Can Consent Be Irrevocable?

1. Introduction

In a video that went viral in 2013, a woman named Jessica is on a canyon swinging trip, preparing to launch herself off a cliff into a hundred-meter bungee fall.¹ She is clearly terrified. Her boyfriend, Clayton, urges her to jump. He counts down from five, but when he gets to one, Jessica still can’t bring herself to jump. He counts down from five over and over again; eventually, Jessica insists that she does not want to jump anymore. She also makes it clear that she does not want Clayton to push her, and he assures her that he won’t. But after continuing to urge Jessica to jump (and being told time and again that she does not want to be pushed), Clayton pushes her and bursts into laughter as she swings into the canyon.

The video is excruciating to watch. I have seen it many times and I still cover my eyes at the end. Jessica’s fear is palpable as she stands on the edge of the cliff, anxious and tearful, begging Clayton not to push her. After he pushes her and starts laughing, Clayton looks like the biggest jerk of all time. Some representative remarks from the comments section of the video include “worst boyfriend ever,” “[s]hame on you for pushing her off and laughing,” and “[t]hat’s just terrifying for that poor girl, you’re a monster.”

However, in an interview conducted after the video went viral, Jessica revealed that earlier that day, she had given Clayton consent to push her off the cliff. She had said to him, “you know, I’m really nervous about this. I don’t know if I’ll be able to jump. If I can’t jump, you need to push me.”² In the interview, Jessica also shared that after she had both feet back on the ground, she was glad Clayton had pushed her. After listening to the interview, Clayton seems like much less of a jerk. After all, Jessica intended for her consent to be irrevocable; she instructed Clayton to push her off the cliff because she had anticipated changing her mind about jumping. She wanted Clayton to push her off the cliff no matter what.

This article considers whether morally valid consent can be irrevocable. Morally valid consent is consent that succeeds in generating a moral permission, making otherwise impermissible conduct permissible. For example, if you give your hairdresser consent to cut your hair, you make it morally permissible for them to do so; without consent, cutting another person’s hair is wrongful. When consent is not morally valid, it fails to generate a moral permission. Some widely accepted conditions for morally valid consent include that it be informed and that it be uncoerced.³ My aim in this article is to show that consent must also be revocable. If I am right, then Clayton wronged Jessica because he pushed her off the cliff without consent. Although she had intended for her consent to be irrevocable when she told him earlier in the day that he needed to push her if she couldn’t bring herself to jump, when she later said she no longer wanted to be pushed, she revoked her consent.

A few clarifications before I begin. First, for the purposes of argument, I assume a performative account of consent and its revocation rather than an attitudinal one. On the performative account, consent requires public communication such as verbal or written speech, or body language. On the attitudinal account, consent simply requires adopting the relevant mental attitudes. This performative approach allows for more straightforward moral evaluation of the cases that follow, as we can assume that permissions are granted and dissolved through public communication rather than private mental attitudes. However, nothing in particular hinges on this assumption. The argument I present may be reinterpreted under an attitudinal view.

Second, my focus in this article is on high-stakes consent: consent that is needed to avoid a serious moral wrong. Consent to push someone off a cliff is high-stakes because it is seriously wrong to push someone off a cliff without consent. I focus on high-stakes cases not only because our intuitions about them are clearer, but because these are cases where we care about consent the most. For instance, we feel very strongly that sexual consent—a paradigm of high-stakes consent—must be informed and uncoerced. But consider a case of low-stakes consent, such as consent to brush past someone on a crowded bus. In this case, we have a far more permissive view as to what counts as informed consent. Philosophical work on consent focuses primarily on high-stakes cases for these reasons, and I will follow suit.

