I. Silencing and Harm

As already indicated, throughout this essay I shall consider silencing to be the prevention of open inquiry. To clarify, silencing is an act by one or more individuals, by a corporate entity, or by norms,Footnote ² that cause one or more other individuals to refrain from engaging in discussion of some topic.Footnote ³ It need not actually stop conversation as other topics may be discussed—perhaps more superficial topics—when people refrain from discussing a topic of concern. Silencing can be intentional, but need not be. It can also be wrongful, but need not be.Footnote ⁴

When a teacher tells students that if they speak they will fail her class, she intentionally silences their discussion. This may be useful as a means of promoting their education but it silences them nonetheless. It may also not be wrongful, and so may not be harmful (as I will explain shortly.) Full silence—complete lack of talking—is not our concern here, though it too may be a problem. We are concerned with cases wherein some specific topic is removed from discussion.Footnote ⁵ This, too, may be rightful in some cases. In a college classroom, for example, we expect that a math professor teaching a trigonometry course will disallow discussion of popular cultural portrayals of romance. This also shows that silencing need not be intentional. In most college math classes, after all, no one attempts to introduce discussion of cultural portrayals of romance and so no one need tell them they should not.

In some cases, of course, it will be wrongful to intentionally silence students. For example, when a professor of a political philosophy class about justice tells students that although the material they will discuss may cause them to think of Marxism or libertarianism, they cannot bring Marxism or libertarianism into the discussion as its been definitively disproven or is otherwise of no value. (Or imagine a professor telling a student that their input is unwelcome as they are stupid or have nothing worth saying—or standing by as one student says this to another.)

That a particular instance of silencing is not intentional also need not mean it involves no wrong. A professor might unwittingly encourage norms in his classroom that hinder learning; this professor’s failure is not rendered innocent by his lack of attention to good pedagogy. For an obvious example, a professor that only calls on male students in class may leave some views unheard while preventing female students from engaging with the material. Even if completely unintentional, this is blameworthy.

What interests us here is when silencing is harmful. For silencing to be harmful, it may be intentional or not, but it must be wrongful.Footnote ⁶ To clarify: harms, for our purposes, are wrongful setbacks to interests—not mere hurts, which are setbacks to interests but not wrongful.Footnote ⁷ Jack’s tripping on a stone in his path and thus skinning his knee is, in the ordinary case, a hurt, but not a harm. While Jack surely has an interest in not skinning his knee, if we assume—what is reasonable—that no one wronged Jack and the stone just happened to be in the wrong place causing him to fall, with no one at fault, there is no wrongfulness and thus no harm understood as a wrongful setback to interests.

If we are concerned, as I am, with the sorts of limits that are used to justify interference, we need a principle that indicates when such limits are proper and when they are not.Footnote ⁸ The principle we are considering here—and the reason for our concern with harms—is the harm principle, originally put as follows by John Stuart Mill:

I won’t directly defend the harm principle here.Footnote ¹⁰ I assume it is clear: absent the credible threat or presence of harm, there is no justification for interference. Interference is (pro tanto) permissible to prevent or rectify harm. This essay is intended as something of a test of this principle. If we think its verdict is correct in the sorts of cases discussed here, we have some reason to endorse it; if we don’t, we have some reason to reject it.

Harms are the sorts of events that often warrant interference or change. Where a harm is credibly threatened or present, we consider whether interference should follow.Footnote ¹¹ If a supposed harm sets back no interests or does so, but does not do so wrongfully, we are unlikely to think there is warrant for interference or change on these grounds.Footnote ¹² So, return to the case of Jack tripping over a rock where no one wronged him. With no wrong, there is no harm and, according to the harm principle, with no harm, there is no reason for interference. Put simply, we don’t think Jack has any justified complaint against anyone with whom we could interfere. By contrast, if you take Jill’s laptop computer without her permission and claim it as your own, you wrongfully set back her interests in the computer, she is justified in complaining, and interference with you to rectify the situation is warranted. Similarly, if Albert stabs Charles in order to satisfy his (Albert’s) desire to see blood, Charles is justified in complaining, and interference with Albert is warranted to rectify the situation. Jack is hurt but not harmed. Jill and Charles are both harmed. According to the harm principle, interference is pro tanto warranted in the latter cases but not the former.

