Sex and the Single Malt Girl: How Voluntary Intoxication Affects ConsentBy Kevin Cole
78 Mont. L. Rev. 155 (2017)
Among the most troubling and practically important questions in defining sexual assault is the role of voluntary intoxication on consent. My focus here is voluntary intoxication short of unconsciousness, where a finding of “consent” would be upheld if not for the woman’s intoxication. While consent secured through surreptitious intoxication is widely criminalized, many jurisdictions, as well as the influential Model Penal Code, do not view voluntary intoxication as vitiating consent. This article addresses the reasons for that position and how these reasons—while important—fail to justify a uniform approach to all cases involving voluntary intoxication short of unconsciousness.
Voluntary intoxication can affect consent in two ways, and discussions of the problem often fail to address the impacts separately. First, intoxication can lead a woman not to understand the sexual activities in which she is engaging. In cases like that, we cannot say that the woman even agreed to engage in those activities—a requisite of simple assent, which is itself a requisite of consent. No special intoxication rule is necessary for this situation; we could simply litigate on a case-by-case basis whether a woman in fact assented. (A special rule could be defended, however, on the grounds that it avoids fact-finding difficulties that would occur if we attempted to determine on a case-by-case basis whether an intoxicated woman actually assented.)
My primary focus is on the second way that intoxication may affect consent—by leading women to give what may be characterized asinauthentic consent. How should we handle intoxication that leads a woman to assent to conduct that she would have rejected if sober, conduct that she would regard (and regret) as out of character?
The discussion is complicated by the scalar notions of capacity and intoxication. One is not simply intoxicated or not, impaired or not, but rather we think of people as being more or less intoxicated or impaired. For simplicity, however, when I use the phrase “significant intoxication” here, I mean the state in which a woman’s consent to sex would seem morally problematic if assessed solely at the time of the sex act in question. People will also have different intuitions about what capacity is needed to consent to sex and about whether mere capacity, rather than some actual subjective deliberative process, is necessary as a moral matter.
Even assuming that the more demanding subjective approach is the correct moral focus, a woman’s failure to engage in that process contemporaneously with assenting to sex does not necessarily mean that we should regard consent as morally invalid in all cases of significant intoxication. As Alan Wertheimer pointed out in his excellent treatment of the subject, prior to becoming significantly intoxicated, a woman may advert to the possibility that she will assent to sex while significantly intoxicated. In essence, the woman may decide to delegate sexual decision-making to her later, significantly intoxicated self. We can refer to the idea that such consent may be adequate as the “tracing” theory. The tracing theory is a strong reason to resist viewing voluntary intoxication as vitiating consent, though I ultimately conclude that the argument should lead us to distinguish between cases of what may be thought of as significant intoxication and extreme intoxication and to hold extreme intoxication at the time of sexual relations to vitiate consent in most cases.
The argument applies (with some adjustments) regardless of whether “consent” is defined in terms of the woman’s subjective willingness or in terms of some objective “token” of consent, like a requirement of “affirmative consent.” Even in an affirmative consent regime, the woman’s intoxication may be argued to vitiate her consent, impairing her mental faculties sufficiently that we should not take even a verbal “yes” as effective. In a “subjective consent” jurisdiction, an intoxicated woman’s subjective willingness to engage in sexual activity may be insufficient in cases of extreme intoxication.
Some may argue that the cases I term “significant intoxication” are a null set. If one thinks that most cases of significant intoxication do not displace a woman’s capacity to consent to sex and that capacity rather than a subjective mental process is all that matters, then one can posit that most cases of voluntary impairment do not involve what I have defined as “significant intoxication” anyway, and so there is no need for the tracing theory. Clearly, not everyone thinks that way, and for those people, some attention needs to be paid to the significance of pre-intoxication thought processes.
Even those who hypothesize that a fairly high degree of impairment should not preclude valid contemporaneous consent may reexamine those views as they think about the many ways sexual encounters unfold. For example, many may find unproblematic that a woman experienced in sex and drinking may have sex with a steady boyfriend after a long evening of mutual consumption and sexual activity, almost regardless of the woman’s ultimate level of impairment. Query whether intuitions change when some unpredictable turn occurs in the evening. Suppose the boyfriend passes out, his roommate comes home unexpectedly, the roommate propositions the woman, and the two have sex. Do we believe that the woman’s decision to have sex with the roommate must constitute valid consent so long as her decision to have sex with her boyfriend would have been valid given the same level of intoxication? If not, the tracing theory may be at work.
Before proceeding to the details of the argument, allow me to disclose a few assumptions I believe defensible but that I will not defend here. I assume the criminal law should be heavily influenced by morality—e.g., what we regard as the conditions for giving morally effective consent. However, practicalities may sometimes lead us to adopt rules that, while influenced by our moral reasoning, focus on morality only indirectly. I also assume the standard we adopt in our criminal law is significant, even if getting the law right is only one (and probably not the most important) problem in addressing problems surrounding sexual assault.
I assume the purpose of the criminal law is not to express solidarity with those who suffer harm, however real. We ought to structure our institutions and responses so as to comfort the afflicted, but among our methods, we should reject criminally convicting those who were not culpable in causing the harm. I assume culpability for purposes of the criminal law entails something more than mere tort negligence—at least gross negligence and probably at least Model Penal Code recklessness.
I also assume that when criminal law embodies vague standards, a jury will most likely apply that vague standard as if it were a negligence standard. In other words, vague standards in criminal statutes often risk punishing the nonculpable in the same way as would explicit recognition of a simple negligence standard addressing the social harm at issue. I defended this position at great length in my article on the American Law Institute’s (“ALI”) draft revisions of the Model Penal Code’s sexual assault provisions. The draft I discussed advocated for an “affirmative consent” provision. I argued that any provision flexible enough to jibe with social norms would be vague enough to constitute a covert negligence standard.
A couple more preliminaries. First, the examples in this essay are not meant to imply that sexual assault is only committed by men and can only be committed against women. The truth that sexual assault occurs in other ways makes the problem even more complicated than this paper assumes. Getting a grip on one frequent set of problems, however, may yield generalizable insights. Moreover, ignoring the historical paradigm may seem to slight insightful feminist critiques of sexual assault law. Second, the title of this piece signals my own concern about the empirical base for a proposal in this area. The allusion is to a work that had prominence well before the maturation of many university faculty and almost all university students. And even I know that most young women are not fans of single malt whisky—indeed, experimentation with alternative intoxicants often challenges the tracing theory of consent I discuss below. The recommendations that follow are subject to revision if further examination shows that they arise from misimpressions about how most women think about sex, the incidence of beneficial or neutral intoxicated sex, the harmfulness of harmful intoxicated sex, or how harmful intoxicated sex typically comes about.
This essay unfolds as follows. Part II briefly discusses two views on why we care about consent in morality and criminal law generally, to provide context for the discussion to follow. Part III canvasses arguments about why voluntary intoxication may affect consent differently from involuntary intoxication. Excessive regulation of intoxicated consent could impair women’s positive autonomy to decide to have intoxicated sex; regulation of either voluntarily or involuntarily intoxicated sex raises vagueness concerns. This Part concludes that the law would better balance the real interests at stake by rejecting a unitary approach to intoxicated sex and instead subdividing the domain into cases of normal and extreme intoxication. Part IV discusses the problems that occur when the criminal law embraces vague standards or precise rules that may not be effectively communicated to those governed by the rules. It argues, however, that these difficulties are muted when the criminal law adopts an intentionally underinclusive approach, so that convictions should remain within the realm of conduct that almost all adequately socialized people will regard as immoral. It then considers some proposed standards in light of these observations. Part V offers some tentative concluding thoughts on the relevance of criminal law theory to campus discipline policies.
II. On the Relevance of Consent
Thinking about how intoxication may vitiate consent requires addressing why consent matters. What is it about consent that performs the “moral magic” of translating “rape into lovemaking” and “trespass into a dinner party” If we can answer that question, we may gain some insight into when intoxication should vitiate consent from a moral standpoint. And that insight should inform our legal rules about consent, even if those legal rules may differ to some degree from our moral views.
One approach focuses on the role of consent in maximizing welfare. When each party to a sexual encounter wishes to engage in sexual relations, we have reason to believe that the transaction will leave both parties to the encounter better off. Of course, people considering sexual activities (or most other transactions) bring different goals to the decision. Leaving these decisions to the participants helps each pursue their preferred ends: erotic pleasure, communicating affection, securing a long-term partner, and the like. And of course, any ex ante assessment of the benefits of a sexual encounter may turn out to be incorrect. But on balance, we think people do a better job promoting their own welfare than we can do by specifying means for them, and we understand that people have different views of what is in their best interests. So consent may be thought to specify those conditions in which we should presume that transactions are welfare enhancing.
Another approach focuses solely on people as holders of rights to personal autonomy. Recognizing an autonomy interest need not be conceived as a strategy for enhancing welfare. Rather, autonomy may be viewed as an essential of personhood. Regardless of why these rights exist, consent may be viewed as specifying those conditions in which a person surrenders a right consistently with her autonomy.
On either of these views, some cases of intoxication may make consent seem morally problematic. Intoxication may alter emotion, desire, and cognition, resulting in a decision that a woman would not have made if sober. This is true even if cognition is not impaired. “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.” When cognition is affected, the decision may change because a woman will weigh the factors differently than when possessed of sober thought processes. From a welfare standpoint, intoxicated consent may be less suggestive that the encounter will be beneficial. From an autonomy standpoint, a woman’s intoxicated decisions may not square with what she regards as her true self.