Finally, my focus is on bodily consent: consent we give others to touch or control our bodies in ways that would otherwise be impermissible. I will not discuss cases involving consent to use or control our property or anything other than our bodies. I think the importance of bodily integrity and the role consent plays in protecting it gives us special to think that bodily consent must be revocable. Indeed, it is not hard to think of cases where property consent is unproblematically irrevocable. For example, one might think of gift-giving as a practice where someone gives irrevocable consent to another person to take her property as theirs.⁴ Although the view I present does not easily permit a unified theory of bodily and property consent, I do not think this is a weakness of the view. Rather, my argument that bodily consent is always revocable because it exists in part to protect our bodily integrity puts pressure on the prevailing assumption that a single theory could account for every kind of consent. There is good reason to think that the norms governing consent to use our bodies and to use our property are different.

The paper proceeds as follows. In section two, I consider two reasons why it might seem as though consent can remain morally valid despite attempts to revoke it: prior consent and subsequent consent. I explain why we should be skeptical of both these explanations. In sections three and four, I consider two arguments against irrevocable consent. On the argument from informed consent, irrevocable consent is not morally valid because it cannot be sufficiently informed. On the argument from bodily integrity, irrevocable consent is not morally valid because we do not have the authority to deny our future selves the ability to protect our bodily integrity: the experience of our bodies as the point of integration between our subjective experience and the rest of the objective world. I explain why the argument from bodily integrity captures the unique problems raised by irrevocable consent and illuminates an undertheorized distinction between autonomy and bodily integrity. On my view, our power to consent is not wholly grounded in our autonomous agency, as consent theorists have largely assumed, but also in our interest in protecting our bodily integrity.

2. Problems with prior and subsequent consent

Broadly speaking, there are two reasons one might think Jessica’s consent remained morally valid when Clayton pushed her off the cliff despite her attempt to revoke it. One is that she had consented to being pushed off the cliff prior to ascending the cliff. Another is that after Clayton pushed her, she was glad he had done so. Let us call consent to X prior to X prior consent and consent to X after X subsequent consent. We can contrast prior and subsequent consent with present consent, or consent a person gives at the time of action.⁵ In what follows, I will first explain why we should be skeptical of prior consent’s relevance to the moral validity of present consent and I will then explain my concerns about subsequent consent.

3. The argument from informed consent

One might argue that irrevocable consent is not morally valid on the basis that it cannot be sufficiently informed. There is widespread agreement that without a person’s informed consent to a high-stakes intervention, the intervention is impermissible even if it would help her, is recommended by experts, or would benefit third parties. Call this the argument from informed consent.

I won’t give an account of precisely the information a person must have or have access to for her consent to be informed—for that, I turn to previous work on informed consent.²⁶ But even without knowing precisely what information a person must have to give informed consent, there are a few observations we can make. One is that how informed consent must be depends on how high the stakes are. For instance, it is much more important that you give informed consent to a surgeon to perform a medical procedure than to a hairdresser to give you a haircut; therefore, you need to know more about the medical procedure than about the haircut to give informed consent. When you give irrevocable consent to an action, the stakes are immediately higher than they would be for revocable consent to the same action. For a lighthearted example, consider shopping at a store with a generous return policy as opposed to one where all sales are final. When you buy a sweater at the first store, you may not feel like you have to think too much about the purchase; you could always change your mind. But if you buy a sweater at the second, you might put in extra effort to ensure it fits, read its care instructions, etc., since the stakes are higher.

One might argue that the stakes of irrevocable consent are so high that it can never be sufficiently informed. For a person’s irrevocable consent to be informed, it seems as though she would need information about all the future situations in which the relevant right will be waived and agree preemptively to cede her decision-making authority in all those situations. This is a tall order, to be sure, but perhaps not an impossible one. Suppose Jessica watches dozens of canyon swinging videos to get a sense of what it will feel like to do it. She speaks to friends who were glad that their canyon swinging partners pushed them off the cliff when they couldn’t bring themselves to jump. Then, she says to Clayton, “You have permission to push me off the cliff no matter what.” Even if Jessica cannot know exactly what canyon swinging is like, it seems plausible that she could know enough for her consent to be informed.

