Author Replies to Critics:
Sigrún Svavarsdóttir, Ohio State University
email: [email protected]
Posted 08/28/01

would like to thank the commentators for their interesting and challenging discussion of the ideas set forth in my "Moral Cognitivism and Motivation" (MC&M from now on). My main objective in that article was to undermine the so-called internalist constraint on accounts of moral judgments. I am happy to see that at least one of the commentators, Jay Wallace, thinks that I have successfully done so, and another, John Skorupski, at least agrees with the conclusion of my main arguments. However, both Wallace and Skorupski take issue with the view of moral motivation-motivation by moral judgments-that I set forth in the article. They pose formidable challenges to my view that require a much longer response than I can give here. Some tentative suggestions will have to suffice. However, first I would like to address Robert Johnson's criticism that targets my main argument against the internalist constraint.

Johnson seems to think that it is problematic to focus on a disagreement between internalists and externalists concerning the possible explanations of the conduct of a single individual, Patrick, on a particular occasion. Johnson insists that since my argument appeals to a methodological principle governing empirical investigation, I should focus on a disagreement between internalists and externalists on how to explain some pattern of phenomena rather than a particular event. This is because "[i]n an empirical investigation, hypotheses are not evaluated by their capacity to explain a single phenomenon. They are evaluated by their capacity to explain patterns of phenomena" (Johnson).

But surely, we may, in empirical investigation, seek to explain particular events: a certain mysterious death, a particular downturn in the economy, the collapse of a bridge, a specific volcanic eruption, etc. In evaluating the possible explanations of a particular event, it is crucial whether the explanatory hypotheses have more extensive explanatory power: whether they, or assumptions on which they rely, can help to explain a whole pattern of observations. If Johnson is making this point, then he is perfectly right. But this does not indicate any flaw in my argument. The problem I raise for internalists has nothing to do with the fact that they must make a case for their explanation of Patrick's conduct by arguing that it is in line with plausible explanations of a whole pattern of observations. Instead, the problem turns on the fact that internalists are committed to excluding the externalist explanation as not even a serious contender. I argue that the burden is on internalists to defend this out of hand dismissal of the externalist explanation. Moreover, I argue that this would have to be done by showing that motivational internalism falls out of the best account of moral concepts and that internalism cannot, therefore, function as a constraint of adequacy on accounts of moral judgments. (I have some problems with Johnson's summary of my argument. For example, it does not rely on the claim that the internalist is committed to rejecting the example of Patrick as incoherent. Indeed, I have carefully presented the example in observational terms acceptable to both internalists and externalists. For a good summary of my argumentative strategy, I refer to Wallace's paper. I also urge the reader to review the presentation of my strategy in section 3 of my article.)

I wonder whether Johnson's dissatisfaction with my argument is really with my claim that the internalist is committed to rejecting the externalist explanation out of hand as incoherent. This would be in line with his worry that "Svavarsdóttir['s] strategy may involve a portrayal of internalism that is not entirely fair" (Johnson). The internalist thesis that is my target-motivational internalism-claims that moral judgments are of conceptual necessity connected to motivation. Surely, this is a familiar position from the metaethical literature, although I do not claim that it is the only thesis deserving of the name 'internalism.' So, what is unfair with my portrayal of internalism? Johnson clearly has in mind that it is unfair to interpret C. L. Stevenson, and kindred spirits, as appealing to this internalist thesis as a condition of adequacy on accounts of moral judgments. (His remarks seem to imply that I have ascribed this thesis to Michael Smith, but that is not true. See MC&M:164.)

I strongly disagree with Johnson's interpretation of Stevenson. In "The Emotive Meaning of Ethical Terms", Stevenson lays down the magnetism requirement as one of three requirements that accounts of 'good' must meet in order to capture the sense of 'good' that is most typical of ethics. Although Stevenson's formulation of the magnetism requirement does not use any modal notions, it is clear from the context that Stevenson intends this as a truth that falls out of the sense of 'good' that is most typical of ethics-that is, as a conceptual truth. This interpretation of the magnetism requirement is further supported the way Stevenson himself applies the requirement when ruling out the view that 'good' means approved by most people. Stevenson takes the mere possibility that a person is not affected by the opinion of the majority to show that this version of subjectivism violates the magnetism requirement. This contention would not make sense if the requirement is merely, as Johnson suggests, that an account of the ethical 'good' should be able to explain the pervasive motivational impact of judgments employing this notion. Also, if Johnson's interpretation were correct, Stevenson should have considered the possibility that, due to conformist tendencies or a need for the approval of others, judgments about what is approved by most people have pervasive motivational impact. Indeed, if the internalist requirement were as weak as Johnson suggests, I do not understand what has been at issue between internalists and externalists about the connection of moral judgment to motivation. No externalist would deny that the connection between moral judgment and motivation is pervasive.

