A Nonideal Theory of Sexual Consent*

Quill R Kukla
Our autonomy can be compromised by limitations in our capacities, or by the
power relationships within which we are embedded. If we insist that real consent
requires full autonomy, then virtually no sex will turn out to be consensual. I ar-
gue that under conditions of compromised autonomy, consent must be socially
and interpersonally scaffolded. To understand consent as an ethically crucial
but nonideal concept, we need to think about how it is related to other require-
ments for ethical sex, such as the ability to exit a situation, trust, safety, broader
social support, epistemic standing in the community, and more.

I. INTRODUCTION

Behind most of our philosophical and lay discussions of sexual consent

lies a loose metaphysical picture: consent is something one gives in an au-
tonomous act of choice. In some versions, consent conceptually includes
the communication of that choice, and in others, it is a cognitive act that
is then communicated. According to the standard picture, if you don’t
have autonomy, due to inherent capacities or coercive circumstances,
then you can’t legitimately give consent, including consent to sex.1
I take autonomy to be a shifting continuum and full autonomy to be
an unreachable and ultimately unhelpful ideal. Our autonomy fluctuates
from context to context, but we always or at least almost always act with
partial autonomy. Compromises to our autonomy are the norm, not the
exception. I will argue that we often legitimately consent to things, includ-
ing sex, while acting with compromised and partial autonomy (for if not,

* I am grateful to my fellow participants at the Villanova workshop on coercion and

consent in March 2019, as well as to audiences at MIT and Brown University, Riki Heck,
Cassie Herbert, and Dan Steinberg, for helpful feedback on earlier versions of this article.
1. Tom Dougherty, “Sex, Lies, and Consent,” Ethics 123 (2013): 717–44; Ruth R. Faden
and Tom L. Beauchamp, A History and Theory of Informed Consent (Oxford: Oxford Univer-
sity Press, 1986); Emily Sherwin, “Infelicitous Sex,” Legal Theory 2 (1996): 209–31.

Ethics 131 ( January 2021): 270–292

© 2021 by The University of Chicago. All rights reserved. 0014-1704/2020/13102-0006$10.00

270
 Kukla Nonideal Theory of Sexual Consent 271
consensual activity would be nearly nonexistent). Of course, I strongly
agree with the standard wisdom that people who have virtually no auton-
omy, because of extreme coercion or incapacitation, cannot consent to sex.
But I will develop and defend a picture designed to make sense of how sex-
ual consent is possible—and, at least as importantly, how it can be fostered
and protected—even under autonomy-compromising conditions. In this
picture, consensual activity is a kind of agential, self-determining activity,
in which everyone involved understands that everyone else is acting in this
agential, self-determining way, and in which everyone involved can and
would stop the activity as soon as it ceased to be agential and self-determining
in this way. This kind of consent may or may not be marked by an explicit
communicative act (although it must be communicated somehow in order
to be legitimate), and it may or may not track an explicit cognitive moment
of choice.
Partial or compromised autonomy can be divided into two kinds.
We can fall short of full autonomy in virtue of our intrinsic capacities, tem-
porarily or permanently. Trauma, dementia, alcohol, exhaustion, some kinds
of disability, and youth may all affect our capacities in this way. Or, we can
fall short of full autonomy in virtue of our external situation. We might be
caught in a power relationship within which we cannot act and choose
freely, or we might be manipulated. We might also be caught in a larger
web of social norms that make it unreasonably difficult for us to act with
full autonomy. Sometimes the compromise of our autonomy might come
from a combination of internal and external factors; for instance, we might
find ourselves in a disorienting context we don’t understand, in which the
possibilities for acting and the meanings and consequences of our actions
are unclear to us. For any of these reasons, our ability to make an autono-
mous decision to have sex might be less than full.
But here’s the issue: We are rarely free of all such compromises. We
almost never have pure, complete decisional autonomy and a perfect
ability to communicate our choices. Our capacities are finite and vulner-
able, and we are all caught up in complex situations that limit and shape
our ability to act and to grasp our own possibilities and their significance.
Our autonomy can be—and typically is—compromised by our relations
of power and subordination to others, our finite and imperfect capaci-
ties, our internalization of and bodily induction into norms that limit
and distort our choices, and the various pressures and dependencies that
make up our situation. Therefore, if we insist that real consent requires
full autonomy, then virtually no sex will turn out to be consensual. Accept-
ing this consequence would either make the notion of consent useless
or turn all sex into rape, and neither is a helpful outcome. Thus, we need
to understand how sex can be consensual, and ethically acceptable, even
when we enter into it under nonideal, substantially autonomy-compromising
conditions.
272 Ethics January 2021
I will note up front that according to the nonideal account of con-
sent that I will develop here, consent, like autonomy, comes in degrees. My
central goal is not to argue that consent is a cleaner or more digital property
than autonomy, but rather that the two are distinct albeit intertwined no-
tions. One can give real consent sufficient to meet the preconditions for
ethical sex without much autonomy, and exercising autonomous choice
is not the same as consenting. My interests here are primarily ethical, not
legal. For legal purposes, if consent is a continuum, we need to be able to
determine whether someone consented enough to sex that it does not
count as rape. I will not take up the question of how to make this legal
determination. But it follows from my argument that having exercised
an autonomous capacity to choose is neither necessary nor sufficient
for meeting this bar.
In what follows, I generally avoid talk of “giving consent.” Elsewhere,
I have argued that it is misleading to understand a typical sexual encoun-
ter as beginning with a request and a granting of this request, which makes
the initiation sound abrupt and one-sided.2 Good sex most often begins
with a dialogical negotiation rather than an abrupt request—a negotiation
that may include invitations, suggestions, reflections on desires and fanta-
sies, and the like, but rarely a flat-out request to which a recipient acquiesces.
In keeping with that argument, I read consensual sex as a kind of collab-
oration rather than a thing to be given to someone else. Moreover, I see
consensuality as a feature of agential activity, in the first instance, rather
than as a feature of a choice that is made at a specific time. As such, con-
sent is not a punctate thing that can be given.
In one dominant philosophical story, valid consent is a chosen act
that turns off someone else’s duty to not do something to the consenter.3
Indeed, Tom Dougherty refers to that picture as “the orthodoxy.”4 I find it
forced and odd to think of people as walking around with duties not to
“do sex” to one another, from which they can be released. This model
makes sex sound asymmetrical, and like something that happens to us
or that we do to others, rather than like something that we actively choose
and do together. It also makes it sound like the default stance toward sex
is rejection. While we may be able to model various consensual and non-
consensual encounters in those terms, I don’t find them helpful for cap-
turing the phenomenological and pragmatic nuances of sexual consent
or its absence.