The problem here is that information about what canyon swinging is like is not the only information Jessica needs for her consent to be informed. Even if she gets to a point where there is nothing more she can learn about canyon swinging, she may not be at a point where there is nothing more she can learn about herself that is relevant to the moral validity of her consent.²⁷ To illustrate this point, suppose a designer offers you a dress to wear for a formal gala that will be held in a year’s time. You try the dress on. You know everything there is to know about the dress: how it fits you, how you feel when you wear it, how well it photographs, etc. You are in an excellent position to agree to wear the dress right now. However, you are not necessarily in a good position to agree to wearing the dress, unaltered, in a year—not because you might learn something new about the dress, but because you might learn something new about yourself. In a year’s time, you might have gained or lost weight, or become pregnant. Perhaps your taste will have changed so much that you no longer like the style of the dress. These are all considerations that are relevant to your authority to commit to wear the dress, and in light of them, you are not well-positioned to agree to wear the dress unaltered in a year no matter what.

We can use the same analysis to show why Jessica is not well-positioned to give irrevocable consent to be pushed off the cliff. I have argued that pain, fear, and anxiety can undermine one’s capacity to consent, yet are emotions that we generally do not feel fully until the time of action approaches. Given our unpredictability—the ways our bodies, desires, values, goals, and emotions can change, sometimes in ways we do not expect, and sometimes suddenly, over a short period of time—it is difficult to say that we can be sufficiently informed about what our future selves will be like to give irrevocable consent.

The high standards for informed irrevocable consent reveal something important: unlike uninformed consent, irrevocable consent does not misfire and fail to generate permissions from the start. If you consent to surgery without sufficient information about what the procedure involves, you did not consent at all; your consent failed to meet a condition for moral validity. However, if you have sufficient information to give informed consent to an intervention, but insufficient information to give irrevocable consent to that intervention, your consent is morally valid until you revoke it.²⁸ Because the information available to you when you consent to an action is often different from the information available to you at the time of action, and because the information available to you at the time of action may be relevant to the moral validity of your consent, it makes sense that there are different standards for the information required to give informed consent to something now and to something later. When Jessica gave Clayton permission to push her before they ascended the cliff, she consented to being pushed off the cliff with the best information available to her at the time. However, her position at the time of action is importantly different from her position when she gave prior consent. At the time of action, she has additional information about how it feels for her to go canyon swinging that is relevant to the moral validity of her consent.

One might object to the argument from informed consent on the basis that it wrongly places the moral burden on Clayton, who is simply acting on wishes that Jessica had previously communicated to him. Why should he be blamed for doing exactly what Jessica asked him to? Sometimes, consent is primarily the responsibility of the consent-giver, not the consent-receiver. If Jessica’s consent to be pushed off the cliff is uninformed, isn’t that on her? Clayton may have a moral obligation to ensure Jessica’s consent is not deceived or coerced, but there is some information relevant to her consent—namely, her feelings about canyon swinging—that only she can access. In some cases, providing information may be primarily the responsibility of the consent-receiver. For instance, in medical cases, physicians might be uniquely responsible for ensuring that the patient has access to all the information they need to give informed consent. But in other cases, the consent-receiver may have no obligation to be the one to provide that information. Jessica and Clayton’s case seems like the latter kind.

I will address this objection in the following section, but for now, we can respond by drawing a distinction between the permissibility of a person’s actions and her praise- or blameworthiness for those actions. Proponents of the attitudinal view of consent—on which consent consists in a person’s inner assent—use this distinction to explain why communication makes a difference to the morality of actions that require consent.²⁹ On the attitudinal view, your internal assent can make it permissible for me to hug you, but I could still be blameworthy for hugging you without having good reason to think that you had consented to my doing so since you didn’t communicate your consent to me. Without endorsing the attitudinal view of consent, we can use the distinction between permissibility and blameworthiness to address the objection raised here. In cases where it is the consent-giver’s responsibility to ensure her consent is sufficiently informed, yet she fails to do so, the consent she gives is not morally valid, and it is therefore impermissible for the consent-receiver to act on the consent. However, the consent-receiver may not be blameworthy for acting on her consent; he had good reason for thinking that she had consented, after all.

