Portal de Periódicos da CAPES

Refusing Protection

2022; Wiley; Volume: 51; Issue: 1 Linguagem: Inglês

10.1111/papa.12226

1088-4963

David J. Clark,

Free Will and Agency

Siblings: Adam, Teresa, and Dan are adult siblings. Jealous about being left out of their mother's will, Adam attempts to break Teresa's leg. The attack can only be stopped only if either Teresa or Dan break Adam's leg first. Dan prepares to harm his brother, but Teresa—who is a committed pacifist and who would rather be harmed than see Dan harm their brother in her name—demands that he stand down. In at least some cases, then, refusal seems to render otherwise-permissible protection impermissible.1 The following essay is an attempt to vindicate this intuition by developing and defending an explanation as to why refusal has this power. I argue that the power to forbid protection is an upshot of a more basic normative power—the power to mitigate specific normative effects of our rights. We have many powers over our rights. We have, for instance, powers to transfer or waive our rights. But in addition to having these more familiar powers, I claim, we have the power to selectively restore some of the rights that one's attacker forfeits by threatening our rights. One way in which this power is exercised is by the refusal of protection. When Teresa refuses Dan's protection, for example, Teresa restores Adam's rights that Dan not harm him—a right that Adam had lost as a result of his threat to Teresa's rights. The restoration of this right is what undermines Dan's permission to protect Teresa. I call this the Liability Contraction Account. The paper proceeds as follows. Section II sketches two general strategies for explaining how refusal undermines the permissibility of protective intervention: either refusal undercuts the permissibility of protection at the “reasons dimension” or it does so at the “liability dimension.” Section III considers the most developed attempt to explain the magic of refusal at the reasons dimension—Jonathan Parry's Exclusion Account. I argue that we should reject this account, and in Section IV I develop an account—the Liability Contraction Account—according to which refusal works its magic at the liability dimension. Section V unpacks some important features and implications of the account. Section VI considers the special case of refusing protection from agents of the state. As we'll see, how we understand the power of refusal has important implications for the ethics of humanitarian intervention. In real-world cases, a population that is threatened by an unjust aggressor will often fail to be of one mind about protective military intervention from a third-party nation. Some members of the population will welcome protection; some will reject it. Does it make a difference what portion or number of the population rejects protection? Some theorists have argued that anything less than the unanimous refusal of protection is irrelevant.2 At the other end of the spectrum, some have argued that refusal by even a minority of the target population can undermine the permissibility of intervention. The Liability Contraction Account takes a position between these two extremes. It does not treat less-than-unanimous refusal as strictly irrelevant to the ethics of war. But, as compared with an account like Parry's, Liability Contraction predicts a much smaller range of cases in which refusal by a subset of the population makes the difference (more on this in Section V). By refusing protection, the target of aggression can sometimes render otherwise-permissible protection impermissible. The question is how exactly refusal manages to do this. Here is a useful place to begin answering this question. We start by getting before us the ingredients that are normally required for permissible protective intervention. We then consider whether refusal undercuts this permission by removing one of these ingredients. In paradigm cases of permissible intervention, there are two central ingredients. The first is attacker liability. Absent a lesser-evil justification, a person may impose defensive harm on an attacker only if the attacker is liable to suffer such harm—that is, only if the attacker lacks a right against such harm.3 Liability is not sufficient for permissible intervention, however. To say that someone lacks a right against some harm is only to say that a certain deontic obstacle has been removed—that a certain reason not to harm him is absent. This is not yet to say that there is good enough reason to harm him. This, then, is the second ingredient: there must be sufficient reason to harm the attacker. These, then, are the two standard conditions on permissible defensive harm: the liability condition and the reasons condition. Thus there emerge two strategies for explaining the power of refusal. One possibility is that refusal works its magic at the liability dimension: it renders protection impermissible by removing or reducing the attacker's liability to