Dignity and Defamation (Visibility of Hate)

2009 OLIVER WENDELL HOLMES LECTURES
DIGNITY AND DEFAMATION: THE VISIBILITY OF HATE
Jeremy Waldron
TABLE OF CONTENTS
INTRODUCTION ....................................................................................................................... 1597

I. WHY CALL HATE SPEECH GROUP LIBEL? .................................................................. 1600
A. The Connotations of “Hate Speech” ............................................................................ 1600
B. The Terminology of “Group Libel” and “Group Defamation” .................................... 1601
C. Civil Versus Criminal Libel .......................................................................................... 1602
D. Can a Group be Libeled?: Beauharnais v. Illinois....................................................... 1605
E. Ways of Assaulting Group Reputation ......................................................................... 1609
F. The Emphasis on Dignity Rather than Offense .......................................................... 1612
G. Beauharnais and Libel in Light of New York Times Co. v. Sullivan....................... 1614
H. Who Needs Protection Against Libel? ......................................................................... 1615
II. WHAT DOES A WELL-ORDERED SOCIETY LOOK LIKE? .......................................... 1617
A. The Look of Hate ........................................................................................................... 1618
B. Hatred and Law in a Well-Ordered Society ................................................................ 1621
C. Political Aesthetics ........................................................................................................ 1623
D. Assurance and Security ................................................................................................. 1626
E. Public Goods .................................................................................................................. 1630
F. The Role of Law and the Role of Individuals ............................................................. 1632
G. Transition and Assurance .............................................................................................. 1633
III. LIBEL AND LEGITIMACY ................................................................................................ 1635
A. The Objection from Autonomy...................................................................................... 1635
B. Viewpoint Discrimination ............................................................................................. 1638
C. Ronald Dworkin’s Argument About Legitimacy .......................................................... 1639
D. Legitimacy: A Difference of Degree .............................................................................. 1642
E. Time and Settlement ...................................................................................................... 1646
F. The Owens Case in Saskatoon ..................................................................................... 1650
G. Islamophobia .................................................................................................................. 1652
H. Distrust of Government ................................................................................................. 1654
1596
2009 OLIVER WENDELL HOLMES LECTURES
DIGNITY AND DEFAMATION: THE VISIBILITY OF HATE
Jeremy Waldron∗
In his three 2009 Holmes Lectures published here, Professor Waldron seeks to describe
and defend laws forbidding group defamation — what we commonly refer to as “hate
speech” — as affirming the basic dignity of each member of society. Part I defends the
characterization of hate speech as group defamation. It argues that hate speech impugns
its victims’ standing as equal members of society. Part II describes hate speech
regulation as the protection of a fragile public good: the assurance offered by each
member of society to all of its members that they can live free of fear, discrimination,
violence, and the like. Part III defends the views articulated in Parts I and II from
various criticisms, particularly those of Professor Ronald Dworkin. Dworkin argues that
forbidding hate speech may result in a loss of democratic legitimacy for other laws. But
Professor Waldron argues that with sufficient safeguards the loss is vanishingly small,
and well worth the concomitant gains. As well, prohibitions on hate speech should only
extend to issues that are “settled,” such as race, rather than issues that are currently
controversial, which should further allay concerns that hate speech regulation will
foreclose freedom or democratic debate.
INTRODUCTION
About two years ago, I published a short piece in the New York
Review of Books, reviewing a book by Anthony Lewis called Freedom
for the Thought that We Hate.1 In it, I expressed some misgivings
about the arguments commonly used in America to condemn what we
call hate speech legislation — legislation of the sort you will find in
England, Canada, France, Denmark, Germany, New Zealand, and in
some of the states of Australia, prohibiting statements “by which a
group of people are threatened, insulted or degraded on account of
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
∗ University Professor, New York University. I am most grateful to Timothy Garton Ash,
Rebecca Brown, Winfried Brugger, Ronald Dworkin, David Dyzenhaus, Noah Feldman, James
Fleming, Charles Fried, Sanford Kadish, Frances Kamm, George Kateb, Henning Koch, Chris-
tine Korsgaard, David Kretzmer, Mattias Kumm, Rae Langton, Charles Lawrence, Anthony Lew-
is, Catherine MacKinnon, John Manning, Jane Mansbridge, Leighton McDonald, Frank Michel-
man, Martha Minow, Peter Molnar, Glyn Morgan, Liam Murphy, Thomas Nagel, Gerald
Neuman, Robert Post, Michael Rosen, Nancy Rosenblum, Michael Sandel, Carol Sanger,
T.M. Scanlon, Robert Silvers, Joseph Singer, Geoff Stone, Mark Tushnet, Roberto Unger, Ajume
Wingo, and the late C. Edwin Baker for their criticisms, suggestions, and (in some cases)
encouragement.
1 Jeremy Waldron, Free Speech & the Menace of Hysteria, N.Y. REV. BOOKS, May 29, 2008,
http://www.nybooks.com/articles/21452 (reviewing ANTHONY LEWIS, FREEDOM FOR THE
THOUGHT THAT WE HATE (2007)).
1597
1598 HARVARD LAW REVIEW [Vol. 123:1596
their race, colour, national or ethnic origin . . . .”2 I said that I thought
there was perhaps more to be said in favor of this legislation than An-
thony Lewis had suggested. But I did not make any very strong asser-
tion. I said: “It is not clear to me that the Europeans are mistaken
when they say that a liberal democracy must take affirmative respon-
sibility for protecting the atmosphere of mutual respect against certain
forms of vicious attack.”3 And I ended the piece with a reminder that:
[T]he issue is not just our learning to tolerate thought that we hate — we
the First Amendment lawyers, for example. . . . Maybe we should admire
some [ACLU] lawyer who says he hates what the racist says but defends
to the death his right to say it, but . . . [t]he [real] question is about the
direct targets of the abuse. Can their lives be led, can their children
be brought up, can their hopes be maintained and their worst fears dis-
pelled, in a social environment polluted by these materials? Those are the
concerns that need to be answered when we defend the use of the First
Amendment to strike down laws prohibiting the publication of racial
hatred.4
I thought that sounded very measured and moderate. Until I
opened my inbox a few weeks after the piece was published to find an
email whose subject line screamed, “YOU ARE A TOTALITARIAN
ASSHOLE.” “You are the type of human excrement that should be
dealt with IF the laws that you propose ever become reality. We do
not [sic] a strong state to support worthless little subsidized parasites
like you.”5 The email left me a little bruised, so I decided that in the
lectures on which this Article is based — lectures dedicated to the
memory of Oliver Wendell Holmes, who himself at one time or anoth-
er took both sides on most free speech issues — I would take the op-
portunity to explain myself.
My purpose is not to make a case for the constitutional acceptabili-
ty of hate speech laws in the United States. I think that is now more
or less an impossibility (though not, as I shall argue, for good reasons).
I will refer to the American debate occasionally, mostly suggesting
ways in which it might be enriched by more thoughtful consideration
of alternative positions. Mostly, what I want to do is offer a charac-
terization of the laws we find in Europe and in the other advanced
democracies of the world, and also as we have found them in America
from time to time. It is important to remember that opposition to hate
speech regulation in this country is by no means unanimous. Apart
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
2 This is the language used in section 266b(1) of the Danish Penal Code. See THE PRIN-
CIPAL DANISH CRIMINAL ACTS 64 (Malene Frese Jensen, Vagn Greve, Gitte Hoyer, & Martin
Spencer eds., DJOF Publ’g 3d ed. 2006).
3 Waldron, supra note 1.
4 Id. (emphasis added) (footnote omitted).
5 Email from Mike Hardesty to author (July 2, 2008, 14:51 EST) (on file with the Harvard
Law School Library).
2010] DIGNITY AND DEFAMATION 1599
from the legal academy, which is divided on the matter, there is divi-
sion among our lawmakers. There were state and municipal laws to
be struck down in R.A.V. v. City of St. Paul,6 in Virginia v. Black,7 and
in Collin v. Smith,8 and there was a state law to be upheld in Beau-
harnais v. Illinois.9 Not everyone in America is happy with the consti-
tutional untouchability of race leaflets in Chicago, Nazi banners in
Skokie, and burning crosses in Minnesota. There has been an honora-
ble impulse among some lawmakers in America to deal with this prob-
lem, and what we have always needed — before rushing to constitu-
tional outrage on behalf of the First Amendment — is to understand
that impulse. Outside the United States, we know that legislation of
this kind is common and widely accepted (though it is certainly not
uncontroversial). For us, that gives rise to a question about what the
European or Canadian or New Zealand legislators think they are
doing with these laws. Why have most liberal democracies undertaken
to prohibit these manifestations of hatred, these visible defamations of
social groups? How do they understand this legislation, how do they
defend it, and how do they position it in relation to concerns — which
they also profess — about freedom of expression?
In Part III, I will focus on one very powerful American argument
against legislation of this kind: an argument made by Professor Ronald
Dworkin and others about the effect that restrictions on free expres-
sion may have on the legitimacy of other laws that we want to be in a
position to enforce.10 I think that critique can be answered, though
not without difficulty. In Parts I and II, however, my argument will
be less defensive. In Part II, I will consider what people ought to be
able to draw from the visible environment of their society so far as
dignity, security, and assurance are concerned, as they lead their lives
and go about their business. I shall argue that hate speech regulation
can be understood as the protection of a certain sort of precious public
good: a visible assurance offered by society to all of its members that
they will not be subject to abuse, defamation, humiliation, discrimina-
tion, and violence on grounds of race, ethnicity, religion, gender, and in
some cases sexual orientation. I will not try to make the case that hate
speech laws actually reduce discrimination, violence, and so on, or that
they make it more likely that hatemongers will give up their bigotry
and become good, tolerant citizens. I hope that will happen, and hate
speech laws may work as part of a broader campaign for equality and
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
6 505 U.S. 377 (1992).
7 538 U.S. 343 (2003).
8 578 F.2d 1197 (7th Cir. 1978).
9 343 U.S. 250 (1952).
10 Ronald Dworkin, Foreword to EXTREME SPEECH AND DEMOCRACY, at v, viii–ix (Ivan
Hare & James Weinstein eds., 2009).
toleration. But I am going to argue that the most important aim of