Of greater interest is Johnson's attempt to turn my example of Patrick against myself. Johnson thinks that Virginia will launch into a blow by blow description of the injustice Patrick and she are up against, much as if she was still trying to convince Patrick of the moral assessment that he has apparently already accepted. Indeed, Johnson thinks that we all will react in this way to cynics like Patrick. According to Johnson, such behavior demonstrates that it is integral to our conception of moral judgments that they are necessarily motivating. Whatever our proclaimed stance in a philosophical discussion, our behavior in circumstances like Virginia's shows that we cannot conceive of people like Patrick as having genuinely accepted our moral judgment. In other words, motivational internalism is true.

I am not convinced, though I accept the challenge. First of all, I doubt that this is the typical reaction to the Patricks of this world. My guess is that a lot of us give up on a lot of them. Second, even if this were the typical reaction, I think there are externalist explanations to compete with Johnson's internalist explanation Virginia�s behavior. In section 4 of MC&M, I attempt an explanation of the pervasive appeal of internalism. I acknowledge the puzzlement that we are apt to experience when encountering people like Patrick and seek to explain it on externalist premises. Some of the puzzlement, I speculate, might have to do with the chilling amazement at people who are so emotionally impoverished that they do not connect motivationally with the kind of considerations that drive moral evaluation. Indeed, it would not be a surprising reaction on the part of the morally committed to start to go through these considerations in the hope that Patrick will emotionally connect to them with the consequence of overcoming his moral cynicism. This provides one possible explanation of the putative behavior of the Virginias of this world. I also speculated that some of the perplexity has its roots in the reasonable assumption that an individual who bothers to make a moral judgment possesses some degree of moral commitment. That assumption and a reluctance to give it up could also be manifested in the sort of behavior Johnson highlights. This provides another possible explanation.

I expect that Johnson finds the internalist explanation more plausible than the two just suggested. I will not try to dislodge that conviction here. Overall, I find the externalist explanations of the various phenomena that bear on this dispute more plausible than the internalist explanations, but I emphasize, once again, that my argument against the internalist constraint does not rest on this assessment. For my argument, all I need is that externalists have an explanation to offer of Patrick's behavior-and, we may now add, of Virginia's behavior. I challenge internalists to give a noncircular defense of a dismissal of that explanation as incoherent.

It is high time to move this discussion beyond motivational internalism and grapple with the issues raised by Skorupski and Wallace. Although neither contests my argument against the Stevensonian internalist constraint, they both articulate and defend an alternative internalist thesis that calls into question my view of moral motivation. Skorupski's tactic is to link judgments of duty or obligation with judgments of blameworthiness, and to link the latter in turn with judgments about what one has the most reasons to do. Skorupski's internalism claims that judgments about what one has the most reasons to do are typically, though not necessarily, in and of themselves motivating to some degree. Since judgments about moral duty and obligation are most often linked to such judgments in the way explained above, they need not engage a desire to be moral in order to be motivating.

The first link is supposed to be conceptual: anyone who understands the concept of the moral will accept that if a person fails to do what he has a moral duty to do, then he is blameworthy, i.e., can be reasonably blamed. The second link is forged via the following conditional: if it is reasonable to blame a person for what he did, then what he did was not what he had the most reason to do. Skorupski suggests that a rejection of this conditional manifests the cynicism of regarding blaming simply as an expedient way of controlling or manipulating people. I am not sure I accept either link as characterized by Skorupski. Of course, there is a long and strong tradition for linking moral duty and obligation with the blameworthy. But is it incoherent to hold that blaming is an unreasonable thing to do under any circumstances, while retaining the notion of our having duties or obligations that we should be encouraged, in other ways, to honor? And is it incoherent to advocate a view of reasons for action according to which a person might have had the most reason to do something, even if it is reasonable to blame him for doing it? Fortunately, we need not settle these matters here because Skorupski's objection to my view of moral motivation does not really turn on his claims about the nature of these links. I certainly have to grant that an individual may hold substantive views in virtue of which his judgment about his moral obligations is linked to his judgment about what he has the most reason to do via his judgment about what is blameworthy. So, if Skorupski's internalism is correct, then a moral judgment need not engage the desire to be moral in order to be motivating. Skorupski's target is the strong claim that the desire to be moral is (psychologically) needed for motivation by moral judgments.