suffer defensive harm. Another possibility is that refusal works its magic at the reasons dimension: it renders protection impermissible by making it so that the defender lacks “good enough” reason to harm the attacker (either because it takes away reasons to defend, or because it contributes countervailing reasons not to defend). On first glance, the appeal to liability seems a non-starter. According to popular theories of defensive liability, an attacker is liable to suffer defensive harm H when the attacker threatens (or is responsible for threatening) to infringe someone's rights and the imposition of H on the attacker is proportionate to, and necessary to avert, that threat.4 Given such a theory, it may seem that the refusal of protection will succeed in undermining an attacker's liability only if it makes it so that the attacker no longer threatens someone's rights. But there is only one way in which the refusal of protection can reliably make it so that the attacker no longer threatens his target's rights—which is by making it so that those threatened rights disappear. The idea is that by refusing protection from an attacker, a person exercises her power to waive her rights. By refusing protection, a target waives those of her rights that are threatened by her attacker, thereby rendering her attacker non-liable to defensive harm, and thereby rendering protective intervention impermissible. Call this the Refusal-as-Waiver Account. The account is untenable. Refusal-as-Waiver tells us that refusal has the same moral effect as consent. But it is clear that refusal has a very different moral effect. Unlike consent, the refusal of protection does not render it impermissible for the target to defend herself, nor does it strip her of her claims to receive compensation if the attack succeeds. Imagine, for instance, a variant of Siblings in which Teresa is not a pacifist, but in which she refuses protection from Dan simply because it matters to her that she be the one to defend herself and that she be the one responsible for the harm to their brother Adam. In such a case, it seems, Teresa would be permitted to defend herself even while Dan is not permitted to defend her. Refusal-as-Waiver predicts otherwise, however, since by waiving those of her rights that are threatened by Adam, Teresa would free Adam of his liability to suffer defensive harm at the hands of everyone—herself included. Likewise, imagine that Teresa fails to defend herself and that Adam's attack succeeds: he breaks Teresa's leg. The fact that Teresa refused Dan's protection surely does not change the fact that Adam now owes Teresa compensation for her injuries. Refusal-as-Waiver predicts otherwise, however. When a person waives their right against some harm, they are not typically owed compensation for that harm. (Illustration: When I decide to donate my kidney and consent to the doctor's removal of the kidney, the doctor does not owe me compensation for my loss.) We should reject Refusal-as-Waiver. And since this may seem like the only game in town for explaining the power of refusal at the liability dimension, it may seem that we should therefore reject the idea that refusal works its magic by undermining attacker liability, and accept that it must therefore work its magic at the reasons dimension. In Section IV I argue that this is too quick, and that there are ways in which a target's refusal can undermine attacker liability without costing the target her rights. But before we get to my positive account, we would do well to examine the only developed attempt to account for the power of refusal at the reasons dimension. Power to Exclude. A person X has the normative power to make it so that the fact that Y's 𝝫-ing would benefit (or harm) X is not a reason for Y (not) to 𝝫.6 Let's call the conjunction of Power to Exclude with the claim that the refusal of protection is one way of exercising this power the Exclusion Account. This account nicely avoids the aforementioned shortcomings of Refusal-as-Waiver. Exclusion permits Teresa to defend herself even as she refuses Dan's protection. This is because powers of reasons-exclusion are person-relative: Teresa's refusal of Dan's protection only makes it so that her benefit does not count as a reason for Dan to defend her; it does not likewise exclude Teresa's well-being from her pool of reasons. Likewise, Exclusion also avoids the implication that refusal undermines the refuser's right to compensation. According to Refusal-as-Waiver, Teresa waives those of her rights that are threatened by Adam. This means that when Adam breaks her leg, Teresa is harmed but her rights are not infringed. And because her rights are not infringed, she is not owed compensation. On Exclusion, by contrast, Teresa does not give up any of her rights when she refuses protection. Thus, when Adam breaks her leg, her rights are infringed. It is for this reason that she is still owed compensation even though she refused protection. Despite these virtues, I