these laws is more immediate. The aim is simply to diminish the pres-
ence of visible hatred in society and thus benefit members of vulnera-
ble minorities by protecting the public commitment to their equal
standing in society against public denigration. In Part I of the Article,
I will sketch some background for this aim, arguing that it helps to
view hate speech laws as protecting vulnerable minorities against the
evil of group defamation. These days we tend to think of defamation
as a tort. But I will argue in Part I that historically the law of crimi-
nal libel has been used to support and express a collective commitment
on the part of society to uphold the fundamentals of people’s reputa-
tions as members of society in good standing — vindicating, as I shall
say, the rudiments of their civic dignity as a necessary ingredient of
public order.
I. WHY CALL HATE SPEECH “GROUP LIBEL”?

What we call a thing tells us something about our attitude towards
it, how we see it as a problem, what our response to it might be, what
difficulties our response to it might bring to light. So it is with the
phenomenon that we call in America “hate speech,” which can cover
things as diverse as cross-burnings, racial epithets, insults to religion,
bestial and other offensive depictions of vulnerable minorities in leaf-
lets, posters, or on the internet, broad-brush ascriptions of criminality
or dangerousness, calls to unite against the members of a hated group,
and neo-Nazis marching in American suburbs with swastikas and pla-
cards saying “Hitler should have finished the job.”
A. The Connotations of “Hate Speech”
When we call these phenomena “hate speech,” we bring to the fore
a number of connotations that are not entirely neutral. If we say we
are interested in restrictions on hate speech, we convey the idea that
the law is proposing to interfere with the spoken word, with conversa-
tion, and perhaps with vocabulary, with our use of racial or ethnic
slurs or epithets — stammered out, as Justice Jackson once put it,
“when the spirits are high and the flagons are low.”11 Speech, in the
sense of the spoken word, can certainly be wounding.12 But I believe
that the expressions of hatred that should concern us include most
prominently those that are printed, published, pasted up, or posted, or
in some other form become part of the visible environment in which
our lives have to be lived. No doubt a speech can resonate long after
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
11Kunz v. New York, 340 U.S. 290, 299 (1951) (Jackson, J., dissenting).
12See, e.g., Charles R. Lawrence III, If He Hollers Let Him Go: Regulating Racist Speech on
Campus, 1990 DUKE L.J. 431, 455.
the spoken word has died away, but to my mind it is the enduring
presence of the published word that is particularly damaging.
The kind of speech we say we are interested in regulating is hate
speech, and that word “hate” can be distracting too. It suggests we are
interested in looking at and regulating the passions and emotions that
lie behind a particular speech act.13 The word “hate” emphasizes the
subjective attitudes of the person expressing the views or publishing
the message in question. It sounds as though it locates the problem as
an attitudinal one and focuses on what motivates the speech in ques-
tion.14 (It is like the phrase “hate crimes” in this respect, and people
may be excused for thinking that the controversy over hate crimes —
over the use of mental elements like motivation as an aggravating fac-
tor in criminal law — is directly relevant to the controversy over racial
expression.)15 The word “hate” suggests — I think misleadingly —
that the task of legislation that restricts hate speech is to try to change
people’s attitudes or control their thoughts.
The restrictions on hate speech that interest me are not restrictions
on thinking; they are restrictions on more tangible forms of communi-
cation. The issue is publication and the harm done to individuals and
groups through the disfiguring of our social environment by visible,
public, and semi-permanent announcements to the effect that in the
opinion of one group in the community, members of another group are
not worthy of equal citizenship.
B. The Terminology of “Group Libel” and “Group Defamation”
In many countries, a different term or set of terms is used by jur-
ists: “group libel” or “group defamation.” Sometimes this is how the
legislation describes itself; it is the terminology used, for example, in
section 130 of Germany’s Penal Code, prohibiting “attacks on human
dignity by insulting, maliciously maligning, or defaming part of the
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
13 It can also bog us down in a futile attempt to define “hate.” Hatred is not an easy idea to
define; Robert Post has a valiant stab at defining “hate” in Hate Speech, in EXTREME SPEECH
AND DEMOCRACY, supra note 10, at 123, because he thinks the crucial issues are determining
when “otherwise appropriate emotions become so ‘extreme’ as to deserve legal suppression,” id. at
123, and “distinguish[ing] hatred from ordinary dislike or disagreement,” id. at 125.
14 Actually, if “hatred” is relevant at all, it is relevant (in many of the statutory formulations) as
the purpose of the offending speech, not as its motivation. The International Covenant on Civil
and Political Rights, adopted Dec. 16, 1966, 999 U.N.T.S. 171, requires signatory states to prohibit
“advocacy of national, racial or religious hatred,” id. at art. 20(2), while the British Public Order
Act, 1986, c. 64, § 18(1), states that “[a] person who uses threatening, abusive or insulting words or
behaviour, or displays any written material which is threatening, abusive or insulting, is guilty of
an offence if . . . (a) he intends thereby to stir up racial hatred, or (b) having regard to all the cir-
cumstances racial hatred is likely to be stirred up thereby.”
15 See, e.g., Heidi M. Hurd & Michael S. Moore, Punishing Hatred and Prejudice, 56 STAN.
L. REV. 1081 (2004).
population.”16 In other countries, “group libel” and “group defama-

