a blog about philosophy in public affairs

Are we socially (and not just legally) obligated to presume innocence?

Content note: this post contains and links to discussions of rape and sexual harassment.

Social attitudes towards rape and sexual violence and harassment have over the last few years been undergoing what Laurie Penny has aptly called ‘rape culture’s Abu Ghraib moment’. From Steubenville, to Jimmy Saville, and academic philosophers, we have been confronted with both how widespread rape, sexual violence and harassment is, and how awfully this is dealt with by the police, courts and institutions. Closer to home for me, a few months ago the Oxford Union president was arrested for rape and attempted rape (the charges were later dropped). This resulted in a campaign to have him resign his position as president and for invited speakers to cancel their appearances until he did. The ‘public intellectual’ A.C. Grayling however refused to cancel his appearance, saying that the president was innocent until proven guilty and should not be tried in the ‘kangaroo court of public opinion’. This has become a common response to accusations of rape (with ‘kangaroo court‘ the favourite and somewhat tired description). The alleged rapist, it is argued, should not be subject to social sanctions and society should reserve judgement because of the principle that people are innocent until proven guilty.

I vehemently disagree with this. But when challenged I have in the past been somewhat unsure of my reasons for disagreeing. One argument is that though innocent until proven guilty is an extraordinarily important principle, it is primarily a legal principle. That means it applies to the courts and the legal process of convicting someone of a crime. If someone is to be subjected to state punishment (from fines, to jail, to being executed), then they have the right to be presumed innocent until proven guilty so that the obligation rests with the prosecution and not the accused to prove guilt beyond reasonable doubt. It is however not clear that public condemnation of an alleged rapist should be subject to the same principle. As has been pointed out the so-called ‘kangaroo court of public opinion’ is not actually a kangaroo court. A kangaroo court (such as white lynch mobs) disregards the standards of a fair trial to punish the accused. Public discussion and condemnation does not (usually) seek to actually replace the legal process and determine guilt and then exact the kind of punishment normally reserved for the state.

But I am unsure of this argument. First, it relies on a kind of reasoning where the legal and social is entirely divorced that I would normally reject. I do not for example accept the absurd argument that women, queer people and people of colour have achieved equal status on the basis that many (but certainly not all) legal discriminations have been removed, because this is undermined by the continued existence of social oppression upheld through patriarchal, white supremacist and heteronormative norms. Second, public condemnation and discussion is not the whole story. Social sanctions, which include being personally or professionally shunned and being removed or temporarily stepping down from public positions, are graver than public condemnation and can approach state punishment in the consequences for the accused. Trying to argue that carrying out these kind of social sanctions does not punish the accused in the way a court does, seems unconvincing. Justifying it requires more than saying that innocent until proven guilty is just a legal principle.

I think the more convincing defence of public condemnation and social sanctions, and thereby overruling innocent until proven guilty, is based on the flawed legal processes and social attitudes that surround rape and sexual harassment and violence. Rape culture and its associated myths infect every step of the legal process from the police to judges. Combined with the social shaming and condemnation of victims, this mean that rape remains (as the graphic above shows) a dramatically under-reported, under-prosecuted and under-sentenced crime. In the absence of a correctly functioning legal system and societal attitudes that support victims I think it is therefore justifiable to publicly condemn and socially sanction alleged rapists and harassers. Of course this will vary from case to case, based on which crime they are accused of and the actions taken by the institutions that are supposed to deal with it, and there is no easy formula for this. I think that these actions are however necessary to challenge the ideas embedded in rape culture and replace them with the kind of norms and institutions that would seriously reduce the prevalence of rape and harassment.

In closing it is worth reflecting why people place so much emphasis on innocent until proven guilty when it comes to rape and harassment. I suspect that this is in fact one more feature of rape culture. At its heart rests the profoundly mistaken view that false accusations of rape and harassment are rife. I think we should remember that insisting that the accused is innocent until proven guilty, is so often based on the assumption that the victim is ‘lying until proven truthful’. To counteract that, I think it is central to believe and support victims. As Stavvers has convincingly argued:

‘Silence is the biggest weapon patriarchy has in keeping rape culture alive, and “I believe her” starts to tear down this wall and encourage and empower survivors to speak out. Because of this, it is crucial that we resist the attacks on this notion, the slurring it as “mobs” and “kangaroo courts”, because it isn’t. It’s solidarity in the face of patriarchy, and we should be proud that it is starting to terrify those who would rather we shut up.’


