Note for readers: I am most grateful for the opportunity to present my work to the Oxford
Criminal Law Discussion Group. I’m just at the beginning of what I expect will be a multi-year
book project on the sexual offenses. The draft I’m giving you is largely a roadmap of issues I
hope to address in the work: I’m painfully aware of its inchoateness. I would welcome all
comments, especially about how to approach questions of autonomy, consent, offense, and
perversion -- key concepts that are dealt with only briefly in this draft. I would also welcome
suggestions for further reading.
Stuart Green
DRAFT/October 17, 2013
What are the Sexual Offenses?
Stuart P. Green*
Our law criminalizes a broad array of presumptively sexual conduct. Among the offenses
that do this (or did until recently) are rape, sexual assault, coercion, human sex trafficking,
female genital mutilation, forced marriage, sexual humiliation, domestic violence, voyeurism,
public nudity and public indecency, obscenity, sexual transmission of disease, selling and buying
sexual services (prostitution), pimping and pandering, statutory rape and child molestation, abuse
of position of trust, child grooming, creating and possessing child pornography, obscenity,
failure to register as a sex offender, fornication, sodomy, adultery, assault by sadomasochism,
incest, bigamy, polygamy, miscegenation, bestiality, necrophilia, and sale of sex toys.
While many of these offenses, taken separately, have generated a significant body of
analysis, there have been relatively few attempts to look at the category of sexual offenses
systematically, across the board.1 I have recently begun work on a book that takes up that
challenge. My goal will be to consider the extent to which the sexual offenses form a coherent
whole, reflecting a distinctive collection of rights, interests, wrongs, harms, and offensive
behavior. This is not to deny that the sexual offenses reflect a tremendous amount of diversity:
The gulf between non-consensual offenses (such as rape) and consensual offenses (such as
adultery) in particular seems especially wide. What I hope to show, however, is that the
consensual and non-consensual offenses do not exist in entirely separate spheres, that in fact they
bear a complementary relationship -- historically, conceptually, and in terms of public policy. I
shall argue that one can’t fully understand one sphere without coming to terms with the other.
The project can be thought of as a kind of “normative reconstruction” of the sexual
offenses. I thus follow the late Neil MacCormick here in attempting to dismantle a wide range of
legal sources and then
to reconstruct them in a way that makes them comprehensible because they are now
shown as parts of a well ordered though complex whole. This requires explanatory
principles establishing criteria of what counts as well ordered and rational. . . . Of course,
it is an intellectual process, involving a new imagining and describing of the found order.
. . . [T]here has to be some discrimination between the parts that belong in the coherent
*
Dist. Professor of Law, Rutgers School of Law, Newark. Email: sgreen@kinoy.rutgers.edu
1
Two exceptions are Richard A. Posner, Sex and Reason (HUP, 1992) (considering sexual offenses from the
perspective of law and economics), and Donal E.J. MacNamara and Edward Sagarin, Sex, Crime, and the Law (Free
Press, 1977) (considering offenses from perspective of criminology). See also Richard Posner and Katharine B.
Silbaugh, in A Guide to America’s Sex Laws (U. Chicago Press, 1996) (compiling statutory provisions).
1
whole and the mistakes or anomalies that do not fit and ought to be discarded or
abandoned or at least revised.2
Consistent with the practice of normative reconstruction, an attempt will be made to develop a
rational and coherent understanding of the underlying norms while acknowledging where such a
conception departs from one or more common formulations of the offenses.
At this early stage of the project, my intent is to (1) identify some of the key conceptual
issues that will need to be addressed, and (2) offer some preliminary hypotheses that will need to
be tested.
Defining the sexual offenses
One way to answer the question “what are the sexual offenses?” would be to look at how
the issue is addressed in positive law. For example, there are extensive collections of criminal
statutes in Anglo-American law that are labeled “sexual offenses,” as well as provisions that
subject a wide range of “sex offenders” to registration and notice requirements.3 We could
simply survey these statutes to see which offenses are included. But this would not be a
particularly interesting way to approach the question. Nor would it tell us why we should care
about these offenses, or give us a basis for comparing schemes from different jurisdictions. And
it would not give us any external criteria for deciding whether such classification was over- or
under-inclusive. To cite just one example: one of the offenses included in the Sex Offender
Registration and Notification Act (SORNA), the federal version of Megan’s Laws, is the
kidnapping of a minor. Is this properly classified as a “sexual offense”? Unless we have some
independent criteria for defining the category, we cannot say for sure. Indeed, looking at the
sexual offenses as a functioning category of positive criminal law only raises further questions:
What do these offenses have in common? How do they differ from non-sexual offenses? Are
there offenses that should be regarded as sexual offenses that are not generally included on the
list? Are there offenses that are regularly included on the list that should be excluded?
So what should count as a sexual offense? The majority of sexual offenses qualify as
such because they have as an element an offender’s engaging in conduct that constitutes a sexual
act. This is true, for example, of rape, sexual assault, statutory rape, abuse of position of trust,
fornication, adultery, sodomy, incest, sadomasochism, bestiality, necrophilia, and sexual
transmission of disease. But what constitutes sexual conduct varies significantly from offense to
offense. Sometimes, the sexual nature of the prohibited conduct is defined with almost clinical
specificity. Other times, it is left vague or is entirely implicit. In some cases, sex functions as a
necessary conceptual element of the conduct prohibited, while elsewhere it seems to be merely
contingent.
2
Neil MacCormick, “Reconstruction after Deconstruction: A Response to CLS,” 10 Oxford J. Legal Studies 539, 556
(1999). See also work by Nicola Lacey.
3
English Sexual Offences Act 2003 (includes rape, assault, child sex offenses, abuse of position of trust, prostitution
and related offenses, indecent exposure, indecent photographs of children, voyeurism, bestiality, necrophilia); Model
Penal Code Art. 213 (includes rape and related offenses, deviate sexual intercourse by force or imposition,
corruption of minors and seduction, sexual assault, and indecent exposure). Sexual offender registration is dealt with
in Part 2 of Sexual Offences Act and in numerous state Megan’s Laws.
2
Also to be included within the category of category of sexual offenses are various crimes
that prohibit not sexual conduct per se, but rather conduct that is related to sexual conduct in a
more attenuated manner – whether by being preparatory of future illicit sexual acts (such as
solicitation, pandering, child grooming, bigamy, polygamy, and failing to register as a sex
offender); infringing on a victim’s future ability to enjoy sex (as in female genital mutilation);
intruding on a victim’s right to sexual privacy (voyeurism) or right to avoid witnessing others’
sexuality (indecent exposure and obscenity); or by ratifying previously occurring illicit sexual
conduct (such as in the case of distribution and possession of child pornography). As is the case
with other “preventive justice”-type criminal statutes, one of the questions to be considered is
whether the conduct prohibited by such provisions is too attenuated from the supposed harms
sought to be prevented to justify criminalization.4
Sexual conduct vs. “having sex”
Defining the sexual offenses in terms of sexual conduct prohibited raises an even more
fundamental question: namely, what should count as sexual conduct in the first place? To put it
more precisely, what necessary and sufficient conditions make an act a “sexual” act rather than
some other kind of act? At some level, of course, the category of sex reflects a “know-it-when-Isee-it” quality. But that approach will not do in hard cases. Unless we have some clear criteria
for defining sexual acts per se, we will have difficulty in saying what constitutes a sexual act in
more specific cases – such as when it is forced (as in sexual assault), performed with someone
other than one’s spouse (as in adultery), bought and sold (as in prostitution), or performed with
an animal or corpse (as in bestiality and necrophilia, respectively).5
For a start, I would distinguish between “sexual conduct” and “having sex.” Consider in
this connection an empirical study published in 1999, not long after Bill Clinton implied, in
grand jury testimony, that he had not “ha[d] sex” with Monica Lewinsky (who, it turns out, had
fellated, but apparently had not had intercourse with, him). In the study, approximately six
hundred American college students were asked what kinds of behaviors they would regard as
“having sex.”6 (Specifically, they were asked: “Would you say you ‘had sex’ with someone if the
most intimate behavior you engaged in was . . . (mark yes or no for each behavior).” While there
were some modest differences between the responses of men and women, a basic hierarchy
emerged: More than 99 percent said they would be “having sex” if they had engaged in penilevaginal intercourse; 81 percent, penile-anal intercourse; 40 percent, oral contact with genitals; 15
percent, having a person touch the genitals; and less than 5 percent, oral or digital contact with
breasts or nipples, or deep kissing.