tion” are terms used in judicial doctrine and among lawyers to describe
restrictions of the kind we would call hate speech restrictions.
The term “group libel” also used to be common in the United
States, and it was alluded to by the Supreme Court in characterizing a
state law that was upheld in 1952 in Beauharnais v. Illinois.17 “Just a
little more than a decade ago,” wrote Harry Kalven in 1964, “we were
all concerned with devising legal controls for the libeling of groups.”18
Five years before Beauharnais, some scholars at Columbia tried to
crystallize debate by publishing a model group libel statute in the Co-
lumbia Law Review.19 It is worth remembering, too, that — as its
name suggests — the Jewish Anti-Defamation League took as its orig-
inal mission “to stop, by appeals to reason and conscience, and if
necessary by appeals to law, the defamation of the Jewish people.”20
C. Civil Versus Criminal Libel
James Weinstein says that the idea of group libel is constructed by
analogy with the tort of defamation; but, as he notes, this analogy is an
oversimplification.21 Libel may be best known today as a tort, but in
the past it has often been understood also as a criminal offense. Crim-
inal libel laws came in various flavors. I suppose the best known are
the laws against seditious libel — of which, for us, the most notorious
example is the Sedition Act, passed by Congress in 1798, making it a
criminal offense to publish “false, scandalous, and malicious writing”
bringing the president or Congress into disrepute or “to excite against
them . . . the hatred of the good people of the United States.”22 This
spectacularly ill-considered piece of legislation has given criminal libel
a bad name in the United States ever since.
Or consider blasphemous libel. William Blackstone observed that
“blasphemy against the Almighty, . . . denying his being or providence;
or . . . contumelious reproaches of our saviour Christ . . . [is an offense]
punishable at common law by fine and imprisonment, or other infam-
ous corporal punishment: for christianity is part of the laws of Eng-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
16 Strafgesetzbuch [StGB] [Penal Code] § 130, ¶ 1, sentence 2. In France, article 29 of the Law
on the Freedom of the Press of 29 July 1881 also prohibits group defamation.
17 343 U.S. 250, 253–54 (1952); see also Joseph Tanenhaus, Group Libel, 35 CORNELL L.Q.
261 (1950) (for a general discussion of this concept).
18 HARRY KALVEN, JR., THE NEGRO AND THE FIRST AMENDMENT 7 (1965).
19 Note, Statutory Prohibition of Group Defamation, 47 COLUM. L. REV. 595 (1947).
20 STRIKING A BALANCE: HATE SPEECH, FREEDOM OF EXPRESSION AND NON-
DISCRIMINATION 326 (Sandra Coliver ed., 1992) (emphasis added) (internal quotation mark
omitted).
21 James Weinstein, Extreme Speech, Public Order and Democracy, in EXTREME SPEECH
AND DEMOCRACY, supra note 10, at 23, 58–59.
22 Alien and Sedition Acts of 1798, ch. 74, § 2, 1 Stat. 596 (expired 1801).
land.”23 Blasphemous libel was held to be a common law offense in

the United Kingdom as late as 1977, when a private prosecution was
brought successfully by Mary Whitehouse against the publishers of
Gay News for a poem that described necrophiliac acts performed upon
the body of Jesus after his crucifixion.24 In 1823, a man was jailed for
three months in Massachusetts for publishing an essay in the Boston
Investigator that denied the existence of God, affirmed the finality of
death, and declared that “the whole story concerning [Jesus Christ] is
as much a fable and a fiction as that of the god Prometheus.”25 There
was also obscene libel — an offense that covered the publication of
virtually any obscene matter. Edmond Curl was found guilty in Eng-
land in 1727 in respect of a book called Venus in the Cloister, about
lesbian love in a convent.26
Notice that, in these senses of libel, we are not really dealing with
offenses that have a whole lot to do with defaming people. Some of
the prosecutions under the Sedition Act involved defamation of those
in power.27 But others involved general subversion of government. In
United States v. Crandell,28 an indictment was laid against Reuben
Crandell for “publishing libels tending to excite sedition among the
slaves.”29 Often the term just goes back to the neutral meaning of the
Latin libellus, a “little book.” For much of its history “libel” could be
used to refer to any published pamphlet, without conveying any judg-
ment about its content.30
When we do focus on defamation, what is consistently emphasized
is the distinction between calumnies that are put about in spoken form,
as speech, through gossip, rumor, or oral denunciation, and those that
have the more enduring presence of something published “by writing,
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
23 4 WILLIAM BLACKSTONE, COMMENTARIES *59.
24 See Corinna Adam, Protecting Our Lord, NEW STATESMAN, July 15, 1977, at 74, available
at http://www.newstatesman.com/200602130044. Gay News was fined 1000 pounds and its pub-
lisher was given a nine-month suspended prison sentence. When a challenge was brought under
the European Convention on Human Rights, the European Commission of Human Rights found
that prohibitions on blasphemous libel could constitute a reasonable restriction on free speech.
See Gay News Ltd. v. United Kingdom, 5 Eur. H.R. Rep. 123 (1982).
25 Commonwealth v. Kneeland, 20 Pick. 206, 207 (Mass. 1838). The Blackstone position on
blasphemous libel was also adopted explicitly by a judge in Pennsylvania in 1824: “Christianity,”
he said, “general Christianity, is, and always has been, a part of the common law of Pennsylva-
nia.” Updegraph v. Commonwealth, 11 Serg. & Rawle. 394, 400 (Pa. 1824).
26 R. v. Curl, (1727) 93 Eng. Rep. 849 (K.B.). See also Colin Manchester, A History of the
Crime of Obscene Libel, 12 J. LEGAL HIST. 36, 38–40 (1991).
27 See United States v. Haswell, 26 F. Cas. 218 (C.C.D. Vt. 1800) (No. 15,324); Lyon’s Case, 15
F. Cas. 1183 (C.C.D. Vt. 1798) (No. 8646). I discussed these cases in Waldron, supra note 1.
28 25 F. Cas. 684 (C.C.D.C. 1836) (No. 14,885).
29 Id. at 685.
30 See THE COMPACT OXFORD ENGLISH DICTIONARY 880 (2d ed. 1991) (listing the first
definition of “libel” as “[a] little book; a short treatise or writing”).
printing, effigy, picture, or other fixed representation to the eye.”31

“What gives the sting to the writing,” said a New York court in 1931,
“is its permanence of form. The spoken word dissolves, but the writ-
ten one abides and ‘perpetuates the scandal.’”32 I believe this distinc-
tion is helpful for our inquiry. Much of what we call hate speech regu-
lation is not directed at the immediate flare-up of insult and offense.
The concept of “group libel” addresses the possibility of racial or reli-
gious defamation becoming established as a visible feature of the envi-
ronment — part of what you can see in real space (or virtual space) as
you look around you.33
Until recently, many countries had laws of criminal libel touching
ordinary individuals. The New Zealand Crimes Act used to specify a
year’s imprisonment as the penalty for any “matter published, without
lawful justification or excuse . . . designed to insult any person or likely
to injure his reputation by exposing him to hatred, contempt, or ridi-
cule . . . .”34 Now why, you may ask, would the criminal law concern
itself with defamation at all, when there was no public issue of sedition
or obscenity or blasphemy? Why not leave it to private law?
One possibility is that certain forms of defamation might be seen as
an attack on public order. It was a matter of keeping the peace, avoid-
ing brawls and so on, in the context of egregious libel flowing over into
fighting words. But public order is a complicated idea, and preventing
fighting or violence from breaking out — that very narrow sense of
keeping the peace — is only one of its dimensions. Public order might
also comprise society’s interest in maintaining among us a proper sense
of one another’s social or legal status. In an aristocratic society, this
meant securing the dignity of great men or high officials with laws of
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
31 PHILIP WITTENBERG, DANGEROUS WORDS: A GUIDE TO THE LAW OF LIBEL 7 (1947)
(quoting a California statute that, in its current version, defines libel as something published “by
writing, printing, picture, effigy, or other fixed representation to the eye,” CAL. CIV. CODE § 45
(West 2007)). The phrase seems to come originally from Odgers on Libel. See Staub v. Van Ben-
thuysen, 36 La. Ann. 467, 468–69 (La. 1884) (“A libel is any publication whether in writing, print-
ing, picture, effigy, or other fixed representation to the eye which exposes any person to hatred,
contempt, ridicule, or obloquy, or which causes him to be shunned or avoided, or which has a
tendency to injure him in his occupation.”) (citing ODGERS, LIBEL AND SLANDER 7, 20 (1st ed.
1881)).
32 Ostrowe v. Lee, 175 N.E. 505, 506 (N.Y. 1931) (quoting Harman v. Delany, (1729) 94 Eng.
Rep. 743 (K.B.) (“Words published in writing are actionable, which would not be so from a bare
speaking of the same words, because a libel disperses and perpetuates the scandal.”)).
33 In the case of R. v. Curl, (1727) 93 Eng. Rep. 849 (K.B.), this was crucial to an understand-
ing of why this obscenity was a matter for the temporal courts rather than for a spiritual tribunal
set up by a bishop. “The Spiritual Courts punish only personal spiritual defamation by words; if
it is reduced to writing, it is a temporal offence. . . . This is surely worse than Sir Charles Sedley’s
case, who only exposed himself to the people then present, who might choose whether they would
look upon him or not; whereas this book goes all over the kingdom.” Id. at 850–51.
34 Crimes Act 1961, 1961 S.N.Z. No. 43, § 211 (N.Z.), repealed by Defamation Act 1992,
§ 56(2), 1992 S.N.Z. No. 105 (emphasis added).
scandalum magnatum,35 to protect nobles and great men from outra-

