Part 1, for those of you old enough to remember, was posted two years ago here. Yes, I'm a very slow thinker.
In Part 2, I want to explore the possibility of extending Scanlon's account of blame to include criminal blame. For both legal and moral theorists writing on these issues, criminal responsibility (and the intelligibility of criminal blame) entails moral responsibility (and the intelligibility of moral blame). I have begun to believe that this assumption is just false, or at least the relation between criminal and moral responsibility is far more complicated than people have believed. In order to explore this idea, I want to look at recent, plausible accounts of blame to see how, if at all, they might ostensibly explain both moral and criminal blame. I begin today by discussing one worry about doing so with Scanlon's view. I hope (sooner this time!) to discuss another worry of doing so, and if I'm energetic enough, I'll then turn to George Sher's account of blame to explore the same issues. If anyone's still awake by then, I may offer a diagnosis of the problem. Warning: long row to hoe ahead!
Start with a brief review of Scanlon's account of blame, according to which my blaming you for Φ rests on a judgment of your blameworthiness for Φ (a judgment that you did something to impair our relationship) and consists in my modifying my attitudes and dispositions with respect to you in a way rendered appropriate by that judgment. So how might this work with respect to state sanctioned punishment of criminals? (Assume that there’s some rough sense of the “state” as being the relevant blaming agent here.) When the state punishes someone for a crime, it judges him to be guilty—-blameworthy—-for that crime, and then, in the standard case, coercively restricts his freedom in response to that verdict, presumably in a way that is rendered appropriate by that verdict (so that, for example, the amount of punishment dished out is “proportional” in some sense to the crime). This may seem tightly analogous to, if not a direct instantiation of, Scanlonian blame, for it at the very least involves the state’s modifying its ordinary dispositions to leave citizens alone to live their lives as they please (to the extent that doing so doesn’t interfere with others doing the same). It also may involve modification of future dispositions, e.g., the default disposition to let its citizens vote. There also may be a sense in which the state’s attitudes have been modified, such that the punishment expresses its disappointment or anger at what the defendant did (perhaps better: it expresses our disappointment or anger?).
There remain serious questions, though, about the extension. In particular, can a Scanlonian extension to criminal blame preserve the two elements fundamental to his account, namely, its emphasis on relationships and meaning? Here I'm just going to discuss the former. Blame, for Scanlon, is a response to relationship impairments, to violations of the norms and expectations that are constitutive of various relationships. The account seems most plausible with regard to intimate relationships—-friendships and loves—-for these are often long-standing and so thoroughly integrated into the very framework of our lives that their impairment is a truly personal blow, and so it can explain how blameworthy activity has a special significance within these relationships that does not extend to outsiders. These relationships are often constituted by norms and expectations specific only to the parties of the relationship—-determined precisely by the parties’ actual attitudes toward one another-—and so the forms blame can take may vary as wildly as the forms personal relationships can take, a fact reflected throughout our blaming practices. The account thus may seem to get less plausible as we move beyond this personal realm to the purely “moral” relationship in which we allegedly all stand to one another in virtue of our capacities for “understanding and responding to reasons.” Here the way in which we stand to one another isn’t determined by any special attitudes we may actually have, and so it may seem that there just is no “relationship” involved for which the relationship-impairment story Scanlon gives could be true. What he says in response to this worry, however, is that relationships don’t always have to depend on any actual attitudes between the parties to generate norms and expectations (e.g., the parenting relationship); rather, the existence of relationships with this generational normative power can merely depend on facts about the parties (parents are required to care for their children in virtue of the fact that they are their children and are dependent on them). And so it may be with the moral relationship: “morality requires that we hold certain attitudes toward one another simply in virtue of the fact that we stand in the relation of ‘fellow rational beings’” (Moral Dimensions, p. 140). We may thus impair this relationship—-defined, for Scanlon, in terms of a kind of mutual concern—-with strangers, just as we may do so with friends, and blame again takes the form of modification of attitudes and dispositions rendered appropriate by the relevant impairment, ranging anywhere from indignation to refusing to enter into business exchanges to failing to take pleasure in the other’s success.
Suppose we grant the extension of the relationship model to include the impersonal moral relationship. Can it also extend (or at least apply) to the criminal realm? If so, what “relationship” does the state stand in to those it “blames”? These are hard questions, not least because it just isn’t clear precisely who the parties of the “legal relationship” would even be. To the extent that the state “is us” in a democratic society, perhaps we might try to say that the criminal has impaired some sort of relationship that all of us qua citizens bear to one another, such that the response rendered appropriate by the impairment is punishment. Call this the "citizenry relationship." What might such a relationship consist in? First, would it just be identical with the moral relationship Scanlon articulates? It couldn’t be. For one thing, there would be jurisdictional differences. It is appropriate to punish only those who have violated laws under our state’s jurisdiction (whether it be municipal, state, or federal), whereas it could well be appropriate to morally blame anyone anywhere. After all, it is only our fellow citizens who may appropriately be subject to our punishment response. For another thing, as the case of strict liability reveals, we punish for violations of the law that do not count as moral impairments: if one engages in “consensual” sex with a minor that one had no way of knowing was a minor, one simply isn’t guilty of violating the demands for mutual concern that, roughly, constitute the norms and expectations of the moral realm, i.e., one just isn’t at moral fault in a genuinely unwitting statutory rape case.