4. The argument from bodily integrity

I have defended one argument against the possibility of irrevocable consent, on which irrevocable consent cannot be sufficiently informed to be morally valid. However, on the argument from informed consent, there is nothing about irrevocable consent in particular that makes it invalid. Rather, the invalidity of irrevocable consent is subsumed under the invalidity of uninformed consent. Even if irrevocable consent cannot be sufficiently informed, we might wonder whether there are independent grounds for thinking that consent cannot be irrevocable. In this section, I argue that consent’s role in protecting our bodily integrity—the point of integration between our subjective experience and the rest of the objective world—explains why consent must be revocable.

When Jessica gives Clayton irrevocable consent to push her off the cliff, she is attempting to “bypass” her future self’s control of the situation. The argument I now propose is rooted in the idea that certain attempts to bypass one’s capacity for self-direction are problematic—specifically, when they compromise one’s ability to protect one’s bodily integrity. The idea that we cannot always bypass our capacity for self-direction might be met with resistance. Although it is certainly problematic to bypass another person’s capacity for self-direction, it may be unclear what it means to bypass one’s own such capacity, let alone for it to be problematic. In this section, I begin by explaining what it means to have the authority to bypass one’s capacity for self-direction, and then I argue that we lack authority to do so when it denies our future selves the ability to protect our bodily integrity.

To understand the idea of bypassing one’s capacity for self-direction, we can look to Derek Parfit’s nobleman.³⁰ As a young man, the nobleman learns that later in his life, he will come into a large inheritance. He currently has socialist leanings, but suspects that, like most people, he will become more conservative as he grows older. To ensure his inheritance will go to the peasants, he signs a document that will automatically allocate it to them and gives his wife the sole authority to revoke the document. He then asks his wife to promise him that she will not, under any circumstances, revoke the document. He says to her, “I regard my ideals as essential to me. If I lose these ideals […] I want you to regard your husband then, not as me, the man who asks you for this promise, but only as his corrupted later self.”³¹

According to Parfit, when the nobleman asks his wife in middle age to revoke the document, she may plausibly regard herself as being committed to two different people. On one hand, she is committed to the young man she married and to whom she made the promise. If she were to revoke the document, she would be betraying him. On the other hand, she is committed to her husband, the man who is now asking her to revoke the document. One need not endorse a time-slice conception of personal identity to appreciate her dilemma. She need not regard her husband as numerically different from the man she married to see him as a different person in the practical sense.³² And insofar as the man she married is a different person, practically speaking, from the person she is married to now, she experiences a genuine conflict of loyalty.

Just as we might blame Jessica for asking Clayton to push her off the cliff no matter what, we might blame the nobleman for putting his wife into this position in the first place. Nietzsche thinks along these lines. On his view, to have the authority to make commitments, a person must be able to “stand security for her own future” and have “a real memory of the will.”³³ When Nietzsche says we don’t have the authority to make commitments when we are unsure about our futures, he is not saying that we lack the capacity to do so. A person who makes a commitment despite being unsure about his capacity to fulfill it has still made a commitment.³⁴ What Nietzsche points out is that there is a sense in which he lacks the authority to bind his future self to the commitment. If the young nobleman had full authority to commit his future self to giving the inheritance to the peasants, his plea to his wife to revoke the document in middle age would not bear the normative weight it does; his wife would be bound to fulfill the promise she made to his young self. But his present plea does bear normative weight. We regard the wife as being committed to two different people; her normative reasons pull her toward different courses of action. As Christine Korsgaard puts it, “if she approaches her husband as one person, she’ll be making a mistake.”³⁵ Generating obligations without full authority to do so is a way of playing fast and loose with that authority. When Jessica asks Clayton to push her off the cliff no matter what, she is similarly playing fast and loose with her ability to generate permissions. She puts Clayton in the difficult position of having to decide whether to betray his past girlfriend who consented to being pushed off the cliff or the one who now wishes to revoke that consent. Perhaps more importantly, however, Jessica puts her future self in a difficult position. In instructing Clayton to push her off the cliff no matter what, she silences her future self’s attempts to protect herself by revoking consent.

