Print This Post Print This Post

I’ve been doing some thinking about Michael Smith’s argument against externalism about motivation. To remind you, externalists about motivation – henceforth, externalists – deny that there is any conceptual or internal connection between making a moral judgment and being motivated (to some extent) to act on it. Internalists about motivation – henceforth, internalists – assert that there is such an internal connection between judgment and motivation.

Here is a rough and ready summary of how the argument proceeds. Smith thinks that externalists about motivation – henceforth, externalists – are committed to an implausibly fetishistic explanation of the connection between judgment and motivation. Why is that? Well, both internalists and externalists can agree, Smith claims, that there is a striking reliability between making a judgment and being motivated to act on it. Smith makes that a conceptual truth by ascribing the reliability to, specifically, ‘moralists’. (See, in particular, Smith, ‘The argument for internalism: reply to Miller’, Analysis, 56.3, July 1996, pp. 175-84.) Moralists are defined as those agents who are appropriately motivated by their moral judgments. The crucial question, for Smith, is this: which of the two sides, internalism or externalism, provides a better explanation of why that might get to be a conceptual truth? Smith argues that internalists can appeal to motivation de re at this point (motivation grounded in a a concern for right-making features), whilst externalists have to appeal to motivation de dicto, or motivation which is grounded in a bare concern with what morality prescribes. And that’s ugly, because, in effect, it’s ‘one thought too many’, to echo Williams.

Lots of people have complained about this argument: for one thing, perhaps motivation de dicto isn’t so morally unattractive after all, and perhaps it doesn’t entirely exclude motivation de dicto. I won’t go back over this ground.

I’m puzzled, rather, by the prior question of how the argument is supposed to work, and, in particular, by how the externalist can be legitimately denied the resources Smith awards to internalists. Let me explain. The moralist, as construed by internalism, makes a moral judgment. For Smith, that is a belief state (see The Moral Problem, ch. 5). To produce action, moral judgment therefore needs to enter into combination with a desire state. What is the content of that desire state? Preumably it’s a desire such that the motivation emerges as de re, not de dicto. It’s a desire to respond to, or instantiate, right-making features. That needn’t have been the case, because, due to Smith’s Humeanism, the belief state and the desire state are independent existences.

Why can’t the externalist say exactly the same thing? Here we have a moralist agent; he makes a moral judgment; and, to produce motivation, that judgment needs to be combined with a desire state. If it’s more attractive to be motivated de re rather than de dicto, then externalists can acknowledge that. If it’s conceptually impossible for the resulting motivation to be de dicto rather than de re, externalists can acknowledge that too. (Though surely they won’t, because it’s clearly not conceptually impossible.) There is nothing remotely embarrassing in any of this to externalists.

What am I failing to grasp?


  1. 1. Posted by Adam Rawlings | July 4, 2007 2:06 pm

    I think the key is in the dispositional analysis of desire and the analysis of normative (thus moral) reasons. Here’s the former from The Moral Problem: “we should think of desiring to phi as having a certain set of dispositions, the disposition to psi in conditions C, the disposition to chi in conditions C’, and so on, where, in order for conditions C and C’ to obtain, the subject must have, inter alia, certain other desires, and also certain means-ends beliefs, beliefs concerning phi-ing by psi-ing, phi-ing by chi-ing and so on.”

    This hooks into Smith’s “anti-Humean” analysis of normative reasons, whereby a normative reason (what it is desirable for you to do) is identified with the desires you would have if perfectly rational. So, whatever you would be disposed to do if perfectly rational.

    The externalist has to say that your ideally rational self would be disposed to do whatever is good because it is good. That’s the use Smith is making of de dicto motivation. If the externalist tries to use a desire (disposition) to do something — say X — which just happens to be good — that is, the motivation is to X not to do good — then it seems that the motivation has become internal. Internalists are motivated de re in that it is the fact that X is X that motivates, not the fact (nor the recognition of the fact) that X is good. (Well, really, I’m being a bit slippery here. Facts don’t motivate for Smith; it’s desires that the facts be realized, or some such. But that complication doesn’t seem to make a difference here.)