While I won’t be discussing the sorts of changes or interferences that might be warranted when a harm is credibly threatened or present, my interest in both freedom of religion and freedom of speech—and freedom more generally—is primarily about this normative issue. Put differently, it is the normative limits of those (and other) freedoms and, importantly, the normative limits of actual imposed limits to those freedoms, that I am concerned with. That is, my overarching concern is with when a particular sort of religious activity, speech, or what have you, may be permissibly limited and when attempts to limit those may themselves be limited. A teacher wrongly limiting his students should be limited (whether retrained or relieved of his duties), a university should adopt policies that do not allow for wrongful limits to speech or religious activity,Footnote ¹³ and a state should be changed if it wrongly limits religious activity or discussion of some specific topic.

Harms, as they must be understood when discussing the harm principle, are events that are wrongful setbacks to interests. So what are the harms of silence? In particular: What harms of silence are involved when a religious group and its practices are otherwise tolerated, but must be kept private? And what are the harms of directly silencing speakers?

In the next two sections, I make clear that there are real harms of silencing. I first discuss those that result from religious toleration that is not as extensive as it should be and then discuss the sorts of silencing that sometimes happens on college campuses. Some of each of these may seem minor. Others will not. All should make us consider whether the silencing in question is morally permissible. If we take the harm principle seriously any silencing that causes a harm is at least subject to further discussion. Absent overriding reason to allow such silencing, it should be prevented.

II. Harms of Silence With Regard to Religious Toleration

Early modern thinkers defended freedom of religion—though with more limits than we would endorse today. Some thought all religious views except atheism or “Papism” should be tolerated. Even those willing to tolerate all religious views sometimes limited the extent of toleration. Some, for example, thought religious groups could be differentially taxed while all were tolerated. Others thought that a religion could be tolerated while its practitioners were not allowed to publicly display their religiosity. Pierre Bayle, for example, wrote:

Bayle is a hero to many who write in defense of religious toleration and toleration more broadly.Footnote ¹⁵ Yet, on his view, the state might have an official religion whose adherents would be free to “walk in Processions thro the streets” and keep “publick Churches” while adherents of other religions would not. Strictly speaking, it might be true that in such a state all religions are tolerated even if public displays of adherence to a variety of religions are not, but those of us committed to freedom more generally would question why those public displays should be forbidden. Certainly, they involve no necessary harms, so would not be subject to justifiable interference according to the harm principle. Moreover, when burdens are differentially placed on religions, one would expect that those religions more heavily burdened would lose adherents while those less burdened would gain adherents.Footnote ¹⁶

Even if the sort of differential toleration Bayle accepted did not itself do direct harm—given the available alternatives of his time (death, imprisonment, and other forms of persecution for adherents of religions not endorsed by the state), it was certainly a kinder, gentler policy than the alternatives—we should not find it acceptable. Promoting one religion (paying for its churches, allowing it public displays and processions, and so on) while only allowing another to be practiced in private may be the quintessential state of affairs that motivates the claim that toleration is insulting to the tolerated—in this case, a failure to recognize that there is no reason to give less respect to one religion than another (promoting one while merely tolerating the other). In such circumstances, those tolerated would rightly say “we do not accept mere toleration; you owe us respect and must fully accept our presence!” Those with the power to tolerate, after all, have the power (and, it seems, reason) to interfere.Footnote ¹⁷

Those of us who believe the ideal liberal state is a state that has maximal toleration (limited only by a narrowly restricted set of principles) have a ready answer to those who claim that when we say the state must tolerate X, this assumes X is unfairly subject to the will of the state. We insist that where the state must tolerate X, it must tolerate all things of X’s type and that, as such, X is no more (or less) subject to the will of the state than other things of its type. If Catholicism is tolerated, so too is Protestantism, Judaism, Islam, Buddhism, Hinduism, and so on. The properly liberal state tolerates equally. It is a neutral state.Footnote ¹⁸ Part of that neutrality, to be clear, is a requirement to allow speech equally to all individuals, no matter their religion or cultural group—and no matter where or how the speech is conducted.Footnote ¹⁹