argue, we should reject Exclusion. My objection comes in three parts. First, I argue that the Power of Exclusion principle has very implausible implications. Second, I argue that Parry's positive argument for Power of Exclusion account fails; the principle is undermotivated. Third, I argue that even if we grant the truth of Power of Exclusion, we should reject the idea that the refusal of protection is generally used to exercise this power of reasons-exclusion. Runaway Boulder. An earthquake has sent a boulder careening down a hill towards Alice. If it hits Alice, it will kill her. You can let the boulder hit Alice, or you can redirect it to the left or right. If you redirect it to the left it will hit and kill Bob. If you redirect it to the right it will merely break one of Charlie's toes. Bob sincerely tells you not to take his interests into account, and asks that you sacrifice him rather than let Alice or Charlie suffer harm. I find this very implausible. Perhaps it is true that you would not wrong Bob by killing him; he has, perhaps, consented to this fate. But even so, it is obvious that you should redirect the boulder toward Charlie rather than Bob. Even if we grant that Bob has consented to be killed, you plainly have more reason to redirect the boulder toward Charlie. Now it might be argued that we have more reason to harm Charlie than Bob, not because of Bob's interests, but because of the interests of others.7 It might be thought, for instance, that the reasons for harming Charlie are grounded in the interests that Bob's friends and family have in Bob's leg not being broken. But these cannot plausibly be the only reasons against turning the boulder toward Bob. The fact that you should not kill Bob is not plausibly contingent on the number of people who care about Bob. (As a test, imagine that there is no one in the world who cares about Bob, but many people who care about Charlie. Still, it seems clear, you should not kill Bob.) But more importantly, it just seems obvious that the strongest reason not to direct the boulder toward Bob is that it would harm Bob. Just as it is plain that you should not direct the boulder toward Bob, it is plain why you should not direct the boulder toward Bob. Perhaps we could bite the bullet in cases like Runaway Boulder if there was a compelling argument in favor of Power to Exclude. But there isn't. Or at least, I'll argue, Parry does not succeed in giving us one. Parry's main line of support for Power to Exclude is that the principle plays a central role in explaining “paternalism's distinctive wrongness.”8 It is largely uncontroversial that when it is wrong to act paternalistically toward someone, this is so because the action would deny the person's autonomy or competency in some way. The hard question is how to make this idea precise. Parry suggests that the best precisification will appeal to something like Power to Exclude. When my benefit does not justify your interfering with my autonomy even for my own sake, says Parry, this is because, “As a result of exercising my will, certain reasons—those pertaining to my good or welfare—no longer contribute to justifying others' actions.”9 Cigarettes: My brother willingly smokes on occasion, but only because he enjoys it, and not because of any addictive compulsion. And when he smokes, he only ever smokes outside, where he is the only person who suffers harm. I care about my brother's health, however, and so I secretly hide his cigarettes from him whenever I find them, letting him think that he absent-mindedly misplaced them. I find this diagnosis implausible. My hiding the cigarettes is wrongfully paternalistic whether or not my brother has “exercised his will” in any way that can be said to be an exercise of his power of reasons-exclusion. He certainly could have performed certain actions whereby he exercised this power. Perhaps he would exercise this power if he told me that he didn't want me to prevent him from smoking. But the point is that he needn't do any such thing for my action to be wrongful. He doesn't need to perform some action that can be plausibly be said to exclude certain facts about his well-being from my realm of reasons in order for it to be impermissible for me to hide the cigarettes. Forms of paternalism like this one are wrongful “by default”; no exercise of powers of reasons-exclusion are needed to make them wrongful. Advice: An undergraduate student asks her philosophy professor where she might apply to graduate school, given that she's interested in programs that are strong in ethics but also interested in programs that are strong in metaphysics. Suppose the professor somehow knows that it would be better for the student if she were to study ethics than metaphysics. And suppose the professor (but not the student) also knows that this student has the following psychological trait: if she hears someone snap their fingers while presenting an option, she will form a strong, irrational preference toward that option. In order