The plausibility of abhorrent views, and why it matters

The plausibility of abhorrent views, and why it matters Calum Miller, University of Oxford ‘Discussing infanticide is not the best way to win friends or secure admiring book reviews.’ So says Jeff McMahan (2007, 132), one of the foremost academic defenders of the permissibility of infanticide. He is not wrong: in a book review of The Ethics of Killing (2002), wherein McMahan defends the permissibility of infanticide, Stephen Mulhall (2002, 16) appears to accuse him of ‘thinking evil thoughts’, of the kind Anscombe (1958, 17) once described as showing ‘a corrupt mind.’ As an empirical point, it is hard to disagree with McMahan’s evaluation here. And, while I think that McMahan is a most gentlemanly and kind person, I think what he defends is indeed truly evil.1 But ought the defence of evil positions elicit contempt in all cases? More radically, should we suppose, with Anscombe and Mulhall, that the mere entertainment of evil thoughts is morally objectionable? I shall argue to the contrary. Not only must we think seriously about evil suggestions as individuals, we should do so as a community, and therefore should do so publicly. There are various pragmatic reasons for allowing freedom of speech, including in academic contexts. One such reason is to allow wicked views to be publicly ‘shown up’ and thereby rendered incredible in the fair light of open debate. Such views could not, in that case, be held by proponents to have been unduly or sinisterly quashed by an authoritarian regime scared of letting out the truth. And this would help to give an accurate picture of the number of adherents: movements based on such views would have less need for secrecy and could not as plausibly claim artificially inflated numbers among their ranks. Another reason for preserving academic freedom of speech is to minimise the risk of applying limits wrongly. Even if it were the case that some views ought not be discussed, why think that the actual authorities would accurately recognise and employ the relevant boundaries? Could a precedent for such boundaries not be exploited to stifle genuinely needed discussion on the grounds that the ‘right’ answer had already been reached by the relevant authorities? The history of governments and zeitgeists, and the moral heterogeneity evident across and within times and cultures – not least our own – do not constitute cause for great optimism about the results of such restrictions. Indeed, most of our views about who counts as human, along with other core moral beliefs, have at some point been doubted or even deemed anathema. As McMahan (2016) himself puts it: It is salutary to recall that the early efforts of those we now recognize as having been in the vanguard of moral progress – abolitionists, campaigners for women’s rights and female suffrage, vegetarians and opponents of vivisection – have always been fiercely resisted and ridiculed by those to whom it was inconceivable that the common sense view at the time might be mistaken. 1 McMahan himself has always extended grace to those with whom he strongly disagrees: the first time we met, I presented a paper arguing for a foetal ‘right to life’ – the conclusion of which he rejected emphatically but met with most helpful and kind comments. There is thus at least some inductive reason to suppose these are not the lucky few decades in history where we have finally achieved moral equilibrium. We have set no such precedent.2 One might, alternatively, support academic freedom of speech on more theoretical grounds. One might think it a simple matter of liberty or autonomy that academic freedom be preserved. Perhaps a restriction on academic free speech would contravene certain human rights. Or perhaps it is simply part of the purpose and nature of universities to allow a completely open exchange of ideas.3 I do not pursue any of these lines of argument here (though they link closely to my main argument). Rather, I suggest that the character of moral epistemology itself suggests no clear connection between the (apparent) heinousness of a view and its obvious falsehood – at the very least, there is no reason to think that wicked views will always be obviously false. This, I suggest, has a number of implications: first, that generally decent people can have wicked views and that having a wicked view should not automatically provoke contempt for the holder. Relatedly, having wicked views is not always culpable, and we should in certain respects and in certain cases be more sympathetic to those advocating them. The second main set of implications concerns not the holders of seemingly wicked views but the extent to which we should entertain such views. The most obvious implication is that some wicked views may be somewhat plausible. I suggest that this – along with sympathy for their proponents – gives some reason to entertain at least some wicked views in academic settings. While I do not go so far as to say that all views should be allowed extensive coverage in academic settings, I suggest that ostensible wickedness alone cannot be grounds for academic censorship, and that the considerations adduced herein lend support to considerably expanded limits of academic freedom. Moreover, alternative criteria for academic restrictions – such as plausibility – do not clearly generate the sorts of academic restrictions often thought plausible by some within academia. While the conclusion I reach may appear radical – and I certainly do not intend to thereby license the intellectual or moral legitimacy of any wicked positions – it is, I suggest, supported by the considerations I adduce here. I understand academic freedom as a broadly negative freedom: the freedoms not to be fired on the basis of arguing for a particular view, not to be disinvited after e.g. a student group invitation, not to have work otherwise involuntarily retracted (e.g. because of complaints to a journal).4 My argument varies in strength for each of these conclusions, and may also support 2 My overall thesis does, in fact, link very clearly with this point. So although it can be seen as an independent reason in favour of academic freedom, I note that it is heavily augmented (and so deserves further reflection) by the arguments developed in this paper. 3 Clearly all these ideas deserve considerably more working out, and I regret that I am unable to offer more than a tip of the hat in their direction in this paper. 