geous imputations on their breeding, their status, their honor, or their
office. The United States abolished titles of nobility in 1787, but it did
not necessarily abolish that sort of concern for status.36 A democratic
republic might equally be concerned with upholding and vindicating
important aspects of legal and social status — only now it would be
the elementary dignity of even its non-officials as citizens — and with
protecting that status (as a matter of public order) from being under-
mined by various forms of obloquy. And that is what I think is the
concern of laws regarding group defamation. They are set up to vin-
dicate public order, not just by pre-empting violence, but by upholding
against attack a shared, public sense of the basic elements of each per-
son’s status, dignity, and reputation as a citizen or member of society
in good standing — particularly against attacks predicated upon the
characteristics of some particular social group.
D. Can a Group Be Libeled?: Beauharnais v. Illinois
Earlier I mentioned that the characterization of hate speech as
group libel is not unknown in the United States. In 1952, what we
would call a hate speech law (dating from 1917) was described and
upheld by the Supreme Court as a law of criminal libel.37 The peti-
tioner was one Joseph Beauharnais, President, Founder, and Director
of the White Circle League of America, who had distributed a leaflet
on Chicago street corners urging people to protect the white race from
being “mongrelized” and terrorized by the “rapes, robberies, knives,
guns and marijuana of the negro.”38 The leaflet had as its headline:
“Preserve and Protect White Neighborhoods! From the Constant and
Continuous Invasion, Harassment and Encroachment by the Ne-
groes.”39 It said: “WE ARE NOT AGAINST THE NEGRO! WE
ARE FOR THE WHITE PEOPLE! We must awaken and protect
our white families and neighborhoods before it is too late.”40 But, it
stated, “The white people of Chicago MUST take advantage of this
opportunity to become UNITED. If persuasion and the need to pre-
vent the white race from becoming mongrelized by the negro will not
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
35 For a discussion of scandalum magnatum, see John C. Lassiter, Defamation of Peers: The
Rise and Decline of the Action for Scandalum Magnatum, 1497–1773, 22 AM. J. LEGAL HIST.
216 (1978); and Edward Veitch, Scandalum Magnatum Is Alive and Well in Canada?, 11 NAT’L J.
CONST. L. 169 (1999).
36 See Jeremy Waldron, Dignity and Rank, 48 EUR. J. SOCIOLOGY 201 (2007); James Q.
Whitman, “Human Dignity” in Europe and the United States: The Social Foundations, in
EUROPEAN AND U.S. CONSTITUTIONALISM 108 (Georg Nolte ed., 2005).
37 Beauharnais v. Illinois, 343 U.S. 250, 253–57 (1952).
38 Id. at 252.
39 Id. at 276 (appendix to opinion of Black, J., dissenting).
40 Id.
unite us, then the aggressions[,] . . . rapes, robberies, knives, guns and
marijuana of the negro, SURELY WILL.”41 The leaflet provided a
tear-off application form, which, if submitted with a dollar, would en-
able the sender to become a member of the White Circle League of
America (provided he promised to try and secure other members as
well).42
On March 6, 1950, Beauharnais was charged under an Illinois stat-
ute prohibiting the publication or exhibition of any writing or picture
portraying the “depravity, criminality, unchastity, or lack of virtue of a
class of citizens, of any race, color, creed or religion.”43 He was con-
victed by a jury and fined a sum of $200.44 His conviction was upheld
on appeal in Illinois and upheld too by a 5-4 majority of the U.S. Su-
preme Court.
Justice Frankfurter, writing for the U.S. Supreme Court, accepted
the Supreme Court of Illinois’s description of the statute as “a form of
criminal libel law.”45 Frankfurter thought characterizing Beauhar-
nais’s leaflet as criminal libel placed it beyond the protection of the
First Amendment: “[l]ibelous utterances,” he wrote, are not “within the
area of constitutionally protected speech.”46 But he did not rest the
decision on a purely formal characterization. He noted that, as a mat-
ter of public order, the state might have a reasonable ground for being
concerned about this type of libel:
Illinois did not have to look beyond her own borders or await the tragic
experience of the last three decades to conclude that wilful purveyors of
falsehood concerning racial and religious groups promote strife and tend
powerfully to obstruct the manifold adjustments required for free, ordered
life in a metropolitan, polyglot community.47
That put it mildly. Justice Douglas, even in dissent, noted that the
Nazis were an example of “how evil a conspiracy could be which was
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
41 Id.
42 See id.
43 Id. at 250 (majority opinion) (internal quotation mark omitted).
44 Id. at 251.
45 Id. at 253 (quoting People v. Beauharnais, 97 N.E.2d 343, 346 (Ill. 1951)) (internal quotation
mark omitted).
46 Id. at 266.
47 Id. at 258–59 (footnote omitted). Professor Nadine Strossen cautions that before we get too
enthusiastic about the ordinance upheld in Beauharnais, we should remember that prior to its use
against this white supremacist group, it “was ‘a weapon for harassment of the Jehovah’s Wit-
nesses,’ who were then ‘a minority . . . very much more in need of protection than most.’” Na-
dine Strossen, Regulating Racist Speech on Campus: A Modest Proposal?, 1990 DUKE L.J. 484,
520 (quoting Joseph Tanenhaus, Group Libel, 35 CORNELL L.Q. 261, 279–80 (1950)) (alteration in
original). In fact, the Jehovah’s Witnesses were prosecuted for what a federal court described as
“bitter and virulent attacks upon the Roman Catholic Church” and “accusations which in sub-
stance and effect were charges of treasonable disloyalty.” Bevins v. Prindable, 39 F. Supp. 708,
710 (E.D. Ill. 1941).
aimed at destroying a race by exposing it to contempt, derision, and

