Liberal Education

The Contours of Free Expression on Campus: Free Speech, Academic Freedom, and Civility

A tension exists on college and university campuses across America today concerning how to pursue liberal, rational, open learning and, at the same time, celebrate a spirit of academic community—in short, how to exercise free expression and maintain civility.

In exploring the contours of free expression on campuses, I begin with an exploration of the boundaries of free speech, especially in the troubling context of hate speech. This boundary must be expansive; difficult, challenging, and, yes, even hateful speech ought to be protected under our system of free expression. This view, largely consonant with accepted First Amendment doctrine, applies to campuses generally, both public and private. This is not a “First Amendment analysis” per se. Rather, our concern is the nature and limits of expression on all campuses.

I will address two issues. First, where is the limit on expression? Where does protected, hateful speech cross over into being a prohibited hate crime? On campuses, this question is not typically about criminal behavior per se. But the question is the same: when does behavior cross over from being protected, however hateful, and become the proper subject of disciplinary action or even expulsion?

The second issue concerns the proper response to hateful speech that is, in fact, protected. To say that it is protected is a threshold issue, not the end of the discussion. This will bring us to the compelling topic with which America generally, and academia particularly, is preoccupied today: the relationship between free expression and civility in the public square.

Two stories to set the context

I begin with two stories to set the context. The first story took place at Williams College, where I was a trustee. A Jewish student complained that a faux eviction notice had been placed on her dorm room door: “If you do not vacate the premises by tomorrow at 6PM, we reserve the right to demolish your premises without delay. We cannot be held responsible for property or persons remaining inside. Charges for demolition will be applied to your student account.” The student understandably felt terrible. The president wanted my opinion on what should be done to those responsible.

The second story occurred on the Brandeis University campus in December 2014, when I was president of the university. It occurred right after the murder of two police officers in “revenge” for the deaths of Eric Garner and Michael Brown. A prominent student member of the campus Black Lives Matter movement tweeted that she had no sympathy with the police officers. Knowing this student, I believe that what she meant was that she was deeply frustrated and troubled that, in her view, vastly more attention had been paid to the deaths of Officers Wenjian Liu and Raphael Ramos in the broader community than was given to the deaths of Garner and Brown. But alas, that is not what she said. And, with “help,” if that is the right word, from one of the sixty or so students who received the tweet and who posted it on what can best be described as an extremist website, her tweet went viral. Not unexpectedly, I received enormous pressure from all sides on this set of events. Some urged that the student be thrown out of school or, at least, lose her financial aid package. Others argued that I should issue a short statement supporting free speech and the right of all members of the community to say what they wish. I will return to the Williams and Brandeis stories later in my remarks.

As we begin, it should be noted that the discussion is fundamentally different from what it would be like in most, if not all, other advanced democracies, which punish pure hate speech. Consider the following definitions of punishable speech excerpted from the statutes of other nations:

  • statements “by which a group of people are threatened, derided or degraded because of their race, colour of skin, national or ethnic background” (Denmark)
  • attacks on “the human dignity of others by insulting, maliciously maligning or defaming segments of the population” (Germany)
  • “threatening, abusive or insulting words, or behavior” intended to “stir up racial hatred” or when “having regard to all the circumstances racial hatred is likely to be stirred up thereby” (the United Kingdom)

Why is this so different for us in the United States?

The context of hate speech and hate crimes is at the intersection of three sets of significant individual and societal rights and interests: (1) freedom of expression, (2) personal safety, and (3) personal dignity. How do we define something as amorphous as “personal dignity”? Consider the concept as developed by Jeremy Waldron in his important book, The Harm in Hate Speech. For Waldron, dignity is concerned with a person’s basic social standing and the interest in being recognized as “proper objects of society’s protection and concern.”1 If the right to one’s safety is inherently individualistic and about liberty, the right to one’s dignity is inherently comparative and about equality—to have one’s dignity respected is to be accorded the same basic social standing as any other member of the society. As Lyndon Johnson is said to have answered a question concerning the moral necessity for the Civil Rights Act of 1964, “A man has the right not be insulted in front of his children.”