4 A few examples of increasingly invasive university/academic actions from the last few years are worth mentioning: the journal Hypatia was berated and encouraged to retract an article in which Rebecca Tuvel argued for a symmetry between transgenderism and transracialism (Weinberg 2017); a number of events put on by Oxford Students for Life, a pro-life group at the University of Oxford, were cancelled or required police intervention to proceed (Oscroft 2017); the University of Strathclyde’s Students’ Association banned a pro-life group (Gray 2016); feminist writer Julie Bindel was banned from a University of Manchester event she had been invited to because of (impolitely expressed) scepticism about transgenderism she had expressed in the past (Palmer 2015); a Sheffield University student was removed from his course for stating on Facebook that homosexuality and samesex marriage were sinful (Grierson 2017); a Wilfrid Laurier graduate student was interrogated by university officials after playing a clip of Jordan Peterson in a class she was teaching (Joseph 2017); Bret Weinstein was told other conclusions. I do not intend here to give a principled argument for unrestricted academic freedom of every variety: I merely offer an argument which lends substantial support to various academic freedoms. Importantly, I note here that the academic freedom I defend does not involve unrestricted teaching freedom: teachers – at least in more objective fields – must be capable of, and willing to, reflect to their students consensuses within their fields. So, for example, a science educator who deceives their students about scientific consensus (including by withholding information) will not, in most cases, be doing their job properly, and so may be liable to a termination of their contract. There are important academic freedoms other than teaching a radical opinion as mainstream, and it is those which concern me here. The arguments I give do lend some support to more liberal teaching freedom as well; but my primary target – especially as this is the domain most under threat at present – is the more liberal marketplace of ideas in university talks, journals, and so on. A case Let us attend to some concrete cases to begin to illustrate the argument. Suppose that there are two competing views, both of similar plausibility, but both of which entail that the other view is not only wrong but utterly wickedly so – or, at least, has wicked consequences. It is not difficult to find real-life approximations, given some mutually agreed background information and values. Let us say that Daisy is accused of murdering Tobias, though in fact she is innocent. Nevertheless, the evidence against Daisy is very strong – Tobias’ own family, known to be honest and reliable, saw Daisy’s uncanny doppelganger Philippa murder Tobias in broad daylight, not knowing Philippa existed. They have no reason to doubt Daisy’s guilt, and so it seems clear that their belief is justified. And let us say that another witness wants to bring important evidence to the table – that Philippa exists, looks identical to Daisy and is, in fact, the killer. Or suppose that someone – also knowing this – defends Daisy’s innocence publicly in the media. Now imagine what Tobias’ family might say: they have every reason to believe that Daisy is guilty and that those trying to defend Daisy are perverting justice, denying the gravity of the crime Daisy has committed, being insensitive to their loss as a family and perhaps even facilitating future murders. To allow the new evidence to be heard might therefore be to take it more seriously than it deserves, to give dangerous and dehumanising speech a platform, and so on. From their perspective, the case is clear: murderers should not be defended, Daisy is very probably a murderer, and so should not be defended.5 by police at Evergreen State College that they could not guarantee his safety after he objected to a day event encouraging white people to stay off campus (Weiss 2017); a faculty member was injured after a protest against Charles Murray prevented his speaking at Middlebury College (Weiss 2017); a Bath Spa University student was prevented from research on transgenderism on the basis that the results might be ‘politically incorrect’ (BBC 2017); and so on. It is not difficult to find many similar cases in the news, although each must be taken with pinch of salt. 5 It may be unclear to you that Tobias’ family are justified in making these claims. If so, very well, for this is essentially my point. The reason it is important is that those who defend the innocence of accused parties in parallel contexts are often met with scorn akin to that of Tobias’ family, even when the evidence is either inconclusive or poor. See e.g. the backlash at football clubs for considering signing footballer Ched Evans even after the overturning of his rape conviction after new evidence came to light (BBC 2016). Along similar lines, see efforts to overturn the presumption of innocence in related cases (Rayner and Gardner 2015; Real Clear Politics 2015): a well-intentioned move, we should be absolutely clear, but again seeming to confuse the heinousness of the alleged crime with the plausibility of its veracity. What should be done in this case? Of course, we have artificially constructed this scenario so that the family are simply wrong, but it is not implausible to suppose that this form of case approximates at least some real cases. And when it comes to deciding innocence or guilt for this sort of crime, the intuitions of most people are clear: everyone has the right to a fair trial – because only when the best case, with all the evidence, is presented for both sides are we most capable of reaching an appropriate judgment. So it is curious that outside the courtroom these sorts of principles are frequently rejected. When a heinous view is espoused in the popular sphere or, sometimes, in the academy, it is frequently met with precisely these kinds of objections: allowing the view a platform is insensitive or offensive or dangerous, helps to defend heinous people or actions, or gives the view the appearance of more legitimacy than it deserves. Even true positions are sometimes barred from discussion on similar grounds: just this week of writing, a UK university barred research into transgender people ‘detransitioning’, seemingly partly on the grounds that it might cause offence, despite there being no concern that the results or conclusions themselves might be false (BBC, 2017). And, of course, given the enormous number of views which have been banned from universities over their history, it is virtually certain on probabilistic grounds that at least some of them have been true. Are there relevant differences between the courtroom case and university cases? On inspection it appears not. One might think that while in a fair court trial both sides have the best possible case presented for them, so that truth is likely to win out, this might not be the case in a university due to lack of resources, making deceit more likely to win than if no discussion were permitted at all. But this is not a very strong reason: in the first place, it is hardly plausible that lawyers always present the case for their defendant well. But, equally pressingly, there is every reason to think that universities should already have at least someone available to represent the side of truth relatively well – at least, if it is a widely recognised truth, which we are assuming is the case if the issue is whether or not to ban its negation. Indeed, if