obloquy.”48 Justice Douglas also wrote that he “would be willing to
concede that such conduct directed at a race or group in this country
could be made an indictable offense.”49 Like two of the other three
dissenters, he had no problem with the idea that group libel laws
enacted to protect the public peace were compatible with the First
Amendment.50 Only Justice Black disputed Frankfurter’s premise out-
right, and for him the problem was the “group” aspect of group libel:
[A]s “constitutionally recognized[,]” [criminal libel] has provided for
punishment of false, malicious, scurrilous charges against individuals, not
against huge groups. This limited scope of the law of criminal libel is of
no small importance. It has confined state punishment of speech and ex-
pression to the narrowest of areas involving nothing more than purely pri-
51
vate feuds.
I think this was a mistake. And I would like to consider Justice
Black’s objection in detail, before addressing a different criticism that
could be made after 1964 — namely, that the decision in New York
Times Co. v. Sullivan52 has removed the whole category of libel from
the list of exceptions to First Amendment protection.53
In my view, Justice Black’s claim that criminal libel provides only
for the “punishment of false, malicious, scurrilous charges against in-
dividuals,”54 not for the punishment of similar charges against large
groups, is misleading for a number of reasons. To begin with, it neg-
lects an important difference between the concern for personalized
reputation in civil cases and a broader social concern for the funda-
mentals of anyone’s reputation or civic dignity as a member of society
in good standing. Unlike civil libel, criminal libel has traditionally
been concerned not with the intricate detail of each person’s reputa-
tion and its movement up or down the scale of social estimation, but
with its foundation. Indeed, it has sometimes been argued that the
civil law of libel and the criminal law of libel work together, to cover
the field as it were. In the case of a civil action for libel, there must be
a defaming of a particular person, or of a group so confined that the
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
48 Beauharnais, 343 U.S. at 284 (Douglas, J., dissenting).
49 Id.
50 Justice Reed in his dissent assumed “the constitutional power of a state to pass group libel
laws to protect the public peace.” Id. at 283 (Reed, J., dissenting). His objection was based on the
vagueness of the terms of the ordinance. Id. Justice Jackson noted that “[m]ore than forty State
Constitutions, while extending broad protections to speech and press, reserve a responsibility for
their abuse and implicitly or explicitly recognize validity of criminal libel laws.” Id. at 292 (Jack-
son, J., dissenting).
51 Id. at 271–72 (Black, J., dissenting).
52 376 U.S. 254 (1964).
53 See infra section I.G, p. 1614–15.
54 Beauharnais, 343 U.S. at 272 (Black, J., dissenting).
allegation descends to particulars. But — so the argument goes — this

does not mean that the law is unconcerned with defamation on a
broader front, only that the problem now becomes the concern of the
criminal law.
Certainly that is what one would conclude from a public order
perspective.55 Consider the eighteenth-century English case, R. v. Os-
borne.56 Osborne was charged with publishing a pamphlet entitled “A
true and surprizing Relation of a Murder and Cruelty that was com-
mitted by the Jews lately arrived from Portugal; shewing how they
burnt a Woman and a new born Infant the latter End of February, be-
cause the Infant was begotten by a Christian.”57 There was an objec-
tion at the trial that “the Charge was so general that no particular Per-
sons could pretend to be injured by it.”58 But the court responded:
This is not by way of Information for a Libel that is the Foundation of
this Complaint, but for a Breach of the Peace, in inciting a Mob to the
Distruction of a whole Set of People; and tho’ it is too general to make it
fall within the Description of a Libel, yet it will be pernicious to suffer
such scandalous Reflections to go unpunished.59
Other reports of the same case say that it was decided as a matter
of criminal libel, but agreed that the public order dimension was key
to that characterization.60 We find the same approach taken in an
American case from 1868. Palmer v. City of Concord61 concerned ac-
cusations of cowardice made against a company of soldiers who had
been engaged in the Civil War. The Supreme Judicial Court of New
Hampshire said this:
As these charges were made against a body of men, without specifying in-
dividuals, it may be that no individual soldier could have maintained a
private action therefor. But the question whether the publication might
not afford ground for a public prosecution is entirely differ-
ent. . . . Indictments for libel are sustained principally because the publica-
tion of a libel tends to a breach of the peace, and thus to the disturbance
of society at large. It is obvious that a libellous attack on a body of men,
though no individuals be pointed out, may tend as much, or more, to
create public disturbances as an attack on one individual.62
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
55 See Tanenhaus, supra note 17, at 266 (“Since criminal libel is indictable at common law be-
cause it tends so to inflame men as to result in a breach of the peace, there is no rational basis for
the exclusion of group defamers from liability to prosecution in common law jurisdictions.”).
56 (1732) 25 Eng. Rep. 584.
57 Id. at 584–85.
58 Id. at 585.
59 Id.
60 For an acceptance of this case as group libel, see Anonymous, (1732) 94 Eng. Rep. 406 (K.B.)
and R. v Osborn, (1732) 94 Eng. Rep. 425 (K.B.). For an ambiguous account of the same case, see
In re Bedford Charity, (1819) 36 Eng. Rep. 696, 717.
61 48 N.H. 211 (1868).
62 Id. at 215 (emphasis added).
That seems to me a viable, or at least arguable, position. But Justice

Black’s dissent in Beauharnais takes the point in exactly the wrong di-
rection with its perverse implication that the very large number of
people defamed in the White Circle League’s leaflet meant that the
leaflet could not be subject to any sort of regulation at all, because it
enjoyed constitutional protection in a way that the defamation of a
single person would not.
E. Ways of Assaulting Group Reputation
How does one libel a group? What aspects of group reputation are
we trying to protect with laws against racial or religious defamation?
The first thing to note is that it is not the group as such that we are ul-
timately concerned about — as one might be concerned about a com-
munity, a nation, or a culture (as distinct from its members). The con-
cern in the end is individualistic. But, as I have already said, group
defamation laws do not concern themselves with particularized indi-
vidual reputation. They look instead to the basics of social standing
and to the association that is made — in the hate speech, in the libel,
in the defamatory pamphlet or poster — between the denigration of
that basic standing and some characteristic associated more or less as-
criptively with the group or class. I do not mean that group member-
ship is in and of itself a liability. But group defamation sets out to
make it a liability by denigrating group-defining characteristics or as-
sociating them with bigoted factual claims that are fundamentally
defamatory. A prohibition on group defamation, then, is a way of
blocking that enterprise.
So, let me count the ways in which a group might be libeled. In
the first instance, the association of defamation with an ascriptive
group characteristic might take the form of a factual claim. That
was important in the Beauharnais pamphlet, with its imputation
that “rapes, robberies, knives, guns and marijuana” were somehow
typical of “the negro.”63 And it is likely to be important, too, in broad-
brush descriptions of Arab-Americans and Muslims as supporters of
terrorism.
Secondly, group libel often involves characterizations that delib-
erately set out to denigrate people — characterizations that probably
fall on the opinion rather than fact side of the distinction sometimes
made in American defamation law. A common example in racist
speech is the characterization of minority members as animals. I re-
member seeing a racist agitator sentenced to a short prison term in
England in the late 1970s for festooning the streets of Leamington Spa
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
63 Beauharnais v. Illinois, 343 U.S. 250, 276 (1952) (appendix to opinion of Black, J., dissent-
ing); see supra pp. 1605–06.
with posters depicting Britons of African ancestry as apes.64 After his

conviction by the jury, he was sentenced by a crusty old English judge,
who (one might have imagined) would have little sympathy for this
newfangled hate speech legislation. But the judge gave the defendant
a stern lecture to the effect that we cannot run a multiracial society
under modern conditions if people are free to denigrate their fellow cit-
izens in bestial terms. There was some shouting from the gallery as
the defendant was taken away. The case made a deep impression on
me.
Thirdly, there are libels that go even beyond opinion but which
denigrate by embodying notices or instructions which can only be un-
derstood as degrading those to whom they are addressed. Thus, for
example, a group and its members can be libeled by signage associat-
ing group membership with prohibition or exclusion: “No blacks al-
lowed.” In 1944, Ontario’s Racial Discrimination Act65 prohibited the
publication or display of “any notice, sign . . . or other representation
indicating . . . an intention to discriminate against any . . . class of per-
sons . . . because of the race or creed of such . . . class of persons.”66
That was quite apart from the prohibition on discrimination itself. I
was interested to learn that in the early days of the Jewish Anti-
Defamation League in the United States, one of the League’s aims was
to put a stop to the poisoning of the social environment by published
declarations of racial and religious exclusion. When the ADL cam-
paigned for legislation preventing stores and hotels from refusing to do
business with Jews, it was not just the discrimination they wanted to
counter, it was the signage: “Christians only.” What concerned the
ADL was the danger that anti-Semitic signage would become an estab-
lished feature of the landscape and that Jews would have to lead their
lives in a community whose public aspect was permanently disfigured
in this way.67
Singly or together, these reputational attacks amount to assaults
upon the dignity of the persons affected — dignity, in the sense of
these persons’ basic social standing, of the basis of their recognition as
social equals, and of their status as bearers of human rights and consti-
tutional entitlements. The moral imperative of respect for human dig-
nity is increasingly understood as a crucial foundation of basic rights
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
64 I discussed this incident at the beginning of Jeremy Waldron, Boutique Faith, LONDON
REV. BOOKS, July 20, 2006, at 22, 22 (reviewing JOHN DURHAM PETERS, COURTING THE
ABYSS: FREE SPEECH AND THE LIBERAL TRADITION (2005)).
65 1944 S.O., ch. 51-1 (Ont.).
66 Id.
67 See Evan P. Schultz, Group Rights, American Jews, and the Failure of Group Libel Laws,
1913–1952, 66 BROOK. L. REV. 71, 95–96 (2000).
and equality.68 The Universal Declaration of Human Rights begins its