Free expression as a core value, which extends to most hate speech

Analysis of hate speech and hate crimes is concerned with legitimate and significant rights on all sides, including the rights of both the speaker and the listener. We must proceed with great caution, protecting rights where we can and limiting rights only where we must.

Campuses are replete with competing interests. Colleges and universities cover a wide range of models and identities. But I believe that most, if not all, schools share a similar mission—to discover and create knowledge, and to transmit that knowledge through teaching and scholarship for the betterment of our local, national, and even international communities. For this mission, free expression and free inquiry are essential.

I thus start from the position that all speech, including hateful speech, is presumed to be protected. By hateful speech I mean that which offends or insults a group along racial, ethnic, national, religious, gender, or sexual identity lines. The definition of the German statute puts it well—attacks on “the human dignity of others by insulting, maliciously maligning or defaming segments of the population.”

I ally myself here with the arguments presented by such scholars as the late Professor Edwin Baker and Dean Robert Post. Baker based his understanding of free expression on a fundamental concept of autonomy. In his essay “Autonomy and Hate Speech,” he wrote, “Law’s purposeful restrictions on [the speaker’s] racist or hate speech violate [that person’s] formal autonomy.”2 Post, in Constitutional Domains and elsewhere, has recognized the harm inflicted by hate speech, but also argued persuasively that the fundamental societal interests of public discourse will almost always outweigh this harm. In America, Post believes “public discourse is an arena for the competition of many distinct communities, each trying to capture the law to impose its own particular norms.” He adds that public discourse in our democracy thus has the “extraordinarily difficult task of ensuring democratic legitimacy in a climate of comparatively severe suspicion and distrust.”3

The right to hold opinions that are offensive to many or most

The normative arguments of Post, Baker, and others find deep resonance in American free expression jurisprudence, which largely protects hate speech. This begins with the underlying premise that a state may not punish a person for holding an opinion, regardless of how obnoxious the opinion may be to the public or even how good a predictor it might be for future anti-social conduct. It is striking that in 1951, Chief Justice Fred Vinson, not a strong advocate of a robust view of the First Amendment, saw no need to provide any support for his assertion that “one may not be imprisoned or executed because he holds particular beliefs.”4

Consider the context of flag burning, which continues to press the limits of the right to express unpopular views. The Supreme Court, even as it has become more conservative over the past three to four decades in its approach to numerous areas of the law, has repeatedly upheld the right to burn an American flag. In 1989, in Texas v. Johnson, striking down the Texas flag burning prohibition, the court held that “if there is a bedrock principle underlying the First Amendment, it is that the government may not prohibit the expression of an idea simply because society finds the idea itself offensive or disagreeable.”5

Similarly, hate speech has generally been held to be constitutionally protected. Beginning in the 1980s, many colleges and universities, concerned over the increase in racial tensions on campuses, adopted speech codes proscribing the expression of bigotry. None of these codes has survived First Amendment challenge. Campus speech codes at public universities have been viewed as prohibitions of speech based solely on the content of that speech. Although sympathetic with the goals of the campus speech codes, the district courts that struck down such regulations as those adopted by the University of Michigan and the University of Wisconsin did so on First Amendment grounds.6

This broad protection of speech on campus, both under the First Amendment and under basic principles of free expression and free inquiry as integral to the academic mission, still permits universities to protect students from being threatened and to protect classes from being disrupted. Where is the line to be drawn?

The flawed “speech vs. conduct” distinction

It is tempting to draw the line as a distinction between speech and conduct: speech is protected, whereas conduct may be regulated or prohibited. This is the distinction that a unanimous Supreme Court relied upon in Wisconsin v. Mitchell, the 1993 case in which the court upheld a Wisconsin bias crime law, the constitutionality of which had been challenged on the grounds that it punished thought or expression. The court held that bias crimes are conduct and may be punished, whereas hate speech is expression and is protected.

The speech-conduct distinction is tempting because it promises a predictable and logical way to draw lines: once we can differentiate speech from conduct, we can effectively protect the former and regulate or even punish the latter. The promise, however, is ephemeral because the speech-conduct dichotomy is far too brittle to work.