4
See generally Andrew Ashworth, Lucia Zedner, and Patrick Tomlin (eds.), Prevention and the Limits of the
Criminal Law (OUP, 2013).
5
Alan Soble, Activity, Sexual, in Alan Soble, Sex from Plato to Paglia: A Philosophical Encyclopedia (Greenwood
Press, 2006), vol. 1, at 15, 15.
6
Stephanie A. Sanders and June Machover Reinisch, “Would You Say You “Had Sex” If . . . ?,” 281 Journal of the
American Medical Association 275 (1999). Although the study was published after the Lewinsky scandal broke, the
data were obtained before.
3
What explains these responses? The authors concede that their study does not answer that
question.7 But we can speculate: Perhaps the young subjects in the study were thinking about
whether they could engage in such contact and still, for better or worse, consider themselves
virgins. Perhaps they were concerned with issues of “fidelity” to boyfriends or girlfriends.
Perhaps their answers varied depending on their sexual orientation. In assessing their responses,
it would be helpful to know what the subjects understood as the costs and benefits (to their
mental health, self-esteem, reputation among their peers, and the like) of labeling some behavior
as “having sex.” Would their answers have differed if they had been asked to make judgments
about the conduct of others, rather than themselves? What if the person they were making a
judgment about was their own sexual partner? Would it matter if the conduct was performed in
the context of a “hook-up” or “one-night stand” rather than in a long-term relationship? What
assumptions did the subjects make based on the minimal description of the conduct given? Did
the subjects assume that the contact was consensual? Would their answers have differed if they
had been told that they had been forced or tricked or coerced into having such contact?
I expect to return to studies of this sort later (in particular, when we look at adultery), but
for the moment, three points are worth making: First, while common or conventional usage is
worth considering, it can hardly be viewed as conclusive in critical projects of this sort. Second,
in deciding what constitutes “sexual activity,” context and audience matter. Perhaps the subject
students’ answers to the question “would you say you ‘had sex’” would have differed had it been
posed in some forum other than a social science study – say, in a late night “bull session,” or in a
doctor’s medical history. Third, and more specifically, the terms “having sex” and “sexual
activity” almost certainly refer to different phenomena: “Having sex” is in all likelihood
understood as a euphemism for sexual intercourse, and as such is best understood as a (centrally
important and in some respects preeminent) subset of what I shall now suggest is the broader
category of “sexual conduct.”
What is sexual conduct?
Historically, theorists have been less concerned with the question “what is sex” or “what
is sexual conduct” than what is “natural” or “normal” or “morally worthwhile” sex or sexual
activity. And they have invariably answered this second question in terms of one or another
purpose or end – whether it is procreation, love, communication, pleasure, or something else.8
Judeo-Christian authorities have historically defined sex in terms of procreation: humans are
commanded to be fruitful and multiply.9 Activity that is directed toward that end, and performed
within the framework of marriage, was regarded as normal or natural sex. Other conduct was
considered deviant, perverse, or unnatural. Thus, for commentators like Augustine, the
7
Id. Though some later studies have attempted to explore these questions. See Ava D. Horowitz and Louise Spicer,
50 J. Sexual Research 139 (2013) (“Having sex” as a graded and hierarchical construct: a comparison of sexual
definitions among heterosexual and lesbian emerging adults in the U.K.). [cite more]
8
For a helpful summary, see Igor Primoratz, Ethics and Sex (Routledge, 1999), at 9-49.
9
Of course, there are significant differences both between the Jewish and Christian views of sexuality and within
the two traditions. For a useful discussion, see David M. Feldman, Birth Control in Jewish Law (Jason Aronson,
1998).
4
descriptive question of what constitutes sex was driven by what were essentially normative
considerations.
More modern, secular scholars have focused on goals other than procreation. For
example, Roger Scruton’s theory of sex focused on notions of intimacy, love, and “mutuality of
desire,” while Thomas Nagel’s focused on sex as a kind of language – a complex, multi-layered
process of mutual perception and arousal.10 Conduct that failed to achieve, or at least aim at,
such ends – for Scruton, masturbation and bestiality; for Nagel, voyeurism and sadomasochism
-- was considered suboptimal, or even perverse.
For present purposes, such a means-end approach presents problems. Although we will
eventually need to deal with the concept of “deviant sex,” our more immediate concern is with
defining the larger category of “sex,” of which “natural” and “unnatural” sex must logically be
subsets.
One particularly influential response to the more general “what is sex” question is
provided by Alan Goldman.11 Goldman says that we should define sex on its own terms, rather
than as a means to something else. According to Goldman’s definition, “sexual desire is desire
for contact with another person’s body and for the pleasure which such contact produces; sexual
activity is activity which tends to fulfill such desire of the agent.”12
Although Goldman’s approach lacks the limitations of Scruton’s and Nagel’s, it
nevertheless presents problems of its own. First, despite Goldman’s disavowal of a teleological
approach, it appears that his approach is itself framed in terms of a means to an end – i.e., having
contact with another’s body as a means to pleasure.
Second, Goldman’s approach to defining sexual activity seems to apply to conduct that
should not properly be regarded as sexual; as such, it is over-inclusive. Goldman himself voices
concern about cases such as contact sports and cuddling with a baby. Goldman concedes that
both involve having contact with another’s body as a means to pleasure, but maintains that “the
desire is not for contact with another body per se, it is not directed toward a particular person for
that purpose, and it is not the goal of the activity.”13 In the case of contact sports, the goal is
“winning or exercising or knocking someone down or displaying one’s prowess.” In the case of
cuddling with a baby, the goal is to demonstrate “affection, tenderness, or security.”14
10
Roger Scruton, Sexual Desire: A Philosophical Investigation (Continuum, reprint ed. 2006); Thomas Nagel,
Sexual Perversion, in Mortal Questions (CUP, 1979), 39. See also Robert Solomon, Sexual Paradigms, 71 J.
Philosophy __ (1974) (describing linguistic theory of sexuality).
11
Alan H. Goldman, Plain Sex, 6 Philosophy and Public Affairs 267, 268 (1977). Primoratz calls Goldman’s the
“best philosophical statement of the hedonist understanding.” Ethics and Sex, at 41. Alan Wertheimer also adopts
Goldman’s definition of sex as his own. Consent to Sexual Relations (Cambridge U. Press, 2003), at 37-38.