2. See Rebecca Kukla, “That’s What She Said: The Language of Sexual Negotiation,”
Ethics 129 (2018): 70–97, for a detailed discussion of the limitations of consent talk.
3. Neil C. Manson, “Permissive Consent: A Robust Reason-Changing Account,” Philo-
sophical Studies 173 (2016): 3317–34; Victor Tadros, Wrongs and Crimes (Oxford: Oxford
University Press, 2016); Victor Tadros, “Consent to Sex in an Unjust World,” in this issue,
293–318.
4. Tom Dougherty, “Sexual Misconduct on a Scale: Gravity, Coercion, and Consent,”
in this issue, 319 – 44.
 Kukla Nonideal Theory of Sexual Consent 273
Consensual sex, for my purposes, is sexual activity in which each party
is participating with agency, takes themselves to be doing so, communi-
cates successfully that they are doing so, and can use communication
to successfully stop the activity or any part of it as soon as continuing
would no longer be an expression of their agency.
Roughly, as I understand things here, someone has agency when
they act in ways that express their own values, desires, and purposes,
whether long-standing or fleeting. So agency requires (1) positive capac-
ities to act and (2) action that hasn’t been hijacked by others’ goals and
desires. It does not require that someone actively or consciously chose
their current action, in an independent act of will. Much agential action
is habitual, impulsive, or automated. And, roughly again, we own our ac-
tion when we identify with it and understand it as an expression of our
agency. A central point for me here will be that we can express agency
even under conditions of unequal power and oppression, or when our
capacities for reflective, higher-order reason and choice are somewhat
(though not completely) impaired—that is, under conditions that com-
promise our autonomy. For the purposes of this article, by autonomy I mean
decisional capacity, or the ability to make a voluntary and independent
choice in light of one’s own values, although I know that the term often
gets used more broadly. I specify this narrower meaning here to distinguish
autonomy from agency. Notice, for now, that nothing in this definition of
consensual agency clearly requires full autonomy, or decisional capacity,
per se.5
Consent requires agency, and agency is partial, vulnerable, and con-
text dependent. It is importantly and distortingly abstract to think of
agency as something that someone simply has or doesn’t have in a given
encounter. Rather, sexual partners can prop up or undermine one an-
other’s agency during sex. Agency can be scaffolded or undermined not
just by other individuals and how they treat one another but by material,
institutional, and social contexts. Ethical sex partners cannot control these
entire contexts, but they can be sensitive to the limits of and possibilities for
agency and consent in a given context and adjust accordingly. Ethically
sound consensual sex requires not only that each partner make sure that
the activities are consensual but also that they respect and respond to
one another as centers of desire, purpose, identity, and action, working
to enable and support one another’s agency, and avoiding undercutting it.
My project here is to develop a realistic understanding of consen-
sual sex, which does not make full autonomy a condition for legitimate

5. I will not be providing rigorous explanations of agency, or of autonomy, or of what

it is to own an action here, as doing so is far too large a project for me to take on in this
article; I hope that these rough pictures will give us enough material to productively distin-
guish autonomy from consent and to analyze the relationships between them.
274 Ethics January 2021
consent. I want to understand how consent can work—and how it can be
violated—under the nonideal, messy conditions under which most sexual
encounters actually happen. Consent, I will argue, requires some minimal
measure of autonomy. But full autonomy is not necessary for consent, nor
is any kind of autonomy ever sufficient for consent. There has been recent
interest in an “enthusiastic consent” model of sexual consent.6 This is the
fairly self-explanatory idea that legitimate sexual consent requires enthusi-
asm. But I think it is important that we can legitimately consent to sex
about which we feel ambivalent, or that we are only agreeing to in order
to please a partner, for instance. We can consent to sex when we are not
really in the mood, just as we can consent to watch a movie we are not es-
pecially up for. Asexual people and sex workers have resisted the idea that
they can’t give legitimate consent to sex despite a lack of felt giddiness
about the whole affair.7 Conversely, if I am arguing that autonomy is not
always necessary for consensual sex, then I certainly need to reject the idea
that autonomy plus enthusiasm is necessary for consensual sex. Thus, I
claim, autonomy plus enthusiasm is also neither necessary nor sufficient
for sexual consent. So the “enthusiastic consent” movement is going in
quite a different direction than I am.
I now turn to examples of impaired autonomy, arising from com-
promised capacities or from a compromising situational context. What
I want to examine is how we can scaffold agency and enable legitimate
consensual sex—understood as the kind of agential sex I described just
above—under conditions of partial autonomy. Likewise, I want to explore
how partners and circumstances can fail to properly scaffold agency when
someone has impaired autonomy. The goal is to understand how good,
properly supported, consensual sex can be enabled and sustained in the
nonideal, imperfect world in which we live as vulnerable and finite beings
enmeshed in power relations.

II. CONSENSUAL SEX WITH INTRINSICALLY COMPROMISED

CAPACITIES: THE CASE OF DEMENTIA

Our ability to make and communicate self-reflective, informed, free de-

cisions to participate in sex can be limited by compromises in our cogni-
tive capacities, whether temporary, permanent, or fluctuating. People with
various forms of mental illness may have fluctuating capacities, and people