Speech and conduct are not merely intermingled; they are inextricable. Thus, the dialectic encompassing speech and conduct precludes not only a neat separation of the two, but also even efforts to determine whether “act” or “expression” is the predominant element in certain behavior. Consider two examples: flag burning, which, as we have already briefly discussed, is constitutionally protected, and draft card burning, which the Supreme Court has held may be punished.7 The court considered the burning of a flag to be expression, whereas the burning of a draft card is conduct.

The slipperiness of the speech-conduct distinction is apparent. Flag burning is surely an expression of political views, but is it not also an act? And what is the conduct in burning a draft card? The conduct of burning? It is at least plausible that, both in terms of the actor’s own understanding of the card burning and in terms of the state’s concern with punishing this behavior, the “conduct” of no longer having a draft card predominates in the act. As Professor John Hart Ely wrote in his classic article on the draft card–burning case, “burning a draft card to express opposition to the draft is an undifferentiated whole, 100 percent action and 100 percent expression. It involves no conduct that is not at the same time communication, and no communication that does not result from conduct.”8 We could say the same of flag burning. And yet one is protected, and one is not.

The point here is that the purported distinction between speech and conduct will not add rigor to any attempt to distinguish protected from proscribable behavior. The flying of a swastika flag or a Confederate flag from one’s dorm room or home cannot be objectively described as expression alone. It is action as well. Distinguishing between conduct and expression ultimately assumes its own conclusions. That which we wish to punish, we will term “conduct” with expressive value; that which we wish to protect, we will call “expression” that requires conduct as its means of communication. The critical decision—which behavior may be punished, and which should be protected—is wholly extrinsic to this process. If a meaningful distinction exists, we must find it elsewhere.

Replacing speech-conduct with focus on the actor’s intent

My proposed distinction looks to basic criminal law doctrine, and the distinction between the criminal act (actus reus) and intent (mens rea). This distinction differs from the speech-conduct distinction because an “act” may include physical activity or verbal activity. Speaking itself is a kind of act. Our focus is on intent—the actor’s mens rea. Is the actor intending to cause harm to a particular victim, or is the actor intending to communicate views, however hateful or unpleasant? This is not to suggest that the speaker’s act of expressing himself or herself is purely deontological; expression has ramifications. Oliver Wendell Holmes Jr. said, “every idea is an incitement.”9 But the expression we should protect does not seek to cause injury.

Several examples illustrate the point. The first two, drawn from Virginia v. Black,10 involve a Virginia cross-burning statute that was struck down by the Supreme Court. The third example is the case I described earlier that occurred at Williams.

The Virginia cross-burning statute began by making it a crime for anyone, “with the intent of intimidating any person or group of persons, to burn, or cause to be burned, a cross on the property of another, a highway or other public place.” The court would have upheld that part of the law. Justice O’Connor wrote that “the First Amendment permits Virginia to outlaw cross burnings done with the intent to intimidate because burning a cross is a particularly virulent form of intimidation.”11 The Court struck down the statute because of what followed: “Any such burning of a cross shall be prima facie evidence of an intent to intimidate a person or group of persons.”12

Not every cross burning is intended to intimidate a victim, and the two cases before the court in Virginia v. Black made the point. These two cases represented the two poles of cross burnings—domestic terrorism and expression of white supremacy. In one case, Barry Black led a Ku Klux Klan rally on private property, after which a twenty-five- to thirty-foot cross was burned. In the other case, Richard Elliott and Jonathan O’Mara were prosecuted for attempting to burn a cross on the lawn of an African American, James Jubilee, who had recently moved next door. Elliott and O’Mara were trying to “get back” at Jubilee because, among other things, he had complained when they used their back yard as a firing range.13

The “prima facie evidence” clause of the cross-burning statute impermissibly blurred the lines between the two meanings of burning a cross. As Justice Souter wrote in his separate opinion, “its primary effect is to skew jury deliberations toward conviction in cases where the evidence of intent to intimidate is relatively weak and arguably consistent with a solely ideological reason for burning.”14 To be constitutional, the statute would have to require proof of an intent to intimidate; proof of a cross burning alone is insufficient.