None of this implies that regretted sex demonstrates a woman’s pre-sex lack of capacity. Even sober sex can be regretted. The preferences of a person will change over time, in light of, inter alia, reflection on experiences. Autonomy is impaired when we treat people as if their preferences will come, at some point, to jibe with our own, and we hence preclude them from making choices that we think they may come to regret later. Likewise, even when preferences remain constant, some risks end up not having been worth taking.
III. The Moral Distinction Between Voluntary and Involuntary Intoxication
The level of intoxication that triggers these welfare and autonomy concerns will be addressed later. For the time being, the reader should assume “significant intoxication” means whatever level of impairment would make contemporaneous consent problematic were we not to consider decisions made significantly in advance of the sexual activity in question. This Part begins by evaluating various arguments about why voluntary significant intoxication may be different from involuntary significant intoxication in terms of vitiating consent. It then runs through a variety of examples that may serve to help us clarify the significance of what I call here the “tracing theory.” It concludes by considering how subdividing the category of significantly intoxicated consent may facilitate better regulation than would be possible under a unitary rule for all cases involving intoxicated consent.
A. Assorted Approaches and Their Difficulties
A few arguments have been offered against taking account of voluntary intoxication in considering capacity to consent. Some do not require much discussion; others are more serious.
One argument focuses on the difficulty of drawing a bright-line between intoxication sufficient and insufficient to vitiate consent. The drafters of the Model Penal Code opined that, in cases of voluntary intoxication, an actor should not be required to ascertain whether his partner was too drunk to engage in sexual relations. As the commentary states, “[f]rom the actor’s perception, at least, this situation is exceedingly difficult to identify and perilously close to a common kind of social interaction.” While it may be difficult to tell exactly when a partner crosses the significant-intoxication line, it is another matter altogether to say that clear cases cannot be carved out for separate treatment—a point developed in more detail later.
Heidi Hurd raises what has been called the “consistency claim.” She reasons from our approach to voluntary intoxication as a defense to crime. We do not recognize a defense; indeed, we often view intoxication as a way toimpute a mental state that does not exist.
On pain of condescension, we should be loathe to suggest that the conditions of responsibility vary among actors, so that the drunken man who has sex with a woman he knows is not consenting is responsible for rape, while the drunken woman who invites sex is not sufficiently responsible to make such sex consensual.
The problem with Hurd’s argument is that our approach to voluntary intoxication as a defense does not provide guidance as to an actor’s responsibility. Rather, it is an example of a case in which we impose liability in the absence of responsibility because of the substantial social harm caused by intoxicated actors, harm that would be even more substantial if the behavior was not punished. No such costs attach to saying that some instances of intoxicated consent to sex do not count.
Alan Wertheimer makes the strongest argument against criminalizing sex with a voluntarily intoxicated partner—what I have called a “tracing theory.” However, that argument, on examination, could support imposing liability in the most severe cases of voluntary intoxication.
Wertheimer’s argument focuses on the two faces of autonomy. The standard feminist critique of consent standards focuses on negative autonomy—the right to avoid unwanted sexual relations. A grudging approach to intoxicated consent protects that interest. But it does so at the cost of positive autonomy—the interest in engaging in desired sexual relations. Some women desire to engage in sexual relationswhile intoxicated. They may find that they are less inhibited and enjoy sex more while intoxicated. Moreover, they desire to engage in intoxicated sex without giving the kind of clear, pre-intoxication consent that may be thought morally sufficient to authorize the activity. Indeed, some women may, at a subconscious level, prefer this strategy in part because it allows them to distance themselves from the decision to have sex. An exacting standard of sober, contemporaneous consent will impair the positive autonomy of these women.
In Wertheimer’s view, the fact that women frequently become intoxicated voluntarily and have sex shows that a nontrivial percentage of them must value intoxicated sex. He believes that the effects of alcohol on capacity are sufficiently well known that many of these women must realize when they decide to become intoxicated that they will make intoxicated decisions about sex that they might not have made if sober. Accordingly, Wertheimer tentatively concludes that we should regard the consent of the voluntarily intoxicated as effective. He recognizes that if the harm of unwanted sex is great enough, then that would be a reason to adopt a different rule even if the vast majority of intoxicated sex is desired ex ante. But in most cases, the consequences of sex are fleeting—unlike getting a tattoo, which Wertheimer believes may require contemporaneous sober consent. Wertheimer’s views are tentative because he recognizes that they turn on empirical assessment subject to rethinking in light of additional evidence.
Some may object that Wertheimer has underestimated the harm of unwanted sex. The psychic damage may in part derive from rhetoric about the propriety of intoxicated consent. Women who, on reflection, find a particular sexual encounter was unwise in light of their values, or who decide their values were mistaken, may understand that experience teaches through mistakes. But some will prefer not to admit that they may have made a mistake. Claiming that women who have intoxicated sex are inevitably victims may cause more women to view themselves as victims, itself a psychic cost. Putting this possibility to one side, however, the legal rule we choose may benefit from a more detailed examination of the cases in which a woman’s decision to drink does not entail her decision to delegate sexual decision-making to her future, intoxicated self.
B. Application and Limits of the Tracing Theory
The attractiveness of tracing theory varies according to a woman’s pre-intoxication thought processes. The inquiry is complicated by the need to consider decisions that may long precede a particular sexual encounter. Given the complexity and the difficulty of the inquiry, as well as the costs to autonomy of case-by-case inquiry, Wertheimer may be right to conclude that we need a default rule based on typical cases.
The tracing theory is strongest when applied to a conscious, fully informed, pre-intoxicated decision to engage in sex. For example, assume a couple that is in a long-term cohabiting relationship in which they are often sexually intimate. The woman, A, has a great deal of experience with both alcohol and sex. She and her partner, B, sometimes have had sex while significantly intoxicated. Right before she begins drinking, she focuses on where the night may lead. She hopes it will end in significantly intoxicated sex. And it does.
Pre-intoxication desire is not indispensable to the tracing theory. Assume A does not begin the night hoping that it will end in intoxicated sex, but she is aware before drinking that she may token consent to B while intoxicated, and she is indifferent as to whether that happens or not. Her intent is conditional on B’s desiring to have sex. Perhaps A is willing to please B even though she herself does not anticipate getting much erotic pleasure from an encounter. While some will dispute whether indifference of this kind should suffice as consent generally, much can be said in favor of permitting sex in such cases, and I will assume that the attitude of indifference suffices substantively. Note, however, that indifference is different from assuming a risk that A hopes will not occur. A may, pre-intoxication, be unwilling to have sex with B regardless of B’s later wishes but understand that drinking with B creates the risk that she will be too drunk to resist later. That pre-intoxication mental state differs from indifference, and tracing consent back to the pre-intoxication mental state would be incorrect.
Now suppose a more plausible scenario. A does not decide pre-intoxication that she is desirous of intoxicated sex with B or willing to defer to B’s desire about whether to have intoxicated sex. But she is willing to permit her later, intoxicated self to decide whether to engage in sex with B, or she is indifferent as to whether her later, intoxicated self does so. We constantly put our lives and safety in the hands of others—bus drivers, carnies. Surely we can decide to put our interests in the hands of our future selves, even if those hands will then be predictably less steady than now. While we know that our intoxicated selves may not proceed exactly as our sober selves would, we also see benefits in delegating decisions to our intoxicated selves—sometimes because they will act differently.
This process need not be fully conscious immediately preceding the relevant instance of intoxication. Throughout their lives, people make decisions about how they will decide. They develop approaches to recurring situations that relieve them of the need to engage in an elaborate cost-benefit assessment each time they face the situation. All of us do this in some situations—maybe we always put our car keys in the same place, having learned over time that doing so is preferable to considering on a case-by-case basis where the best place is to put the keys based on when we are next going out, whether we are too tired to go to the relevant spot, and the like. Many of us may think unwise a decision to delegate sexual decisions to our intoxicated or significantly intoxicated selves. But “we must be on guard against the tendency to attribute greater value to characteristics which are more likely to be found in twentieth-century intellectuals than in other groups or cultures.” For some, the unexamined life is not as boring.
Of course, if A and B were to break up, A could, pre-intoxication, decide to delegate sexual decisions to her intoxicated self while at a singles bar, a frat party, or the like. Given widespread knowledge of the effects of alcohol on decision-making, Wertheimer’s impression that many women make this decision seems justified, especially when one takes into account the habits one develops over time. Some women may impose pre-intoxication limits on how much they will drink; others will avoid places where they fear they will encounter people or situations that their intoxicated self will not be able to navigate.
Rather than adopting a flat rule about the effect of intoxication on consent, we could, in theory, apply the tracing theory by inquiring on a case-by-case basis whether a woman’s intoxicated consent was preceded by the pertinent kinds of pre-intoxication thought processes. As a matter of determining whether the woman gave morally effective consent, that approach would not be troubling. In articulating a legal standard about how potential sex partners must treat the woman’s token of intoxicated consent, a case-by-case approach confronts difficulties. To ask questions about the woman’s thought process is to offend her autonomy. If intoxicated consent is problematic, why should a potential sex partner be permitted to rely on representations made by the intoxicated woman—might not she make a nonresponsible decision to deceive? And if her subconscious goal, in part, was to create some psychological distance between her true self and her decision to have sex, the inquiry would frustrate that preference. These considerations may explain why the law usually speaks in terms of “capacity” when considering whether consent is valid.
