Printer Friendly

Consent to Sexual Relations.

By Alan Wertheimer. Cambridge: Cambridge University Press, 2003.

     A. Wertheimer' s Central Claims
     B. Problems Internal to the Argument
         1. A brief note on the reliance on evolutionary psychology
         2. Coercion versus the redistribution of power
     A. Is Autonomy Theory Useful in Thinking About Sexual Consent?
     B. Is Welfarism Coherent?


Reading Alan Wertheimer's work is always richly rewarding. He is never the least bit trivial or esoteric; his books invariably tackle vital conceptual issues with direct relevance to legal and social practice. He has for instance, in the past, provided us a cogent account of how to think about coercion (1) and illustrated clearly how distinct conceptions of what coercion is, how being coerced might be distinct from being constrained, or how coercion may not excuse the coerced party of responsibility, play out in a host of legal domains. He has analyzed what has long been (outside the unduly limited, and rather unpersuasive, Marxist tradition) the rather vague concept of exploitation (2) and demonstrated with extraordinary clarity how we can bring these analytical insights about exploitation to bear on questions about unconstitutional conditions, unconscionable contracts, or the propriety of sexual relationships between psychotherapists and their patients. In his most recent book, Consent to Sexual Relations, (3) Wertheimer once more addresses a set of unquestionably significant problems. Why should we legally prohibit or (merely) morally condemn sexual contacts between men and women (4) in cases where the woman has not given any token (in words or deeds) that she consents to the sexual contact? More significantly, when is the "morally transformative" capacity of the consent-token to legitimize the sexual contact compromised by the circumstances in which it is given? That is, when is she (unduly) coerced, deceived, or incompetent (most interestingly because of age, cognitive dysfunction, false beliefs, or intoxication)?

Not only is Wertheimer willing to tackle significant issues, he is also terrifically adroit at clarifying the scope and shape of the problems he addresses. He is both a superb issue spotter and a great organizer. Even if one feels that he's gone astray in his arguments at some point, the reader is almost certain to admire how well he has identified (and clearly illustrated) the range of problems someone who wants to come to grips with a field must confront and how he has laid out the arguments in such a way that one feels that such arguments probably should be confronted in just the order he has chosen. He also has a fabulously no-nonsense style: no matter how politically uncomfortable it might be to dismiss a conventional claim, he'll just go ahead and do it. Moreover, he is all-too-unusually modest about what he has and has not fully figured out: he will often set out a framework for resolving a problem and note, with far more restraint and wisdom than most of us are able to muster, that resolving what the best practice might be requires making quite tough judgments on a host of still-unresolved empirical and conceptual points.

All these virtues characterize Consent to Sexual Relations. Once again, Wertheimer is never dismissive or disrespectful about claims other authors have made, but he remains refreshingly direct and brave. When, for instance, he confronts the frequent, often-politically charged claim that "rape is about violence, not sex," he does his best to understand sympathetically the distinct concepts its proponents might mean, but, in the end, he demolishes many variants of the claim unapologetically, without fear of political incorrectness (pp. 70-80, 91-92). (5) When he analyzes whether cognitively impaired women can give consent to sex that either legally or morally justifies sexual contact for those to whom they give consent (with or without greater restrictions than women who are not cognitively impaired have on the range of men to whom they can give consent or the situations in which they can give legitimizing consent), he concludes with a terrific framework of questions, but nothing like a final answer (pp. 223-26).

Above all, though, both the book in its entirely and each particular chapter are masterfully organized. I can scarcely recall reading a work in which the sequence of questions the author feels should be confronted is more lucidly or persuasively established or one in which the reader can more cleanly recreate how the author has answered those questions. In fact, I will organize Part I of this Book Review by essentially restating Wertheimer's superbly well-structured argument, noting the claims I will not pursue, before providing more extensive comments on several problems I would classify as internal to the work.

And there are problems. In fact, for all its quite considerable virtues, Consent to Sexual Relations can be a remarkably infuriating book, and not simply, I think, because any intellectually provocative book on a politically charged subject might infuriate those who don't agree with all of the author's conclusions. What makes the book so infuriating at times is that for all of Wertheimer's bottom-line modesty and caution--he really is unsure, for instance, whether and when sex with intoxicated or cognitively impaired women is problematic and is very willing to admit it--he can be remarkably smug and non-self-critical about drawing any number of conclusions that warrant a great deal more qualification. At times, the impact of this needlessly casual self-confidence on his overall argument is rather slight. For instance, while I find his nearly three-chapter-long (pp. 37-88, 103-05, 112-18) rehashing of the standard picture of male and female sexuality in fin-de-millennium evolutionary psychology (6) extraordinarily unsubtle and his account of its critics both inadequate and misleadingly focused on their purportedly untoward motivations, I discuss (a tiny portion of) my (quite extensive) reservations only briefly, because this particular picture of human sexuality does (relatively) little work in advancing his most important claims.

At times, though, his penchant for dismissing what he sees as unduly sentimental arguments can get him in real trouble. Wanting (quite rightly) to reject the expansive claim that consent can never legitimize sex if the consenting woman feels subjectively constrained or did not, objectively, have the full range of life-options that would bear upon her decision whether to consent or not--options that she might have in a more just, egalitarian world (7)--he concludes his chapter on coercion by stating: "It is difficult to defend principles that prevent people from consenting to transactions that will move them from an unjust or unfortunate situation to a better situation" (p. 192). But the statement, though brave and clear, is plainly untrue. Wertheimer himself defended just such principles several pages earlier, even in the sorts of complex cases in which consenters face not simple right-violating threats but what might be seen as offers to improve their unjustifiably constrained situations. (8) Similarly, I will note that Wertheimer's account of the capacity of mentally retarded women to consent is needlessly oblivious to his own discussion of deception (9) and discuss in a bit more detail that his willingness to accept certain limits on the legitimizing capacity of intoxicated consent, at least given certain factual presuppositions, is inconsistent with his reluctance to adopt more expansive accounts of when consent is a product of illegitimate coercion. (10)

In the final analysis, though, I think the most nagging problems in the book are simply inevitable given the limited traction we should expect within the standard philosophical frameworks we have available to deal with issues of sexual consent (and political theory issues more generally). It is these limits, external to the book's argument, which ultimately seem most perplexing to me. I discuss this concern in Part II. Fundamentally, I claim that what is available to us are two varieties of "liberal" political theories--theories that emphasize welfare maximization and theories that emphasize the need to protect autonomy by establishing a core of inviolable rights designed to permit individual self-determination by agents acting in accord with their rational faculties--alongside decidedly illiberal "perfectionist" theories, premised on the view that action ought to be taken to ensure that people act "virtuously" or in a fashion that manifests "true human nature" or permits "human flourishing" (though the virtues people should manifest or the pictures of human nature differ radically for distinct perfectionists, both in content and in level of specificity).

Wertheimer is seemingly drawn to both sorts of conventional liberal arguments, discussing ways in which his preferred norms might comport with distinct views of the dictates of welfarism by diminishing experienced harms and comport as well with distinct views of the dictates of autonomy theory. Looking at the book carefully, though, I think one would find that Wertheimer is ultimately rather skeptical that autonomy theorists provide any useful practical guidance, except to the degree that autonomy theory slides into a sort of perfectionism (that Wertheimer most plainly rejects) by claiming that, unless a man has sex with a woman who has particular sorts of reasons for consenting, he is acting inconsistently with her autonomy. I am quite sympathetic to Wertheimer's suggestion that autonomy theory is fundamentally best thought of as either empty or so deeply internally contradictory that it suggests nothing interesting about valid social practice. I am sympathetic as well to his claim that strong perfectionist arguments about what sorts of acceptable reasons there are to have sex are both undermotivated and rather creepy. (11) But I think welfarism is far more infused with a weaker form of perfectionism than Wertheimer acknowledges, (12) and his failure to acknowledge that compromises the ultimate force of most of his claims.


A. Wertheimer's Central Claims

What I see as the core of Wertheimer's argument (13) begins with two observations that are telling--both because they are simultaneously obvious and yet easy to ignore:

(1.) Consent to (heterosexual) sex is an issue because there will be many occasions on which a woman does not want to have sex with each man who desires to have sex with her and because (at least some) men (on at least some occasions) nonetheless desire to have sex with a woman despite the fact that she does not want to have sex with them (p. 38). (14)

2. This dichotomy is a serious issue because women experience undesired sex as acutely aversive (in welfarist terms) (p. 38).

What makes nonconsensual sex both subjectively aversive and morally problematic is at core the way in which women experience it rather than the fact that it violates an abstract fight to autonomy: being kissed when one does not want to be kissed equally violates "rights" to make only those decisions consistent with one's own rational aims and "bodily integrity," understood as a genetic "tight" but is a far less important problem entirely because of distinctions in hedonic consequences (p. 109). (15)

Wertheimer's main preliminary task, he believes, is to explain not only why women don't always desire sex and why men desire it even when the woman with whom they desire to have sex doesn't, but also why we are likely to see frequent clashes between men and women as a result. Why, alternatively, is the gap between the sex men desire and that which women want to have likely to be severe and felt intensely enough that some men may use coercion or deceit, or be willing to have sex with an incompetent person, in order to obtain the sex they desire but could not obtain through ideal consent? Why are men more specifically prone to use violence when their sexual desires cannot be met noncoercively? (In the second view, men are specifically predisposed to rape. In the first, rape is merely an instrumental response to the frustration of a compelling desire--the unsatisfied desire for sex. Raping is akin to using robbery as an instrumental technique to meet the otherwise-unfulfilled desire for material goods (p. 81). (16) His second task is to explain why women experience nonconsensual sex so aversively: one cannot explain the profundity of the reactions simply by noting that people obviously (almost tautologically) will be bothered by something happening to them that they don't want to happen. His claim is that:

3. Evolutionary psychology gives us the tools to understand both why perpetrators act as they do and why women experience nonconsensual sex so aversively.

I set out the author's claims about evolutionary psychology and a bit of my quite skeptical response to the claims in more detail below, (17) but for now, I want merely to note what a modest role this rather substantial portion of the book plays. While it (mildly) bolsters the normative argument that we should try to minimize the level of harm to women by regulating male behavior rather than using social control mechanisms to try to induce women to "get over" their suffering if we find that women's profound reaction to nonconsensual sex is, in some sense, deeply biologically programmed, the truth is that one could proceed with the meat of Wertheimer' s argument merely by noting the fact that some men seek sex with unwilling women, that those men are not (invariably) totally turned off to the idea of sex simply by the women's unwillingness, and that women suffer if the sex occurs, whatever the reason for these plainly observable phenomena. It is of no moment to the ultimate argument, for instance, whether women find nonconsensual sex so awful because, as Wertheimer claims, they (at least during the years in which they could reproduce (18) must be picky in choosing mates who were likely to have both the resources and the desire to ensure that their children reach the age at which they too can procreate (19) (and coercion compromises the capacity to be picky), or, to take just one of innumerable possible alternative explanations, because it is a painful reminder of their relative powerlessness in the world, particularly their sexual powerlessness. (20)

Wertheimer then argues that we should temporarily set aside questions about when a person who indicates her consent to sex might not be in a position to grant the sort of consent that makes sex legally or morally unproblematic (e.g., because of deception or intoxication) so that he can explore why the most unproblematic forms of consent (21) make the decision to engage in sexual contact morally and legally permissible. His argument is that expressions of consent (22) are morally transformative because:

4. Those who have sex with partners who do consent are likely to make those people better off than they otherwise would be; so, setting aside cases in which consensual sex has untoward effects on third parties (p. 131), consensual sex is good because it meets the core goal of ethics: to ensure mutually beneficial interaction (pp. 124-25). (23)

5. Those who have sex with partners who consent are respecting their partner's autonomy in the rather weak ways that Wertheimer finds relevant: they are ensuring that the person with whom they are having sex is self-governing (pp. 126-30) (24) and acting for her own reasons; they are also treating her as an end, not a means, in the relevant sense, by respecting her right to use her unique human facility, the facility to reason, to determine whether and how she will contribute to the events that occur (p. 127).

At the same time, Wertheimer rejects "stronger" views of what is required of a person who seeks to respect the autonomy of others. Wertheimer rejects the ideas that a person illegitimately uses another as a means unless the person consents to sex within a committed marriage relationship (25) or for a particular "acceptable" reason (e.g., to cement a "full" relationship or because he is satisfying her intrinsic desire for sex). (26) Wertheimer, like me, leans toward consensual minimalism: he believes that an autonomous person may consent to sex even though she, in some sense, "doesn't want to have sex" because people may have a "pro attitude" towards something, all things considered, even though they wish they could separately reject the thing they not only assent to, but also seek out. For example, the fact that I may enthusiastically consent to chemotherapy treatment does not mean that I like or want chemotherapy for its own sake (pp. 141, 157-58). Most of the hard problems with consent occur, in my view, not because there is a confined list of good reasons to have sex. Rather, they occur, first, because some of the good reasons to have sex are operative only because we tolerate women having unduly constrained option sets (27) and, second, because it is impossible to judge when we should accept a woman as competent (generally) or competent in a particular situation (e.g., when intoxicated) unless we also impose perfectionist canons of "good sex."

The true heart of the book, in my view, is the discussion of when we should be wary that a woman who has unambiguously given a token of her consent still does not give the sort of consent that gives a man reason to believe that the sex is morally or legally permissible. At bottom, Wertheimer is suspicious of consent that arises from coercion, deception, or incapacity. Although I will argue he is misleading in this regard, he rarely thinks the categories overlap. He claims instead:

6. A person is sexually coerced if she must choose between consenting to sex and moving to a worse position than her baseline. The baseline could be defined in terms of (a) the status quo ante (i.e., does the person seeking her sexual consent permit her to remain in the same position she was in before the request if she refuses consent?); (b) her statistical expectations (i.e., does the person seeking sex permit her to end up in the position that most people in her position would end up in, even if she does not consent to sex?); (c) her subjective expectations; or (d) as Wertheimer would define it, her "moralized expectations" (p. 167) (i.e., will she get everything to which she is morally entitled, (28) even if she refuses to consent to sex?).

Wertheimer rightly rejects the pure forms of the first three possible understandings (29) in favor of the moralized view. He also rightly rejects an understanding of coercion that focuses on the woman's subjective feeling of constraint or her strong compulsion to make a particular choice. We may feel more constrained or compelled to choose chemotherapy for cancer (or even when we choose to consent to sex with a hypothetical lecherous millionaire who offers us a staggering sum to have sex) than we feel when we choose to hand over money to someone who threatens to beat us up, but we are only coerced in the last case.

7. Though Wertheimer does not conflate the "moralized expectations" view with the view that A coerces B in a fashion that negates the legitimizing force of her consent if and only if B's legal rights will be violated if she does not consent to sex, the question of whether it would be sensible to argue that B is legally entitled to an end-state is often critical for Wertheimer, especially, but not exclusively, in distinguishing criminal rape from morally dubious behavior. (30)

Thus, Wertheimer is quite restrictive in defining impermissible coercion given two critical facts: first, because we have few affirmative legal duties to aid one another, A's refusal to do something affirmative for B unless she has sex with him rarely renders consent problematic; (31) second, many forms of injurious conduct (or threats of injurious conduct) are protected by law so long as regulating the threatened conduct directly seems unwise. (32)

While Wertheimer appears anxious to be fairly permissive about the legitimizing force of consent in situations in which some might see problematic coercion, he thinks (as I do) that both the criminal law and, to an even more striking extent, common morality are unduly tolerant of deception (pp. 193, 199, 213). Tort law, which requires proof that a particular putative plaintiff has been harmed by the putative defendant's action, may be tricky to invoke in deception cases because it is almost invariably difficult to prove both "decision causation"--that is, the putative plaintiff would not have made the decision to consent to sex had she known more about the situation she was in (pp. 200-01)--and "injury causation"--that is, the decision to have sex that she would not otherwise have consented to led to any measurable sort of injury--particularly if the injury is not a direct consequence of sex (e.g., a sexually transmitted disease) but of conditionally undesired sex (pp. 201-04). (33) There will also be innumerable proof and line-drawing problems if we attempt to impose tort damages or criminal penalties on those who are responsible for creating (through commission or omission) false beliefs that affect a woman's decision to consent to sex (pp. 198-99). (34) Despite these limitations, Wertheimer is very wary about decisions to consent that are grounded in misinformation.

8. "As a general principle, we might think that A's deception should generally undermine the moral and legal transformative power of consent because it precludes B from being able to decide whether engaging in sex with A is in her interests or compatible with her values" (p. 193).

Finally, Wertheimer deals with capacity. Generally speaking, in deontological terms, a person has the capacity to consent when she is able to make particular focused decisions (35) that meet her second-order reflective preferences (that is, she consents to a particular act X for reasons that make sense to her, given her long-term goals and purposes). In welfarist terms, she has the capacity to consent when she is prudent, so that her predictions about the hedonic consequences of her decision to do act X are likely to be correct. A person is imprudent (lacking capacity) when her decisions are systematically likely to be so wrong that it is probable that her interests would be served best by taking the decision out of her hands, even though the individual is the one who is both most motivated to make decisions that serve her interests and most likely to know her own idiosyncratic tastes (p. 215). Generally speaking, it is easier to treat sexual consent as being illegitimately compromised by incapacity--effectively interfering with a woman's positive autonomy interests in having the sex she desires in order to protect her negative autonomy interests in not having sex without "good reasons"--when there are other instantiations of the same "self" and the strong possibility of intertemporal clashes of the will. (She is young now, but will likely someday be able to consent to sex when she can do so reflectively and more prudently; she is intoxicated now, but can consent when sober; she may regret the sex she has now when older or when sober, and the regret will itself be dysphoric.) It is harder to justify delegitimizing consent when the person is unlikely ever to be able to make second-order-reflective or hedonically prudent judgments or to look back at a prior decision from a more considered viewpoint. (36) Wertheimer is quite open to the notion that working out usable principles to deal with young, mentally retarded, and intoxicated girls and women who token consent is very difficult; he basically suggests, however tentatively, the following resolutions: (37)

9. Youth: The question of whether post-pubescent adolescent girls should be able to give legitimizing consent to sex (especially when there is an age gap between the girl and her partner) turns largely on the empirical question of whether a substantial number of the girls end up suffering significant hedonic loss (psychological or physical) if they have sexual relations (p. 220). (38)

10. Cognitive dysfunction: Because cognitively impaired women, unlike the young, will never be able to grant legitimizing consent to sex if we deem them incompetent, concern for their positive sexual autonomy should make us wary of denying them the right to give such consent, even when those who propose to have sex with the women may, in some sense, "take advantage" of them and even in situations in which the cognitively impaired women may be misled about the intentions of those who suggest sex (pp. 224-26).

Peculiarly, Wertheimer does not refer back to his discussion of deception when he describes the scenario in which a young football player falsely promises that he will be the boyfriend of a mentally retarded girl if she will allow his friends to watch him have sex with her. Though he finds the case problematic largely because she might later be distressed to learn that she was deceived, he shares the common (but difficult to defend) intuition that the case is somewhat troubling even if she never experiences any negative feelings. (Wertheimer does not seem to consider why she might experience no negative effect: Is she too cognitively impaired to realize that the football player lied? Or, would she simply not be bothered to learn she has been fooled unless we helped her see she should be bothered? Would she not be bothered if she learned he has lied because she doesn't understand what the football player thinks of her or because she doesn't care what he thinks of her? (39) But Wertheimer does not pay much heed to the fact that he was predisposed toward finding deception unacceptable more generally, even if the deceived person never learned of the trickery or experienced regret (pp. 202-03).