We may now return to the case at Williams College involving the faux eviction notice that had been placed on a student’s dorm room, in imitation of the notices placed on Palestinian homes that are to be demolished by Israeli authorities due to the connection between residents and acts of terrorism. The college president asked what I thought should be done to those responsible for the notice. “This,” he said to me, “is not just speech. This is actual conduct. Can we sanction these students?”

We talked about Virginia v. Black and the role of intent. When he asked how we would know the student’s intent, I suggested looking into the way the notices were posted. Were only the leaders of a Jewish student organization targeted? For that matter, were only Jewish students targeted? As it turned out, every student in that dorm, regardless of affiliation, received one. That the complaining student honestly felt intimidated is not the issue. The issue is the actual intent of those who posted the notices—to intimidate and threaten individual Jewish students or to make a dramatic statement about their views concerning the Israel-Palestine conflict.

Verbal assaults

We may apply a similar analysis to cases of pure speech. Words alone can be harmful, and their use can even constitute a crime. Behavior designed to instill serious fear may be criminalized, and it does not matter whether it involves spoken words, physical conduct, or some combination of the two. Many states have some form of assault law that proscribes the creation of fear or terror in a victim.15 These laws—variously enacted as “menacing,” “intimidation,” and “threatening” statutes—may be violated through the defendant’s use of words alone.16 Reviewing courts have upheld various forms of verbal assault statutes, if sufficiently narrow in focus. For example, “intimidation” statutes that criminalize words used to coerce others through fear of serious harm are constitutional, so long as they clearly apply only when the words are purposely or knowingly used by the accused to produce fear and the threat is real.17

Thus, even pure speech may, in some cases, cross the line between protected expression and that which may be sanctioned or punished. But punishment is only appropriate, whether verbal or physical behavior is involved, when the purpose of the behavior is to instill fear of imminent serious harm. A racial epithet, when screamed at another student in a menacing manner, or a Confederate flag, when brandished on the lawn of a black student fraternity to terrorize them, is no longer protected expression; it has crossed over into that which may be punished by the university.

Responding to protected hateful speech

To say that we ought to protect most hateful verbal activity as free expression is not the end of the matter. How should we respond to hateful speech on campus?

Robert Hughes, writing about the controversial exhibition of photographs by Robert Mapplethorpe in the early 1990s in Cincinatti, Ohio, provides an instructive approach.18 Hughes observed that the questions concerning the photographs had become largely constitutionalized, focusing on whether, as a matter of a constitutional right, a museum may exhibit this work, or whether a city may, as Cincinnati did, shut down such an exhibition.19 Hughes wrote that the focus on questions of constitutional limits precluded the discussion of an arguably more important question: as a matter of art criticism and aesthetics, is this art any good?

The constitutional and jurisprudential questions that have occupied us thus far are critically important. But they are best seen as threshold issues, not as the ultimate societal issue. To address them, there must be a context for a moral response to constitutionally protected hate speech, just as there must be room for aesthetic questions as to the merits of constitutionally protected art. This is especially true of residential campuses, where the very mission of the institution includes building a community and preparing future citizens.

The required response to hateful speech is to describe it as such and to criticize it directly. Supreme Court Justice Louis D. Brandeis wrote in Whitney v. California that, except in those rare cases in which the harm from speech is real and imminent, the answer to harmful or hateful speech is not “enforced silence” but, rather, “more speech.”20

This allows us to return to the story I shared at the outset from my own campus at the time. Recall the student tweet of “no sympathy” with the murdered New York City police officers. Strictly speaking this is not hate speech, but the case remains relevant. I rejected the idea of expelling the student from the university, or of pursuing any student disciplinary charges or other sanctions, such as terminating financial aid. That would have been to engage in “enforced silence.” I believed her tweet to be protected speech. But I also believed that this was a case that called for more than a mere statement confirming her rights. In the same statement that defended her freedom of expression and her academic freedom, I added a criticism of my own, saying that, in my view, her comments were contrary to the highest values of the university and that I found them to be abhorrent.