If we did engage in a case-by-case assessment of whether a particular woman’s intoxicated consent was preceded by the relevant kinds of thought processes, we would find cases where it was not so. For example, a young person inexperienced in sex and intoxicants may not have considered the factors relevant to concluding that she made an autonomous decision to delegate sexual decision-making to her future, intoxicated self. If we reject the case-by-case approach, a flat rule distinguishing between extreme and significant intoxication holds some promise since cases of extreme intoxication may be those in which positive autonomy concerns are less prominent. Many empirical intuitions support this idea. Women may more often decide to become significantly intoxicated than to become extremely intoxicated—in part because of the especially debilitating consequences of extreme intoxication. If extreme intoxication is often not anticipated while in a responsible state, then often no responsible decision will have been made to defer sexual decision-making to a later, extremely intoxicated self. Relatedly, if women do not often become extremely intoxicated, then even women who decide to defer sexual decision-making to an extremely intoxicated self may often lack much information about how extreme intoxication will affect their future decisions, a relevant factor in determining whether the decision to defer was robustly autonomous. These considerations suggest that fewer women who have extremely intoxicated, rather than significantly intoxicated, sex can have their decision traced to an adequate pre-intoxication decision process. Accordingly, a rule forbidding extremely intoxicated sex will less impair positive autonomy than would a rule forbidding all intoxicated sex or even just significantly intoxicated sex. 
This argument does not make significant intoxication irrelevant to the consent determination. My focus has been on when a woman’s decision to have sex is suspect. But intoxication is also relevant to determining whether a woman has made that decision at all. For example, if a woman is confused about what is transpiring because of intoxication, we may conclude that she did not decide to engage in the sexual activity in question or that her apparent token of consent was ambiguous. Precluding all instances of intoxicated or significantly intoxicated consent can reduce the risk we will make mistaken assessments of whether the woman has decided to have intoxicated sex, but doing so may impair the positive autonomy of women who actually do decide to engage in intoxicated sex.
C. The Benefits of Dividing the Category of Intoxicated Sex
If what has been said previously is correct, then as Wertheimer posited, we need a rule about intoxicated consent that balances a woman’s desire for positive and negative autonomy and/or effects net welfare benefits, taking account the possibility that even if most women prefer having the ability to token significantly intoxicated consent, sufficient harm to the minority of women who suffer by tokening significantly intoxicated consent could justify a rule against it. Contrary to Wertheimer’s assumption, however, we need not view all intoxicated consent as created equal. Indeed, we may achieve a better resolution by distinguishing among degrees of intoxicated consent. Doing so will not eliminate all problems related to intoxicated sex, but it will address many of the serious ones.
If we assume that we must treat all intoxicated consent alike, Wertheimer may be correct that intoxicated consent should be considered valid out of respect for women’s positive autonomy and to promote the welfare gains anticipated by women desiring intoxicated sex. However, the previous discussion, examining cases in which intoxicated sex is more or less likely to be traceable to an adequate pre-intoxication decision, suggests that we may benefit by following the approach the criminal law takes to vulnerable victims, assaults on police officers, and felony murders by subdividing the category of intoxicated consent into cases of significant and extreme intoxication.
Assume that, among monolithic approaches, the balance between positive and negative autonomy would be better served by a rule permitting all intoxicated sex than a rule forbidding all intoxicated sex, as Wertheimer argues. It does not follow that this rule would strike a better balance than a rule permitting significantly intoxicated sex but forbidding extremely intoxicated sex. As suggested above, cases of extremely intoxicated sex are far less likely to involve an exercise of positive sexual autonomy. Treating cases of extremely intoxicated sex differently allows us to permit the positive autonomy benefits from significantly intoxicated sex without suffering the negative autonomy costs from extremely intoxicated sex.
This conclusion depends in part on whether women who token extremely intoxicated consent to sex are more likely to have experience with extreme intoxication than are most women who are experienced drinkers. Some may fit into that category. But in many cases, extreme intoxication likely sneaks up on women who did not reflect, pre-intoxication, on the likelihood of becoming extremely intoxicated. Some women with experience with one intoxicant will be surprised by the different effects of another. Some women will get drunker than they intended pre-intoxication because they fail to consider how alcohol will affect their post-intoxication decisions to continue drinking. And some women with experience getting extremely intoxicated will not have done so in situations that would have alerted them to how they may get isolated from others and find themselves in sexual situations. Indeed, one suspects that a higher percentage of women would, pre-intoxication, have different attitudes toward significantly intoxicated and extremely intoxicated sex, as extreme intoxication puts a woman at greater risk of making decisions at odds with her core values. Accordingly, inferring a woman’s robustly autonomous pre-intoxication decision to have extremely intoxicated sex is less plausible than inferring a woman’s robustly autonomous pre-intoxication decision to have significantly intoxicated sex.
All of this suggests that even if Wertheimer is right to be concerned with women’s positive autonomy to engage in intoxicated sex, the value of protecting that interest is greatest in cases short of extreme intoxication. Of course, how women reflect pre-intoxication on whether to delegate sexual decision-making to their significantly intoxicated selves is an empirical question, as is the harm that women suffer in cases in which they do not so choose. Instincts about this question will be influenced partly by opinions regarding the point at which we draw the significant-intoxication line for purposes of sexual decisions.
The challenge then becomes to determine whether we can draw a legally acceptable line between significantly intoxicated and extremely intoxicated consent—a line that avoids the problems of vagueness. Before turning to that question in the following section, a couple of points are worth making.
A rule upholding significantly intoxicated consent may fail to protect those without much experience or information about drinking or sex. These women are particularly likely not to engage in the kind of pre-intoxication thought process presupposed by the tracing theory. To some degree, these problems are reduced by the statutory rape laws that set an age of consent high enough that most persons above that age probably will be able to make sensible decisions about drinking and sex. But some will not. Education on these matters is critical—and not only if significantly intoxicated sex is usually permitted. While criminalizing significantly intoxicated sex may deter some of it, we cannot expect perfect deterrence. Punishing offenders does not make victims whole. We need to educate inexperienced women about the risks of drinking and sex early enough that they can avoid harm.
Even if the above arguments lead us to conclude that significantly intoxicated sex should be legal, the morality of the conduct is a different matter. For example, even if we do not punish a man who has sex with a partner above the age of consent who seems so unsophisticated as to not understand the consequences, we may criticize the decision. So too with a man who has sex with a significantly intoxicated woman in circumstances in which he is aware of a substantial risk that she has not made a pre-intoxication decision to defer sexual decision-making to her significantly intoxicated self. Even if we do not punish because of the problems of requiring partners to know everything about each other before engaging in sex, we may well criticize a man who knows that his significantly intoxicated partner had so little experience with and information about the relevant intoxicant as to have contemplated the sexual consequences of drinking. Educating men on why a woman’s assent may sometimes be morally problematic is as worthy an undertaking as educating women on the consequences of drinking and sex.
IV. Defining Extreme Voluntary Intoxication
Thus far, I have argued for the plausibility that many cases of significantly intoxicated sex should be considered legal because of the likelihood that the partner made a pre-intoxication decision to delegate sexual decision-making to her significantly intoxicated self and that the likelihood of that decision is reduced in cases of extreme intoxication. I have not tried to define “extreme” intoxication except by reference to why it is relevant—because it is the switching point at which negative autonomy gains from criminalizing the conduct more than offset the positive autonomy losses. Law-abiding people will have different conceptions of when that point is reached, posing a definitional problem in drafting an acceptable legal rule.
This Part begins by addressing some possible approaches to distinguishing significantly intoxicated sex from extremely intoxicated sex, examining several approaches suggested by the ALI draft amendments to the sexual assault provisions of the Model Penal Code and the drafters’ concerns about identifying an approach that escapes the problems of vagueness. It then considers when vagueness problems are most significant, arguing that greater vagueness is permissible when it occurs beyond the line that any law-abiding person would recognize as distinguishing moral from immoral conduct. Deploying specific rules to avoid vagueness concerns raises its own set of problems, but those problems are less important when the rules are significantly underinclusive in proscribing immoral conduct. This section then addresses some possible objections.
A. Possible Approaches, the ALI Draft, and the Vagueness Problem
How might we fashion the law to accomplish the goals discussed above—adequately balancing women’s interests in positive and negative autonomy and permitting decisions that will be, on balance, welfare enhancing. Like many jurisdictions, the 1962 Model Penal Code did not view voluntary intoxication as negating capacity to consent to sex. However, like many jurisdictions, the Code did explicitly criminalize some sex obtained through involuntary intoxication. Those provisions may provide guidance on how to handle involuntary intoxication.
The Code equated sex via surreptitious drugging with sex through threat of death. An actor was guilty of rape if he had sexual intercourse with a woman and had “substantially impaired her power to appraise or control her conduct by administering or employing without her knowledge drugs, intoxicants or other means for the purpose of preventing resistance.”
Similarly, the 2015 ALI draft imposes rape liability on an actor who engages in sexual penetration with a person who:
lacks substantial capacity to appraise or control his or her conduct because of drugs, alcohol, or other intoxicating or consciousness-altering substances that the actor administered or caused to be administered, without the knowledge of such other person, for the purpose of impairing such other person’s capacity to communicate, by words or actions, his or her refusal to engage in such act.
The central focus of the 1962 Code and 2015 ALI draft provisions is on whether the partner has “substantial capacity to appraise or control his or her conduct.” Even in the case of involuntary intoxication, the question was not whether the woman would have consented to sex but for her impairment. Instead, the provisions imply that a woman’s impaired decision to consent nevertheless may be a valid one. It is only when the impairment substantially impairs the woman’s cognitive or volitional abilities that resulting consent is invalid. This standard is similar to the Code’s standard for when involuntary intoxication is a defense to crime. It does not matter that you would not have stolen a wallet but for the fact that someone slipped you a mickey, just as it does not matter that you would not have stolen a wallet but for the fact that someone left it unattended. More than but-for causation is required to render the wrongdoer irresponsible.
