A Journal of Philosophy, Applied to the Real World

The Duty to Remove Statues of Wrongdoers

Stockholm University



This paper argues that public statues of persons typically express a positive evaluative attitude towards the subject. It also argues that states have duties to repudiate their own historical wrongdoing, and to condemn other people’s serious wrongdoing. Both duties are incompatible with retaining public statues of people who perpetrated serious rights violations. Hence, a person’s being a serious rights violator is a sufficient condition for a state’s having a duty to remove a public statue of that person. I argue that this applies no less in the case of the ‘morally ambiguous’ wrongdoer, who both accomplishes significant goods and perpetrates serious rights violations. The duty to remove a statue is a defeasible duty: like most duties, it can be defeated by lesser-evil considerations. If removing a statue would, for example, spark a violent riot that would risk unjust harm to lots of people, the duty to remove could be outweighed by the duty not to foreseeably cause unjust harm. This would provide a lesser-evil justification for keeping the statue. But it matters that the duty to remove is outweighed, rather than negated, by these consequences. Unlike when a duty is negated, one still owes something in cases of outweighing. And it especially matters that it is outweighed by the predicted consequences of wrongful behaviour by others.

1. Introduction

This paper argues that if someone perpetrates serious rights violations, then this gives the state grounds to remove public statues of that person. By ‘public statues’, I mean statues on display in, for example, public parks, public squares, shopping centres and public buildings.1 (I suggest that there may be an exception for statues on display in certain types of public museums or galleries, and that it can be permissible to move statues of wrongdoers to those public locations.) While I focus on statues, my arguments also apply to portraits and other depictions of historical figures that are on public display. Much of what I say also applies, perhaps with some adjustment, to other types of cultural property that do not depict historical figures but are in some way connected to them, such as buildings named after such figures.

I argue that a person’s being a serious rights violator is a sufficient condition for a state’s having a duty to remove a public statue of that person. Keeping public statues of serious rights violators is incompatible with the state’s duties to condemn and repudiate serious wrongdoing. We should recognise that a range of features can ground a duty to remove a public statue. There are, for example, lots of reasons to remove statues of Confederate soldiers: amongst other things, such statues cause distress, (Timmerman, forthcoming) give credence to white supremacist views, (See ‘Take Em All Down’) and affirm existing social injustices (Moore, 2017).

The duty to remove a statue is a defeasible duty: like most duties, it can be defeated by lesser-evil considerations. Consider the following familiar example. I have promised to meet you for lunch, but pass a child drowning in a pond on my way to the restaurant. I can easily save the child, but not without missing our lunch date. While my promise to you confers on me a duty to meet you, that duty is defeated by my more stringent duty to save the child’s life. I thus justifiably fail in my duty to meet you for lunch, given the harms at stake. This is not to say that my duty to meet you was not a genuine duty—I still seem to owe you an explanation for my absence, for example, and perhaps an apology. The claim is rather that it was all-things-considered permissible for me to fail to discharge my duty in this case. I take this to be the general structure of lesser-evil justifications for failing in one’s duties. Contrast this with a case in which I promise to meet you for lunch, but then realise that you have no intention of showing up (perhaps you mistakenly send me an email revealing that our lunch date is a cruel joke on your part). In this case, my duty to meet you seems to be negated rather than outweighed—that is, I simply cease to be under a duty to.

If removing a statue would, for example, spark a violent riot that would risk unjust harm to lots of people, the duty to remove could be outweighed by the duty not to foreseeably cause unjust harm. This would provide a lesser-evil justification for keeping the statue. But it matters that the duty to remove is outweighed, rather than negated, by these consequences. Unlike when a duty is negated, one still owes something in cases of outweighing, as when I justifiably fail to discharge my duty to meet you for lunch in order to save the child. As I will argue, it especially matters that the duty is outweighed by the predicted consequences of wrongful behaviour by others.

In Section Two, I defend the claim that public statues of persons typically express a positive evaluative attitude towards the subject. I defend this claim against the view that statues are primarily non-evaluative historical records and against the view that statues honour achievements rather than persons. In Section Three, I argue that a subject’s having engaged in serious rights violations gives us grounds to remove a public statue. In Section Four, I argue that states have duties to repudiate their own historical wrongdoing and to condemn other people’s serious wrongdoing. Both duties are, I argue, incompatible with retaining public statues of people who perpetrated serious rights violations. In Section Five, I argue that this also applies in the case of the ‘morally ambiguous’ wrongdoer, who both accomplishes significant goods and perpetrates serious rights violations. In Section Six, I reject the view that we ought to remove statues of wrongdoers only when those statues cause harm. In Section Seven, I argue that the removal of statues is typically preferable to their recontextualization through explanatory plaques and the like. Section Eight concludes.

2. Statues as typically evaluative

My account is partly motivated by the view that public statues of historical figures typically express positive evaluative attitudes towards that figure. We build and display public statues primarily as a means of honouring people, and such statues thus express a claim that the depicted figure is worthy of admiration or respect. Not all statues do this: some statues are intended, for example, to mock or degrade the subject.2 My account does not support a duty to remove statues that critically depicts a wrongdoer performing a wrongful act, where the statue is clearly condemnatory. But the vast majority of statues of wrongdoers are not like this: they merely depict the person (often in a manner meant to convey their heroic or otherwise admirable status), and they are erected as expressions of esteem. Statues of historical figures are thus distinct from other types of commemorative monuments. A public statue commemorating a massacre, for example, need not express a positive evaluative attitude towards the massacre. But a public statue of a person who participated in a massacre is inappropriate precisely because such a statue would denote a positive evaluation of that person.3 As I will argue, it is because statues of historical figures are typically evaluative in this way that their being publicly displayed is typically incompatible with the duties to condemn and repudiate wrongdoing, and we thus have duties to remove them.

The positive evaluation view has two components: first, that statues honour, and, second, that the object of this honouring is the depicted person. I defend each component in turn.

2.1 Statues as honouring

The positive evaluation view can be contrasted with what I will call the historical record view, which holds that statues are merely or primarily historical records. According to this view, statues convey only importance or significance, and have no connotations of admiration or respect (several commentators defend views that are variations on this theme, see Kuznar, 2017 and Beard, 2015).

If the historical record view were true, the dearth of public statues of, say, Hitler in Britain would be baffling. It’s hard to imagine a more important historical figure in British history than Hitler. And yet the absence of such statues is far from baffling. Rather, it is straightforwardly explained by the fact that we do not tend to build statues to people whom we believe engaged in serious wrongdoing, even if those wrongs were of monumental historical significance.4 The same goes for less significant, but still widely-known, wrongdoers, such as the serial child killers Ian Brady and Myra Hindley. Brady and Hindley—the notorious ‘Moors Murderers’—are part of Britain’s collective national memory, but it would be incontrovertibly inappropriate to build statues of them. Importance is perhaps a necessary condition of our building a public statue of someone, but it is surely not sufficient. We build statues only to those people whom also we think it fitting to admire.

This plausibly explains why, just as it has never seemed appropriate to build statues of Hitler, we would not now build a public statue to Cecil Rhodes. Colonialism was widely admired when Oriel College, Oxford erected its now-infamous statue of Rhodes in the early 1900s. Rhodes was lauded for his part in the violent theft of land from native black Africans for ‘civilising’ use by white Europeans. By contrast, we now regard Rhodes’ colonialism as part of a wider practice of serious rights violations. If statues are mere records, Rhodes’ wrongdoing gives us no reason not to build new public statues of him. And yet his wrongdoing seems to give us decisive reason not to build new statues to him.