11. Intoxication: Wertheimer persuasively dismisses a number of arguments about whether women who are voluntarily intoxicated can give legitimizing consent to sex. (40) He rejects two common claims: (1) it is proper to think of women as responsible for the decisions they make while drunk largely because they are responsible for getting drunk (pp. 240-42), (41) and therefore, their consent must validate the man's sexual contact (pp. 244-46); (42) and (2) it would be inconsistent to hold men criminally responsible for having sex when they are intoxicated while rejecting the transformative power of women's consent (pp. 243-44). (43)

Wertheimer is also reluctant to focus too narrowly on the woman's incapacity to reason well at the moment she tokens consent to sex. The most obvious exception to the principle that intoxication negates valid consent is the case in which the woman wants to have sex at some particular point, believes she will only do so if she gets drunk enough to shed her fears or inhibitions, and then gets drunk and has the planned-upon sex (pp. 234, 236, 238). Wertheimer's reluctance might also be justified in cases in which women are aware (in a classic "assumption of the risk" sense) that they are more likely to token consent once drunk but still drink voluntarily in situations in which such circumstances are likely to happen (p. 257).

Instead, he once again focuses largely on harm: the key issues are the probability, gravity, and irreversibility of the harm that women will suffer if they act on one of their state-inconsistent preferences. (Harm will occur if the imprudent drunken self chooses sex, and, over the course of the woman's life, the experience proves more dysphoric than hedonically beneficial.) But Wertheimer has a particular technique--familiar to those who have read the law-and-economics literature on the relationship between wealth maximization and consent--for measuring whether allowing intoxicated consent to be transformative is likely to be in the welfarist interests of a group. If members of the group (while sober) would typically prefer their consent to be valid (while drunk), it should be valid (pp. 250-53). He expects, in this regard, (but is not committed to the conclusion) that sober people would prefer not to be allowed to consent to getting tattooed while drunk: the possible harm is relatively irreversible, albeit not enormously weighty (p. 249), and (I might add, following on his arguments about whether there will be other opportunities to give "better" consent (44)) the chances that they will gain a good deal by getting the tattoo at a particular moment (when drunk) are low. At the same time, he expects (but notes that he does not have a high stake in the conclusion) that most women would want their intoxicated consent to sex to count--both because he suspects that the psychological losses which occur when one has had sex that one wouldn't have had but for inebriation are fairly low and because the refusal to allow such consent to be operative would deprive women of pleasurable sexual experiences that they couldn't simply replace with other experiences (pp. 250-51, 257) (both because alcohol may make sex more pleasurable and because each sexual experience is separable). (45)

B. Problems Internal to the Argument

1. A brief note on the reliance on evolutionary psychology

Wertheimer does not so much respond to critics of the pop evolutionary psychology account of male and female sexuality that he advances--and the associated proposition that knowledge of the evolutionary psychology of sexual desire helps us explain both why men may seek nonconsensual sex and why women may find it so aversive--as he disparages their motives (pp. 62, 66). To the extent that critics of "pop evolutionary psychology" of rape and sex indeed rely, as Wertheimer asserts, on the ideas that evolutionary psychology is amoral, I would agree that they are indeed mistaken. If some such critics believe that evolutionary psychologists maintain that it is impossible to blame people for manifesting sexually aggressive behavior toward which they are "naturally predisposed," Wertheimer is correct that they are wrong. Just as evolutionary psychologists would argue that we can control our weight even though we may be "hard-wired" to enjoy high-calorie sweets and fats, so too would they argue that we can control dispositions that might make sexual aggression more likely than it would be if we had distinct predispositions (pp. 63, 86-87). (46) In my view, though, most critics of the accounts of sexuality and rape upon which Wertheimer relies believe, above all, that the accounts are "bad science," and Wertheimer should have acknowledged that it is commonplace among quite mainstream evolutionary biologists and philosophers of biology, with no particularly strong ideological axes to grind, to think that the literature Wertheimer extols is undertheorized, conceptually vague, and above all, nearly bereft of rigorous empirical support. (47)

The literature on evolutionary psychology and its misuses is so rich and complex that I am by no means confident that Wertheimer should have felt obliged to alert his readers to any particular aspect of the controversies. But I think it might well have been helpful if Wertheimer had explicitly recognized that he had implicitly bought into three quite distinct propositions and that such propositions are legitimately controversial, with each being more controversial than the last:

First, just as physical phenotypic traits may emerge as a result of natural selection, (48) so too might cognitive capacities and affective reactions. Thus, at a very general level, the program of evolutionary psychology--which attempts to identify cognitive/affective capacities that are adaptations, rather than merely to identify adaptive behavior, as human behavioral ecologists have historically attempted--is a sensible program because observed human capacities are significantly a product of selection pressures, and capacities may persist even when they have become dysfunctional. (49)

Second, while evolutionary psychologists recognize that the adaptive brain must be capable of learning, solving new problems, and reacting to unforeseen circumstances, their view is that cognition and emotion are relatively "modularized." (50) Cognitive and emotional processes are best described as highly modularized (or fine-grained) when they are increasingly domain-specific (e.g., there is a disposition better described as a "disposition to rape" rather than a "disposition to be willing to use force to get things one wants"); (51) mandatory (e.g., men are attracted to a certain "look" thought to signal fertility and health, even when they know the object of lust is infertile); opaque (e.g., men do not know why they are attracted to the particular look); and informationally encapsulated--that is, we cannot readily combine knowledge from outside the module with knowledge inside the module (e.g., a man cannot combine knowledge that a particular "nubile" woman is infertile--he sees her birth control pills--to reformulate his sense that he is drawn to her because she has fertility markers"). (52)

Few evolutionary psychologists believe that the brain is fully modularized, (53) nor do any of the critics of evolutionary psychology deny the possibility that the brain uses heuristic shortcuts. The debate between critics and proponents is, at core, both a theoretical and an empirical one. At the theoretical level, critics of evolutionary psychology are much more prone to point out why highly modularized brain development is implausible (54) and to be wary, empirically, that many of the problems that evolutionary psychologists believe to be best solved by a modularized response are in fact accurately characterized in that way.

Third and most problematically, Wertheimer uncritically accepts the claims by evolutionary psychologists that they have, in relationship to sexual drives and aversions, properly identified the actual modules that govern sexual behavior--assuming there are relatively specific modules. (55) They claim to have done so through a combination of two methods: first, empirical observation of (purportedly) near-universal human traits, and second, imaginative construction of the traits that must exist, given natural selection pressures, regardless of whether we can directly observe them or not.

To say that these claims are controversial within mainstream biology would be an understatement: it is startling that Wertheimer seems to treat resistance to the detailed claims of people like Buss, Symons, and Thornhill about sexuality as if it were the leftist/feminist version of evangelical creationist resistance to evolutionary theory. Take what is (arguably) the only evolutionary psychological claim that does any real work for Wertheimer, the claim that women find rape extraordinarily aversive because it is adaptive to do so because rape deprives them of the opportunity to select mates likely to have the resources and interest to increase the odds that the women's offspring will survive (pp. 113-15). The (purportedly) "observed" trait, for which compromised reproductive selectivity is the adaptationist explanation, is that women of reproductive age are more traumatized by rape (and especially potentially impregnating vaginal rape rather than other forms of sexual coercion) than women who are not of reproductive age (pp. 115-16). The claim is enormously problematic. First, it is simply not true that there is anything resembling dependable data that reproductive-age women are more traumatized by rape than other women (56) or even that men who are raped by other men do better than women of child-bearing age even though there are no direct (57) reproductive consequences to men being raped. (58) Second, it is not at all clear why evolutionary psychologists should even expect this trait to be observed if their modularized adaptationist explanation were valid: the posited "revulsion at rape" module in the brain might be triggered by hormones that accompany ovulation rather than by sex-linked factors that are present over the whole course of the life cycle. (59) The typical inattention to translation from (supposedly) functional behavior/attitudes/outcomes to a concrete neural mechanism that could result in the behavior, coupled with the usual false faith in optimal adaptationism, leads them to assume (for no particular reason) that women would not have developed "wasted" or "excess" revulsion. (60)

2. Coercion versus the redistribution of power

Wertheimer, not atypically, models the problem of coercion in the following terms: A sexually coerces B when he forces her to elect between sexual contact and a worse-than-appropriate baseline state. Thus, classic, easily identified coercion occurs when B chooses between: (1) a violent beating (a state worse than the appropriate baseline state of remaining unbeaten); or (2) consenting to sex with A, the person who has made (1) her other option.

Wertheimer is surprisingly silent about what is wrong with coerced sex (perhaps because it seems so obvious). He does recognize that the coerced person is acting rationally in choosing option (1), so the problem is not that A has stripped B of the power to act in accord with reasons that make sense to her (p. 165). One suspects that what is driving Wertheimer is a simple "minimize harm" welfarist principle: both states (1) and (2) make B worse off than she was (entitled to be) prior to A's decision to create the election, and there is no justification (e.g., in terms of gains to A or to third parties) that makes it acceptable for A to inflict this harm. But I will argue that if Wertheimer is indeed driven by the desire to implement some sort of rough-and-ready utilitarian calculus, thinking about coercion as he does is the wrong way to go.

As I noted, (61) too, Wertheimer believes that the appropriate baseline state should be defined not in terms of the status quo ante or the statistically expected state but in terms of the state to which the putatively coerced person B is morally entitled. The end-state to which she is entitled includes, but is not limited to, a state in which all of her separate legal rights are respected. (62) The great difficulty Wertheimer faces is figuring out why B is entitled to anything else besides an end-state in which her separate legal rights are respected and to what in particular she might be entitled. (63) But starting from this framework pushes Wertheimer in the direction of figuring out when, in general, A should be obliged either to do something for B (because of prior contract or affirmative general obligation) or be obliged not to do something to B. (64) What I will argue is that this inquiry is misleading because there are many reasons to restrict A's affirmative or negative obligations that do not go to the issue of whether we believe we will meet reasonable ends (e.g., maximizing utility) if we allow consent in a particular circumstance to legitimize sexual contact. (65)

I think it is far preferable to approach these problems by thinking about power and the redistribution of power rather than coercion. Here is the heuristic I will use in exploring these issues: there are three possible end-states, and B can readily order her preferences among the states. Her favored state need not be available to her; however, she is, in some sense, a more "powerful" person if it is available to her (a person gains power when she gains more options). There may be mechanisms--it is of little moment to me whether we call them redistributive or describe them as establishing a framework that generates an initial distribution of power--that we can employ to increase her power, to ensure that she has more power than she would have if these mechanisms were not in place. Using the mechanisms, though, is costly--for others, and perhaps even for her--and we must assess whether, on balance, the use of the mechanisms is sensible.

Start with the simplest paradigm case, a case outside the domain of sex:</p> <pre> Income Distribution: B's preference order is (1) necessities + goods that are helpful to increase a sense of belonging in the community and to give some degree of social status; (2) necessities; (3) helpful-for-belonging goods. </pre> <p>B's power would plainly increase if she had access to (1). And there are mechanisms available to us to (re)distribute power in this situation so that (1) is potentially available; for example, we could tax away some of the resources held by A so he would not have available his preferred option ("nearly everything he wants to consume") but rather a less desirable option ("most things he wants to consume"). For simplicity's sake, I will assume we evaluate the wisdom of redistributing power through this simple tax-and-transfer mechanism using utilitarian, social welfare function maximizing methods. (66) We assess, in the first instance, whether in short-run static terms B gains more welfare from the goods she can now consume than A loses when he is unable to consume what he previously was able to. We also account for dynamic costs: Will the tax on A affect his incentives to produce in ways that cause welfare losses to him and others? Will using a pure transfer mechanism discourage B from developing the capacity to be self-supporting? And will this transfer both decrease her long-term access to goods and cause independent hedonic losses associated with the absence of the self-respect that arguably comes from not relying on transfers? (67)

It is important to realize, too, that it is not inevitable that we can conceive of mechanisms that permit B to gain the power she wants. (We don't know how to let her fly on her own power, no matter how much she wants to.) This is true even in situations in which a particular person A seems to establish the option set B faces, and B bristles at the choices that person A appears to offer her, wishing she had the power to expand the option set. An additional example:</p> <pre>

Fast Swimmer: (68) B's preference order is: (1) swim much faster + no long practices; (2) swim much faster + long practices; (3) swim no faster + no long practices. </pre> <p>Her coach, A, tells her that she cannot swim faster unless she practices more. B may feel constrained here. She may feel she really does not want to practice more, or she may even subjectively feel that her coach has bullied her into practicing more. But assuming that practice is the only thing that makes people swim faster, the limit on her power is not socially constructed in any relevant sense: there is no alternative social arrangement that gives her the power to realize her first choice. (69)

It is important to analyze several more cases, including some in which Wertheimer is wary to find that manifest sexual consent does and does not legitimate the sexual contact, to see why the distribution and coercion models may differ:</p> <pre> Standard Threat-of-Violence Coercion: (70) B's preference order is: (1) no sex with A + no nonsexual injury; (2) sex + no nonsexual injury; (3) no sex + nonsexual injury. </pre> <p>The mechanism we use to ensure that B has access to (1) is to take away A's power to eliminate the option from the choice set by threatening nonsexual injury. We can both penalize the threats directly and penalize sex that occurs after the threats are made. In either case, although penalization will not always make option (1) available to all women on all occasions--deterrence is plainly imperfect--we would still assume that (1) will be available more often if we employ the mechanism. And, again, to do a simplistic social welfare function maximizing analysis, we believe the gains to B of doing so would outweigh the losses to A (especially if we launder preferences in constructing a social welfare function so that A's immoral gains do not "count") if we employed this mechanism. Our normative goal is not to "eliminate coercion" but to figure out whether a particular method--conventionally understood as eliminating coercion--serves the end of redistributing power in a welfare-enhancing way. (71)

But let's look at some harder cases. Consider the isolated motorist. B is stuck in the desert after her car has broken down. A mechanic (A), with relevant tools and parts, comes by. He knows she is out of cell phone range, is unlikely to run into anyone else who can fix her car or even help her find someone to fix her car, and is likely to suffer a great deal if she is stuck in the desert until a rescuer comes by in the ordinary course of events. Her preferences involve:</p> <pre> Isolated Motorist: (72) B's preference order is: (1) no sex (or, no payment of exorbitant, super-market price for repair) + repair (at the ordinary market price for the repair in nonemergency

situations); (2) repair + sex (or exorbitant price); (3) no sex (no exorbitant price payment) + no repair. </pre> <p>There are a number of mechanisms that we might use to ensure that B has the power to realize option (1). (73) We could establish a general legal or moral obligation for A to repair B's car at nonemergency market prices. But there are a host of reasons this policy may be a bad idea. (Again, the point is not to explore the tricky evaluative questions but merely to signal what they might look like.) We may think the market price will not adequately compensate each and every qualified A in these situations, so that the utility losses to A might outweigh the gain to B; we may think legal regimes that establish positive duties which require people to offer aid will discourage people from noticing that others need aid or that even encouraging people to help leads to crippling guilt and anguish over all the remediable suffering one tolerates in one's daily life. If we follow the coercion model--as Wertheimer often does, albeit incompletely--we will tend to conclude that because it is (likely) not sensible to say that A is violating B's moral expectations by depriving her of option (1), the consent is unproblematic.

This would be a mistake, and we would see that once we recognized that we could use another mechanism that is likely to increase the availability of option (1). (74) We could say that B's consent to sex does not legitimize A's sexual contact with her (setting aside whether that illegitimacy gives rise to legal penalties or merely moral opprobrium). What will happen to the Bs of the world then? It depends, in significant part, on what we expect the As of the world to do if deprived of the option of obtaining legitimate sex for repairs. (75) It is certainly plausible that most mechanics would still be willing to offer repairs at market prices, and most women would thus have the option of (1). But, plainly, at least some mechanics would withdraw from the market, leaving only the choice of (3). (76) Whether women would be helped or hurt by this mechanism (ignoring the losses to the As of the world) (77) depends on an empirical fact about how the redistributive mechanism will work out in practice (just as the wisdom of tax-and-transfer programs is sensitive to empirical facts about the behavior of those who are taxed and those to whom resources are redistributed). But what is key is that a utilitarian may well think that the redistributive mechanism--delegitimizing sexual consent in this case--is worthwhile, with little regard to the wholly separate question of whether establishing a more general obligation to repair at market prices is appropriate.

Finally, let's look at problems of intoxicated consent not through the lens of competence but through that of power. While I will soon relax the assumption that all women have the same preferences (in order to perform a welfarist analysis with a bit more nuance), let's first assume that all women share B's preferences:</p> <pre> Fraternity Parties: B's end-state preference order is: (1) drinking at frat parties + no sex, even if she tokens consent to sex once drunk; (2) drinking at frat parties + sex if she tokens consent once drunk; (3) not attending frat parties or not drinking at them + no sex. (78) </pre> <p>The fact that Bs go to parties and drink, knowing that they are "risking" having sex, does not tell us that the sex is, on balance, desired. Wertheimer believes that the fact that women seem to go to parties with frat boys knowing that they risk tokening drunken consent means that women prefer a world in which their drunken consent will count (pp. 252-53), but that assumption seems plainly wrong. We really do not observe the women's behavior in a world in which they could go to the parties anyway without having their consent immunize the men from legal (or quasi-legal disciplinary) liability or moral opprobrium. The behavior could merely tell us that the women value going to parties more than they devalue the expected losses from the sex. They could be better off still if they could decouple sexual risk from a social life.

What are the mechanisms for empowering B? How can we increase the probability that (1) will be an available option? The case is significantly like the Isolated Motorist case: it is plain that we don't want to oblige frat boys to hold sex-free parties for drunken girls, but we may think it sensible to delegitimize (legally or morally) drunken consent. Some of the costs, structurally, of using the "delegitimizing sexual consent" mechanism look the same as in the Isolated Motorist case: we need to know how many parties will be cancelled, how badly Bs feel when their experience is restricted to option (3) (as they will be when some parties are cancelled), and how often, instead, boys will just add (1) to the option list anyway.

What is different about this case, though? Just as we worried about the impact of resource redistribution on B herself in the Income Distribution case, we may think that a mechanism that focuses exclusively on the frat boys will harm Bs more than one that either focuses on forcing the women to exercise more self-control in granting consent once drunk or one that facilitates their capacity to make the same choices when drunk as they would make in advance. (This may be a matter of telling Bs that they had better focus, even when drunk, because the men will not be deterred from following their expressed wishes; it may be a matter of finding a designated sober girlfriend to keep them from following frat boys to their rooms.) Women-focused mechanisms may (depending on the empirics) seem superior if we believe that women do not have uniform tastes: a rule that focuses on changing A's behavior will cause welfare losses for the subset of women who gain from having sex while drunk, while one that focuses on changing B's behavior will permit those who gain utility (over the course of their lives) by having inebriated sex to do so.