12
Id.
13
Id. at 269.
14
Id.
5
Perhaps. But even if Goldman is right about contact sports and baby cuddling, that leaves
the problem of (ordinary, non-“happy ending”) massage, which Goldman himself does not
consider. Surely, getting a massage is an “activity which tends to fulfill” the “desire for contact
with another person’s body and for the pleasure which such contact produces.” To be sure, what
the massage recipient desires is for the massage therapist to have contact with the recipient’s
body, not the other way around. But the same can be said of presumptively sexual activities like
receiving oral sex. And while massage can certainly have therapeutic value, it is hard to deny
that the basic point of a massage is the pleasure of physical contact. So, either massage is a form
of sex or Goldman’s account is too broad.
Third, Goldman’s physical-contact-leading-to-pleasure definition of sex is also underinclusive inasmuch as it would seem to exclude rape and prostitution. It must be a very rare case,
if indeed there are any, in which being the victim of forced intercourse “tends to fulfill” the
desire for contact with another person’s body. Even some rapists might not be interested in
sexual pleasure per se; they might be motivated purely by a desire to dominate their victims
(presumably, this is what some feminists mean when they say that rape is a crime of violence,
not a crime of sex). From a purely descriptive perspective, however, it seems odd to say that the
intercourse involved in rape is not sex, or that the rapist is engaged in a sexual activity while the
victim is not.15 Many prostitutes are also unmotivated by a desire for pleasure while they are
having sex, and there also may be cases in which committed lovers do not experience sexual
pleasure, as when one engages in sex merely to please the other. Indeed, in each of these
contexts, Solomon’s hedonic approach seems reminiscent of the traditional, means-end, onlygood-sex-constitutes-sex approach.
Another way in which Goldman’s approach is under-inclusive follows from his
requirement that sex must involve a “touching” of “another person.” There are some kinds of
presumptively sexual activity that do not involve touching: think of flirting, seduction, phone
sex, voyeurism, exhibitionism, and viewing pornography. There are also forms of presumptively
sexual activity that do not necessarily involve another person, such as masturbation, and, again,
viewing pornography.
Goldman himself recognizes the difficulty his definition of sexual activity has in applying
to such conduct. His solution is an awkward one:
While looking at or conversing with someone can be interpreted as sexual in given
contexts it is so when intended as preliminary to, and hence parasitic upon, elemental
sexual interests. Voyeurism or viewing a pornographic movie qualifies as a sexual
activity, but only as an imaginative substitute for the real thing . . . The same is true of
masturbation as a sexual activity without a partner.16
15
Though some feminist scholars have indeed made this claim. See, e.g., Greta Christina, Are We Having Sex
Now or What?, in Alan Soble and Nicholas Power (eds.), The Philosophy of Sex: Contemporary Readings (5th
ed.2008), 23, 27-28.
16
Id. at 270.
6
I’m not sure I understand what Goldman means when he says that certain kinds of
activity “qualif[y]” as sexual activities, but only as “imaginative substitute[s] for the real thing.”
It seems to me that masturbation and pornography viewing are such ubiquitous activities that
they should be able to stand on their own as discrete forms of sexual behavior. It seems odd, in
other words, to insist that “real” sex necessarily involves physical contact with a partner when so
much sex-like behavior seems to involve neither.
A subjective approach to defining sexual activity
In critiquing Goldman’s hedonic account of sex, I have focused on the fact that it would
include presumptively non-sexual activities like massage, and exclude, or at least downgrade to
“substitutes,” presumptively sexual activities like masturbation, viewing pornography,
voyeurism, phone sex, and flirting. Can I offer an account that accords more closely with our
intuitions and common usage?
I agree with Goldman that the definition of sex should be closely tied to the notion of
pleasure, but I would propose two major changes to his account. Goldman says that “sexual
desire is desire for contact with another person’s body and for the pleasure which such contact
produces.” I would modify this to say that sexual desire is desire for sexual pleasure; sexual
activity is activity which tends to fulfill such desire of the agent. I would thus eliminate the
requirement that desire be for contact with another person’s body and specify that sexual desire
is desire for sexual pleasure.
So what do I mean by sexual pleasure? Isn’t it circular to define sexual desire as a desire
for sexual pleasure? How exactly would my account distinguish the pleasure of a lover’s caress
from the pleasure of a massage therapist’s effleurage? (Let us assume, for purposes of
discussion, that both the lover and the therapist are touching the same, non-sexual-organ part of
the body – say, the hands or neck.) I would say that the difference between massage and the
lover’s caress is a phenomenological one; it is a matter of how each activity is intended, and
perceived. Unlike massager and massagee, caresser and caressee feel sexually aroused.
Moreover, such subjective feelings of arousal typically manifest themselves in a range of
objectively measurably ways. In men, arousal typically involves the swelling and erection of the
penis and changes in hormone levels. In women, sexual arousal involves vaginal wetness,
swelling and engorgement of external genitals, internal enlargement of the vagina, and increases
in testosterone levels. Other changes include an increase in heart rate and blood pressure, a
feeling of being hot and flushed. Evidence of sexual arousal has also been observed in images of
the amygdala and hypothalamus.17 This is not to say that every sexual thought, image, or activity
will trigger all of these responses in all people, or that these physical symptoms cannot in some
unusual cases result from non-sexual feelings.18 It is only to recognize that there is an
unmistakable subjective feeling that virtually every person on earth can recognize as sexual and,
17
Sexual Arousal, Wikipedia, http://en.wikipedia.org/wiki/Sexual_arousal
18
Simon Blackburn offers the case of priapism as involving the physical manifestations of lust or desire, but
without the usual psychological triggers. Simon Blackburn, Lust (OUP, 2004), at 16.
7
which, as neuroscience progresses, will increasingly be subject to empirical observation. Under
this approach, what is perceived as sexual is sexual.19
If I am right, then this would explain why neither massage nor baby cuddling should
normally qualify as sex. Though physically pleasurable, neither of these activities is typically
attended by the subjective feeling of sex or the physiological indicators of sexual activity. In
those unusual cases in which they are accompanied by such feelings, however, it would be
proper to regard them as “sexual.”
And what about the prostitute, the victim of rape, and the person who engages in sex
merely to please his or her partner? If such acts were accompanied by a physiological or
psychological response that was sufficiently similar to that triggered during genuinely
pleasurable forms of sex,20 then on my account they should be regarded as sexual.
My account would also eliminate Goldman’s requirements of touching and sharing. The
question here is largely a definitional one. We could say, along with Goldman, that that
masturbation and viewing pornography are merely “imaginative substitutes” for genuine sexual
contact. And at some level that is certainly true. But the fact is that both masturbation and
pornography involve subjective feelings and physiological responses that are very similar to
those experienced during intercourse, foreplay, and other shared forms of sexual contact and
precursors to sexual contact.
Another advantage of the subjective approach is that it allows us to account for objects
and situations that are not normally regarded as sexual, but which evoke an idiosyncratic sexual
response in one or more persons -- such as fetishes and paraphilia. Often, these involve everyday
objects and situations that are not thought of as sexual by the general public – say, “Guardsmen
socks” or frotteurism -- but become viewed as sexual by individuals. When these objects and
situations trigger a sexual response, it makes sense to think of them as sexual.
What is so special about sexual conduct?