there is no one on campus available and capable of presenting the case for the widely recognised truth at least reasonably well, one wonders whether such a view deserves its widespread acceptance and a platform hegemony. Another relevant difference might be that lawyers tend to debate empirical facts, while perhaps agreeing about the evaluative facts. They don’t disagree that murder is wrong or on the appropriate punishment: they just dispute whether Daisy committed murder. Could one not say that the disputants should consider each other doxastically mistaken but not thereby culpable? Perhaps they agree on the core values and it is only the secondary prescriptive views (based partly on different empirical views) which differ – with terribly unfortunate consequences. But nor is this as incisive as it might seem. For those who think that some academic answers should be off-limits will surely include some empirical disputes which are clearly politically charged: whether the Holocaust happened, whether some races are better on average at some things than others, or (in the Bath Spa case) how many post-op transgender persons regret their decision and decide to ‘detransition’ afterwards. In any case, it will not always be clear in advance that any given dispute does not turn, in part, on an empirical matter. In advance of giving the opposing view a detailed hearing, then, there can be no reliable way of ensuring that the dispute is entirely evaluative. On the other hand, there are surely some legal (or otherwise accepted) cases where evaluative facts are in question: whether or not a certain consideration should be deemed exculpating, for example. So this factor is not going to satisfactorily account for our discrepant judgments. So take an academic example – in particular, an example where the heinousness runs both ways. The abortion debate is one of the best examples of this. There are those who think that foetuses have the same moral value and rights as adult humans, and who are thereby inclined to think that abortion is tantamount to murder.6 On the other hand, there are those who think that foetuses have significantly fewer and less pressing rights than adult humans, such that abortion is permissible. In particular, it has been argued persuasively that the most plausible position in support of abortion rights also entails an extreme view – the permissibility of infanticide (McMahan 2002; Giubilini and Minerva 2013; Singer 1993; Tooley 1972). As Tooley (1972, 38) puts it, ‘One of the interesting ways in which the abortion issue differs from most other moral issues is that the plausible positions on abortion appear to be extreme positions.’ Dworkin (1993, 10, as quoted in Greasley 2017, 15) similarly comments, ‘Selfrespecting people who give opposite answers to whether the fetus is a person can no more compromise, or agree to live together allowing others to make their own decisions, than people can compromise about slavery or apartheid or rape…’. Finally, although she neither takes a pro-life view nor endorses infanticide, Kate Greasley (2017) has recently argued compellingly that if the foetus is granted personhood, there can be no plausible defence of abortion’s legality except in the most extreme of circumstances (e.g. if both the pregnant woman and foetus are already destined to die, and abortion is likely to save the woman). What can we say about this sort of case? In the first place, it seems that both sides are likely to act heinously (even if not culpably) if they are wrong: if the abortion opponents are wrong, they are likely to gravely inhibit the autonomy of pregnant women for no objective benefit,7 while if the abortion defenders are wrong, they are likely to commit or endorse something not too far from murder (this becomes clearer in the infanticide cases). In fact, it is likely that one of these extreme views, both of which are frequently characterised as evil, wicked, corrupt, and so on, are likely correct: that abortion is immoral and should be illegal except when the mother’s life is in danger and the foetus is likely to die anyway; or else that abortion and infanticide are widely permissible.8 So if there are honest, non-culpable and rational (I call such agents ‘sincere agents’ for shorthand) proponents of both views, it seems that some such people are liable to have a heinous opinion (namely, either that women should be deprived of fairly basic freedoms of bodily autonomy, or that we should condone the killing of members of the human moral community) and probably commit heinous acts on that basis. And it is difficult to deny the premise. Jeff McMahan – who endorses the permissibility of abortion and infanticide – is undeniably honest and informed. As he himself notes, there is little vested interest in publicly and explicitly condoning infanticide. And, while I think he is wrong and that my own position is more reasonable (as most of us presumably feel about our own opinions), it is simply not credible to insist that he is straightforwardly and culpably irrational. His position is extremely well thought out and relies on some credible intuitions about certain thought experiments. Again, I think he is very and dangerously wrong. But to dismiss him as straightforwardly 6 Let us suppose, for the sake of argument, that they reject Thomson-style violinist arguments which suggest that even adult humans would not have a substantive right to life if dependent on another human. Reasonably comprehensive responses to such arguments can be found in Greasley (2017) and elsewhere. 7 Plausibly, too, they are liable to the charge of ‘speciesism’ – perhaps on a par with racism – for prioritising human life above animal life. 8 Of course, not everyone agrees that one or the other extreme view is correct. Unfortunately, I do not have space to defend this view. But in any case, it is not strictly necessary, since all I am arguing is that both these extremes have opinions which might be heinous if wrong. In fact, perhaps most debates regarding who counts as a person count as similar instances. irrational is not plausible (even if we think, as I do, that the arguments opposing his view are stronger). Yet it is equally hard to deny that there are sincere opponents of abortion. For some of us – especially in the medical profession – vocal opposition to abortion is liable to make us exceptionally unpopular and probably inhibit our career prospects. I can vouch at least for myself when I say that I do not enjoy the unpopularity that comes from voicing my position, and that I would prefer to be more popular. My vested social, professional and pecuniary interests all lie in adopting the prevailing moral sensibility (although without going as far as condoning infanticide). There are plenty of us – myself included – who have studied the debates in detail and who were only convinced after reading the academic literature on the topic. And, again, it is difficult to simply dismiss us as fundamentally irrational. Just as there are some strong