preamble with the strong assertion that “recognition of the inherent
dignity and of the equal and inalienable rights of all members of the
human family is the foundation of freedom, justice and peace in the
world,”69 and the International Covenant on Civil and Political Rights
follows up on that by stating that the rights it protects “derive from
the inherent dignity of the human person.”70 Legal and political theo-
rists are finding that dignity provides a useful and compelling perspec-
tive on the foundations of constitutional rights,71 and also on the foun-
dations of ideals such as democracy and the rule of law.72 As well as
these broad ideals, dignity is also increasingly appealed to on particu-
lar issues in legal controversy — on the death penalty (where its use is
well known),73 on antidiscrimination law,74 on issues relating to abor-
tion,75 and — as we shall see — in scholarly controversies on this mat-
ter of hate speech.76
Dignity is a complex idea, with philosophical as well as political
and legal resonances. In the sense I am using the term, dignity is not
just a Kantian philosophical conception of the immeasurable worth of
humans considered as moral agents.77 It is a matter of status — one’s
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
68 There is an excellent discussion and critique of the legal concept of human dignity in Chris-
topher McCrudden, Human Dignity and Judicial Interpretation of Human Rights, 19 EUR. J.
INT’L L. 655 (2008).
69 Universal Declaration of Human Rights, Preamble, G.A. Res. 217A, at 71, U.N. GAOR, 3d
Sess., 1st plen. mtg., U.N. Doc. A/810 (Dec. 10, 1948).
70 International Covenant on Civil and Political Rights, Preamble, adopted Dec. 16, 1966, 999
U.N.T.S. 171.
71 See, for example, the fine discussion in Gerald L. Neuman, Human Dignity in United
States Constitutional Law, in ZUR AUTONOMIE DES INDIVIDUUMS: LIBER AMICORUM SPI-
ROS SIMITIS 249 (Dieter Simon & Manfred Weiss eds., 2000).
72 I pursued the theme of the relation between dignity and the rule of law in my 2009 Tanner
Lectures at Berkeley. See Jeremy Waldron, Dignity, Rank, and Rights, in 29 THE TANNER
LECTURES ON HUMAN VALUES (Suzan Young ed., forthcoming 2010), available at
http://ssrn.com/abstract=1461220; see also Jeremy Waldron, The Concept and the Rule of Law, 43
GA. L. REV. 1, 28, 36 (2008).
73 See, e.g., Furman v. Georgia, 408 U.S. 238 (1972); see also HUGO ADAM BEDAU, KILLING
AS PUNISHMENT: REFLECTIONS ON THE DEATH PENALTY IN AMERICA 110–41 (2004) (ar-
guing for a dignitarian interpretation of the Eighth Amendment).
74 See, e.g., Denise G. Réaume, Discrimination and Dignity, 63 LA. L. REV. 645 (2003).
75 See Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 846–53 (1992) (plurality opinion);
see also Carol Sanger, Decisional Dignity: Teenage Abortion, Bypass Hearings, and the Misuse of
Law, 18 COLUM. J. GENDER & L. 409 (2009); Reva B. Siegel, Dignity and the Politics of Protec-
tion: Abortion Restrictions Under Casey/Carhart, 117 YALE L.J. 1694 (2008).
76 See generally, e.g., STEVEN J. HEYMAN, FREE SPEECH AND HUMAN DIGNITY (2008);
Alexander Tsesis, Dignity and Speech: The Regulation of Hate Speech in a Democracy, 44 WAKE
FOREST L. REV. 497 (2009). For an expression of concern about the use of “dignity” in this con-
nection, see Guy E. Carmi, Dignity — The Enemy from Within: A Theoretical and Comparative
Analysis of Human Dignity as a Free Speech Justification, 9 U. PA. J. CONST. L. 957 (2007).
77 See IMMANUEL KANT, GROUNDWORK OF THE METAPHYSICS OF MORALS 42–43
(Mary Gregor trans. & ed., Cambridge Univ. Press 1998) (1785).
status as a member of society in good standing. It validates the legal

position of the ordinary individual both as an equal and (paradoxical-
ly) as the possessor of a very high-ranking status.78 And it generates
demands for recognition and treatment that accord with that status.
Philosophically we may say that dignity is inherent in the human per-
son — and so it is. No law or social practice can take it away. But as
a social and legal status, dignity has to be nourished and maintained
by society and the law, and this — as I argue in Part II — is a costly
and difficult business and something in which we are all required to
play a part. At the very least, we are required in our public dealings
not to act in a way that undermines one another’s dignity in this socio-
legal sense — and that is the obligation that is being enforced when we
enact and administer laws against group libel.
F. The Emphasis on Dignity Rather than Offense
This brings me to an important distinction. There is a big differ-
ence between protecting individual dignity from defamation (based on
some denigration of group characteristics) and protecting people from
offense, even when the offense goes to the heart of what they regard as
the identity of their group. The argument I want to make is not con-
cerned with offense. It is concerned with what happens to the stand-
ing of persons and groups in society, not with any element of their dis-
tress unconnected with that.
This distinction is particularly important for understanding laws
prohibiting hate speech against religious minorities.79 Consider, for
example, worries about expressions of Islamophobia. The group of all
Muslims in society, the group of all followers of Islam, is a group of
people committed to the one God, to his Prophet, Mohammed, and to
the holy writings of the Koran. On the account that I am developing,
individual Muslims are entitled to protection against defamation, in-
cluding group defamation as Muslims. A published allegation, for ex-
ample, that all Muslims are terrorists or are properly suspected of in-
volvement with terrorism would rightly fall foul of well-drafted hate
speech legislation. But that does not mean that the law should aim to
protect the founders of the religion, or the reputation of God as Mus-
lims understand Him, or the creedal beliefs of the group.80 The civic
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
78 See Waldron, supra note 36, at 215–19.
79 See, for example, Public Order Act, 1986, c. 64, Part 3A (Eng.), amended by Racial and Re-
ligious Hatred Act, 2006, c. 1 (Eng.), which prohibits the publication of threatening material in-
tended to stir up religious hatred. I discuss this provision further in Part III.
80 Earlier I mentioned the 1977 prosecution in England of Gay News for blasphemous libel
against Jesus. See supra note 24 and accompanying text. It is no part of my argument that hate
speech law should support this sort of prosecution. That example was offered simply to illustrate
the diversity of historic uses of “libel” in the law.
dignity of the members of a group stands separately from the status of