The important point here is that when there is broad consensus about a person’s wrongdoing, the question of building a statue of that person does not arise. Hence, the claim that statues have positive evaluative connotations seems sound. The problematic statues are built precisely because, at the time they are built, most people do not believe that the subject is a serious wrongdoer. But the fact that statues have these evaluative connotations supports not only refraining from building new statues of wrongdoers, but also taking existing statues of wrongdoers down. The fact that a statue already exists does not make it any less evaluative.5

While I argue that we ought to remove statues from most public spaces, I think we can make principled exceptions for some public museums and galleries.6 This is because museums, unlike parks or squares, are explicitly and primarily intended as places of historical record (see Sears, 2018). This has two relevant implications. First, it means that museum exhibits typically lack the evaluative dimension of other public displays. Museum exhibits are not typically restricted to things we admire—rather, we use museums to display all kinds of objectionable things, such as medieval torture implements and relics from concentrations camps. The positioning of statues in prestigious locations, such as parks, town squares, town halls and so on is an important part of what makes public statues expressive of positive evaluative attitudes. The Mayor of New Orleans, Mitch Landrieu, emphasises the social significance of putting Confederate soldiers “literally […] on a pedestal in our most prominent places of honour.” (Landrieu, 2017). When a community chooses to display a figure in a prominent, central location, where its members must regularly confront it, this choice expresses (or is at least reasonably interpreted as expressing) the community’s view of what represents them.7 This is especially true given that such prominent places are, by their nature, limited in number, which means that there is inevitably a comparative dimension to the decision about who or what occupies them. Note too that the decision to continue to display a statue is not a historic decision, taken by now-dead inhabitants. Past generations may have decided to erect a statue; the decision to keep the statue—to continue to display it in the centre of the town square, or at the entrance to the public park or town hall—is made by the current generation.

Second, museum exhibits are usually accompanied by detailed explanations of their significance, and displayed with other relevant artefacts that give a broad contextual setting. As Matthew A. Sears puts it, “[at] their best, museums convey history in all its messy complexity, and encourage visitors to ask questions of the past, including how it’s being used to shape the present. Monuments, on the other hand, are blunt-force objects, and frequently discourage nuance, deliberately distorting the past to convey something in the present that may or may not be an outright falsehood.”(Sears, 2018). A statue of a slave-owner alongside examples of the tools used to punish and restrain slaves, for example, not only shows that respect or admiration would be misplaced, but also genuinely informs the viewer about the historical wrongs of slavery, including the fact that slavery was once so widely accepted that people built public statues of slave-owners. Displaying a statue in a museum in this broader context does not express admiration for the depicted person.

With respect to (at least some) art galleries, displaying a statue is expressive, but it’s usually expressive of an aesthetic evaluation of the statue qua sculpture, rather than of a positive evaluation of its subject. Since, again, the context makes this shift in expression clear, it could be permissible to display a statue of a wrongdoer in an art gallery. These considerations explain why it can be permissible to move a statue to a museum or gallery, even if, as I will argue, they ought not to be displayed in other public places.

2.1.2 Meaning as socially constructed

It might be objected that the expressive value of a statue is not settled by its creators or commissioners. For example, Lawrence A. Kuzner claims that, “these pieces of metal and stone have only the meaning we assign to them, and that meaning can take any form we like. They can be revered or reviled; honoured or ridiculed; or co-opted for a new purpose.”(Kuznar, 2017). On this view, the fact that a statue was intended as an expression of admiration by its creator or commissioner does not entail that it does, in fact, express admiration.

It’s true that the meaning of public statues is socially constructed. Such statues could have had a different meaning, and we could, over time, shift our attitudes such that building public statues is no longer expressive of positive evaluative attitudes. It might also be true that the meaning of a statue is not settled by what its creators or commissioners intend (see Burch-Brown, 2017, for a related discussion of the possibility that statues can change their meaning). As I argued above, we can shift the expressive value of a statue by moving it to a museum or gallery. But, as the earlier discussion of building new statues of Rhodes shows, it’s nevertheless true—albeit contingently so—that public statues of people do typically express positive evaluative attitudes, and that we regard only those worthy of admiration as fitting subjects. Thus, the idea that we (whoever ‘we’ are) can simply make it the case that some particular public statue does not express a positive evaluative attitude towards its subject is unpersuasive. For as long as it is true that we build statues only to people whom we think worthy of respect, it is implausible that we can selectively withdraw this connotation, or change the statue’s meaning simply by insisting that its meaning has changed.8

It is especially implausible that we might do this when there is no consensus about whether the subject merits respects. Given that some people believe that Rhodes statues rightly honour Rhodes, and that Confederate statues rightly honour confederate soldiers, it’s unclear how ‘we’ can make it the case that such statues no longer honour their subjects. The mere fact that the meaning of a type—in this case, public statues—is socially constructed does not entail that we (or some subset of ‘we’) can change the meaning of tokens of that type while leaving them in situ, and while there is no more general shift in our practices regarding public statues.

It is illuminating here to compare our attitudes to other honouring practices, such as the awarding of an honorary doctorate. Honorary doctorates recognise a person’s outstanding contribution to a field outside of the usual academic context. J.K. Rowling received an honorary doctorate from Edinburgh University for her contributions to literature. Now imagine that we discover that Rowling did not write the Harry Potter books. We would not think that since the meaning of a doctorate is socially constructed, we can strip this particular doctorate of its honorific connotations and thus allow Rowling to keep it. If you’re honoured in light of your putative achievements, but the relevant achievements turn out not to be yours, we withdraw the honour. We do not let you keep the award, but somehow strip it of its honouring. We likewise strip athletes of medals when they cheat, rather strip their medals of their meaning. Similarly, if we build a statue of someone to honour them for their actions, but then realise that their actions are not, in fact, the kind of thing we ought to honour, we cannot simply insist that this particular statue no longer honours. Its meaning is determined by our wider practices, and cannot be selectively withdrawn.

2.1.3 Statues as historical records

It is even harder to sustain the idea that statues are primarily historical records once we recognise the selective nature of public monuments. Setting aside statues of royalty and fictional characters, only 3% of public statues in Britain depict women.(BBC, 2018) The same is true in Australia (indeed, Australia has more statues of animals than of women (Spicer, 2017)). The United States does marginally better at 10% (Peled, 2017). Prior to the erection of a statue of Mary Seacole 2016, there were no statues of historical black women in the UK at all. All the permanent statues in Trafalgar Square are of white men. As Madge Dresser argues, “[m]any Black Britons today, for example, feel personally excluded by the public commemorative conventions of their country. If monuments are about remembering, who or what gets ‘forgotten’ in the public discourse can be just as significant.” (Dressler, 2008; also Hirsch, 2017). The idea that public statues are primarily historical records, and ought to be preserved as such, looks implausible in light of this selective approach to who we depict. Any account of history based on existing statues would be utterly inadequate and misrepresentative.

Of course, we might think that the appropriate response to these concerns is simply to build more statues. If the record is incomplete, we can correct this by building more statues of people from historically marginalised groups. But, again, if we care only about the record, we should also build public statues of Hitler, Stalin, and Brady. And, as I suggested above, the fact that we’re not prepared to do that suggests that statues are typically evaluative in the way I have claimed. This also explains the intuitive objection to attempting to ‘correct’ or complete the record by, say, putting a statue of Nelson Mandela alongside a statue of Rhodes. Rather than presenting a more rounded historical record, such a display would imply that Mandela and Rhodes are merely two sides of the same story—their actions on a moral par, their views equally reasonable, both worthy of respect. This implication is morally objectionable. Both are important historical figures about whom we need to educate ourselves. But only one merits the kind of positive evaluation expressed by a public statue.9

One objection to removing statues combines the claim that statues honour with the claim that they are historical records. According to this objection, the presence of the statues records the specific historical fact that these people were honoured. So, proponents of this view grant that statues honour, but think that it’s important to keep them as a record of that fact. For this argument to go through, then, we need to know why it is important to keep such a record. The most likely suggestion is that removing the statues will somehow lead us to repeat the mistakes of history. We need to remember that people used to honour colonialists and slave-traders, lest we forget how widely such practices were accepted. But if what I have argued here is correct, keeping these statues up will not merely record that they were historically honoured, but also continue to honour their subjects. As I argue below, we have reason not to honour wrongdoers. And, as I argue in Section Six, we have a range of methods for remembering serious wrongdoing. It is not plausible that public statues of the perpetrators of serious wrongs are necessary for remembering either that serious wrongdoing took place, or that some such wrongdoing was widely accepted. Given this, I doubt that the underlying claim—that keeping statues is necessary for avoiding future wrongdoing—is sound, and thus that it can defeat our reasons not to honour wrongdoers.