To respond to this objection I have to address issues concerning the nature of judgments about reasons for action, their relation to moral judgments, and their connection to motivation. These are issues I deliberately avoided in MC&M. I will not be able to develop or defend an account of such judgments here. Let me, however, go out on the limb and air some tentative ideas.

Accepting that an evaluative judgment provides one with a (non-derivative) reason to act amounts, first and foremost, to a readiness to have it enter into one's practical deliberations, readiness that manifests a commitment to the relevant value domain. Having such a commitment involves having a cluster of conative and emotional attitudes, at least some of which are engaged by judgments distinctive of the value realm in question. The motivation that is linked to the acknowledgment that an evaluative judgment provides one with a reason for action is best seen as stemming from these conative states. In other words, such judgments about what one has reasons to do are strongly connected to motivation because an agent would not make them unless he was committed to the values that are seen as grounding these reasons. If the value domain in question is morality, the motivation stems from the desire to be moral and other attitudes integral to the moral commitment. The desire to be moral is, therefore, involved in moral motivation, even when moral judgments are linked to judgments about reasons for action. To this, I should add that agents who pass from a moral judgment to the judgment that they have the most reason to do what that judgment favors demonstrate that they have an overriding commitment to morality. Of course, I owe a fuller account of moral commitment, as well as a development and defense of my ideas about judgments about reasons for action and relations of these judgments to evaluative judgments and to motivation.

Early on in his essay, Skorupski poses a question for me. He asks why the moral disposition (the disposition to be motivated by one's moral judgments) has to be grounded in some one type of intentional state, or even in any intentional state at all. This is a fair, deep, and difficult question that presses us towards thinking more generally about the nature of motivation. I agree with Skorupski that it would be silly to think that any variation in how one is motivated by a judgment is best explained by postulating a desire engaged by that type of judgment. The variation between scenario A and B in Skorpuski's example of the driver is not plausibly explained in this way. And the variation in moral motivation between Virginia and a severely depressed person is not plausibly explained in this way. I readily grant Skorupski that the fact that the motivational impact of a judgment may be affected by factors other than the judgment itself does not alone justify the claim that a desire is involved when the judgment motivates. However, certain variations in motivation generally, and in moral motivation more specifically, are best explained by postulating a variation in an underlying conative state. These variations in motivation are manifested in patterns of similarity and difference not only in conduct, but also in various cognitive and emotional dispositions. Having a desire is not merely a matter of being disposed to behave in a certain way, but also of being disposed to experience life in a certain way, think in a certain way, and emotionally react in a certain way. Admittedly, I need to identify the features of such patterns that indicate that the variation in how a judgment motivates should be traced to a variation in an underlying conative state. This is a major task and cannot be fully completed without digging deep into the general issue of motivation. It would, therefore, be premature to consider my view of moral motivation established. All I claim for my discussion in MC&M is that it has shown that the view has some plausibility as a working hypothesis.

Wallace objects to that hypothesis on the grounds that it makes "moral motivation seem altogether optional and arbitrary" (Wallace). It goes against "the intuition that moral judgment is non-contingently related to moral motivation" (Wallace). According to Wallace, this intuition is the starting point for internalists and should be honored. He claims that the version of internalism that I target in my paper, though not a straw man, does not do justice to this intuition, and he seeks to articulate a more adequate version of internalism that does not fall prey to my argument. His version of internalism-his motivational requirement-reads as follows: "agents are necessarily motivated to act in accordance with moral requirements, to the extent they are deliberating correctly and are otherwise practically rational" (Wallace). My account of moral motivation is supposed to be at odds with this requirement, since it makes moral judgment contingently related to moral motivation.

My view makes moral motivation out to be a conceptually and psychologically contingent matter: it is psychologically and hence conceptually possible that moral judgments fail to motivate an agent who is fully rational and deliberating correctly. Wallace makes clear that he has no problem with this claim. The modality in his motivational requirement is deontological: the normative necessity of must or ought. But why is my account of moral motivation then in conflict with the motivational requirement? Assume that we ought, as rational beings capable of correct deliberation about moral matters, to be motivated by our moral judgments; if we are not, we can be criticized for irrationality or for deliberating poorly. Assume also my view of moral motivation: the desire to be moral grounds moral motivation. It follows that we fail a normative requirement unless we have the desire to be moral. We need to have it in order to have the psychological dispositions we ought to have. So, from the normative perspective that gives rise to the motivational requirement, this desire should not "seem altogether optional and arbitrary," even if it is a purely contingent psychological fact that one has it. (Compare: from the normative perspective of theoretical rationality, it does not "seem altogether optional or arbitrary" that one form the belief that q from the belief that if p then q and the belief that p; however, it is an entirely contingent psychological matter whether one forms that belief in the given context.)