    I think that’s what Smith’s driving at. I don’t find it plausible, but my disagreements with Smith are deeper than this issue. Certainly it seems to block the way you’re trying to save externalism within Smith’s account.

    Another way to save externalism, of course, is to attack the stability of the three underlying distinctions here: between de re and de dicto, between world-to-mind and mind-to-world directions of fit, and between motivating and normative reasons. None of them, as Smith presents them, strike me as particularly stable.

  2. 2. Posted by Richard Chappell | July 4, 2007 2:24 pm

    Internalists holds that moral judgment necessitates motivation, right? So, unlike the externalist, they do not need to appeal to any pre-existing desire. They can say that moral judgment gives rise to a new (de re) desire for the particular thing that has been positively appraised.

    Externalists lack this resource, and so must instead appeal to a general pre-existing desire that will cover whatever their particular moral judgment might turn out to be — hence, the desire must be de dicto in nature.

    Does that sound right?

  3. 3. Posted by Gerald Lang | July 4, 2007 5:46 pm

    Let me think further about Smith’s dispositionalism. In reply to Richard’s post, I don’t see how Smith can accommodate anything like motivational necessity (even though many internalists have wanted to do just that). On the motivational side, he’s a Humean. Beliefs and desires are distinct existences. Yes, since he’s talking about moralists, then we know in advance that the right kind of motivation is going to follow the making of a moral judgment. That is indeed guaranteed. But that’s just an upshot of the way he’s defined the term ‘moralist’. It’s a guarantee that holds, not just for internalists, but for externalists too. As far as I can see, both sides are in the same boat.

  4. 4. Posted by Gerald Lang | July 4, 2007 6:17 pm

    Now for Adam’s post. Let me fasten on the following passage:

    ‘The externalist has to say that your ideally rational self would be disposed to do whatever is good because it is good. That’s the use Smith is making of de dicto motivation. If the externalist tries to use a desire (disposition) to do something – say X – which just happens to be good – that is, the motivation is to do X not to do good – then it seems that the motivation has become internal.’

    I don’t see why internalists are entitled to earmark motivation de re from the very beginning. It seems to me that externalists can allow that certain agents may well be motivated by right-making features, rather than a concern for rightness simpliciter, without its being incumbent upon them to admit that these are the conceptual arrangements that must obtain for any agent who makes a (sincere) moral judgment. (Some will go de dicto, some may be left cold.) It is only this further admission that would force externalists to acquiesce to internalism.
    Let me point out here that I myself am unsympathetic to externalism. I’m just trying to figure out why Smith thinks he’s wrestled externalists to the ground.

  5. 5. Posted by Adam Rawlings | July 4, 2007 9:32 pm

    I suspect that Smith may want to dig in his heels on this point and insist that any externalist who accepts that motivation can come about from the right-making feature directly, rather than from the rightness of the right-making features, is really an internalist.

    The argument for this is in Chapter 3 of The Moral Problem. Smith defines two slightly different internalisms, both of which he wants to say are true. The first, “rationalism”, is “if an agent judges that it is right for her to phi in circumstances C, then either she is motivated to phi in C or she is practically irrational.” (He also notes this is supposed to be a conceptual claim.) The second, “the practicality requirement on moral judgement” (clearly, concise names are not Smith’s strong suit), is “If it is right for agents to phi in circumstances C, then there is a reason for those agents to phi in C.” Since he believes the latter entails the former, the latter can just be called “internalism” for the purposes of this discussion.

    With regards to the de re/de dicto argument, Smith defines “strong externalism” is the view that “it is a contingent and rationally optional matter whether an agent who believes that it is right to act in a certain way is motivated to act accordingly.” According to Smith, it’s the strong externalist who elevates a moral fetish into the motivation for the good and strong-willed person.