Consider now the case that occurred at the University of Oklahoma in March 2015. Members of the Sigma Alpha Epsilon fraternity, on their way to a fraternity “Founders Day” event, engaged in horrific racist chanting that included the use of the n-word and celebrations of violence. Two student leaders of the fraternity were expelled from the university. I am highly sympathetic to the impulse for this expulsion and share in full the sentiment expressed by the university’s president, who said in a statement that he was “sickened” by the event. But I question the case for the expulsion. Had the context been different, had this occurred outside of a predominately African American fraternity house, for example, this could have been a case of verbal assault, warranting full punishment. But with the actors chanting on a chartered bus in the presence solely of their own members, this was an instance of protected hate speech—vulgar, disgraceful, and indeed sickening, but also protected. There was no intent to threaten or cause direct harm to anyone. The well-intended impulse to punish the leaders stems, in large part, from a correct sense that this behavior required the strongest possible condemnation, but also from an incorrect assessment of the possible responses.

We bind ourselves to an impoverished choice set if we believe that we can either punish speech or validate it. There is a middle position, expressed in Brandeis’s dictum of “more speech,” that allows us to respond without punishing. In the face of hate speech, the call for more speech is not merely an option; it is a professional or even moral obligation.

When I criticized the student tweet, and on one similar occasion when I protected but criticized a faculty listserv that included vulgar and disgusting language directed at, among others, my predecessor and the State of Israel, I was accused by some of creating a “chilling effect” on their right and ability to express themselves. My response will be surprising to some: not all “chilling effects” are bad. Some are cases of the type of enforced silence of which Justice Brandeis spoke; this is classically what we mean by a chilling effect, and these are pernicious and contrary to our system of free expression. But then there are those that are cases in which we influence each other for the good, when we are touched, as Lincoln said in his first inaugural address, “by the better angels of our nature.” We should seek that effect on each other. Having that kind of effect on each other, especially through the ways in which
we discuss and disagree, is at the heart of the academic enterprise.

Campus leaders must search for respectful ways to disagree, whether we debate and discuss in person or virtually. I would advance three principles for respectful disagreement:

  • We should look for common ground, even when we disagree, and articulate that common ground as part of the discussion.
  • We should assume the best in each other, and not suspect the motives of those with whom we disagree.
  • We should disagree without attacking each other personally—dispute, without delegitimizing.

Charles Black was a legendary figure in constitutional law at the Yale Law School. He was one of the architects of the legal arguments attacking segregation and an advocate of judicial activism. His colleague, the equally legendary Alexander Bickel, argued for judicial restraint. When Bickel passed away, Black wrote in tribute that they had “agreed in everything but our opinions.”21 It is as powerful a statement of respectful, even loving, disagreement as I know. If we have lost the ability to say to those with whom we disagree that we “agree in everything but our opinions,” we have lost something very precious and, perhaps, irreplaceable. But if we can strive to do so, we will be building the most important kind of community there is—and one worthy of the great shared mission America’s colleges and universities.


1. Jeremy Waldron, The Harm in Hate Speech (Cambridge, MA: Harvard University Press, 2012), 5.

2. C. Edwin Baker, “Autonomy and Hate Speech,” in Extreme Speech and Democracy, ed. Ivan Hare and James Weinstein (New York: Oxford University Press, 2009), 143.

3. Robert Post, “Hate Speech,” in Extreme Speech and Democracy, ed. Ivan Hare and James Weinstein (New York: Oxford University Press, 2009), 133, 137. See also Robert C. Post, Constitutional Domains: Democracy, Community, Management (Cambridge, MA: Harvard University Press, 1995).

4. American Communications Associate v. Douds, 339 U.S. 382, 408 (1950). This is the same Justice Vinson who applied the “clear and present danger” standard to permit the prosecution of leaders of the Communist Party in Dennis v. United States (341 U.S. 495 [1951] [Vinson, C. J., plurality opinion]).

5. Texas v. Johnson, 491 U.S. 397, 414 (1989).

6. See Doe. v. University of Michigan, 721 F. Supp. 852 (E. D. Mich. 1989); UWM Post, Inc. v. Board of Regents of the University of Wisconsin, 774 F. Supp. 1163, (E.D. Wis. 1991).