6. See, e.g., Jaclyn Friedman and Jessica Valenti, eds., Yes Means Yes! Visions of Female
Sexual Power and a World without Rape (New York: Seal, 2019); and Gaby Hinscliff, “Consent
Is Not Enough: If You Want a Sexual Partner, Look for Enthusiasm,” Guardian, January 29,
2015.
7. James A. Roffee, “When Yes Actually Means Yes: Confusing Messages and Criminal-
ising Consent,” in Rape Justice: Beyond the Criminal Law, ed. Anastasia Powell, Nicola Henry,
and Asher Flynn (Basingstoke: Palgrave Macmillan, 2015), 72–91.
 Kukla Nonideal Theory of Sexual Consent 275
with early or midstage dementia may have fluctuating or permanent com-
promises in their capacities. Alcohol and drugs can temporarily compro-
mise capacities. Developmental disabilities can do so permanently. It is
often treated as a social truism that people with such compromises can-
not consent to sex. For example, it is frequently treated as obvious that
drunk or high people cannot consent. Someone who is incoherently black-
out drunk almost certainly doesn’t have enough awareness or self-control
to have consensual sex, but in fact a huge swath of perfectly normal, fun
sex happens after a few drinks or tokes. While such situations raise ethical
and social challenges and risks, it is unrealistic to count all such sex as
nonconsensual, and hence in effect rape.
I want to focus on the challenging and murky case of people with
moderate dementia, many of whom live in long-term care facilities, and
the question of whether and how we can understand them as able to have
consensual sex. My goal in this section is not to argue for or against a par-
ticular vision of what consent with dementia looks like. Rather, here I want
to use dementia as an example of a condition that compromises one’s ca-
pacity for autonomous decision-making and then tease out just how murky,
complex, and situationally dependent issues of consent are in such non-
ideal conditions.
In 2015, a seventy-eight-year-old man was arrested for having sex
with his wife in her nursing home. Her doctors had declared her incapa-
ble of consensual sex because of her dementia. By everyone’s report they
had a loving relationship with a long history and no signs of abuse. She
had very little discursive ability left but always manifested pleasure and af-
fection when her husband showed up.8 It is impossible to know, from a
newspaper story, what exactly happened, and there is no point in specu-
lating from a distance about the details of their private encounter. I want
to think about what in this story might be conducive to or undermining of
the possibility of consensual sex. We know that elderly people with de-
mentia crave and take pleasure from touch, including sexual touch, long
after they have lost other desires and pleasures. Indeed, a significant num-
ber of dementia patients become sexually aggressive and demanding. More-
over, sex later in life is correlated with lower rates of depression, more self-
care, and higher markers of general well-being. Dementia is associated
with an increase in libido in 14 percent of patients, and with a decrease
in almost none.9 While experiencing sexual desire and benefiting from
sex certainly do not add up to consenting to sex, this all seems to be pow-
erful prima facie evidence that enabling people with dementia to have
8. Pam Belluck, “Sex, Dementia, and a Husband on Trial at Age 78,” New York Times,
April 13, 2015.
9. Evelyn M. Tenenbaum, “To Be or to Exist: Standards for Deciding Whether Demen-
tia Patients in Nursing Homes Should Engage in Intimacy, Sex, and Adultery,” Indiana Law
Review 42 (2009): 675–720.
276 Ethics January 2021
consensual sex if possible is important to their well-being. And yet the ma-
jority of nursing home staff report that it is not important for the residents
to have the opportunity to remain sexually active.10 The standard view of
nursing homes and their staff is that sexual activity among residents is a
legal risk and probably nonconsensual, and hence best prevented; there
are a handful of explicitly sex-positive nursing facilities, but they are rare
and contentious.11
Nursing homes are typically places where standard intimate connec-
tions have been severed or lost and touch is highly regulated. Many resi-
dents come to nursing homes following the death of a longtime spouse or
partner. When nursing home residents with dementia try to start sexual
relationships with one another, these are typically highly discouraged
and institutionally interrupted, including through physical and spatial
mechanisms, such as doors that can’t be locked and a pointed lack of
access to private spaces. Many nursing homes are run by religious organi-
zations with sex-negative and sex-shaming commitments and attitudes.
Thus, the entire nursing home environment is in an antagonistic relation-
ship to its residents’ need for sexual intimacy. Moreover, our more wide-
spread cultural norm of treating elderly people as generally asexual, and
indeed treating the entire topic of sex among the elderly as disgusting
and off-limits, contributes to an environment in which elderly people have
few hermeneutic resources for understanding and interpreting their own
sexual desires and sorting the healthy from the dysfunctional ones. This
in itself impedes their sexual agency.
In an antagonistic environment of this sort, it is harder for sex with
someone with fluctuating and marginal capacities to actually be consen-
sual, I want to suggest. Consensual sex, as I have delineated it, is possible
only when the activity is agential and self-determining, and everyone in-
volved recognizes this, and everyone can communicate successfully if
they want to stop all or part of the activity, and any such communication
will be respected. Under the best of circumstances, a frail person in an
institution with moderate dementia needs a lot of social scaffolding in or-
der to meet these conditions. They need to be around people and an en-
vironment that keeps them safe, makes communication of their needs
and desires as easy as possible, and offers them easy and straightforward
ways to exit any activity or encounter. When the only way to have sex is in
secret and in hiding, under threat of being found out and “punished,” it
is impossible to get the kind of social support that would make sex safe and
that would provide a comfortable and easy place to go for redress if some-
thing goes wrong. Stress enhances cognitive deficits, so residents who are
sneaking around and breaking rules are also not likely to be at their most

10. Melissa C. White, “The Eternal Flame: Capacity to Consent to Sexual Behavior
among Nursing Home Residents with Dementia,” Elder Law Journal 18 (2010): 133–58.
11. White, “Eternal Flame.”
 Kukla Nonideal Theory of Sexual Consent 277
competent. There may well be no one trustworthy to go to with questions
or concerns if they have any. If they are already hiding their sexual activi-
ties, it may be harder for them to exit a sexual situation or relationship
if they want to, especially if there are also mobility issues involved.
So the typical nursing home undermines its residents’ capacity for
consent and is set up to prevent sexual encounters. But can someone with
moderate dementia have consensual sex, under the right conditions? We
have every reason to think that often sex is really what they desire. Of
course, they are also particularly vulnerable to abuse and manipulation.
But given their real desire for sexual contact and the benefits of having it,
it would be good to figure out conditions under which they can have con-
sensual sex, rather than just focusing on preventing it.
We can’t know what exactly went on between the seventy-eight-year-
old man and his wife. But there is every reason to think that a long-term,
caring partner will be best able to detect subtle signs of desire, discom-
fort, agitation, and pleasure in someone whose ability to communicate
and remember are compromised, and that that person will be most likely
to care about their partner’s well-being and getting it right. We would
hope that he would be ready to stop immediately if she showed any sign
of fear, discomfort, or disengagement. There is also every reason to think
that someone is more likely to be participating willingly and agentially in
sex if it is with someone with whom they have chosen to have sex many
times in the past. Consensual sex expresses the agency of everyone in-
volved and requires ongoing effective communication during the activity
itself. A longtime couple has lots of practice in this joint, dialogical agency.
By doing something together that they have done many times, they are typ-
ically each better able to act with agency.
Of course, it is completely possible, as we know all too well, for hus-
bands to impose sex nonconsensually on their wives. Elder abuse is also a
huge problem, and it’s possible for a husband of a person with dementia
just to have his way with her, rather than being responsively and respon-
sibly engaged in supporting her agency. But given what we do know about
elderly dementia patients’ general desires for sexual intimacy and its con-
nection to their well-being, in a case where the sexual activity is continu-
ous with their past agential action, we should not presume that such sex is
nonconsensual. We should also recognize that it would be easier for such
sex to be consensual if it didn’t have to be snuck on the sly, under fear of
shaming and punishment if it is discovered.

