II. CRIMINAL OFFENSES
Cases of what appear to be harmless wrongdoing raise a well-known difficulty for the harm-oriented, utilitarian conception of crime. But though the difficulty is broadly recognized, it usually arises in the gray areas at the periphery of criminal liability; when it comes to core crimes, such as homicide, battery, or rape, it seems that harm plays a decisive role. But this is not always the case. One category of felonies, rape by deception, provides a conspicuous example. Consider State v. Minkowski,6 in which the defendant, a gynecologist, was accused of raping during medical examinations a number of his female patients, who on recurrent visits had not realized what was going on. Most would agree that these women were indeed raped even before discovering the violations, despite the fact that at that stage it would have been difficult to find either physical or psychological harm. It is easy, of course, to condemn the defendant’s conduct and justify criminal conviction on obvious rule-utilitarian grounds. But doing so would miss the target. Minkowski’s actions are reprehensible acts of rape all by themselves, and they should be treated as such out of concern for the unsuspecting victims, quite apart from any broader welfare-related ramifications of condoning Minkowski’s conduct.
Minkowski wronged his victims even if he did not harm them. How? It is natural to answer this question by appeal to the value of autonomy. Even in the absence of harm, the familiar story goes, the women were wronged because they were subjected to nonconsensual sex, in derogation of their autonomy. But this account runs up against another familiar conundrum: the limited role that victims’ consent plays in assessing criminal liability. In State v. Brown,7 for example, the defendant habitually beat his wife when she drank alcohol, allegedly as part of an agreement to help her overcome her alcoholism. In convicting Brown, the court rejected a defense of consent. How are we to assess this decision? Once again, it is easy to marshal public-policy arguments, this time oriented toward the protection of autonomy. There is, for example, good reason to be suspicious in general of agreements such as the one alleged. But here, too, as in Minkowski, we face the gap between the generic offense to which such considerations pertain, and the specific token, to which they may not. Even when we consider this case in isolation and assume that in this particular instance the wife did consent, we might still conclude that the beating is wrong.8
If neither the notion of harm, and relatedly human welfare, nor the notion of consent, and relatedly individual autonomy, by themselves provide a satisfactory account for these cases, what does? An account of criminal offenses based on the notion of human dignity can close the gap. The puzzle presented by Brown can be solved by recalling the discussion of slavery in the previous chapter. I have argued that slavery is an affront to dignity even when it does not adversely affect a particular slave in other ways: the meaning of slavery denies the slave’s equal moral worth. How does this meaning accrue? One answer is that a meaning attaches to an action or a practice by virtue of certain empirical characteristics and consequences it typically has. After all, it is not a mere coincidence that slaves’ lives are usually wretched and their autonomy is trampled. Exploiting a person by disregarding his own needs, interests, and desires is a paradigm of disrespect. But though the meaning of slavery has an empirical basis, the meaning that attaches to slavery as an insult to dignity is retained even in the situation we imagined, in which the typical derogatory effects on the slave’s welfare and autonomy are stipulated away. A similar account applies to the Brown case. The fact that physical violence ordinarily hinders both welfare and autonomy is reason enough to render it a blatant manifestation of disrespect. But here, too, the meaning of violence can outrun the reasons for ascribing that meaning to it. Although for the most part their significance is not attached to actions arbitrarily or at random, the connection between the reasons for ascribing to an action- type its significance as expressing disrespect and the tokens of that action need not be tight. Once an action-type has acquired a significance by reason of the disrespect it typically displays, all tokens of that action will possess the same significance and communicate disrespect even if that reason does not apply to them.
These observations buttress our judgment that the beating inflicted in Brown involves an affront to the wife’s dignity despite her consent: when it comes to the meaning of these actions, the typical case of violence casts its shadow over the exceptional, thus amounting to an affront to the victim’s dignity. But this explanation raises a further query: how widely is that shadow cast? Two kinds of examples will serve to illustrate the problem and help complete the account: medical treatment and contact sports. In both areas, the level of permissible violence far exceeds the level, if any, that would be tolerable in a Brown-type scenario. Why does the nasty record of the typical case of violence define the meaning of the beating in Brown, but not that of surgery or a boxing match?
Recall the case of Mary, discussed in the previous chapter, who cuts open John’s chest and mutilates his body in countless other ways, resulting, let us now suppose, in John’s death.9 A grisly homicide? Far from it; just an ordinary, if unsuccessful, open-heart surgery. No district attorney is ordinarily alerted to such a case or would take notice of it. But why? Isn’t Mary’s action a prima facie case of assault, and given the results, of homicide as well? The prima facie case would not, of course, end the DA’s inquiry, only trigger it. But once the inquiry is started, applying to this case the requisite criminal law doctrines is incongruous, and the results are uncertain. Given the initial charge, Mary must be able to raise some defense, and two seem to be available to her: consent and justification. But as Brown reminds us, consent does not in general exonerate when serious assault or homicide is involved. She may fare not much better under a justification defense. Depending on its precise formulation, the lesser-evil standard may not be met in this case, since as things turned out, the operation ended up shortening the patient’s life rather than prolonging it as hoped. Though doctrinal waters in this area are too muddied to allow for any general and definitive statements, it is at least possible to view this ex post circumstance as marking the line between a genuine case of justification, and one in which the justification is only partial or shifts the burden of proof. These considerations are unsettling. Bringing to bear on medical treatment the doctrinal machinery of the criminal law would result in doctors routinely scrambling for cover. But as we have already noted, in practice, the first step that would lead us down this treacherous doctrinal road is not in fact likely to be taken: we ordinarily refrain from subjecting medical treatment to the categories of the criminal law. But how can such a gruesome affair as I’ve described fail to at least induce an initial criminal investigation? Part of the answer is linguistic. Surgeons don’t use knives but scalpels; they don’t slash or rip their patients’ organs, but perform incisions instead; and so on. These linguistic markers together with other factors, such as the surgeons’ green drab, invoke a comprehensive system of images, attitudes, and norms that constitute medical practice, thereby erecting a barrier, at once conceptual and psychological, that separates the goings-on within the practice from what happens outside. Consequently, the meaning of what Mary does in performing surgery is radically discontinuous with the beating administered in Brown, and the nasty connotations of violence do not ordinarily apply.
Boxing (and other contact sports) can be distinguished from Brown along similar lines. A case of consensual wife-beating is still a case of wife-beating, and it draws its offensive meaning from the typical cases in which there is no consent. But that meaning does not carry over to a practice we recognize and label as “boxing.” Moreover, since within boxing violence is not demeaning to the participants, no disrespect will be conveyed by a boxer’s punches even if the individual boxer holds the opponent below contempt and harbors the most disparaging emotions toward him.
-  This is not to deny, of course, that the theater of operation can sometimes serve as a stage ofcrime. A murderer can dress up as a surgeon, and kill her victim; a surgeon may deliberatelyfumble the operation, etc. But these departures from ordinary practice should be seen for