    So, the idea is this. A good and strong-willed person is one who, ceteris paribus, desires to do (is disposed to do) what his ideally rational self desires to do (is disposed to do). According to strong externalists, a good and strong-willed person’s (GSWP) dispositions to do these things are a contingent matter. The GSWP could be disposed to do very different things. According to internalists, though, this is an impossibility: the GSWP is (usually) disposed to do what is right, and thus is (usually) disposed to do what he has reason to do. Else he would fail to be a GSWP.

    Now, here comes the de re/de dicto business: the GSWP is, if internalism is true, motivated de re to do what is right. I think we can agree that Smith has gotten us that far.

    If externalism is true, though, could the GSWP be motivated de re to do what is right? It’s hard to see how, given that the connection between the normative reason (the ideally rational disposition to do what is right) and the motivating reason (the actual disposition to do what is right) is a contingent matter under externalism, but should be reliable in a GSWP. The moral fetish — motivated to do right de dicto instead of de re — fits the bill as far as providing a reliable bridge between judgement and motivation, but fails, says Smith, because it is a fetish and thus a vice, hence contrary to the nature of the GSWP.

    So, overall, I think it’s down to the way the way we understand the GSWP. It’s not that an externalist has absolutely no resources to claim that a person can be motivated by the right de re. They certainly can be. The problem is when we’re confronted with the GSWP. [Aside: I suspect the GSWP is invoked by Smith as a counterpart to the amoralist. There’s certainly, to my eye, structural similarities between the arguments.] The GSWP reliably does what is right: his dispositions are usually in accord with the dispositions of his ideally rational self.

    The internalist can accommodate this within the analysis of moral judgement. A moral judgement carries motivation within itself; so, a GSWP, who usually has correct moral judgements will usually be motivated to do the right thing. The externalist’s analysis of moral judgement has a real problem, though, because they have to account for why what they claim to be a contingent matter — the connection between judgement and motivation in the GSWP — is nonetheless reliable in the case of the GSWP. As said, it seems that all Smith can come up with is the moral fetish of de dicto motivation, and that’s just no good.

    If the externalist were to claim that a GSWP can be reliably motivated by the right de re, I suspect Smith would ask for an explanation of the reliability of this connection between the GSWP’s judgements and his dispositions to act. I suppose one could look for something in the neurobiology that accounted for the reliable connection, but it would then be hard to see the GSWP as a GSWP: we’d end up with a person who was just “wired” such that particular judgements caused particular bodily movements. The virtues of the GSWP seem to drop out of the picture.

  6. 6. Posted by Jussi Suikkanen | July 17, 2007 7:11 pm

    Hi Gerald,

    sorry I’m a bit late on this. I want to have a go at the argument for what it’s worth. Start from the idea that moral judgments are judgments about reasons. Now, the internalist wants to say that there is a connection of rational (or conceptual) necessitation between the latter judgments and the de re desires. Smith gives an explanation for this in terms of the content of the reason judgments – that’s what A+ would advice you to want.

    At this point, the idea is that coherence and a disposition towards it is constitutive of rationality. Unless you desire the object de re your psychology is less coherent and thus more irrational. So, if you are rational, your disposition towards coherent psychology kicks in and produces the right de re desire. Lack of that desire, on the other hand, marks you off as irrational.

    So, why cannot the externalist say the same? Because, according to them, the reason-judgment does not rationally necessitate the de re desire. Once rationality or coherent psychology is not in question when one has made the reason-judgment and lacks the de re desire. But, often we get the de re desire. How? Well, if we first desired whatever there is reason to do de dicto, then it would be irrational for us not to have the de re desire that would cohere with that general desire. In this case, the desire can be explained through a rational process. The only problem is that the de re desire is derivative from the general desire for whatever is reasonable. It’s not what the agent directly cared for in the first place – it’s just an instrumental desire to satisfy the more central desire in the agent’s psychology. That’s at least how I’ve understood the argument. I do recommend Teemu’s excellent paper on the topic from the Aristotelian Society Proceedings few years back.

Post a comment

Name: (required)

Email Address: (required) (will not be published)