In the twenty-first century these issues are largely settled. The liberal states of the West might have official religions, but (perhaps other than special tax privileges, which should be rejected), adherents of other religions receive equal treatment. No religious group is condemned to the shadows. Problems of this sort, though, persist.Footnote ²⁰ For one quick example, in May of 2018, a man was fined 210 Swiss francs for greeting a friend in public by declaring “Allahu akbar.” Officials were worried this might have caused others to become “afraid or shocked.”Footnote ²¹ Being Moslem is tolerated in Switzerland, but speaking Arabic too loudly is apparently not. Silence seems preferred.

Pierre Bayle’s brother was arrested by the French Catholic authorities because of Bayle’s “heretical” writings and died in prison. The French at the time (partly) tolerated Protestants, but Bayle had converted to Catholicism and then returned to his earlier Protestant faith. For this, he drew the scorn of the Catholic authorities (as well as important Protestant theologians). Bayle himself had to leave France. There can be no doubt that the French government harmed Bayle’s family, especially his brother.Footnote ²² There is little reason to go through the many other sorts of harms that have occurred due to refusal to tolerate religious difference. The history here is clear. Augustine recanted his defense of toleration, after all, when he saw the Donutest heresy (which he despised) disappearing—as Donutests were killed or unable to transmit their views to others for fear of being killed. The effect of such persecution is to silence minority religions. That silencing is itself a harm—the wrongful setting back of people’s interests in discussing their religion—separate from the harm of imprisonment or death. It is a harm of silence.

Bayle’s crime wasn’t (or wasn’t solely) that he was public about his non-endorsed religion, but that doesn’t make the point less clear. Being able to practice one’s non-Catholic Christianity, one’s Judaism, or one’s Islam (and so forth) in private but not in public has a similar effect. Indeed, even if the punishment for failure to abide by the law is relatively minor—say having to pay a fine rather than go to jail—the effect is silencing. People are less free to discuss the merits of their religion as opposed to the merits of another religion—especially that endorsed by the government. When this is the case, people are less likely to learn about competing religions, so less likely to decide to choose a different religion or even become familiar with the beliefs of others. Of course, there could be discussion in private, but this leaves everyone with limited numbers of people with whom to discuss religion. They can discuss religion with friends and family, but are unlikely to have friends and family who outright reject their religious views or who accept radically different views. To encourage discussion with people who reject or depart widely from one’s religious views requires being able to discuss it publicly—to invite discussion with people previously unknown to oneself. Such discussion is of value and many (arguably all) of us have interests in its occurring—both in speaking and in hearing about different religions. Absent a good justification, these interests are wrongfully set back if government or others do not tolerate public discussion of religion.Footnote ²³ The failure to tolerate, in other words, results in harms of silence.

III. Harms of Silence on College Campuses

On college campuses in the United States and elsewhere, student groups have sought to “de-platform” speakers with whom (they think) they disagree.Footnote ²⁴ Some of these speakers, admittedly, should not be invited to college campuses in the first place. Milo Yiannopoulos, a political commentator that worked at Breitbart News with no reputation as an academic, comes to mind. While inviting him to a campus is likely a mistake—as it would provide little opportunity for serious learning—uninviting or de-platforming him causes other problems. It can, for example, easily give conservative students the impression that their voice is not respected. (A better solution: explain to students in advance why an invitation to such a speaker is not worthwhile and have them consider alternative speakers whose presence would be.)

Consistent with the harm principle, my view is that we should each be generally free to express ourselves as we wish unless our doing so is, in that instance, a harm. (This does not require the provision of a forum, but does require not interfering with available fora.) Silencing is otherwise prima facia wrong. Yet, in recent years there have been attempts on college campuses—perhaps the most important locations of what should be unfettered discussion—to silence some who would participate in that discussion.