Sex plays a highly valued role in our law, and in our lives generally.21 Most people regard
the decision whether, when, how, and with whom to have sex to be among the most personal,
19
A somewhat similar subjective approach is suggested in Margo Kaplan, Sex-Positive Law, 87 NYU Law Review
(forthcoming, 2014), though Kaplan would apparently go even further and include in the definition of sex not just
sexual pleasure produced by physical, visual, or auditory stimuli, but also “mere thoughts and fantasies without any
external stimulation.” (p.7).
20
See, e.g., Diana T. Sanchez, et al., Eroticizing Inequality in the United States: The Consequences and
Determinants of Traditional Gender Role Adherents in Intimate Relationships, 48 J. Sex Research 168 (2012)
(postulating the male rapists typically experience sexual pleasure).
21
A recent study suggests the median number of sexual thoughts per day for men was 18.6 (quite often, though far
less than suggested by the urban myth that men think about sex every seven seconds). The median number for
women was 9.9. See Terri D.Fisher, et al., Sex on the Brain?: An Examination of Frequency of Sexual Cognitions as
a Function of Gender, Erotophilia, and Social Desirability, 49 Journal of Sex Research 69 (2012).
8
private, and meaningful choices they make in their lives.22 The fact that an act or interest is
sexual also carries special weight in the law, as can be seen in the case of sexual assaults,
invasions of sexual privacy, and sexual exploitation of children, each of which is treated as more
serious, other things being equal, than non-sexual assaults, invasions of non-sexual privacy, and
non-sexual exploitation, respectively.
A major reason sex is so valued is that it holds the potential for a range of significant
benefits that ordinarily cannot be obtained by other means. Without sexual intercourse, of course,
we could not reproduce; and under evolutionary theory, we would expect that human beings
would place a priority over controlling their reproductive autonomy.23 Sex also provides
significant hedonistic benefits -- both as a relief from the carnal demands of sexual desire, and in
what Richard Posner has called the more refined sense of “ars erotica, the deliberate cultivation
of the faculty of sexual pleasure; the analogy is to cultivating a taste for fine music of fine
wine.”24 Finally, and perhaps most complexly, sex can satisfy deep-seated needs for human
connection, intimacy, and communication (as well as, in some cases, for domination and
submission).
Tabulating the many benefits of sex, however, probably doesn’t fully account for the
importance we place on it in our lives. For one thing, not all sexual acts produce these benefits
equally. Obviously, only sexual intercourse has the potential for procreation; and only certain
kinds of sexual activity performed with a partner hold the potential for human connection and
intimacy. Yet, even those sexual acts with few clear benefits are still considered to be within the
realm of privileged conduct. I have not yet worked out an adequate explanation for why this
should be, but I would assume that such an explanation would take account of both innate and
cultural aspects of human sexuality.
Hierarchies in sexual conduct
Given the wide range of activities that would qualify as sexual on my account, it is worth
asking whether, within the category of sexual conduct, some activities should be regarded as
more central or highly valued than others. The answer to that question, of course, will depend on
the criteria by which such acts are judged. Some sexual acts are presumably more conducive than
others to pleasure, intimacy, procreation, communication, or other goals. In a system that judged
the value of a given sexual act by its ability to produce φ, those sexual activities that produced
more φ would be judged as more valuable than those that produced less.
These judgments are important because they affect not only how we choose to conduct
some of the most important aspects of our lives, but also how we classify and grade the sexual
offenses, and whether various offenses should be crimes to begin with. In the case of
22
For an interesting take on the decision to abstain from sex, see Elizabeth F. Emens, Compulsory Sexuality, 66
Stanford LR (forthcoming, 2013).
23
See Jeffrie G. Murphy, Some Ruminations on Women, Violence, and the Criminal Law, in Jules L. Coleman and
Allen Buchanan (eds), In Harm’s Way: Essays in Honor of Joel Feinberg (Cambridge U. Press, 2004), 209, 214.
24
Posner, Sex and Reason, at 111.
9
nonconsensual sexual offenses such as rape and sexual assault, we can observe the following
pattern: the more central or significant or highly valued the form of sex in which V is compelled
to participate, the greater the infringement, and, consequently, the more serious the offense.25 For
a variety of reasons -- cultural, physiological, hedonic, and historical -- sexual intercourse seems
to enjoy a special moral and legal status. It almost surely involves a higher level of intimacy
than, say, kissing or fondling. So forcing V to have sexual intercourse will be viewed as entailing
a more wrongful and harmful act than forcing V to kiss or touching V in a sexual way without
consent.
In the case of consensual or aconsensual sexual offenses, such as voluntary adult incest
and sadomasochism, a different sort of pattern exists. Here, we might say that, the more valuable
the form of prohibited sex is to its practitioner, the greater we would expect the burden on the
government to justify the prohibition.
Sexual conduct in sexual offenses
The question of what constitutes sexual activity turns out to be quite relevant within the
context of the sexual offenses. Many offenses specifically enumerate those activities that will be
considered “sexual.” Others leave that determination to the finder of fact, based on the
defendant’s motives or intent.
The Sexual Offences Act 2003 does both, though in different contexts. Consider Section
4, which makes it a crime to cause a person to engage in “sexual activity” without consent.
“Sexual activity,” in turn is defined quite explicitly. It consists of nonconsensual: penetration of
V’s anus or vagina, penetration of V’s mouth with a person’s penis, penetration of a person’s
anus or vagina with a part of V’s body or by V with anything else, or penetration of a person’s
mouth with V’s penis. This provision of the statute could hardly be more specific or categorical
about what kind of “sexual activity” is covered.
But what constitutes “sexual activity” in other provisions of the Act is defined in a much
less categorical manner. For example, the offense of sexual assault (Section 3) criminalizes nonconsensual touchings of any part of the victim’s body with any part of the offender’s body or
with anything else, provided that the “the touching is sexual,” which in turn is defined (under
Section 78) as being of a sexual “nature” or because its “circumstances” or “purpose” are sexual.
Similarly, statutory rape, incest, necrophilia, and bestiality provisions, at least in some
jurisdictions, prohibit any “sexual” “contact” or “activity” with a child, familial relation, corpse,
or animal, as the case may be.
So what does it mean for a touching to be of a sexual nature, have a sexual purpose, or
occur in sexual circumstances? To decide that, I would apply something like the test articulated
above. I would ask whether the activity is the sort that “tends to fulfill” the “desire for sexual
pleasure.”
Consider the English case of Court.26 D pulled V, a 12-year old girl across his knees and
smacked her buttocks with his hand through her shorts. Had V been a parent or schoolmaster, he
25
Gardner and Shute make a similar point, which is discussed below.
26
[1989] AC 28, 44.
10
might have been doing so for disciplinary purposes. But, in the actual case, D was a store clerk
and V was a customer, and when asked by the police why he had done what he did, D admitted
that he had a “buttock fetish.” For this reason, D was held to have had a “sexual” motive, and
therefore to have committed indecent assault. This seems to me the right result.
In some cases, a fact-specific, “tends to fulfill” test is in fact preferable to a categorical
approach to defining sexual activity. Consider the case of Tabassum.27 D carried out what may
have been genuine research into breast cancer, aware that women he examined mistakenly
assumed he was a medical doctor. The court held that their mistake vitiated the women’s consent
and that, even assuming the research was genuine and done without a sexual motive, the
examinations were inherently sexual and therefore constituted indecent sexual assaults rather
than mere assault.