to benefit the student, the professor exploits this trait, and subtly snaps his fingers only while mentioning the ethics programs. The lesson: we cannot give a general explanation of the wrongfulness of wrongful paternalism by appeal to Power to Exclude. And thus we cannot motivate this principle on the grounds that it plays a central role in explaining paternalism's distinctive wrongness. Easy Defense: Attacker attempts to murder Target. Defender can protect Target by lightly punching Attacker. But Target is such a strict pacifist that she condemns even a mild harm such as this, and for this reason refuses Defender's protection. Exclusion, however, delivers the same verdict in Easy Defense as in Siblings. Just like Teresa, Target has refused protection. She has thereby excluded her well-being from Defender's reasons for protecting her. Defender thus has no reason to protect Target (apart from reasons that are grounded in the interests of others), and he has some reason to not protect Target—namely, that this would cause mild harm to Attacker.11 Exclusion thus implies that Defender lacks the permission to intervene. I find this result extremely counterintuitive. Even if there is such a power as a power of reasons-exclusion, Target is not plausibly exercising that power by refusing protection. The Exclusion account claims that the power of refusal is located at what I've called the “reasons dimension.” Specifically, it claims that refusal undermines the permissibility of protective intervention by excluding certain reasons that would normally favor such intervention. I've argued that we should reject this account. But this isn't the only way that refusal might make a difference at the reasons dimension. Rather than excluding some of one's reasons for protective intervention, refusal might instead contribute additional reasons against protective intervention. I think it's clear that refusal does contribute such reasons. Consider our central case of Siblings. When Teresa refuses Dan's protection, Teresa reveals certain of her preferences to Dan. Since it is (pro tanto) in Teresa's interest that her revealed preferences are satisfied, this constitutes one reason for Dan to refrain from protecting her. Likewise, it may be (pro tanto) in Teresa's interest that others exhibit respect for her deeply held values and commitments, such as her pacifism. This too would give Dan some reason to refrain from protecting her. But while the refusal of protection will almost always give the would-be defender some reasons to refrain from intervention, I think it is rare that such reasons are weighty enough to tip the scale. It is in Teresa's interest that her revealed preferences are satisfied and that others exhibit respect for her values and commitments. Still, her interest in not being seriously harmed greatly outweighs these interests. This is especially clear in a version of Siblings where Adam will kill Teresa unless Dan intervenes. Even if we grant that Teresa has an interest in preference-satisfaction and respect for her pacifistic commitments, it is strongly in Teresa's all-things-considered interest that Dan intervene. So while refusal certainly gives the refuser more of an interest in the defender's refraining from intervention, this is not usually sufficient on its own to render otherwise-permissible protection impermissible. There must be more to the story. I think we must look beyond the mere interests at stake and consider the possibility that refusal throws additional rights-based obstacles into the mix. Rights have the sort of profile we need: they are the sort of things that impose very weighty constraints on action—constraints that can outweigh even the saving of a person's life. (I am not permitted, e.g., to kill one person so as to harvest his organs to save the lives of two others.) If refusal contributes a right such that protective intervention would infringe that right, this would nicely explain why refusal often undermines the permissibility of such protection. In this section, I articulate one version of such an account. To introduce the account we should start by noticing an important feature of our rights against harm: when these rights are threatened by an attacker, some of the attacker's own rights against harm are thereby lost. Specifically, the attacker comes to lose those rights that “stand in the way” of the defense of your right: he loses his rights against (necessary and proportionate) harm that is a means or side-effect of preventing him from infringing your right.12 Importantly, an attacker does not just lose such rights with respect to his target; he loses these rights with respect to any potential defender. When Adam attacks Teresa, for instance, he loses his right that Teresa not harm him as a means or side-effect of averting his attack—but so too he loses his right that Dan not harm him as a means or side-effect of averting his attack. Adam makes himself liable from all sides, so to speak. This is the default