More fundamentally, though, the facts giving rise to criminal norms and expectations are just different from those giving rise to the Scanlonian moral norms and expectations. In the latter case, the relevant facts are about the capacities of the parties, namely, their abilities to understand and respond to reasons generally. When these conditions are met in two people, they relate to one another as “‘fellow rational beings,’” and it is ostensibly in virtue of these conditions that the norms and expectations of the moral relationship are instantiated between them. In the criminal realm, what are the relevant fact(s) giving rise to the legal norms and expectations? They certainly aren’t facts about our facilities with reasons. Instead, they have more to do with external facts about our status as citizens and adults (where that borderline is specified at different points with respect to different laws and locales). There are also internal facts about competence that matter here: competence to stand trial, or competence with respect to grasping the law (where someone lacking these competences could still have a basic facility with reasons). Our question, then, is this: if these external and internal conditions were met in two people, would this be sufficient to put them in a “citizenry relation” with one another defined in terms of the norms and expectations articulated in the law?
I cannot see that it would, but the reason for this is surprising: there is no such relationship, at least in the terms we would need to extend Scanlon’s story about blame to the realm of criminal responsibility. What makes blame appropriate is one party’s impairment of the relevant relationship, be it a friendship or a moral relationship. This impairment consists in lacking the sorts of attitudes demanded by the norms of friendship or moral relating. But in any event, we may think of these norms as providing the basic structure for our relating to one another, so that when one party violates the norms it weakens the structure of that relationship in some respect. This is not to say that the content of the norms will always consist in rules for how the parties deal just with each other—it might specify, for instance, how the parties are to deal with entities outside the relationship as well—but the norms at least must provide tendrils between the parties, articulating the variety of ways in which they are connected. This way of thinking helps clear up the notion of impairment a bit. When the norms are in place and adhered to, they provide connections that are weakened via transgression by one or more parties. The modification of attitudes and dispositions in which blame consists is thus appropriate insofar as it really involves a re-envisioning of the nature of the relationship itself, seeing it more clearly—if it is to go on—as different, weaker, than it was before. Consequently, what distinguishes the norms of a relationship from regular old non-relationship norms (e.g., “Men should take their hats off in church”) is that the former are directed obligations, whereas the latter are not. That is, the former, but not the latter, are owed to others, such that those to whom things are owed have a claim on those obligated to fulfill their obligations. These directed obligations are symmetrically mutual in most relationships, though not all (parenting is one example where they are not). Having a directed obligation connects one to those to whom the obligation is owed; violating such an obligation impairs that relationship by weakening the relevant connection.
Are there such directed obligations in the criminal law? I fail to see that there are, although I admit that this is a controversial position. When, contrary to the legal prohibition against stealing, I steal some money out of a store’s till, the legal wrongness of my doing so does not consist in the store’s proprietor having a claim against me that I not steal from her. This is hard to see, of course, because in such a case legal obligation overlaps with moral obligation, so it’s easy to conflate our thought that the store’s proprietor does indeed have a claim against me that I not steal with the illegality of doing so. To see the case more clearly, then, we have to focus on a couple of things. First, suppose that I steal $99, and suppose further that, due to various pragmatic difficulties in prosecuting petty crimes, it is only illegal in our society to steal items worth $100 and above. Nothing about the proprietor’s claim against me would seem diminished or changed in such an imagined society, despite its not contributing to an illegality there. And adding the criminal illegality to such theft in our actual society fails to involve any additional claims, as far as I can see. Would, for example, others in society have a claim on me that is distinct from the proprietor’s moral claim? This would be quite odd, stretching as it does the notion of directed claims to the breaking point: surely I don’t legally wrong others when I steal from the proprietor.
Second, consider criminal regulatory obligations, those involving no mala in se (evil/harm in itself), e.g., tax fraud. These are very clear cases of nondirected duties: my obligation not to cheat on my taxes isn’t owed to anyone; no one has a claim on me not to do so. It would seem, then, that criminal obligations with respect to mala in se crimes need not be thought of in any way as a fundamental difference in kind from criminal regulatory obligations, as suddenly a codification of directed duties; rather, it makes much more sense to think of what’s going on simply as an extension of nondirected duties to harm or victim-based cases.
If this is right, then the norms of the criminal law do not create a lattice of connections between citizens, for they don’t have the structure of directed duties. This means either that there is no citizenry relationship or any relationship that does obtain is not of the sort that grounds Scanlonian blame. Absent the relationship model within which Scanlon’s notion of impairment makes sense, criminal punishment cannot count as (Scanlonian) blame.