Following my suggestion that an act should be regarded as sexual if and only if it is
“tends to fulfill” the agent’s desire for sexual pleasure, the holding in Tabassum seems to me
mistaken. Touching a woman’s breast for the purpose of unconsented-to medical research does
not appear to qualify as a sexual act. Indeed, there are all kinds of acts that involve contact with
“sexual” bodily parts that are not, in the normal course of things, sexual (gynecological exams,
urological exams, mammograms, etc.). Touching those parts without informed consent is
certainly a violation of the victim’s rights to bodily autonomy, but it is not a violation of
specifically sexual rights.
Sex as a distinguishing factor in sexual offenses
In considering the role that sexual conduct plays in defining various sexual offenses, one
of the questions we will want to consider is the extent to which the sexual nature of such conduct
is necessary, or merely contingent.
In some cases, the infringement of a right to sexual autonomy serves to inform the
offense and distinguish it from otherwise similar non-sexual crimes. For example: Sexual assault
differs from other assaults in the particular wrongs it entails: unjustifiably touching someone
who does not want to be touched is morally wrong, but forcing someone to engage in those most
intimate of relations is worse, and is almost invariably treated as a distinct offense. Female
genital mutilation follows a similar pattern: although non-lethal mutilation involving non-sexual
organs is a very serious crime, mutilation of the sexual organs has physiological and
psychological effects on a victim’s life that are distinctive, and for this reason FGM deserves to
be treated as a distinct offense. All voyeurism involves an infringement of the victim’s privacy,
but voyeurism that intrudes on a victim’s sexual privacy is arguably distinctive and worthy of
specialized legislation. Similarly, offensive exposure of any sort can make observers
uncomfortable, but indecent exposure that involves sex arguably involves a distinctive kind of
wrong. And though there are many ways for teachers, coaches, and clergy to exploit the young
people over whom they have authority, using their position to obtain sex seems especially
wrongful and therefore worthy of special criminalization.
27
[2000] 1 Cr. App. R. 328.
11
But there is also a range of offenses that, though sometimes classified as sexual, prove,
on inspection, to be only contingently so. For example, using sex to transmit disease is wrongful,
but it is not clear that it is qualitatively worse than doing so by means of a (non-sexual) blood
transfusion, provided that both underlying acts are voluntary, or involuntary, as the case may be.
Necrophilia is also typically classed as a sex offense, but I am skeptical that corpse desecration
involving sex is qualitatively different from non-sexual forms of corpse desecration, such as
dismemberment. And to the extent that bestiality is properly thought of as a form of cruelty to
animals, it is not clear that it should be viewed as involving any greater cruelty than subjecting
an animal to beating, abandonment, or confinement.
Looking at sexual conduct normatively
So far, we have considered the concept of sexual activity descriptively. I now want to
begin to consider sex normatively. There are many perspectives from which value judgments
about sexual activity can be made. We can judge such activity from an aesthetic or emotional
perspective: for example, we can say that sex was exciting or dull, pleasurable or painful,
meaningful or not. We can also judge sex from a functional standpoint: for example, that orgasm
was or was not achieved, that it led to a wanted (or unwanted) pregnancy, the consummation of a
marriage or other intimate relationship, the transmission of a disease, or emotional trauma.28
From a moral perspective, human conduct is normally classified as obligatory,
permissible, supererogatory, wrongful, or amoral.29 It would be an interesting exercise to identify
cases in which sexual activity could be classified as obligatory or supererogatory. But, for
present purposes, I shall be primarily concerned with identifying sexual acts that are wrongful or
blameworthy. From the perspective of the retributive theory, that will be a prerequisite for
criminalization.
When is sexual conduct morally blameworthy?
When should a sexual act be considered morally blameworthy? Given the larger project,
it seems appropriate to limit the discussion of this question in two ways. First, I intend to focus
solely on the kinds of sexual conduct that are, or traditionally were, subject to criminal sanctions.
Second, I intend to consider sexual activity types rather than tokens. The inquiry will thus
involve a kind of generalization about criminal conduct, similar to the sort of inquiry that a
legislature makes (or ought to make) in determining whether a certain kind of conduct justifies
criminal penalties, and, if so, what kind and what amount. The question is, in effect, to what
extent the sort of defendant who commits theft can be said on average, or typically, to be at fault.
In such an assessment, it is normal to assume that no applicable exculpatory conditions, such as
mistake, duress, or justification, apply; and the individual motivations or personal histories of
particular defendants who commit such acts are not ordinarily a concern. Thus, I put to the side
cases in which an act is deemed wrongful because, for example, the actor’s motives were
insincere or because he was insensitive to his partner’s feelings.
28
Sex is also sometimes also spoken of as being “natural” or “unnatural” (or “deviant,” or “perverse”), though, as
we shall see, this is probably best understood as a particular way of talking about its moral content.
29
The locus classicus is J. O. Urmson, Saints and Heroes, in A. Melden (ed.) Essays in Moral Philosophy, (Seattle:
University of Washington Press, 1958).
12
Of course, the line between types and tokens is not always clear. For example, should we
regard vaginal intercourse as an activity type and intercourse performed with someone who is not
one’s spouse (adultery) as a token, or should we regard adultery itself as a separate activity type?
In making such judgments, I shall take my lead from the way such acts have traditionally been
dealt with by legislatures and courts, while not of course accepting such treatment as necessarily
correct.
The Harms, Wrongs, and Offensiveness of Sexual Conduct
In considering the extent to which a given form of sexual conduct should be considered
blameworthy, my focus will be on harms and wrongs, as well as on the concept of offensiveness.
Harmfulness reflects the degree to which a criminal act causes, or risks causing, what Feinberg
called a “significant setback to another’s interests.”30 Wrongfulness reflects the extent to which
the act involves the violation of a freestanding moral norm, rule, right, or duty. The concept of
offensiveness functions as a kind of alternative to harmfulness. Though not causing significant
setbacks to interest, certain kinds of conduct nevertheless entail widely disliked mental states
such as disgust, revulsion, shock, shame, embarrassment, and anxiety; and it is these kinds of
behaviors that sometimes cause what Feinberg called “offense.” For present purposes, I offer no
argument that such elements comprise a set of necessary and sufficient conditions for
criminalization. Nor do I mean to suggest that these elements do not overlap to some significant
degree; indeed, we shall see that they often do. Instead, I hope to show merely that they are
relevant to determining whether an actor’s conduct should be regarded as blameworthy.
Let me offer a preliminary example of how the distinction between harms and wrongs
would play out in the context of the sexual offenses. Let us look first at wrongless harms.
Imagine a case in which X and Y engage in consensual sexual intercourse. Despite good
intentions and reasonable precautions on X’s part, Y nevertheless finds herself with an unwanted
pregnancy or a sexually transmitted disease brought on by the sexual act. To the extent that X has
caused Y to suffer a setback to her interests, we should say that X has caused Y harm. We should
not, however, conclude that X has been wronged, since X committed no violation of a
freestanding norm, rule, right, or duty.
Finding a case in which X is wronged but not harmed is a bit more controversial, at least
in the context of sexual intercourse. Gardner and Shute ask us to imagine a case of what they call
“pure rape.” As they put it:
It is possible, though unusual, for a rapist to do no harm. A victim may be forever
oblivious to the fact that she was raped, if, say, she was drugged or drunk to the point of
unconsciousness when the rape was committed, and the rapist wore a condom. . . . [In
such cases,] we have a victim of rape whose life is not changed for the worse, or at all, by
the rape. She does not, in MacKinnon’s phrase, “feel violated.” She has no feelings about
the incident, since she knows nothing of it. Indeed, the story has no prospective
30
Feinberg, Harm to Others.
