My aim in this lecture is to reclaim the idea of proportionality for a rights-based theory of interpersonal morality and law. The topic gets its interest from two directions. Recent philosophical writing about proportionality analyses it in terms of some weighted balance of goods to be achieved against harms that are created in the process. There is a great sophistication in this contemporary writing about proportionality, and many important distinctions between different occasions, in particular, between proportionality in the use of defensive force, and proportionality in the context of a broader judgment about the justification of so-called ‘lesser evils’.1 Prominent writers in this area reject consequentialism, restricting both the catalogue of goods to be weighed and the range of morally significant difference in weight. I will argue for a fundamentally different way of thinking about proportionality, which focuses not on the relation between good results and bad side effects, but instead on the restrictions on the use of means of enforcement that are otherwise available. Like many of these contemporary writers, I will treat the example of proportionality in the use of defensive force against wrongdoing as a central example. In addition to this philosophical interest, the topic has recently acquired a legal and moral urgency, as a result of the recent development of what purports to be a more narrowly rights-based perspective on the use of defensive force. In a number of American states, the law of self-defence has been modified to include what are called ‘stand your ground’ provisions, which entitle someone when faced with aggression or wrongdoing to use lethal force to prevent it. As the name suggests, under these regimes, there is no duty to retreat when faced with a wrongdoer. Instead, you can stand your ground against him. There are many objections that might be made to the application of these laws – the extent to which they depend upon the defender's subjective beliefs about the danger posed by the supposed aggressor,2 and the way in which on-going patterns of discrimination seem to be reinforced by them. Without in any way meaning to suggest that these are minor difficulties, I hope to use a discussion of proportionality to raise a different objection to them; in particular, focusing on the way in which they misunderstand the nature of rights and the nature of their enforcement. Stand your ground laws raise issues about necessity as well as proportionality; where traditional Anglo-American ideas about defensive force suppose it to be necessary only when retreat is impossible, stand your ground laws permit force as an alternative to retreat. The central issue is proportionality: the laws in question stand the defender's entitlement to stand his ground (not merely to preserve his life) as sufficient to justify the use of lethal force. In order to defend the idea that proportionality is about the acceptable use of means rather than the relation between good results and bad side effects, I will, as advertised in the title of this lecture, apply some philosophy. In particular, I will focus on the very abstract philosophical distinction between constitutive and regulative principles. The basic form of my argument is that the operation of proportionality in cases of defensive force depends upon a distinction between the principles governing the rights that individual human beings have as against each other with respect to the security of their bodies, and the independence of those bodies from other people's actions, on the one hand, and a regulative principle governing the way in which a public authority can authorise the enforcement of these first-order rights. I will argue that a different set of norms applies to the public authorisation of enforcement of rights. These norms do not change the underlying rights; although I cannot use disproportionate force to prevent you from touching me, that limitation does not show that you are permitted to touch me. Nor do the norms governing enforcement show that the rights are themselves restricted in any way, or that each person's right to the security of his or her body is a matter of degree, depending on how bad it would be for others to stop them from interfering with it. To suppose that these two levels of practical thought must be identical is to make the mistake that Kant, in the Critique of Pure Reason, characterises as ‘subreption’, taking a principle that properly governs one level of thought and applying it to another. The philosophers who think that rights somehow take harm as their object introduce proportionality as an element of a univocal concept of permissibility, collapsing the distinction between a right and the grounds of its enforcement. The defenders of ‘stand your ground’ laws suppose that, because rights are morally fundamental, nothing can restrict their enforcement. Both conceptions seem to me to misunderstand the distinction between right and the manner of its enforcement. In focusing on defensive force, I do not mean to suggest that it is unique, or even paradigmatic to thinking about proportionality. Proportionality figures prominently in many other areas of practical thought. Many accounts of the moral emotions and many accounts of virtue incorporate ideas of proportionality; the properly regulated person does not overreact to things. Where virtue meets the theory of right – when you ask yourself the moral question of whether you should stand on some right that you have – virtue often speaks the language of proportionality. The question of what you are entitled to do is already settled; the question of what to do looks beyond right. It would be in some way excessive for you to claim this debt, or to expel this trespasser, or to proceed against this wrongdoer. Sometimes you might decline to stand on your right because it does not seem worth the bother; other times you might do so to avoid being what Aristotle once described as a ‘stickler in the bad sense’, the one who views interactions with others exclusively in terms of claims of right; in still others you might decline to stand on the right because too much is at stake for the other person and too little for you. In all of these instances, some assessment of proportion – how much – speaks to whether you should stand on your right. Indeed, one of the significant tasks of raising children is to teach them both that they have rights on which they are always entitled to stand and that they should not always stand on them. These cases of proportionality raise many interesting questions, but they are different from the question that I want to consider, not least because the question I want to consider seems to be a question of right: it is not merely that using excessive force to defend your right is a case in which virtue urges you not to stand on that right. In such cases, it is not that you should not stand on your right, but that you can be prevented with force from, or punished for, standing on it. The person who always stands on his or her rights may be callous or cruel; the one who uses disproportionate defensive force against a genuine attack is a criminal. There are also other instances of proportionality that are internal to right: on many prominent views, punishment is supposed to be proportional to the seriousness of the wrong being punished, and the law and morality of war raise issues of proportionality with respect to the grounds of going to war and with respect to what is euphemistically called ‘collateral damage’. Proportionality also figures in constitutional rights analysis, restricting the occasions on which a right may be infringed for a properly public purpose.3 I believe that the approach developed here contains resources for explaining how proportionality finds its way into these moral ideas, but I do not propose to do so here. I want to begin my explication with a characterisation of the way in which individual human beings have rights as against each other. Although I will begin by characterising these rights in moral terms, you may quickly begin to suspect that I am actually smuggling in legal ideas. That is not an accident: I actually think that the morality of interpersonal rights looks a lot like the legality of it, not merely because of a significant overlap in subject matter and substantive norms, but because of a common form of a moral problem. Sophisticated legal systems take positions in a way that has important implications for a wide range of moral issues, but they do so in a distinctive way. A legal system operates through its officials, specifying the questions they are empowered to answer and specifying the grounds on which they are supposed to answer them. Sometimes this is a way of achieving or enabling results that stand in no systematic relation to the form that their answers take; a policy concerned with climate change may be implemented by assigning tasks to particular decision-makers, although its subject matter is not concerned with decision-makers, but rather with solving an environmental problem. Others, however, take who is in charge of what question as their subject matter. The law of property is a familiar illustration of this structure, as it specifies the powers of owners, that is, sets out the range of questions an owner is entitled to answer. Closer to my immediate topic, the same structure shapes the law of persons. The moral question about our rights as against each other looks just like a typical legal question: who gets to decide what happens with your body, both the purposes for which you act and the ways in which others may affect it? The moral answer is that you do; the morally appropriate legal answer is the same, not because that rule is the most effective way of bringing about the desired moral outcome, but because it just is the moral outcome. This moral problem means that the morality of the situation needs to be given effect through legal institutions in order to be fully adequate to itself. The basic moral idea that I will suggest organises interpersonal rights to bodily security against the acts of others is that no person is in charge of any other person. I have argued elsewhere that this same idea organises the morality of private rights more generally,4 but nothing I say here depends on that broader claim. Instead, I will limit myself to the claim that it structures the rights that each person has to his or her own body. The thought is that nobody is in charge of anyone else's body. This has two familiar moral implications, in addition to several more that are less familiar but inseparable from them. The first familiar implication is that nobody else is allowed to so much as touch your body without your authorisation. You get to decide the purposes that you will pursue with your body (restricted by whatever rights others have their bodies, and whatever rights whoever it is has with respect to external objects, that is, property and perhaps shared public spaces). Nobody else gets to do so. Conversely (but ultimately just equivalently) you do not get to decide what purposes other people's bodies will be used for. That is why I cannot, for example, run my fingers through your hair in order to satisfy my curiosity about its texture. I am not allowed to do so, even if you are currently asleep or preoccupied, and so will not notice that I am doing so. This basic moral thought requires a distinction between two distinct ways in which I might wrong your body without your authorisation. The first of these is subordinating it to my particular purposes by literally using it. Running my fingers through your hair is among the most minimal instances of this first way, and it contrasts with a variety of ways in which I might affect your body, including, for example, bumping into it as I make my way through a crowd. I do not wrong you by bumping into you if I am jostled as I make my way through a crowd, but I do wrong you if I have the exact same effect on you by using your body for a purpose that you have not authorised. There are also some intermediate cases – tapping you on your shoulder to get your attention – that could be classified either as using your body without authorisation or, more plausibly, as akin to speaking to you; which it will be will depend on a variety of contextual factors. These difficulties of classification will figure later in my analysis; for now I merely want to draw attention to the fact that the apparent exceptions – jostling you and tapping you on the shoulder – seem to be categorically different, rather than different in degree. The second way in which I might commit a wrong in relation to your body is by being insufficiently careful around it, by doing the kinds of things that characteristically injure other people's bodies. It is not that I wrong you whenever I turn out to injure you. It is only if I injure you by doing something that characteristically causes injury – the type of thing about which you might, after the fact say, ‘you should have thought of that’ or ‘you should have been more careful!’ or ‘look where you are going!’ This contrasts with a non-wrongful type of injury that has found favour in philosophical discussions – as you flip the switch, you cause, through a pathway that nobody could have anticipated, a power surge that burns me. Not only is there no conceptual space for me to tell you should have been more careful, but there is also, equivalently, no space for me to complain that you have acted in a way that is inconsistent with my independence from you. These two thoughts – you should be careful, and you should look out for my safety – are equivalent, because in a system in which no person is in charge of anyone else, each person needs to restrict his or her conduct in light of the claims of others to independence. If I wrong you by injuring you, I subordinate your safety to my particular purposes. This is very different from the way in which I wrong you if I touch you without your authorisation; it is not that injuring you as a side effect of what I am doing uses you for my purposes; instead, it simply ignores your independence of me, by acting as if you weren't there. So although both types of wrong violate your independence, they do so in different ways. One of the most obvious ways in which someone can act that typically injures another is setting out to injure that person. But I want to suggest, perhaps initially counterintuitively, that deliberate injury is not wrong because it involves an intention to cause a bad result. Instead, deliberate injury is wrong because it is wrong in both of the two ways just outlined: it is (trivially and necessarily) the kind of thing that tends to injure. Second, typically when one person deliberately injures another, it is in the service of producing some further result. Most crimes, for example, are committed in the process of the criminal trying to gain money, a purpose that both law and morality regard as unobjectionable if pursued in appropriate ways. The criminal is not a criminal because of the end he has in view, but because of the means he uses in pursuing it; a murderer is someone who kills or uses force in a way that makes death a near certain consequence, as a way, that is, a means, of accomplishing some other thing. More generally, if I injure you as a way of achieving something, I have acted in a way that is inconsistent with your independence. But that inconsistency is objectionable because of the means that I used to achieve my end, not because of the (so far unspecified) end that I pursue. This brings me to the first of the less familiar features of the system of interpersonal rights. They are concerned exclusively with the means that you use, and not the ends for the sake of which you use them. I cannot touch your body without your permission because your body is not available to me as a means for pursuing my purposes. The requirement that I avoid injuring you restricts the ways in which I can pursue my purposes, not the content of those purposes. The second, less familiar feature is that this idea that no one is in charge of anyone else also leads to the conclusion that nobody needs to use his or her own body in a way that best suits the preferred purposes of others. Just as you don't need to let me run my fingers through your hair, so, too, you don't need to do anything to assist me in whatever other purposes I might have; that is, just as you do not need to make your body available for me passively (allowing me to run my fingers through your hair) so, too, you do not need to do so actively by assisting me or by accommodating me, organising your affairs so as to enable me to succeed at whatever and I have set myself. All you need to do is avoid wronging me. This may seem like a surprising conclusion, but let me pause and make it a little bit more familiar. The basic thought is that the restrictions on the means a person can use are formal – they set out the issue of who gets to decide about the purposes for which a particular body will be used. But that entails that the significance of those particular purposes – neither yours nor mine – does not figure in any way in determining what we may or may not do in relation to each other. That, in turn, entails that I am not allowed to touch your body or act in ways dangerous to it no matter how much is at stake for me, or how little for you. You do not need to be more careful around me when the stakes for me are higher – drivers do not need to drive more carefully when in wealthy neighbourhoods than in poor ones because an accident would cause more income loss,5 and they do not get to drive more dangerously when the stakes for them are personally higher. Conversely, it means that you do not need to assist me with respect to any purpose I might have, regardless of how important that purpose is to me or how easy or difficult it would be (in relation to your purposes) for you to assist me. That does not mean that as a matter of virtue you should not be more careful when the stakes are higher for others or lower for you. Nor does it mean that a public law regime cannot take account of these very factors, requiring you to move out of the way of emergency vehicles, demand that you render assistance in emergencies, and, to get to the concern of the next section, require you to stand down when the only way in which you could enforce your of right is by using disproportionate force. This simple picture is a picture of interpersonal rights, and I want to suggest it has much to recommend it, because it enables us to think of individual human beings as free in a morally significant sense, that is, that each is independent of each of the others. The simple picture also gives us the beginnings, but only the beginnings, of a picture of enforcement. There are three familiar forms of enforcement of rights, which we might characterise as prevention, correction, and punishment. I want to focus only on prevention, as it is the one that relates directly to my topic of defensive force. The simple picture suggests a simple account of protection: you are entitled to use force whenever it is required in order to prevent a wrong. I will argue that the simple picture is incomplete in what follows, but before I do so, I want to suggest that it must to be taken seriously. The basic thought is that what you are entitled to do is limited to what is consistent with the rights of others. If so, the question immediately arises about what may be done when someone acts in a way that is inconsistent with your rights (or the rights of other others). The simple answer to this question is that you can prevent them from doing it, as Kant puts it, to ‘hinder a hindrance to freedom’.6 On this view, the enforcement of a right isn't something over and above the right that is being enforced; it is just stopping the violation of the right. This view is not the view that two wrongs somehow manage to make a right; instead, it is a matter of doing what is necessary in order to prevent a wrong from occurring; force deployed to prevent illicit uses of force is part of a normative order in which all uses of force can form a consistent set. The simple picture includes a requirement of necessity – it only authorises the use of force in order to prevent wrongdoing, and so in cases in which the use of force is not required to prevent wrongdoing, or in cases in which more force is used than is necessary to prevent the specific wrong in question, the simple picture does not authorise it. But if it has a view of necessity, the simple picture has no conceptual space for any idea of proportionality between wrongdoer and defender. That is because the question of how easy or difficult, or how a permissible or forbidden act relates to the particular purposes of a particular person, simply cannot be expressed with the moral and conceptual resources of the simple picture. It does, arguably, have space for a narrow version of what Jeff McMahan has called ‘wide proportionality’,7 prohibiting uses of defensive force that would harm third parties. If the only way that you can stop me from running my fingers through your hair is to break both of my arms, the simple picture says that you do no wrong by doing so. But if the only way that you can do so involves also, either as a side effect or as a means, breaking the arms of other people who are nearby, you are not entitled do so, simply because you can not use your means in ways that use or injure those others. You are morally prohibited from doing something to enforce a right, but not because of any right on the part of the wrongdoer.8 But this is a narrow sense of wide proportionality because the urgency of your end doesn't figure in the prohibition. Instead, the fact that you are using force to defend yourself doesn't entitle you to use or endanger others in ways that are otherwise forbidden. I have been calling the simple picture simple, but I do want to draw attention to the fact that it has defenders. I mentioned the ‘stand your ground’ laws in the United States; George Fletcher has argued that something like the simple picture provides the only adequate analysis of rights enforcement. In a number of places he has drawn attention to a German criminal law case from close to a century ago in which a farmer shot at apple thieves.9 The farmer was acquitted. Fletcher applauds this decision, characterising its principle as ‘right never yields to wrong’. But although he applauds the principle, Fletcher also recommends what he calls ‘structured’ legal thinking, in which the morality of right gives way to a morality of what he calls ‘altruism’. Although right does not yield to wrong, Fletcher contends that the farmer should have been more considerate, given how significant the stakes were for the apple thieves, and that the law might in some situations demand such altruism. Fletcher is right to emphasise the structured rather than flat nature of legal thought, but mistaken in supposing it requires a departure from the realm of right.10 I will make this point by focusing on the concept of enforcement by a public authority. Before explaining why enforcement requires a public authority, I want to first make the thought that it does intuitively plausible. Let me do so by drawing your attention to two types of misconduct that are helpfully characterised by reference to the concept of a public authority. The first of these is vigilantism. The vigilante seems to do something wrong, even if – as is only sometimes the case – the vigilante punishes someone who is actually guilty, and gives that person just the punishment that he or she deserves. Although films sometimes celebrate vigilantes, they are morally troubling, because of the way in which they take matters into their own hands. The celebratory films typically manage to do so by using the second type of misconduct – corrupt officials or, as John Gardner nicely encapsulates it, ‘criminals in uniform’11 – as their foil. I do not think that it is an accident that viewers identify with the vigilante and find the movie satisfying because of the contrast with the corrupt official. From old Westerns through the latest Star Wars movies, the good vigilante is called upon to take up the role that the corrupt official has not merely forsaken but violated, by hiding wrongdoing behind a façade of right. Rising to that sort of occasion has a sort of nobility to it, which shines even brighter because of the contrast with the baseness of the corrupt official's conduct. The good vigilante does more than is required, at least of him, because the corrupt official does so much less. Nonetheless, there is something troubling about vigilantes and vigilantism. There is a question about what to do when you are confronted with a completely corrupt public authority, but I will not answer it here. Instead, what I want to do is draw attention to what is especially troubling about the corrupt official, and the mirror image flaw in the vigilante. The corrupt official uses the power accompanying a public role for purely private purposes, selling protection to the highest bidder, accepting bribes to look the other way, and so on. The vigilante is the mirror image of this, the one who takes it upon him- or herself to be the enforcer of law. As I said, when enough officials are sufficiently corrupt, perhaps vigilantism may become acceptable or at least defensible, or perhaps the concept of ‘official’ lacks application. But the antecedent of the conditional identifies the difference in standpoint; it is not that a vigilante stands morally on all fours with an officer of the law acting properly within his or her mandate. To the contrary, even the best vigilante is a defective substitute for a public-minded official. There is a distinctively public standpoint that public officials are supposed to take – they are not supposed to side with good people as against bad people, or to decide for themselves who is and is not without merit. Instead, what they are supposed to do – as police vehicles in Toronto put it – is ‘serve and protect’. Not just protect the virtuous, but protect everyone. And not just protect the law-abiding, and not, further, classify people into law-abiding/not law-abiding with respect to anything except the transaction immediately in progress. As Gardner puts it in another wonderful formulation, ‘one is only ever as law-abiding as the last law one abided by’.12 I want to use these two examples now to suggest that there is a general problem about private enforcement of rights, and that this problem is best expressed in ideas that are formally similar to the ideas that animate what I called the simple picture of rights, that is, both the general way of framing moral issues in term of who is in charge of what question, and the specific normative principle that says no person is in charge of any other. There is an important difference, however. The simple picture, which I have suggested governs the rights that we have as against each other, essentially says nothing about the grounds on which we should decide what to do. Virtue may advise us as to whether to stand on our rights, but right is silent on the matter, and, indeed must be if virtue is to have anything to say. But when it comes to what I will now call the public standpoint of enforcement, things look very different. A police officer dealing with a conflict has a mandate, a distinctively public one. The public mandate is the thing that the corrupt official disregards; it is the thing that the vigilante claims not to need in being able to take matters into his own hands. The public official differs from the vigilante precisely because the public official is supposed to be acting not in his own private capacity, but as an officer of the law, acting within a particular mandate. It is this mandate that introduces questions of proportionality. It might be asked why this public standpoint is required. I noted earlier that the moral problem with which I began looks strikingly similar to a legal problem. Let me fill that thought out by drawing attention to what I will call the generic legal problem. On the view that I will defend, law concerns itself with a fundamental moral problem about authority. Rather than using authority in order to get people to do what (legal officials think) they should do, the organising structure of law is that it fixes who is in charge of what questions, and the grounds on which the person in charge of a question may answer it. In so doing, it sets out a system of mutual restrictions on conduct, and the violation of those restrictions is, in a sense I will explain, by its nature coercive. Law thereby governs coercion by prohibiting people from acting in excess of their authority, and in so doing specifying who can respond when someone does so, and the manner in which the responder can respond. A systematic and public articulation of who is in charge of what questions thus contains both a specification of restrictions on conduct and a coordinate specification of what may be done to uphold that specification if it is violated. If the main business of the law is determining who is in charge of what question, then its task is also, secondary to this first task, determining when force can be used, how much, and by whom – when one person has committed or is in the process of committing a wrong, what can be done to address it, and by whom? I want to emphasise that this task is a distinctively moral one: it is not that we have an empirical problem of not knowing who is in charge of what; empirically, maybe everyone is in charge of everyone else, or nobody in charge of anyone. But of course ‘in charge of’ is already a moral concept. Instead, if you start with the thought that no person is in charge of any other,