2.3 People or achievements?

We might doubt my claim that statues of historical figures honour people in light of their acts, arguing instead that statues express a positive evaluation only of some particular achievement, and express nothing about the depicted person more generally. Indeed, our explanation of why we have a statue of a person will typically be that he or she did something in particular—led the fight for votes for women, or defended Britain against the French, for example—rather than that he or she was an all-round good thing. Insofar as the celebrated achievement itself was not wrongful, nor tied to wrongdoing, perhaps we need not worry about whether the person also engaged in unrelated wrongdoing.

It is true that some honouring practices celebrate only specific achievements. Literary prizes, for example, honour a particular accomplishment rather than the person: they are awarded for a book, or a poem, or for a body of work. The same is true of scientific prizes or music awards.

Statues are not like this. Consider the myriad ways in which we can and do mark achievements, in addition to the kinds of awards just mentioned. We use, amongst other things, fountains, pillars, buildings, sculptures, bridges, parks, and trees. The fact that we choose to build a statue of a specific historical figure, rather than one of these other types of memorials, is not just the luck of the draw, then. It is meaningful—that is, it is meant to express something about that person, rather than merely mark an achievement. There is a difference in kind between what is expressed by a fountain marking the abolition of slavery, such as the Buxton Memorial Fountain, and a statue of William Wilberforce. The Buxton fountain expresses an attitude about the abolition of slavery. A statue of Wilberforce expresses an attitude about him, in light of his connection to abolition. This is further evidenced by the fact that we typically build statues of people after their death, and not immediately after their achievements.10 It would be inappropriate, I think, to strip someone of a book prize because he or she was unveiled as a child abuser. Such prizes make no claims about the general admirability of the person (but see Archer and Matheson, 2019, for a defence of the view that honouring a person’s achievements encourages us to condone their wrongdoing). But it would be appropriate to remove the author’s statue from the library in their home town on those grounds, precisely because statues express these more general evaluative attitudes. As Joanna Burch-Brown puts it, our assumption when we encounter statues is that “the figures involved are broadly positive and appropriate sources of pride.”(Burch-Brown, 2017, 75).

3. Serious rights violations

I contend that states ought to remove public statues of people who engaged in serious rights violations. I will assume that the category of serious rights violations includes violating the rights not to be killed, enslaved, raped, tortured and so on, as well as the violation of certain political rights, such as the right of self-determination that is plausibly violated by colonialism and unjust wars. I do not further define the notion of a serious rights violation here. I focus on serious rights violations, which threaten individuals’ basic interests, because when one perpetrates a wrong of this sort, that fact becomes the dominant feature of one’s moral record, in a way that, say stealing a car or lying to one’s spouse does not. Serious rights violations are thus what we might call defining wrongs: they are always salient to our evaluation of the perpetrator, and typically negate their morally good deeds. Put bluntly, it does not matter how much money a person raises for charity if he also sexually abuses children.11 I leave it open here whether the perpetration of less serious wrongs (perhaps on a wide scale) can also ground a duty to remove a public statue.

3.1 Indirectly violating rights

I take it that the case for removing a public statue is clearest when the subject directly perpetrated serious rights violations herself—that is, when she killed or enslaved and so on. But it is implausible that only the direct perpetrators of wrongs violate rights: rather, those who indirectly participate in seriously wrongful practices can also thereby commit serious rights violations. Someone who owns no slaves, but runs a slave market, for example, is engaged in serious rights violations. The Mafia boss who orders hits, but never kills anyone himself, is engaged in serious rights violations.

Of course, indirect participation in a practice such as slavery is widespread, especially when slavery is legal. Participants in the United States slave trade in the 18th and 19th centuries include not only slaveowners, catchers and traders, but also legislators who passed (or blocked) relevant legislation, individuals who knowingly bought goods produced by slaves, those who informed on escaped slaves and so on. It also includes Confederate soldiers who fought for the legal right to own slaves. There is often similarly broad indirect participation in an unjust war: vast swathes of the German population contributed to the Nazi campaign of aggression and genocide during the Second World War, for example.

As above, my account does not commit us to a moral sainthood standard for statues, where only those who live utterly morally pure lives are consider fit for depiction. My claim is only that we ought to remove statues depicting people who participated in serious rights violations of the sort described above. Nevertheless, given the scale of participation in many seriously wrongful practices, the implications of my thesis are revisionary, compelling states to remove many of their public statues. This is especially so given that it applies to statues honouring serious wrongdoers not only in light of their wrongful actions, but also in light of actions or achievements unrelated to wrongdoing. The implications of the thesis are broader still if one believes that (culpable) omissions can constitute serious rights violations.

Exactly how revisionary the thesis is, then, depends on the correct account of participation in a serious rights violation. For example, many people reject the claim that civilians engage in serious rights violations by knowingly contributing to unjust wars or genocides (see for example Haque, 2017, 57, 71; Fabre, 2009). They might also deny that those knowingly buying slave-produced goods violate the basic rights of slaves. They might propose a narrower account of what it is to violate basic rights. Perhaps only those who directly perpetrate serious wrongs (e.g. Haque, 2017.; for rebuttal see Frowe, 2019, 129), or whose contributions to those wrongs pass a certain causal threshold12, commit serious rights violations. Alternatively, we might draw upon existing accounts of accomplice liability to set the boundaries (the literature here is vast, but see, for example, Goodin and Lepora, 2013; Dressler, 2008, 429; Kutz, 2007, 289—305; Bazargan-Forward, 2017).

My own view is that any plausible account of participating in serious rights violations is going to include lots of people who indirectly contribute to those violations, and thus generate the result that we should remove many—perhaps most—of our public statues of historical figures. Thankfully, though, we need not identify the correct account of participating in a serious rights violation here. Rather, the important point for our purposes is that my account is revisionary only insofar as we grant that those who indirectly participate in serious wrongdoing do themselves commit serious rights violations. This is compatible with the view that not all participation in serious wrongdoing constitutes a serious rights violation (this idea is explored in literature on collective harms e.g. Kagan, 2011, 105-141; Nefsky, 2011, 364—395). But when participation in a wrongful practice does constitute a serious rights violation, it strikes me as very plausible that states ought to remove statues of those participants. The implausible position would, I suggest, be to believe both that the participants engaged in serious rights violations and that it is nonetheless appropriate to depict them in public statues. As I argued above, such statues express a positive evaluative attitude towards the subject. As I’ll argue below, this means that such statues conflict with the state’s duties to repudiate and condemn wrongdoing.13

4. Grounding the duty to remove

I focus here on states—that is, state actors—as the bearer of the duty to remove statues. This is, in part, because I am considering public statues, the presence of which (at least ordinarily) falls under the authority of the state. Since it is the state, acting for its members, who decides what occupies our public spaces, it seems helpful to begin with the question of whether state actors have a duty to remove certain types of public statues. This is compatible with thinking that private individuals may also have certain duties with respect to statues of wrongdoers—to campaign for their removal, for example. But it is at least less obvious that private citizens have duties to remove public statues (not least because many will lack the ability to remove public statues).

My focus on states also reflects the fact the state is sometimes uniquely or best situated to discharge certain duties, or express certain sentiments. There can, for example, be times when it is appropriate for a state representative to condemn wrongdoing even if individual citizens need not do so (for example, see Cunningham, 2014). This is in part because, especially in the case of widespread serious wrongdoing, condemnation should be public, and expressive of the wider community’s rejection of the wrongdoing. Insofar as state actors speak for their citizens, their actions can have a significance that individual actions lack. There is, in general, something especially powerful about official rejection or condemnation of wrongdoing. Consider, for example, the significance (beyond the prospect of punishment) that victims of crimes attach to a perpetrator’s being found guilty of wrongdoing by a court. Even when the victim already knows that the perpetrator is guilty, the formal recognition by the state of this wrong is deeply meaningful.

