance) with other things they care about. So then the big problem, as it were, is with the person who just doesn't care about anything. And you might say, "that's inhuman," because it is inhuman in the sense that normal, even marginally adjusted humans are just not like that. Maybe the test case then is whether there is some kind of "external" reason to care about anything at all. If you're a theist, for example, then probably there is. But if you're not, then perhaps there isn't (that is, not even an external reason to be an egoist). Even so, if this kind of point just amounts to the idea that there's no getting a sociopath to care about anything--that there's no reasoning with such a person about it--then it all seems less of a threat to the actual things we do when "doing ethics." So, what we need is to be good as seeing connections, rather than finding magical "external" reasons that serve as reasons regardless of one's cares (or pro-attitudes). And where we find cares that, excluding sociopaths, have broad occurrence in various times and cultures, then we have some basis for treating those things as very strong reasons, where the contingency on a particular, but also diverse "we" sort of disappears. That is, there's contingency and then there's contingency. And if some reason is responsive to general features or cares of being human, then it's not some kind of huge mistake, or myth, to see it as a categorical kind of reason. Of course, someone like Williams might say that there aren't many such considerations ("one has to have a special reason to kill someone..."), but that's for another time. I should be grading right now, whether I care to or not...

  2. Thanks. This all sounds right to me, but almost too right. I'm struggling to think how Joyce would respond. He talks about rational beings as such and aliens that eat human brains but are still rational, and so on. Maybe what he says is right about those cases, but wrong about the relevant ones. But then it seems suspiciously easy to dismiss his position.

    He says that folk morality (that's my term, I believe) says that even people who cannot keep their promises (because they are so selfish and have been so badly raised) ought to do so. (So much for ought implies can.) So they are supposed to have a reason to keep their promises (i.e. a moral reason) even though they have no desire to keep them or belief that they ought to keep them. But that's crazy. They have no such reason. So folk morality is wrong.

    I have a few problems with this. One is that I have a hard time accepting that such people exist, i.e. people who cannot keep promises. Another is that there usually are game theory-type reasons for keeping your promises (tit for tat, etc.). And another is that many people think promises are not as categorically binding as Joyce seems to think people think they are. Roughly speaking, I think he's all wrong (but because of this I have a hard time persuading myself to re-read all that he says about it carefully enough to be sure I haven't missed something important). On the other hand, he is careful to make modest claims about making his position plausible and so on, rather than proving it.

  3. To turn (and distort) a phrase: too right doesn't make a wrong...

    I'm inclined to agree that this notion of "folk morality" is something of a straw man. My students are contextualists about a lot of things. At the same time, I can find things where we all agree, say, that it's wrong, terrible, etc. (like the Holocaust). But there's a lot a vagueness in between. Maybe some folk--say moral conservatives--are more like Joyce's "folk." But that hardly makes them representative. And going all the way to error theory if the real target is something like absolutism seems to involve a confusion of "metaethical" and "normative ethical" levels. Of course, the idea that there are two fully separable levels here is also wrong.

  4. Agreed--too right isn't wrong. I just wonder whether I've misrepresented Joyce's position somehow, although, of course, I don't think I have.

    I agree also that there's something wrong with folk morality as he conceives it, although I think what he objects to is the idea of 'moral wrongness' as such, and I have some Anscombean sympathy with the idea that there is something wrong with that idea. But I'd much sooner become a virtue ethicist as a result than an error theorist.

  5. But I'd much sooner become a virtue ethicist as a result than an error theorist.

    Agreed. And there's a small, inchoate pragmatist who lives inside me (inchoate because my studies of pragmatism are limited, esp wrt Dewey, of whom I should read more) who reminds me that giving up notions like intrinsic value and categoricality isn't the end of the world. (With the consolation being something to the effect that not all circles are vicious.)

  6. Yes, I think there's (probably) something good in pragmatism that is relevant to all this, but I don't know enough to be sure. Joyce's attitude toward practice, use, etc. seems a little odd to me.