intuitions in favour of the importance of psychological capacities for rights, so there are strong intuitions in favour of the importance of human equality and against the permissibility of infanticide. I think the former intuition can be accommodated readily within my framework, and McMahan thinks the latter intuitions can be accommodated readily within his own (though ultimately rejecting at least the latter as false). But we can at least understand why each other’s (plausible) intuitions lead to the beliefs they do. Both sides have at least some plausibility behind them. And yet one of us must be heinously wrong – despite our honesty, education and rationality. One might be tempted, again, to insist that this is not really a dispute about fundamental ethical facts: both sides might agree that murdering persons is wrong – they just disagree regarding who counts as a person.9 Sincere agents would not disagree about the fundamental ethical facts – and it is disputes about those which we should forbid. But this is not a feasible move: one could just as well say that Nazis and non-Nazis both agree on the fact that murdering persons is wrong – the former just disagree that Jews (and others) are persons. Obviously if any views are to be banned, it ought to be ones like these – and so one cannot appeal to this reasoning to justify our tolerance of this debate. There is a sort of middle ground case in this vicinity: Peter Singer was recently protested for his views on disability, since he thinks (roughly) that certain cognitive features are determinants of value – features which some disabled people do not have. His opponents think that he has abhorrent views about disabled people by counting them as less than human, while someone of his view might well plausibly suppose that his opponents have abhorrent views regarding the extent to which we force each other to look after disabled people (e.g. by forcing taxes to help care for them, or by discriminating in favour of them against cognitively comparable animals). Clearly, some people think his views are literally intolerable: they should not be given a place at the academy. Others10 who find his views contemptible think that they nevertheless should be allowed a place in universities. And even many who find his views on disability repulsive find themselves indebted to his work on global poverty and animal suffering.11 So we have a useful spectrum of cases – infanticide, disability discrimination, and Nazism – where the personhood of ostensible persons is denied, and where there are differing levels of resistance to the permissibility of these views in academia. If we cannot find significant formal differences between these cases (there may be differences in plausibility, but Of course, it is equally possible that they disagree on more fundamental ethical facts too – I am just highlighting a possible informative case. 10 See for example George and West (2017), a public statement which, though not explicitly mentioning the Singer case, appeared to be in response to it and was signed by Singer himself. 11 I am grateful to Donald Downs for making this point to me. 9 I will later argue that this is likely to be person-relative and is not a good justification for banning some), then perhaps the discrepancy should lead us to re-examine our thoughts about limiting academic speech. Perhaps you are not convinced by the infanticide example. It might seem obvious to you that infants are persons. It might seem obvious to you that they are not. But for any of us, there are going to be at least some marginal cases of personhood. Of course, it would be impossible to settle on an uncontroversial example, since we all have different ideas of personhood. But, for example, suppose you think that certain cognitive traits are necessary for personhood. There is such a wide spectrum of cognitive life – a spectrum that is close enough to continuous – that it is virtually inevitable that, somewhere between the two ends (persons vs. non-persons), there will be a grey area admitting of at least some agnosticism. Denying the equality and personhood of some agents in this grey area, on the one hand, risks a heinous crime against them. But granting personhood ‘just in case’ only pushes the problem back a step: what about the agents just below them on the cognitive scale? And in any case, granting personhood ‘just in case’ might well require considerable deprivations of liberty elsewhere – by forcing everyone else to look after them through taxes, and so on. There is simply no easy answer, and risks abound each side.12 Maybe there is no entirely uncontroversial example of disagreement among sincere agents regarding heinous acts. In that case let us think more abstractly. Let us call the formal situation I have described a contemptible dilemma – where both sides of a dilemma are, if wrong, committed to a contemptible or heinous position. And, where there are actual proponents on each side of a contemptible dilemma, we have a contemptible dispute. I have argued that some such disputes are sincere – that is, they involve sincere agents. Just as McMahan honestly believes that restrictions on abortion are undue violations of women’s bodily autonomy, so I honestly believe that abortion is the unjust taking of a human life. Neither of us have much incentive to lie in these ways and every incentive to take more moderate positions. The evidence of sincerity is compelling. The evidence of education and broad rationality is likewise strong, and so we seem to have such cases. I suggest that there is no theoretical reason why contemptible disputes should be readily soluble, or why all sincere agents should end up on one side of the dispute. What would need to be the case for this to be false? We would need a principle along the lines of: ‘any heinous view must be obviously false to all sincere agents.’ But why think that such a principle is true? Do all heinous views have an epistemic probability of 0? There is certainly no reason in probability theory why this should be obvious. Nor is it clear how any laws of psychology or moral epistemology should lend themselves to this conclusion. Those views which require some reflection or inference trivially require at least some deliberation regarding their truth – and so clearly must permit at least some debate (even if privately in one’s own head), allowing for the initial possibility that the heinous view is, in fact, true. I suspect most of us would concede that there are at least some heinous views which we have ruled out by such inference. It appears to follow that such deliberation over whether a In case you are tempted to think that this could be escaped with a different criterion of personhood – being a human being, for example – run through the exercise again, with hypothetical cases if necessary. There will always be difficult cases. In that case, what about ET? He is not biologically human but nevertheless plausibly counts as a person. But then our criterion includes human-like organisms. And again, the relevant likeness will admit of both so many degrees and so much agnosticism that the problem is bound to recur. 