Related to the above objection is Wallace's complaint that my account of moral motivation (like the intuitionistic account) "leaves it entirely open whether the considerations that move us are ones that merit our interest and concern; in this respect, morality might well turn out to be a kind of collective fetish", i.e. "an irrational investment of interest and attention in objects that are not intrinsically worthy of such a response" (Wallace: 9). This observation is perfectly right, but I do not see how this is a problem with the view. I do not think that we should expect normative insights from an account of the psychological states underlying moral motivation. As I acknowledged when setting forth my view (MC&M: 171), it is a crude psychology in that it gives us the most superficial understanding of why someone is motivated by his moral judgment, and it certainly gives us no understanding of the merits of being so motivated. I agree with Wallace that we need to understand the nature and substance of moral judgments, including their normativity significance, in order to gain such normative insights. The issues to which I limit my attention in MC&M are not the only issues moral philosophers need to address.

I doubt these responses will satisfy Wallace. He contends that an investigation into the normative dimension of morality yields the motivational requirement and reveals shortcomings with my view of moral motivation. Let's, therefore, review how Wallace comes up with the motivational requirement. He infers it from two assumptions. The first claims that it a condition of rationality that one is motivated in accordance with judgments that one acknowledges to have normative significance. This assumption, as I understand it, lays down a normative condition for evaluating an agent's practical rationality. The second assumption claims that an agent is not deliberating correctly unless he acknowledges the normative significance of moral requirements that he acknowledges as obtaining. Wallace eventually offers support of this claim in the form of another assumption, namely that moral considerations have, in fact, normative significance. He hesitates to assign to this claim the status of a conceptual truth, allowing that it is possible to accept the truth of a moral judgment while denying that it has normative significance. However, Wallace is clearly committed to the view that the normative significance of moral requirements is accessible to anyone who deliberates correctly.

How are these ideas at odds with the view that the rationally required motivation, when forthcoming, is rooted in a desire? Why couldn't the motivation that should come with acknowledging that a consideration has normative significance be rooted in a desire? Wallace claims that my view of moral motivation cannot get right the intentional perspective of the person who acknowledges the normative significance of moral considerations and is thereby motivated to act on them: those who are morally motivated "characteristically take it that their actions are recommended or justified by the fact that they are morally required or otherwise morally valuable, and this in turn renders their moral motivation intelligible to them, as a fitting response to the considerations that they take to have normative significance. This is something that gets left out on the interesting reconstruction of the intentional perspective of the moral agent that Svavarsdóttir offers." (Wallace). This is an important challenge to my account of moral motivation.

Wallace is completely right to observe that by postulating a desire to be moral to account for motivation by moral judgments, I seem to be reconstructing the intentional perspective of people who are morally motivated as that of people simply attracted to the idea that they do the morally right or good thing. And I concur with him that there is, at least often, more going on. Of relevance here, is an aside that I make in MC&M: "I believe an agent can see himself as doing something because it is ß without thinking that its being ß carries much justificatory force or gives him anything like a [normatively loaded] reason for action" (MC&M:202, n. 60). This would be the case when, desiring something sweet, I reach out for something I think of as sweet. But Wallace is right that when an agent acknowledges the normative significance of moral considerations, he sees himself as acting on a consideration that justifies his action, that gives him normatively loaded reason for action. His intentional stand is different from the agent who is motivated by a judgment he does not acknowledge as having any normative significance. Thus, the intentional states, which underlie their intentional perspectives, must differ. However, I am unconvinced that this demonstrates anything more than that my view of moral motivation is incomplete.

I agree with Wallace that elucidating the purported normative significance of moral considerations is one of the most important issues in moral philosophy. MC&M was, indeed, meant as a modest step towards approaching that difficult issue. My goal was to create room for novel approaches to understanding the purported normativity of moral requirements and the action-guidingness of moral judgments without running into facile objections from motivational internalists. The idea that normativity is to be understood in terms of the motivational role of moral judgments has been prevalent in the metaethical literature, but it has always struck me as wrongheaded. This is not to deny that an account of the purported normativity of moral considerations and an account of the pervasive motivational impact of moral judgment have to go hand in hand. And this is not to deny that a full account of moral motivation has to incorporate an elucidation of how an acknowledgment of the normative significance of moral considerations interacts with a commitment to morality, incorporating the desire to be moral, so as to give rise to an intentional perspective quite different from that of the person who fancies something sweet.