4.1 The duty to repudiate

Plausibly, the state’s duty to remove statues is especially stringent in cases of state collusion in wrongdoing. The state can collude in wrongdoing in at least two ways. First, it can openly endorse the wrong by legalising it, as in the cases of slavery, the denial of women’s equal status, the forced adoption of Aboriginal children, and the persecution of homosexuals. Second, state actors might deliberately conceal wrongdoing by those in positions of authority and power, as with the sexual abuse of children in Catholic church, or the murder of black detainees in police custody.

There are at least three reasons why the duty to remove statues is plausibly most stringent when the state colluded in the relevant wrongdoing. First, such wrongs seem most likely to cause wider and more significant harm: if a wrong is legal, for example, it may well be much more pervasive than if it attracts criminal sanctions. Second, a wrong’s being facilitated by a state or state actors has an expressive function, since it conveys public sanctioning of the act, which seems to further wrong the victims (for discussion of how laws can constitute expressive harms, see, for example, Hill, 1999; Anderson and Pildes, 2000, 1503-1575; Blackburn, 1999, 467-491). It is one thing to be wronged; it is a further wrong to have one’s state fail to recognise that one is being mistreated. Whilst this most obviously applies in the case of legal wrongs, it plausibly also applies in other forms of state or state actors’ collusion in wrongdoing. For example, it seems worse for state officials to destroy or conceal evidence of wrongdoing compared to private individuals (and many states treat misconduct in a public office as an aggravating feature of wrongdoing). Third, having been implicated in these wrongs confers on the state a stringent duty to repudiate them now. It is this third feature that I explore here.

As I will understand it here, the duty to repudiate is distinct from the duties to condemn and punish wrongdoers, and prohibit or prevent wrongdoing. Repudiation is about rejecting one’s own past wrongdoing: it requires one to disavow certain attitudes and beliefs in virtue of one’s past behaviour. To see the appeal of this view of repudiation, consider the case of Muslims who are asked to publicly reject the wrongs of Islamic fundamentalist terrorists.14 I suggest that these requests are objectionable because, whilst they are often framed in the language of condemnation, they are, in effect, implicit demands that Muslims repudiate the wrongdoing of terrorists—that they disavow the beliefs underpinning such attacks. And yet repudiation is appropriate only when one is somehow implicated in that wrongdoing.

Repudiation in the context of historic wrongdoing requires the state’s acknowledging its past complicity in, or sanctioning of, wrongdoing, and explicitly rejecting the attitudes or values that underpinned that wrongdoing. For example, the British government’s responsibility for recognising injustices resulting from slavery, and rejecting the racism underlying slavery, is distinct from its usual responsibility to aid its citizens when they are the victims of other people’s wrongdoing. As the legal representatives of the British state, the current government is responsible for recognising the harms of slavery in part because the state was partly responsible for those harms. Given its historical participation in slavery, the government must take steps to disavow slavery that it need not take with respect to wrongs in which it did not participate. Repudiation requires, amongst other things, a public and sincere declaration of an act’s wrongfulness, and a commitment to not facilitate such wrongdoing in the future.15 And this, I suggest, is incompatible with publicly honouring the perpetrators of the wrongdoing. Hence, the duty to repudiate confers on states an especially stringent duty to remove certain cultural artefacts.

We can helpfully draw here on the philosophical literature surrounding apology, particularly political apology. For example, Allison Don and Per-Erik Milam argue that an effective political apology must, “respond to blame by repudiating the misconduct, and profess a change in quality of will that shows the apologiser to be relevantly different from the offender.” (Don and Milam, draft manuscript). This notion of reform is key to several accounts of apology: someone who apologises for her wrongdoing only to commit the same wrong again shows herself to be insincere in her apology, since the repetition suggests that she fails to properly recognize the wrongness of her conduct (see Bennett, draft manuscript; Matheson, draft manuscript).

Amongst individuals, showing the apologiser to be relevantly different from the offender requires a change in the offender’s quality of will—we want her to change her attitudes or beliefs, in order to count her apology as sincere. But this does not work at the level of states: we are not looking for the apologiser to show that they are not the person they were, but rather to show that the state has reformed. This demands visible, publicly-accessible actions rejecting the past wrongs. And, as Don and Milam argue,

“[t]here is a genuine risk that a poorly executed political apology can deepen the divide between victim and offender by demonstrating the opposite of what they intend—e.g. that the state does not have due regard for the victimised group and is not sufficiently committed to avoiding similar misconduct in the future. However sincere its expression, the force of an apology can be undermined by doubts about attitudes of the collective and people on whose behalf it is being made. Victims may rightly feel that a spoken apology, unsupported by a concrete and significant demonstration of sincerity, does not do justice to the offence, just as material reparation without an acknowledgement of responsibility is also inadequate.” (Don and Milam, draft manuscript, 9).

I suggest that, in the absence of a lesser-evil justification for keeping it, the refusal to remove a public statue to someone who perpetrated the wrongs in question undermines the idea that a state has reformed and genuinely repudiates its role in that wrongdoing. It thus gives victims legitimate grounds for believing that their complaints are not taken seriously by their state and co-citizens.

4.2 The duty to condemn

Even when it lacks a duty to repudiate its own past wrongdoing, the state still has a duty to condemn other people’s serious wrongdoing. As I will argue, the duty to condemn is also incompatible with continuing to display statues of wrongdoers. Like the duty to repudiate, this duty is grounded in what is owed to the victims of wrongdoing. As I argued above, those who are wronged with the help of their state, or state actors, are distinctively wronged, given the expressive function of state collusion. And the duty to remove statues is plausibly more stringent in collusion cases—that is, it might be less easily defeated by lesser-evil considerations—since we have particularly strong reason to recognise and make good harms for which we are responsible. Nevertheless, the duty to remove statues still obtains in the absence of state collusion in the relevant wrongdoing. It would be objectionable to remove a statue of someone who committed rights violations with the collusion of the state while leaving in place a statue of someone who committed similar wrongs without such collusion. States plausibly have duties to properly recognise and condemn wrongs done to their citizens irrespective of whether they facilitated or concealed those wrongs.

There are straightforward instrumental reasons to condemn wrongdoing—for example, that so doing deters future wrongdoing. But there are also less instrumental reasons to condemn wrongdoing. Condemnation is primarily an expressive act. Its value does not rest solely on its capacity to deter future wrongdoing or prevent harm arising from past wrongdoing. It rather reflects our intrinsic reasons to affirm victims’ moral standing in the face of actions that have denied that standing by publicly asserting the wrongfulness of those actions.

In this respect, the duty to condemn is like the duty to apologise: whilst apologies can have instrumental effects, smoothing social relations and so on, whether one has a duty to apologise does not seem to depend on whether apologising will, or is expected to, produce these effects. I can owe you an apology even if, or perhaps especially if, relations between us are irreparably damaged. The apology is owed in light of the wrong you have suffered, not in light of the beneficial effects of apologising. Similarly, states’ duties to condemn wrongdoing are grounded in the reasons we have to publicly reject wrongdoers’ implicit or explicit claims about the moral status of their victims. Serious rights violations deny their victims’ equal moral standing; official, public condemnation of the wrong helps re-affirm that standing. Thus, unlike the duty to apologise (and the duty to repudiate), the duty to condemn obtains with respect to other people’s wrongs.

Importantly, there can be duties to condemn wrongdoing even if there is no duty to blame the perpetrator (although, of course, there will be cases in which both condemning and blaming is appropriate). Moral standing can be equally challenged by blameless and culpable wrongdoing. We can, for example, imagine that some American slaveowners genuinely believed that their slaves were inferior beings, and were better off being enslaved rather than free. Perhaps slaveowners who held such beliefs were less culpable than slaveowners who were wholly indifferent to their slaves’ wellbeing, or even non-culpable. It hardly follows that enslavement was less degrading for their slaves compared to the slaves of more culpable slaveowners. Thus, even if, for example, some slaveowners and Confederate soldiers were not blameworthy for their actions, this does not show that we may display public statues of them. The duty to condemn is not grounded in the claim that the perpetrators deserve blame, but rather in the claim that the victims are owed appropriate recognition of the wrong (as argued by Burch-Brown, 2017, 77).