12 heinous belief really is heinous is not necessarily concerning: it is a natural part of our moral knowledge. And there is no clear reason to prohibit communities from aiding this deliberation. Perhaps there are some moral beliefs we rightly hold without reflection. But is it the case that every single heinous belief possible is excluded by these? And is it the case that every single one of these beliefs are known intuitively by every sincere agent, and known with such conviction that they require no deliberation at all? Again, while this is not conceptually impossible, there is no reason to think it is the case. And it seems prima facie relatively implausible. Either way, some of our core moral beliefs will be held on the basis of some reflection. And both reflective and non-reflective moral beliefs are subject to disagreement among sincere agents. And so it follows that, very probably, there are some contemptible disputes in which a sincere agent has a considerable degree of agnosticism. Practical implications We have contended that there are contemptible disputes, that some such disputes are sincere, and that there is no epistemological reason for thinking that they should be readily soluble. Now let us draw out some implications. The first implication of these theses is that it is extremely unlikely that all morally wicked views are obviously wrong or implausible – and especially unlikely that all morally wicked views are obviously implausible to all sincere agents. This has a further implication uncomfortable for many: that some morally abhorrent views are nevertheless plausible, or even probable (given limited evidence sets). This should not surprise us too much: most of us have probably held views we later thought to be abhorrent. It is possible that we were simply being irrational at those times – but it is also possible that some of us simply lacked some of the relevant evidence or had different intuitions which were not demonstrably false. In any case, why assume the current zeitgeist is entirely immune to the same limitations that have pervaded human moral thought for so many centuries? And if we are not, then it is likely that even the ethical elite largely hold to some abhorrent views – and perhaps on matters in which they think precisely the alternative is abhorrent. That the relations between moral facts, moral knowledge and sincere agents generate this uncomfortable structure plausibly has further implications. In the first place, it should prompt us to re-examine any moral discourse which divides people into the essentially decent and the wicked. Sincere agents likely, in some cases, hold abhorrent views, and they should not thereby be discounted as thoroughly wicked or bigoted people on those grounds alone. Likewise, holders of wicked views are not necessarily culpable for their doing so. So, in the first place, there should be greater sympathy and openness to those with ostensibly wicked views on account of their plausible sincerity. This must obviously be tempered, however. This sympathy and openness is not intended to diminish the seriousness of abhorrent moral views and actions. If we are confident that someone is suggesting something abhorrent, we should be clear about this and, in our actions and policies, oppose them just as robustly as otherwise.13 For example, I am confident that infanticide is very seriously wrong – so I will continue to labour as an activist against it, I will continue to decry it as evil – including in academic settings, and I would support strong legal 13 This obviously admits of substantial variety. From my own Christian tradition, grace and forgiveness are always to be extended for even the most heinous of crimes, and vengeance is strictly forbidden. But for others, heinous crimes are to be met with very severe punitive measures. Nothing in my argument turns on this. punishments for those who commit it. I am not intending a kind of moral scepticism or moral paralysis. How our convictions should prescribe our actions is notoriously difficult to determine (see e.g. Buchak 2014). But what seems at least clear is that we should still act on at least some of our convictions. And those cases in which we are most confident that we have a strong obligation seem like optimal candidates. The sympathy is therefore strictly limited: we should try our best to listen to and understand their heinous views and the reasons for them. And, at least in some cases, we should be more hesitant to denounce them as a thoroughly wicked person. For it is simply a fact of life that (relative to average humans, at least), some decent people will have some abhorrent views. Our sympathy should be for hearing and understanding the ideas, and (at least in some cases) for the person – not being too quick to judge them as a whole. This does not make one a ‘sympathiser’ in the sense that one enjoys, is partial to, or encourages their views. It is an objective hearing in a (hopefully) objective academic context. If the view is truly in correct and therefore detestable, one would hope that an academic setting would provide the best opportunity of soundly refuting it, rather than letting it fester and gain popularity in less critical environments. And if not – on the off chance that it is correct – academia might well be best placed to recognise this and allow it to pick up speed, as well as refining it as appropriate. So we have, too, at least the rudiments of a rationale for being particularly interested in academic freedom of speech. These are surely some interesting results. That sincere agents are liable to have contemptible views from time to time – although inevitable on reflection – is nevertheless surprising. And given the plausible implications – that having a contemptible view, or maybe many contemptible views, does not mean one should necessarily be entirely discredited as a moral thinker or agent14 – we have substantial work to do in contemporary discourse. This has generated some reason, therefore, to preserve a relatively liberal conception of academic freedom. There are many practical and theoretical reasons for preserving academic freedom (in a broad sense), some of which I identified at the beginning of this article. But the recognition that the geometry of moral knowledge itself draws a fine – and sometimes evidentially or rationally indeterminate – line between the morally abhorrent and the morally obligatory should give us very strong reason to. For there are a number of reasons to engage with sincere disputants who seem to us to hold contemptible views. The first is that such engagement is good for them. As described, it is possible that such disputants are still otherwise reasonably good people. And in that case, we ought to help them come around to the more probably correct view – for their own sake and for others (indeed, this is probably true even if they are not otherwise good people – but their being otherwise good is extra motivation). Simply banning their view from discussion is more likely to breed resentment, and is a great injustice to them if they are, in fact, correct. 