III. CONSENSUAL SEX WITH SITUATIONALLY DIMINISHED

AUTONOMY: THE CASE OF PATRIARCHY

Our autonomy may be compromised by our external situation, even if we

are intrinsically capable of autonomous decision-making. This applies to
278 Ethics January 2021
our ability to autonomously choose to have sex as well. Most obviously,
someone may be forced or threatened into sexual activity, in which case
the activity is unequivocally nonconsensual. But there are more interest-
ing cases for our purposes, in which autonomy may be partial but legiti-
mate consent may still be possible.
There are all sorts of benign reasons why it might be impossible to
make a fully informed autonomous choice about sex. You might be decid-
ing whether to try something very new and different, and you might be
unsure how you will feel about it—proceeding might involve a leap into
the unknown. Consider a first same-gender sexual experience, or an ex-
ploration of an unfamiliar kink, for example. Sometimes you might find
yourself in a disorienting setting, and the uncanniness of the surround-
ings may throw off your sense of control.
Power differentials between potential sex partners can undercut au-
tonomy, even if no one is abusing their power. These differentials might
come from institutional roles (such as advisor and student), a large age or
status difference, or a difference in social stability (such as a citizen and a
refugee), for instance. It is not clear at what point having sex with some-
one who has less power becomes wrongdoing. Moreover, even if we think
that a sexual relationship across a power differential is ethically vexed and
constitutes wrongdoing on the part of the more powerful partner, this is
quite a bit weaker than saying that it is nonconsensual, which is a very
strong charge indeed. It is all too easy to just condemn sex across power
differentials because of their impact on autonomy. However, this would
rule out basically all sex, since it is rare for two people to have exactly
equal social power. So we need to be able to say something more nuanced
about these cases, and we need to understand how we can enable and sup-
port the possibility of consensuality in the context of power differentials.
That is, if we are to understand how actual, messy, nonideal sex works, we
need to understand how and when it can be ethically acceptable despite
power differences. And more specifically, if we are to understand how
nonideal consent works, we have to make room for the fact that consen-
sual sex can happen even when there are pressures and inequalities that
constrain our options, impose unfortunate risks on us, and make our
choices less than fully autonomous. This is difficult to think about, but
it is essential, given the reality of the dense and inescapable web of power
relations we all live within.
Some unequal relationships are direct and asymmetrical enough,
with enough potential for harm, that we might feel comfortable saying
that there really just is no room for consensual sex. The risks from turn-
ing the more powerful person down, or exiting a relationship, may be just
too high and too concrete. This might be so no matter how much the
more powerful person promises not to use this power, and no matter
how many other safeguards are in place. I am tempted to think that this
 Kukla Nonideal Theory of Sexual Consent 279
is true for faculty members and their graduate student advisees, for in-
stance. But there are much less clear cases in which there are still stark
power differentials, such as a faculty member and a grad student in the
same discipline but at a different school perhaps. Here there may be sex-
ual encounters that are under the shadow of possible exercises of power,
or desires distorted by the lure of status. We may not think that these en-
counters are a great idea, but I am loath to say that such encounters are
automatically nonconsensual.
I now want to turn to what is perhaps the most difficult and pervasive
example of sex across power differences, namely, heterosexual sex under
patriarchy. Similarly to the previous section, my goal in this section is not
to answer the question of when sex under patriarchy is consensual, but
rather to show the messy complexity of the question under the nonideal
circumstances that patriarchy creates. Not only is there a standing power
differential between men and women, but there are also all sorts of rea-
sons to think that women’s autonomy—and especially our sexual auton-
omy—is substantially diminished under patriarchy, particularly but not
only in our interactions with heterosexual men. As has been discussed
and explored in depth, patriarchy diminishes our ability to effectively re-
fuse sex.12 It impedes our credibility as speakers and reporters of our ex-
perience,13 especially when it comes to rape and sexual assault.14 Growing
up under patriarchy involves developing attitudes and habits that are tan-
gled deep within a pervasive set of social norms that train us to be passive
about sex and to be quiet about our own sexual desires. It generally dis-
courages us from being assertive and from thwarting men’s social de-
mands. And indeed, when we resist men’s attempts to interact with us,
we lay ourselves open to violence and verbal abuse. Patriarchy is funda-
mentally structured by norms that demand that women serve men’s needs,
and sexuality is the paradigmatic arena where such servitude is demanded.15
Patriarchy undermines women’s self-trust, which directly harms our abil-
ity to have reflective control over our actions.16 Many feminists have argued
that women’s sexual subjugation to and domination by men is built into

12. Rae Langton, “Speech Acts and Unspeakable Acts,” Philosophy and Public Affairs 22
(1993): 293–330; Catharine A. MacKinnon, Feminism Unmodified: Discourses on Life and Law
(Cambridge, MA: Harvard University Press, 1987).
13. See, among many other sources, Miranda Fricker, Epistemic Injustice: Power and the
Ethics of Knowing (Oxford: Oxford University Press, 2007).
14. Kristie Dotson, “A Cautionary Tale: On Limiting Epistemic Oppression,” Frontiers:
A Journal of Women Studies 33 (2012): 24–47; Katharine Jenkins, “Rape Myths and Domestic
Abuse Myths as Hermeneutical Injustice,” Journal of Applied Philosophy 34 (2017): 191–205.
15. Kate Manne, Down Girl: The Logic of Misogyny (Oxford: Oxford University Press,
2017).
16. Kate Abramson, “Turning Up the Lights on Gaslighting,” Philosophical Perspectives
28 (2014): 1–30.
280 Ethics January 2021
the conceptual essence of “women” as a gender.17 Whether or not one
accepts this strong conceptual claim, it is hard to deny that patriarchy has
organized social roles around gender subjugation in deep ways. I recap all
of this here not to lecture my readers on basic Feminism 101 but to high-
light the main ways in which patriarchy specifically works along multiple
axes to curtail and undermine women’s sexual autonomy.
But this implies that if we believe that sexual consent always requires
full decisional autonomy, then women could never consent to heterosex-
ual sex under patriarchy. In effect, all heterosexual sex would be rape.
This is, of course, pretty much what some second-wave feminist theo-
rists, such as Andrea Dworkin, have argued.18 If we presume a traditional
model linking the capacity to consent directly with full decisional auton-
omy, it is a conclusion that’s hard to avoid, I believe. But I take this con-
ditional to be a reductio: any view of sexual consent according to which
the mere fact of patriarchy precludes women’s consent to heterosexual
sex is on the face of it the wrong view of consent. Otherwise, the notion
of rape becomes nearly useless and unprosecutable, for one thing. For
another, it is unacceptably insulting and paternalistic to tell straight and
bisexual women that none of the sex we are having with men, including
the sex we are actively seeking out, is consensual. Finally, for all the many
problems and pitfalls surrounding heterosexual sex, there’s plenty of
good, happy, wanted, fulfilling heterosexual sex that actually happens;
it’s just not useful or illuminating to say that all such sex is rape even though
it is welcomed and delightful. Indeed, it makes a mockery of the serious
wrong of rape to do so. Thus, we need to understand how legitimate consent
to heterosexual sex under patriarchy is possible for women. More generally,
we need to understand how consensual sex is possible across and within
power relationships and webs of autonomy-limiting norms.
Whatever the answer turns out to be, it’s crucial that we recognize
that it is not just going to be a matter of an individual man choosing
not to avail himself of the power and privileges given to him by patriar-
chy. That an individual man decides that he will take no for an answer
without responding abusively, or that he will attend and be responsive
to a woman’s expressed desires and discomforts, is not enough to re-
store her autonomy to her. Women are trained over the course of years
to acquiesce to men’s requests and to fear hurting their feelings. This is
embedded in our bodies. Many of the forces that reduce our sexual au-
tonomy are structural, not individual. Men cannot suspend the force
of patriarchy at an individual level. Choosing not to avail yourself of
power doesn’t make that power go away.