Perhaps the clearest case of attempts to silence academic participants is the de-platforming of Charles Murray at Middlebury College in Vermont in early 2017 (a case in which a professor from that school was physically hurt enough to go to the hospital).Footnote ²⁵ Murray is known as a politically conservative thinker. Some academics defend Murray’s work and others think it indefensible, but no matter one’s assessment of its merits, it is academic work—published in academic outlets with peer review. A talk on campus by Murray would not be worthless. He was invited by a student group to speak at the campus. The group—The American Enterprise Club—presumably hoped that his presence would spur honest discussion. Instead, opposed students acted in ways that would embarrass faculty and administrators at any college.Footnote ²⁶ They were rude to the speakers before Murray and then, as Murray began speaking, stood, turned their backs to him, and spoke loudly in unison against him so that he could not be heard. He may have said things they would want to argue against. He may have said things they would agree with. They could not know. (If they simply did not want to know, of course, they could have stayed home.) This, it seems, shows a clear lack of respect for those who attended the talk to hear Murray, as well as those who issued the invitation—and the effects of both are hard to undo. Of course, others with more or less mainstream views have also faced de-platforming.Footnote ²⁷ De-platforming is obviously about silencing.

In some cases, speakers have declined invitations to speak at college campuses because of protests.Footnote ²⁸ In still other cases, speakers have had their invitations revoked.Footnote ²⁹ In all of these cases, open inquiry was prevented. The prevention was intentional and, too often, wrongful.Footnote ³⁰ There are many reasons to worry about these sorts of silencing activities. I have discussed some of them elsewhere.Footnote ³¹ Here, the point is only that there were some that had an interest—in most of these cases, a clear and stated interest—in engaging with the silenced speakers (who also have an interest in engaging with them). Absent a justification that overrides those interests and the other problems with silencing—significant at a collegeFootnote ³²—these silencing activities are harms.

Some might argue that “no platforming” is permissible because colleges are not meant to provide speakers about every topic. No university that I know of, for example, brings in speakers to discuss the relationship between Charles Schultz’s Peanuts and the weather. There is (I assume) no reason for such discourse. So too, though, there is likely no reason to bring in speakers advocating a “flat earth” theory, a theory about how vaccines cause autism, or a theory about how the Nazis did not kill anyone during World War II. We can agree, though, that there is no reason to bring in such speakers and still think that “de-platforming” is unacceptable.

Note that I switched, in the last paragraph, from talking of “no platforming” to talking of “de-platforming.” The former may be the original term—and is used to indicate university policies against inviting speakers of certain sorts. Such policies are justified, presumably, on the simple basis that universities are not committed to teaching students all views, but only academic and plausible views. This is not a limit to free speech, we can agree, but an exercise of academic freedom. Academic freedom, in my view, involves recognizing that the professoriate serves as a gatekeeper, encouraging discussion of academic and plausible views even if they are fringe, disturbing, or opposed, but not encouraging discussion of other views—perhaps views by those with no accredited expertise. It is the faculty that should determine the sorts of discussions encouraged on college campuses; it is the mission of colleges to provide a broad forum for such discussion with policies that vest control in the faculty. Footnote ³³

Notice, though, that such policies are not at issue in the Charles Murray case or other de-platforming cases. In these, we have speakers who have been invited—usually by student groups—and given a platform but then made unable to use it. They have been de-platformed, not no platformed. Put differently, they have been invited and brought to the campus given the campus policies but then not allowed to dialogue—by students, not faculty or administration.Footnote ³⁴

Just as journal editors and referees serve as gatekeepers, preventing (usually) the publishing of low quality work in academic journals, professors (should) serve as the gatekeepers on campus. They not only determine what is taught in the classroom—fully responsible for the content of their own courses, but also, via service on hiring committees and college curricula committees, responsible for what sorts of courses are offered at all—and also often determine what speakers are brought to campus. They may bring speakers for talks targeted primarily at faculty, but they may also bring speakers for talks aimed at students and the public. In that capacity, they seek to invite speakers—and set policies for others to invite speakers—whose talks would provide value according to the mission of the university (this need not be limited to those doing academic work).Footnote ³⁵ The important point here is that it is the professoriate, not the students, who have the qualifications to decide who should have a platform at the school.