If this standard is sound, it holds implications for voluntary intoxication. It is hard to see how a but-for test could be rejected for involuntary intoxication but accepted for voluntary intoxication. Even if a woman did not fully appreciate how she would be affected by alcohol, she would be no worse off than if unaware that she had become intoxicated. Note also that the involuntary intoxication standard focuses on capacity, not on what the woman actually thought about. Again, if capacity is the pertinent question in cases of involuntary intoxication, capacity ought to be the pertinent question in cases of voluntary intoxication. I am doubtful, as a moral matter, whether capacity should be the focus in either case, though capacity may be embraced as a useful legal proxy.
Even if we should not require greater capacity in voluntary than in involuntary-intoxication cases, does the standard for involuntary intoxication also set an appropriate standard for assessing consent in cases of voluntary intoxication? Here we confront the tracing theory described previously. Do women often voluntarily and substantially impair their own capacity to appraise or control their conduct at the time they token consent to sex, and if so, should not our legal standards accommodate this expression of positive autonomy? In other words, does the “substantial capacity” standard adequately capture the idea of extreme voluntary intoxication discussed above so that the same standard should be applied to both cases of voluntary and involuntary intoxication? Women may get drunk anticipating that they will have sex that they otherwise would not, but do they get drunk anticipating that they will end up lacking “substantial capacity” to control themselves? (Even if we applied the same standard to voluntary and involuntary intoxication, we may still punish involuntary-intoxication cases more seriously, of course.)
The ALI drafters, however, raise a concern other than striking a balance between protecting interests in positive and negative autonomy: the vagueness problem. They criticize both the Model Penal Code’s involuntary intoxication standard and other tests relating to voluntary and involuntary intoxication on the same grounds. The ALI’s April 2015 draft criticized the standards for attempting the “impossible task of drawing an identifiable line between intoxication that makes compliant behavior inauthentic and intoxication that does not.” The drafters thought the task had led to “the vagueness of applicable law in this area.” It could be tolerated in the cases of involuntary intoxication because the requirement of surreptitious administration “eliminates at one stroke the potential for overly broad liability; indeed, the . . . [surreptitious administration] requirement owes much of its support to its ability to keep the legal standard at safe distance from any slippery slope.”
Other approaches were likewise deemed problematic. For example, the commentary critiques a “typical formulation” that states, “rather unhelpfully,” that incapacitation means the partner “lacks the judgment to give a reasoned consent.” “Even more vacuously,” other statutes define incapacity as a condition rendering the victim “incapable of giving consent.” Statutes and case law like this “offer no coherent standard at all.” Efforts to provide more guidance fare no better. A California case defining incapacity in terms of whether the partner “would . . . have engaged in intercourse with [defendant] had she not been under the influence” would “transform many happy couples into serial sex offenders; a test of this sort in effect gives juries license to convict either party almost any time alcohol has mixed with sex.”
The same court said that incapacity would be shown if the victim was “unable to make a reasonable judgment as to the nature or harmfulness of the conduct,” a standard criticized as “permit[ting] convictions under a benchmark with little content.” The court did attempt to flesh out this standard. It said that “a poor judgment is [nonetheless] a reasonable judgment so long as the woman is able to weigh and understand the physical nature of the act, its moral character and its probable consequences.” But this elaboration too comes in for criticism: “by adding . . . that capacity requires understanding of the moral character and the consequences (perhaps including the emotional consequences) of intercourse[,] the court returned to a concept of alcohol-induced incapacity that is preposterously broad.”
As an alternative to “some difficult-to-define capacityto appraise or control his or her conduct,” the draft undertook a quest less “metaphysical and . . . quixotic”: to determine whether the partner’s voluntary intoxication was such as to render the partner “unable to express by words or actions his or her refusal to engage in” the sex act in question. The focus was on the inability to communicate and whether the defendant was at least reckless regarding that inability—i.e., consciously aware of a substantial and unjustified risk that the woman was unable to communicate. An actor would become aware of such a risk if a woman did not speak or otherwise communicate or spoke only incoherently in the period leading up to the sex act in question. Of course, an extended period of uncommunicative behavior would signal this risk while a shorter one may not.
While this “inability to communicate” standard largely addresses the drafters’ vagueness concern, the standard changed in the project’s September 2015 draft. Concerned that the prior formulation would miss too many problematic cases, the September draft revised the provision to focus on whether the partner “is passing in and out of consciousness or is in a state of mental torpor as a result of intoxication.” And in a subsequent draft, the standard was massaged in a less obvious way: a sentence was added to the commentary on consent (not the commentary on capacity to consent), seemingly incorporating by reference aspects of the 1962 Code’s general conditions for establishing consent.
The addition was surprising in light of the general structure of the reform project. If taken seriously, however, the comment seems to alter the standard for intoxicated consent in ways that are in considerable tension with the draft’s earlier criticisms of vague standards in this area. The 1962 Code’s consent provision—not central to its treatment of sexual assault cases—declares consent invalid if “given by a person who by reason of . . . intoxication is manifestly unable or known by the actor to be unable to make a reasonable judgment as to the nature or harmfulness of the conduct charged to constitute the offense.” Putting aside the mens rea language, the draft’s incorporation of which is unclear, the remaining standard’s focus—on whether the partner is “unable to make a reasonable judgment as to the nature or harmfulness of the conduct”—is identical to the language that, when employed by a California court, was deemed to be “a benchmark with little content.”
Whether future commentary on the intoxication provisions will address this inconsistency is a matter of speculation. Assuming the commentary was intended to change the intoxication standard, the reasons may relate to other changes in the draft. The insertion occurred shortly after the draft claimed to move from an “affirmative consent” to a “contextual consent” model. It may be that the drafters felt a greater need to draw the intoxication line closer to the line identifying morally questionable conduct in light of this change. The drafters argued their affirmative consent approach reduced the need to undertake the murky inquiry into how intoxication affected a person’s capacity to consent to sex: when a person had not given affirmative consent, “there is no need to determine whether they are ‘incapable of giving consent’ because, whatever their capacities, they clearly have not given consent.”
Regardless of whether one follows an affirmative consent model, a question exists as to whether a person has made a decision to allow any particular sex act—an essential of simple assent. On a subjective approach to consent, intoxication is relevant to whether other evidence indicates that in fact the partner was willing, as apparent acquiescence has evidentiary force in relation to an act only if the act was foreseen, and intoxication can impair the ability to figure out what is going on. On an affirmative consent approach, even an affirmative “yes” to a question about penetration is significant only if the partner understands the question, another matter affected by alcohol.
Apart from how intoxication may bear on assent, however, the capacity issue remains. What if the partner understood what she was agreeing to—and even was an active participant in it—but the conduct was so out of character that she would predictably regret it when sober? If this possibility does not exist or is not a problem, then the same should be true even when intoxication is induced surreptitiously. So the drafters should have had concerns about this issue even under the affirmative consent approach.
Vagueness problems in the capacity to consent area may be even greater than the ALI commentary implies. As noted earlier, by its nature, questions about capacity are scalar, while our decisions about legal consent must be binary—either the intoxicated consent is valid or not. For that reason alone, different people will draw the line between valid and invalid consent differently.
Add to that problem the possibility that we may draw the line differently for certain decisions about sex than about others. Do we think that the same capacity is necessary to consent to sex with a long-time partner as to consent for the first time to a heterosexual or homosexual encounter or to consent to sex for the first time with a particular person? Is the same capacity needed to consent to each particular sex act that may occur? Wertheimer convincingly argued that a different capacity should be required based on context—greater capacity to get medical treatment or a tattoo than to have sexual relations. If that is true, then we should not assume that sex itself is an indivisible category. And if that is true, then we will draw the intoxication line differently based on the precise sexual decisions we are analyzing. All of which is to say that reducing our moral intuitions to an acceptably clear legal standard is likely impossible.
One may object that the Model Penal Code’s definition of recklessness—central to its definition of criminality—is just as vague as the standards that may be embraced to distinguish authentic from inauthentic consent. Note, however, that while the definition includes some imprecise elements—like the question of whether a risk of harm is “substantial” or is “unjustifiable”—those terms are constrained by the requirement that criminal liability be found only when the risk is of a type that disregarding it “involves a gross deviation from the standard of conduct that a law-abiding person would observe.” Thus, the Code’s approach to recklessness actually supports the approach advocated here.
B. Why and When Vagueness Matters
Obviously, if we ask whether a woman’s consent after voluntary intoxication was valid, whether she lacked capacity to consent, or some similar “vacuous” formulation, we would invite arbitrary decisions by prosecutors, judges, and juries. The same could be said if we simply asked whether the sex partner was extremely intoxicated. Vagueness poses other problems too. A vague criminal standard risks punishing law-abiding persons who lack culpability, simply because a particular jury represents a particular collection of moral instincts common among law-abiding persons. We can attempt to avoid this problem by articulating a more precise rule to govern decisions, but the rule cannot create culpability among actors who are not aware of it, and skepticism is justified regarding how many rules we can expect any particular actor to know.