One might object that Jessica’s situation is unlike the nobleman’s in an important way. Unlike the nobleman, Jessica anticipates that once she ascends the cliff, she will become irrational—that her judgment will become clouded by fear. This is precisely why she wants her consent to be irrevocable. On the other hand, the young nobleman anticipates a change in his values, not in his capacity for reason. Because of Jessica’s irrationality, it might seem as though Clayton ought to prioritize Past, Rational Jessica’s wish to be pushed off the cliff. I briefly addressed this objection in section two, when I argued that we cannot appeal to an agent’s previous “soundness of mind” to justify acting on her prior consent in light of her present dissent because doing so is too morally risky. But I think there is a deeper reason why we ought not act on a person’s prior consent in light of her present dissent, rooted in the tension that can arise between respecting a person’s autonomously-formed preferences and respecting her bodily integrity.

In section two, I considered how our power to consent stems in part from our interest in protecting our autonomy: our capacity to determine how we act, in accordance with our genuine desires, free from external influence. I propose that our power to consent is closely related not only to our interest in protecting our autonomy, but also to protecting our bodily integrity: the point of integration between our subjective experience and the rest of the objective world. Unlike autonomy, which endows a privileged status upon attitudes that one endorses in a cool, calm, and collected state, even if one does not have those attitudes at a particular moment, bodily integrity pertains to one’s experience of one’s body at a particular moment.

To understand the distinction between bodily integrity and autonomy, we can consider the following, not uncommon scenario. Suppose a person values her life when she is in a “cool, calm, and collected” state. Perhaps she tells her family that if she were ever on the brink of suicide, she would want medical help, even if that meant committing her to a hospital against her will. Now, suppose she experiences a mental breakdown and her family involuntarily commits her to a hospital. She begs her family not to do this. Her pleas become so desperate and violent that the doctors have no choice but to lock her in a room by herself, where she continues in anguish to plead with them to release her.

On any account of autonomy, the patient is not acting autonomously. She is in a highly disturbed mental state that has undermined her capacity for self-governance; her pleas to her family and doctors to release her are made in spite of herself. Therefore, her family and doctors do not violate her autonomy by committing her to the hospital involuntarily. Their actions reflect their care for her and her autonomously-formed preferences; they want her to recover from the breakdown before she hurts herself. However, if you have ever been one of the family members or doctors in this scenario, you will know that it leaves you with an awful feeling. When someone begs you to refrain from doing something that you must do for her own good, it doesn’t feel like you are respecting her. Indeed, the more desperate her pleas—the more she lacks control over her actions—the less it feels like you are respecting her. If a person’s power to consent is grounded solely in her autonomy, then committing her against her will is merely unpleasant; the patient’s pleas, while desperate and violent, are not utterances she has control over, but something like mere sounds she is making in a perturbed mental state.