17. See, e.g., Sally Haslanger, Resisting Reality: Social Construction and Social Critique
(Oxford: Oxford University Press, 2012).
18. Andrea Dworkin, Intercourse (New York: Free, 1987).
 Kukla Nonideal Theory of Sexual Consent 281
When we think about the many ways in which patriarchy undercuts
women’s agency, especially in the sexual arena, it can feel hopeless to
hold on to a vision of legitimate heterosexual consent. No matter how
much scaffolding is in place, we will never undo the fact that we are all
embedded within concrete norms that create a serious power asymmetry
between men and women. Here, I think it is illuminating to consider a
different domain in which legitimate consent is a crucial and much-
discussed ethical precondition for interaction, namely, medical treatment.
In bioethics, how to enable and ascertain informed consent before inter-
vening on someone’s body is a keystone topic. In this domain, however,
there is no confusion over the fact that in medical settings, doctors and pa-
tients do not have symmetrical power. Doctors have more institutional, so-
cial, and cognitive authority in the medical setting than do patients. They
also have the ability to grant or withhold important goods and the power to
gravely harm or substantially help the patient; neither of these powers is
reciprocal. Moreover, patients often go to doctors for help at times of
heightened vulnerability. All this is basically unavoidably built into the
structure of the encounter. This is so even when the patient is an otherwise
highly authoritative and agentially empowered person.19 This power imbal-
ance can indeed easily undercut patients’ ability to act as self-determining
agents capable of legitimately giving or withholding consent to treatment.
However, bioethicists take it as given that the way to promote the pos-
sibility of legitimate consent and refusal is not to erase or overcome these
power differentials but to find ways of scaffolding patients’ agency within
them. We build decision tools and institutional checks and balances; we
think about where, with whom, and when patients should make deci-
sions, with the help of what kinds of conversations. Patients, as a matter
of near necessity, make choices and agree to treatment paths under non-
ideal conditions. None of this scaffolding works perfectly. But the con-
versation in the medical domain is about how to build up agency despite
power inequities and situational forces that chip away at autonomy, not
about how to get rid of these things and enable perfect autonomy. Intu-
itively, none of us think that power imbalances and vulnerabilities in the
medical domain make legitimate consent impossible. I am suggesting
that we take a similar approach to thinking about sexual consent under
the nonideal conditions of patriarchy and other situational factors that
undermine full autonomy.20 People with memory problems, mildly drunk

19. The play Wit (Edson, Margaret, Dramatists Play Service, Inc., 1999) explores this
dynamic powerfully.
20. Past a certain point of compromised autonomy, of course, we don’t think that pa-
tients can consent on their own behalf. A patient with advanced dementia, or an infant, or
someone in a persistent vegetative state has no meaningful autonomy and cannot consent.
In the medical domain, at that point we assign a proxy or surrogate decision-maker. This is
282 Ethics January 2021
people, people out of their element, and people at a power disadvantage
should not be assumed to be incapable of real sexual consent. At the same
time, we should recognize that we need to take steps to enable and support
that capacity to consent. I address the question of what those steps may be
in the next section.

IV. RELATIONAL AGENCY AND SCAFFOLDED CONSENT

I am finally in a position to try to articulate more directly and systemati-

cally how we can scaffold sexual agency and the capacity to consent, un-
der nonideal conditions in which decisional autonomy is compromised
owing to internal or situational reasons (or both).
Many feminist theorists have developed versions of “relational au-
tonomy,” all of which are variations of the insight that autonomy is not
a property that inheres in individuals, but rather an enacted capacity that
is essentially supported by other people and by the environment. Catri-
ona Mackenzie helpfully explains,

Relational conceptions hold that autonomy is a complex compe-

tence, the development and exercise of which requires ongoing in-
terpersonal, social, and institutional scaffolding. Relational theo-
rists seek to analyze the nature of this competence; to understand
the kind of social scaffolding required to develop and exercise it;
and to show how autonomy can be thwarted by social oppression
and injustice. . . . While emphasizing the crucial role of social rela-
tionships in shaping our self identities and developing the skills in-
volved in autonomy competence, relational theorists argue that
some social relationships, such as those characterized by domina-
tion, abuse, coercion, violence or disrespect, provide hostile condi-
tions for autonomy.21

While these accounts seek to broaden the term ‘autonomy’ so that it be-
comes a relational capacity, for my purposes, what they are describing is
better framed as a relational theory of agency. Remember that, by stipu-
lation, I reserve ‘autonomy’ for the kind of individual decisional capacity
that is more traditionally seen as essential for consent. I also don’t need
to sort through the details of the wide variety of accounts of relational au-
tonomy here. What I want to take from these accounts is a picture of how
other people and well-designed institutions and environments can build

an important disanalogy between the two cases. In the sexual domain, there are never any
circumstances in which consent should be given on behalf of someone else, although, hor-
rifyingly, that has been proposed in the case of dementia patients. See, e.g., Tenenbaum,
“To Be or to Exist,” which rightly rejects any such proposal.
21. Catriona Mackenzie, “Autonomy,” in The Routledge Companion to Bioethics, ed. John
D. Arras, Elizabeth Fenton, and Rebecca Kukla (New York: Routledge, 2014): 277–90, 285.
 Kukla Nonideal Theory of Sexual Consent 283
up and hold in place people’s agency, understood as their ability to ex-
press and enact their desires, maintain their integrity and sense of self,
and determine their own activities and narratives.
The imaginative path into a relational understanding of agency that
I have found most helpful is Hilde Lindemann’s concept of holding in per-
sonhood.22 Her core idea is that many of the roles, identities, and activities
that are most central to our sense of self and our integrity require social
and material support and participation; when we become less able to en-
act these things for ourselves, other people can take over some of the la-
bor of maintaining us in our identities. She writes, “Even as none of us
can form a personal identity without the help of many others, so none
of us can maintain our identities all by ourselves. Many self-constituting
social roles, for example, require others to take up reciprocal roles with
respect to them: I can’t be a mother if I don’t have a child. To be held in
my identity as a mother, my family and my friends and co-workers need to
interact with me as someone who occupies that role.”23 In this sense, all of
us need to be held in our identities through the uptake of others. But in
particular, as our competence or our ability to hold onto ourselves wavers
and becomes vulnerable, whether this is because of our internal capaci-
ties or our situation, others need to step up and take over more of the la-
bor of holding in selfhood for us: “Torn out of the contexts and condi-
tions in which we can maintain our own sense of ourselves, we run the
risk of losing sight of who we are—at least temporarily—unless someone
else can lend a hand.”24
This does not mean that others should act for us or instead of us. For
someone to act in place of me is to co-opt my identity rather than to hold
me in it. Rather, others help me to act as myself, through the kind of up-
take and recognition they give me. Lindemann focuses mostly on people
who are in medical crisis or have dementia. She gives examples of every-
day forms of recognition and maintenance: picking out a record for some-
one that accords with their tastes and adds to their beloved collection, car-
ing for and giving them updates about a pet that is dear to their heart and
entwined with their identity, giving them a manicure or a haircut, or call-
ing them by their professional title. These are small acts of self-care and
self-maintenance that others can do for us. There is a slippery line be-
tween holding someone in personhood and imposing an understanding