Expressions of condemnation are particularly important when there is no question of punishing the perpetrator, as is usually the case with the subjects of public statues. Punishment serves multiple functions of condemning, blaming and penalising. When the perpetrator is dead, and escaped punishment during their lifetime, we have especially strong reason to independently condemn their wrongs, precisely because we cannot punish the perpetrator and express our condemnation of the wrong in that way.

The duty to condemn is owed most obviously to the victims of the particular wrong in question. Assuming that we can have duties to the dead, this includes victims who no longer exist. It also includes secondary victims—that is, those who suffer harm as a result of historical wrongs to others (such as the harms arising from the legacy of slavery that befall people who were not themselves enslaved). But it is also, I think, owed to citizens in general, both in their capacity as potential victims of other wrongs, and in light of their interest in living in a society in which each person’s moral standing is taken seriously by their state. We all have reason to want our states to refrain from honouring wrongdoers. As I described in Section 2.1.1, our public monuments—and our decisions about whether to keep them—express a community’s values. I have a claim that my government remove public statues of people who engaged in serious wrongdoing, irrespective of whether I am a victim of such wrongdoing, since I have a legitimate interest in not being part of society that expresses positive evaluative attitudes towards violent racists, slaveowners, misogynists and other wrongdoers; plausibly, I also have a responsibility to discourage the expression of such attitudes.

In order to count as discharging the duty to condemn, condemnation must be sincere. But one cannot sincerely condemn wrongdoing if one simultaneously honours the wrongdoer, as one does by retaining public statues of them. There’s an element here of the familiar idea that justice should not only be done, but also be seen to be done. As Adam Omar Hosein has recently argued, justice requires not only that states actually weight citizens’ interests equally, but also demonstrate that they do so, doing “a sufficient amount to secure for each member the confidence that her rights and interests are being given equal weight.” (Hosein, 2018) Similarly, states need to give citizens the confidence that they sincerely condemn wrongdoing. Denouncing slavery as a moral evil whilst keeping public statues of slaveowners, or of people who fought to sustain slavery, undermines a state’s claim to sincerity.

It seems to me that retaining public statues of wrongdoers is straightforwardly inconsistent with showing that one properly recognises the gravity of serious rights violations. This is true even if the state enacts other relevant measures, such as offering compensation. But even if one denies that these things are straightforwardly inconsistent, it’s certainly plausible that one might reasonably interpret a refusal to remove a statue as indicative of a failure to properly grasp or attend to the wrongs in question, especially when those wrongs concern historically marginalised groups (see also Archer and Matheson, 2019).

4.3 Shifting norms

Reflecting on the duties to repudiate and condemn gives us grounds for rejecting the shifting norms objection to removing controversial monuments. Roughly, this objection holds that someone who defended slavery or colonialism a century or so ago cannot be held to the moral standards that we endorse today, because those things were legal and widely accepted back then. Thus, there is no duty to condemn such people. Hence, the duty to condemn cannot support a duty to remove public statues.

This argument is unpersuasive because, as above, there can be duties to condemn wrongdoing even if there is no duty to blame the perpetrator. The shifting norms objection seems particularly wrongheaded in the case of publicly-sanctioned or concealed wrongs. For example, the fact that social conditions in the United States might have made beliefs in the inferiority of enslaved people credible is partly what the state needs to publicly repudiate by removing the statue. It’s precisely because the state facilitated these wrongs by normalising them, by making them legal, or concealing them, or failing to punish them that the state now has a duty to repudiate those wrongs. That these excusing conditions existed cannot, therefore, be a reason for retaining public statues of wrongdoers.

5. ‘Moral ambiguity’

We might grant that there is a duty to remove statues of wrongdoers when the subject is honoured for their wrongdoing, or for achievements that were connected to that wrongdoing. Statues of Rhodes, for example, are particularly egregious because what is being commemorated—his role in the violent expansion of the British Empire (and, in the case of the statue at the University of Cape Town, ‘giving’ land to the university )—is the very thing that makes Rhodes morally objectionable. Similarly, statues of Confederate soldiers commemorate, specifically, their fight to retain the legal right to own slaves.16

More controversial is the claim that we should remove statues of people who perpetrated serious rights violations but also accomplished significant moral goods. For example, some people think that we should keep statues of Thomas Jefferson in light of his role as a Founding Father, even though Jefferson was a slaveowner and a rapist (Danielle, 2017). The duty to remove is similarly controversial with respect to at least some people who used unjust means to achieve significant goods. For example, Winston Churchill is largely credited with defeating Nazi aggression in Europe—clearly, an enormously significant moral good. But Churchill also authorised the use of unjust means, such as the firebombing of Dresden, as part of this defeat. He is also alleged to have been responsible for serious wrongs elsewhere, such as the Bengal famine of 1943. Let us assume that Churchill was, in fact, so responsible, and that this famine violated the basic rights of the Bengalis who were thereby harmed, and that the bombing of Dresden violated the rights of at least some of the people who were thereby killed. Does it follow that we should not to have public statues of Churchill?

I think it does: it is wrong to honour those who engaged in serious wrongdoing, even if they also achieved significant goods. It is objectionable not only to honour a person for her wrongdoing, but also despite her wrongdoing. To do so either ignores the fact that someone perpetrated serious rights violations, or implies that our duties to the victims of those violations are outweighed by some other consideration. The first option seems like a straightforwardly disrespectful dismissal of the harms inflicted on the victims of the wrongdoing, and of their claims to appropriate recognition of those wrongs. The second is perhaps a better description of what people have in mind when they suggest that we ought to keep statues of Churchill even if they grant that he committed serious wrongs. But we must be careful not to mistake or misrepresent the considerations at stake. We are not weighing the duties to the victims against the importance of the good that the person achieved: it is not a question of whether, for example, the duty to condemn the Bengali famine is trumped by the good of defeating the Nazis.

Rather, we are weighing the duty to condemn against the good of having the statue. And it’s very unclear what this good consists in, such that it could weigh against the claims of the victims of serious wrongdoing in this way. It seems very unlikely that anyone would suffer any setback of their important interests because of the absence of a statue of Churchill. Holding that erecting or keeping the statue is nonetheless more important than our duties to the victims thus seems to objectionably diminish their claims.

We might think that it’s important to honour people for achieving substantial moral goods: that we have a debt of gratitude to Churchill, for example, that trumps our duties to condemn and repudiate. But if this is true, it should hold across the board. There should, for example, be a genuine question about whether our gratitude for the millions that someone raises for charity outweighs our obligation to condemn their abuse of children, given that this money could plausibly save many lives, which is a significant moral good.17 Even if raising a couple of million doesn’t do the trick, there should be some point at which one raises enough money, or saves enough lives, that the resultant debt of gratitude outweighs the duty to condemn the abuse of children, or other serious wrongs. I am sceptical that engaging in these trade-offs can support the permissibility of keeping or erecting statues of wrongdoers. Moreover, we need not deny that we owe gratitude to Churchill in order to satisfy our duties to condemn. We need only refrain from having public statues of Churchill. This is compatible with the permissibility of alternative expressions of gratitude that do not express the general admiration of the person in the way that statues do. A different sort of memorial can celebrate Churchill’s winning of the war, just as the Buxton Fountain celebrates the abolition of legal slavery.

I have suggested that serious rights violations dominate the perpetrator’s moral record, making public statues of them inappropriate. But what should we say about someone who engages in serious wrongdoing, but is later fully repentant, perhaps duly punished, and works hard to lead an admirable life of good deeds? Such cases raise philosophical questions concerning the persistence of personhood across time and changes of character that I cannot address here. But we should notice that reformed wrongdoers are importantly different to the ‘morally ambiguous’ characters we are considering here. Reformed wrongdoers are those who undergo some change in their quality of will, ceasing their wrongful actions. Jefferson’s wrongdoing, in contrast, was contemporaneous with his achievements; the slave-owning Jefferson was the Founding Father Jefferson. So too with Churchill.18 So, even if we think we may have public statues of those who renounce their wrongdoing and go on to achieve great things, this does not undermine the claim that we may not keep public statues of Jefferson and Churchill.