effect of posing a threat to another's rights against harm. Earlier we noted an uncontroversial way in which it is in a target's power to mitigate these effects: she can waive the right that is threatened. By waiving that right, she thereby restores all of the rights that had been lost by the attacker. He regains those rights he had lost with respect to his target as well as those rights he had lost with respect to other potential defenders. The power to waive our rights offers us a form of control over our rights, but it is a very all-or-nothing form of control. We enjoy better control over our rights, however, if we also enjoy more fine-grained powers with which to mitigate their protections.13 If considerations of control justify admitting into our normative theory the power to remove the complete package of protections afforded by rights, these same considerations should justify admitting in the power to remove only parts of the package. And it often matters to us that we have this more fine-grained form of control. Consider, for instance, the case of Siblings. Teresa is a pacifist, who does not want others to defend her and who does not want to defend herself either. Still, she does not want to be harmed, and she wants to be compensated if she is harmed. Rather than waive her right in its entirety, Teresa would prefer the power to remove the default permission that others have to protect her without also giving up her deontic and compensatory protections. Or to take a different case: imagine a professional mixed martial artist is attacked on the street; an onlooker can defend her, but the fighter can just as well defend herself with the same amount of defensive force. And suppose it matters to the fighter that her life be characterized by self-sufficiency and that she be the one to defend herself. The fighter would not want to waive her rights in this case, since she wouldn't want to give up her permission to defend herself. She would only want the power to give up the permission that other people have to defend her. I propose that we have just such a power to remove some—and only some—of the liability protections of our rights without also removing the deontic and compensatory protections of our rights. Specifically, we have the power to restrict the domain of persons with respect to whom our attacker is liable. Where the normal effect of an attack is to make the attacker liable to any potential defender, this effect can be mitigated by removing the attacker's liability with respect to only some potential defenders. It is this very power, I claim, that is typically exercised by the act of the refusal of protection, and that explains the bulk of refusal's moral power. When Teresa refuses Dan's protection, for instance, she does not waive her right. But she does mitigate one of the normal effects of the threat to her rights: she makes it so that Adam is no longer liable to be harmed by Dan. This is why Dan lacks a permission to defend Teresa. And because Teresa only removes Adam's liability with respect to Dan, Adam remains liable to be harmed by Teresa. This is why Teresa—unlike if she had simply waived her right—retains her right against Adam's attack, retains her permission to defend herself, and retains her conditional right to compensation if the attack succeeds. Call this the Liability Contraction Account (“Contraction” for short). The basic idea of Contraction to hand, let's unpack some of the implications and distinctive of the account. Suppose Dan disregards Teresa's demand that he not protect her. He breaks Adam's leg, thereby preventing Adam from breaking Teresa's leg. Contraction tells us that Dan acts impermissibly, since Adam is not liable to have his leg broken by Dan. Contraction, then, implies that Dan violates Adam's rights. This implication will give many readers pause. If Dan wrongs anyone, isn't it Teresa rather than Adam? Slasher: Jill validly consents to have sex with Jack later that night. Bloggs knows what Jill has said to Jack, but due to certain sexist assumptions has the belief that Jill lacks the power to waive her right not to be intimately touched by Jack. To “protect” Jill, Bloggs slashes the tires on Jack's car so that he can't drive to Jill's. Given Contraction, Dan would wrong Teresa in this same way by disregarding her demand that he not protect her: Dan would treat Teresa as lacking powers she has to mitigate some of her own rights-based protections. Contraction, then, nicely accommodates the intuition that Dan's intervention would infringe Teresa's rights. But what should we say about the implication that Dan also infringes Adam's rights? This will seem counterintuitive to many, in light of the further implications this may seem to have. It may seem to follow from the fact that X infringes Y's rights that (i) Y has the standing to complain about the infringement, (ii) Y would have permission to defend himself from the infringement, and (iii) X owes Y compensation for the infringement. But it seems