22. Hilde Lindemann, “Holding on to Edmund: The Relational Work of Identity,” in

Naturalized Bioethics: Toward Responsible Knowing and Practice, ed. Hilde Lindemann, Marian
Verkerk, and Margaret Urban Walker (Cambridge: Cambridge University Press, 2009), 65–
79; Hilde Lindemann, “Holding One Another (Well, Wrong, Clumsily) in a Time of De-
mentia,” Metaphilosophy 40 (2009): 416–24; Hilde Lindemann, Holding and Letting Go:
The Social Practice of Personal Identities (Oxford: Oxford University Press, 2016).
23. Lindemann, “Holding on to Edmund,” 72–73.
24. Lindemann, “Holding on to Edmund,” 74.
284 Ethics January 2021
of them on them; good holding has to be truly generous and subtly re-
sponsive to the past and present reality of a person. You can see this dis-
tinction clearly when you see people interacting with family members
with dementia, for instance. You can watch some family members sloppily
impose a sense of self on the person with dementia (“YOU LIKE THE
DANCING, DON’T YOU, GRANDMA?” shouted in grandma’s general
direction), while others offload some of the difficult, subtly responsive la-
bor of maintaining that person’s identity onto themselves.
Whereas Lindemann speaks of our holding one another in our iden-
tities, I am interested in the somewhat more local project of holding each
other in agency. Just as I can help someone to be themselves by holding
them in their identity, I can also help someone act as themselves, and to
express themselves in action, especially when their own capacities and cir-
cumstances are insufficient for doing this on their own. Holding some-
one in their agency or their identity is a nonappropriative way of letting
them be, not by leaving them alone but by actively enabling them to be.
Holding in agency can directly contribute to holding in identity, but it
might be a smaller-scale project.
Nursing home facilities, as we saw, often provide no private spaces or
time alone, disorient their residents by failing to create a homelike and
familiar environment, and presume that sex is a pathological sign of dys-
function rather than an expression of agency. Such institutions will be ill-
equipped to sustain someone’s agency, particularly if sexual intimacy is
what they really seek and is in keeping with their values and sense of self.
Any sex they have will be furtive and risky, and their actions will be medi-
calized if discovered, which is pretty much the exact opposite of helping
integrate them into agency. Likewise, a culture that brands women who
want sex as “sluts” who must have some sort of ulterior motive, or which
takes women who enjoy sex as lacking in self-respect, inherently under-
mines their positive sexual agency. A culture and justice system that pre-
sume that nonconsensual sex is rare and only really happens to women
who did something wrong will automatically be unable to give competent
uptake and support to women who are trying to express their agency
through sex.
A great deal of holding others in their agency is a matter of giving
them competent uptake. Kristie Dotson has explored in detail the kind
of harm we can do to people when we do not give reliably competent up-
take to what they say and express.25 When we think about incompetent
uptake and epistemic injustice in the sexual domain, we tend to focus on
the case of men’s incompetence at hearing and understanding women’s

25. Kristie Dotson, “Tracking Epistemic Violence, Tracking Practices of Silencing,”

Hypatia 26 (2011): 236–57.
 Kukla Nonideal Theory of Sexual Consent 285
refusals of sex or attempts to exit sex. This is indeed an important ex-
ample of pernicious incompetence, when it occurs.26 But, especially when
someone’s autonomy is compromised or vulnerable, or when there are
power relations at work, competent uptake has a much larger role to play.
A partner can protect a vulnerable partner’s agency by giving competent
uptake across the board: being a good, skilled, and caring interpreter of
their expressions of pleasure, pain, desire, embarrassment, comfort, fear,
and the like. This means that one and the same person—someone with
dementia, for instance—may well be able to have consensual sex with one
partner, who has these skills and competencies and is committed to exercis-
ing them, and not with another. Thus, like agency, consent is also a relational
capacity.
When it comes to an intimate encounter, holding someone in their
agency involves, among other things, being competently responsive to
their gestures and their nascent expressions, while also respecting and
understanding who they are, what they value, and so forth. It may quite
literally involve feeling and seeing what their body is doing and respond-
ing appropriately. At the same time, partners cannot generally hold peo-
ple in agency all on their own. As we have seen, the social, institutional,
and material surroundings may help or hinder holding. As Lindeman
so beautifully puts it, “It’s not only other people who hold us in our iden-
tities. Familiar places and things, beloved objects, pets, cherished rituals,
one’s own bed or favorite shirt, can and do help us to maintain our sense
of self.”27
When people are free from any significant situational or internal im-
pediments to their autonomy, they may choose to participate in a sexual
activity that is substantially out of step with their prior experiences and
desires. Sexual exploration is often an important part of sexual flourish-
ing—of figuring out the space of our possible pleasures. Sometimes we
try something and don’t like it, which is part of why it’s so important to
have clear and reliable exit conditions in place. But there’s no reason our
past tastes and activities should determine our present choices. However,
when someone’s autonomy is compromised, it is not the time for exper-
imenting with edgeplay. One of the main ways in which we can hold
others in their agency when that agency is fragile is by being richly respon-
sive to whether the desires they are expressing are in keeping with their
patterns, values, and sense of self. It behooves partners of vulnerable

26. Although, interestingly and depressingly, there is good empirical evidence that
men only feign such incompetence along the path to overriding women’s refusals by coercive
force. See, e.g., Rachel O’Byrne, Mark Rapley, and Susan Hansen, “You Couldn’t Say ‘No,’
Could You? Young Men’s Understandings of Sexual Refusal,” Feminism and Psychology 16
(2006): 133–54. Many thanks to Riki Heck for pointing me to this body of empirical literature.
27. Lindemann, “Holding One Another,” 263.
286 Ethics January 2021
and compromised people not to exploit what might be a passing impulse
that is not grounded in that person’s self.
It is important and surprisingly rare to remember that there is no
general reason to assume that we better protect someone’s agency by
not having sex with them than by having it. (This is part of why the idea
that sexual consent is a release from a standing duty not to have sex with
them sits poorly with me.) It is completely possible that someone may
genuinely want a sexual encounter, and that that encounter may be a
good expression of who they are, their values, and the kind of thing they
enjoy, even if they are somewhat drunk, or have an impaired memory, or
have limited reflective reasoning skills, or if they have less power than
their partner, or are stuck within a system of misogynist norms that dis-
count their sexual pleasures. A good partner may use responsive uptake,
a rich understanding of their partner’s past, a suitable material setting,
clear exit conditions, and the like in order to help give them the stability
and safety they need to express this agency. Conversely, someone who de-
means or humiliates their partner, or insults their intelligence or self-
esteem, does not thereby directly undermine consent (although these
are other ways in which sex can be unethical). But they do risk chipping
away at their partner’s ability to act with agency, by undermining things
like self-trust and a stable sense that one’s own desires are worth express-
ing and protecting. These are ways of failing to hold the other in agency.
Now that we have a reasonably rich, socially embedded picture of
agency on board, we can extract some of the things that prop up and scaf-
fold agency and the possibility of consent from our discussion so far and
list them a bit more systematically. Many of the tools that can help do
this are also components of ethical sex more generally—components
that do not reduce to autonomous choice or consent. I claim that these
can all work in tandem, or in any combination. It is important to note that
these are not a list of necessary and sufficient conditions for consent. On
the contrary, my picture is much messier than this. These are ways in
which we can scaffold consent and help make it possible. It is precisely
when some of them are lacking that we need more of the others. Notice
that some of them depend on partners and others depend on broader
social institutions.