What is especially bothersome about Wertheimer's discussion of coercion is that in talking about intoxication, an issue he believes goes to competence, Wertheimer seems to recognize that the ideal way of judging whether it is worthwhile to legitimize a consent token is to ask whether women (79) would prefer the token of consent to be treated as valid, all things considered, but he does not generally follow the same procedure in talking about coerced consent. (80) If he is willing to admit that women's intoxicated consent to sex should not count, at least so long as they would prefer to have the option that it does not count, (81) even though they are not morally entitled to a world in which boys host sex-free parties for them, it is unclear why he thinks that moral entitlement does any work in thinking about "coercion" and power. (82)


In this last Part, I do the following: (1) provide some very hasty (and self-consciously idiosyncratic) accounts of three supposedly opposed, basic foundational positions that dominate Anglo-American political/legal philosophy: welfarism, deontological rights theory, and perfectionism; (2) provide a brief overview of how Wertheimer helps us see, in relationship to sexual consent, why standard deontological theories (with their particular accounts of the right and staunch opposition to aggregation) are empty unless they rely on the welfarist and perfectionist ideas they purportedly abjure to gain content; (3) provide a brief review of why I think our understanding of the dictates of what I take to be one of the two most significant aspects of welfarism (i.e., its account of the good rather than its allegiance to aggregation across persons) is ultimately parasitic on the perfectionist views it ostensibly defines itself in opposition to; and, finally, (4) bring these discussions back to bear on a question that Wertheimer treats all too hastily: How should we ascertain whether certain people with cognitive disabilities are competent to consent? Do conventional rape laws that negate the force of consent by cognitively impaired persons survive the antiperfectionist message of the Supreme Court's decision in Lawrence v. Texas? (83) Underlying that question, I will suggest, is a deeper question that illustrates my central preoccupation in this Part: Can we establish a legal regime that both protects cognitively impaired persons against (something we would consider) exploitation while permitting them the opportunity to act on sexual desires without understanding the uneasy interplay between welfarist, perfectionist, and deontological (autonomy-protecting) theories?

For present purposes, I define welfarists as being: (1) consequentialists (judging action in terms of its effects rather than conformity to some limiting prior principles); (84) (2) committed to the belief that the ideal evaluative metric is the occurrence of some sort of positive hedonic experience, as judged by each subject herself, whether that experience is called "welfare," "utility," "happiness," or "life satisfaction"; and (3) committed to the practice of some method of aggregation of positive (and negative) hedonic experiences across persons (i.e., committed to the construction of a social welfare function from individual welfare functions). At the same time, I define the subset of deontological autonomy theorists in which I am interested as theorists who are: (1) nonconsequentialists; (2) unwilling to aggregate across persons (believing, above all, in the separateness of persons and the inviolability of each individual); and (3) committed to the notion that the "prior principles" that limit our action are grounded in the need to promote "fairness" (85) and to protect "individual autonomy" (rather than to increase welfare levels, although autonomous action might indeed generally be welfare enhancing for each individual). Finally, I define perfectionists as committed to the belief that action ought to be taken to ensure that people act "virtuously" or in a fashion that manifests "true human nature" or permits "human flourishing" (86) (though the virtues they should manifest or the pictures of human nature differ radically for distinct perfectionists, both in content and in level of specificity). What I will call strong perfectionists tend to have a rather specific view of what actions people ought to take (whether these actions make them more "satisfied" or not or whether they express their autonomous will or not). Weaker perfectionists, on the other hand, are more prone to believe that a wide range of diverse actions are appropriate given the diversity of tastes and personalities, provided that people manifest certain more general virtues in making specific choices or evaluate the choices that they make from certain vantage points.

A. Is Autonomy Theory Useful in Thinking About Sexual Consent?

Deontologists suggest not only that it is impermissible to sacrifice one person's legitimate interests to benefit another (nonaggregation) (87) but also that we can distinguish legitimate interests worthy of protection (generally conducive to some form of autonomy) from mere assertions of (welfarist) interest or desire. I want merely to comment on the ways in which Wertheimer helps us realize that, in the context of the establishment of rights against sexual violation, autonomy theory derives whatever content it has from welfarist and/or perfectionist theory. (88)

Wertheimer alerts us to several basic structural reasons why autonomy theory does little work when facing the real, practical problems that arise in delineating the boundaries of what should be legally prohibited as rape or sexual battery (or even what should be thought of as more or less morally problematic, though legally tolerated, as inadequately autonomous sex). Two are discussed in more detail here.

First, autonomy theorists must be committed both to protecting negative autonomy--the capacity to be free from sexual contacts that do not advance the subject's autonomous will--and positive autonomy--the capacity to render it permissible for others to engage in sexual relations with her. The commitments to each form of autonomy are, however, fundamentally irreconcilable and can be prudentially "balanced" only outside the confines of autonomy theory. A more stringent and demanding view of the preconditions for truly autonomous choice will protect a woman's negative autonomy only by encroaching on her capacity to realize her own goals; it will preclude sexual contacts that might not be made by ideally rational agents under conditions that permit her to manifest her rationality (pp. 3, 36, 125, 251). Closely related, but not identical, to the conflict between negative and positive views of autonomy is the conflict between demanding and permissive views of what rational agents choose. It is perfectly consistent with autonomy theory to be mildly perfectionist--i.e., to claim that autonomous choices are only those choices made for certain considered reasons by persons with option sets adequately rich enough to allow them to reject certain bad reasons for action (p. 128). It is equally consistent with autonomy theory to be more agnostic and pluralistic about what autonomous agents might reasonably think are good reasons to do things and even suspicious of attempts to interfere with another's efforts to do the best she can given her circumstances (pp. 128-29, 191-92). We might, for instance, describe impulsive choices as nonautonomous (i.e., inconsistent with second-order preferences about the sorts of preferences we would like to manifest) or describe a regime in which we interfere with impulsive choices as inconsistent with the principle that disrespecting a person's choice mechanisms substitutes our view of how she should live her life for her autonomous view. (89)

What I think is most vital to note is that autonomy theory has no internal tools to resolve these disputes. Instead, a decision in a particular case to protect, say, negative autonomy can only be justified by arguing that either (1) the welfarist losses that those who have "imperfect" sex will suffer outweigh the losses that will occur because those who cannot give legitimizing consent when consent conditions are imperfect do not have access to sex that might be "good enough," (90) or (2) it is inconsistent with our notion of human flourishing to tolerate the "imperfect" sex that we choose to regulate, even if it is hedonically satisfying or neutral. (91)

Take the case of mental retardation: there is simply no doubt that those with cognitive impairments are imperfectly able to make choices that express a rational life plan and that, if we are more stringent about the preconditions for competent consent, we will preclude them from being able to consent. At the same time, cognitively impaired persons have (some) reasons to consent to sex, and stringent conceptions of the level of autonomy that one needs to demonstrate to give legitimate consent to sex will interfere drastically with their positive autonomy (pp. 224-25). Below, I question Wertheimer's argument that the resolution is wholly welfarist (92)--are mentally retarded individuals hurt or helped in hedonic terms by being allowed to have sex? Nonetheless, his critical argument (i.e., that autonomy theorists have only a set of contradictory injunctions to throw out) seems quite convincing to me.

Second, autonomy theorists have nothing helpful to say about inconsistency of the will. While welfarists can (at least in theory, depending on empirical facts) reject an intoxicated woman's sexual consent because the decision causes more hedonic regret than it produces positive hedonic experience, Wertheimer seems (rightly) skeptical that autonomy theorists have anything coherent to say about which will governs when a person's preferences vary over time. (I think, however, he is less clear than he ought to be that the fundamental problem posed by the intoxication cases is the intertemporal instability of the actor's preferences.) Plainly, we do not require that all choices be made by a person whose choices are consistent over time--in fact, we couldn't do that because to forbid the choice that a person makes at time T1 from governing because he will renounce his preference at time T2 is to allow T2's choice to govern even though it, too, is a time-inconsistent preference. Wertheimer rightly suspects, once more, that we forbid particular choices which are inconsistent with choices that would be made at distinct times and under distinct conditions when (and only when) the choice we want to forbid has an expected negative consequence (or, I might add, a range of outcomes, some of which are so terribly negative that a risk-averse person would not want the choice to be permitted, even if the expected value of the outcome were positive) (pp. 248-53).

Take the case of intoxicated consent. It raises a very complicated question of how we should conceive of why women, when drunk, may sometimes token consent to sex to which they would not consent when sober. (93) Broadly speaking, one could argue that intoxicants impact emotion, desire, and/or cognition. To the degree they merely impact emotion or desire, intoxicants simply pose problems of inconsistent wills. A woman who consents when intoxicated may use the same reasoning process about sex as she would when sober. She weighs the same factors (e.g., how it feels, how likely it is to cement or establish a relationship, how likely she is to feel bad afterwards, how high the risk of STDs or pregnancy), but she is more likely to choose sex because, for instance, it physically feels better when she is drunk. (94) If that is the case, though, it is hard to believe that autonomy theory has much to say about whether intoxicated consent is valid or invalid; on the contrary, a perfectionist who thinks that relying (too much) on "mere" physical sensation imperfectly realizes our natures as humans (or is simply worried that it fails to realize his view that women's nature is relatively chaste) is likely to have a strong view.

To the degree that intoxicants also pose problems of cognition, one might argue that an intoxicated woman is simply incapable of totaling up costs and benefits as well as a sober woman. True enough. But it is still not clear how an autonomy theorist ought to react to that argument. One possibility, consistent with the idea that one has not simply substituted welfarism for autonomy theory, is to say that we believe a woman is incompetent to make a decision not simply when the expected hedonic outcome is negative (in which case competence and autonomy are doing no work at all analytically), but when there is a sufficient risk, which differs across persons, that granting consent will have a negative outcome in welfarist terms. The autonomy theorist would argue that the intoxicated woman is simply not competent to make such an important decision that is (adequately likely) to be wrong for her. I am dubious that anyone sincerely adopts this strategy. To do so would require ascertaining, for instance, whether the inebriated Ivy League college girl is still "smarter" than most women--all permitted to legitimately grant sexual consent--are when sober. Alternatively, of course, competence is doing no work at all: the fact that the expected value of the decision is negative is completely driving the decision that she could not be making her choice competently. (95)

B. Is Welfarism Coherent?

In an article published in Philosophy and Public Affairs, entitled "Hedonic Psychology and the Ambiguities of 'Welfare,'" I argued that we cannot understand whether a person's welfare has increased simply by reference to her subjective situation without looking (to some extent) at our outside observer's views of what a well-lived life is. (96)

There are three basic steps to the argument that welfare, as we conventionally understand it, is a radically incomplete concept. First, if we define welfare as a particular set of sensations (e.g., the net of "pleasure" over "pain"), we will fail to accurately measure how "well off" people who care about sensations or end-states different from pleasure and pain really are. (97) Second, preference utilitarians attempted to solve this problem by extolling not the presence of any particular set of feelings, but the satisfaction of desires or preferences. What is problematic about this formulation is that the satisfaction of preferences is not intrinsically welfare promoting; preferences are at core merely predictions about future hedonic states. Predictions may be wrong. Preference utilitarians tried to deal with the problem of error by respecting only "informed" and "prudent" choices, believing these alone are likely to predict future hedonic states accurately. But unless we know whether a choice will make us happy, the choice is not plainly adequately informed. And unless we choose what in fact makes us happy, it is not clear that we've been prudent. But unless we know what happiness is, how can we have information about the welfare impact of a choice? (98) Third, hedonic psychologists tried to avoid the problems of both hedonic and preference utilitarianism. Rather than declaring, as hedonic utilitarians had, that people should seek a particular end-state, the hedonic psychologists merely asked, using a technique known as "ecological momentary assessment," whether one was content with whatever was happening in one's life (that is, wanted it to persist) or wanted to avoid it. Rather than relying, like preference utilitarians, on potentially mistaken predictions of future states, hedonic psychologists relied on reports of reactions to existing states. But ultimately, their account is no less (weakly) perfectionist than the account proffered by the conventional hedonic utilitarians:</p> <pre> [T]he new hedonics has failed. It has not avoided the problem of inadequate ecumenicalism. It has simply supplanted one idiosyncratic, psychologically reductionist account of "objective happiness" (Bentham's sensory pleasure) with another (maximizing the amount of time spent in experiences that the subject wishes, moment by moment, would persist, rather than those she wishes would end, or towards which she is indifferent). It is possible that some people experience discrete periods of their lives as nothing more than the sum of momentary pains (avoidance) and pleasures (attraction) within those periods, and experience life as a whole as simply the sum of the hedonic valence of a string of discrete periods. But as a descriptive account of what most people care about, it seems no more plausible than Bentham's. For some, their sense, at particular reflective moments, of how their life has been going generally best captures the "true" quality of their experience. They may, for example, place a higher value on feeling proficient or ethically worthy than on experiencing sustained sensations of physical pleasure or other sensations associated with hedonic comfort. It is not clear that the higher-order meaning people attach to life experience, which inevitably reflects a construal of events

over a long time trajectory, can ever adequately be captured in any moment-to-moment account. If this is right, then elicitation devices like "ecological momentary assessments" are no more value-neutral

than Bentham's utilitometer. Anyone asking people to evaluate their

experience asks them to do so from only one of the many vantage

points from which they might conceivably evaluate it; in so doing,

each of us is implicitly valuing certain attitudes about life more

than others, and to that extent, is acting as a weak perfectionist. (99) </pre> <p>Wertheimer attempts to use welfarist principles to resolve the issue of whether a cognitively deficient woman who is not so cognitively impaired that she is unable to understand that she has the capacity to say "no" without bad consequences or understand the possible physical consequences of sex should be allowed to consent to sex with no more stringent limits than more cognitively ordinary women can. His efforts, however, flounder precisely because of the incoherence of the welfarist concept. (100) It might be helpful to consider Judge Berzon's partial concurrence and partial dissent in Anderson v. Morrow, (101) an opinion that, while largely using the language of autonomy rather than welfare, closely tracks (and perhaps even better elucidates) Wertheimer's suspicion of perfectionism:</p> <pre> At the time of the sexual encounter for which Anderson was prosecuted, ([102]) JH was 26 years old and lived with her mother. She is 4'7" tall, has the emotional maturity of a 6- [to] 8-year-old child, and is moderately mentally retarded. She has "a little" vision in one eye, and "fairly good" vision in the other. She also has a heating deficiency but does not always wear a heating aid. It is difficult to understand her speech. As an adult, JH has had at least two boyfriends with whom she had sexual relationships. After both relationships had ended, with one boyfriend she "continued a sexual relationship ... a couple of times a year." She has also received counseling at the Benton County Mental Health Center, where she learned about using condoms. Prosecution experts testified that, "She understands that a man is on top and the woman is on the bottom and the penis is inserted into the vagina," and that she understands the health and pregnancy risks of not using condoms, but that she does not understand dating norms, such as the circumstances in which kissing or holding hands is appropriate dating behavior. .... ... The jury acquitted [Anderson] of [kidnapping, forcible rape, and forcible sodomy]. Thus, the jury rejected JH's version of the incident, as it concluded there was no forcible rape or sodomy. Instead, the basis for Anderson's conviction was non-forcible sexual activity with a person legally incapable of consent. .... The question, then, is whether the sexual liberty interest outlined in Lawrence regulates the manner in which a state drafts and applies its statutory rape law as applied to adult victims. I believe it does. Before proceeding, it is crucial to note the interrelationship between JH's sexual liberty interest and Anderson's: If JH has, in certain circumstances, a constitutionally protected right to consent to sex and she does in fact consent, then there is no constitutionally legitimate basis, under Lawrence, to preclude Anderson from having sex with her in those circumstances. .... As written and as read to the jury at Anderson's trial, the language of section 163.305(3) is susceptible to at least two constructions that could ... impermissibly limit JH's ability to consent to sex: (1) the construction, ultimately proffered by the prosecution in this case after some wavering, that JH is never able legally to consent to sex; and (2) a construction that invites law enforcement, including police officers, prosecutors, judges, and juries, to impose its own sexual mores upon JH when deciding whether her consent was valid. .... The first reason why section 163.305(3), as written and as read to the jury at Anderson's trial, could implicate constitutionally protected conduct is that it is susceptible to a construction that JH will never be able legally to consent to sex. The jury could have easily interpreted the phrase "incapable of appraising the nature of the conduct" to mean that either JH is always capable of consenting to sex or she never is. This binary view of mentally retarded individuals generally and JH in particular might well be an unconstitutional imposition on their sexual liberty.... [T]here is clear consensus among experts in the field of mental retardation that mentally retarded individuals experience sexual desire and can meaningfully consent to sex in some situations. The prosecution expert witness conceded as much at trial. Moreover, JH has evidently engaged in voluntary sexual intercourse on a number of occasions in a manner that offended

neither her mother nor the state. Given the general clinical belief

that mentally retarded individuals desire and can "ethically" consent to sex, as well as JH's own desire and demonstrated capacity

to understand and appreciate sexual contact, it could well be unconstitutional for Oregon law to hold that JH can never legally

consent to sex. .... ... [T]he statutory provision alternatively invites those applying the law to invoke their own sexual mores and override JH's sexual choice when deciding whether JH is capable of consent in a particular instance. During the pretrial hearing, the prosecutor first interpreted the statutory scheme in precisely this contextual manner. The prosecution's initial attempt at interpreting the statute was to propose a rule that JH's mother could provide consent on JH's behalf. Realizing the problem with that interpretation, the prosecution suggested that JH's consent would be legally valid in the context of a "boyfriend-girlfriend" relationship. ....

Expert testimony at trial also suggested an invitation to apply

one's own moral framework to JH's sexual choice. In explaining why

JH's consent was not valid, the prosecution's non-medical expert on sexually abused, mentally retarded individuals testified that

whereas JH sees "sex" as merely a physical act, "If you ask, you

know, anyone else what sex was or what intercourse is you see an

entire picture. You see the candles, the wine, the dating, you know,

whatever else goes on. With her sex is just one quick spur of the

moment thing." That the state may not burden a particular sexual choice out of distaste or disagreement is the central holding of Lawrence. .... ... [W]hile the state surely has a very strong, legitimate interest in ensuring that the consent of a mentally disabled individual is knowledgeable and truly voluntary, and in disregarding that consent in situations where the alleged victim does not understand either the circumstances and consequences of sexual conduct or the extent of her ability to refuse sex, the state has no legitimate interest in imposing sexual mores on retarded individuals or their consensual partners. When considering how to construct a regime that both respects the sexual choice of the mentally retarded and protects them from predation, others have recognized this distinction between an appropriate voluntariness inquiry and

inappropriate moralizing. (103) </pre> <p>The problem with both Judge Berzon's argument in Anderson and Wertheimer's conceptually parallel arguments that we can think intelligently about whether cognitively impaired women are experientially helped or harmed by sex without having views external to their experience about what "harmed" and "helped" mean is that they seem deeply blind to the degree to which they depend on unanswerable questions about whether an agent has adequate information (and adequate "prudence" to reflect on the "information" she ostensibly "possesses"). (104) The problem of information is particularly acute in this case. Can a cognitively impaired person gain (and process) "enough" information without becoming such a different person that it is not clear that "simply" correcting information flow leaves us with the same "subject"?

At some level, what we must know about JH is that she is capable, to some extent, of knowing what it has been like to have sex in past circumstances. At the same time, it is also plain that we have very little way of telling whether JH lacks knowledge (or true understanding) of the emotional consequences of the decision to have sex in these circumstances. Does it matter? Does it matter only if these unknown emotional consequence are bad? If so, from which--and whose--point of view should the emotional consequences be judged?