The latter problems—culpability problems with vague standards and clear-but-unknown rules—are most troubling when we try to draw our legal line close to the moral line. Traditional mistake-of-law doctrine is tolerable when applied to conduct nearly everyone recognizes as immoral. Likewise, even a vague standard or clear-but-unknown rule is tolerable when it will only reach conduct that nearly everyone recognizes as immoral. We can tolerate greater vagueness in a standard when it requires line-drawing well away from the contestable line between moral and immoral conduct—and clearly within the domain of the immoral. While the standard risks arbitrary enforcement, it does not raise the culpability concerns that arise when we attempt to come closer to the moral-immoral line. Regrettably, those who (in my opinion erroneously) view criminal punishment as a form of compensation to the victim will be tempted to come close to this line.
Accordingly, the question we should ask about a proposed voluntary intoxication approach is whether it safely avoids the contestable moral-immoral line, remaining well within the immoral category. We can construct such a rule, and we can expect the rule to do some good. But as with any underinclusive protection of social interests, it will need to be supplemented by something else—in this case, education about how people can avoid being victimized by conduct closer to the moral-immoral line. Such education is essential regardless of whether we attempt to make illegal everything that we may consider immoral, but it is even more important if we are to avoid the perils of vagueness that accompany an attempt to equate immorality with illegality.
When it comes to scalar notions like what capacity or actual thought processes are necessary morally to authorize certain conduct, people will inevitably disagree. For example, Joan McGregor has suggested that consent should be invalid whenever a woman is “drunk or high on drugs.” In another work, she mentioned the standard at which driving a vehicle is prohibited, an approach that implies a surprising view about the physical coordination and sensory processing needed for a safe and satisfying sexual encounter. On the other hand, though Shlomit Wallerstein would view voluntarily impaired consent as invalid, he “stress[es] that [he is] not talking about someone who is tipsy having had one or two glasses of wine, but rather of ‘a person who has had a lot to drink’ (for example, binge drinking)” or someone who is “really drunk.” As if a single standard isn’t complicated enough, Christine Chambers Goodman advocates a sliding-scale for consent to sex that varies depending on the degree of a woman’s intoxication. Seidman and Vickers claim that because “there is no bright line test for determining how much alcohol or drugs inhibits a person’s ability to consent, there must be a bottom line”; they advocate that “if alcohol is present, non-consent must be presumed unless a woman makes an explicit verbal statement she wishes to engage in sexual intimacy that includes penetration.” Like the scholars who have weighed in on the issue, we may well expect jurors to reflect a range of views regarding what constitutes problematic intoxication.
We can allow juries to identify social norms in a setting like this by simply giving them a general standard—like those found in the Model Penal Code’s involuntary intoxication provision. When does a woman lack “substantial capacity to appraise or control” her conduct? Just as different academics have after extended study, different defendants will perceive different lines and different jurors will too. Unless jurors are told specifically to convict only those whose moral perceptions grossly deviate from their own, as they would be under a gross negligence standard, or those who were aware they may be violating a legal or moral norm, as may be theoretically possible, they will likely construe “capacity” in terms of their own views of good conduct. The vague standard imperils an actor not unlike many other law-abiding citizens just because the actor ended up with a sex partner who objected, after sober reflection, to her intoxicated assent, and a jury with a different view of where the moral line should be drawn.
We can accomplish some good in regulating intoxicated sex without incurring this problem. In other words, we can regulate sex with a voluntarily intoxicated partner without incurring the perils of a generally murky affirmative consent requirement.
The ALI drafters’ first proposal regarding voluntary intoxication, focusing on the partner’s inability to communicate “by words or actions,” satisfies this criterion. People will disagree about whether a person’s failure to communicate implies that she was “unable” or just that she had decided not to do so. But if a person in close proximity has not communicated in any way for any significant period of time and circumstances suggest that the partner is impaired by intoxicants, few people would fail to perceive a substantial risk that the partner had reached a level of intoxication she did not anticipate while competent. Accordingly, wide consensus likely exists that a person should not take advantage of the partner’s condition. Likewise with that aspect of the drafters’ revision focusing on whether the partner was “passing in and out of consciousness,” so long as it is remembered that unconsciousness is different from being asleep.
Being “in a state of mental torpor,” on the other hand, may be sufficiently vague to become a conduit for jurors’ debatable conceptions of whether the partner lacked “substantial capacity to appraise or control” her conduct. The drafters provide the following explanation for the “torpor” standard:
Oxford English Dictionary defines “torpor” as “[a]bsence or suspension of motive power, activity or feeling” and Merriam-Webster defines it as “a state of mental and motor inactivity with partial or total insensibility” and “a state of lowered physiological activity . . . .” “Mental torpor” also corresponds to the level of intoxication that some refer to as “stupor,” at which a person loses significant response time, has trouble moving, vomits, and may lapse in and out of consciousness. These conditions are meant to capture an extreme point of intoxication at which consent becomes highly unlikely, and at which any passivity or non-responsiveness on the part of the other person should presumptively be deemed due to intoxicants, rather than a signal of consent. The focus of the inquiry, therefore, is not on the question whether the complainant had some difficult-to-define capacity to appraise or control his or her conduct; instead the inquiry is concerned solely with the question whether the degree of intoxication was so extreme as to effectivelypreclude the expression of unwillingness.
The stated goal of identifying an “extreme point of intoxication” is sound, not just because those are the cases in which an inference of simple assent is problematic, but also because, as argued above, the consequences of “extreme intoxication” are far less likely to have been foreseen pre-intoxication by the sex partner, and hence her positive autonomy interests are less likely to be impaired by a rule invalidating consent in these situations. The examples provided about how “torpor” may manifest certainly involve extreme intoxication—indeed, one could fashion a rule explicitly invalidating consent by a person whose intoxication is to a degree as to render her nonresponsive to stimuli or incapable of walking safely without assistance. “Torpor” may be construed, however, to sweep more broadly. Even sticking with the drafters’ definitions, an “absence . . . of . . . activity” or a “state of . . . motor inactivity with partial . . . insensibility” could permit reintroduction of an affirmative consent requirement—and even trump affirmative consent when indicated—whenever the partner had consumed enough alcohol to bother a jury’s sensibilities.
My goal is not to advocate in favor of any particular extreme-intoxication rule. Reasonable people will disagree on which rules would be narrow enough to avoid transforming an incapacity standard into the equivalent of a simple negligence standard for criminal liability, leaving law-abiding people at risk of being told after the fact that their norms about capacity differ from those of a jury. Instead, I hope to have illustrated the considerations that may inform a standard that seeks to deter violations of a woman’s negative autonomy while maximizing her positive autonomy and avoiding the other perils of vagueness that concerned the ALI drafters.
I illustrate the position graphically below, with the line between “significant intoxication” and “extreme intoxication” depicting the true “moral” line and the “legal standard” oval
Pre-intox. Significant Intox. Extreme Intox.
depicting how even vague or unknown legal regulation can avoid punishing law-abiding people who may disagree on where exactly to draw the moral line.
C. Some Objections
Some may object to the proposal that extremely intoxicated voluntary consent, but not significantly intoxicated voluntary consent, should be criminalized. Start first with objections to permitting significantly intoxicated consent to stand. If this rule underprotects women’s negative autonomy, one may ask why we should worry about the positive autonomy of women who desire to engage in intoxicated sex without clearly signaling that fact in advance of becoming intoxicated. After all, a “no-means-no” rule infringes on positive autonomy too—the desire of some women to engage in what may be called “coy sex” after purporting to signal unwillingness. I have previously supported the ALI draft’s recommendation of a no-means-no rule. One might question whether that position is consistent with my position on intoxicated consent.
The difference is empirical and is subject to dispute on those grounds. I suspect that a significant group of women who have sex while significantly intoxicated are comfortable with their decisions but would not want to signal their intentions clearly before becoming intoxicated. If that is correct, no other rule will protect that interest. Moreover, other means are available to minimize the impact such a rule will have on negative autonomy. Young women—well before college age—can be educated on the nature of sex and intoxication, reducing the number of cases in which women’s autonomy is infringed when significantly intoxicated sex occurs (and, one hopes, also the incidence of unwanted extremely intoxicated sex).
To protect the positive autonomy interest in coy sex, on the other hand, disables all women from having ready access to a clear signal to end a sexual encounter. When a man persists after a woman says “no,” the woman, often smaller physically, may well view that conduct as an implicit threat that the sex act will be consummated regardless of the woman’s unwillingness. Accordingly, the cost to negative autonomy seems high in this setting. And my own instinct is that the percentage of women who wish to have sex after saying “no” and without doing something later that revokes that “no” is small, diminishing the positive autonomy concern so long as the no-means-no rule permits the “no” to be revoked.
Now consider a positive autonomy objection to the extreme-intoxication rule. What is a woman to do who enjoys having sex while extremely intoxicated? Of course, we could articulate a special rule permitting extremely intoxicated people to have intercourse, so long as they signal their decision to do so pre-intoxication. To tweak the Jimmy Buffet song, she might ask her partner, pre-intoxication, “Why don’t we get extremely drunk and screw?” Even if explicitly authorizing such conduct is a good idea, it still would not protect a woman’s positive autonomy interest in engaging in extremely intoxicated sex without expressing agreement to do so in advance. My guess is that this preference is fairly rare, especially if the “extreme intoxication” standard is set at a very high level.
Finally, I return to a point raised earlier. One could object that this scheme is unduly complicated. Better to identify serious social harms—having sex with a person whose consent is inauthentic—and forbid consciously running a substantial and unjustified risk of bringing about those harms. I normally agree with that formula for criminalization. Here, however, the usual approach would present unusual problems. A sensitive person would always perceive some risk that a particular woman’s decision to have significantly intoxicated sex is inauthentic. Inquiries aimed at reducing that risk would themselves be insulting to a woman’s autonomy, inconsistent with many women’s preferences, and of dubious value, as they would require reliance on communications that may be inauthentic (because by hypothesis the woman is significantly intoxicated). It is no answer to avoid the admittedly clunky, rulified approach to criminal law defended here by simply asserting that women have sufficient capacity to consent contemporaneously to significantly intoxicated sex, or even to extremely intoxicated sex. If those positions are something other than ipse dixit, I suspect it is because of something resembling the analysis above.