7. See United States v. O’Brien, 391 U.S. 367 (1968) (upholding conviction for burning of a draft card); Watts v. United States, 394 U.S. 705 (1969) (upholding facial validity of 18 U.S.C. §871, which criminalizes threats of violence against the president of the United States; specific threat in Watts held to be insufficient to satisfy requirements of section 871).

8. John Hart Ely, “Flag Desecration: A Case Study in the Roles of Categorization and Balancing in First Amendment Analysis,” Harvard Law Review 88, no. 7 (1975): 1495.

9. Gitlow v. New York, 268 U.S. 652, 673 (1925) (Holmes, J., dissenting).

10. 538 U.S. 343 (2003).

11. 123 Sup. Ct. at 1549.

12. Va. Code. Ann. §18.2-423 (Michie 1991) (enacted in 1950). The prima facie provision was added to the statute in 1968.

13. Id. at 1543.

14. Id. at 1561 (Souter, J.).

15. See, for example, Model Penal Code §211.1(1)(c) (Official Draft 1985) (“A person is guilty of assault if he . . . attempts by physical menace to put another in fear of imminent serious bodily injury”); ibid., §211.3 (one is guilty of a “terroristic” threat if one “threatens to commit any crime of violence with the purpose to terrorize another”); ibid., §250.4(2) (one is guilty of harassment if one taunts another in a manner likely to provoke a violence response). For additional examples, see Iowa Code §708.1(2) (West 1989); Fla. Stat. §784.011 (West 1992).

16. See Kent Greenawalt, Speech, Crime, and the Uses of Language (New York: Oxford University Press, 1989), 90–104 (generally), 298 (speech that is intended primarily to hurt the listener has limited expressive value and may properly subject the speaker to criminal punishment); Rodney A. Smolla, Free Speech in an Open Society (New York: Knopf, 1992): 48–50 (government interest in restricting speech is highest where that speech threatens physical harm).

17. See, for example, Mont. Code Ann. 45-5-203 (1991); the Montana Intimidation Statute, for example, provides as follows: “(1) A person commits the offense of intimidation when, with the purpose to cause another to perform or to omit the performance of any act, he communicates to another, under circumstances which reasonably tend to produce a fear that it will be carried out, a threat to perform without lawful authority any of the following acts: (a) inflict physical harm on the person threatened or any other person; (b) subject any person to physical confinement or restraint; or (c) commit any felony. (2) A person commits the offense of intimidation if he knowingly communicates a threat or false report of a pending fire, explosion, or disaster which would endanger life or property.” An earlier version of this statute required only a threat without any requirement that there be a reasonable tendency that the threat would produce fear. This earlier version was held to violate the First Amendment in a federal habeas corpus proceeding. See Wurtz v. Risley, 719 F. 2d 1438 (10th Cir. 1983). The statute was amended to conform with the court’s decision and has not been challenged since. See also State v. Lance, 721 P. 2d 1258 (Mont. 1986) (upholding section (1)(b) of the un-amended statute).

18. Robert Hughes, “Art, Morals, and Politics, The New York Review of Books, April 23, 1992, 21–27.

19. See Eric Harrison, “Mapplethorpe Display Brings Smut Charges,” Los Angeles Times, April 8, 1990,; Isabel Wilkerson, “Cincinnati Gallery Indicted In Mapplethorpe Furor,” New York Times, April 8, 1990,; Isabel Wilkerson, “Cincinnati Jury Acquits Museum In Mapplethorpe Obscenity Case,” New York Times, October 6, 1990,

20. Whitney v. California, 274 U.S., 357 (1927).

21. Charles L. Black Jr., “Alexander Mordecai Bickel,” Yale Law Journal 84, no. 2 (1974): 200.

To respond to this article, e-mail, with the author’s name on the subject line.

Frederick M. Lawrence is secretary and CEO of the Phi Beta Kappa Society, senior research scholar at Yale Law School, and visiting professor of law at Georgetown University Law Center. This article is based on an address delivered at the 2017 annual meeting of the Association of American Colleges and Universities.

Previous Issues