their beliefs, however offensive an attack upon the Prophet or even
upon the Koran may seem.81 The specific concern about group libel
does not encompass these things.
In general, a dignitarian rationale for laws against group defama-
tion differs from an approach based on the offense that may be taken
by the members of a group against some criticism or attack. If space
permitted, I would want to rehearse arguments I made a long time
ago, in relation to the work of John Stuart Mill and in relation to the
Salman Rushdie affair (The Satanic Verses) that being disturbed by a
shocking attack on one’s views, even one’s deepest and most cherished
religious convictions, is not something people have an interest in being
protected against.82 For the purposes of the present Article, let me just
say this: it is an advantage of the “group libel” formulation that it con-
veys this distinction. Libel and defamation generally are never orga-
nized to protect people from being offended: they are organized to pro-
tect the dignity and reputation of the persons themselves, not to
impose an aura of untouchability around their convictions.
I do not deny that the distinction is a delicate one. And I do not
mean to convey indifference to the subjective or felt aspect of assaults
on dignity. Civic dignity is not just decoration; it is sustained and
upheld for a purpose. It is an important part of my argument in Part
II that the social upholding of individual dignity furnishes the basis of
a general assurance of decent treatment and respect as people live their
lives and go about their business in public. And an assault on indi-
vidual dignity is bound to be experienced as wounding and distressing;
unless we understand that distress we do not understand what is
wrong with group defamation and why it is appropriate to prohibit it
by law.
Not only that, but in anyone’s reaction to any particular incident of
hate speech, there are going to be a whole lot of factors all mixed up.
The phenomenology of this sort of assault is complex and tangled.83 It
is not easy to differentiate the offense from the insult, or the immediate
wounding of an epithet from the perception of a threat, or the outrage
from the humiliation, or the anger from the shame of having to explain
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
81 I do not mean to deny how distressing an attack on (say) the Koran might be. The delib-
erate infliction of that distress might be wrong and unlawful in other contexts, for example as a
way of abusing detainees in the war against terrorism. See Jeremy Waldron, What Can Christian
Teaching Add to the Debate About Torture?, 63 THEOLOGY TODAY 330, 341 (2006).
82 See Jeremy Waldron, Mill and the Value of Moral Distress, 35 POL. STUD. 410 (1987), re-
printed in JEREMY WALDRON, LIBERAL RIGHTS 115 (2003); Jeremy Waldron, Too Important
for Tact, TIMES LITERARY SUPPLEMENT, Mar. 10–16, 1989, at 248, 260, reprinted in WAL-
DRON, LIBERAL RIGHTS, supra, at 134 (discussing the Salman Rushdie The Satanic Verses af-
fair in a chapter entitled Rushdie and Religion).
83 See WALDRON, LIBERAL RIGHTS, supra note 82, at 115–16.
to one’s children what is going on. The reactions are all mixed up,
and it will often seem that the law, in responding to one set of phe-
nomena, is also responding to others. I think this is likely to be true of
hate speech also. When a racial group is hatefully denigrated, there
will be fear, hurt, vehement disapproval to the point of outrage, humil-
iation, shame, anger, offense, and so on. And it will be hard to disen-
tangle; often there will be no point in doing so.
But it is important in this context to try. In American discussions
of hate speech, it is often assumed that hate speech laws are an
attempt to protect people from the immediate wounding effect of vi-
cious slurs and epithets. I have no doubt that the wounding effect of
slurs and epithets is considerable. Professor Charles Lawrence has
done a tremendous amount to convey the trauma that such wounding
words — assaultive hate speech — might cause,84 and I can imagine
an honorable legislative attempt to protect people from this and to
prohibit the infliction of this harm. But that project is different from
the dignity and reputation rationale that I am considering here.
G. Beauharnais and Libel in Light of
New York Times Co. v. Sullivan
I want to return now to Beauharnais v. Illinois, where we left it in
section I.D. It is remarkable that in the half century since it was de-
cided, Beauharnais has never explicitly been overturned by the Su-
preme Court. In a few cases, lower courts have expressed doubts
about the precedent,85 and among First Amendment scholars there is
considerable doubt whether the Supreme Court would nowadays ac-
cept the idea of group libel as an exception to First Amendment pro-
tection. Many jurists — better informed than I am in the ways of the
Justices — say they probably would not.86
Some attribute faulty reasoning to the Justices in making this pre-
diction. (That of course does not mean the prediction is false as a pre-
diction.) Anthony Lewis says that the basis of Beauharnais was un-
dermined by the 1964 Supreme Court decision in New York Times Co.
v. Sullivan,87 where the Court held that public figures cannot recover
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
84 See Lawrence, supra note 12, at 452–56.
85 See, e.g., Dworkin v. Hustler Magazine, Inc., 867 F.2d 1188, 1200 (9th Cir. 1989); Collin v.
Smith, 578 F.2d 1197, 1205 (7th Cir. 1978); United States v. Handler, 383 F. Supp. 1267, 1277–78
(D. Md. 1974); see also Garrison v. Louisiana, 379 U.S. 64, 82 (1964) (Douglas, J., concurring);
MICHAEL J. SANDEL, DEMOCRACY’S DISCONTENT 83–86 (1996). But see Smith v. Collin, 439
U.S. 916, 919 (1978) (Blackmun, J., dissenting from denial of certiorari) (providing interesting and
not unfavorable comments about the Beauharnais precedent).
86 Professor Laurence Tribe, for example, has observed that “subsequent cases seem to have
sapped Beauharnais of much of its force.” LAURENCE H. TRIBE, AMERICAN CONSTITU-
TIONAL LAW 926 (2d ed. 1988).
87 376 U.S. 254 (1964).
damages for libel unless they can prove that a false statement of fact
was made maliciously or recklessly.88 The Court argued that robust
discussion of public issues, to which the United States has “a profound
national commitment,” is bound to include “vehement, caustic, and
sometimes unpleasantly sharp attacks on government and public offi-
cials.”89 The idea was that when they take on public responsibilities,
state and federal officials have a duty to develop a thick skin and suf-
ficient fortitude to shrug off public attacks. Lewis is right that the
Court no longer regards libel per se as an exception to the First
Amendment. But it is not at all clear why the reasoning in New York
Times Co. should protect the defendant in the Beauharnais case. The
African Americans libeled collectively in Beauharnais’s “obnoxious
leaflet”90 were not public officials who had taken on the burden of of-
fice. They were ordinary citizens who may have thought that they had
a right to be protected from scattershot allegations of the most severe
criminal misconduct — the “rapes, robberies, knives, guns and mari-
juana of the negro.”91
Still, as an empirical matter, the naysayers are probably right: Jo-
seph Beauharnais’s conviction would not be upheld today.92 Lewis’s
fallacious reasoning is common, and if constitutional scholars are taken
in by it, there is no reason to suppose the present Justices are immune.
But my argument in this Article is not about the desirability of using
the group libel idea as a constitutional strategy in the United States,
but about what might be involved as a matter of principle in thinking
that group defamation is a problem, and what insight may be available
from this characterization for those willing to take the risk of appear-
ing thoughtful in these matters.
H. Who Needs Protection Against Libel?
In the New York Review of Books piece that I mentioned at the
outset, I asked: What is it that we believe now that we did not believe
in the days when we had laws against blasphemous libel and seditious
libel? And how much (if any) of that change of belief on that topic in
fact carries through to also disqualify laws protecting the reputation of
groups, laws of the kind I have been discussing in this Article?93
We know that prosecutions for attacks on Christianity faded away
much more quickly than prosecutions for political speech. The logic of
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
88 See id. at 279–80; see also LEWIS, supra note 1, at 159.
89 New York Times Co., 376 U.S. at 270.
90 Beauharnais v. Illinois, 343 U.S. 250, 287 (1952) (Jackson, J., dissenting).
91 Id. at 252 (majority opinion).
92 See, e.g., SAMUEL WALKER, HATE SPEECH: THE HISTORY OF AN AMERICAN CON-
TROVERSY 77–100 (1994).
93 See supra note 1 and accompanying text.
prosecuting atheists always sat uncomfortably with the American posi-