V. From Criminal Law to Campus Policies
Let me offer some preliminary thoughts on how campus policies may justifiably depart from sound criminal law principles. The heightened procedural protections for criminal defendants and the reluctance to impose liability for simple negligence and under vague provisions or unknown-but-clear rules are based on the purposes of the criminal law and its effects. The criminal law condemns, and it deprives people of liberty. Campus policies are different. Their goal in part is to ensure a safe environment for students. Requiring the protections of the criminal process, including proof beyond a reasonable doubt, seems excessive before dealing with a threat to others. A school may also prefer the negative sexual autonomy of its students over their positive sexual autonomy, regardless of how the criminal law strikes the balance in governing a populace less skewed toward the inexperienced. To ensure that negative autonomy is more fully protected, schools may prefer the kind of vague standards that are inevitable if the goal is to deter (or discipline in) all cases in which negative autonomy is violated.
Schools should recognize that, while they do not send people to prison when they find violations of sexual assault policies, they do inflict a serious harm. Presumably, expelled students have trouble continuing their educations. Those disciplined in other ways may still have the discipline disclosed if they apply for graduate schools or certain jobs. And to the extent that schools wish to support those whose negative autonomy has been impaired, means other than “conviction” are available to do so.
While a case may be made in favor of vaguer standards of capacity in school disciplinary codes than those that would be optimal in a criminal code, in many cases students who are merely negligent in discerning the norms of their community will not pose dangers. The filing of charges in itself has an educational effect. Remedies short of “conviction” may be appropriate in such cases to ensure that the student understands what is expected of him. Of course, this “diversionary” approach will not be justified for offenders who engaged in sex with a woman who would be regarded as “extremely intoxicated” under a standard or clear rule that would satisfy criminal norms. Nor would it be justified for repeat offenders or others who can be shown to have violated a school’s norms with full knowledge of what they are, though we must understand that if these norms are sufficiently out of touch with the morality of the student body, insisting on these norms injects an element of arbitrariness into whom we declare dangerous.
Whether schemes such as this would have evolved naturally, we may never know. The federal government’s intrusion into this area has encouraged schools to embrace unfair disciplinary systems. Among the worst features of which I am aware are requiring accused students to face charges without the meaningful assistance of counsel and drawing hearing panels from volunteers who are then trained by university bureaucrats. Perhaps the recent wave of litigation against schools by those accused of sexual misconduct will result in beneficial rethinking of current practices.
The frequency of sexual assaults involving intoxicated victims demonstrates the importance of careful attention to the question of how intoxication affects consent. But because of the frequency of intoxication and desired sex, no simple solution is available. Maximizing the negative autonomy interest in avoiding unwanted sex comes at the cost of intruding on the positive autonomy interest in having desired intoxicated sex, and vice versa. Many jurisdictions have resolved this dilemma by treating voluntary intoxication as usually irrelevant in addressing issues of consent. This article has argued that the law can better reconcile these positive and negative autonomy interests without doing violence to important norms against vague criminal prohibitions. Those norms are not fully applicable to campus disciplinary proceedings, but they supply useful points of reference even in that setting.
* © 2016, Kevin Cole. Professor of Law, University of San Diego School of Law. Thanks to Larry Alexander and Don Dripps for their helpful comments on an earlier draft, and to Asal Alipanah and Taylor Leuschen for their excellent research assistance.
 Cf. Alan Wertheimer, Consent to Sexual Relations 233 (2003) (also excluding consideration of those who are semiconscious at the time they token consent).
 Model Penal Code § 213.1(1)(b) (1962).
 Kimberly Kessler Ferzan & Peter Westen, How to Think (Like a Lawyer) About Rape, Criminal Law and Philosophy (forthcoming) (manuscript at 10), available at https://perma.cc/968C-GF2J.
 For a different approach, seemingly focused only on the presence of assent, see Michal Buchhandler-Raphael, The Conundrum of Voluntary Intoxication and Sex, Brook. L. Rev. (forthcoming) (manuscript at 58), available at https://perma.cc/4ZPD-9FWN) (drunkenness should not obviate express consent).
 The distinction between these ways in which intoxication can affect consent is similar to the distinction drawn when the law allows intoxication to affect a defendant’s criminal liability. In cases of voluntary intoxication, when the law recognizes the effect of intoxication, it focuses on how it impacts cognition, not behavioral control. Intoxication’s effect on volition is considered only when the intoxication is involuntary. See Donald A. Dripps, Rehabilitating Bentham’s Theory of Excuses, 42 Tex. Tech L. Rev. 383, 402–06 (2009).
 See Ferzan & Westen, supra note 3, at 40–41:
The most challenging capacity-to-deliberate cases are ones in which (i) incapacity is scalar, that is, a matter of degree along a spectrum, rather than binary; (ii) the incapacity is difficult to measure ex ante; (iii) society perceives some value in sexual intercourse along the spectrum; and (iv) risks of harm are nevertheless genuine. A good example is sexual intercourse with a person who has intentionally ingested alcohol. Intoxication is scalar, ranging from a mild buzz to complete unconsciousness.
 Wertheimer, supra note 1, at 234.
 Apart from concerns with the authenticity of consent, the problem of determining assent does not completely disappear under an affirmative consent standard. Presumably, most advocates of such standards would not want a woman’s “yes” in response to a question she did not understand to constitute consent, so long as the woman’s partner understood the woman’s confusion. Intoxication is relevant to whether people understand the world around them.
 Alternatively, different intuitions about this example may reflect the view that a different capacity is required to choose a partner than to choose the particular acts to engage in with a partner identified pre-intoxication. See infra text accompanying note 75.
 I elaborate on this view in a forthcoming piece, Kevin Cole, Mushy Morality and Exacting Criminal Theory (forthcoming) [hereinafter Cole, Mushy Morality].
 See, e.g., Allison C. Nichols, Out of the Haze: A Clearer Path for Prosecution of Alcohol-Facilitated Sexual Assault, 71 N.Y.U. Ann. Surv. Am. L. 213, 263–65 (2015) (recommending changes in investigatory and prosecutorial practices).
 Kevin Cole, Better Sex Through Criminal Law: Proxy Crimes, Covert Negligence, and Other Difficulties of “Affirmative Consent” in the ALI’s Draft Sexual Assault Provisions, 55 San Diego L. Rev. 506, 507 (2016), available at https://perma.cc/C2LU-VNYH) [hereinafter Cole, Better Sex].
 The drafters subsequently modified their standard to some degree but retained the objectionable objective (negligence) qualities. In May 2016, however, the ALI membership adopted a motion after floor debate to eschew the drafters’ objective definition of consent. This important project continues to work its way through the ALI process, and in this piece I discuss how the draft addresses intoxication—a matter that has not yet been discussed on the ALI floor.
 See Helen Gurley Brown, Sex and the Single Girl (1962). The book inspired a hit movie by the same name released by Warner Brothers in 1964.
 It is worth noting, however, that which way the empirical investigation may push is unclear. A proposal to invalidate consent by the extremely intoxicated imposes increasing costs to positive autonomy if evidence shows that a growing number of women intentionally become “black-out” drunk with the plan of hooking up.
 See Heidi Hurd, The Moral Magic of Consent, 2 Legal Theory 121, 123–24 (1996); cf. Michelle Madden Dempsey, Victimless Conduct and the Volenti Maxim: How Consent Works, 7 Crim. Law & Phil. 11, 12 (2013) (consent not the whole picture; “[t]urning a trespass into a dinner party requires, at least, dinner—while turning rape into lovemaking requires, at least, love.”).
 See, e.g. , Richard A. Posner, Economic Analysis of Law 10 (4th ed. 1992) (“The third basic principle of economics is that resources tend to gravitate toward their most valuable uses if voluntary exchange—a market—is permitted.”).
 See, e.g., Gerald Dworkin, The Theory and Practice of Autonomy, at 3 (1988) (canvassing use of “autonomy” as “equivalent of” liberty, self-rule, sovereignty, freedom of the will, dignity, integrity, individuality, independence, responsibility, and self-knowledge).
 Mark Kelman, Thinking About Sexual Consent, 58 Stan. L. Rev. 935, 977 (2005).
 But see Kelman, supra note 19, at 976 (“autonomy theorists have nothing helpful to say about inconsistency of the will.”).
 See David Bowie, Young Americans, on Young Americans (RCA Records 1975) (“He kissed her then and there/She took his ring, took his babies/It took him minutes, took her nowhere.”).
 For example, the idea that a drunk woman runs the risk of intoxicated consent seems no more compelling than the argument that someone who walks about the city in the evening runs the risk of being mugged.
 Model Penal Code § 213.1 cmt. 5(a) (1962).
 Id. § 213.1 cmt. 5(a).
 See Wertheimer, supra note 1, at 233.
 Model Penal Code § 2.08(2) (1962) (“When recklessness establishes an element of the offense, if the actor, due to self-induced intoxication, is unaware of a risk of which he would have been aware had he been sober, such unawareness is immaterial.”).