Two arguments on Scottish Independence, one for and one against


Defending Quotas


  1. Bruno, thanks a lot for this thought provoking piece. I find the two arguments you offer for not confining the presumption of innocence to the legal domain convincing. I was therefore less convinced by the case you made for making rape allegations an exception. I agree that there is something fundamentally wrong in a society and a legal system where victims of rape fear reporting rape, feel shame or think it futile to report it. And I agree that it is crucial to offer and show support and protection to victims so as to empower them to speak out and seek trial. What is crucial then, as you point out, is to counteract the presumption that the victim is “lying until proven truthful”. My question is whether presuming the contrary, that the “victim is telling the truth until proven lying”, commits you to rejecting the presumption of innocence of the alleged rapist/aggressor. Can we not hold both? That the victim is truthful and has a case that deserves to be heard while at the same time maintaining that we should not take action that might have serious impact and cost on the accused before we have proof of his/her guilt? You seem to think that we can only counteract the lying presumption by giving up the innocence presumption. I am not sure this is necessarily true.

  2. Hi Bruno, thanks for this post, which raises important (and timely questions). Can you clarify one point, which suggests itself if one looks at the picture above, but which you don't quite seem to endorse: namely that someone accused of rape who is not legally convicted is, with a relatively high likelihood, nonetheless guilty? One way of reading this discussing could be along the lines of: "how likely is it that someone is wrongly accused of being a rapist vs. how likely is it that someone is rightly accused, but not legally condemned (for example because of lack of evidence)?" The different factors (patriarchy, silencing, etc.) could all play a role in determining these likelihoods. But there could also be deeper underlying questions, and I'd be interested where you see those.

  3. Thanks Siba for the thoughtful comments. Perhaps they do not have to necessarily conflict, but they certainly can. For example in cases where the victim and the accused share an environment, such as a university campus, and it is deeply upsetting for the victim to be constantly faced with their attacker (e.g. cases at Columbia and Oxford). In these cases it seems that we do have to choose between believing the victim or the accused, because the victim has asked us to support them against the accused by for example excluding the accused from the campus or otherwise shunning social interaction with them. I'm not sure that we can hold both here?

  4. Thanks Lisa, you are right that this was underlying my argument and I should have addressed it explicitly. I do think the statistical likelihood plays an important part in why we should believe the victim. The Crown Prosecution Service report I linked to for example lists 5,651 rape prosecutions compared to 35 prosecutions for false allegations in the same 17 month period, and a man is apparently more likely to be killed by an asteroid than be falsely accused of rape.

    My other reasons apart from the statistical likelihood are related to the (1) personal effect on the victim and the (2) wider strategic effect. (1) I think that not believing a victim adds considerably to the harm they have already experienced. (2) I think publicly saying you believe a victim can contribute to combatting rape culture and opposes the efforts of rape apologists who at the same will be trying to blame and shame the victim.

  5. Thanks for the post, Bruno. One question I would be interested to ask you here regards the connection between the means and ends. If I understand you correctly, the thought is that the relaxing/dropping “innocent until proven guilty” in the social context for some period of time is to be valued as a means of correcting background culture. One concern that came to my mind about this approach is what kind of spiral effects might result: if we relax/drop the principle in some cases, will it lose some of its (important) general force or be easier to bypass it in other cases? May people feel more widely licenced to attempt to correct injustice via this route or, very worryingly, might governments subvert the line to argue that the principle can/should also be relaxed/dropped in order to combat terrorism? I honestly do not know the likelihood of these scenarios, but hopefully you see my worry. It leads me to wonder whether there might be alternatives to social sanctions that could have the same effects, but avoid some of these potential pitfalls. You say that these actions might be “necessary” to challenge ideas embedded in rape culture. I wonder how firmly you think they are necessary – that alternatives simply will not do the work that is needed?