13
dimension for the victim, except possibly a hangover in the morning; otherwise the
victim’s life goes on exactly as before.31
Such a case, it seems, would involve a wrong to Y, but not, it seems, any harms. [We will return
to this potentially controversial contention in due course.]
In considering the blameworthiness that underlies the sexual offenses, we will also have
to look at the concept of offensiveness. As noted, this involves behavior that causes in its victim
mental states such as disgust and revulsion, rather than any setbacks to interest. Such conduct is
often wrongful, in the sense that it violates someone’s rights, though not all behavior that causes
offense will necessarily involve a violation of rights: Like harmfulness and wrongfulness,
offensiveness and wrongfulness are conceptually separable, or so I shall argue. Whether the fact
that a given type of conduct has a tendency to cause wrongful offense should be taken to satisfy
the blameworthiness requirement of retributive justice is, as we shall see, a topic of considerable
complexity and controversy.
A preliminary taxonomy of sexual offenses
Rather than simply consider each sexual offense seriatim, I propose that we look at them
within the context of a taxonomy. But what criteria should be used to classify the sexual
offenses?
One possibility would be to focus on harms. But the harms caused by sexual offenses are
extremely diverse. For example, the potential harms engendered by rape include physical pain,
unwanted pregnancy, sexually transmitted diseases, injury to internal organs, emotional distress,
and shame.32 Female genital mutilation can cause pain, risk of hemorrhage, ulcers, abscesses,
septicemia, tetanus, gangrene, shock, and death, chronic urinary tract infections, infertility, and
complications in labor, in addition to sexual dysfunction and psychological harm. Incest can
potentially cause abuse of family members, intra-familial sexual jealousies and rivalries, and
genetic abnormalities. Necrophilia arguably causes posthumous harms. The analysis of harm in
sexual offenses is complicated even further by the distinction between offenses that cause direct
harm (such as rape, voyeurism, exhibitionism, female genital mutilation) and those offenses that
involve endangerment (such as failing to register as a sex offender, pornography, and
prostitution). In addition, offenses like public indecency, obscenity, and soliciting sex arguably
do not cause harms, in the sense of setbacks to interest, but they may give rise to mental states
associated with offense, such as disgust, revulsion, and shock. To use these harms and offense
grounds as a basis for offense classification would lead to a highly fragmented taxonomy.
In contrast to harmfulness, the concept of wrongfulness seems to offer the opportunity for
a more focused approach. But what is it that makes sexual conduct wrongful? In making a first
cut, I propose that we focus on the role of consent. [Part of the larger project here will be to
31
Shute and Gardner, the Wrongness of Rape, at 5.
32
See Michelle Madden Dempsey and Jonathan Herring, Why Sexual Penetration Requires Justification, 27 Oxford
J. Legal Studies 467, 475-81 (2007).
14
determine exactly why consent, or its absence, play such a large role in defining the moral
content of individual sexual offenses.] For the moment, I rely on a common sense understanding
of consent as a means for dividing my list of presumptively sexual offenses into three broad
categories: “nonconsensual,” “consensual,” and “aconsensual.”
The nonconsensual sex offenses are those in which an offender (whether actually or
constructively) forces, coerces, or deceives a victim into participating in sexual activities in
which the victim does not, or cannot, consent, and thereby infringes on the victim’s sexual
autonomy. These offenses include rape and sexual assault, public indecency, and voyeurism,
statutory rape and abuse of position, production of child pornography, and female genital
mutilation. [These acts normally entail harms.]
The consensual sex offenses are those in which an offender engages in sex that, while
involving the consensual participation of a second party, is nevertheless regarded, or traditionally
was regarded, by society as in some sense “deviant” or “perverse.” Examples are sodomy,
fornication, adultery, sadomasochism, and prostitution. [These acts are not normally harmful,
though they may in some cases be offensive.]
The aconsensual sex offenses are those in which an offender engages in sex-related
conduct that does not necessarily involve a second party, and therefore does not involve a
question of consent. Examples are bestiality, necrophilia, obscenity, possession of child
pornography, and sale of sex toys. [These acts are also not normally harmful, though they are
often offensive.]
Taken together, the consensual and aconsensual sex offenses are often referred to as
“morals” offenses.
Two points about these categories are worth noting at the outset: First, whether a given
offense belongs in one category or another will depend on a range of factors, such as how the
offense is defined in a specific jurisdiction, which aspects of the definition are taken to be
controlling, and how the offense is applied to the facts of a given case. For example, when incest
involves consensual sexual conduct between consenting adults, it would seem to fit into category
2; but when it involves sexual conduct between an adult and a juvenile, it will likely fall into
category 1. Similarly, prostitution and polygamy can involve conduct that is either consensual or
nonconsensual, depending on the circumstances.
Second, the offenses contained in categories 2 and 3, on the one hand, and category 1, on
the other, can be understood as mirror images of each other: The category 1 (forced sex) offenses
all involve ways in which the government seeks to protect potential victims from having their
sexual autonomy infringed by others. The consensual and aconsensual offenses, by contrast,
involve ways in which the government itself seeks to control citizens’ private sexual conduct. In
recent years, largely owing to the feminist critique, society’s view of what constitutes forced sex
has broadened, and the number and variety of such offenses has proliferated. Meanwhile, largely
as a result of the liberal critique, the notion of what kinds of private conduct the government may
legitimately regulate, and what kinds of sex are regarded as “deviant” in the first place, have
15
narrowed, and the number of still-viable offenses involving consensual and aconsensual conduct
has consequently been reduced.
Sexual autonomy and the wrong in nonconsensual offenses
What is there about engaging in a sexual act without the consent of one’s partner that
makes it morally wrongful? How does the wrong in sexual assault differ from the wrong in nonsexual assault? In evaluating the moral content of cases in which D had sex with V without V’s
consent, should it matter whether D forced V to have sex, coerced her into doing so, or deceived
her into doing so? And how does the wrong of having sex without V’s consent compare to the
wrong of forcing her to undergo genital mutilation, enter into a force marriage, witness the
offender’s sexuality (as in indecent exposure), or be spied upon (as in voyeurism)? [lots of
questions here!]
The wrong in rape is frequently described in terms of a violation of the victim’s “sexual
autonomy” – famously referred to in Coker v. Georgia (holding that rape is not sufficiently
heinous to merit capital punishment) as the “privilege of choosing those with whom intimate
relationships are to be established.”33 Jed Rubenfeld offers an overview:
To Patricia Falk, the “central value protected by sexual offense provisions is sexual
autonomy . . . the violation of which represents a unique, not readily comparable, type of
harm to the victim.” Stephen Schulhofer has argued extensively in favor of “sexual
autonomy” and the “right to sexual self-determination.” Philosopher Joan McGregor
concludes that the “moral wrongness of rape consists in violating an individual’s . . .
sexual self-determination and the seriousness of rape derives from the special importance
we attach to sexual autonomy.” The citations could be multiplied.34
It is important, however, to consider exactly what “autonomy” or sexual “selfdetermination” means in the context of rape and sexual assault law. A good place to start is with
Isaiah Berlin’s famous distinction between two kinds of liberty – negative and positive.35
Negative liberty is essentially freedom from interference by external bodies. It involves freedom
from “deliberate interference of “other human beings within the area in which I could otherwise
act.” “You lack political liberty only if you are prevented from obtaining a goal by human
beings.” “Mere incapacity to attain a goal is not a lack of freedom.” “[B]eing free in this sense
means not being interfered with by others.”