 Hurd, supra note 16, at 141. Interestingly, the American Law Institute draft would not apply the Model Penal Code’s usual intoxication rules so as to impute recklessness to an actor regarding facts that would have been known if the actor has not been voluntarily intoxicated. Model Penal Code § 213.3(2)(c) (Discussion Draft No. 2, Apr. 28, 2015) [hereinafter April 2015 Draft]. However, this change would not avoid Hurd’s argument. The draft does not displace the Model Penal Code’s usual position that an actor’s voluntary intoxication does not create a defense based on the actor’s lack of capacity to control his conduct; it only addresses the cognitive aspect of the usual rule.
 Cf. Larry Alexander, The Moral Magic of Consent (II), 2 Legal Theory 165, 172 n.11 (1996) (objecting more generally to Hurd’s reliance on standards for excuses). Wertheimer makes a similar point, as well as some others of which I am less sure. See Wertheimer, supra note 1, at 245. Hurd herself may now be open to the idea that the law regarding intoxication as a defense does not hew to an acceptable line of moral responsibility. See Michael S. Moore & Heidi Hurd, Punishing the Awkward, the Stupid, the Weak, and the Selfish: The Culpability of Negligence, 5 Crim. L. & Phil. 147, 181–82 (2011) (rejecting tracing strategy as justification for current rules on intoxication as a defense).
 Wertheimer, supra note 1, at 251.
 Id. at 250–51.
 Id. at 251.
 Id. at 251–52.
 Id. at 252.
 Id. at 249.
 Id. at 256. For more recent evidence indicating the prevalence of such pre-intoxication planning, see Lori E. Shaw, Title IX, Sexual Assault, and the Issue of Effective Consent: Blurred Lines—When Should “Yes” Mean “No”?, 91 Ind. L.J. 1363, 1385–89 (2016) (canvassing studies).
 See Donald Dripps, For a Negative, Normative Model of Consent, with a Comment on Preference-Skepticism, 2 Legal Theory 113, 120 (1996):
[P]reference-skeptics need to account for the possibility that even wrong choices contribute to the quality of an individual’s life by making the individual a better person. If this possibility is ever credible, it is credible in the context of sexual choices. Mistakes about sex teach difficult but valuable lessons about deferring gratification, about empathy for others, about individual vulnerability, and about identity itself. When society refuses to accept consent as a justification for sexual behavior, it short-circuits the learning process for the sake of preventing undesirable conduct.
 Dworkin, supra note 18, at 17.
 Cf. It’s a Wonderful Life (Liberty Films 1946), available at https://perma.cc/6NZM-HKWQ (“Why don’t you kiss her instead of talking her to death? . . . Ah, youth is wasted on the wrong people!”).
 A different result may be justified in a case in which a woman tells a man before going out for drinks that she does not want to have sex with him that evening. In such a case, a well-fashioned “no-means-no” rule may intervene. But if the woman, after drinking, rescinds the “no,” we face the question of whether the rescinding (or more precisely, the decision to rescind) took place prior to or after arriving at the serious-intoxication line.
 The argument bears some similarity to the discussions of precommitment strategies. A goal of those discussions was to articulate a non-paternalist and non-perfectionist justification for legislation—like drug prohibition. The argument is that rational people may prefer to preclude themselves from making later decisions (to consume drugs, for example) under circumstances in which their deliberations may be impaired by temptation and the like. The kind of precommitment discussed here—the decision to delegate decision to a future self whose impairment is anticipated—is somewhat different, in that honoring it requires making the future conduct legal rather than illegal. Of course, a precommitment argument could be marshalled for why consent should be vitiated by intoxication that impairs but does not preclude rationality, if sex while intoxicated deviates from what people desire when fully rational. For a review of the literature and a discussion of the difficulties of preferring prohibition to freedom in these settings without slipping into paternalism or perfectionism, see Donald A. Dripps, Precommitment, Prohibition, and the Problem of Dissent, 22 J. Legal Stud. 255, 255–56 (1993).
 Wertheimer himself explicitly excepts “semi-conscious” women from his analysis, see Wertheimer, supra note 1, at 233, but does not explain what he means. So he too arguably would subdivide the category of intoxicated consent, though not necessarily in the way suggested here.
 The analysis tracks arguments about how we can deter crime more efficiently by dividing a social harm in ways that identify potential wrongdoers with different incentives to commit a crime. For example, we may impose a sentencing enhancement on those who pick out vulnerable victims (since they are easier to victimize) or assault police officers (because of the special incentives that criminals have to resist arrest). See Kevin Cole, Killings During Crime: Toward a Discriminating Theory of Strict Criminal Liability, 28 Am. Crim. L. Rev. 73, 114–15 (1990) (making a similar argument why strict liability for homicide could be rejected except for those engaged in robbery and the like).
 Model Penal Code § 213.1 cmt. 5(a) (1962).
 See, e.g., Nichols, supra note 11, at 221 (“Today, almost all jurisdictions explicitly recognize a situation where the defendant administered a drug or an intoxicant to a victim without her knowledge or consent as meeting the elements of the state’s rape or sexual assault statute.”) (citing Patricia J. Falk, Rape by Drugs: A Statutory Overview and Proposals for Reform, 44 Ariz. L. Rev. 131, 173 (2002)).
 Model Penal Code § 213.1(1)(b).
 April 2015 Draft, supra note 27, § 213.3(1)(c).
 See Model Penal Code § 2.08(4) (using similar language in setting standard for when involuntary or pathological intoxication establishes a true defense to a criminal charge (as opposed to merely negating mens rea)).
 While I disagree with Hurd’s position that the legal rules on intoxication as a defense should lead us to adopt the same rules on the issues of capacity to consent, see supra text accompanying note 22, I agree with her view that how we assess moral responsibility for criminal acts is germane to how we assess moral responsibility for the acts or intentions that arguably constitute consent. See Hurd, supra note 16, at 146. As for responsibility for criminal acts, I incline toward Douglas Husak’s view that the important question focuses on a defendant’s subjective thought processes. See Douglas Husak, Ignorance of Law: A Philosophical Inquiry 136 (2016): “[R]eason-responsiveness should be construed internally or subjectively in order to justify holding wrongdoers fully blameworthy for their conduct . . . [I]t is hard to see why rational persons should respond to reasons of which they are unaware . . . . The person who is reckless . . . deserves some amount of blame for her wrongdoing, but less than that of the person who fully appreciates his conduct is wrongful.”
See also Moore & Hurd, supra note 28, at 150, 179–80 (after extensive investigation, rejecting argument that capacity to have adverted to risk is a proper grounds of criminal responsibility and rejecting tracing arguments in situations in which the prior decision was not itself sufficiently culpable from a subjective perspective).
For an interesting example of an instance in which we arguably examine a person’s actual thought processes, rather than their capacity to engage in certain thought processes, to determine whether consent exists, see Case Study, Doctor Knows Best? Tubal Ligation in Young, Childless Women, Hastings Center Report 9–10 (Sept. 21, 2016) (canvassing subjects a doctor should address before holding a woman’s consent valid).
I address some of the justifications for enacting criminal laws that deviate from moral norms in Cole, Mushy Morality, supra note 10.
 If we treated cases of voluntary and involuntary intoxication identically, we would avoid the inevitable problems of sorting out which cases belong in which categories. Cf. Model Penal Code § 2.08(5)(b) (intoxication is self-induced if actor “ought to know” substance’s “tendency . . . to cause intoxication.”).
 Kevin Cole, Intoxication and the Sources for Defining “Consent” Under the ALI’s Draft Sexual Assault Provisions, CrimProf Blog (May 9, 2016), https://perma.cc/8MFA-K4Z4 [hereinafter Cole, Defining “Consent”].
 April 2015 Draft, supra note 27, at 65.
 April 2015 Draft, supra note 27, § 213.3(2)(c) & cmt., at 64–65.
 Id. at 65.
 Id. at 64.
 People v. Giardino, 98 Cal. Rptr. 2d 315, 321 (Cal. App. 4th Dist. 2000).
 People v. Smith, Cal. Rptr. 3d (Cal. App. 3d Dist. 2010).
 April 2015 Draft, supra note 27, at 65.
 Id. at 66.
 Id. at 67.
 Id. § 213.3(2)(c).
 Model Penal Code: Sexual Assault and Related Offenses § 213.3(2)(c) (Preliminary Draft No. 5, Sept. 8, 2015) [hereinafter Sept. 2015 Draft].
 The sentence reads, “[o]f course, all circumstances remain relevant, as specified in subsection (3)(c), in making the contextual assessment of whether the person’s behavior both communicates willingness and qualifies as legally effective consent.” A footnote appended to that sentence reads, “[s]ee Model Penal Code § 2.11(3) (1962) (specifying conditions in which assent does not constitute legally effective consent).”
 See Cole, Defining “Consent,” supra note 54.
 Model Penal Code § 2.11(3)(b).
 April 2015 Draft, supra note 27, at 64–65.
 For a skeptical view of the significance of the purported change, see Kevin Cole, Backpedalling in Place: The ALI’s Move from “Affirmative” to “Contextual Consent”, available at https://perma.cc/JQ9N-LFJR and Kevin Cole, Like Snow to the Eskimos and Trump to the Republican Party: The ALI’s Many Words for, and Shifting Pronouncements About, “Affirmative Consent”, 1, available at https://perma.cc/HMS7-RARV.
 April 2015 Draft, supra note 27, at 66 (footnote omitted; emphasis in original).
 However, in some cases, a person may consent to acts that she did not fully anticipate. See Cole, Better Sex, supra note 12, at 540–46.
 See Buchhandler-Raphael, supra note 4, at 58.