What is more bothersome still is ascertaining whether it would matter if she is likely to regret the sex more if we (are able to and do) explain more to her. It is not implausible that she is especially incapable of "reading" whether her sexual partner thinks of her as an appealing person or merely as a dupe who is sexually "available." If she doesn't know her sexual partner thinks she is a dupe, is she adequately informed to make a welfare-enhancing choice (or an "autonomous" one)? This problem is suggested in part by reflecting on the following issue: What is information anyway? In theory, more information is always better (except to the degree that processing information is costly). But is welfare-relevant (or autonomy-relevant) information only that information which, if known, could lead to higher welfare levels? And if this is the relevant sort of information only if it could make her happier, we must face (at least) two questions. First, what if learning this information wouldn't possibly make her happier? (Imagine believing that there is no subsequent state of the world--choosing to have sex with Anderson or choosing not to, knowing his contempt for her--that would make her "better off," at least given certain conceptions of what it is to be better off, than she'd have been had she remained uninformed. (105)) Second, and more important, what do we mean by "happier" anyway, without regard to a theory of human flourishing? What if we believe that if we really work to make JH as conscious and acute as she can be---even about the contemptible attitudes of those around her, even when she is the object of their repulsive feelings--she will feel a satisfaction with herself for developing her capacities that is deeper than the satisfaction she would get from "mere" sensual pleasure? Or, to put it precisely the other way around, what if we're just ruining her good time? (106)

The problem of identity transformation may be especially acute here, but it is present in more masked forms in any case. Does it matter if we believe that she would be upset if she knew how she were viewed by others, or is she incapable of being a person who would be upset in that way? What if she is merely unlikely to be upset in that way if left to her own devices? How would we figure out whether JH is "better off' in welfarist terms (or more "autonomous") if we let her think that all of her sexual contacts have been socially unproblematic (thus causing her to have lots of sexual contacts which more or less feel good), rather than push her as hard as we can to recognize that some of these contacts are "bad" (causing her to have fewer contacts, but which survive the greatest level of reflectiveness we can help her achieve and give rise to "deeper" feelings of satisfaction)? (107) Could we rely on her meta-references about how to evaluate her welfare position (e.g., "I'd like to be the sort of person whose life is lived so as to maximize X sort of happiness rather than Y sort") if she lacks the cognitive capacity to understand questions about the evaluative perspective she'd most like to use?

In the final analysis, I think, we can read Lawrence as justified in forbidding strong perfectionist views on sexual practices; Judge Berzon is perfectly right to be up in arms at the social worker who implicitly tells the jury that sex without candlelit dinner associations is simply no good. But we can also read it to show that the demands of deontological theory are too vague and that the demands of the welfarist theory (which typically gives deontology content) are too dependent on weak perfectionism to separate the question of whether JH is being harmed from the question of whether she will be most satisfied with her life only when she not only feels, but also becomes, most self-aware.


Consent to Sexual Relations does a good deal more than establish a superb framework for identifying critical issues that those who think about the legal and moral regulation of sexual contacts must face, though it would be a considerable achievement were that all it did. Wertheimer's arguments about the limits of a framework that separates issues of sexual autonomy from issues of experiential harm are especially telling, but the analytical acumen he brings to bear throughout the book, particularly on problems of intoxication and deception, is quite praiseworthy as well.

The long sections purporting to demonstrate that evolutionary psychologists have given us a useful picture of both rapists' motivations and victims' responses seem highly problematic. Not only does Wertheimer fail to clarify which of his normative arguments are and are not sensitive to adopting the evolutionary psychology perspective or particular conclusions, but he is also alarmingly uncritical about some (justifiably) quite controversial pop science.

While I deeply admire Wertheimer's efforts to analyze issues of consent through a fundamentally welfarist framework, I think there are two significant problems. First, I think he backs off his commitment in the most significant class of cases he deals with, cases that he describes as raising issues of coercion. Wertheimer argues that consent legitimizes sexual contact even if the person elects to have sex rather than face some (unwanted) state X, unless the putative victim is independently legally or morally entitled to state not-X. I think this is the wrong question to ask because there may be reasons not to impose a duty on those who deal with the putative victims to create state not-X that do not affect whether we should try to empower the putative victim through distinct means. The right question is whether we can empower the putative victims and expand their option set without undue costs to others (or, more indirectly, to themselves). Second, I think that, like most liberal political philosophers, he believes it possible to distinguish sharply between "subject"-centered welfarist theories and "objective" illiberal theories that rely on a stronger perfectionist theory of the good. I think that claim is wrong because we cannot elicit information from subjects about their welfare rather than information about particular forms of welfare that we, as objective observers, care about. Given that ambiguity, I suspect the best account of competence to consent will be quite different from the account that Wertheimer offers. It is not coherent to ask, for instance, whether cognitively impaired women are happier if their sexual consent counts unless we know whether we value, say, sensory pleasure more than maximizing our capacity to reflect on the world and our decisions.




(4.) The book focuses on problematic heterosexual contacts between men and women who are, at the very least, considerably less desirous of sex than the men seeking sexual contact, explicitly dismissing the possibility that heterosexual women ever try to force or pressure reluctant men into having sex (pp. 38-39, 45, 79) or that heterosexual men can generally really fully empathize with women's profoundly horrific experience of nonconsensual heterosexual sex (p. 83). I think Wertheimer is largely, though not entirely, empirically right in this regard. (Survey evidence indicates that heterosexual men are pressured by women into having unwanted sexual relations, both psychologically and occasionally physically, a good deal more often than Wertheimer implies but are far less likely to react nearly so aversively to the experience as do similarly victimized women. See, e.g., Cindy Struckman-Johnson, Forced Sex on Dates: It Happens to Men, Too, 24 J. SEX RES. 234, 234-39 (1988) (finding empirical support for the contention that men who are coerced into sex by women most often fail to report the assault, but that sexual coercion is much less emotionally traumatic for male victims than for female victims).) I am skeptical that his evolutionary psychological explanations for these "facts" are convincing. See infra Part I.B.1.

I don't think Wertheimer ever gives an adequate account of his failure to discuss problems of nonconsensual same-sex contact, though he briefly notes (however unpersuasively) that homosexual rape by males may merely be a nonadaptive byproduct of the often-frustrated, reproductively driven male desire for casual sex with multiple female partners (p. 86). For discussions of the prevalence of forced sex with male victims, see, for example, Elizabeth J. Kramer, When Men Are Victims: Applying Rape Shield Laws to Male Same-Sex Rape, 73 N.Y.U.L. REV. 293, 295 (1998) (explaining that if one includes same-sex rape in prisons, men are roughly twice as likely to be raped as women; even excluding prison rapes, and accounting for the fact that men may well be even less likely than women to report having been raped, Department of Justice figures show more than 25,000 rapes of males a year, compared to about 150,000 rapes of women). For a general discussion of nonconsensual sex within lesbian relationships, see LORI. B GIRSHICK, WOMAN-TO-WOMAN SEXUAL VIOLENCE: DOES SHE CALL IT RAPE? (2002); frequency rates are discussed in Linda A. Bernhard, Physical and Sexual Violence Experienced by Lesbian and Heterosexual Women, 6 VIOLENCE AGAINST WOMEN 68, 75 (2000).

(5.) Some of his arguments on this issue are, in my mind, more persuasive than others: the key point for now is that they are straightforward and unapologetic.

(6.) Wertheimer draws heavily on the two popular scientific texts on evolutionary psychology and sex. See DAVID M. BUSS, THE EVOLUTION OF DESIRE (rev. ed. 2003); DONALD SYMONS, THE EVOLUTION OF HUMAN SEXUALITY (1979). Wertheimer draws to a lesser extent on the most mainstream evolutionary psychological text on rape in particular, and the most renowned law review article on the same subject. See RANDY THORNHILL & CRAIG T. PALMER, A NATURAL HISTORY OF RAPE (2000); Owen D. Jones, Sex, Culture and the Biology of Rape: Toward Explanation and Prevention, 87 CAL. L. REV. 827 (1999).

(7.) Think of the canonical trope on this one: the woman who consents to marry someone and have sex with him only because she cannot get market work that pays enough to support herself (or herself and the children she might want).

(8.) As written, this section would permit consent that was responsive to unambiguous threats of violence to legitimize sexual contacts because refusing to do so would preclude women from moving from an unjust situation (in which they would face nonsexual violence unless they consented to sex) to another unjust one they preferred (unwanted sex without violence). The whole point of Wertheimer's chapter on coercion--and everyone's writing on coercion--is that coerced choices do, when made, improve the coerced party's position over what it would be if the "unjust" situation that they initially faced persisted without "trading" sex for an improvement in that situation. As I note, see infra Parts I.A, I.B.2, the fact that coerced women (as a group) may be less likely to face injustice if their consent is deemed nonlegitimizing may mean that, for the group, allowing consent to legitimize the contact may not improve the position of the group, but for any individual woman at a particular time, it might well be the case that the refusal to legitimize her consent precludes her from improving her position. (Conventional rape law--legally delegitimizing consent that is responsive to threats of physical force--may make women as a group better off by diminishing the number of threats of nonsexual violence women face, but a particular woman might prefer to "offer" her sexuality to a man who threatens nonsexual violence than to be physically forced to submit to intercourse or beaten up while "successfully" resisting unwanted sex, as she might be forced to do in the "unjust" world.)

(9.) See infra note 37.

(10.) See infra notes 78-80.

(11.) Most of the traditional arguments that gay sex is unnatural or perverted depend on certain perfectionist notions, for example that body parts have "natural" uses which represent their best and highest use. See, e.g., Michael Levin, Why Homosexuality Is Abnormal, 67 MONIST 251, 251 (1984) (arguing that the "natural" use of the erect penis is to penetrate a vagina; it is most suited to that end and "evolved" for that purpose so that homosexuality entails the failure to make the best, most evolved human use of the body). But even less overtly bigoted perfectionist arguments about sex can make one's skin crawl, both because they place such stringent and fanciful demands on legitimate sexual desire and because the stringent demands always seem to involve a remarkable high-mindedness for an activity that rarely seems so terribly high-minded. See, e.g., Thomas Nagel, Sexual Perversion, in PHILOSOPHY AND SEX 326, 332 (Robert B. Baker et al. eds., 3d ed. 1998) (arguing that nonperverted sex involves, and only involves, a desire that one's partner be aroused by the recognition of one's desire that he or she be aroused).

(12.) I make the argument that welfarism must be weakly perfectionist in some detail in Mark Kelman, Hedonic Psychology and the Ambiguities of "Welfare," 33 PHIL. & PUB. AFF. 391 (2005). Part II of this Book Review is, at core, merely an attempt to bring the framework I tried to develop in that piece to bear on the particular issue of consent to sexual relations, especially in the context of whether cognitively impaired women are competent to consent.

(13.) I omit in this Book Review discussion of Wertheimer's brief account of legal issues in rape law (pp. 11-36). I should note, though, that his discussion of why it is of little moment whether we believe a defendant has not committed the actus reus of rape or has made an exculpatory mistake about the presence of objective offense elements (p. 24) is plainly wrong in some classes of cases (such as ones involving those charged with aiding and abetting another's rape, who would not be exculpated if the principal had made a mistake that they themselves did not make) and analytically misleading in all classes of cases. I also omit discussion of one interesting aspect of Wertheimer's discussion of competence and "false preferences" (pp. 226-31) and what strikes me as a not-very-interesting set of speculations on whether less desirous women may be duty-bound under certain conceptions of justice to have sex they don't intrinsically desire with more desirous men with whom they are in long-term relationships (pp. 258-76).

(14.) Wertheimer explains this "fact" about men in evolutionary psychological terms--to oversimplify, men with a genetic predisposition to desire sex with the maximum number of fertile women who reveal through their "beauty" that they are likely to be healthy enough to bear and raise the man's child will out-reproduce those men who do not have such a predisposition, at least so long as they account adequately for ecological constraints and other resource constraints that would preclude them from maximizing the number of children they can expect to raise to child-bearing maturity if they simply maximize pregnancies. But he never really demonstrates the "fact" he wants to explain. He gives a great deal of evidence (whose quality I will set aside for now) that men are "attracted" to the sorts of "nubile" women who would have appeared fertile to early humans and that men are more interested than women in "casual" sex but provides nearly no evidence that there is a widespread disposition to continue to desire sex (as much? at all?) once it is perceived as unwanted (pp. 38-46, 56-59). Ogling a model and fantasizing about having sex with her does not tell us the ogler/fantasizer would continue to desire sex (as strongly? at all?) if he actually met, propositioned, and got rebuffed by the model, or if he would desire it more, less, or the same amount knowing it to be unwanted.

(15.) Though, as I noted, see supra notes 11-12 and accompanying text and infra notes 89-91 and accompanying text, Wertheimer frequently invokes ideals of autonomy, these sorts of arguments are typical of a large number of claims that subtly bespeak the degree to which he is fundamentally drawn to welfarist accounts of moral reasoning.

(16.) Distinctions between these two views go to a vital set of questions about evolutionary psychology that Wertheimer unduly skims over: the degree to which our cognitive and affective processes are highly modularized (to oversimplify radically, do men have a narrow disposition to rape or is rape the occasional vector outcome of broader processes--the disposition to desire sexual contact, a disposition that plays out differently for distinct individuals in distinct settings, and the disposition to make instrumental, context-dependent cost-benefit calculations in seeking to satisfy desires?), and if they are significantly modularized, the degree to which evolutionary psychologists can properly identify the level of fine-grained granularity at which the modules exist. (Again, to oversimplify, if men have a "sexual desire" module, is it a very specific, fine-grained module that directs them towards reproductive success, at least that would direct them to such success under the conditions that purportedly existed during the environment of evolutionary adaptation, or is it a broader sex-seeking module?) I address Wertheimer's casual acceptance of claims by evolutionary psychologists that most significant cognitive and affective processes are modular and that they have identified the modules properly. See infra notes 50-54 and accompanying text. Note, though, that if nonconsensual sex is merely the byproduct of excess, unfulfilled male sexual desire, rather than a "rape" module, Wertheimer has a lot more explaining to do. First, conceptually, it is not clear whether he believes (and if he does, why he does) that rape proclivity should be a function of the level of sexual frustration. Wertheimer believes that nearly all men want sex more often and with more women than they can actually obtain (pp. 40-43), but he asserts that the "fact" that men who are less likely to have as much voluntary sex as they desire are more likely to rape strongly bolsters the claim that rape (or at least rape propensity) is the expected outcome of unfulfilled desire (p. 76). Making that assertion implies all sorts of (unproven, even unstated) things about the "pent-up" nature of unfulfilled desire--it is not clear why, within this picture, it is more frustrating to be unable to have sex with many women than it is to be unable to have sex with the particular unwilling woman one desires, and it is certainly not clear that the functional relationship between "frustration" and "number of unavailable women" has any mathematical properties that we could intelligently assess.

Second, the (purported) "fact" that frustrated men rape more is "proven" incredibly indirectly: men whom evolutionary psychologists predict would have more trouble attracting women (because they are lacking in resources) rape more (pp. 76, 81). But while evidence in this area is inevitably less than fully persuasive--because it is grounded in self-reports on issues in which we'd expect self-presentation concerns to be strong--most of the direct evidence that we do have shows that rapists have no less voluntary sex than control-group felons who have not committed sexual crimes (and there is no evidence that rapists have had less sex than men of the same age in the general population). See, e.g., DIANA SCULLY, UNDERSTANDING SEXUAL VIOLENCE 71-72 (1990) (reporting that eighty-nine percent of rapists, compared to ninety-one percent of nonrapists, had engaged in consensual sex at least twice a week before entering prison; forty-two percent of rapists compared to thirty-seven percent of nonrapists had consensual sex daily; forty-nine percent of rapists had lived with three or more women and sixty-two percent of rapists had already fathered children); K.P. Gwee et al., The Sexual Profile of Rapists in Singapore, 42 MED., SCI. & L. 51, 53 (2002) (reporting that sixty percent of rapists, compared to forty-four percent of nonrapists, had had more than ten sexual partners in their lifetime; thirty-one percent of nonrapists, compared to only twenty-one percent of rapists, had no steady sexual relationship at the time of the offense that resulted in imprisonment). There is some evidence that rapists have less sex relative to their desire for sex than nonrapists, even though they may have more sexual contacts. See, e.g., MARTIN L. LALUMIERE ET AL., THE CAUSES OF RAPE 72-76 (2005). But telling a convincing evolutionary psychological story that explains distinctions in desire, let alone distinctions in desire for sex with women who are less likely to want to raise one's children if they get pregnant, is no mean feat. There is also plenty of evidence that poorer men as a group engage in all sorts of antisocial behavior whose connection to sexual selection is at best attenuated. Most rapists commit an enormous number of nonsexual crimes and show across-the-board antisocial dispositions; rape rates typically correspond to more general rates of violent crime, both across time and across communities. Id. at 62, 72-76.

(17.) See infra Part I.B.1.

(18.) Wertheimer endorses the claim made by Thornhill that women of reproductive age are substantially more aversive to rape than those who are older or younger (p. 115). I use the claim to illustrate much of what I think is worst in the practice of evolutionary psychologists. See infra notes 59-60 and accompanying text.

(19.) See pp. 113-18 for this mainstream evolutionary psychological explanation.

(20.) One way to highlight the rather puny role that the evolutionary psychology material--whether right or wrong--is playing here is to reflect on male reactions to being raped. It is hard to imagine that anyone would seriously argue that our strongly negative moral and legal reactions to raping men should waver because we cannot (even purport to) connect this averse reaction to the loss of reproductive control. (In that sense, the aversion is not likely to serve a role that would be strongly adaptive; though here, as always, a weak, just-so adaptationist story could be readily constructed: raped men might be stigmatized in ways that make them unlikely to be able to attract sexual partners later.)

(21.) Wertheimer shares my predisposition that consent is not problematic simply because it is not given in explicit words or given contemporaneously with sex. While it is plainly impermissible to have sex once consent is withdrawn (pp. 159-61) and manifesting willingness to have sex at some future point does not give rise to contractual obligation (pp. 120-21,160), consent can be valid even if sex occurs after a party indicated (without an explicit "yes" or similar words) that she would say no if and when she wanted sexual contact to cease (p. 121). I think Wertheimer is also plainly right that a person can agree to do something later that she is no longer contemporaneously able to consent to; obviously, we consent to surgery though we will be unconscious when it occurs, and it does not seem problematic that someone could soberly reflect on a decision to have sex but do so only after getting too drunk to make a rational choice (p. 156).

He is also plainly correct that demanding the use of formal words is needless if we simply want to ensure that the woman has tokened consent (p. 151). As he says, "I think that nothing problematic will follow from construing any form of behavior or omission as a token of consent so long as its meaning is clear and so long as [the woman] can indicate to the contrary if it is not" (p. 153). We do communicate through words, deeds, and selective silence. In my view, the strongest argument for demanding higher levels of verbal formality/contractualism has far less to do with the problem of communicative ambiguity than it does with power and the ambiguous nature of coercion. Assume, as I think is true, that there are many women with the following end-state preference-ordering regarding declining a proposition to have sex:

(1) No sex + no awkward moments in which the woman needs to be verbally assertive. (Note that volunteering that she does not desire to have sex forces the woman to assume that it is appropriate to "turn down" a sexual proposition when it might still be ambiguous that she is being asked to have sex, and she thus risks embarrassment if the man says he did not desire her, forcing her to be confrontational about expressing her own desires.) (2) Sex + no such verbal confrontation. (3) No sex + verbal confrontation.