Of course, student groups may be given the right to invite speakers, but that right is one that should be granted by the faculty of the college and subject to monitoring and revocation by that faculty. In the current state of academia, it is more often administrators that decide who can be brought to campus and which student groups can invite speakers. Often, today, student groups have much wider latitude (in policy) to invite speakers than I advocate.Footnote ³⁶ On my view, partially because faculty uninviting speakers would create problems akin to those of de-platforming, the monitoring should be significant and applied in such a way as to avoid the invitation of any speakers the faculty would deem unacceptable.

Some may think that this insistence on the professoriate having the final control over who can be invited (platformed) to speak on a campus is illiberal—that having anyone able to censor students by limiting who they invite is a problem for those of us committed to liberalism. Those with this worry, of course, should have the same problem with administrators or student government leaders in such a gatekeeping role. They would likely suggest, though, that neither administrators nor student government leaders act in the way I am suggesting faculty ought to act. More reasonably, it might be thought that neither administrators nor student government leaders would interfere with student invitations. Granting that they could, even if they often do not, interfere in troubling ways, it is immediately clear that this worry is somewhat limited in that there is little reason to think faculty would abuse this power more than administrators or student government leaders.

The worry is not, of course, dispelled by merely indicating that others could act in the same worrisome way that concerns our interlocutor. The worry, in fact, may seem similar to an important and long standing theoretical concern: the paradox of toleration is, put simply, that if we are committed to toleration, we may need to tolerate non-tolerating behavior by others. The interlocutor here might suggest that we should tolerate student activities even if they involve non-tolerating behavior (or intolerant speakers). My view, by contrast, suggests there is some such behavior that we ought not tolerate—student invitations to Milo Yiannopoulos, for example.

A more apt consideration is the paradox of liberalism: as liberals, we believe our states ought to be committed to extensive toleration but might wonder whether that requires that we tolerate those committed to overthrowing that same system of extensive toleration. The paradox, in my view,Footnote ³⁷ is answered in that if we endorse a particular principle about the limits of power, that principle will indicate what sorts of behavior we do not have to tolerate. If we accept the harm principle, as I do, then we can interfere with behavior that is harmful. We can (and should), for example, tolerate talk of overthrowing the system of extensive toleration, but not actions that harm others, including such actions that are part of a plan to overthrow the system. The air of paradox is dispelled by the principle because remaining in accord with the principle is necessary for the institution to remain. Of course, the case of a college is different than the case of a liberal state. (Though colleges too should be concerned to limit harm and so should endorse the harm principle, at least as a disjunct of their policy.) Still, any institution will have policies against those seeking the destruction of the institution. The Catholic Church, for example, reasonably refuses to tolerate priests that advocate atheism.

The principle or policy that indicates what sorts of behavior to tolerate or not on a college campus will depend on the college’s goals. It is perfectly reasonable, for example, for a small Baptist College to have a mission that is narrower and different than a small secular liberal arts college and for both to have different missions than that of a research university. (Colleges can brand themselves.) I would suggest, though, that each should be transparent about its mission. I would not help my child attend a college that promised to have no speakers that were experts in evolution, but others might. I would also not help my child attend a college that promised to have no discussions of theology, but again, others might. It is important that colleges adhere to a requirement that they be truthful about what they are and what they permit on their campuses. I believe the best colleges not only allow, but provoke open inquiry. In any case, if a college is genuinely committed to open and honest inquiry into all domains of the liberal arts and sciences, the faculty should set policies for platforming that reflect that mission. The mission of the college, then—presumably set by the faculty, which it also helps to constitute (via hiring policies)—is what sets the limits to the platforming policies faculty ought to work with. If a speaker’s talk is likely to be inconsistent with a college’s mission, the faculty of that college have good reason not to provide her a platform. Such no platforming is thus justified and, so, not harmful. De-platforming someone who is platformed in accord with such a policy, on the other hand, is unjustified and thus potentially harmful—that is, likely to cause a harm of silence—if it sets back the interests of others. (At the sort of college I would prefer, there would be few limits other than those imposed by limited resources—speakers are often paid after all, and come with other costs—and the need not to hinder open and honest inquiry conducive to learning.)Footnote ³⁸

So again, on the view defended here, one role of the professoriate is to serve as gatekeepers to those who might speak on campus. This is important not only to protect the integrity of the institution but also because being given a platform provides others second-order evidence that the speaker’s talk is of value—that the speaker ought to be taken seriously.