This is a fairly “thin” sense of the concept of liberty. It should be contrasted to a much
“thicker” and more complex concept that Berlin calls “positive liberty.” This involves not just
freedom from restraints but positive freedom, understood as self-realization, self-determination,
or self-mastery. Berlin had in mind specifically the idea that a citizen should have a role in
33
433 U.S. 584, 597 (1977).
34
Rubenfeld at 1394 (footnotes omitted).
35
Isaiah Berlin, Two Concepts of Liberty (1958), published in Berlin, Four Essays on Liberty (Oxford U. Press,
1969).
16
choosing who governs the society of which he is a part. (In contexts that are not directly relevant
here, Berlin suggested that negative and positive liberty could sometimes be in conflict with each
other.)
Some criminal laws are intended to protect people’s sexual autonomy in the negative
sense. Such autonomy is infringed when a person is forced to engage in sex or sex-related
activity without her consent. This can happen in a variety of ways. A victim can be physically
forced to engage in sex, or coerced or tricked into doing so. In voyeurism, the victim does not
even have an opportunity to consent because she is normally unaware of the voyeur’s activities.
In statutory rape, the victim is deemed too young to consent to sexual relations in the first place.
Other criminal laws are intended to protect other aspects of negative sexual autonomy. FGM
laws protect girls from having their sexual organs mutilated, indecent exposure laws protect
people from being forced to witness others’ sexuality, and Peeping Tom laws protect people
from having their sexual privacy invaded.36
Other criminal laws have exactly the opposite effect: here, it is the laws themselves that
tend to infringe people’s sexual autonomy. For example, prior to Lawrence, homosexuals were
prohibited from engaging in homosexual acts; in Brown, defendants were prohibited from
engaging in sadomasochistic acts; various miscegenation prohibited people of different races or
religions from having sexual relations; the law in some American states prohibits people from
buying sex toys; and military law prohibits personnel from engaging in adulterous relationships.
Autonomy in the positive sense is more often the subject of civil law, rather than criminal
law. For example, laws allowing same sex marriage have the effect of enhancing sexual
autonomy. Other laws diminish positive autonomy by prohibiting marriages between people of
the same sex or of different races. (Miscegenation laws thus infringe both negative and positive
autonomy, first by prohibiting people of different races from having sex, and second, by
prohibiting them from marrying.) Both negative and positive sexual autonomy can be important
to the achievement of a fully-lived life. But protecting sexual autonomy in the positive sense is
mostly beyond the scope of the criminal law.
Criminalizing consensual and aconsensual offenses
Despite the obvious difficulty of developing a theoretically sound account of exactly why
the conduct underlying the nonconsensual sex offenses is morally wrong, and despite public
policy and scholarly debates over what should be the precise scope of such offenses, the basic
idea that rape, sexual assault, voyeurism, public indecency, and sexual transmission of disease
are morally wrong and that they should be crimes is uncontroversial. (Female genital mutilation,
given the breadth and persistence of its practice in some cultures, may be an exception here.)
36
It is also worth asking whether bestiality and necrophilia should be regarded as nonconsensual offenses. In
bestiality, the offender has sexual relations with a “victim” – a non-human animal -- that is deemed incapable of
giving consent to humans. [This is a curious claim, though, given that certain kinds of animals are regularly bred
with mates essentially chosen for them by their human owners.] The case of necrophilia is even murkier.
Presumably, corpses are incapable of giving consent (though we might wonder if people can give consent, prior to
death, to have their bodies treated in particular ways after they are dead).
17
The situation is quite different in the case of the presumptively consensual and
aconsensual crimes like sodomy, fornication, prostitution, polygamy, incest, sadomasochism,
and adultery. There are genuine debates not only about whether some or all of this conduct
should be criminalized, but also about whether it is morally wrong to begin with.
Historically, such offenses were justified largely, though not exclusively, on moralistic
grounds.37 An appeal was made to society’s collective moral judgment that such conduct, though
often not harmful, was morally wrong and that society would somehow “disintegrate” (in
Devlin’s word) unless that conduct was prohibited and even criminalized.38
Sodomy and fornication have largely been decriminalized in Western societies, but the
other offenses mostly remain on the books. Is there any way to justify this? I can think of six
theories under which criminalization could potentially be justified:
Harmless wrongdoing. Justifying the criminalization of consensual or aconsensual
offenses under the rationale of legal moralism would require three steps. First, the conduct must
be judged deviant or perverse. Second, deviance or perversion must be shown to constitute a
form of moral wrongfulness. Third, such moral wrongfulness must be shown to provide a
sufficient basis for criminalization. I am highly skeptical that the third step could ever be
satisfied in a liberal state.39 But I believe the first two questions are very much open to debate
and are worthy of our consideration.
Paternalism. Justifying the criminalization of conduct such as selling sexual services and
being married to a man with multiple wives would require proof (1) that the conduct tends to
cause “far-reaching, potentially dangerous and irreversible” harms to self (in Gerald Dworkin’s
phrase), and (2) that the state is justified in using coercive methods to protect people from
causing themselves such harm.40
Problems with consent. Some sexual offenses, while ostensibly consensual, may, on
further reflection, be found to involve background conditions that make genuine consent suspect.
Assuming such conduct is also harmful, criminalization would be appropriate. For example,
much incest seems to lack the proper indicia of consent. Similarly, many cases in which the
37
I have dealt with the complex history of some of these laws in Stuart P. Green, Vice Crimes and Preventive
Justice, Criminal Law and Philosophy (October 2013).
38
Patrick Devlin, The Enforcement of Morals (OUP, 1965), at 10.
39
See Joel Feinberg, Harmless Wrongdoing. But see Alan Brudner, The Wrong, the Bad, and the Wayward, in
Rethinking Criminal Law Theory: New Canadian Perspectives in the Philosophy of Domestic, Transnational, and
International Criminal Law (Francois Tanguay-Renaud and James Stribopoulos, eds.) (Hart, 2012).
40
See Gerald Dworkin, “Paternalism,” 56 The Monist 64 (1971), at ___; see also Gerald Dworkin, “Paternalism:
Some Second Thoughts” (1983); Joel Feinberg, Harm to Self (1986); Douglas Husak, “Recreational Drugs and
Paternalism,” 8 Law and Philosophy 353 (1989).
18
offender is a customer of prostitutes or husband to multiple wives seem to raise the possibility
that ostensibly consensual behavior is actually coercive.41
Consent not possible. There are other offenses involving conduct that, while ostensibly
aconsensual, may, on further inspection, be found to involve a morally relevant actor from
whom consent is difficult or impossible to obtain. These offenses are analogous to traditional
nonconsensual offenses like statutory rape and abuse of position of trust. I have in mind here
necrophilia, which arguably causes posthumous harms to the decedent,42 and bestiality, which
arguably causes harm to animals.43
Harm to downstream third parties. Some conduct, even if aconsensual, or consensual as
between A and B, may nevertheless pose risks of harm to downstream third parties. For example,
adultery, though consensual as between participating parties, may in some cases harm the spouse
or family of the cheating spouse. Sale of sexual services is also said to harm family life and,
when solicitations are made in public, degrade the quality of community life. The question here
will be whether such harms are sufficiently serious and unattenuated to justify criminalization.