A legal/moral requirement of verbal formality might mean that she gets to choose (1)--since a "must ask/must receive permission" system removes the burden of verbal assertion from women--while in a world in which there is no such requirement, she would not only token (2) but also subjectively prefer it to (3). Her silence would indicate that she prefers sex to the available alternatives. I amplify on why Wertheimer more generally underestimates the problem of coercion--or what I think is more aptly characterized as the problem of undue powerlessness--infra Part I.B.2. For one fuller account of what I have reduced here to stylized choice-sets (e.g., no sex, no awkward moment), see MARKUS D. DUBBER & MARK G. KELMAN, AMERICAN CRIMINAL LAW: CASES, STATUTES, AND COMMENTS 794-95 (2005).

(22.) I am persuaded by Wertheimer's argument that a person who has sex with someone who has not communicated her consent, even if she has a positive attitude in her mind towards the sex so that she personally is not harmed, has committed a wrongful act in part because the woman's uncommunicated consent could not possibly give the man a reason to act. For example, the fact that, unbeknownst to me, you may want me to steal your car because you intend to file an insurance claim if I do can't possibly give me a moral reason to take the car from you (much less from the insurance company) (pp. 146, 148). One can also, perhaps more profitably, think of this issue in "attempt" law terms: a person who does something that is harmful in most circumstances (and having sex with people who don't communicate their consent will usually result in harm) has revealed himself as immoral and dangerous, even though in the particular case no harm ensued (because the woman secretly wanted the sexual contact) (see pp. 96, 100, 102, 111).

(23.) In standard parlance, this is a preference-utilitarian argument for the legitimizing force of consent. As Wertheimer puts it, "if people typically consent only to those interactions that will improve their expected welfare, and if people typically make fairly good judgments about such matters, then consensual interactions will leave both parties better off than they otherwise would be" (p. 125). This leaves open the three broad classes of questions about the utilitarian account of sexual consent. First, what does one do if one thinks that people don't know precisely what it is they are choosing (the problem of deception--and to a lesser degree incapacity--in rape law or of information in preference-utilitarianism)? Second, what does one do if there are classes of persons (the young, the cognitively deficient, the intoxicated) who may not make good judgments (incapacity in rape law, imprudence in preference utilitarianism)? Third, what does one do about the fact that people may maximize their welfare given available options but do not have all the options they might ideally have (cognate to the problem of coercion)?

(24.) If deceived about the quality and nature of one's own actions, the actions may not conform with one's values. If forced to act, it is the forceful person's will, not the subject's, that determines outcomes.

(25.) Kant himself, if not most latter-day Kantians, believed that sexual desire is so powerful that we care nothing for our partner when we are in the grip of passion, so that all sex is at risk of violating the maxim requiring us to treat others as ends, not means (see pp. 132-33). Rather quaintly, Kant felt that the risk was adequately obliterated by marriage (and only marriage). Wertheimer posits that Kant held this belief since the long-term commitment implied a sufficiently secure regard for the general interests of the other; Hampton believes that Kant thought that marriage cures the ordinary defects in sex because each married partner "owns" the other, and this unity of wills dissolves the possibility that someone who could really be described as "other" is being used as means. Jean Hampton, Defining Wrong and Defining Rape, in A MOST DETESTABLE CRIME 118, 139 (Keith Burgess-Jackson ed., 1999). To be honest, I cannot even vaguely fathom what Kant thought he was getting at in his almost brutally dull passages on marriage.

(26.) Kittay and Chamalass have both endorsed the idea that women are not treated as autonomous agents unless they consent to sex because of desire. See Martha Chamalass, Consent, Equality, and the Legal Control of Sexual Conduct, 61 S. CAL. L. REV. 777, 842 (1988) ("Although I believe that sexual encounters which have pleasure or intimacy as their purpose come closest to the egalitarian ideal of good sex, even these encounters are not risk free for women."); Eva Feder Kittay, Ah! My Foolish Heart: A Response to Alan Soble's "Antioch's 'Sexual Offense Policy': A Philosophical Exploration," 28 J. Soc. PHIL. 153, 158 (1997) ("A sexual encounter with another, which has the recognition of the other's desire (and so the other's sexual agency) as a sine qua non, is the sole way in which we can engage in sex without reducing the other to an object."). Anderson and Hampton both believe that men do not treat women as autonomous, even when the women "consent" to sex, if the sex is not being used to affirm an intimate relationship, see ELIZABETH ANDERSON, VALUES IN ETHICS AND ECONOMICS 154 (1993), or if the sexual relationships men have with women mean to demonstrate some way in which they devalue them, see Hampton, supra note 25, at 134-40. Wertheimer rejects the idea that an autonomous person cannot choose to have sex for reasons distinct from these supposedly "preferred" reasons. He rejects these views using the same sorts of arguments that all liberal anti-perfectionists use: because people are diverse, it is silly to think that there are only a small number of good reasons to have sex (e.g., it is reasonable to choose sex for companionship, to comfort others, to get pregnant, etc.; it is also reasonable to choose to have uncommunicative rather than communicative sex) (see pp. 136-38, 141).

(27.) Again, thinking about easy coercion cases clarifies this: avoiding a gunshot is a good reason to have sex, but we shouldn't treat sex-responsive-to-threats-of-gunshots as legitimizing the sex. We use the law to expand women's option sets, to redistribute power from one group (sex-seeking threatening men) to another (women). I analyze coercion from the perspective of standard theories of redistribution at infra Part I.B.2.

(28.) Wertheimer believes that the woman does not get the opportunity to receive what she is entitled to when she consents to sex because the man proposes to violate her rights if she does not consent or because he proposes to do what he is obliged to do in any case only if she consents (p. 167). But Wertheimer acknowledges that the content of the moralized baseline is left open by this formulation: What is the man obliged to provide her? What proposals threaten rights violations? (See p. 169.)

(29.) A person may be coerced, even though she is asked to choose between an option and a state no worse than the status quo, if the person proposing the option is obliged to make her better off still. For example, a professional lifeguard, contractually bound to save people, can't morally rely on a drowning victim's consent to pay huge sums of money (or have sex) before being saved, merely because the lifeguard does not propose to make her worse off than she was before the proposal (pp. 167-68). At the same time, the fact that a criminal defendant may legitimately consent to a "lighter sentence" in exchange for a guilty plea though the sentence she is agreeing to puts her in a worse position than she was in means that one isn't always entitled to insist on the status quo ante (p. 168). (I am not fully persuaded that the status quo can reasonably be thought to exclude statistical expectations, but this seems a mere definitional quibble: at the time the defendant agrees to a short prison term, he might not be in prison, but his status quo state includes, at a minimum, his feelings about anticipated events.) Similarly, the fact that a person might expect an employment relationship to continue in an employment-at-will regime doesn't mean that she does not consent to an employer's proposals to do more work than she'd prefer to do in order to keep her job. Id. In discussing coercion, I note that statistical expectations may end up beating heavily on what the moralized baseline may be. See infra note 76 and accompanying text.

(30.) Thus, he argues that the boyfriend who will spread malicious rumors that his girlfriend is a slut if they don't have sex but will keep quiet if they do is not criminally coercive because spreading malicious gossip is not a legal wrong (p. 182). But the focus on legal rights infects the discussion of the dictates of morality as well: while he thinks it might violate sensible criminal prohibitions designed to encourage informants to punish a person who says he will not inform the IRS of B's tax dodging if she has sex with him, he feels it is not a sexual offense (either morally or legally) because B is not legally entitled to dodge her taxes. Id.

(31.) Thus, Wertheimer believes that a mechanic encountering a stranded motorist in a broken-down car, with no affirmative duty to aid the motorist, should be allowed to negotiate a legally enforceable contract for super-market payments or be immunized from rape charges if he demands sex in exchange for repairs (pp. 170, 176). I analyze this case in some detail, see infra Part I.B.2; it is, in my mind, one of two critical restrictions on the scope of his account of coercion that does not survive scrutiny from his own welfarist perspective.

(32.) Wertheimer believes that consent obtained because women find the explicit or implicit alternative that they will be verbally abused legitimizes the resulting sex (unless the verbal abuse signals the possibility of nonsexual violence) simply because we wouldn't directly legally protect women against verbal abuse (p. 187). Again, I analyze this (and milder cases in which the man does not even threaten "verbal abuse" but merely makes it uncomfortable for the woman to avoid sex by giving her few "easy" ways out of a sexual situation) in more detail later in this Book Review. See infra note 82. This is, in my view, the second key indefensible restriction in Wertheimer's scheme.

(33.) In his discussion of the harms of having been duped into having sex, Wertheimer is atypically sympathetic to rights-based, rather than welfare-based, accounts of wrongdoing. Obviously, as Wertheimer acknowledges (p. 202), someone duped into having sex may experience regret when she learns she was duped, but he is sympathetic to the argument that she is wronged even if she never discovers it, just as our lives are "truly" better if we are actually admired than if we wrongly believe we are admired (pp. 95, 202). Whether the intuition that nonveridical experiences are inferior to veridical ones would survive in the absence of our experience that we all bear some risk of discovering we have been duped is unclear. It is also unclear whether we all bear a probabilistic welfare loss that we have been duped (unduly often) unless duping is banned (p. 203). (Think in this regard of our reaction to unknown privacy intrusions--e.g., anonymous government agents who read our embarrassing e-mails but never let us know that they have done so. Our negative reaction to this practice could be defended without regard to welfare loss, but it could be explained as well as grounded in the recognition that we all do know there is some needlessly high probability that we are being snooped on, even if the precise incidents of snooping remain undetected and there is no single moment of crystallized shame, or it could be based on our experience-based refusal to believe the claim that we will never learn that others have learned our secrets.)

(34.) A man's statement that he is not married when he actually is may induce consent that would not otherwise be given, but so would a statement that he is in love, that he will not be a distracted jerk in the morning, or that he has never met someone so appealing. We have trouble, however, drawing the lines that would tell us which of these lies is permissible to obtain consent. Moreover, efforts to prove that he stated he was unmarried will give rise to even more clashing "he said/she said" testimony that is difficult to resolve. I would add another point: I think it would be hard to resolve whether we think a lie about, say, marital status is always so material (for those who have sex when they are relying merely on the response to a question about marital status, rather than a far longer-term observation of their would-be partner's life habits) that we could call such deception rape, without regard to the question of whether it affected the particular victim's decisionmaking. But an inquiry into whether the particular woman would have made a different decision had she been fully informed threatens to turn into an inquiry about whether she is "loose" or not, and we have rightly spent decades purging rape trials of such inquiries into victim "character."

(35.) Wertheimer persuasively notes that the capacity to consent or make responsible decisions is domain specific: a youngster might have the capacity, say, to decide whether to get an abortion once pregnant or to have sex with someone her own age without being able to make prudent decisions about whether to have sex with somewhat older boys or men. A cognitively dysfunctional woman might be able to make decisions about whether to have sex without being able to make prudent decisions about health care or financial matters (pp. 217-19, 221-22, 226). (It might also be the case that we think end-state preferences about, say, health care vary little across persons, so a surrogate decision-maker will largely make superior means/ends decisions if she substitutes her judgment for the judgment of the person with cognitive impairments, while we suspect reasons for having sex vary widely across persons.)

(36.) See, e.g., pp. 224-25 (describing the case of mental retardation); see also pp. 24647 (explaining that while we wouldn't readily allow a drunken middle-age person to consent to hip replacement surgery since he'll have plenty of other opportunities to reflect more capably on its costs and benefits, we should be loathe to set an ultrahigh competence standard, in terms of reflectiveness and prudence, if we are to permit assisted suicide because anyone considering assisted suicide will, irreversibly, be in a less than fully competent state).

(37.) Because I largely agree with Wertheimer's treatment of deception, I pretty much ignore the issue in this Book Review, except to make the following point: I think he radically understates the overlap between deception and competence issues. A mentally retarded (or young, or intoxicated) woman does not merely lack the ability to act autonomously (i.e., to make her present decisions for reasons she thinks of as relevant) or to make prudent decisions likely to maximize her welfare; she knows less about the situation she is in. She is less likely to know facts that are often misrepresented by those we accuse of immoral or illegal deception: she is less likely to know if she will get pregnant (just as a woman deceived by a man who claims to have had a vasectomy is misinformed); she is less likely to perceive accurately what the man she chooses to have sex with "really" thinks of her (just as someone who is lied to about feelings or intentions is making a misinformed decision). Wertheimer does not really explain why he is generally so wary of deceit-based consent and so reluctant to find women incompetent because the overlap between the cases seems largely to escape his attention. Obviously, there are reasons to differentiate the cases (and some of these reasons are ones he brings up in other contexts). We can "outlaw" (or hope to affirm moral codes that condemn) deceit without interfering much with women's positive sexual autonomy, but we could not do so if we negate the ability of some women to give legitimizing consent. The deceived woman could have sex in the future without being deceived, while, say, a mentally retarded woman might be considered to be permanently retarded. Moreover, a judgment that a woman is incompetent may be demeaning to her, while the prohibition of deceit largely passes judgment on the deceiver and not on the deceived (though it implies a lack of self-protective ability to sort out lies from truth). The deceived woman could have sex in the future without being deceived, while, say, a mentally retarded woman might be considered to be permanently retarded.

At the same time, as I argue in Part II, the problem of what it means for a cognitively dysfunctional woman to be truly informed about the sexual interactions she consents to is still more complicated than I am suggesting now: resolving what we mean when we describe a person as adequately informed requires us, I will argue, to break down the conventional boundaries that Wertheimer relies upon between liberal theories (autonomy-oriented or welfarist) and illiberally perfectionist ones.

(38.) My sense--but it is only a sense--is that Wertheimer thinks most of what is written about whether adolescent girls are harmed by sex they have consented to or even sought out is driven by predisposition and ideology (a battle between feminists who are preoccupied with what they see as the demeaning obsession with female chastity and those preoccupied with the omnipresence of male sexual exploitation) and should instead be driven by psychometric research. See pp. 217-18, 220. Wertheimer also argues persuasively for bright-line rules in this area, rather than efforts to judge on an individualized basis whether a particular youngster is adequately reflective, competent, or unlikely to be harmed by a sexual encounter (pp. 221-22).

In Part II, I return (in discussing intoxication) to an issue that I think Wertheimer inadequately sorts out in making reference to "harm." A pure welfarist might say that young women (or intoxicated women) cannot consent to sex because, as a matter of fact, social welfare will be decreased if their consent is considered valid. But, for reasons I will detail, an autonomy theorist might still make reference to harm even though such a theorist would be reluctant to forbid X from consenting to sex merely because Y and Z would be hurt more than X herself would be helped, or even to bar X from harming herself if her choice were indeed an autonomous choice. The autonomy theorist might nonetheless argue that while people with certain defects (youth, inebriation, mental retardation) are competent to make certain judgments, they are not competent to make judgments about "risky" activities, and sex is a hedonically risky activity. Such a theorist might believe this because he thinks that while the inebriated woman is not truly competent to make any decisions, we simply won't worry about incompetent, not truly autonomous choices unless harm is a possible outcome. Or such a theorist might believe that choices that involve risks of harm are more cognitively complex than ones that don't, so that such choices are simply more difficult to make.

(39.) Issues that I reraise in Part II about what it means to make an informed or prudent choice require us to think carefully about these questions.

(40.) At the same time, he recognizes that many intoxicated women will be incapable of giving clear tokens of consent in many of the drunken "date rape" cases we actually worry about (p. 232).

(41.) For example, if we ingest a substance that we should know will make us uncontrollably violent, we will be responsible for the violence.

(42.) Wertheimer argues, convincingly, that this argument is unpersuasive. First, one can be "responsible" for creating a problem without having to bear all of its consequences: the fact that smoking may make us somewhat responsible for our lung cancer doesn't mean that we must pay for its treatment. Second, the fact that a victim is irresponsible doesn't mean he can't be a victim: someone giving intoxicated consent to a medical procedure has not given the sort of valid consent that permits the doctor to operate even if the intoxicated party is at fault for losing the capacity to make considered judgments.

(43.) First, exercising the judgment needed to obey the criminal taw's norms may well require less cognitive capacity (and be less impaired by alcohol) than making good long-term judgments about whether having sex is a good idea. Second, while the drunken criminal (think about nonconsensual assault here) harms another without regard to the victim's action--so we want to pressure him to be as responsible as possible--the irresponsible consenter doesn't force the putative rapist to do anything at all, so we needn't focus exclusively on pressuring her to face further "bad" consequences for her misbehavior.

(44.) See supra Part I.A.9-11.

(45.) As I discuss in infra Part I.B.2, Wertheimer touches on, but does not lucidly explore, the relationship between judging the propriety of consent by reference to whether women would choose ex ante to call their consent legitimizing and his discussion of coercion: if (as per Wertheimer's discussion of coercion) women are not invariably permitted to choose the option set they would most prefer (unless the failure to have that option set available violates some independent moral/legal right), why might women in this setting be permitted to choose the option set they most prefer, without paying heed to the question of whether refusing them that option violates an independent moral or legal right?

(46.) Moreover, if it turned out that certain behavior with horrific consequences were truly unavoidable, it would be worth knowing that fact; shooting the (truthful) messenger doesn't do us much good (p. 68). Similarly, if critics of Wertheimer's preferred evolutionary psychological account believed that we should reject evolutionary biological propositions about sex because the widespread belief in its teachings--whether it is right or wrong, whether it actually implies moral irresponsibility for sexual aggression--would lead men to believe they could not control their aggressiveness, then these critics should supply some evidence; neither I nor Wertheimer have ever seen evidence suggesting that those who have these beliefs in fact behave less morally (p. 69).

(47.) Good, entirely mainstream works expressing varying degrees of suspicion of the methods of evolutionary psychology generally can be found in KEVIN N. LALAND & GILLIAN R. BROWN, SENSE AND NONSENSE: EVOLUTIONARY PERSPECTIVES ON HUMAN BEHAVIOR 176-96 (2002) and KIM STERELNY & PAUL E. GRIFFTHS, SEX AND DEATH: AN INTRODUCTION TO PHILOSOPHY OF BIOLOGY 324-32, 341-53 (David L. Hull ed., 1999). Claims that there are (what I describe in the text as) "highly modularized adaptations" for fertile women to find rape especially aversive or for men to be willing to force nonconsenting women to have sex with them when they are unable to woo them have been subject to even greater levels of derision by perfectly mainstream evolutionary biologists. See, e.g., STERELNY & GRIFFITHS, supra, at 316-17; Jerry A. Coyne, Of Vice and Men: A Case Study in Evolutionary Psychology, in EVOLUTION, GENDER, AND RAPE 171 (Cheryl Brown Travis ed., 2003).