There are many sorts of evidence about whether what a speaker says is true. First order evidence is evidence about the claim in question. Second order evidence, by contrast, is evidence one has for believing some other evidence about the claim. A speaker presents evidence in virtue of speaking. Whether that evidence is reliable is a second order question. If one has reason to believe the speaker is mentally incapacitated in some way, one has second order reason to disbelieve the first order evidence provided by the speech. A professoriate’s conferring a platform on a speaker provides (or is, in any case, often taken to provide) second order evidence that the talk is of value—that the speaker is to be trusted and taken seriously (though not necessarily to conclude that the speaker is correct).Footnote ³⁹

Faculty ought not give a platform (or let students give a platform) to an anti-vaccination crusader whose credentials include nothing certifying any expertise in the biological sciences. (The faculty may want to talk to such a person themselves, but giving them a platform might lead students and others to take the view more seriously than they ought.) They ought not give a platform to someone claiming aliens began invading the earth in the 1930’s and that they know this (only) because they were abducted by aliens (absent, I suppose, further evidence of that abduction). They might, on the other hand, want to give a platform to Avi Loeb, Chair of the Department of Astronomy at Harvard University and Chair of the Board on Physics and Astronomy of the National Academies, who takes seriously the idea that the asteroid Oumuamua might be an alien spacecraft. Loeb’s discussions of such things would be of value to a university, though the supposed abductee’s would (likely) not.

Students, I submit, are not in the same sort of position as faculty to know whether speakers are worthy of their attention. They are not in position, that is, to decide, absent any oversight, who should receive a platform, nor to overrule other students, faculty, or administrators in determining when a speaker is acceptable. Their de-platforming a speaker is unacceptable even if a university or its faculty can acceptably no platform others. Students are not in a suitable position to decide that their colleagues should not be able to listen to Loeb or, to return to an earlier example, Charles Murray. They can, of course, refuse to listen themselves. Claiming no one ought to be exposed to a speaker the faculty approved, however, disrespects the faculty and fails to show the humility appropriate to their status as students.

We should not overstate the concern; de-platforming does not occur that frequently. Students at many colleges cannot afford the time to engage in the sort of behavior those at Middlebury can. Many have to balance childcare and employment while simultaneously doing school work and so cannot get involved in extracurricular campus activities, including those involving de-platforming speakers. Nonetheless, de-platforming does happen and is not the only sort of silencing that takes place in the hallowed halls of higher education. I discussed other sorts of cases in Section I above—cases involving how professors sometimes silence students. There are also cases involving faculty being silenced on their own campuses.

There are cases where one or more members of an academic department or college actively seek to prevent one or more other members from pursuing certain intellectual pursuits. This might be a matter of trying to prevent the acceptance of funding for an academic project from a particular source. That funding might help a student group that wishes to bring in speakers of a particular sort, it might help the faculty member travel to give talks on a particular topic or take the time needed to finish writing a paper or book, or it might be funding to hire academics working in a specific area or to start or promote an academic program. In these and other sorts of cases, faculty who hold minority views may well get the impression that their voice is not respected by their colleagues even while they continue to publish in respected venues.Footnote ⁴⁰ Some, of course, will go on doing what they do regardless of the condemnation. Others, though, may leave the profession, may choose to leave the college (or find a way to switch departments), or simply alter their research and teaching so as appear not to have minority views.Footnote ⁴¹ All of these cases involve silencing—and at least some are simply unjustifiable instances of setting back the interests of faculty with minority viewpoints (and those genuinely committed to open inquiry).Footnote ⁴² They are, that is, harms of silence.