6. Harm as a constraint on removal

We might think that there can plausibly be a duty to remove statues of serious rights violators only if there currently exist, or will exist, people who are harmed by their wrongs, or if the statues themselves cause harm. That is, we might think that there is a harm constraint on a duty to remove statues. One possible attraction of such a constraint is that it seemingly limits the revisionary implications of my thesis. Most writers on this topic are keen to show that their view does not require taking down vast numbers of statues, particularly very old statues (see Timmerman, forthcoming, 7; Burch-Brown, 2017, 78). For example, Johannes Schulz argues that Germans have no reason to feel degraded by Roman statues of Julius Caesar, despite Caesar’s brutal treatment of Germanic tribes, because the “hierarchy between the citizens and soldiers of the Roman Empire and the allegedly ‘barbaric’ tribes in Germania and Gaul no longer has any correspondence in the social reality of present-day Europe.” (Schulz, 2018, 5-6).

Thinking about genocide gives us good reason to reject a harm constraint on a duty to remove statues.19 Imagine a successful genocidal campaign that kills all the members of a certain group, but causes no harm to other people. Supporters of the genocidaires erect a statue in their honour, to which everyone else is indifferent. If we have reason to remove statues only when there are current or future victims of the relevant wrongs, or when the statue causes harm, it would follow that there is no duty to remove a statue honouring the genocidaires. The same concern speaks against Schulz’s suggestion that we ought to remove statues that express degrading ideologies only when they are connected to existing wrongful social hierarchies. If one simply erases the members of the disadvantaged group, there is no persisting wrongful hierarchy. But the case for removal of honorific statues seems stronger, rather than eliminated, in such cases.

Thinking about Schulz’s example also suggests that a harm constraint might not, in fact, limit our obligations to remove statues. Much depends on how we identify the relevant degrading ideology. We might, like Schulz, describe Caesar’s ideology as the view that members of Germania and Gaul were inferior to Romans, and agree that since there is currently no wrongful social hierarchy between these groups—perhaps the tribal groups don’t even exist anymore—there is no reason for Germans to feel degraded by statues of Caesar. But we might equally offer a broader reading of Caesar’s ideology: he believed that not all humans enjoy equal moral standing, and that those of lesser standing may be murdered and enslaved. This ideology certainly corresponds to current social realities. Of course, on this broader reading, Germans have no special reason to feel degraded by statues of Caesar. But anyone who is currently harmed by a wrongful hierarchical system might legitimately object to, and feel degraded by, the commemorating of someone who perpetrated the belief that some types of people are inferior, and may be treated as such. It strikes me as wholly appropriate for, say, an African-American to feel degraded by statues to racial supremacists in general, rather than only by those who believed specifically in the inferiority of blacks. Moreover, it seems wholly appropriate for people in general to object to the presence of public statues of racial supremacists, irrespective of whether they themselves are harmed by the presence of racist hierarchies. The Germans might have no special complaint about the Caesar statues, but they may complain nonetheless.

7. Why removal?

We might grant the force of the duties that I have articulated, but deny that they require us to remove statues from public display. Proponents of keeping statues of wrongdoers often claim that monuments can trigger conversations about the serious wrongs of the past, thereby presenting opportunities to educate ourselves about those wrongs (see also Beard, 2015; Schulz, 2018). If we remove the statues, the thought must go, we remove these valuable opportunities. Even though I have not ruled out the permissibility of keeping statues in museums, we might think that many people do not visit museums, and thus we would be denying valuable educational opportunities to, say, those from less privileged backgrounds.

Schulz argues that whether removing a controversial statue is appropriate depends on whether removal, rather than “various kinds of contextualisation”, is most likely to “further the establishment of relations of respect” between citizens (Schulz, 2018, 12). Schulz cites the decision not to rename the IG-Farben-Haus building at the University of Frankfurt as an example of a successful alternative to removal. Schulz describes how, “students and the administration agreed on contextualising the building, rather than renaming it. A large plaque […] now alludes to the involvement of IG Farben in the Holocaust. A critical analysis of the company’s as well as the university’s actions during the Nazi period is part of campus walking tours.” (Ibid, 15). Schulz concedes that where a historical figure is “ambiguous”, in the sense described above, greater contextualisation may be needed. For example, if the University of Virginia wants to keep its statues of Jefferson, its students might be required to take classes, “on the history of slavery and Jefferson’s relation to it”. (Ibid, 21).

But the use of these contextualising measures—plaques, critical commentaries, classes on slavery—does not require, and therefore cannot justify, keeping statues of serious rights violators or naming buildings after them. A re-named building can have a plaque noting that it used to be named after a company that participated in genocide, thus sparking the same conversation about past injustices that Schulz describes. A campus tour can make the same observation. It can also point to where statues used to be, before they were removed or replaced. The historical fact of, for example, a building’s having been named after Holocaust collaborators is not rendered inaccessible by renaming the building (see also Burch-Brown, 2017, 75; Timmerman, forthcoming, 5). The historical facts that justify compelling students to take classes on slavery persist even if we take down statues of Jefferson.

Once we recognise that these ways of informing ourselves about past injustices are available even in the absence of these commemorations, it is unclear why keeping the commemoration but improving its context is preferable to removing it. It is an illusion to think that statues, or a building’s having a particular name, provide valuable opportunities to educate ourselves about past wrongs. Since those opportunities are no less available in the absence of those commemorations, their provision by statues and the like is not valuable. Since the names and the statues play no essential role in facilitating the conversations that Schulz thinks can promote relations of respect, his argument does not support a permission to keep them.

This might seem at odds with the explanation in Section Two of why it could be permissible to display a statue of a wrongdoer in a museum, which points to the wider contextualisation that such venues provide. But recall that it’s not the mere contextualisation that renders such displays permissible—or rather, it’s not merely the lack of contextualisation that makes public displays impermissible. The location of public statues is also morally significant: having a statue of a slaveowner on public display—particularly, for example, in the centre of your campus, or at the entrance to your library—is also expressive of a positive evaluative attitude towards that person. The comparative dimension is also significant: to keep the statue of Jefferson is to judge that having a contextualised statue of a slaveowner and rapist is better than having a statue of someone who did not perpetrate serious rights violations.

Of course, there is a range of recontextualization measures that one might employ, in addition to the explanatory plaques and so on that Schulz has in mind. These include defacing public statues, as happened with the Rhodes statue at the University of Cape Town and many Confederate statues. Such actions can certainly express a rejection of the subject’s wrongdoing. But, crucially, they do not, ordinarily, constitute a rejection of that wrongdoing by the state. Rather, they are usually unauthorised, illicit acts of vandalism carried out as acts of resistance to the state’s refusal to remove the statue. I think they are permissible, but this permissibility is partly explained by the fact that the state ought to remove the statues. In other words, whilst defacing public property is usually wrong, it might not be wrong when the state has failed in the duty that I have defended—namely, to remove the statue from public display.

Of course, the state could sanction such defacing as a means of repudiating and condemning the wrongdoing. But, first, it is not clear to me why this is better than removing the statue (and, say, replacing it with a memorial to those who suffered the harms of slavery or colonialism). And, second, it seems to me that state-sanctioned acts of vandalism typically lack the expressive power of illicit, rebellious defacement.

Alternatively, the state might install the ‘counterbalancing’ statues that I discussed in Section 2. These installations would be state actions, and so avoid the worry about non-state rejection of wrongdoing that arises in the case of vandalism. But I am sceptical that such counterbalancing is desirable, even if one could construct a display that avoided the ‘moral par’ objection that I raised with respect to Rhodes and Mandela. The deeper worry here is that, in many cases, the reason why recontextualising a statue might seem preferable to removing it—might best foster relations of respect between citizens, as Schulz puts it—is that some people have objectionable attitudes that make them want to keep the statues of wrongdoers, and we predict that some of those people will act wrongly in the face of attempts to remove them.20 This wrongdoing might manifest in, for example, violent resistance to attempts to remove the statue or, more insidiously, as heightened social tensions that exacerbate the wrongful attitudes that many people already hold.