(48.) See pp. 48-53. Evolutionary psychologists--and Wertheimer, to some extent (p. 50)--obviously recognize that not all traits we observe emerged as a result of selection pressure. At core, traits may be thought of as adaptations (traits that emerged as a result of selection pressure and remain behaviorally adaptive in the current environment), exaptations (traits that did not emerge as a result of natural selection but have been co-opted to meet the organism's needs), past adaptations (once functional, selection-pressured traits that now fail to further reproductive success), and dysfunctional or nonfunctional traits (traits that were neither specifically selected for nor enhance reproductive fitness). They also recognize, to some extent, that even traits that have been subject to selection pressure are not globally optimal adaptations, but they may emerge from a restricted pool of potential traits that can realistically develop given plausible mutation patterns subject to developmental biological limitations. For lucid discussions of these points, see LALAND & BROWN, supra note 47, at 132-35; Ron Amundson, Two Concepts of Constraint: Adaptationism and the Challenge from Developmental Biology, 61 PHIL. SCI. 556 (1994); Stephen Jay Gould & Elisabeth S. Vrba, Exaptation--A Missing Term in the Science of Form, 8 PALEOBIOLOGY 4 (1982); Elliot Sober, Six Sayings About Adaptationism, in THE PHILOSOPHY OF BIOLOGY 72 (David L. Hull & Michael Ruse eds., 1998).

What makes even this rather benign statement of the core belief of evolutionary psychologists more controversial than Wertheimer acknowledges is that many mainstream evolutionary biologists believe evolutionary psychologists overemphasize both the role of selection pressure and the possibility of globally optimal adaptations. See, e.g., Elisabeth. A. Lloyd & Marcus W. Feldman, Evolutionary Psychology: A View from Evolutionary Biology, 13 PSYCHOL. ENQUIRY 150 (2002). Critics allege that evolutionary psychologists are especially vulnerable to this problem when they are trying to induce which traits people have (to ascertain human traits not so much by observing them but by figuring out what traits people must have developed, given the selection pressures they would have faced in the "environment of evolutionary adaptation," or EEA). There are two problems. First, we know considerably less about what adaptive pressures our ancestors faced than evolutionary psychologists imply. See LALAND & BROWN, supra note 47, at 177-82. Moreover, even if we assume (counterfactually) that we can describe the trait that would be optimal given adaptive pressures, we cannot simply presume that trait exists and in fact emerged, given that many traits are nonadaptive and that globally optimal traits may not emerge. See, e.g., id. at 188-89; STERELNY & GRIFFITHS, supra note 47, at 341-42.

(49.) Evolutionary psychologists--and Wertheimer by implication--are wedded to the idea that human evolution is slow, so that traits which are plainly of no evolutionary advantage in environments that may have dominated recorded human cultural history may readily survive because there has not been enough time to alter the traits that were more optimal in the LEA (pp. 280-81). (Thus, even if it were tree that barely any rapists actually impregnate their victims and may not have done so for thousands--but not hundreds of thousands--of years, it would tell us little about whether a tendency to rape could be present in the brain.) While all sensible evolutionary biologists agree that there may be adaptive lag (i.e., that some traits may persist in an environment in which they are no longer adaptive), Wertheimer once again fails to acknowledge controversial aspects of this claim. Wertheimer should at least acknowledge that evolutionary theorists generally classified as gene-culture co-evolutionists argue forcefully that people whose traits "fit" their cultural environments may, in rather few generations, supplant those who do not. Take, for instance, the trait of lactose intolerance: it is extraordinarily common (eighty percent rates) in cultures that did not rely on dairy farming and far less common (less than ten percent rates) in those that did. Standard gene-culture co-evolutionary theory posits that this could happen in the rather brief (300-generation or 6000-year) period in which dairy farming has been significant in some cultures. See, e.g., WILLIAM H. DURHAM, COEVOLUTION: GENES, CULTURE AND HUMAN DIVERSITY 242, 282-85 (1991). For a quite accessible summary, see LALAND & BROWN, supra note 47, at 241-86. Given the relative rapidity of gene-culture co-evolution, the tendency to rape could either have appeared or disappeared after the LEA, assuming it was once a disposition (or once wasn't, but has become one now).

(50.) The classic account of brain modularity is set out in JERRY A. FODOR, THE MODULARITY OF MIND (1983). See also THE ADAPTED MIND: EVOLUTIONARY PSYCHOLOGY AND THE GENERATION OF CULTURE (Jerome H. Barkow et al. eds., 1992).

(51.) One would expect, and hope, that a philosopher like Wertheimer would be acutely aware that it is possible to describe all traits at distinct levels of generality so that there would almost surely be controversies over the degree to which cognitive or affective traits had been properly labeled at the apt level of granularity. Wertheimer himself sees the granularity problem perfectly clearly outside the scientific domain. See p. 206 (explaining that the legal distinction between fraud in the inducement and fraud in the factum is philosophically suspect because it turns on arbitrary decisions about whether to describe a decisionmaker's intentions in finer- or broader-grained fashion; a woman might be said to be informed that she is consenting to intercourse when she is misled that she needs it as medical treatment--a broad-grained description--but be misled that she is consenting to intercourse with someone with particular motives and which will have particular results--a finer-grained description). The problem of descriptive granularity is familiar to lawyers more generally. For example, in talking about foreseeability, we recognize that an event looked at in very fine-grained terms (for example, death by heavy bleeding in a precise portion of the brain) is not foreseeable but, when looked at in broader-grained terms (death from a blunt blow to the head), may seem a good deal more probable and foreseeable. See Michael S. Moore, Foreseeing Harm Opaquely, in ACTION AND VALUE IN CRIMINAL LAW 125, 125-27 (Stephen Shute et al. eds., 1993); see also Mark Kelman, Interpretive Construction in the Substantive Criminal Law, 33 STAN. L. REV. 591,640-42 (1981).

One could conceivably "prove" domain specificity--i.e., identify a single process that is not part and parcel of a broader web of cognitive and affective processes--only by demonstrating the existence of cell-level physiological processes of the following form: particular environmental inputs, in particular domains, either uniquely activate some reasonably small number of brain cells, but only those; or these cells are differentially activated only in the presence of these inputs. For a variety of reasons, such a picture of brain functioning seems implausible at worst, or very difficult to prove at best, especially in the domains in which evolutionary psychologists are most interested. (It is highly implausible, for instance, that even if men are "naturally" attracted to a certain female body type, they recognize the body type with a set of cells otherwise unused in visual processing.) Still, the notion that there are "dedicated" brain cells is not especially consistent with some facts about brain hardware that we do know: while genetic coding precludes the "hijacking" of just any old cell for whatever purpose seems apt--e.g., lung cells just won't record visual images, no matter what--many brain cells do appear capable of performing a variety of tasks. See, e.g., J. Sharma et al., Induction of Visual Orientation Modules in Auditory Cortex, 404 NATURE 841 (2000) (explaining that while visual images are generally transmitted to a particular part of the cortex, other brain cells can receive the same images if the usual receptor cells are damaged, though these other cells, ordinarily, are used for quite distinct purposes).

But one should note--contrary to the claim of some philosophers of science who have been especially harsh critics of modularity--that the absence of such an account of brain hardware does not disprove modularity (as understood by evolutionary psychologists). Such claims are made, for instance, in David J. Buller & Valerie Gray Hardcastle, Evolutionary Psychology, Meet Developmental Neurobiology: Against Promiscuous Modularity, 1 BRAIN & MIND 507 (2000). We simply do not know enough about the neurological hardware of cognition to say that our brains are either compatible or incompatible with the theory. And the fact that cells have many purposes does not disprove the idea that selection pressure has shaped a particular modularized use: the fact, for instance, that the penis both transports waste and impregnates does not mean its morphology was not subject to selection pressures related to its role in sexual reproduction.

(52.) In the standard evolutionary psychological picture of sexuality that Wertheimer adopts, male desire-provoking pathways are typically more modularized than female desire-provoking pathways, though the distinctions in levels of modularity are neither adequately highlighted nor explored. Men are purportedly predisposed to desire to impregnate those who appear most likely to be fertile and to survive to care for the offspring (though the adaptation that leads to the tendency to impregnate, mediated through sexual desire, more or less spills over into sexual desire for nonprocreative sex; theorists differ as to whether the desire for nonprocreative sex is best thought of as having independent adaptationist explanations or is best thought of as a byproduct). Women's purported basic sexual preference--for resource providers who will stick around to help take care of their children--is plainly, on its face, considerably less cognitively or perceptually modularized. Modern men are still supposed to find lustrous hair, facial symmetry, and a particular hip-to-waist ratio sexually irresistible because it signaled fertility to their ancestors; women are supposed to find men with expensive cars attractive--rather than men with the body types associated with successful resource gathering in the EEA--because they seek "resources." But seeking resources is on its face a less modularized cognitive/perceptual task than seeking facial symmetry (though, perhaps, feeling that one's in love once one cognitively processes, through less modularized perceptual processes, that one is in the presence of a guy with resources, is conceivably more modularized.)

(53.) See LALAND & BROWN, supra note 47, at 186 (explaining that evolutionary psychologists concede that the brain is not fully modularized, although some evolutionary psychologists advocate a quite highly modularized view). See generally Johan J. Bolhuis & Euan M. MacPhail, A Critique of the Neuroecology of Learning and Memory, 5 TRENDS COGNITIVE SCI. 426 (2001) (critiquing the neuroecological approach and finding that the "evolutionary or functional considerations cannot explain the neural mechanisms of behaviour in general").

(54.) One way of thinking about the modularity issue at the biotheoretical level is to think about the distinction between "mosaic" traits, which can develop independently of the rest of the organism's phenotype, and "connected" traits, which cannot. One way of recognizing a mosaic trait is to discover that the trait exists in some, but not all, closely related species. Thus, for instance, some closely related bears engage in certain forms of hygienic behavior while others do not. But almost all primates have two lungs, and having only one would necessitate a host of other body changes: as a result, it is unlikely that the number of lungs in the body is a mosaic trait. See STERELNY & GRIFFITHS, supra note 47, at 320-21. There are substantial reasons to believe that most cognitive and emotional traits are connected traits, which Wertheimer never even seems to recognize. It is rather unlikely that we have very specific mechanisms for hunting or for mate selection; some of the mental processes we use in hunting are helpful in storytelling, and some of the mental processes we use in mate selection help us solve a host of other social coordination problems. Id. at 321. If many of the problems involved in mate selection involve complex interpretive capacities that are useful in a variety of situations, it is doubly unlikely that the traits are modularized: first, because it is unlikely that the agent could gain adequate information based on a small number of inputs that are invariably available when he needs to make judgments, and second, because the traits that are useful in this task are also used in others. Think about what an encapsulated mechanism to judge whether one has reason to fear (present or future) sexual infidelity might look like. For more on this point, see id. at 330-33.

Moreover, while evolutionary psychologists laud the neuron capacity-saving advantages of modularization, critics believe they radically understate the vulnerability of a hypothetically modularized brain to exploitation by other agents who become aware of the modular perception properties, making modularization unstable in a game-theoretic sense. See John R. Krebs & Richard Dawkins, Animal Signals: Mind-Reading and Manipulation, in BEHAVIOURAL ECOLOGY: AN EVOLUTIONARY APPROACH 380 (John R. Krebs & N.B. Davies eds., 1984). For a more general account of evolutionarily stable strategies that account for the actions of other agents, see Richard Dawkins, THE SELFISH GENE 69-86 (1976). Take an easy hypothetical case: if some women developed a modularized feeling of sexual desire for all men who smiled and said, "I love you," treating this action as a guarantor of their reliability as child protectors, they'd get exploited by men who'd say that and dupe them into bearing their kids. (Presumably, one reason men might treat certain sorts of "looks" as fertility guarantors, and have modularized desires for women with this "look," is that appearance cannot readily be "faked.") But men with Thornhill and Palmer's sort of modularized "rape gene" are similarly stuck employing a strategy that could quickly become a losing strategy depending on the response of other agents: if females engage in even moderate levels of infanticide when they are impregnated by rapists or dominant men (the ones who can woo the women through noncoercive means) band together to kill rapists, the strategy no longer has much fitness value. If such a hypothesis were true, one would expect men who were predisposed to rape would have developed a less modularized capacity to evaluate the perils and rewards of nonconsensual sex.

(55.) It is quite unclear how modularized Wertheimer's theorists think "sexuality" is: the problem of "mate selection" could be broken down into very fine-grained modules (we could have a module to tell us when to remain faithful or to cheat on our usual partners; we could have a module that tells us when to desert the usual partner; we could have a module to tell us how much time to spend helping our gene-sharing siblings find a mate; we could have a module to tell us how to punish infidelity); midgrained modules (all of these problems could be solved by a brain with a handful of operative principles; for example, male principles that attempt to maximize the number of women impregnated and minimize resources expended on those who are not genetically related); or less "mate selection" specific mental capacities (the capacity to detect infidelity, and then to react to it punitively, is just a subset of the mind's ability to detect and punish defectors from cooperative games). Again, the reason for this ambiguity is philosophical and conceptual, not biological: it is very difficult to ascertain why a particular evolutionary psychologist describes a trait at the level of generality that she chooses to describe it rather than a broader or narrower level.

(56.) Wertheimer relies a great deal on a single nonreplicated study, originally done by Thornhill and Thornhill and subsequently diffused in the literature, which is enormously problematic. See Nancy Wilmsen Thornhill & Randy Thornhill, An Evolutionary Analysis of Psychological Pain Following Rape: I. The Effects of Victim's Age and Marital Status, 11 ETHOLOGY & SOCIOBIOLOGY 155 (1990). The study has a number of minor but nontrivial problems. It is unclear that a sample of raped women who seek emergency care is representative. The measurement of trauma level is subject to enormous criticism--reactions of prefertile girls are often given by their caretakers rather than by self-report or (what are usually thought of as more reliable) projective tests. Data on fertile and menopausal women are also generated from self-reports of the level of trauma five days after the incident, despite the fact that long-term and short-term consequences often diverge and that "trauma" is not a single concept but a complex one. See, e.g., THOMAS W. MCCAHILL ET AL., THE AFTERMATH OF RAPE 62 (1979) (stating that while severe violence during the rape does not predict poorer immediate reactions, it does worsen victims' suffering one year out); Dean G. Kilpatrick et al., Factors Predicting Psychological Distress Among Rape Victims, in TRAUMA AND ITS WAKE 113, 114-15 (Charles R. Figley ed., 1985) (noting that many, but not all, rape victims improve substantially within six to twenty-one days and within three months after their rape experiences); see also Stephanie A. Shields & Pamela Steinke, Does Self-Report Make Sense as an Investigative Method in Evolutionary. Psychology?, in EVOLUTION, GENDER, AND RAPE 87 (Cheryl Brown Travis ed., 2003) (discussing problems in the use of self-report data on trauma levels).

Far worse, even if we accepted the Thornhills' rather loose measures of trauma, they (and Wertheimer) misread their own data: it is true that fertile women are more traumatized than nonfertile women by rape, but it is not true that they are more traumatized than menopausal women. It is only by aggregating the prefertile girls (whose level of trauma was reported by caretakers) with menopausal women that Thornhill can conclude that fertile women are more traumatized than nonfertile women. This charge was first made in Jerry A. Coyne & Andrew Berry, Rape as an Adaptation: Is This Contentious Hypothesis Advocacy, Not Science?, 404 NATURE 121 (2000). Thornhill persistently denies the charge is true. See, e.g., Randy Thornhill & Craig T. Palmer, Rape and Evolution: A Reply to Our Critics, 4 PSYCHOL., EVOLUTION & GENDER 283, 294-95 (2002). But his denials are at best misleading. The overall quality of the underlying study is so poor--the authors do not regress trauma levels against a range of rape and victim characteristics but merely present gross cross-tabularized comparisons of a number of negative reactions to rape by age group--that there is absolutely no reasonable basis to conclude either that postmenopausal women are or are not more adversely affected by rape than fertile women. (One lucid example of the impact of the failure to do regression analysis is that the Thornhills blithely report that prepubescent girls are radically less scared to be out on the street after a rape than older women. Thornhill & Thornhill, supra, at 165. It is doubtless true that nearly none of the girls were raped on the street by a stranger, so their finding may have little to do with age.)

At any rate, if one looks at the entire set of postmenopausal and fertile women in the Thornhills' sample, one finds that there is no statistically significant distinction in the reactions of older women in nine of the eleven domains they mention. Id. Postmenopausal women have a significantly more adverse reaction in one domain (change in fear of known men) while fertile women do significantly worse in one domain (change in heterosexual relations). Moreover, it is not the case that the menopausal women invariably have worse reactions, but the reactions are not statistically significantly worse merely because of small sample size. In fact, in the nine cases in which there are two dozen or so menopausal women, the postmenopausal women have more adverse reactions along four dimensions, the fertile women along five. By shameless data mining of the sample--eliminating married women from the sets of both fertile and postmenopausal women (but holding no other factors constant)--the Thornhills can indeed argue that at least a subset of postmenopausal women do worse. It is this table that Thornhill cites in defending his claim from Coyne's attack. Id. at 168. There must surely, though, be subsets of fertile women who do worse as well given how nearly identical the aggregate data for the groups are.

(57.) As I noted, supra note 4, rape could indirectly compromise men's reproductive success if having been raped made a man a less desirable sexual partner for women or led to social ostracism that stripped him of the capacity to call on others in a kinship group for support in caring for his young.

(58.) Data on male reactions to same-sex rape are limited, but those that there are available suggest that levels of trauma are similar to, but arguably more intense than, levels of female trauma, especially among heterosexual men who are raped. See, e.g., Patricia A. Frazier, A Comparative Study of Male and Female Rape Victims Seen at a Hospital-Based Rape Crisis Program, 8 J. INTERPERSONAL VIOLENCE 64 (1993).

Note that this carelessness in presenting "observed universal traits" is a consistent problem in the literature upon which Wertheimer relies. Citing Buss, he notes, for instance, that women are likely to be drawn to men with resources (pp. 57-58), but Buss draws this conclusion on the basis of cross-cultural questionnaires that show that women rate "access to resources" as only the twelfth most important factor in selecting a mate. See DAVID M. Buss, THE EVOLUTION OF DESIRE 68-69 (rev. ed. 2003). The fact that women find it more important than men is also wholly predictable given gendered income inequality, without regard to adaptation. Similarly, Wertheimer notes that men outreproducing other men given selection pressure (those more wary in the EEA of investing care-giving resources in others' kids) will be highly sexually jealous (pp. 60-61). But if you look at Buss's data, not just the conclusions he draws from his data, you see that we can predict (certain sorts of) sexual jealousy rates far more accurately by looking at the respondent's country of origin than his or her gender: northern European men, for instance, scarcely care about virginity at all. See BUSS, supra, at 68-89. Buss argues that those who care little for virginity still care about future fidelity; if men have modularized brains, it is not clear how they would have been able to shed their concerns over virginity, the most tractably delimited input that would have signaled future fidelity in the EEA.

Similarly, Wertheimer accepts the "finding" that certain female body shapes (a particular hip-waist proportion) are invariably attractive, presumably because they (once) signal(ed) fertility (p. 57). But he simply wishes away empirical studies that find that these findings are extraordinarily sensitive to the use of particular forms of stimulus figures (that arguably confound judgments about weight, hip size, waist size, and fecundity). See Louis G. Tassinary & Kristi A. Hansen, A Critical Test of the Waist-to-Hip-Ratio Hypothesis of Female Physical Attractiveness, 9 PSYCHOL. SCI. 150 (1998). Other biologists question both the association between particular body shape and fertility and whether, in an evolutionary game-theoretic sense, it would be plausible that all men should develop the same tastes in women, which would lead to many men being frustrated in their desires to mate. (More generally, mainstream evolutionary biology does not lead to the prediction that adaptive traits are the same traits in each member of the species; quite to the contrary, in a game-theoretic sense, we would expect an equilibrium to develop in which a mixture of traits develops in a population, with some organisms developing certain traits based on the traits they expect others will develop. For a further explication of this point, see Lloyd & Feldman, supra note 48, at 152.)