To be clear, individuals in emotionally disturbed states who pose threats to themselves or to others should be taken care of, even if that means committing them to a hospital involuntarily. However, even if the patient’s family and doctors acted rightly, all things considered, the awful feeling they are left with seems to have moral significance. Rather than mere unpleasantness, the awful feeling is something like guilt: a reactive attitude in response to a moral infraction.³⁶ Bernard Williams has argued that when we are faced with conflicting prima facie moral obligations, even if we act rightly in the end, all things considered, the other obligation leaves a “remainder” that even fully admirable moral agents often feel in the form of regret.³⁷ However, apt responses to moral remainders need not be limited to regret. Guilt is another response one might have to violating a prima facie obligation even though one has acted rightly, all things considered. It is not uncommon to feel guilty and for that guilt to be apt even though one has done nothing wrong. For example, suppose you and your spouse promise never to lie to one another, but protecting the well-being and privacy of a dear friend requires that you lie to your spouse. Any guilt you feel for lying to your spouse demonstrates your responsiveness to the requirements of your commitments and to the moral remainder left by your actions. The aptness of your guilt is reflected in the fact that it would not be inappropriate for you to apologize to your spouse for having lied to them, even though you did not act wrongly.³⁸ In the hospitalization scenario, the guilt experienced by the patient’s family and doctors is responsive to the fact that they have wrongfully infringed upon the patient’s bodily integrity, committing her to the hospital without consent, even though they did the right thing, all things considered. The feeling of anger that fills viewers who watch Clayton push Jessica off the cliff in the canyon swinging video is also more like a reactive attitude toward a perceived moral infraction than a mere feeling of unpleasantness. When viewers call Clayton a “monster,” they are responding to the fact that he is unmoved by Jessica’s pleas not to be pushed. This awful feeling is dampened with the knowledge that Jessica had previously asked Clayton to push her, but it doesn’t entirely go away. Even if pushing Jessica off the cliff was the right thing to do, all things considered, pushing her despite her attempts to protect her bodily integrity by revoking consent leaves a moral remainder that calls for some response on Clayton’s part.

There is something morally questionable about treating a perturbed person’s pleas as mere sounds rather than as genuine appeals. When understood as reactive attitudes, the feeling of guilt the patient’s family and doctors in the case above and the feelings of anger and horror viewers of the canyon swinging video experience provide evidence that we should not ignore the pleas of someone in a perturbed mental state even if they lack the capacity for autonomous choice. I propose that bodily integrity can explain why this is the case. When the patient begs her family and doctors not to commit her, and when Jessica begs Clayton not to push her off the cliff, they are protecting something morally significant: their bodily integrity. On my view, individuals should retain the capacity to protect their bodily integrity even if they cannot make autonomous choices and even if exercising that capacity would undermine the desires and values they identify with in a cool, calm, and collected state.

On some views, bodily integrity is simply a right that is guaranteed to a person in virtue of her autonomy. For instance, Martha Nussbaum has argued that a person’s right to bodily integrity—to have the decisions she makes for her body treated as sovereign—is based on the idea that autonomous, rational agents’ choice-making abilities, including choices about their bodies, ought to be protected.³⁹ However, on other views, the right to bodily integrity is not merely the right to non-interference in decisions about one’s body, but the right to exclusive control over one’s body. According to Jonathan Herring and Jesse Wall, the right to bodily integrity should not be understood as protecting a person’s bodily autonomy, but as protecting “the point of integration between a person’s subjectivity and the remainder of the objective world.”⁴⁰ A person’s bodily integrity consists not only in her rationally-formed preferences and values for her body. If it did, then the body would be no more than a “necessary (but ultimately uninteresting) casing” for her mind.⁴¹ By incorporating the notion of subjectivity into our understanding of bodily integrity, we can explain why a person’s desires and dignity matter even if she is not in a sound state of mind, and why interference with bodily integrity requires further justification than interference with autonomy.

If the power to consent were grounded solely in a person’s autonomy, consent given in a cool, calm, and collected state to an action that will take place when the agent is later in a perturbed mental state could be irrevocable. Her prior consent represents her autonomously, rationally-formed interests, which would take priority over her present, irrational dissent. But if there is a notion of bodily integrity separate from autonomy, and if a person’s power to consent is grounded in part in her interest in protecting her bodily integrity, then she does not have the authority to give irrevocable consent. The perturbation she experiences matters because her body is the point of integration between her subjective experience and the rest of the world, and she experiences her body as being under threat. Her present dissent matters. And the guilt the patient’s family and doctors are left with after they involuntarily commit her to the hospital and the anger audiences feel toward Clayton for pushing Jessica off the cliff are responses to what they perceive as disrespect for bodily integrity.