The fact that some people will engage in wrongful behaviour in the face of attempts at removal is, as I suggested in Section One, morally significant. We should be sensitive to whether our actions will cause others to inflict wrongful harms, and the threat of sufficiently grave wrongful harm can defeat the duty to remove a statue (for a discussion of our responsibility for harms imposed by others, see Frowe, 2014, ch.5). But such facts should not make their way into our deliberations about whether there is a duty to remove the statue, even if they bear on the all-things-considered permissibility of removing it. Those who require the retaining of statues of wrongdoers as a condition of their not engaging in wrongful behaviour cannot thereby undermine the claims of the victims of serious rights violations.

By way of comparison, consider the British government’s policy of compensating slave owners for the ‘losses’ inflicted upon them by abolition. Compensating someone for taking away their slaves strikes us as objectionable, even if this policy best facilitates certain valuable ends, such as the end of slavery, because it constitutes a failure to properly recognise the wrong done to slaves. Indeed, it serves to legitimise slaveowners’ claims to own their slaves by legally reinforcing the notion of people as property. Given this, compensating the slaveowners was a pro tanto wrong, albeit one that was plausibly justified as the lesser evil. Likewise, even if retaining statues is sometimes justified on lesser-evil grounds, there is still an important sense in which keeping them is wrong. And, in both cases, the choice between evils is forced by wrongdoing.

There is thus a further wrong involved in cases in which a duty cannot be discharged because of the predicted wrongdoing of others, compared to cases in which there is no such predicted wrongdoing. Compare a case in which removing a statue will unavoidably cause some toxic material to be released from its core with a case in which removing a statue will cause those who want to keep it to riot, thereby inflicting harm on innocent people.21. In the riot case, there’s an important sense in the state’s failure to discharge its duty to remove the statue is unnecessary, since harm could also be avoided by everyone’s abiding by their moral obligations. This is not true of the toxic material case, in which harm can be avoided only by keeping the statue in place. The rioting case thus involves a wrong that is absent in the toxic material case, since it is wrong for the prospective rioters to gratuitously render the state unable to discharge its duties. Note that this wrong does not arise only in the more extreme cases of rioting or other threats of physical harm. As I mentioned above, more subtle responses, such as holding increasingly hostile attitudes to members of certain groups, are also wrongful. This explains why, for example, African-Americans are entitled be aggrieved by the keeping of Confederate statues even if they grant that taking them down would overall worsen race relations in the United States, and that the statues should therefore stay. We should not have to keep statues of slaveowners or Confederate soldiers in order to have conditions for relations of respect, just as we shouldn’t have to compensate people in order to get them to give up their slaves.

8. Conclusions

I have argued that the state has a duty to condemn and repudiate serious wrongdoing that is incompatible with retaining public statues of historical figures who perpetrated serious rights violations. Public statues of such figures are typically evaluative: they express a positive attitude about the depicted person that undermines the state’s claims to be sincerely condemning or repudiating their wrongdoing. I argued that the duty to remove public statues applies not only to statues honouring people for their wrongdoing, but also to statues that honour people despite their wrongdoing. We cannot weigh morally significant achievements against serious wrongdoing in order to justify public statues of wrongdoers. The duties to condemn and repudiate wrongdoing that underpin the duty to remove persist even when, in light of lesser-evil considerations, it is all-things-considered permissible to keep the statue. These lesser-evil considerations typically arise as a result of threatened wrongdoing; such threats of wrongdoing do not negate the state’s duties to condemn and repudiate wrongdoing.


Versions of this paper were presented at a workshop on resistance, co-hosted by the Stockholm Centre for the Ethics of War and Peace and the Philosophy Department at Stellenbosch University, at the Philosophy Department at St. Andrews University, as a Royal Institute of Philosophy Lecture at the University of Kent, at the Filosofidagen conference at Umea University, at the Philosophy Research Group at the University of Roskilde and at the Society for Applied Philosophy Annual Conference in Utrecht. I am very grateful to those audiences. Thanks also to Derek Matravers, Tom Douglas, Frej Klem Thomsen, Sune Lægaard and two anonymous referees for helpful written comments. Work on this paper was supported by the Knut and Alice Wallenberg Foundation, grant no. 1521101.


Anderson, Elizabeth S. and Pildes, Richard H., (2000) ‘Expressive Theories of Law: A General Restatement’, University of Pennsylvania Law Review 148: 1503-1575.

Archer, Alfred and Matheson, Benjamin (2019) ‘When Artists Fall: Honoring and Admiring the Immoral’, The Journal of the American Philosophical Association 5/2: 246-265.

Bazargan-Forward, Saba, (2017) ‘Complicity’, in Marija Jankovic & Kirk Ludwig (eds.), Routledge Handbook of Collective Intentionality, New York: Routledge, pp327-337.

BBC News, (24 April 2018) ‘Reality Check: How many UK statues are of women?’ BBC, available at https://www.bbc.co.uk/news/uk-43884726 (accessed 20/11/19).

Beard, Mary ‘Cecil Rhodes and Oriel College’, Times Literary Supplement, 20th December 2015, available at https://www.the-tls.co.uk/cecil-rhodes-and-oriel-college-oxford/ (accessed 20/11/19).

Bennett, Christopher, ‘Sorry’s Not Good Enough’: Apology as Normative Power and Expressive Action’, (draft manuscript).

Blackburn, Simon, (1999) ‘Group Minds and Expressive Harms’, Maryland Law Review 60, No. 3: 467-491.

Burch-Brown, Joanna (2017) ‘Is it Wrong to Topple Statues and Rename Schools?’ Journal of Political Theory and Philosophy 1: 59-88.

Cunningham, Michael (2014) States of Apology, New York: Manchester University Press.

Danielle, Britni (2017) ‘Sally Hemings wasn’t Thomas Jefferson’s mistress. She was his property.’ The Washington Post, 7 July 2017, available at https://www.washingtonpost.com/outlook/sally-hemings-wasnt-thomas-jeffersons-mistress-she-was-his-property/2017/07/06/db5844d4-625d-11e7-8adc-fea80e32bf47_story.html (accessed 29 November 2019).

Don, Allison and Milam, Per-Erik ‘National Apology’, (draft manuscript).

Dressler, Joshua (2008) ‘Reforming Complicity Law: Trivial Assistance as a Lesser Offense’ Ohio St. J. Crim. L. Vol 5, pp 427-448.

Fabre, Cecile (2009) ‘Guns, Food, and Liability to Attack in War’, Ethics 120/1: 36-63.

Frowe, Helen (2014) Defensive Killing, Oxford: Oxford University Press.

———(2019) ‘Civilian Liability’ Ethics 129/4: 625-650.

Goodin, Robert E. and Lepora, Chiara (2013), On Complicity and Compromise, (Oxford: OUP).

Haque, Adil Ahmad (2017), Law and Morality at War, (Oxford: OUP).

Hill, B Jessie, (1999) ‘Expressive Harms and Standing’, Harvard Law Review 112, No. 6, 1313-1330, at p.1314.

Hirsch, Afua, (2017) ‘Toppling Statues? Here’s Why Nelson’s Column Should be Next’, The Guardian 22 August 2017 https://www.theguardian.com/commentisfree/2017/aug/22/toppling-statues-nelsons-column-should-be-next-slavery (accessed 20/11/19).

Hosein, Adam Omar, (2018) ‘Racial Profiling and a Reasonable Sense of Inferior Political Status’, Journal of Political Philosophy, 26/3: 1-20.

Kagan, Shelly (2011) ‘Do I Make a Difference?’ Philosophy and Public Affairs 39/2: 105-141.

Kaiserman, Alexander (2016) ‘Causal Contribution’, Proceedings of the Aristotelian Society, 116/3: 387-394.

Kutz, Christopher (2007) ‘Causeless Complicity’, Criminal Law and Philosophy, 1/3 289-305.

Kuznar, Lawrence (2017) ‘I detest our Confederate Monuments. But they should remain.’ Washington Post 18th August 2017 available at https://www.washingtonpost.com/opinions/i-detest-our-confederate-monuments-but-they-should-remain/2017/08/18/13d25fe8-843c-11e7-902a-2a9f2d808496_story.html?utm_term=.28dd745bc81f; (accessed 20/11/19).