14 I leave open the question of whether it can ever be sufficient to discredit one as a moral thinker or agent. Certainly, one contemptible view can cause great suffering and lead to absolutely heinous acts – that is not in dispute. And certainly we should treat the doer accordingly. But my point here is to emphasise that ethics can be heavily compartmentalised – evil though the one view may be, it simply does not entail in every case that the person is thoroughly evil. And my view here does take seriously the idea that we should appreciate the good in someone who might have committed terrible acts. The second, more pressing reason is that such engagement is good for the pursuit of truth. This was captured most favourably by JS Mill (1859, 33): Were an opinion a personal possession of no value except to the owner; if to be obstructed in the enjoyment of it were simply a private injury, it would make some difference whether the injury was inflicted only on a few persons or on many. But the peculiar evil of silencing the expression of an opinion is, that it is robbing the human race; posterity as well as the existing generation; those who dissent from the opinion, still more than those who hold it. If the opinion is right, they are deprived of the opportunity of exchanging error for truth: if wrong, they lose, what is almost as great a benefit, the clearer perception and livelier impression of truth, produced by its collision with error. For Mill (and I think he is probably right), the chance that the minority opinion-holder is right is an obvious reason in favour of free speech: convincing the academic community thus would be a great service to it, and probably to the whole world if it should be accepted more widely. But even if it is wrong, Mill suggests, our knowledge is edified, since the nature of its wrongness helps to illuminate the truth. Returning to our earlier examples, suppose we think that McMahan is wrong about infanticide. Nevertheless, his work on the topic is extremely sophisticated, and his discussion may help us to navigate moral territory better. Suppose we think he is right, for example, that certain psychological connections in an agent (which he thinks is an embodied mind) give that agent some interests to be respected, but find that his arguments elsewhere are lacking. This gives us considerable reason to rethink our attitude towards animals and the harmful ways in which we treat them. Locating our disagreement most precisely gives us moral and practical clarity in other domains, and so serves truth. Even cases where our disputant is most obviously wrong – say, in denying the Holocaust – thinking about what exactly is so contemptible about such a position (it could be the result of culpable ignorance or intellectual laziness, or it could be subconscious racism, or it could be simply correlated with antipathy towards Jews, and so on), prompted by such discussions, may help us in other areas of ethics. So truth, too, is served. Finally, returning to the possibility that our disputant is correct, it helps all of us. For it might just correct our views and might, indeed, prevent us from committing heinous acts on the basis of them. Of course, this will be very unlikely in some cases. And it might be that our views are relatively harmless even if false, and so we are playing a low risk game. But there is always at least the risk of the injustice of moral carelessness and wrongful condemnation. Since this is in some cases extremely unlikely, this benefit may reasonably be weighed against the risks of academic freedom. Fortunately, for the reasons I gave earlier, inter alia, the risks of academic freedom seem low in almost all cases, and indeed there are other benefits to academic freedom. Nevertheless, I concede there is some reason against academic freedom, and that the reason I have given for academic freedom here is of varying strength depending on these various other factors. And indeed, my argument does not suggest that academic freedom should be maximised at the expense of other resources: universities ought not spend large amounts of resources on inviting disputants in many cases, for example. But I continue to note that what has seemed incredible to past generations has turned out to be both true and morally crucial. There have been plenty of societies where it has been inconceivable that members of other races or tribes or groups have been just as morally valuable as one’s own, for example. And, nevertheless, showing the wrongness of this view is a crucial foundation of modern liberal moral thought. If the worst that can happen is not so bad, if there are other benefits to academic freedom, and if or when there is a relatively low cost to allowing such views to be heard, the chance seems well worth it. Objections Let me conclude with a few thoughts on our intuitive resistance to academic freedom and, by doing so, clarify my position. I am no less repulsed by some opinions than everyone else, and I have not forgone the instinct to simply declare some opinions anathema and not give them the time of day. So let us briefly consider some possible bases for such dismissal. One possible thought behind no-platformers, protestors, and those who hold to narrower conceptions of academic freedom is that whether a view should be discussed is a function of how appalling the view is, in which case any view – no matter how plausible – which is probably appalling should not be discussed. The anxiety may be that by discussing a view, we are tacitly conceding that it is not really appalling. Of course, my previous discussion attacks at precisely this juncture: there simply is no reason to think that every heinous opinion will be very improbable. So when can we say that an opinion is too heinous to be discussed? If the probability of its heinousness if 0.51, is that sufficient? But then there is a perfectly reasonable chance that we are completely wrong. In that case, it seems clear that to prohibit its discussion would be imprudent. So appallingness simply cannot be a sufficient criterion for banning a view from the academy. But this suggests that the principle itself is simply wrong. It is simply not the case that to think a view should be discussed is to tacitly concede that it is not really appalling. Perhaps whether a view is worth discussing is not just a function of how appalling it is but of how plausible it is. Might heinousness combined with a certain degree of implausibility ground academic restrictions? There are a number of reasons why this might not be so simple. First, there is the problem of arbitration. Who decides how implausible a view is? Sincere experts might disagree, and in any case it is not commonplace for experts in relevant fields to be in charge of policy on academic