 Cf. Sharon Cowan, The Trouble with Drink: Intoxication, (In)capacity, and the Evaporation of Consent to Sex, 41 Akron L. Rev. 899, 909 (“the presence and validity of consent is also related to the question of the relationship between the parties—the issues may be very different in relation to a couple who have been intimate over a long period of time, as opposed to two people who have just met or barely know each other. In that sense any rigid rule that a certain level of intoxication would cancel out consent, contravenes principles of autonomy.”).
 Some of the troubling cases about intoxicated consent involve situations in which one may imagine that the woman’s extremely intoxicated consent involved a sex partner she may not have imagined pre-intoxication. See, e.g., Shlomit Wallerstein, “A Drunken Consent Is Still Consent”—Or is It? A Critical Analysis of the Law on a Drunken Consent to Sex Following Bree, 73 J. Crim. L. 318, 322 n.12 (2009) (reporting on case in which “a security guard . . . was asked to escort the complainant to her dormitory because she was very drunk and her friends feared she would get lost on the way. They had sex in the corridor of her dormitory.”).
 See Wertheimer, supra note 1, at 249; Kelman, supra note 19, at 968 n.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.
 Model Penal Code § 2.02(2)(c).
 Concerns about vague criminal provisions also support recent proposals to incorporate a mistake-of-law defense in federal criminal law.
 Some would say that the rule cannot create culpability even if it is known. See, e.g., Michael S. Moore, Legal Moralism Revisited, A paper presented to the Conference on Legal Moralism, Institute for Law and Philosophy, University of San Diego, San Diego, California, at 5 (May 20–21, 2016) (copy on file with author) (“I am one of those ‘new anarchists’ who deny that law qua law obligates citizens, even prima facie.”) (citing Michael Smith, Heidi Hurd, and Joseph Raz as fellow travelers).
 I discuss these issues in the context of a no-means-no proposal in Cole, Better Sex, supra note 12, at 525–26, 553, 556.
 One may defend current campus discipline codes as educational, but the oft-overlooked distinction should be drawn between education and indoctrination. One would hope that women informed as to data on intoxication and sex that other women regretted would be better positioned to make their own decisions about how to conduct themselves, and that men so educated would be better equipped to avoid causing harm to others. But “[t]hose who maintain a standard of conduct by threat produce conformists when they succeed, and deviants when they fail.” Dripps, supra note 38, at 119.
 Joan McGregor, Force, Consent, and the Reasonable Woman, In Harm’s Way: Essays in Honor of Joel Feinberg 244–45 (Jules L. Coleman & Allen Buchanan, eds., 1994).
 Joan McGregor, Is It Rape? On Acquaintance Rape and Taking Women’s Consent Seriously 150 (2005).
 Wallerstein, supra note 78, at 332.
 Christine Chambers Goodman, Protecting the Party Girl: A New Approach for Evaluating Intoxicated Consent, 2009 BYU L. Rev. 57, 94, 97 (2009). Goodman’s concern seems primarily to be with the question of whether the woman actually assented rather than with whether the assent was authentic.
 Ilene Seidman & Susan Vickers, The Second Wave: An Agenda for the Next 30 Years of Rape Law Reform, 38 Suffolk U. L. Rev. 467, 486 (2005).
 One student author argues that consent should be vitiated when “the complainant’s ability to affirmatively communicate willingness or unwillingness to engage in the act is hindered by reason of intoxication with alcohol or other substances substantially enough to cause observable physical weakening or impaired verbal ability.” Karen M. Kramer, Note, Rule by Myth: The Social and Legal Dynamics Governing Alcohol-Related Acquaintance Rapes, 47 Stan. L. Rev. 115, 152 (1994). Even modest intoxication would appear to be covered by the standard. Another student author recommends statutory language addressing a person “who is mentally incapacitated because of the influence of an intoxicant, such that the person is unable to meaningfully understand the nature and consequences of the act.” Nichols, supra note 11, at 252.
Without addressing the vagueness concern, Cowan appears to prefer that incapacity be further defined but simply in terms of “extreme drunkenness.” Cowan, supra note 77, at 921 (“Parliament should introduce legislative reforms that would include consent given in circumstances of extreme drunkenness as one of the situations where consent is not present”). She occasionally focuses on specific physical manifestations of such intoxication, like vomiting, inability to speak or move, passing in and out of consciousness, and memory blackouts. See, e.g., id. at 917. But she does not appear to recommend defining incapacity in those terms; she supports the creation of a rebuttable presumption of nonconsent that can be overcome by “evidence of consent or evidence as to a belief in consent that is reasonable in all the circumstances”; id. at 909 n.45 (suggesting that consent is not inconsistent with the physical symptoms she cites).
Patricia Falk advocates “the use of descriptive words such as informed, knowing, or reasoned, in characterizing the nature of the consent required in sexual assault cases.” Patricia J. Falk, Rape by Drugs: A Statutory Overview and Proposals for Reform, 44 Ariz. L. Rev. 131, 202 (2002).
Andrew Ashworth recommends instructing juries about capacity by focusing their attention on general principles of awareness, understanding, and ability. See Andrew Ashworth, Comment, Rape: Consent—Intoxication, 2007 Crim L. Rev. 900, 903.
 Cf. Samuel W. Buell, Culpability and Modern Crime, 103 Geo. L.J. 547, 595 (2015) (“The book on criminal law does not include a chapter about the problem of mens rea toward social norms.”).
 For a discussion of the problems with affirmative-consent standards, see Cole, Better Sex, supra note 12, § II.
 April 2015 Draft, supra note 27, § 213.3(1)(a), (b), (c), (2)(c).
 The draft itself recognizes the distinction. See id. § 213.3(1)(a) (criminalizing penetration of a person who “is sleeping, unconscious, or physically unable to communicate”) ; see also Unconsciousness—First Aid, MedlinePlus, U.S. National Library of Medicine (last visited Nov. 6, 2016), available at https://perma.cc/8TE4-34CQ (“Being asleep is not the same as being unconscious. A sleeping person will respond to loud noises or gentle shaking. An unconscious person will not.”). Sleep is nevertheless relevant to issues of consent. A person just awakened, for example, may be more prone to confusion, and hence a conclusion that the person assented to sexual contact shortly after awakening may require different evidence, regardless of whether the person even expresses a verbal “yes.”
 Sept. 2015 Draft, supra note 68, at 85.
 Other dictionaries also suggest that “torpor” may be viewed as consistent with moderate intoxication or simple languor. See Dictionary.com, https://perma.cc/2E6W-Y5FG (last visited Nov. 6, 2016) (defining torpor as: “sluggish inactivity or inertia”; “lethargic indifference; apathy”); The Free Dictionary, https://perma.cc/Z5QB-QEVB (last visited Nov. 6, 2016) (defining torpor as: “[a] state of mental or physical inactivity or insensibility”; “Lethargy; apathy”).
 A standard recommended by Lori Shaw in the Title IX setting may prove instructive, even though largely motivated by concerns against arbitrary enforcement and notwithstanding the author’s apparent recognition that students may not understand the detailed rules. See Shaw, supra note 37, at 1395 (“Obviously, students do not pull out the code of conduct for a quick review before hooking up, but what our standards say and omit to say impacts the culture of every campus.”). Shaw proposes the following provisions regarding when voluntary intoxication vitiates consent:
(5) The person making contact knows or reasonably should know that the other person is:
a) unconscious or drifting in and out of consciousness;
. . .
c) physically helpless (i.e., physically unable to communicate, effectively resist, sit, stand, or walk unassisted, and/or leave)
. . .
e) so disoriented by the use of alcohol and/or other controlled or intoxicating substances that he or she cannot understand the nature or consequences of the sexual contact.
Shaw’s definitions include:
(1) Disorientation exists when a person exhibits significant confusion or lack of awareness about what is happening around him or her and how it relates to him or her. A person is disoriented if he or she is:
a) incoherent (i.e., unable to converse clearly and logically); or
b) noticeably confused about his or her name, address, or current location, or the date and time.
(2) Lack of understanding of the nature or consequences of sexual contact exists when a person is:
a) unaware the contact is taking place or is sexual in nature;
b) unaware of who is making the contact at the time it is made . . .
See Shaw, supra note 37, at 1421–22.
 See Cole, Better Sex, supra note 12, at 524–25, 556, 575.
 Cf. Wertheimer, supra note 1, at 234 (quoting the Jimmy Buffett song and discussing a hypothetical involving explicit agreement between partners to have sex while one is under influence of date rape drug); April 2015 Draft, supra note 27, § 213.9(5), (7)(a) (affirmative defense for spouses and “intimate partners” to charge of having sex with an unconscious or sleeping person if, “in light of the specific facts and circumstances of that relationship and the context surrounding the disputed act, the actor honestly and reasonably believed that the act was welcome”; “imtimate partners” limited to co-habiting relationships ).
 For an excellent, extended treatment of the campus situation and the government pressures that gave rise to it, see Jacob Gerson & Jeannie Suk, The Sex Bureaucracy, 104 Cal. L. Rev. 881, 883–84 (2016).
 If a school’s standards deviate from the views of its students, we can expect those standards to lack much moral sway with students and to be violated consensually on many occasions. Students punished under such standards, then, would sometimes not differ from a large number of students whose presence on campus is never regarded as causing a problem. They would simply be the unlucky ones caught up in the disciplinary process.
 Gerson & Suk, supra note 100, at 931–32.
 See, e.g., Office of the Provost, University of Pennsylvania, Student Disciplinary Procedures for Resolving Complaints of Sexual Assault, Sexual Violence, Relationship Violence and Stalking § II(G)(1), available at https://perma.cc/8GGN-E8VR.