implausible that any of these things are true with respect to Adam and Dan. If Dan disregards Teresa's demand and intervenes to protect her, it seems that Adam has no standing to complain, no permission to use counter-defensive force against Dan, and no right to compensation if Dan breaks his leg. In fact, however, this is compatible with the claim that Dan infringes Adam's rights. First, it does not follow from the fact that X infringes Y's right that Y has the standing to complain about the infringement. Suppose I regularly break promises that I make to others. For this reason, I will lack the standing to complain if you break a promise to me: to complain would be to hold you to standards to which I do not subject myself. Nonetheless, this does not mean that you fail to infringe my rights by breaking your promise. The fact that I regularly break promises to others does not make it impossible for you to make binding promises to me—to make promises that impose on you a directed duty (and on me a correlative right) that you do the thing you promised to do. I think the very same considerations are true of Adam. Even on the assumption that Dan infringes Adam's rights, we should think that Adam lacks the standing to complain. He lacks the standing to complain that his leg is broken for the very same reasons I lack the standing to complain about your broken promise: to complain in these circumstances would be for Adam to hold Dan to standards (“don't break someone's leg without justification”) to which he does not subject himself. Miners: Three miners—Alpha, Beta, and Gamma—are deep in a mine when they encounter a harmful gas. If exposed to the gas for an extended period, they will suffer moderate lung damage. The three miners quickly put on their gas masks. Alpha, however, is jealous of Beta's recent promotion, and attacks him, destroying his gas mask. Beta now must choose between suffering lung damage or stealing Gamma's mask (it is impossible for him to take Alpha's mask by force). He decides to attack Gamma and steal her mask. Now it is Alpha who faces a choice: he can defend Gamma either by knocking Beta to the ground (in which case Beta will suffer the lung damage) or by giving up his own gas mask to Beta (in which case Alpha will suffer the lung damage). I think Adam is forbidden from defending himself against Dan in Siblings for the same reasons that Alpha is forbidden from imposing defensive harm on Beta in Miners. Alpha has a duty to assume the costs of Beta's attack on Gamma because of facts about his responsibility for the conditions of that attack. Alpha is responsible for breaking Beta's gas mask. And by breaking Beta's mask, he puts Beta in a position in which Beta can act permissibly (i.e., not attack Gamma) only at great cost (i.e., damage to his lungs). This fact greatly mitigates Beta's own responsibility for the threat that he poses to Gamma. This difference in responsibility between Alpha and Beta, I believe, is what explains why Alpha has a duty to assume the costs of Beta's threat to Gamma (i.e., why he is not allowed to let those costs fall on Gamma or to redirect them onto Beta). We get a similar difference between Adam and Dan. Adam is highly responsible for his threat to Teresa. By posing this threat, Adam puts Dan in a position in which the latter can act permissibly (i.e., refrain from protecting Teresa) only at great cost (i.e., a broken leg for his sister). This fact, I think, mitigates some of Dan's responsibility for his wrongful threat to Adam. Given this difference in comparative responsibility between Adam and Dan, Adam has a duty to assume the costs of Dan's attack. He isn't liable to be targeted. But once he has been targeted, Adam then takes on a duty to assume the harm. And here this duty doesn't require Adam to actively redirect any sort of costs toward himself (as in the Miners case); it requires him only to abstain from redirecting the threat back toward Dan. This is why Adam is not permitted to defend himself from Dan's wrongful threat. This is also why Adam is not owed compensation when his leg is broken. The reasons why Adam has an ex ante duty to assume the costs of Dan's attack—namely, the fact that it is more just if the costs fall on the party most responsible for those costs—are the same reasons why Dan lacks an ex post duty to assume the costs of Dan's attack. It's important to notice, however, that these considerations do not generalize to every case of refused protection. This is because the attacker won't always bear a greater share of responsibility than the defender. Suppose, for example, that Adam's responsibility is also mitigated: a villain has promised to harm his mother if he does not attack Teresa. We can imagine that the villain's threat is so serious as to make Adam no more responsible than Dan for the costs that must now be distributed. Under these conditions, it doesn't seem that Adam would have a duty to assume the costs of Dan's threat. Thus, given Contraction, Adam woul