Wertheimer repeats the common claim that one reason we know that rape is mostly a reproductive strategy for males is that they typically rape those who are fertile (p. 77). But the claim is highly problematic: prepubescent girls are raped at rates far higher than their representation in the population generally. See, e.g., Coyne, supra note 47, at 171. Moreover, though the data is highly imperfect, male rape victims appear more likely to be young adults in their early twenties, just like female rape victims. See Paul J. Isley & David Gehrenbeck Shim, Sexual Assault of Men in the Community, 25 J. COMMUNITY PSYCHOL. 159 (1997). These data suggest that a host of nonreproductive factors might explain (additionally? instead?) why relatively young people are sexually victimized (i.e., they are the same age as typical criminal perpetrators and most likely to be those whom the perpetrators confront; they are also most likely to be out and about in public alone).

(59.) Again, if Wertheimer were interested in presenting why wholly mainstream, nonideological biologists are skeptical of the writings he cites, he would note that in order to make any sense out of the claims about reproductive-age women's "special" aversion to rape, one must distinguish between what would generally be referred to as the organizing and activating effects of sex hormones. Broadly speaking, the organizing effects of sex hormones occur mostly at a sensitive stage of development--well before birth in humans--and determine whether the brain and body will develop as a female or male. Activating effects (which last hours, weeks, or even months longer than the hormones remain in the body, but not indefinitely), on the other hand, can occur at any time in life, when a hormone temporarily activates a particular response. The distinction between the two kinds of effects is not absolute; early in life, hormones exert activating effects even while they are organizing body development, and during puberty, hormones can induce long-lasting structural changes as well as activating effects. But to the degree there are hormonal effects and they are best thought of as "organizing effects," we would certainly expect that postmenopausal women, and perhaps prepubescent women, would have the same reactions to rape as women of childbearing age. See Arthur P. Arnold & S. Marc Breedlove, Organizational and Activational Effects of Sex Steroids on Brain and Behavior: A Reanalysis, 19 HORMONES & BEHAV. 469 (1985); Christina L. Williams, A Reevaluation of the Concept of Separable Periods of Organizational and Activational Actions of Estrogens in Development of Brain and Behavior, 474 ANNALS N.Y. ACAD. SCI. 282 (1986).

Oddly, the Thornhills' article does not even mention this problem. The impact of this observation on the data Wertheimer so casually accepts is ambiguous. On the one hand, one might argue that even if one accepts what strikes me as the unavoidable truth that the Thornhills' "data" is wholly unconvincing twaddle, it is still possible that the ultimate explanatory cause of female rape aversion is a narrowly modular adaptation that serves to protect reproductive choice; the aversion is simply present over the whole life cycle of females. On the other hand, it could be true that reproductive-age females react more aversively to rape (i.e., the Thornhills are right although the data they offer to prove their claim does not demonstrate that), but such a finding might not in any way reflect the fact that rape aversion is a specific biological adaptation. The adaptation (if it exists) could only plausibly be mediated through sex hormones, and if these hormones are (dominantly) "organizing," then (even genuinely) observed distinctions in reactions would not reflect adaptationist pressures. Rather, an especially aversive reaction could result from a host of other factors: for instance, reproductive-age women (rightly) feel that as a matter of (culturally contingent) fact, they will be blamed more for having been raped than younger or older women would be, or young girls might lack the purely cognitive capacity to appreciate the impact of rape (or many other life events).

One should also recognize that if the affective reactions that the Thornhills (implicitly) posit are triggered by ovulation-related hormones (and once more, one wonders what else they might believe), we should expect that the (adverse) reaction of (postpubescent) twelve-to seventeen-year-old girls would be just like that of premenopausal women and not at all like those of prepubescent girls. The Thornhills do not separately present data for the early adolescent group, but if one looks at the book that summarizes the interviews on which they rely, one finds that the eighteen-and-older age group has substantially more adverse reactions to rape than the group of twelve- to seventeen-year-olds, though along most dimensions, the adolescents do worse than the still-younger girls. See MCCAHILL, supra note 55, at 40.

Further, one would expect that if the proximate cause of rape revulsion is sex-hormone induced activation of a revulsion reaction, as the Thornhills must implicitly assert, one would expect that women's level of revulsion to, or resistance to, rape would vary over the course of the menstrual cycle and would diminish not only at the end of their reproductive years but also when they are pregnant and/or lactating (and amenorrheic). Thornhill and Palmer do implicitly endorse the "activating hormone" theory when they present data that women are less likely to be raped during the portion of the menstrual cycle when they are most fertile. See Thornhill & Palmer, supra note 56, at 101-02. Once more, though, the evidence they cite is of extremely low quality. There is some suggestive evidence that women's resistance to rape is higher when they are most fertile. See S.M. Petralia & G.G. Gallup, Effects of a Sexual Assault Scenario on Handgrip Strength Across the Menstrual Cycle, 23 EVOLUTION & HUM. BEHAV. 3 (2002) (finding that women in the ovulatory phase of the menstrual cycle showed a marked increase in handgrip strength after reading an essay depicting a woman walking to her car pursued by a strange man, but handgrip strength did not increase for women in any other phase of the menstrual cycle who read this essay, nor for ovulating women who read an essay in which a woman was walking to her car with many other people around). But the (properly unpublished) evidence that Thornhill and Palmer cite both about rape rates over the course of the cycle and rape-avoidance behavior over the course of the cycle are nearly worthless, for reasons well explored in Daniel M.T. Fessler, Rape Is Not Less Frequent During the Ovulatory Phase of the Menstrual Cycle, 5 SEXUALITIES, EVOLUTION & GENDER 127, 131, 134-38 (2003).

(60.) It is especially peculiar that Wertheimer so blithely buys into the idea that women would not develop "excess" revulsion (beyond that needed to protect reproductive choice) when he simultaneously blithely accepts (and must accept, given the data) that men have "excess" desire for nonconsensual sex from a reproductive vantage point. One-third of rapes are of women of nonreproductive age, twenty percent do not involve vaginal penetration, and fifty percent do not involve ejaculation in the vagina. See Coyne & Berry, supra note 56, at 121. Plainly, the data on nonreproductive male sexual coercion cast doubt on a very finely grained modular view of male sexual desire (and even more on an even more finely grained disposition tu rape when frustrated in one's desire to reproduce); it is incumbent on Wertheimer to explain why he thinks that scientists suspicious of the unexplained move from extreme modularity in women to relative nonmodularity in men on this issue must be motivated by ideology rather than reason.

Moreover, Wertheimer selectively cites data to suggest that reproductive-age women do not experience "reproductively" excess revulsion (p. 116). While it may be true that fertile women are less traumatized by forced oral sex than forced vaginal intercourse, there is plenty of (uncited) evidence that forced anal intercourse is associated with more intense trauma (e.g., ongoing nightmares) than any other form of sexual assault. See, e.g., MCCAHILL, supra note 55, at 66.

(61.) See supra Part I.A.6.

(62.) Wertheimer notes, again quite typically, that if B chooses sex rather than the violation of a "trivial" legal fight--for example, sex to stop A from briefly trespassing on her property--she will not be thought of as coerced into sexual conduct because it would not be reasonable for B to succumb to the threat to worsen her position below the baseline (pp. 165-66, 184).

(63.) Wertheimer acknowledges this problem: "To say that we should evaluate the coerciveness of A's proposal by reference to B's moralized baseline is completely neutral with respect to the specification of that baseline" (p. 169).

(64.) Many theorists who broadly model the sexual coercion problem in the same manner as Wertheimer identify B's entitlement baseline with her legal rights to an even greater extent than Wertheimer does. See, e.g., Keith Burgess-Jackson, A Theory of Rape, in A MOST DETESTABLE CRIME 92 (Keith Burgess-Jackson ed., 1999).

(65.) One way of thinking about one key aspect of the point I am about to make, albeit one phrased somewhat differently than I will frame it, is to note that the question of whether A is obliged to give some particular valued good (G) to B is a separate question from the question of whether A is allowed to transfer G to B at a higher-than-market price. Thus, as I will explain, it may be proper to say that an auto repairman who happens upon a stranded motorist has no obligation to repair her car, but, if he chooses to repair, he might not be free to charge a monopoly price (in money terms) or to demand an even higher psychological "price" (unwanted sex). This way of making the point is emphasized in BARBARA H. FRIED, THE PROGRESSIVE ASSAULT ON LAISSEZ FAIRE 61, 112-13, 246 n. 186 (1998).

(66.) The utilitarian-style analyses in the text are meant to be nothing more than rough-and-ready parodies of a sophisticated social welfare function maximizing analysis; they are merely designed to illustrate a style of analysis rather than be analysis. I also set aside temporarily the issue I will return to in Part II: Is welfare a coherent concept?

(67.) Not only is the utilitarian analysis in the text self-consciously sketchy, it is also vital to note that the "redistribution" paradigm does not commit us to welfarism at all: one could assess the propriety of this redistribution through the lens of rights theories. Libertarians, for instance, would reject the permissibility of taxing A; certain sorts of positive autonomy theorists and those committed to increasing people's capacity to flourish rather than their hedonic state would describe the gains to B not in terms of welfare gains but gains in her capacity to live an adequately self-governing life.

(68.) This case is parallel in most important respects to Wertheimer's "Gangrene" (p. 172).

(69.) Compare this case with:</p> <pre> Fastest Swimmer: B's preference order is: (1) win races + no more practice; (2) win races + more practice; (3) lose races + no more practice. </pre> <p>At some level, the "necessity" to practice in order to "win" is partly socially constructed--the existence of a cultural set of conventions in which competitive swimmers all practice means that if one wants to beat other swimmers, one must practice too. But it is vital to note two points: First, A is fundamentally superfluous to whatever mechanism we might choose to use to empower B, if there is one, though A is the one who announces the limits that B faces (i.e., "if you want to win, you must practice"). There is little A can do to change the general social conventions, and telling him not to alert B to her limited option sets doesn't really empower her. Second, of course, it might not be a good thing to use any mechanisms that would empower B in this way. We may think that the world in which swimmers train is a better world, given a host of distinct ways of evaluating what makes it better.

(70.) This case is parallel to Wertheimer's "Weapon" (p. 164).

(71.) It seems appropriate to note at this point that there may be "additional" (essentially welfarist) reasons to eliminate something that looks more like the sort of coercion Wertheimer describes, which go beyond the desire to expand the "victim's" power and option sets: the victim may bear particularly acute or distinct losses when she is unable to get something she desires because another person is acting wrongfully. Return to the Fast Swimmer/Fastest Swimmer cases: a swimmer who was told by a coach that she'd have to practice more if he is to serve as her coach--not because it will help her swim faster or win races but rather because he simply likes to exert whatever power he can over helpless youngsters--will (deservedly) feel worse that she is constrained to practice more (if she wants to keep working with the otherwise-excellent coach).

I raise this issue, above all, because I think it is quite relevant to the debate between those (like me) who think that antidiscrimination and accommodation requirements differ. In my view, the employer or public-accommodations owner who resists a demand to accommodate is essentially like the coach who tells the swimmer she will have to practice more to swim faster: he is merely announcing a fact about the world (how many resources must be devoted to social inclusion), and while the swimmer (and unaccommodated patron or job applicant) may wish she had the power to get what she wants (social inclusion without resource trade-offs), the coach (and accommodation resister) bears no special role, let alone a role as a wrongful actor, in making a desired option unavailable. I discuss this point in more detail in Mark Kelman, Defining the Antidiscrimination Norm To Defend It (Sept. 2005) (unpublished manuscript, on file with author).

(72.) This case is close to two cases raised by Wertheimer, "Greedy Mechanic" (p. 170) and "Rescue" (p. 176).

(73.) One thing that is interesting to note is that the mechanisms focus on A, rather than third parties, because he is relatively uniquely situated to empower her. Compare the case with a variant on Wertheimer's "Lecherous Millionaire" (p. 128):</p> <pre> Lecherous Millionaire: In my terms, B's preference order is (1) no sex with A + money to feed her hungry kids; (2) sex with A + money; (3) no sex + no money. </pre> <p>In this case, A is not uniquely situated to provide B with money (the tax-and-transfer system might do better), and we may think there are no good reasons to use mechanisms that especially involve him. Now in the Isolated Motorist case, we could empower B by establishing a collective system (akin to a tax-and-transfer in the Lecherous Millionaire case) that would send out lots of public rescue trucks so that she would never face a private monopolist. It is because the social costs of such a system seem so high that we feel the mechanism we use, if any, will focus on A himself.

(74.) Just as criminalizing threats of nonsexual violence does not guarantee the presence of option (1) but merely increases it, we may use a mechanism in this case that merely increases the frequency of option (1) availability.

(75.) Wertheimer recognizes this problem to some degree (p. 181), but he is unable to integrate it into his account of coercion, which focuses on "moralized expectations" rather than assessing the consequences of a particular redistributive mechanism. As a result, he seemingly concludes, far too hastily, that we should allow consent (to high payment or to sex) to legitimize these transactions because they permit the woman to do the best she can in an unjust world (pp. 190-92). Note, too, that in the Lecherous Millionaire case, there is no realistic chance that A would offer option (1) if he could not offer (2) (p. 128).

(76.) One can translate this concern into the coercion framework (and at times I think Wertheimer may be struggling to do so), but it is labored. One could argue that in a subclass of cases, like this one, A coerces B because he deprives her of her statistically based expectations (repair without sex) or argue that while Wertheimer treats the moralized and statistical expectations baselines as alternatives, that in fact the moralized baseline is the statistical baseline when the sole reason Bs will not obtain the statistical baseline is that A sees the possibility of eliciting otherwise-unwanted sex.

(77.) I also set aside a number of quite important issues in this discussion that suggest that even if women prefer (2) to (3), (2) should not be a permissible option: Even if we do not have a canon of good reasons to have sex, do we believe certain reasons are so bad that they should be taken off the table, for the sake of a woman's interests in her long-term flourishing? Similarly, are there widespread social costs to women treating their sexuality as an overt "bargaining chip asset" (either because such behavior discourages the development of other assets or encourages men to ignore other assets)? Some of these issues are discussed well in KIMBERLEY A. YURACKO, PERFECTIONISM AND CONTEMPORARY FEMINISM 66-70 (2003).

It is worth noting in this regard that the reasons we might give for thinking it unwise to permit Bs to treat their sexuality as a bargaining asset might well be situation-sensitive, in ways that are skewed to the categories the "coercion" framework encourages us to employ. Thus, both the student who trades sex for a higher, undeserved grade (pp. 178-80) and the woman who trades sex for a huge sum of money from a lecherous millionaire are not morally entitled to the "good" they bargain for (the higher-than-deserved grade; money from a stranger), but we may well believe that sexual bargaining over money and status is less demeaning (whether directly, as prostitution, or indirectly, as dating the repulsive rich) and does less to undermine women's sense of the importance of their own nonsexual virtues (and men's sense of women as whole persons) than does allowing bargaining in school and at workplaces.

(78.) Thus, they think they will hedonically regret drunken consensual sex--they are not the girls who drink in order to have sex or those who enjoy it more than they regret it. At the same time, a party-less life seems dismal to them.

(79.) I set aside the thorny problems of how to account for distinctions in preferences among women.

(80.) He recognizes the overlap just once (p. 181).

(81.) Wertheimer believes it should count because he makes the empirical assumption that they want it to count because the pleasures of intoxicated sex so outweigh the regret losses.

(82.) From this perspective, the reason that we would not want to delegitimize consent to sex premised on the statement "Have sex or we will break up" (p. 151) is not that B has no moral entitlement to a continued relationship. Rather, it is that any mechanism needed to give B her preferred option radically more often ("no sex + continued relationship with A") will harm both A's interests (in utility terms and autonomy terms) and probably harm B so much more than it helps her that she would likely want her consent to count. (If he merely breaks up with her because they are not sleeping together and he is not allowed to tell her that he intends to do so, such a situation doesn't do her any good. So she might well choose to live in a world in which he can announce his intentions and still act on her subsequent manifestation of consent.) Obviously, these tests converge if we think of the redistributive mechanism as the imposition of an affirmative duty on B to continue dating A: in that case, the redistributive perspective merely tells us why we shouldn't impose a moral entitlement to a particular end-state.

At the same time, it is simply inadequate to say, as Wertheimer does (p. 187), that sex obtained through various forms of verbal pressure (assuming it is not interpreted as conveying the willingness to use physical force if the verbal pressure is resisted) is acceptable, merely because taunting or abusive language is not itself regulated. B may not have an independent entitlement that A refrain from calling her a "prude" or a "tease," but if we (morally, if not legally) attempt to delegitimize consent obtained in a situation in which the woman cannot reach her optimal option state (no sex + no taunts), we may well think that women's lot improves a good deal more than men's lot declines by restricting taunts. Just as we may think it improper to impose an independent affirmative duty to repair on the mechanic, we may think it improper to impose an independent negative prohibition against the use of taunting words. But that does not tell us that we should not change men's incentives to taunt by stripping them of the power to get sexual access by taunting (and implicitly or explicitly promising to stop if women consent to sex).

(83.) 539 U.S. 558 (2003).

(84.) For these purposes, I set aside the degree to which rule utilitarianism, for instance, blurs the line between welfarist and deontological theory along this dimension by forbidding consequentialist assessments of each action standing on its own.

(85.) Their commitment to individual autonomy, in the relevant sense, buttresses (among other things) a deep antipaternalism, in opposition to both perfectionist and welfarist challenges. Contra one sort of paternalism, subjects should not be obliged to take any set of actions that fulfills some perfectionist observer's notion of human flourishing; contra other forms of paternalism (those implied by welfarism), we should not overturn properly autonomous choice just because subjects will be "happier" if their decisions are trumped. The question of whether the ostensibly rejected paternalism is merely reintroduced in the guise of creating boundary conditions for admission into the protected group of autonomous agents whose will should be free from interference is a question I hope we will better understand at the end of the argument in this Book Review.

(86.) It is important to distinguish perfectionists from paternalists, though their concrete political programs might often converge. Paternalists may well, at core, question the capacity of subjects to meet their own ends and, therefore, guide individuals to take concrete actions that advance ends still of the subject's ultimate choosing. By contrast, perfectionists believe it is important to ensure that people act "virtuously" or "excellently," regardless of whether virtuous actors are helped by such conduct or not. One of the reasons the line is blurry, of course, is that its neat distinction between ends and means is itself anything but neat. A paternalist asymptotically approaches a perfectionist to the degree to which she believes that a person who does not seek virtue does not yet understand her own truest, most reflective ends. This point, I hope, will be clearer as we work through the text of the Book Review more generally, since it is central to many of the arguments.

My sense is that most theorists that one would conventionally label perfectionists care above all about what Rawls so aptly refers to as "the realization of human excellence in the various forms of culture." JOHN RAWLS, A THEORY OF JUSTICE 25 (1971). These perfectionists are less concerned with insuring a multiplicity of lives manifesting "virtue" (however defined) than insuring the social presence of the best representatives of human capacity. I think (very tentatively) that the distinction between the perfectionists I describe in the text and the sort of perfectionism Rawls associates with Aristotle and Nietzsche at core refers to disputes about the aggregation and distribution of the teleological good within the system. Id. at 325. (Those drawn to utilitarianism--who treat subjective utility as a teleological good in the same way that perfectionists treat "virtuous conduct" as a teleological good--also dispute how utility is best distributed across persons and how to balance distributive "gains" against gains in the "quantity" of the desired good.)