freedom. What we would need is for the view to be implausible to all sincere agents, and this simply casts the net too wide to generate much more than minimal restrictions on academic claims. Secondly, showing that a view is implausible in the first place will require some engagement with it, and so some academic engagement with implausible heinous views is necessary to begin with. There is, moreover, the problem of new evidence and arguments: it is always possible that new arguments may overturn even a very implausible view, and unless we are prepared to countenance those arguments, we cannot reasonably maintain that no strong counter-arguments to our position exist. The problem is that once we have accounted for these (e.g. by finding cases where all sincere agents do agree), there will not be much left to censor – and it is particularly unlikely that reasonable censorship will line up with what many contemporary activists are attempting to censor (for example, debates on abortion or transgenderism).15 One wonders if – even aside from questions of how to ground the restrictions – there is a more transparent inconsistency in the way meta-ethical views are treated: we suggest it is heinous and perhaps censorable to deny that rape is objectively wrong if one is a moral realist, but it is less common to suggest that moral nihilism is morally heinous or censorable, even though it entails precisely the same denial. 15 What seems more reasonable than trying to consistently censor views on more nuanced grounds than this, therefore, is simply to be clear from the outset that discussing an appalling view is not to concede that it is worth discussing, plausible, or less than appalling. If this proviso is understood from the outset, we can circumvent the problem of heinous and implausible academic views being seen to have academic credibility or moral legitimacy. Mulhall (2002, 16) describes a second possible objection as anticipated by Anscombe: ‘She thereby expressed a (highly controversial but hardly unintelligible) fear not only of evil actions, but of thinking evil thoughts – a fear of the dulling and degrading of moral sensibility that such acts of contemplation can encourage and express.’ It is not difficult to agree that there is something unnerving about ruminating on evil actions. The Christian tradition – shared by Anscombe, Mulhall, and myself, as well as many of the foremost respondents and critics of the unpopular views of Singer and his company – is clear about fleeing from sin and about the importance of purity in one’s thoughts just as well as one’s deeds. But this intuition may be preserved – in the Christian tradition and outside of it – without suggesting that we literally need not think about evil propositions at all, or in detail. For there is a huge difference between deliberating over evil actions (perhaps pleasurably) and thinking about them in a descriptive, analytical way. The former warrants avoidance; it is far from clear that the latter does too. Nevertheless, it is still open to wonder whether discussing such views dullens our moral sensibilities, as if encouraging us to consider them as academically respectable or mere differences of opinion. This is a reasonable concern that I share. It might even be that discussing the views is not for everyone – those who are worried about this effect might want to voluntarily abstain from the debate. But that is no reason to prevent anyone from taking part in the debate, especially not those who are able to hold their condemnation of evil views and actions quite firmly in the context of academic debate, as many of us are. Finally, we might offer some further ways of resolving this tension. One of has been indicated already: part of the solution is surely to insist on being clear on the heinousness of the view one is addressing. Another solution might be to resist the strong division of academia and polemics. This will need to be done carefully, since there are clearly strong advantages to the civility and (relative) objectivity of academia. But for those who think there are equally strong advantages of polemics against and stigmatisation of truly evil views, I note simply that my thesis is consistent with them. Practical considerations aside, I doubt very much whether it is even possible to avoid thinking at least some evil thoughts. In fact, there is likely to be at least some rudimentary cogitation about virtually all our moral beliefs (even our belief that heinous beliefs are false requires at least some thought about them). And, plausibly, there are inevitably some cases where detailed thought about evil ideas is required: where important prima facie inconsistencies arise in our moral thinking, for example (see e.g. McMahan 2013). Thinking about evil opinions is inevitable, and we cannot piously avoid it. And it is plausibly safer to do so in a community where they can be corrected and shown for what they are. All these considerations – that we can be clear about a heinous view while discussing it, that moral rumination is inevitable, and so on – count finally against a final intuitive thought: that, if my thesis is correct, then whether or not we should do some evil act is ‘up for debate’. This is, indeed, an uncomfortable thought. But, in addition to the arguments I have given for my view, we might also give an alternative explication of this thought which helps to smoothen any inconsistency. One such interpretation might be to say that, once we have arrived at a conclusion16 regarding e.g. the dignity and rights of a certain category of humans, the issue is no longer ‘up for debate’ in the sense that we allow such humans to be mistreated or violated by others just in case those others disagree with us on their status. We could say that there is no ‘debate’ when it comes to how to treat them, given their humanity. We do not allow court defences based on their lack of humanity, nor do we suddenly revoke their rights if we have transient or shallow doubts about their status. So there are credible interpretations of the intuition which are consistent with genuine epistemic uncertainty about an ostensible person’s rights and their grounding. In any case, it is difficult to see what alternative there is. All of us feel deeply, deeply disturbed by the contention of a certain demographic group’s humanity, and at the need to defend it. But given that some people will inevitably disagree – sometimes sincere agents, no less – disturbance is inevitable. The issue is how we respond to it. One possibility is to simply refuse to endorse their concerns with debate. This is terribly dangerous, especially if the movement gains ground and becomes more popular. And we cannot get away with saying that such views never would become popular: they have done so all too recently and devastatingly. The alternative is to engage in the uncomfortable debate. As I suggest, this might require the inclusion of polemics or stigma in our academic discourse. This is perhaps not ideal either: but if it is