The point here would be not to “dilute” or “define down” the harm principle,44 but rather to
rethink the context in which such offenses are committed and the harms they may cause.
Conduct is offensive. Conduct underlying offenses like public indecency and obscenity,
even if determined not to be harmful, may cause serious offense to third parties. The task here
will be to determine the nature of such offense and whether it provides an adequate basis for
criminalization.
Sexual Deviance and Perversion
Sorting out which, if any, of the consensual and aconsensual sexual offenses can
justifiably be criminalized, and on what grounds, will require a complex, offense-by-offense
analysis. Historically, the concept of sexual perversion played a significant role in justifying the
regulation of such conduct. Whether it should still be regarded as relevant in this regard, or at all,
remains to be seen. I set out a few preliminary, and somewhat speculative, thoughts here:
At one time or another, in many cultures, homosexuality, oral sex, incest, fetishism,
sadomasochism, transvestism, bestiality, coprophilia, exhibitionism, voyeurism, and
masturbation were all regarded as perversions. Attitudes about at least some of this conduct have
shifted. Some of these acts, even when regarded as perverse, were not criminalized. And at least
some sexual acts that have been criminalized (including adultery, sexual transmission of disease,
and possibly prostitution and even rape) are not normally thought of as perverse, even if they are
immoral.
41
Thom Brooks, “The Problem with Polygamy,” 37 Philosophical Topics 109, 116 (2009).
42
Feinberg’s discussion of posthumous harms is the classic account, though, curiously, he has almost nothing to
say on the subject of necrophilia in any of the four volumes of Moral Limits of the Criminal Law.
43
Neil Levy, “What (if Anything) is Wrong with Bestiality?,” 34 J. Social Philosophy 444, 450 (2003).
44
See Bernard Harcourt, The Collapse of the Harm Principle, 90 J. Criminal Law & Criminology 148 (1999).
19
What qualities are there in these acts that have caused them to be viewed as perversions?
Perverse sex is often spoken of as being “unnatural.” But this of course raises the question of
what it means for sex to be “natural.” One approach would be to say that a sexual act is unnatural
when it is statistically unusual. This would be a normatively-neutral way to think about
perversion, but it is not clear that it would be a particularly workable one. How unusual would an
act have to be before it could be considered perverse? Would have to be universally unusual, or
only unusual within a given population? Would an act committed regularly by a small percentage
of the population be unusual? How about an act committed rarely by a large percentage of the
population? Even if we could agree on what counts as unusual, there would remain the problem
that while some sexual acts traditionally thought of as perverse are engaged in by a relatively
small number of people (e.g., incest, coprophilia, bestiality), other traditionally perverse acts
(such as oral sex and masturbation) are quite common. All that said, it appears that the
unusualness of a given sexual act nevertheless should play some role in deciding whether it is
perverse. In general, the more common and familiar behavior becomes, or at least the more
aware we are of its being performed, the less likely we are to continue to regard it as perverse.
Another way to think about naturalness and unnaturalness is in terms of functionality.
This approach would require us to posit some function that sex in general is meant to serve, and
then ask whether a particular form of sexual behavior serves that end. This is similar to the
approach we saw taken by church authorities, as well as by modern scholars such as Nagel and
Scruton. Thus, if the purpose of sex is, say, procreation, conduct that fails to further that end will
be regarded as dysfunctional, or perverse. A similar analysis would apply if the goal of sex was
love, communication, intimacy, or simply sexual pleasure.
Robert Gray offers an interesting variation on this view.45 For him, the concept of
perversion must be understood in terms of the “natural adaptive function of sexual activity.”46
Because sexual activities are those acts that “give rise to sexual pleasure,” a sexual perversion
would be an act that produced pleasure but was inconsistent with the evolutionarily adaptive
function of sexuality. But Gray nevertheless avoids the natural law view that natural sex must
always be procreative. Even supposing that one function of sex is reproduction, he says, success
in raising new generations of people requires the maintenance of “fairly stable male-female
reproductive pairs” and “well-organized, stable societies.” Thus, any sexual activity that
contributes to (or at least does not interfere with) either dimension of reproduction might avoid
the label “perverse.”47
A few writers have contended that the concept of perversion is essentially empty and
should be abandoned. For example, Michael Slote has argued that sense cannot be made of
“unnatural” desires or practices, since anything in the world is, by virtue of being there, natural.48
45
See, e.g., Robert Gray, Sex and Sexual Perversion, 75 J. Philosophy 189 (1978).
46
Id. at 197.
47
See Sarah Hoffman, Sexual Perversion, in Alan Soble (ed.), Sex from Plato to Paglia, at 767, 771 (characterizing
Gray’s argument).
48
Michael Slote, Inapplicable Concepts and Sexual Perversion, in Robert B. Baker and Frederick A. Elliston (eds.),
Philosophy and Sex (Prometheus, 1st ed. 1975), at 261.
20
And Igor Primoratz says that the concept of perversion is simply too plagued by “inconsistency
and confusion” to be worth preserving.49
I confess to being unsure about exactly what role the concept of perversion should play in
a book on the sexual offenses. Let me conclude by suggesting four possibilities:
First, if nothing else, understanding what has traditionally been understood as sexually
perverted in various cultures seems essential to understanding why certain kinds of consensual
and aconsensual sexual conduct were criminalized in the first place. For example, even Aquinas
recognized a distinction between sexual acts that he regarded as immoral because they are
perversions and sexual acts that he regarded as immoral because they are coercive.50 Without an
understanding of what sexual perversion meant to previous generations, our analysis will lack
historical and cultural context.
Second, at least some conduct that has traditionally been regarded as sexually perverted
tends to elicit a response of disgust and revulsion, and may therefore be subject to
criminalization under an “offense”-focused rationale. I have in mind here conduct such as
indecent exposure, obscenity, bestiality, necrophilia, and incest. What exactly is there about such
conduct that makes many people find it disgusting? Why is its performance regarded as
offensive? Can such conduct be regarded as offensive even in cases where it is performed in
private? It seems impossible to understand these kinds of conduct without reflecting on what it
means for them to be perverse.
Third, some conduct regarded as perverse might be self-harming. One question is
whether the fact that it is perverse adds anything to the analysis of paternalistic criminal laws.
Finally, apart from the question whether sexually perverse conduct is offensive or selfharming, there is a question whether it is morally blameworthy. Does it really make sense to
speak of sexual conduct that is performed in private, whether alone or with a validly consenting
party, and which has no tendency to harm others, as immoral? Or is perversion just an outdated
aesthetic category? Given the retributive requirement that conduct be blameworthy before it can
be criminalized, analysis of the sexual offenses requires that we decide if sexual perversion is
morally wrong.
49
Igor Primoratz, Sexual Perversion, 34 American Philosophical Q. 245 (1997).
50
Thomas Aquinas, Summa Theologiae, IIaIIae, ques. 154, art. 11, discussed in Alan Soble, Philosophy of Love
and Sex, at72-73. Aquinas distinguished between (1) acts that are “incompatible with the sex act,” such as “selfabuse,” “intercourse with a thing of another species,” “acts with a person of the same sex,” and acts in which “the
natural style of intercourse is not observed, as regards the proper organ or according to the rather beastly and
monstrous techniques”; and (2) sexual acts that are morally wrong in spite of being “natural” insofar as they
“conflict with right reason . . . with respect to the other party.”
21