(87.) Don't many ostensibly autonomy-enhancing rules in fact trade off the interests of a subset of the ostensibly protected group for the interests of another in the very formulations of the rule? Assume, for illustrative purposes, that the world has ten women. Five of them can best be described as women who drink in order to have sex that they deeply want to have beforehand, never regret afterward, and cannot have sex without the inhibition-reducing power of alcohol. (They do not merely elect to risk having hedonically "bad" sex; they drink earlier knowing they will have "good" sex that they would not have unless they drank.) Five of them are women who deeply regret sex that they will in fact have if intoxicated. Assume, too, that there is no way that any man can recognize ex ante which sort of woman he is with. Finally, assume that we are able to make interpersonal utility comparisons and that the losses to the women who regret the sex far outweigh the gains to those who self-consciously drink in order to disinhibit themselves. (This problem arises in rape law not just in relationship to intoxication but to verbal assent: surveys of college students reveal both that many women say "no" when they mean "yes" and many others do not express verbal assent or nonassent at all. It is easy for aggregating welfarists to explain why men should not--legally and morally--act on the supposition that any particular "no" is insincere, even if the insincere "no" is quite prevalent. The losses to those whose "no" was sincere substantially outweigh, in welfarist terms, the gains obtained by women who enjoy sex more if they have sex after ritual protest or silence than they do if they must speak honestly about their sexual desires.) Naturally, one can imagine altering the proportions of women with each trait and/or the hedonic experiences that those hurt and those helped by the practice of having sex with inebriated women experience. That plainly matters for a welfarist. The question remains: Is there any way of thinking about autonomy that makes it irrelevant?

What seems plain is that a rule against sex with inebriated women (or those who the men think may well be "ritual protesters") sacrifices both the utility gains (conventionally understood) and interests in autonomous expression (conventionally understood) that those who drink to disinhibit (or those who ritually protest to increase their comfort with the sexual encounter) have. As I try to illustrate in the remainder of this Part, the rule cannot be justified as necessary to preserve sexual autonomy in some general sense--it seemingly runs roughshod over the sexual autonomy of the disinhibitors and ritual protesters. It is either perfectionist (the right way to make sexual decisions and express sexual desire is soberly and verbally) or welfarist. But the key point for now is that any rule--a rule permitting or forbidding sex in these circumstances--is aggregative because it must be: the people taking action (the men who do or do not have sex with inebriated women or those who say no) are inevitably dealing with an aggregated group because distinctions within the group are unobservable. If we tell them to treat all people the way only some of the group members would want to be treated, we are inexorably asking those who get treated in the undesired fashion to sacrifice for the sake of others.

(88.) It is possible, at the very general level, to attempt to explain or justify the existence of the prohibition against (acts that are unquestionably) rape in either deontological or welfarist terms. For deontological rights theorists, control over the integrity of one's body and one's sexuality may seem as close to a core "right" as one can imagine: a rapist unquestionably treats his victim's body as a means to the gratification of his own ends, without regard to his victim's agency or her capacity to choose ends. Welfarists, in turn, will argue, in regard to battery generally and in regard to sexual assaults in particular, that the utility gains to attackers are almost certainly lower than the utility losses to victims. Naturally, there are questions about whether welfarists (must) "count" the welfare gains of attackers at all in constructing social welfare functions or whether they must "launder" such "illegitimate" preferences. To the degree that immoral preferences do not count, the distinction between deontological or perfectionist and welfarist theories further blurs.

(89.) Generally, those committed to more restrictive views of when choices are autonomous are more protective of negative autonomy.

(90.) Wertheimer shows unquestioned sympathy for the possibility of resolving "empty" disputes about the dictates of autonomy theory by recourse to suppositions about welfare effects on a number of occasions. See, e.g., pp. 108-09 (arguing that, because the purpose of creating rights is merely to protect interests that are important to people, distinctions between nonconsensual kissing and intercourse are a matter of welfarist harm, not distinctions in degree to which autonomy is violated); p. 111 ("[I]f sexual relations with minors that were otherwise consensual did not frequently result in aversive experiences, then I believe we would and should say that they are not particularly harmful."); pp. 251-52 ("[T]he emphasis on the objective harm or wrong of violations of sexual autonomy can take us only so far [when discussing why intoxicated consent may not be problematic]. We do not think that the right to control access to one's own body on each potential occasion is of monumental moral importance independent of the more tangible physical and emotional harms.").

(91.) Wertheimer usually purports to disclaim such perfectionist resolutions. See, e.g., pp. 130-41 (questioning "whether mutual sexual attraction and intimacy should be the sine qua non of morally permissible sex"). At times, he shows some sympathy for them. See, e.g., pp. 225-26 (arguing that mentally retarded women who never experience welfare losses from consensual sex may still be "objectively" harmed). I argue below that it is not truly conceptually possible to reject variants of "perfectionism."

(92.) He states his commitment to a welfarist resolution of this problem in unambiguous terms: "If retarded females typically end up feeling very hurt in such cases because they do not understand how sex will affect them, then there would be reason to regard their consent as invalid" (p. 224).

(93.) Wertheimer acknowledges he has a limited view of how intoxicants affect behavior (pp. 236-37).

(94.) Even if she weighs the same factors differently--for example, she is more responsive to physical feelings than to the need to cement a relationship when drunk although she can reason about both--it is hardly clear that the distinct weighing is anything more than an inconsistent will. Hypotheses about the mechanisms by which alcohol changes decisions are complex and contested, but take, for instance, the fairly commonplace "Inhibitory Conflict Model," initially associated with Steele and Southwick. See Claude M. Steele & Lillian Southwick, Alcohol and Social Behavior h The Psychology of Drunken Excess, 48 J. PERSONALITY & SOC. PSYCHOL. 18 (1985). Assume that we find that inebriated women are most likely to change their decisions about whether to have sex in a situation of high inhibitory conflict--i.e., when they find a man both attractive and "sexually risky," in terms of relationship potential, pregnancy, or disease transmission. Women who are inebriated still perceive the risks as well as sober women, though they tend to give lower estimates of the risks and elevate estimates of positive outcomes (relationship potential). Inhibitory cues are being suppressed, but it is not clear whether the seemingly cognitive outcomes (lower risk estimates) motivate or merely rationalize a desire to pursue the attractive men. Interestingly, when there is low inhibitory conflict (risky but unattractive men, attractive and less risky men) decisionmaking may change very little. See Sheila T. Murphy et al., Inference Under the Influence: The Impact of Alcohol and Inhibition Conflict on Women's Sexual Decision Making, 24 PERSONALITY & SOC. PSYCHOL. BULL. 517 (1998).

(95.) Think about two cognate cases: inebriated women who choose to go out dancing in public despite their usual shyness and women who masturbate only (or more often) when drunk. Ignoring the question of who might have a duty to interfere with the second activity, should we treat these decisions as problematic? If we are not worried, why not if we worry about drunken consent to sex with others? The unabashed welfarist or perfectionist has certain stylized responses (though the welfarists may be willing to dispute empirics, and perfectionists may do the same as to the "brand" of perfectionism). Drunken sex with others is bad to welfarists because its expected value is negative (or it is unwise for risk-averse people to choose given the possibility of calamity), and it is bad to perfectionists if there is a canon of "apt" human sexuality with which intoxicated sex does not jibe. But for autonomy theorists, the cases are quite messy: we might think the dancer or masturbator is inadequately autonomous, but we may simply not care. The possibility of serious harm is a precondition for moral concern (though most autonomy theorists are quite concerned with cases in which they assert that no experiential harm will occur--e.g., sex with comatose victims or operations performed without informed consent to which patients would have consented--and they purport to believe this without regard to the question of whether the putative defendant has risked, if not caused, harm). Alternatively, we might believe what makes the woman unable to figure out whether sex with others is okay is that there are many considerations to be brought to bear and that these considerations are difficult to balance once she is drunk; the calculus for the would-be masturbator is simpler precisely because few issues must be balanced. But then, of course, we have the question of (1) whether there are women too unintelligent to be allowed to masturbate or go out dancing (if we could stop them) and (2) whether really smart, inebriated women are still "smart enough" to calculate whether to have sex with others.

(96.) Kelman, supra note 12, at 391-97.

(97.) "Welfare was classically defined by (Benthamite) hedonic utilitarians as a sensation (pleasure net of pain). But the claim that subjects' welfare is increased when and only when they achieve a particular kind of sensation is incompatible with respect for the diversity of life ends." Id. at 391. In particular, if a subject seeks (and achieves) what she sees as religious salvation rather than "pleasure," why should we not describe her as well off?

(98.) The usual strategy that preference utilitarians (and Wertheimer, using somewhat different terminology, see, e.g., p. 125) employ is to argue that people are extraordinarily likely to achieve satisfaction if their informed and prudent preferences are met. So, the issues of competence that Wertheimer is rightly concerned with--deception, youth, mental retardation, and intoxication--go to the (mixed) questions of information and prudence. (Thus, for instance, an intoxicated woman may be imprudent in the sense that she is unable to align her manifest preferences with her long-term stable preferences and may also be unable, cognitively, to process instrumentally relevant information--e.g., about the risks of pregnancy or STDs or the risks that she is sleeping with a jerk who will shun her come the daylight--that she would find relevant given the preferences that she maintains although inebriated.) The problem is that we cannot know whether a choice is informed or prudent unless we know that it made us happy, and without a robust account of what happiness is--without a robust account of the very thing Bentham wrongly claimed to provide--we cannot know what it would mean to say that a choice makes us happy. Thus:</p> <pre> Consider the problem of inadequate information. At least two different kinds of (better) information might be thought to lead to hedonically better choices: information about the nature of the goods a person is evaluating and information about the causal relationship between desire-satisfaction and ultimate well-being. Correcting for inadequate information in the first sense plays a morally innocuous role in preference utilitarianism. It does not require that we "correct" a person's desires, but only his reasoning about the actions required to satisfy those desires. </pre> <p>Kelman, supra note 12, at 395-96. For example, individual S already knows that he prefers water to poison, but when he looks at the liquid in glass A, he mistakenly believes it is water and so desires to drink it. In this case, "[c]orrecting [the] factual misapprehension is entirely consistent with honoring preferences, since [S]'s true preference here, sensibly understood, is" to drink water, not to drink the liquid in the glass, whatever it turns out to be." Id. at 396. Furthermore:</p> <pre> Correcting for inadequate information in the second sense, however, ultimately collapses preference utilitarianism into hedonic utilitarianism. If preference satisfaction improves welfare only when preferences accurately predict the hedonic states we will experience if they are met, then any preference whose satisfaction makes us unhappy is, by definition, inadequately informed. Ascertaining whether satisfying any given preference, or preferences in general, is in fact welfare-enhancing therefore requires us to "peek ahead" to end-states, to see whether the realization of a particular desire will have turned out to be hedonically satisfying. But we can carry out this exercise only by answering the questions that preference utilitarianism hoped to avoid: what do we mean by happiness, or well-being, and how do we measure it? In short, it can be answered only by reintroducing all the descriptive and normative difficulties of hedonic utilitarianism.

Similar problems arise in correcting for choices made "imprudently." New hedonics psychologists have demonstrated repeatedly that people's choices vary with context. For example, people make different decisions, depending on whether they are in a "hot" (or emotionally "aroused") state or in a "cold" (non-aroused, dispassionate) state.... Thus, the typical shopper will buy more food if he is hungry when he enters the grocery than he would if he entered it satiated.... [But it is not clear why--on purely procedural grounds, divorced from reference to the hedonic outcomes of the distinct choices--we would say that buying more when hungry is imprudent. After all, focusing just on procedure, one could readily argue that one will eat when hungry so that one's purchases in a satiated state more poorly reflect the desires one will have at the relevant point of consumption.] The conventional presupposition that [purchasing] decisions made in a "cold" state are more prudent[, hedonically speaking,] must import some substantive intuition that "hot" choices turn out poorly, hedonically. [For example, Americans eat or buy too much for their own good.] This, like other intuitions about which procedures lead to "prudent" decisions, may or may not be a defensible empirical hunch, given certain (unstated) assumptions about what decisions are in "people's own good." For present purposes, the key point is that the only way to determine which of those preferences is more prudent is, again, to peek at the answer to the question that preference utilitarianism hoped to avoid: is the subject actually happier with the outcome chosen in the "cold state"? </pre> <p>Id. at 396-97 (footnotes omitted).

(99.) Id. at 405-06. There is no reason to think that any particular person should prefer a life in which, at many particular moments, she is happy with what she is doing rather than one in which she considers her life to be well lived from a variety of distinct, longer-term perspectives. Even the experience of something as seemingly "banal" as short-term physical pain may not be best understood as the sum of the moment-by-moment levels of pain as it is understood as a reaction to an "event chunk" whose meaning is assessed in terms of whether the incident on the whole was bearable or ended as badly as one feared.

The key point, though, is that welfarism is suffused with weak perfectionism. Again, here's how I put it:</p> <pre> I do not wish to defend the idea that the "true self" is reflective rather than caught in the moment; quite the opposite. The idea of the reflective self, like the idea of the disintegrated subject elicited by moment-to-moment assessments, rests upon a particular normative ideal of the self. Rather than defending any particular ideal, I want to suggest that endorsement of some such ideal is unavoidable, so that the choice of elicitation methods for measuring happiness is a choice of what sort of subjective experiences (all of which might arguably fit under the rubric of "happiness") people should value. That (normative) choice has to be made by criteria that are external to welfarism itself, and will inevitably reflect the evaluator's judgments about what makes a life worth living. Many of those judgments seem more naturally the province of deontological or perfectionist (autonomy-based or other) arguments than welfarist ones. What we find when we go looking for evidence about the relationship between people's circumstances and their well-being will depend on how we make these judgments. To note just one example: if we try to figure out whether more material consumption brings happiness, we may get one answer if we look to preferences as revealed in leisure/work tradeoffs; quite a different one if we privilege instantaneous approach/avoidance ecological momentary assessment; and still another one if we privilege deathbed

confessionals, as the final summing story that people tell about

their lives. </pre> <p>Id. at 410-11 (footnotes omitted).

(100.) I largely set aside the following (nontrivial) question that relates more to the theory of the proper contours of the criminal law than it relates to the question of whether the putative victim's interest in freedom from "harm" (or from interferences with her autonomy) would be met by punishing the defendant's behavior. It is possible, of course, that defendants ought to be punished criminally without (exclusive) regard to the question of whether they have harmed another individual or individuals or interfered with their private rights. The prohibition on cruelty to animals, for instance, need not depend in any way on suppositions about whether animals experience pleasure and pain in a manner that is commensurate with human welfare or on questions about whether the animals are rights holders. Instead, the defendant may be seen to reveal his antisocial disposition (for "treatmentists," like the drafters of the Model Penal Code who believed that criminal acts were simply diagnostic signals of the antisocial disposition that required "peno-correctional" intervention) or may be seen to disrupt social order (what was conventionally referred to as "the King's peace").

I thus set aside an issue that I find extremely difficult. Assume that we simultaneously believe that a cognitively capable man who has sex with cognitively deficient women is justly punished because he has revealed himself to be an antisocial exploiter of vulnerable women and believe that the woman has some right to have sex. Have we impermissibly burdened her "formal" right to have sex by subjecting the people necessary for her to realize this right in practice to the risk of criminal penalties? Does it matter if there are (a significant number of) men with whom she might have sex who would not be at substantial risk of being adjudged to have revealed the relevant sort of antisocial disposition? Clearly, we do not truly protect a right if we make it criminal to exercise it. A law that decriminalized obtaining or seeking an abortion while criminally punishing those who performed abortions would not be a regime that was truly protective of abortion rights. By contrast, it is not so clear that the presence of a "criminal" is an inevitable correlate of a cognitively dysfunctional woman's exercise of her (putative) right in the same way that an abortionist is inevitably present when a woman aborts a pregnancy.

(101.) 371 F.3d 1027 (9th Cir. 2004) (Berzon, J., concurring in part and dissenting in part).

(102.) The statute under which the defendant was prosecuted states that a person who has sexual intercourse with another person is guilty of first-degree rape if "the victim is incapable of consent by reason of mental defect, mental incapacitation or physical helplessness." OR. REV. STAT. [section] 163.375(1)(d) (2003). The statute further states: "'Mentally defective' means that a person suffers from a mental disease or defect that renders the person incapable of appraising the nature of the conduct of the person." [section] 163.305(3).

(103.) Anderson, 371 F.3d at 1037-43 (footnotes and citations omitted).

(104.) Alternatively, of course, we could believe that we can both measure the aggregate amount of pleasure and pain the cognitively impaired woman felt over her lifetime that was caused by the sexual encounter and believe that all she should care about is pleasure and pain. I take it Wertheimer would share my sense that this sort of primitive Benthamite calculation is both infeasible and unacceptable.

(105.) There is nothing special about JH's cognitive impairment here. Think about telling your best friend that her husband is cheating on her so she can make a more informed decision about whether to divorce.

(106.) How can we know if your friend (whom you first met, however briefly, in supra note 105) is "better off" in a marriage founded on a lie than living life as a facer-of-hard-truths when we know that she will seem more or less better off depending on how we elicit information from her about her welfare? Do we, as weak perfectionists, merely sum up the number of pleasant days she has experienced, or do we put a great deal of emphasis on whether she can describe having experienced the form of pride she might reveal in describing herself as someone who faced up to the world, warts and all?

(107.) Think too about the problem of hedonic adaptation. Should we be crossing our fingers that people simply never learn that they are oppressed so that they will meet their low expectations of their lives or compare their rotten situations only to the rotten situations of those in their reference group? Or are autonomous people counterfactual people, once we recognize that autonomy theories are not impervious to perfectionist correction? For discussion of the pro-hedonic impact of lowered expectations, see Kelman, supra note 12, at 400. For discussion of the tendency to be much better off when one knows things are bad than when one holds out hope of a better outcome, see id. at 408 n.43.

Mark Kelman, William Nelson Cromwell Professor of Law and Vice Dean, Stanford Law School. I am grateful to Nick Kelman for research assistance, particularly, but not exclusively, on issues touching on evolutionary psychology and the impact of alcohol on cognition and emotion, and to Ari Simon for further research. Thanks, too, to Sharon Dolovich, Dick Fallon, Barbara Fried, Mark Greenberg, Karen Parker, Seana Shiffrin, and participants at the Public Law Workshop at Harvard, the Legal Theory Workshop at UCLA, and the faculty workshop at Stanford for helpful comments on portions of this piece and some cognate works that are incorporated to a significant degree in this paper.
COPYRIGHT 2005 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kelman, Mark
Publication:Stanford Law Review
Article Type:Book Review
Date:Dec 1, 2005
Previous Article:Fixing FACA: the case for exempting presidential advisory committees from judicial review under the Federal Advisory Committee Act.
Next Article:Separation of powers and the criminal law.

Related Articles
Sexual wisdom.
Franklin E. Zimring, An American Travesty: Legal Responses to Adolescent Sexual Offending.
Tensions in the struggle for sexual minority rights in Europe; que(e)rying political practices.
The age of consent; young people, sexuality and citizenship.
Chronic Pain and Family: A Clinical Perspective.
Sex crimes; perpetrators, predators, prostitutes, and victims, 2d ed.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters