Jump to ContentJump to Main Navigation
The Constitution of the Criminal Law$

R.A. Duff, Lindsay Farmer, S.E. Marshall, Massimo Renzo, and Victor Tadros

Print publication date: 2013

Print ISBN-13: 9780199673872

Published to Oxford Scholarship Online: May 2013

DOI: 10.1093/acprof:oso/9780199673872.001.0001

Criminals in Uniform

(p.97) 6 Criminals in Uniform
The Constitution of the Criminal Law

John Gardner

Oxford University Press

Abstract and Keywords

This chapter addresses the relationship between the character of the state and the constitution and the justification of harm-causing action. It takes as a starting point the author's debate with Malcolm Thorburn about the Diceyean interpretation of English constitutional law; specifically the way in which we should understand the contrasting moral positions of citizens and public officials with respect to the justifications that they may offer for their harmful actions. It suggests that a defender of the Diceyan doctrine can consistently believe that public officials, or some of them, are in a special moral position because they are officials. By considering the role of police officers in some depth, the special characteristics of official conduct and misconduct are illustrated whilst endorsing a fundamental continuity in the kinds of consideration that are relevant to assessing their conduct and the conduct of ordinary citizens. Such a discussion invites us further to reflect on what it is to be a citizen and how the criminal law should be constituted in the light of such a notion.

Keywords:   English law, criminal law, constitutional law, Malcolm Thorburn, police officers, policing

I. Dicey’s Doctrine

In England the idea of legal equality, or of the universal subjection of all classes to one law administered by the ordinary courts, has been pushed to its utmost limit. With us every official, from the Prime Minister down to a constable or a collector of taxes, is under the same responsibility for every act done without legal justification as any other citizen. The Reports abound with cases in which officials have been brought before the courts, and made, in their personal capacity, liable to punishment, or to the payment of damages, for acts done in their official character but in excess of their lawful authority.1

It is often suggested that this account of the legal position in England, even if true at the start of the twentieth century, was falsified during the course of that century by the rise of modern administrative law. But the rise of administrative law is totally irrelevant to Dicey’s point. Administrative law adds new ways to hold public servants and public authorities responsible in law, but it does not take away any of the old ones. Police officers and tax inspectors may still be sued in the civil courts for trespass, defamation, and breach of confidence, just like the rest of us. Prison guards and traffic wardens may still be prosecuted in the criminal courts for theft, perverting the course (p.98) of justice, and blackmail, just like you and me. That MI5 officers, immigration officials, and local authority librarians are banned from murder, torture, and kidnapping comes of the fact that everyone is banned from murder, torture, and kidnapping. That at the time these people were or purported to be occupying their official roles, or fulfilling their official functions, does not in itself give them any legal protection against any criminal charge or cause of action to which they would otherwise be vulnerable.2

Of course it is true that many public officials have extra legal permissions that are incidents or constituents of their role. Police officers, for example, have extra permissions to enter onto land, seize goods, and detain persons, all of which give them extra scope to defend their actions in a criminal or civil court beyond what would be possible for an ordinary member of the public in the same circumstances. Dicey never denied this. His point was only that any such extra permissions need to be pleaded as a defence in the ordinary way in an ordinary court in the face of an ordinary charge or claim for an ordinary crime or tort. Both police officers and employees of private utility companies have special permissions to enter onto land. Both customs officers and private bailiffs enforcing judgment debts have special permissions to seize goods. The legal question, when any of these people is sued for trespass or conversion, or charged with burglary or theft, is simply whether he or she had the special permission that he or she claims to have had, not whether the office in virtue of which he or she had it was in the public or the private sector.

This is the ‘citizens in uniform’ doctrine. There is some poetic licence in the name, because the literal presence of a uniform is neither here nor there,3 and nor, strictly speaking, is anyone’s citizenship. Yet the name conveys the core of the doctrine. It can be expressed as an admonition: ‘Don’t think that when you step into your official role (your “uniform”) you stop being yourself and can abdicate responsibility in your capacity as an ordinary member of the public (a “citizen”) for the things that you do. You still answer to the law as yourself, and you can’t hide behind your public role when you do it.’

Dicey famously idealized this doctrine, which he rightly took to be a settled doctrine, not only of English law, but also of the British Constitution. (p.99) He cast correspondingly infamous aspersions on the Napoleonic alternative,4 in which (he claimed) the existence of a special droit administratif5 with special courts of its own was coupled with extensive immunities for public officials from the ordinary processes and liabilities of criminal and private law. As is well known, he thought that English law here conforms to the ideal of the rule of law in a way in which French law does not. In this he was both right and wrong. It depends on how one understands the words ‘in a way in which’.

Let me explain. The rule of law, like all credible political ideals, has universal6 and parochial aspects. Any legal system that does not include an effective way of holding public officials to account in its courts has on any credible view a very serious rule-of-law deficit. It might even be reasonable to claim, on the strength of this deficit alone, that the rule of law does not prevail in that legal system, or in the civilization of which it is the legal system. But the French legal system is clearly not like that. It has an alternative way of holding public officials to account in the courts, namely by providing special courts in which to do so. Given other features of the French legal system, or of French civilization, this may well be the best way of conforming to the ideal of the rule of law in France. By the same token, given other features of English law, or of English civilization, the English way may be the best way of conforming to the ideal of the rule of law in England. We would have to know a lot of other things about each country and its law before we could judge whether things would be better than they are now, from the point of view of the rule of law, if the English adopted the French model, or if the French adopted the English model. It is absurd to think that the rule of law gives its universal blessing to one model over the other. Clearly it allows that different legal systems may find their own ways of conforming, possibly with equal success, to the desideratum of judicial accountability for public officials.

(p.100) So Dicey is right that the English constitution conforms to the ideal of the rule of law in a way in which the French does not, if that is taken to allow for the possibility that the French model also conforms to the ideal, but does so in a different way. But the statement is wrong if it is taken to mean (as Dicey might have read and approved it) that the English constitution conforms to the rule of law in a respect in which the French does not. In one and the same respect—namely in respect of holding public officials accountable before the courts—French law and English law conform to the ideal, but each system has its own way of doing so. That allows us to say, if we like, that there is a French ideal of the rule of law and an English ideal of the rule of law, but that both are parochial renditions of the same ideal, viz the universal ideal of the rule of law. (I bracket the vexed question of whether Dicey got the French position entirely right.7)

Malcolm Thorburn and I have argued about the Diceyean interpretation of English constitutional law before.8 He denies that English law includes the ‘citizens in uniform’ doctrine. He also denies that it would be a good thing, from the point of view of the rule of law, if it did. (I argued with him only about the first point, but as he gleaned from my tone, I tend to disagree with him about the second as well.) One very interesting feature of Thorburn’s challenge is that he does not follow Dicey in contrasting the ‘citizens in uniform’ doctrine with the supposedly Napoleonic alternative, the model of liability in droit administratif combined with immunity from various criminal and civil processes and liabilities. Instead, Thorburn contrasts the ‘citizens in uniform’ doctrine with what I will call the ‘officials in plain clothes’ doctrine. According to this doctrine, there is a body of rules belonging to the general part of the criminal law that are tailored primarily to regulating public officials in the exercise of their official powers. In this respect, criminal law is a type of public law, a sibling of administrative law.9 To be more exact: using a cluster of ‘justification’ defences to criminal charges, the law licenses officials to arrest, to detain, to seize, to search, and otherwise to engage in diverse acts of public defence and public necessity, which are incidents of their specifically public roles. It is true, agrees Thorburn, that versions of the same justification defences are sometimes available to non-officials too. The criminal law permits me, as an ordinary member of the public, to defend and protect people, including myself, in ways that would otherwise incur criminal (p.101) liability. In some cases I am even permitted (and empowered10) to make an arrest. When such defences are available to me, says Thorburn, their aim and effect is to make me a public official pro tempore.11 By hypothesis, I have no standing role as a public official, but the criminal law is licensing me to act as a public official right now, in the absence of anyone nearby who has that standing role. So whereas Dicey and I think of the arresting police officer as a citizen in uniform, Thorburn thinks of the ordinary self-defender or defender of others as a stand-in plain-clothes police officer.

In recent work, Thorburn has renewed and redoubled his criticisms of the ‘citizens in uniform’ doctrine. He now writes:

According to the Diceyan ideal, there is nothing particularly special about the work of public servants. What makes a police officer a police officer, on this account, is little more than his pay cheque: he is paid to do certain things that he would be permitted to do anyway. The aspect of this story that proponents of the Diceyan account like to emphasize is that police officers are not privy to any special treatment – they are held accountable in criminal law in precisely the same way as the rest of us. But the obverse of the same story is the recognition that there is nothing special about the public provision of certain services: nothing of principle would be lost if we were to privatize all public functions: policing, prisons, even the military. Of course, advocates of the Diceyan position might suggest that there are pragmatic reasons for keeping some of these services in public hands, but they would not have any objections in principle to our doing so. I have argued elsewhere that there are important reasons of principle to object to what Paul Verkuil calls ‘outsourcing sovereignty’ – but they are not reasons that the Diceyan account can articulate.12

One curiosity of this ‘privatization’ objection is that, if sound, it would seem to count against the ‘officials in plain clothes’ doctrine no less than it counts against the ‘citizens in uniform’ doctrine. On Thorburn’s own view, what is a legally approved act of self-defence or defence of others by an ordinary (p.102) member of the public if not a temporarily privatized performance of a police duty? And if such selective privatization of policing is warranted, then why is it not mainly a ‘pragmatic’ question how far it should be extended—for example, what the balance should be between police arrests and citizens’ arrests?

I will not pursue this tu quoque response to Thorburn here. Instead I want to ask whether he is right about the moral and political implications of the Diceyan view. I ask because I am in an awkward position if he is right about these implications. If he is right, my views about the best way to implement the ideal of the rule of law in the UK, my own country, (viz by retaining the citizens in uniform doctrine), do not square with my other political ideals. These ideals include a very deep-seated opposition to the transfer or contracting-out to non-public-sector bodies of various hitherto public-sector activities, including (but certainly not limited to) policing, tax inspection and collection, border and immigration control and enforcement, planning and licensing, welfare benefit assessment and payment, diplomacy and consular services, the work of the intelligence agencies and the armed forces, criminal prosecution, the prison and probation services, and—perhaps most crucially of all—the administration of justice in the courts. I will not attempt to defend this opposition here. Instead I will set myself the more limited task of showing that a defender of the Diceyan doctrine can consistently believe that public officials, or some of them, are in a special moral position because they are officials. In other words, they can consistently believe that there is something ‘special about the work of [at least some] public servants’ when they are occupying their official roles.

To that end, my main thesis will be that, if one is killed or injured or robbed or abused (etc), it makes things morally worse that one is killed or injured or robbed or abused (etc) by a police officer on duty. Consequently, the officer has a ceteris paribus harder job of defending his or her actions than he or she would have if he or she were an ordinary member of the public who did the same thing. You will notice soon enough, if you have not done so already, that the special moral position I will be sketching is that of a police officer qua police officer, not that of a police officer qua public servant. In adumbrating this special moral position I will be assuming, not arguing, that the job of a police officer is and should remain in the public sector. That is what I mean when I say that I will stop short of arguing against the privatization of the police. Nevertheless, I believe that once the special moral position of police officers qua police officers is appreciated, it is a fairly short step to seeing why that job should be kept out of private hands, and (in more specific opposition to Thorburn) why the rest of us should not be regarded, when (p.103) intervening with legally recognized justification in matters that also warrant police intervention, as police officers pro tempore.

II. The Morality of Policing

When passer-by Ian Tomlinson was deliberately pushed from behind during the London G20 protests in 2009, and died soon afterwards, the action of the man who pushed him was made more morally lamentable, and hence harder to excuse, by the fact that he, the pusher, was a police officer on duty.13 When commuter Jean Charles de Menezes was mistaken for a would-be suicide bomber at Stockwell tube station in July 2005, and then killed by bullets fired at close range while he was restrained, the already shocking killing was made more shocking, and put in need of more thorough scrutiny and more compelling justification or excuse, by the fact that the killers were police officers on duty.14 In explaining why, I will be focusing on these and similar fatal actions by police officers on duty. But let me emphasize from the start that the considerations I will be adducing are of far wider import. They apply no less to police collusion in phone hacking by journalists, to conspiracy between police officers and corporations or politicians to thwart lawful protests or other dissent, to sexual exploitation by police officers of those with whom they come into official contact, to arrest and detention on trumped-up charges as a technique of intimidation, and to obstruction by police officers of the investigation of possible crimes committed by police officers.15 They also apply to various officials who are not strictly speaking police officers, such as immigration staff, tax inspectors, customs officers, prison guards, intelligence operatives, and soldiers engaging with civilians, to the extent that their work is akin to that of the police. For simplicity I will use the label ‘police officers’ loosely to refer to the whole range of officials with policing duties. For our main interest is in explaining what, morally speaking, those duties (or some of them) are.

(p.104) A. The duty to protect

We all have moral duties16 not to kill people, and police officers are no exception. But police officers have an additional moral duty to protect people against a wide range of misfortunes, and inter alia to protect them against killing. As one young woman wrote after escaping from the Norwegian island of Utφøya, where many of her peers were killed by a lone gunman: ‘Just think of it, he dressed himself in a police uniform, the symbol of safety and support.’17 The suggestion here is not that an imposter dressing up as a policeman thereby acquires the moral duties of a policeman (although maybe he sometimes does). The suggestion is only that this killer lulled his victims into a false sense of security by creating the expectation that he would protect them from killing (there being news of a killer on the loose nearby). This expectation of protection is not a specifically Norwegian one. In places (unlike Norway) where police officers routinely fail to provide such protection, so that people come to expect and even accommodate the failure, there remains a sense in which the police are still failing to meet people’s expectations. They are failing to meet people’s expectations of how police officers should behave. These expectations are normative ones. The police force exists to protect people from (inter alia) killing and thus the moral duty to do so goes with the job, irrespective of how often it is breached and irrespective of whether holders of the job, or other people, ever recognize that fact.18

(p.105) There may come a point, to be sure, at which failures to protect, or other moral failures by police officers, become so endemic that doubts start to arise about whether they should still be regarded and treated as police officers. (‘What a pathetic excuse for a police force!’ ‘Call yourself a policeman?’) Do such doubts take us back to the problem of the moral duties of imposters? Not quite. There is a difference (although there are certainly borderline cases) between gangsters who dress up as police officers and police officers who behave like gangsters. There are non-moral criteria, in other words, for someone to count as a police officer. There must be. Otherwise those who fail badly enough in the moral duties of police officers (individually or collectively, as you like) are not police officers, and so do not have those duties, and so cannot fail in them.

Now, wearing a police uniform is clearly not one of the non-moral criteria for being a police officer, for in that respect police officers and imposters may be indistinguishable. How about working in the public sector? That would be a turn up for the books—privatization of the police is conceptually impossible!—but I doubt whether it could be sustained. What is among the non-moral criteria for being a police officer, it seems to me, is that police officers have special legal powers to do certain police-characteristic things, which remain their legal powers even when they are systematically abused. With such a criterion in place, ‘call yourself a policeman?’ and similar reproaches can be taken to convey that, in the opinion of the speaker, the addressee is not fit to be a police officer, or is a degenerate example of a police officer, or is a police officer not worthy of the name, all without denying that a police officer is still what he or she is. Likewise, when we doubt whether such degenerates should be ‘regarded or treated’ as police officers we may simply mean: don’t think of them as your protectors or saviours, don’t make them cups of tea, don’t help them with their inquiries, don’t give them the time of day. Don’t credit them, in short, with a fitness for their office of which they are not worthy. Compare the line we should take with bad laws: they are still laws, but errant examples that exert no moral hold over us and so should be ignored as guides to action (and perhaps also derided or subverted).19

(p.106) We will return in a moment to the special relationship of police officers to the law. For now our interest is in their moral duty to protect, whether this is reflected in the law or not. There are interesting questions about the stringency of this duty, and in particular about how much protection people should expect from the police. To what extent should people be expected, and hence presumably permitted by law, to protect themselves rather than relying on the authorities? Should those facing more specific or more immediate or more serious dangers get enhanced protection? The cases that concern us here do not, however, raise these questions. It is one thing to be a protector whose protective measures are found wanting. The question can then arise of what further protective measures, if any, should have been taken—of how far the duty to protect reaches. As the Utøya survivor’s comments bring out, however, it is quite another thing to be a protector who does the very thing from which he or she is supposed to be providing protection. If one has a duty to protect someone from killing, one breaches it in an especially grave way by killing that same someone oneself. For doing so is not a mere failure in, but rather an inversion of, one’s duty as protector. Killing A is as far away from protecting A from being killed as one can get. And the further away one gets from doing one’s moral duty, all else being equal, the morally worse one’s breach of duty is. In that dimension of moral evaluation, killings by police officers are among the worst there can be.

I say ‘among’ because police officers (even using that title as broadly as we are using it here, to include a range of police-like officials) are clearly not the only people with duties to protect others from being killed. Such duties—albeit owed to much smaller constituencies of people than the duties of the police—are normal incidents of some personal relationships, such as those between parents and their children, and between one spouse or partner and another. Killings of children by their parents or vice versa, and killings of spouses by spouses, share the morally exacerbating feature that I have just associated with killings by police officers. One is not only killed; one is not only not properly protected from being killed; one is killed by the very person, or one of the persons, who had the job of protecting one from being killed. That person did the opposite of their duty. So the class of worst killings, in the dimension of moral evaluation that concerns us here, is not limited to killings by police officers. Nevertheless, killings by police officers do belong to that class. In this dimension, being killed by a police officer is morally akin to being killed by one’s mother or one’s husband.

(p.107) Surely this claim cannot be extended to the killing in, for instance, the de Menezes case? True enough, the officers who killed Mr de Menezes could hardly have done a worse job of protecting Mr de Menezes. But they killed him—didn’t they?—in the course of performing their moral duty to protect the rest of us. Doesn’t that neutralize the special moral awfulness supposedly attaching to their actions? Some people think that an action that was one’s only way to perform one’s moral duty can never be at the same time a breach of one’s moral duty (or at any rate a breach of that same moral duty).20 Some people think (a different proposition) that a morally justified action can never be a breach of a moral duty.21 Some people think, moreover, that the killing of Mr de Menezes does not count as a breach of duty thanks to one or more of these propositions. But the propositions are both false, and are anyway irrelevant to the de Menezes case.

That the propositions are both mistaken is the galling lesson of Sophie’s Choice. ‘You choose just one of your children to send to his or her death’, says the Auschwitz Kommandant sorting those who will live from those who will die, ‘or I will send both of them to their deaths’.22 As a parent, Sophie has a duty to protect her children—both of her children—from being killed. Neither the fact that she was justified in handing over one of her children in order to save the other, nor the fact that handing over her daughter was the only way she had to do her duty to protect her son, entails by itself that she did not thereby breach that same duty as owed to her daughter. That thanks to his proposal Sophie could not but breach her duty to at least one of her children was the main point of the Kommandant’s sadistic plan. Indeed the plan, we can now see, was even more sadistic than that. The Kommandant did not simply make sure that Sophie violated her parental duty; he ensured that she violated it in a particularly egregious way. He ensured that she condemned one of her children to the very killing from which she, as a mother, was duty-bound to provide protection. He was out to destroy Sophie as a parent by actively involving her in the killing of her own daughter. That she was morally justified in being, and morally bound to be, so involved did not reduce the special moral awfulness of what the Kommandant got her to do.

You may say that not every killing by a police officer fits the tragic Sophie’s Choice model. Surely there are some cases in which a police officer no longer (p.108) has his or her usual moral duty to protect a particular person whom he or she has in his or her sights? We will come back to that question shortly. For now, my point is only that the special moral awfulness of a breach of duty, when one does the very opposite of one’s duty, is not neutralized by the mere fact that one had a conflicting duty to do as one did, nor by the mere fact that the breach was justified.

Still less, we should now add, is it neutralized by the mere fact that one was trying to do one’s duty, or by the mere fact that one’s breach would have been justified if the facts had been as one thought them to be. These are excusatory considerations, and they are the best that the luckless police officers who mistakenly killed Mr de Menezes can hope to draw upon. Looking more closely it would be seriously misleading for them to claim, as I put it before, that ‘they killed him in the course of performing their moral duty to protect the rest of us’. They can only say that they killed him in the course of trying (as assiduously as any police officer could?) to perform what they (on reasonable grounds?) took to be their moral duty to protect the rest of us. And this excuse, even with the parenthetical words included and the question marks deleted, clearly does not in any way diminish the gravity of the breach of duty which it is supposed to excuse. On the contrary, the gravity of the breach is one of the main factors that determines how hard it is going to be to make the excuse, for it is one of the main factors that determines what would count as acceptable steps to ascertain the facts, how much self-restraint or caution would be reasonable, etc.

We could sum up: that doing the opposite of one’s moral duty is a particularly grave breach of that duty does not entail, or even suggest, that it is a particularly blameworthy one. It may be wholly justified or excused, and thus entirely blameless. No doubt police officers occasionally have powerful justifications or excuses for killing someone whom they have a duty to protect, and are thus morally exonerated. Their powerful duties to others may be what give them these powerful justifications and excuses. My point, however, is that powerful justifications or excuses are exactly what they need if they are to be exculpated, because the breach of moral duty involved in killing someone whom one has a moral duty to protect is a particularly egregious one, one that calls for (as I put it before) more thorough scrutiny and more compelling justification or excuse than other killings, all else being equal. And even with such compelling justification or excuse, at least some police officers must, like Sophie, carry on their consciences these deadly breaches of duty. Exculpation, as surely they would tell you, is not the same as absolution; a faultless or blameless failure is not the same as no failure at all.

(p.109) B. The duty to uphold the rule of law

As well as being protectors, police officers are officers of the law, and have a moral duty to uphold the law, as well as to protect people, in their work. The two duties overlap, but each extends beyond the other. The role of the police as protectors is not limited to protecting people against breaches of the law. The role of the police as upholders of the law, meanwhile, is not limited to providing protection, but also includes upholding laws (often very stupid laws) which do not protect anyone. This reflects the fact that the duty of police officers to uphold the law is really but one aspect of their duty to uphold the rule of law, the moral ideal according to which everyone is ruled by, and hence answers to, the law. Not all of us have this moral duty. Indeed those of us who are not bound to the law by our occupations (or by other special relationships with it23) do not even have a general moral duty to obey the law, never mind to uphold it. The police are therefore in the awkward position of being morally bound to uphold the law, and to make it the case that people answer to it, even when (as the police themselves often well know) the law is an ass and has no legitimate hold over those same people. That is another harsh burden of office. It is one that police officers share with prosecutors, judges, and other officers of the law. They have a moral obligation to subject people to the law even when those same people have no moral obligation to submit to it.

The most obvious way in which a police officer can fail in his or her duty to uphold the (rule of) law, already hinted at in these remarks, is by acting illegally himself or herself. Breaking the law is not quite the opposite of upholding the law. The opposite of upholding the law is undermining it, whether or not by breaking it. But breaking the law is nevertheless a long way from upholding it, and is thus a serious breach of moral duty on the part of a police officer, even when the law she breaks is a stupid one. I should emphasize, as I did in connection with the duty to protect, that the fact that the breach is a particularly serious one does not entail that it is a particularly blameworthy one. It need not be blameworthy at all. There may be morally acceptable justifications or excuses for police law-breaking. Sometimes, in a country with laws that are not merely stupid but immoral, civil disobedience or conscientious objection by officers of the law, sometimes by judges as well as by police officers, is called for. My point is only that, since they are then (p.110) doing something of special moral gravity, their justification or excuse for breaking even such a terrible law needs to be an especially powerful one.

A twist in the tale is that sometimes the law itself will recognize the justification or excuse in question. The law is rather unlikely to recognize civil disobedience or conscientious objection as a justification or excuse, but it might well recognize self-defence or public necessity or duress or mistake of fact. When such defences are recognized by the law, and enable a police officer to be acquitted or not prosecuted according to the proper procedures for determining such things, there is no illegality in the relevant sense. To that extent, the moral duty to uphold the rule of law has not been breached, and no further justification or excuse for breaching it is called for.

There is a strong moral case, however, for the courts to insist that officers of the law produce particularly compelling justifications and excuses before being acquitted in such a situation.24 Such officers need to show themselves fit for their role as upholders of the rule of law. An officer who pleads provocation to a charge of murdering a suspect, for example, had better not be held to the ordinary standard of self-control applicable to the rest of us. To uphold the rule of law often requires tremendous reserves of self-control and someone who is only ordinarily self-controlled is not fit for police work.25 Equally, to uphold the rule of law often requires high epistemic competence. A police officer must be particularly free from bias, superstition, gullibility, and prejudice. She needs to be the sort of person who does not maintain easy assumptions or jump to conclusions. In the de Menezes case, the English law of mistake in self-defence and defence of others gave the police officers (just as it would have given you and me) very great latitude in respect of errors of perception and deduction in the lead-up to the decision to kill. Indeed, the reasonableness of the officers having mistaken Mr de Menezes for a would-be bomber is neither here nor there to English law, so long as they really did so mistake him.26 That degree of excusatory latitude for any of us, but especially for police officers, is in my view indefensible. The effect is that the police officers who were party to the killing of Mr de Menezes could not have been (p.111) convicted of murder in England. So the killing did not represent, under the heading of police illegality, a breach of the police duty to uphold the rule of law. But the fact that the law did not allow for their epistemic competence to be tested to make sure that they were fit to be police officers is a bad reflection on the state of the law.27

I say ‘under the heading of police illegality’ because, to repeat, the moral duty of police officers to uphold the rule of law is not only a duty to avoid their own illegality. It also includes, for example, a duty to apprehend law-breakers and to take them through the first stages of accountability to the criminal law: arrest, search, questioning, and charge. This duty to apprehend offenders can make it tempting for police officers, often abetted by politicians and journalists, to regard themselves as existing to protect ‘law-abiding citizens’ against the ‘criminal classes’. But this self-understanding on the part of police officers is already antithetical to the rule of law. Under the rule of law, even the most notorious and dangerous offenders are entitled to the law’s protection, just like anyone else. Police officers, as officers of the law, are there to protect the bad guys as much as the good guys, except to the extent that specific protections are withdrawn by the legal mechanisms of accountability to the law itself (arrest, search, detention, charge, remand, trial, sentence) or by the legal recognition of extreme exigencies with which the ordinary legal mechanisms (of arrest, etc) cannot be expected to cope. Subject to these strict exceptions, the police are there to protect looters against shopkeepers no less than shopkeepers against looters, anti-capitalist protestors against corporate interests no less than corporate interests against anti-capitalist protestors, burglars against householders no less than householders against burglars, paedophiles against neighbours no less than neighbours against paedophiles, and illegal immigrants against security guards sent to deport them no less than security guards against illegal immigrants.28 Indeed, the rule of law is not consistent with the police dividing the world up into ‘good guys’ and ‘bad guys’, or into ‘law-abiding people’ and ‘criminals’, or into ‘respectable women’ and ‘common prostitutes’, or into any other classes of people supposedly more and less entitled to the protection of the law and hence of the police as officers of the law. So when vigilantes say that they are merely law-abiding people out to protect themselves against criminals, (p.112) the police should reply that one is only ever as law-abiding as the last law one abided by, so that it is always an open question, in any situation of potential conflict, who, if anyone, is (going to turn out to be) a criminal, and hence subject to arrest and other legal consequences.29

These are but further aspects of the principle that Dicey calls ‘legal equality’. Unlike the more parochial ‘citizens in uniform’ doctrine, they are aspects that remain even where that principle is not pushed to its ‘utmost limit’. The duty to uphold the rule of law is breached when—and hence the rule of law does not prevail to the extent that—police officers hold themselves not to owe the same protection to some person or people (B or the Bs) on the basis that the police exist mainly to protect some other person or people (A or the As) against B or the Bs. So even though the officers who killed Mr de Menezes did not break the law, it is still possible that they breached their moral duty to uphold the rule of law in another way, say by regarding ‘terrorist suspects’ in general, or Mr de Menezes in particular, as a ‘target’, less entitled to the protection of the law than other people because of the ongoing threat that they or he supposedly posed to ‘ordinary Londoners’ or ‘decent people’. In the fevered days after the tube and bus bombings of 7 July 2005, and in the morally corrosive climate of the so-called ‘war on terror’, it would not be surprising (although it would not follow that it would be morally excusable) if at least some police officers fell into this trap, mistaking policing (which has nobody as its enemy) for war (which is waged only against an enemy).30

These remarks bring us back to the question, postponed a few pages back, of whether there are cases in which a police officer no longer has his or her usual moral duty to protect a particular person whom he or she has in his or her sights. Our discussion may seem to suggest a negative conclusion. But in fact the answer is more complex and invokes some distinctions which can be hard to draw sharply. Under the rule of law, nobody is an outlaw and nobody should be treated by the police as either above or below the law, or as being either more or less entitled to its protection. Nor can the law itself, consistently with the rule of law, make an outlaw of anyone, so as to license the police to treat him that way. But it does not follow that the police officer’s moral duty to protect (the one discussed in A above) is the same duty in all circumstances. It seems to me that, in respect of many circumstantial (p.113) variations, this duty probably resembles the ordinary moral duties not to kill and injure that all of us have. Probably there is a relevant moral difference, for example, between incapacitating wrongdoers in the course of their wrongdoing, and incapacitating others. Killing or injuring bystanders, hostages, passengers, and other innocents in the elimination or mitigation of a threat can sometimes be justified, but it usually remains a serious breach of duty even when it is.31 Those who are morally implicated in the same threat, on the other hand, may thereby lose some of their rights not to be killed or injured in its elimination or mitigation.32 This distinction, or something like it, applies to you and me in connection with our ordinary moral duties not to kill and injure people. Probably it, or something like it, applies also to police officers in connection with their special moral duty to protect people from killing or being injured. Thus, police officers who kill or injure for public protection are often, but not always, in Sophie’s Choice situations. If Mr de Menezes really had been a suicide bomber ready to detonate his explosives, then it seems to me that any attending officer’s moral duty to protect Mr de Menezes from being killed would have been abrogated to the extent urgently necessary and proportionate to the cause of eliminating the threat he posed to other people on the tube. Always bearing in mind, of course, that each officer would still have the distinct moral duty to uphold the rule of law in doing so.

III. The Last Line of Protection

I said that being killed by a police officer is morally akin to being killed by one’s mother or one’s husband. Each of these people is duty-bound to protect one from being killed, and each strays as far as it is possible to stray from performing that duty by becoming one’s killer himself or herself. In this (p.114) dimension, I claimed, killings by the police are among the morally worst there can be. But can things get worse still? Yes. There is a further aggravating feature in at least some domestic cases which also extends to many if not all killings by police officers.

Sometimes people have nobody else to turn to for protection beyond the supposed protector who turns on them. Think of children who are sexually abused by their parents, where one parent is the abuser as principal and the other is complicit, usually by turning a blind eye to the abuse, but sometimes in even more degenerate ways. Or think of teenagers hoodwinked into thinking they are going on a family holiday but in fact being kidnapped for forced marriage by their parents. Or think of spouses facing systematic domestic violence from which, gradually deprived by their spouse of any social infrastructure attuned to their urgent needs, they have (or reasonably believe that they have) no realistic means of escape. In such cases it is often not only that the supposed protector does the very things that he or she is supposed to be protecting against; often it is also that there is nobody else to protect against the protector-turned-abuser. It is no good to say: ‘Call the police!’ ‘Tell your teacher!’ ‘Get a court order!’ The victims may reasonably fear that outside interventions will only make things worse, and so do not think of them as providing a further line of protection. Often that judgement is encouraged and then exploited by the abuser. But it need not be encouraged or exploited in order to add yet more to the enormity of his or her wrong. That the abuser is de facto the last line of protection against abuse makes his or her protective duty all the more stringent, and his or her violations of it all the worse.

The victims’ fears of outside involvement are not always misplaced. Because of ways in which the rule of law ties their hands, or for other institutional reasons, police officers (and other officers of the law) may be unable to provide very effective protection against errant protectors such as these. Sometimes that may itself be a failure of duty on the part of the police, calling for strong justification or excuse (justification or excuse in which the fact that the rule of law ties their hands surely plays a large role). But note that the call for justification or excuse here is made more pressing by the fact that the ‘last line of protection’ consideration applies a fortiori to the police themselves. Under the rule of law, the police are there, inter alia, to give everyone somewhere else to turn, a final line of protection when other lines of protection, such as those that are supposed to be built into family relationships, break down. That applies equally to wrongdoers seeking protection from rough ‘community’ justice. They too are supposed to have the police at their disposal as a last line of protection against the ‘community’. So when police officers do the very thing that they are supposed to have protected one (p.115) against, say by leaving one to the mercies of the baying mob, there is a further aggravation that we have not yet detailed. They not only fail in their protective duty (bad enough); they not only turn that duty on its head (worse); if the rule of law prevails, they also leave those whom they are there to protect, with nowhere else to turn (the worst so far).

You may say that this is an exaggeration. One paltry police officer is rarely one’s last line of protection in the relevant sense, for one paltry police officer is not the police. True, if a police officer kills one in an isolated spot then in a sense he or she personally was one’s last line of protection; but in that sense any killer who kills one in an isolated spot is one’s last line of protection. So that is not the relevant sense. The relevant sense of ‘line’ here is the ‘thin blue line’, the police understood as an institution. In the Tomlinson case, for example, the deadly mistake may turn out to have been that of a single officer from whom the police as a force failed, admittedly, to provide protection, but whose actions were not, we may dare to hope, taken on behalf of the police force as a whole. Contrast the de Menezes case, in which—so it appears—the officers present acted in a concerted way and more or less correctly implemented police operational guidelines covering (what they took to be) the situation.33 Mr de Menezes was not just up against a rogue officer, or even just a rogue team of officers. He was up against a whole system that had, by the time of his death, turned from his protector into his assassin. He had nowhere else to turn because the system that had turned against him was, according to the principles of the rule of law, his last line of protection. That seems to me to make the de Menezes case, in one respect, even more morally troubling.

But here we hit a worry. I said that, if the rule of law prevails, the police are the last line of protection. So, if the rule of law prevails, the fact that it was the police who failed to protect one is a matter of particular moral moment. But if the police fail to protect one, in the systematic way just described, doesn’t that entail that the rule of law doesn’t prevail? In which case, surely, the failure is promptly rendered less morally momentous, because the condition for its special moral moment (‘if the rule of law prevails’) is not met? There are several reasons for doubting this argument. Most important, it seems to me, is this: those who have the moral duty to uphold the rule of law (or any other moral ideal) cannot rely on the fact that it is not upheld as any kind of mitigation for their failure to uphold it. The police are duty bound, as upholders of the rule of law, to maintain themselves as the last line of protection, and to prevent rogue forces (vigilante groups, private militias, (p.116) neighbourhood gangs, mafia protection rackets, renegade intelligence units, and such like) from usurping that position by providing a further line of protection operating above the law. When the police do not provide the relevant protection, it is no excuse that they never do provide it because they are no longer what Robert Nozick calls the ‘dominant protective agency’.34 It is their duty to be that agency. That fact forges a close connection, one of several, between the moral duty to protect that we discussed in section II A and the moral duty to uphold the rule of law that we discussed in section II B.

IV. Dicey’s Doctrine Revisited

It may seem that, in all this reflection on the moral position of police officers and police forces, we have lost sight of our original plan. I promised to explain how a defender of the Diceyan ‘citizens in uniform’ doctrine could consistently believe that police officers are in a special moral position because they are police officers. I have given extended attention to that special moral position, at times analogizing it to the moral position of parents, spouses, friends, and so on, without ever explaining how any of this was meant to relate to the Diceyan doctrine. All I established was (a) that I believe in the Diceyan doctrine as a way of implementing the rule of law in England, and (b) that I believe that the police are in a special moral position. I did nothing, you may say, to show that these two beliefs are consistent, never mind mutually conducive.

But is that true? My discussion was meant to show the appeal of the view that the differences between the duties of police officers and of other people, in other roles, are ordinary moral differences. Although police officers as such are indeed in special moral positions, there is no distinct ‘political’ morality applicable to them that displaces ordinary moral judgement. Morality is just morality, and it applies to people. It applies to public officials (judges, soldiers, parliamentarians, police officers, local authority librarians, etc) because they are people. They do not stop being people and hence do not stop being bound by morality when they put on their uniforms, or otherwise go on duty.

Of course when they go on duty there are some adjustments in their moral positions, some new moral duties and some new moral permissions, but that is simply because everyone’s moral position is affected by the activities they engage in and the roles and relationships they are in when they engage in (p.117) them. In this respect, becoming a police officer or a soldier is just like becoming a lover, an architect, a pen-friend, a journalist, a plumber, a member of cabin crew, a polar explorer, a TV chef, a hillwalker, or a foster-parent. It is a morally relevant change in circumstances; there are new tasks to perform and hence new moral duties to fulfil and, sometimes, new permissions to enable them to be fulfilled well. I italicized ‘in this respect’ to remind you that the various roles on this list are wildly diverse in other respects. It is very important not to think, for example, that police officers owe protection to the public on the same relational basis on which Sophie owes protection to her own (more than she does to other people’s) children. A police officer who stops to ask whether the people that need to be rescued from that derailed train or hijacked aircraft are ‘our’ people has lost his or her way, and should not be working for the police.35 But that is a problem distinct from the one we have been discussing. I did not suggest here that there is a single basis or rationale for moral duties, or that they all have a standard structure (eg second-personal, relational). I only suggested that they are all part of morality, and that morality binds police officers on duty just as it binds anyone else. Nobody can evade it by saying: ‘I work for the government now’. For morality goes with one wherever life takes one, and is never displaced by the supposed demands of one’s work, which always, on the contrary, answer to morality and give one, in themselves, no excuse when one violates it.

This view about the unity of morality seems to be what Thorburn has trouble accepting. For him it is ‘not at all obvious why ordinary citizens should ever have decision-making authority over their fellow citizens’.36 What he finds ‘particularly troubling about the exercise of such decision-making powers is that they seem to be wielded by individuals with no special status that could explain their authority to make such decisions’.37 And the main reason why he finds this troubling is that ‘if literally anyone can make decisions about others’ most basic interests in life, liberty, security, and property, then the law’s claim that each person is sovereign over herself and her basic interests is hollow indeed’.38 In other words, he thinks, there is one morality for the rest of us, and there is another for the authorities (even if we sometimes cross the line and stand in for them pro tempore). The two moralities have different subject matters. One is the morality of authority and coercion. The other is the morality of, I suppose, everything else. But this underestimates the role of both authority and coercion, their justified role, in (p.118) ordinary life (and not only with children). Meanwhile it overestimates the role of authority and coercion in the work of public officials. We deliberately did not focus attention, here, on the use of coercion or authority by police officers. We focused attention instead on killings by police officers. That they kill (or injure, exploit, lie, intimidate, conspire, discriminate, steal, etc) in their role as police officers makes a number of moral differences, some of which I sketched. But note that the moral differences as I sketched them are always ordinary moral differences, of a type that we all encounter in our daily lives as we move among our various roles and relationships and try to adapt to their circumstances. As Thorburn anticipates that I will say,39 being a police officer—and, I would add, a public official of any kind—is just another of those circumstances to adapt to.

Thorburn’s resistance to the Diceyan ‘citizens in uniform’ doctrine comes mainly, I think, from his denial of the unity of morality. My relative ease with the Diceyan doctrine is consonant, in turn, with my acceptance of the unity of morality. In the unity view, to repeat, we are still people, and still answer to morality, even when we put on our uniforms. Likewise, according to the Diceyan doctrine, we always answer to the law as ourselves, and we can’t hide behind our public roles when we do it. That doctrine about responsibility before the law parallels closely the moral picture that I have tried to sketch out. Of course, to repeat a point I made in section I, I don’t believe that the Diceyan doctrine is the only way to go about regulating public officials that is compatible with that moral picture. The Diceyan doctrine is a particularly radical solution that emphasizes the fundamental similarities between the moral position of public officials and the rest of us, and plays down some of the local differences. In particular, it doesn’t attach importance to the mere fact that one is a public official in the way morality in the raw (sometimes) does. Dicey did not disguise the fact that his doctrine was a particularly radical solution. On the contrary, he relished it and trumpeted it. Recall his perhaps excessively proud words: ‘In England the idea of legal equality, or of the universal subjection of all classes to one law administered by the ordinary courts, has been pushed to its utmost limit.’


(*) University of Oxford. I was lucky to have detailed written comments on earlier drafts of this chapter from Matt Clayton and Antony Duff. Both advised me to drop what was my focal example of official breach of duty from section II, and doing so has, I think, much improved the chapter. Many thanks to both (without landing either of them with any liability for the new section II). For other helpful criticisms and comments I would like to thank Niki Lacey, Jeff McMahan, Victor Tadros, François Tanguay-Renaud, Friederich Toepel, and numerous other participants in the Stirling criminalization conference (September 2010) and its Warwick follow-up (March 2011).

(1) A V Dicey, An Introduction to the Study of the Law of the Constitution, 6th edn (London: Macmillan and Co, 1902) 189–90. I understood our project title The Constitution of Criminal Law to be inviting reflections on the relationship between criminal law and constitutional law. Hence my broadly Diceyan themes.

(2) Although it certainly adds some, notably misconduct in a public office, which is both a crime and a tort at common law. At common law, bribery (in its ‘receiving bribes’ form) was also an offence that could be committed only by holders of a public office, but this restriction has recently been removed in England and Wales by statute. For some interesting and pertinent reflections on the reform, see Peter Alldridge, ‘Reforming the Criminal Law of Corruption’ (2002) 11 Criminal Law Forum 287.

(3) On uniforms more literally, see Christopher Kutz, ‘The Difference that Uniforms Make’ (2005) 33 Philosophy and Public Affairs 148.

(4) Dicey, Introduction, n 1 above, eg at p 190.

(5) As Dicey says (Dicey, Introduction, n 1 above, at p 486), what French lawyers call ‘droit administratif’ does not correspond to what English lawyers call ‘administrative law’. So I follow Dicey in leaving the expression untranslated.

(6) By speaking of the universal aspects of the ideal, I do not mean to suggest that the rule of law should prevail everywhere. It should prevail, in my view, only where there is or should be a legal system (or something very close to one), which is not everywhere. The rule of law has universal aspects in the less ambitious sense that I am after here so long as there are common desiderata that make it the same ideal, despite variations in its parochial interpretation or implementation, wherever it prevails or should prevail. For analysis of familiar doubts about whether the rule of law has universal aspects even in this sense, see Jeremy Waldron ‘Is the Rule of Law an Essentially Contested Concept (in Florida)?’ (2002) 21 Law and Philosophy 137.

(7) Cf Dicey, Introduction, n 1 above, at p 322, n 1.

(8) Thorburn, ‘Justifications, Powers, and Authority’ Yale Law Journal 117 (2008), 1070; Gardner, ‘Justification under Authority’ Canadian Journal of Law and Jurisprudence 23 (2010), 71.

(9) For further elaboration of the family resemblance, which for Thorburn goes beyond the particular feature I foreground here, see Thorburn’s ‘Criminal Law as Public Law’, in Antony Duff and Stuart Green (eds), Philosophical Foundations of Criminal Law (Oxford: Oxford University Press, 2011).

(10) Thorburn emphasizes powers but (I have argued) what the criminal law cares about must be permissions. Either can exist without the other. See Gardner, ‘Justification under Authority’, n 8 above, at 85–9.

(11) Thorburn, ‘Justifications, Powers, and Authority’, n 8 above, at p 1076. (‘It is only insofar as they are performing a public function that ordinary citizens have the authority to make such judgments, and accordingly, they are bound by similar normative constraints when deciding what conduct is justified as public officials would be in the same situation.’)

(12) Thorburn, ‘Two Conceptions of Equality Before the (Criminal) Law’, in François Tanguay-Renaud and James Stribopoulos (eds), Rethinking Criminal Law Theory (Oxford: Hart, 2011) 12. I have omitted the footnotes in which Thorburn refers to Paul Verkuil, Outsourcing Sovereignty: Why Privatization of Government Functions Threatens Democracy and What We Can Do About It (New York: Cambridge University Press, 2007), and to his own ‘Reinventing the Night-Watchman State?’ (2010) 60 University of Toronto Law Journal 425.

(13) ‘Ian Tomlinson unlawfully killed, inquest finds’, The Guardian, 3 May 2011. The officer was subsequently acquitted of manslaughter: ‘Ian Tomlinson death: Simon Harwood cleared of manslaughter’, The Guardian, 19 July 2012.

(14) Independent Police Complaints Commission, Stockwell One: Investigation into the shooting of Jean Charles de Menezes at Stockwell underground station on 22 July 2005 (London: IPCC, 2007). There has been no prosecution of individual officers involved in the shooting but the Metropolitan Police Service was convicted of an offence under the Health and Safety at Work Act 1974. ‘Met police guilty over De Menezes shooting’, The Guardian, 1 November 2007.

(15) Which was also a feature of the de Menezes case: Stockwell One, n 14 above, at pp 165–6.

(16) I put the word in the plural because—although nothing turns on the point here—I believe the moral duty not to kill intentionally is distinct from the moral duty not to kill accidentally (ie where the death is a side-effect, even a known one). Breaches of the second duty are in principle amenable to a wider range of justifications and excuses than breaches of the first duty. It does not follow that intentional killing is morally worse, even ceteris paribus: see John Finnis, ‘Intention and Side-effects’ in R G Frey and Christopher W Morris (eds), Liability and Responsibility (Cambridge: Cambridge University Press, 1991) 32 at 60–1.

(17) Emma Martinovic, ‘Norway Attacks: a survivor’s account of the Utφya massacre’, The Guardian, 27 July 2011.

(18) It is tempting to trace the moral duties under discussion here to the oath of office, or the contract of service, of a police officer. Although such oaths and contracts are indeed morally binding (so long as they do not purport to abrogate moral duties that exist apart from them and are not extracted by immoral means) it is implausible to think that they are the only or even the primary basis for attaching moral duties to roles, or for holding people to be bound by those duties. For the most part such oaths and contracts exist to solemnize, and hence to reinforce morally and psychologically, and sometimes to give legal effect to, moral duties that belong to the role anyway, thanks to the rationale for the role’s existence. So the police duty to protect cannot be avoided simply by refusing to undertake it. A police commissioner cannot relieve her officers of the duty by avoiding any mention of it in their oaths or contracts. Nor can a police officer avoid acquiring it by evading the taking of his oath of office, or by deleting the relevant words from his contract of service before accepting the job. For excellent discussion see Michael Davis, ‘Thinking Like an Engineer: the Place of a Code of Ethics in the Practice of a Profession’ (1991) 20 Philosophy & Public Affairs 150 at 156ff, and John Kleinig, The Ethics of Policing (Cambridge: Cambridge University Press, 1996) 238ff.

(19) This is all consistent with the claim, which I endorse, that someone who doesn’t understand the moral duties of a police officer doesn’t understand what it is to be a police officer (‘Call yourself a police officer? You don’t know the meaning of the word!’), or that someone who doesn’t understand how legal systems ought to be, morally, doesn’t understand what a legal system is. It is one thing to claim that understanding what an X is necessitates understanding how an X ought to be. It is quite another thing to claim that nothing is an X unless it already is how an X ought to be. In his Natural Law and Natural Rights (Oxford: Oxford University Press, 1980) at 11, John Finnis rejects the second claim about law, and yet he gives repeated succour to those who accept it by the curious way in which he expresses himself thereafter (viz by sometimes but not always reserving the word ‘law’ for cases of morally successful law).

(20) See eg Barbara Herman, ‘Obligation and Performance’ in Owen Flanagan and Amélie Rorty (eds), Identity, Character, and Morality (Cambridge, Mass: MIT Press, 1990) 311 at 324.

(21) See eg John Skorupski, Ethical Explorations (Oxford: Oxford University Press, 1999) 170–1; Stephen Darwall, ‘Morality and Practical Reason: a Kantian Approach’ in David Copp (ed), The Oxford Handbook of Ethical Theory (Oxford: Oxford University Press, 2006) 282 at 286.

(22) The scenario is from William Styron’s Sophie’s Choice (New York: Random House Inc, 1979).

(23) For example, naturalization as a citizen. For discussion, see J Gardner, ‘Relations of Responsibility’ in Rowan Cruft, Matthew Kramer and Mark Reiff (eds), Crime, Punishment, and Responsibility: The Jurisprudence of Antony Duff (Oxford: Oxford University Press, 2011) 87 at 95.

(24) Here I am building on my earlier discussion of the point in J Gardner, Offences and Defences (Oxford: Oxford University Press, 2007) 124–6.

(25) In 2000, a UK-wide police recruitment campaign (‘Could you?’) stressed that not everybody is fit to be a police officer, and in particular that extremely high levels of courage, patience, scrupulousness, and self-control are called for. For a report of the campaign, see http://news.bbc.co.uk/1/hi/902853.stm. Numerous recent cases of unprofessional police behaviour towards public protesters suggest that not all recruits, and maybe not all recruiters, took the point. See eg Moos and McClure v Metropolitan Police Commissioner [2011] EWHC 957 (Admin); R v Barkshire [2011] EWCA Crim 1885, to mention only examples where the unprofessional behaviour was also illegal.

(26) Settled by R v Williams (1984) 78 Cr App R 276 and Beckford v R [1988] AC 130, and put on a statutory footing by the Criminal Justice and Immigration Act 2008, s 76.

(27) To say that is not to advocate a departure from the ‘citizens in uniform’ doctrine because, as I explained, that doctrine only says that the police are answerable like the rest of us before the ordinary criminal courts on ordinary criminal charges, and cannot hide behind their uniforms as a way of defending their actions. It does not say that the standards of reasonableness used by the courts in assessing their guilt cannot be adjusted to reflect the higher expectations we have of police officers (or of others occupying professional roles, public or not). See Offences and Defences, n 24 above, at pp 128–9.

(28) I explored this point, or some applications of it, in my Offences and Defences, n 24 above, at ch 11.

(29) I have benefited greatly from reading James Edwards’ largely unpublished work on this topic (and on several others nearby).

(30) Here I am implicitly rejecting Bruce Ackerman’s attempt, in ‘This is not a War’ (2004) 113 Yale Law Journal 1871, to establish a tertium quid between policing and military action. To my mind, a fatal muddying of the waters, and a gift to enemies of the rule of law everywhere.

(31) Cf Michael Bohlander’s reaction to the famous hypothetical discussed by the German Federal Constitutional Court at 1 BvR 357/05 (2006) in which an air force pilot, now on policing duties, is ordered to shoot down a hijacked airliner in order to prevent its being used as a missile by the hijackers against a civilian target. Bohlander writes: ‘there is no balancing exercise; [the passengers] are, to put it bluntly, already dead’. See his ‘In extremis—hijacked airplanes, “collateral damage” and the limits of criminal law’ [2006] Criminal Law Review 579. Even Bohlander describes this view as ‘harsh’. It is hard to see why he would find it harsh if he really believed it. Not surprisingly, and rightly, the Federal Constitutional Court rejected it, although they bent over too far backwards in doing so, and concluded that the shooting down must be unjustified if the duty to the passengers still holds. Not so. This reasoning reflects the pernicious effects of the propositions about conflicts of duty and justified breach of duty that I rejected earlier (at n 20 and n 21 above).

(32) There is a very large literature on the scope and basis of this loss of right. François Tanguay-Renaud and I have published some reflections on the topic in ‘Desert and Avoidability in Self-Defense’ (2011) 122 Ethics 111, where citations to other recent treatments can also be found.

(33) Possibly they had not been given the order to implement them, itself a very serious breach which I am ignoring here. Stockwell One, n 14 above, at p 121.

(34) Nozick, Anarchy, State, and Utopia (New York: Basic Books Inc, 1974) eg at 54.

(35) For further reflections on this topic, see Gardner, ‘Relations of Responsibility’, n 23 above.

(36) Thorburn, ‘Justifications, Powers, and Authority’, n 8 above, at p 1118.

(37) Thorburn, ‘Justifications, Powers, and Authority’, n 8 above, at p 1125.

(38) Thorburn, ‘Justifications, Powers, and Authority’, n 8 above, at pp 1125–6, emphasis in original.

(39) Thorburn, ‘Two Conceptions of Equality’, n 12 above, at p 6.