what is required to defend the humanity of those in doubt, it might well be warranted. Conclusion These considerations may not be sufficient to justify unrestricted academic freedom, but they do constitute a good reason to significantly expand the range of views we are willing to entertain.17 And supposing we find no reasonable threshold of plausibility with which to adjudicate, or supposing that almost any view may be plausible to at least one sincere agent, perhaps the most promising solution is to extend academic freedom unrestrictedly. The most sensible and sensitive way to do this is to undermine the implication that discussing a view tacitly assumes its decency. I suggest that my initial considerations serve this purpose as well. In short, a view’s being appalling is not good reason to think it is also extremely improbable. Yet contemporary discourse – probably from all political vertices – by its attitude towards those with ostensibly contemptible views, and especially its failure to interact with those views, does not often faithfully represent the fact that some abhorrent views are plausible. My argument therefore recommends informal sympathy and a hearing for sincere agents expressing contemptible views, and moreover it recommends that such hearings extend to the academy. By giving plausible alternative interpretations of common intuitions and slogans 16 Recall that in at least some cases there will inevitably be some reflection or inference, so one cannot respond to this by suggesting that we do not need to ‘conclude’ that certain of our core and important moral beliefs are true. It is simply a fact – whether comfortable or not – that some of our most important beliefs about human dignity are genuinely concluded rather than e.g. directly perceived or intuited. 17 I leave open for now the question of the extent to which other factors might be important: for example, the otherwise creditable nature of a disputant’s work or career. My thanks to Donald Downs for drawing my attention to the importance of this point. against academic free speech, one can hold the spirit of these consistently with my proposals, even if this requires some modification of academic discourse. These debates are uncomfortable, no doubt. And I do not hereby intend to credit the depraved – morally and intellectually – views which surely exist in the academy. But I do hope to have shown that despite our discomfort and despite their depravity, debate simply cannot be avoided on consequential or principled grounds. So let us have the debates and subject our core moral beliefs, along with depraved ones, to such academic tests. If we are indeed right, we can be confident that we will pass them.18 References Anscombe, G.E.M. 1958. “Modern Moral Philosophy.” Philosophy 33 (124): 1-19. doi:10.1017/S003181900037943. British Broadcasting Corporation. 2016. “Ched Evans: Footballer Found Not Guilty of Hotel Rape.” BBC News, October 14. http://www.bbc.co.uk/news/uk-wales-37659009 British Broadcasting Corporation. 2017. “Bath Spa University ‘Blocks Transgender Research.’” BBC News, September 25. http://www.bbc.co.uk/news/uk-41384473 Buchak, Lara. 2014. “Belief, Credence, and Norms.” Philosophical Studies 169 (2): 285-311. doi:10.1007/s11098-013-0182-y Dworkin, Ronald. 1993. Life’s Dominion: An Argument about Abortion and Euthanasia. London: Harper Collins. George, Robert, P., and Cornel West. 2017. “Truth Seeking, Democracy, and Freedom of Thought and Expression – A Statement by Robert P. George and Cornel West.” Princeton University website, March 14. https://jmp.princeton.edu/statement Giubilini, Alberto, and Francesca Minerva. 2013. “After-Birth Abortion: Why Should the Baby Live?” Journal of Medical Ethics 39 (5): 261-263. doi:10.1136/medethics-2011-100411. Gray, Jasmin. 2016. “University of Strathclyde Student Union Bans Pro-Life Group For Violating ‘Safe Space.’” Huffington Post, December 5. http://www.huffingtonpost.co.uk/entry/university-of-strathclyde-student-union-bans-pro-lifegroup-for-violating-students-safe-space_uk_5845420ce4b061fb97e5eddd Greasley, Kate. 2017. Arguments about Abortion: Personhood, Morality and Law. Oxford: Oxford University Press. Grierson, Jamie. 2017. “Christian thrown out of university over anti-gay remarks loses appeal.” October 27. https://www.theguardian.com/uk-news/2017/oct/27/christian-felix-ngole-thrownout-sheffield-university-anti-gay-remarks-loses-appeal 18 I am grateful to George Tarr, Donald Downs, and Joshua Blanchard for their detailed and helpful comments on this paper. Joseph, Rebecca. 2017. “Wilfrid Laurier admits it mishandled Lindsay Shepherd academic freedom case.” Global News, December 18. https://globalnews.ca/news/3923478/wilfridlaurier-no-complaint-lindsay-shepherd/ McMahan, Jeff. 2002. The Ethics of Killing: Problems at the Margins of Life. Oxford: Oxford University Press. McMahan, Jeff. 2007. “Infanticide.” doi:10.1017/S0953820807002440. Utilitas 19 (2): 131-159. McMahan, Jeff. 2013. “Infanticide and Moral Consistency.” Journal of Medical Ethics 39 (5): 273-280. doi:10.1136/medethics-2012-100988. McMahan, Jeff. 2016. “Philosophical Critiques of Effective Altruism.” Effective Altruism Forum, May 3. http://effectivealtruism.com/ea/x4/philosophical_critiques_of_effective_altruism_by/ Mill, John Stuart. 1859. On Liberty. 2nd ed. London: John W. Parker and Son, West Strand. Mulhall, Stephen. 2002. “Fearful Thoughts.” London Review of Books 24 (16): 16-18. Oscroft, Alex. 2017. “WomCam protesters try to shut down anti-abortion talk.” Oxford Student, November 2. http://oxfordstudent.com/2017/11/02/womcam-protesters-try-shut-anti-abortiontalk/ Palmer, Ewan. 2015. “‘Transphobic’ feminist Julie Bindel banned from Manchester student union talk on free speech.” International Business Times, October 8. http://www.ibtimes.co.uk/feminist-julie-bindel-banned-manchester-student-union-talk-freespeech-1522670 Rayner, Gordon, and Bill Gardner. 2015. “Men Must Prove a Woman Said ‘Yes’ Under Tough New Rape Rules.” The Telegraph, January 28. http://www.telegraph.co.uk/news/uknews/lawand-order/11375667/Men-must-prove-a-woman-said-Yes-under-tough-new-rape-rules.html Real Clear Politics. 2015. “Hillary to Sexual Assault Victims: ‘You Have The Right To Be Heard, You Have The Right To Be Believed.’” Real Clear Politics, September 14. https://www.realclearpolitics.com/video/2015/09/14/hillary_clinton_women_should_be_belie ved_when_they_claim_rape_have_to_increase_prevention.html Singer, Peter. 1993. Practical Ethics. 2nd ed. Cambridge: Cambridge University Press. Tooley, Michael. 1972. “Abortion and Infanticide.” Philosophy & Public Affairs 2 (1): 37-65. Weinberg, Justin. 2017. “Philosopher’s Article on Transracialism Sparks Controversy (Updated with response from author).” Daily Nous, May 1. http://dailynous.com/2017/05/01/philosophers-article-transracialism-sparks-controversy/ Weiss, Bari. 2017. “When the Left Turns on Its Own.” The New York Times, June 1. https://www.nytimes.com/2017/06/01/opinion/when-the-left-turns-on-its-own.html