1. Background trust between everyone involved—trust that every-

one cares about the other’s safety, desires, boundaries, and com-
fort; that everyone is alert to everyone else’s expressions and
communications, including those that are nascent or subtle, and
ready to give them competent uptake; and that beyond the bound-
aries of this encounter everyone will protect one another’s well-
being and respect their privacy and integrity. Notably, how much
trust and the type of trust involved depend on the context and
 Kukla Nonideal Theory of Sexual Consent 287
the players. A one-night stand depends on a different kind of
trust than sex within an extended relationship. Trying some-
thing emotionally challenging about which you feel ambivalent
requires a different kind of trust than does having familiar sex
with a longtime partner. When power between partners is unequal,
this power imbalance cannot be removed by the more powerful
partner sincerely promising not to exercise that power. But a long
track record at being trustworthy at not exercising it can help
shared, dialogical agency to flourish.
2. The concrete ability to exit an activity at will, without recrimina-
tion or extended negotiation, in a way everyone involved can
immediately recognize. We all have more agency when we are
provided with direct, reliable tools for exiting a situation when
we want to. For instance, there’s a big difference between a male
partner telling me, “Don’t worry, I won’t do anything you don’t
want me to do,” with the full social weight of his ability to do other-
wise behind him, on the one hand, and our establishing a safe
word or another easy and unambiguous way to stop the action
between us, on the other.
3. Competent uptake from each partner, including skills at under-
standing and responding to what each is communicating, and
the ability to read bodily responses and adjust and fine-tune
on the fly. While I am a firm believer that one-night stands and
casual hookups can be lovely and ethically sound, a couple that
develops and establishes competent communication over a
long period of time can consent to a wider and more boundary-
pushing range of activities with one another.
4. A broader social context that does not undermine the agency of
people who engage in the sexual activities at hand. Slut shaming
and strict norms controlling who we may have sex with and how,
for example, remove the external scaffolding that supports agency
within sex.
5. Avoidance of activities that are agency undermining in virtue of
their content. Sexual activities should not be degrading, gas-
lighting, or objectifying, for example. Degrading or objectifying
role-play may be parts of consensual, ethical sex if everyone in-
volved wants them to be, but actual dehumanization is incom-
patible with supporting agency.
6. Meaningful epistemic agency, not just in the sense that partners
must listen to and believe one another leading up to and during
sex, but in the sense that everyone must have basic testimonial
credibility outside this encounter as well. People need to be be-
lieved, within reason, when they report that sex was nonconsen-
sual. If they don’t have the ability to meaningfully report violations
288 Ethics January 2021
of consent, this undermines the possibility of consensuality. Con-
versely, someone with reduced capacities who plausibly reports
that sex was consensual needs to be believed as well. People need
to be heard and believed, according to the norms of decent episte-
mic practice. The #MeToo movement has not undone patriarchy,
but arguably one important point of it is to help build up the pos-
sibilities for victims of sexual assault to be competently heard and
recognized (even if not legally), pushing back against the system-
atic epistemic injustice that is perpetrated by the myth that sexual
assault is rare and that its victims are unstable and abnormal. This
serves as a tool for supporting agency within sex, by undoing some
of the gaslighting that surrounds sexual violence.
7. Relatedly, in order to participate consensually in sex, people
need redress if their consent is violated. They need to be able
to hold their partners accountable in meaningful ways for not
respecting their agency.
8. The last several points depend on everyone involved being so-
cially connected—having a support network, a check on reality,
and a community that holds people accountable, outside of the
couple. Accordingly, a sexual partner who uses their power to
socially isolate a partner undercuts their agency.
Remember that I am trying to build a nonideal theory of consent.
The above list does not form necessary or sufficient conditions for legit-
imate consent. Consent might be legitimate with some but not all of these
conditions fully in place, or it might fail to be legitimate with all of them
in place. Under messy, autonomy-compromising, nonideal conditions,
these are some of the tools and conditions that can help to scaffold
and enable consent, especially when other things from the list are lack-
ing. My list of possible scaffolds is not intended as an exhaustive list.
The capacity for consent is not all or nothing, according to the account
I have been building. Rather, it is bolstered or undermined by these sorts
of conditions. When decisional autonomy is compromised, other forms
of scaffolding become especially important.
In an earlier article, I argued that consent is just one dimension of
ethical sex, but one that has been given an inflated pride of place and
made to do too much work in our discussions about the ethics of sexual en-
counters.28 Items on the above list such as trust and epistemic justice can
bolster the possibility of legitimate consent, but they are not themselves
conceptual components of consent. If consensual sex is sex that expresses
and successfully communicates the agency of all involved, then things
like trust and epistemic justice are important background conditions that

28. Kukla, “That’s What She Said.”

 Kukla Nonideal Theory of Sexual Consent 289
can support and protect that agency and help enable good communica-
tion. Consensuality ensures that our sexual activities express our own
agency, rather than being imposed on us by others. That a sexual encoun-
ter is consensual also leaves open other important ethical questions about
it, such as whether it is degrading; whether it was initiated for good rea-
sons; whether it constitutes a betrayal; whether it will put one or more
partners at unacceptable risk; whether it was negotiated in a way that
was respectful, welcoming, and free of shame and other harmful emo-
tional weapons; and whether it is complicit in oppressive social norms.