  6. That's a valid concern Andrew. Especially in the light of people like Boris Johnson arguing that any Briton returning from Syria and Iraq should be presumed guilty of terrorism. But I'm not sure whether changing how we approach the social presumption of innocence with rape cases would really affect the legal presumption of innocence with terrorism. I agree that if we changed the legal presumption in rape cases this could make it easier to open in other cases, but I would be against that.

    I also think that socially presuming guilt is already an established practice. People take a stand on people's guilt all the time with various other crimes, before there has been actual trial. For example we think that arms manufacturers are guilty of gross human rights abuses, and take action against them even though they have not (and are unlikely to ever) faced trial. So I'm not sure that it represents such a significant departure from what we already do in public discourse and political action.

  7. Hi Bruno, thanks for the clarification. Although I’m sympathetic to your overall line, I think it is a somewhat dangerous line to take. For one thing, it is not clear that the likelihood of wrong accusations stays the same as it historically was (and ironically such a shift could come hand in hand with an an overall positive shift in culture, as an ugly side-effect). On the other hand you might say that the likelihood would still be very small, and much smaller of the likelihood of a reverse error. And cases of false accusations would be punishable; there was a case in Germany where I think a woman ended up in jail (I think) for having falsely accused a man a rape, who had ended up in jail because of that. What worries me a bit is that the stain of being accused of a rape is something very hard to get rid of, especially in times where such things can be found on the internet years later. So the bullet you have to bite – in terms of how unfair it could be to innocent individuals who are falsely accused – is quite a bullet…

  8. Just a quick comment on the arms manufacturers analogy: I would say this is somewhat different. Arms manufacturers act within a legally established framework (or let’s assume so for now) in what is a controversial area on which many would say that the legal rules are outdated and morally deficient. Committing a rape is clearly against the law, and there is also little controversy about the moral wrongness of rape – which is why, in turn, there is so much controversy of what should count as rape.

  9. Perhaps it is quite a bullet, but there are also significant harms associated with socially presuming an alleged rapist innocent, namely the harm to the victim in not believing her/him. Since being innocent is so much (much) less likely than guilty, and the harm in being falsely accused is bad but not as bad as being disbelieved, do you not think this outweighs it?

    Also, these concerns are also the consequence of the legal process of being accused of rape. The fact that you have been arrested for example is made public. The only way to avoid that would be anonymity for alleged rapists. Would you support?

  10. But the analogy does work if we think arms manufacturers are guilty of legal crimes as well (which they undoubtably do to sometimes e.g. bribing foreign governments, circumventing arms embargoes etc).

  11. As I said, I think I'm with you on the general line – so yes, I think I would bite the bullet as well. A lot depends on the concrete circumstances, I think, and sometimes concrete, case-by-case considerations may in fact be easier than general abstract claims. For example, if someone is accused of sexual harassment and might be a risk to other potential victims, then public condemnation seems unavoidable, if only to warn those who might become victims.

  12. Thanks for this great post, Bruno.

    Like you, I think that any justification for not presuming innocence will have to make reference to the failures of the legal system, for if the legal system were doing its job properly, it seems reasonable to defer to its judgments. I suppose your post highlights two possible justifications that we could appeal to in order to support your conclusion: (1) it may be that not presuming innocence will help to reform the legal system in the right direction, (2) it may be that not presuming innocence here and now goes some way towards correcting for the failures of the current legal system. It may be helpful to keep these two distinct lines of argument separate, since they may be susceptible to different kinds of objections.

    Also: Have you seen Thomas Vinterberg's 'The Hunt'? Its an excellent film and deals with several difficult, related issues.

    Finally: Since I am just about to head off to teach a seminar on Locke, it may be worth mentioning that, when discussing the inconveniences of the state of nature, Locke famously writes about the problems of allowing each citizen herself to judge and to punish wrongdoing.