Although bodily integrity and autonomy are distinct concepts, they are not necessarily unrelated to one another. What an autonomous agent consents to being done to her body will often coincide with her interest in protecting her bodily integrity. Even in a cool, calm, and collected state, we do not want to subject ourselves to painful or dehumanizing experiences. However, it is not uncommon for considerations of bodily integrity and autonomy to conflict because violations of bodily integrity are precisely the kinds of interventions that can compromise a person’s capacity to make autonomous choices. How How can a person be expected to remain cool-headed—to act in ways that reflect what she truly values—when her bodily integrity is under threat? One can’t be expected to think clearly and rationally when one’s body is being violated. The idea that a person’s body and not merely her autonomous agency must be respected may be hard to accept, especially for philosophers who, following Plato, assume that the body is subordinate to the rational mind, “an evil” that is “mingled with our souls,” filling us with “passions, and desires, and fears, and all manner of phantoms” that impede our search for truth.⁴² But I hope that the examples and analysis I have offered demonstrate the need to integrate bodily integrity into our conversations about consent. Recognizing the distinction between bodily integrity and autonomy allows us to take subjective experiences of violation seriously.

5. Conclusion: asking for help

I have argued that prior consent and subsequent consent do not bear on the moral validity of present consent and therefore cannot ground irrevocable consent. I then presented two arguments that consent cannot be irrevocable: the argument from informed consent and the argument from bodily integrity. On the argument from informed consent, irrevocable consent cannot be sufficiently informed because we do not have access to information about how we will feel at the time of action that is relevant to the moral validity of our consent. On the argument from bodily integrity, consent cannot be irrevocable because we do not have the authority to deny our future selves the ability to protect our bodily integrity: the point of integration between our subjective experience and the rest of the objective world. Because our power to consent is grounded not only in our interest in being autonomous but also in protecting our bodily integrity, the fact that a person is in a perturbed mental state and unable to make autonomous choices does not diminish her right to protect her bodily integrity by revoking consent.

To conclude, I want to consider one last objection to my view that I find especially evocative and gesture toward some responses to it, though I will leave full treatment of the objection for a future project. One worry one might have about my conclusion that consent must be revocable is its implication that we can never ask for help to do the things we want but cannot will ourselves to do. Jessica may not be able to will herself to jump off the cliff. But jumping is what she “really wants”: in a cool, calm, and collected state, she values jumping. Can’t we sometimes ask for that extra push (in Jessica’s case, literally an extra push) to get what we want when our will isn’t strong enough?

One response to this concern points to an idea I have already discussed, that when a wrong action has good consequences, we tend to forget that the action was wrong in the first place. We may even become grateful that it happened. Jessica can be glad that Clayton pushed her off the cliff; perhaps being pushed off the cliff is what she “really wants.” But this does not change the fact that Clayton seriously wronged her by pushing her off the cliff after she revoked her consent. I suspect that we wrong others all the time when we act in their best interest and help them achieve their genuine, autonomously-formed goals. We push people past their comfort zones to get them to do what we think (or even know) they really want to do. Sometimes people are happy they were pushed, sometimes they are not; these subsequent attitudes do not change the moral status of the pushes themselves.

Second, it is worth noting that there is something undesirable about leaving the execution of our wills in others’ hands. Imagine a person who cannot will herself to do any of the things she really wants to, and always relies on other people to “push” her to do them. There seems to be a sense in which she is failing as an agent, a sense in which her life is seriously deprived. She is like Harry Frankfurt’s wanton, a slave of her passions, except that she delegates the work of achieving her “true” goals to others.⁴⁸ We may ask for an extra push when our wills fall short, but perhaps we should not always want to ask in the first place. After all, what Jessica really wants is not to be pushed, but to jump off the cliff.