V. MINIMAL DECISIONAL AUTONOMY

I have been arguing that consensual sex is compatible with compromised

autonomy, in cases where the person’s agency is held in place and sup-
ported by their partner and by the situation. I have also on several occa-
sions dropped a reminder that if someone really has little or no autonomy,
then consensual sex is impossible. So how much autonomy is “enough”?
This, of course, depends on the other forms of support I have been discuss-
ing, but we can still think in general terms about what sort of partial auton-
omy is minimally required for consensual sex to be possible. By now it
should be clear that this minimum bar will be a necessary but not a suffi-
cient condition for the possibility of consent.
Agnieszka Jaworska has proposed an account of “minimal auton-
omy” designed for the medical domain. According to her picture, an agent
has minimal decisional autonomy sufficient for informed consent in a
medical context if (1) the attitude that guides the decision properly repre-
sents the agent’s self, (2) the agent acts in light of reasons prescribed by
this attitude, and (3) the agent is open to a fresh perception of reasons that
push against theirs.29 She argues that these needn’t be interpreted in any
intellectualist sense that requires high-level discursive cognitive compe-
tencies or control. The first condition basically just says that the choice is
not directly manipulated, coerced, or extracted by someone else. The sec-
ond (which I think she phrases in language much more cognitivist than
she needs or wants for her purposes) says that the action flows from the
agent’s own motivations. The third, which is the most cognitively demand-
ing perhaps, captures the requirement that the agent has enough norma-
tive responsiveness and reflective capacities to be open to recognizing rea-
sons not to do what they are choosing to do. Their action cannot be simply
unstoppably impulsive. These three conditions strike me as setting a

29. Agnieszka Jaworska, “Caring, Minimal Autonomy, and the Limits of Liberalism,”
in Naturalized Bioethics: Toward Responsible Knowing and Practice, ed. Hilde Lindemann, Mar-
ian Verkerk, and Margaret Urban Walker (Cambridge: Cambridge University Press, 2009),
80–105.
290 Ethics January 2021
plausible minimum bar for how much capacity for decisional autonomy
someone needs in order to be potentially competent to engage in consen-
sual sex.
The third component is an especially important condition for being
able to engage in consensual sex, because someone who cannot be open
to counter-reasons in this minimal sense also can’t reliably recognize and
respect their partner’s refusal, needs, or desire to exit if these are in ten-
sion with their own desires. Someone who cannot recognize their part-
ner’s needs and desires as possibly conflicting with their own and de-
serving of respect should under no circumstances be sexually active. We
spend a lot of time worrying that people without autonomy might be ex-
ploited and have nonconsensual sex imposed on them, but we should
also worry that people without autonomy cannot be trusted to choose not
to impose themselves on someone else who has not consented.
But notice that a person who is mildly drunk, or who has moderate
cognitive disabilities or fluctuating memory, could meet all of these con-
ditions for minimal autonomy. So could someone who for structural rea-
sons has limited opportunities, a curtailed ability to assert themselves,
and so forth. So this minimal bar for autonomy does not rule out the
kinds of cases we have been looking at. But again, meeting this bar is
not sufficient for enabling legitimate consent. This just establishes a min-
imum level of competence that makes consensuality—which requires the
fuller agential participation we have been looking at—a possibility. More-
over, as I have emphasized, consensuality is in turn not the only ethical
dimension of sex.

VI. EPILOGUE: SEARCHING FOR GOOD SEX

As I was writing this article, I searched around for media portrayals of

good-quality, scaffolded consent under nonideal conditions of compro-
mised autonomy. I was looking for explorations of good, ethical sex in-
volving drunk people, people in unequal power dynamics, and people
with cognitive deficits, in particular. I found such representations hard
to track down. Partly, I think, this is because we are uncomfortable with
the whole idea of consent under such nonideal conditions. When some-
one has compromised capacities, we generally think of sex as bad and
dangerous and shameful for them. When they have reduced situational
power within a sexual relationship, we either simply disapprove of that re-
lationship or tell myths that deny that power differential—myths of se-
ductively controlling women, or oversexualized people of color, or love
that is somehow so magical that it erases institutional power differentials.
My casting about was not scientific, and there may well be positive and
responsible portrayals of consensual sex under nonideal conditions of
compromised autonomy out there to find. But what I found was that in
 Kukla Nonideal Theory of Sexual Consent 291
most cases such sex was either treated as obviously unethical and abusive
or treated cavalierly as requiring no special scaffolding.
Perhaps one of the most thoughtful and impressive developed ex-
plorations of failed consent under nonideal conditions is the Lars Von
Trier movie Breaking the Waves (1986). In this movie, Bess, a socially iso-
lated woman with limited cognitive capacities and a submissive and trust-
ing nature, is asked by her newly disabled husband to have sex with multi-
ple men in their remote northern town, for his own sexual gratification.
Although she hates the idea, she does so willingly, because she loves him
and is deeply dependent on him, and she values being obedient to him
and pleasing him above almost anything. She is humiliated and slut-
shamed ruthlessly by her other partners and by people in the community,
sometimes violently, and she takes no sexual pleasure in the encounters.
Because her family didn’t approve of the marriage, she has no social
network.
Bess has at least minimal autonomy. Her capacities don’t automati-
cally preclude consensual sex. And she does in fact choose, voluntarily, to
have sex. But her husband does the opposite of enabling her agency or
holding her in personhood. In the face of her vulnerability, he uses his
power over her to build her into a being who acts and desires as he wants
her to; he gives no uptake whatsoever to her expressions of her needs, de-
sires, or well-being, and instead uses her as a palate on which to project
his fantasies. Tearing her out of any meaningful social context or commu-
nity, she is a victim of radical epistemic injustice. She has no power to
speak and be heard or seen, and as her sense of reality corrodes, there
is no one there to help stabilize it for her. She collapses into a nonself.
It is a hard movie to watch. But it is not, I submit, a movie about how peo-
ple with her capacities cannot have consensual sex; rather, it is a movie
about how people like her can have their agency undone by other people
and by their situation, even with their minimal autonomy intact.
The 2004 film 50 First Dates, directed by Peter Segal and billed as a
romantic comedy, also represents a cavalier exploitation of a character’s
compromised autonomy. But in sharp contrast to Breaking the Waves, the
character’s manipulability is here played for laughs, with no apparent
awareness of or concern for how her compromised capacities impact
her ability to give legitimate consent. In the film, Lucy, a young woman
with amnesia, wakes up every morning with a blank memory slate. When
Henry falls in love with her and decides to woo her, he shows her a video
every morning designed to convince her she is already in a relationship
with him. Henry tries to get Lucy to agree to sex by completely controlling
her narrative understanding of her relationship with him. He has and
uses the power to represent to her what her desires are, and he uses it
to talk her into sex, even though she has little to no ability to push back
or have confidence in any alternative narrative. This is pretty much the
292 Ethics January 2021
exact opposite of holding her in her agency and identity. This is hijacking
her agency, and the movie is in effect a series of rapes, portrayed as a light-
hearted romance.30
It seems to me that it is important for us to develop, curate, and ex-
plore powerful examples of well-scaffolded, consensual sex under non-
ideal conditions. We should be as well versed in how to make good, eth-
ical sex possible as we are in how sex can go ethically awry.

30. If Henry and Lucy in fact had a stable relationship prior to Lucy’s accident, and
Henry was using the videos to help anchor her in a preexisting reality, then this could have
been a powerful example of holding in agency rather than of hijacking agency.