  13. I agree Lisa, a lot will come down to specifics. And I wouldn't rule out the possibility that there might be cases where it might be better to socially presume innocence. I think I mean that the approach taken above should be the default way to respond. And I definitely agree that the case you mention is the strongest example of when we should publicly condemn.

  14. I think thats a good summary of the argument. Your suggested division is also useful, since it differentiates between whether not presuming innocence is only aimed at (1) counteracting the injustice of the legal system, or if it is also aimed at (2) reforming it to make it less unjust. I suppose I think that both of these reasons apply, but (1) would be sufficient to justify not presuming innocence.

    I haven't seen The Hunt, but it looks very relevant and will add to my list.

  15. Thanks for the further thoughts, Bruno. I am also unsure about the analogies. I am a little inclined to see the social and legal presumptions as somewhat more interdependent and I am also not sure the (legal and social) relaxation of this presumption in recent terrorism cases is so disconnected from a wider context in which it has become more common that social positions on guilt are taken before/independent of court proceedings, especially in high profile cases.

    That said, I perhaps mis-weighted my comment. I did not mean so much to question these effects. As I say, I find them very hard to predict. I meant really to lead to the question I posed towards the end of my comment: do you think that social sanction is a ‘necessary’ means of correcting the legal injustice, or might you think that we could pursue other avenues, such as investing our campaigning resources in reforming social/legal norms through educative programmes? One possibility is that such an alternative may demonstrate to a victim how seriously we take the injustice she has experienced without some of the potential difficulties with socially presuming guilt. I was interested to know whether your case rests, in part, on thinking that such alternatives are simply not enough?

  16. Andrew I'm also sure about wether its necessary. Such judgements are always hard. I tend to think that not socially presuming innocence does form a pretty important part of changing the social/legal norms. Perhaps it is a matter of how much they need to be changed. So perhaps in a period (like now) when the dominant social/legal norms are heavily biased against the victim, not socially presuming innocence is necessary. Then perhaps the more the norms are changed the less focus we put on this, because we begin to have more trust in the legal process working correctly?

  17. Thank you for the highly relevant and thought provoking argument.I have read it several times and still cannot quite decide what position to conclude. My intuition and experience supports your position but I also share the concern expressed by several people that it is unfair and dangerous to presume guilt in this particular. I was wondering whether we could draw an analogy to another crime which is rampant, under-reported, hierarchical and involves – most often – a very vulnerable and marginalised victim group – pedophilia. In cases of accuasations, which like rape are most often of someone in the same environment, there is a tendency to suspend the alleged criminal and keep them away from the alleged victim ex. teaching contract not cancelled but suspended with or without pay, there name is not made public but the victim feels recognised and heard because something public is done to acknowledge the potential crime. Couldn't some middle-ground such as this be possible with regard to rape? On another note, I thought it might make sense to link this post to another blog that has discussed this issue: http://feministphilosophers.wordpress.com/.

  18. That's an interesting possibility Anya. I like the fact that it gives a public signal of how seriously the crime is being taken. But my concern would be making accused rapists anonymous, because this is often how other victims hear that they are being charged and come forward. The accusations around various entertainment stars in the UK (Jimmy Saville etc) for example brought forward literally hundreds of other victims. Also what happens if the victim and alleged pedophile/rapist do not share an environment and the alleged does not have to be suspended from their work, does the victim still get some public acknowledgement of the seriousness of the crime?

    I think it would be great to link it to the Feminist Philosophers blog!

  19. Iain Crawford

    I think Grayling was right. To ostracise an accused man is to possibly prejudice the result of his trial…. ‘members of the jury, the accused has already lost his post, had speaking invitations cancelled…’
    What is the source of the graphic? Other estimates put false accusations at between 2 and 10% of the total.

  20. Thanks for this blog. This will serve as an eye-opener to those accused of rape. Like other types of cases, it’s true that we should not put emphasis on innocent until proven guilty. However, this will always depend on the situation. Perhaps there should be an analogy about the crime.

Leave a Reply

Your email address will not be published. Required fields are marked *

Powered by WordPress & Theme by Anders Norén