Landrieu, Mitch as reported in the New York Times, ‘Mitch Landrieu’s Speech on the Removal of Confederate Monuments in New Orleans’ New York Times, 23 May 2017, available at https://www.nytimes.com/2017/05/23/opinion/mitch-landrieus-speech-transcript.html (accessed 20/11/19).

Matheson, Benjamin, ‘Apologising for Atrocities’, (draft manuscript).

——— (2019) ‘Should Aung San Suu Kyi’s Nobel Peace Prize Be Revoked?’, 18th March 2019, Bij Nader Inzien, available at https://bijnaderinzien.org/2019/03/18/should-aung-san-suu-kyis-nobel-peace-prize-be-revoked/ (accessed 20/11/19).

McMahan, Jeff (2009) Killing in War, Oxford: OUP.

Moore, Michael (2017) ‘Quess?’ ‘What the Removal of New Orleans’s White Supremacist Monuments Means to My Students’, Artsy 15 May 2017 https://www.artsy.net/article/artsy-editorial-removal-new-orleanss-white-supremacist-monuments-students (accessed 20/11/19).

Nefsky, Julia (2011) ‘Consequentialism and the Problem of Collective Harm: A Reply to Kagan’, Philosophy and Public Affairs, 39/4: 364-395.

Nguyen, C. Thi, (2019) ‘Monuments as Commitments: How Art Speaks to Groups and How Groups Think in Art’, Pacific Philosophical Quarterly https://onlinelibrary.wiley.com/doi/abs/10.1111/papq.12279, p. 10.

Peled, Shachar ‘Where are all the women? New effort to give them just due on monuments, street names’, CNN online 8th March 2017 at https://edition.cnn.com/2017/03/08/us/womens-monument-project-trnd/index.html (accessed 20/11/19).

Sartorio, Carolina (2019) ‘More of a Cause?’, Journal of Applied Philosophy online at https://doi.org/10.1111/japp.12370 (accessed 20/11/19).

Schulz, Johannes (2018) ‘Must Rhodes Fall? The Significance of Commemoration in the Struggle for Relations of Respect’, The Journal of Political Philosophy, 27; 2, pp. 166–186, https://onlinelibrary.wiley.com/doi/10.1111/jopp.12176.

Sears, Matthew A., (2018) ‘Monuments aren’t museums, and history suffers when we forget that.’ Macleans, 10 August 2018 available at https://www.macleans.ca/opinion/monuments-arent-museums-and-history-suffers-when-we-forget-that/ (accessed 20/11/19).

Shiffrin, Seana (2009) Reparations for U.S. Slavery and Justice over Time, UCLA Public Law & Legal Research Paper Series, Research paper no. 15-46.

Spicer, Tracy (2017) ‘Why aren’t more women immortalised in stone?’ Sydney Morning Herald 28 Sept 2017 at https://www.smh.com.au/lifestyle/tracey-spicer-why-arent-more-women-immortalised-in-stone-20170928-gyqm59.html (accessed 20/11/19).

Tadros, Victor (2018) ‘Causal Contributions and Liability’, Ethics 128: 402-43.

Timmerman, Travis, ‘A Case for Removing Confederate Monuments’, in Bob Fischer (ed.) Ethics Left and Right: The Moral Issues that Divide Us (forthcoming).

‘Take Em All Down’ at https://campaigns.organizefor.org/efforts/take-em-all-down-2 (accessed 20/11/19).

1.I understand ‘public’ to here mean something that is either owned by the state, or receives (substantial) financial state support. I do not address the issue of statues on private property.

2.Although not a statue, Marcus Harvey’s painting of Myra Hindley is a good example of a what we might call a ‘mere’ or even critical depiction of a historical figure, where there is clearly no expression of respect or admiration. The public reaction to Harvey’s painting demonstrates our (albeit rather selective) aversion to depicting wrongdoers.

3.The statue of Josiah King, a Union soldier, in Minnesota is a case in point. King sits atop a statue honouring Union soldiers. But Native American activists have recently drawn attention to the fact that King also participated in a 1863 massacre of Native Americans. Since his role in the massacre has been publicised, there have been calls for the statue to be removed. See https://www.mprnews.org/story/2018/09/27/iconic-minnesota-soldier-part-of-atrocity.

4.The ‘believe’ is important here: clearly, the claim is not that we do not build statues of wrongdoers, but that we don’t build statues of people whom we believe, at the time, to be wrongdoers. More on this, including on wrongdoers who also achieve important goods, in Section Five.

5.As I discuss in 2.1.2, it’s clearly the case that, at least sometimes, the removal of statues is controversial because some people do not accept that the subject was a wrongdoer. However, this is not plausibly true of all those who argue that we ought to keep public statues of e.g. Rhodes.

6.I say ‘some’ because some museums and galleries are e.g. primarily propaganda tools, rather than places of historical record or aesthetic appreciation.

7.This is, I think, compatible with some community members’ wanting to remove a statue, since the protesters object to being perceived to endorse these values, even if they do not endorse them. As C. Thi Nguyen argues, “[I]t is because monuments present themselves as expressing group commitments that it matters so much to the group members that they get it right.” (Nguyen, 2019, 10)


8.I address the separate issue of recontextualization in Section 6. My argument here is directed at those who claim that since ‘we’ no longer approve of, for example, colonialism, the meaning of a Rhodes statue has thereby been changed.

9.Note that Mandela’s participation in justified violent actions in the struggle against Apartheid does not make him a serious rights violator, even if innocent people were thereby killed. Justified transgressions infringe, rather than violate, rights.

10.Thanks to Tom Douglas for this point.

11.I consider some issues surrounding reform and rehabilitation in Section 3.3

12.E.g. Fabre, 2009, 43 -4; McMahan, 2009, 225. For dissent, see Frowe, 2014, 202-205. Of course, it is unclear how we should assess the significance of causal contributions: several metaphysicians express scepticism about the idea that one can, for example, distinguish between contributions in terms of size. See e.g. Sartorio, in press; Tadros, 2018, 402-43. For a defence of degrees of causal significance, see Kaiserman, 2016, 387-394.

13.Burch-Brown also suggests that removing statues can be a means of condemning and repudiating wrongdoing, drawing on Seana Shiffrin’s work on reparations (Shiffrin, 2009). See Burch-Brown, 2017, 69.

14.Ilhan Omar recently objected to being asked to issue a statement on female genital mutilation. See https://edition.cnn.com/videos/politics/2019/07/23/rep-ilhan-omar-calls-question-appalling-muslim-conference-sot-vpx.cnn

15.I do not address here what else repudiation involves (e.g. the disgorging of the benefits of injustice).

16.It’s worth noting here that some statues of Confederate soldiers are not of specific individuals, but are instead generic statues depicting a type. Since these figures do not depict specific persons, but are rather intended to represent Confederate soldiers in general, they cannot be honouring ‘the person’ as a whole. However, here it’s very clear that what is honoured is the wrongful practice: having these statues is akin to having a celebratory memorial of a massacre. Thus, there seems to be good reason to remove these statues as well as those depicting named individuals.

17.The Savile case is the obvious comparison here, although that case is complicated by the connection between Savile’s abuse of children and his fundraising activities (that is, he used the position he obtained in a children’s hospital as a result of his charity work as a means to abuse the children). But I doubt that in an otherwise similar case in which the abuse was not enabled by the charitable works we would be any keener to engage in the kind of weighing described above.

18.The converse can also occur, of course. Whatever we think about building statues of people who commit wrongs before reforming and achieving significant goods will probably also settle the question of building statues to those who achieve goods and then ‘regress’ and commit wrongs. The Myanmar politician Aung San Suu Kyi is an example (for discussion of this case, see Matheson, 2019)

19.Thanks to Derek Matravers for suggesting this.

20.This is not to say, of course, that all those who oppose removal do so on the basis of objectionable attitudes, or threaten violent opposition to their removal. But it seems plausible that many of those who are willing to resort to violent means in order to keep e.g. a Confederate statue are motivated by more than e.g. a concern about whitewashing the historical record.

21.Thanks to an anonymous referee for this example.