tion on religion. Christian belief might appear vulnerable to public
denunciations; it might seem in need of the law’s support; but it was
not clear that this was support that the law was constitutionally en-
titled to give. Often the concerns were not so much free speech as
anti-establishment considerations.94 Since Christianity could no longer
be seen as part of the organized apparatus of social control, then, vul-
nerable or not, it would just have to fend for itself in the unruly mar-
ketplace of sacred and profane ideas.
So far as political speech is concerned, there is a different story to
be told.95 In 1798, federal authority looked precarious; it was at the
mercy of public opinion and public opinion was looking well-nigh un-
governable. Public agitation led to political violence and brief upris-
ings in some of the states. George Washington was denounced as a
thief and a traitor;96 John Jay was burned in effigy;97 Alexander Ham-
ilton was stoned in the streets of New York;98 a Connecticut Federalist
was attacked with fire tongs in the House of Representatives;99 and
Republican militias armed and drilled openly, ready to stand against
Federalist armies.100 Over everything, like a specter, hung fears of
the Jacobin terror in France.101 It was by no means obvious in those
years — though it seems obvious to us — that the authorities could af-
ford to ignore venomous attacks on the structures and officers of gov-
ernment, or leave these critics’ publications unmolested in the hope
that they would be adequately answered in due course in the free
marketplace of ideas. That government could survive the published
vituperations of the governed seemed more like a reckless act of faith
than basic common sense.
However, in the two centuries since then we have learned that the
state does not need legal protection against criticism. It is strong
enough to shrug off our attacks, strong enough to dismiss our denun-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
94 See, e.g., State v. West, 263 A.2d 602 (Md. Ct. Spec. App. 1970).
95 See generally John R. Howe, Jr., Republican Thought and the Political Violence of the
1790s, 19 AM. Q. 147 (1967).
96 See GORDON S. WOOD, EMPIRE OF LIBERTY: A HISTORY OF THE EARLY REPUBLIC,
1789–1815, at 198 (2009); cf. James D. Tagg, Benjamin Franklin Bache’s Attack on George Wash-
ington, 100 PA. MAG. HIST. & BIOGRAPHY 191 (1976) (noting that an effigy of Jay was guillo-
tined, filled with gunpowder, and exploded).
97 See WOOD, supra note 96, at 198; Matthew Schoenbachler, Republicanism in the Age of
Democratic Revolution: The Democratic-Republican Societies of the 1790s, 18 J. EARLY REPUB-
LIC 237, 254 n.25 (1998).
98 See WOOD, supra note 96, at 198; Monroe Johnson, Washington Period Politics, 12 WM. &
MARY C. Q. HIST. MAG. 159, 162 (1932).
99 See WOOD, supra note 96, at 229; Eric M. Uslaner, Comity in Context: Confrontation in
Historical Perspective, 21 BRIT. J. POL. SCI. 45, 66 (1991).
100 See Howe, supra note 95, at 147.
101 See generally WOOD, supra note 96, at 239–75.
ciations as not worth the effort of suppression (though strong enough

to make the effort if it wants). In 1919, the jurist for whom the lec-
tures on which this Article is based are named began supporting free
speech ideas. In his seminal dissent in Abrams v. United States,102 Jus-
tice Holmes predicated his position on the derisory impotence of what
he called the defendants’ “pronunciamentos”103 and on a greater confi-
dence in the state’s ability to resist their destructive effect. “[N]obody,”
he said, “can suppose that the surreptitious publishing of a silly leaflet
by an unknown man, without more, would present any immediate
danger that its opinions would hinder the success of the government
arms . . . .”104
So what about group defamation? Prosecutions for seditious libel
began to seem inappropriate when we realized that the government
had become so powerful that it did not need the support of the law
against the puny denunciations of the citizenry. Does that apply to
vulnerable minorities? Is their status as equal citizens in the society
now so well assured that they have no need of the law’s protection
against the vicious slurs of racist denunciation? Prosecutions for blas-
phemous libel began to seem inappropriate when religion came to
be regarded as a private matter. Is that true of the status of embattled
minorities? Is their position in society — the respect they receive
from fellow citizens — a matter of purely private belief, with which
the law should have no concern? The state and its officials may be
strong enough, thick-skinned enough, well enough armed, or sufficient-
ly insinuated already into every aspect of public life to be able to shrug
off public denunciations. But the position of minority groups as equal
members of a multiracial, multiethnic, or religiously pluralistic society
is not something that anyone can take for granted. It is a recent and
fragile achievement in the United States and the idea that law can be
indifferent to published assaults upon this principle seems to me a
quite unwarranted extrapolation from what we have found ourselves
able to tolerate in the way of political and religious dissent.
II. WHAT DOES A WELL-ORDERED SOCIETY LOOK LIKE?

I chose as the title of the second of my Holmes Lectures a rather
technical philosophical question: what does a well-ordered society look
like? The idea of a well-ordered society is “a highly idealized” abstrac-
tion from the philosophy of John Rawls.105 Rawls wants to consider
the possibility of a society whose basic structure is regulated (and
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
102 250 U.S. 616 (1919).
103 Id. at 626 (Holmes, J., dissenting).
104 Id. at 628.
105 JOHN RAWLS, POLITICAL LIBERALISM 35 (1996); see also id. at 46.
known to be regulated) by principles of justice and inhabited by

people with an effective sense of justice; he wants to posit this concep-
tion for the purpose of asking certain questions about such an im-
agined possibility — for example, whether it could exist stably under
conditions of religious and philosophical diversity where there is no
body of deep foundational ideas affirmed by all of its citizens.106
I am not going to go into any of the technical detail of Rawls’s
theory. Instead I want to use one central element of his conception of
a well-ordered society — an element that I shall call assurance — to
cast some light on our problem of what to do about hate speech, par-
ticularly when it takes the form of group defamation.
A. The Look of Hate
A society which permits such publications may look quite different
from a society that does not. Its hoardings and its lamp-posts may be
plastered with depictions of members of racial minorities, characteriz-
ing them as bestial or subhuman. There may be posters proclaiming
that members of these minorities are criminals, perverts, or terrorists,
or leaflets saying that members of a certain race or followers of a cer-
tain religion are threats to decent people and that they should be de-
ported or made to disappear. There may be swastikas celebrating the
genocidal campaigns of the past. There may be signs indicating that
the members of some minority are not welcome in certain neighbor-
hoods or in polite society generally, or flaming symbols intended to in-
timidate them if they remain. That is what a society may look like
when group defamation is permitted. And my question is: is that what
a well-ordered society looks like?
I ask because it is assumed by many liberal constitutionalists, par-
ticularly in the United States, that a free society with a constitution
will not permit laws prohibiting speech like this, on the ground that
any such prohibition is precluded by our commitment to the First
Amendment principle of free speech. They may acknowledge that the
social environment that results from their toleration of hate speech is
likely to look unpleasant; they may say that they do not like the look of
these billboards, placards, or flaming crosses any more than we do.
But, they say, the society that permits them may still be well ordered,
even though it presents this ugly appearance, precisely because the so-
ciety is one in which racists are allowed to speak their minds like
everyone else.107 Some go further and are inclined to celebrate the di-
versity and unruliness of the various messages and speeches milling
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
106 See id. at 35–40.
107 See, e.g., Dworkin, supra note 10, at vi–vii. I discuss Dworkin’s argument at great length in
sections III.C–III.E.
around in the marketplace of ideas. They love this richness and unti-
diness: let a thousand flowers bloom, they say, even the poisonous
ones. For of course some of the ideas are foul and distasteful. But if
you blur your eyes a bit, you can abstract away from the distasteful
content, and what you see is a glorious splash of moving, variegated
color — ideas interacting openly and unpredictably with one another.
And that, they say, is surely a feature of a well-ordered society (even if
the men, women, and children who are the targets of the foul and dis-
tasteful messages have difficulty maintaining this lofty perspective).
Of course, if the racist appearances correspond to a racist reality,
everything is different. If the signs saying “Christians only” are ac-
companied by discriminatory practice against Jews, if Muslims are
beaten up in the street, if minority members are not protected against
the discrimination advocated in racist posters, or if those in power
treat people in the unequal and degrading ways that the racist leaflets
call for — then there is something to worry about. And that would
show that the society is not well ordered. But if it is just signage, they
say, there is no cause for concern.
That is the position I want to test, by focusing on this issue of ap-
pearances. My question — what does a well-ordered society look like?
— is not a coy way of asking what makes a society well ordered or
what a well-ordered society is like. I am interested in how things lit-
erally look, the visible environment. How important is the look of
things? Is it unimportant compared to how things actually are? Or is
it an important part of how things actually are? And if it is an impor-
tant part of how things are, what in particular should we be looking
for? The colorful, unruly diversity of a free market in ideas? Or the
absence of visible features that are at odds with the fundamental
commitment to justice with which a well-ordered society is supposed
to be imbued?
If it is the latter, then can we use this as a way of understanding re-
strictions on hate speech and group defamation? In our interpretive
characterization of these laws where they exist, can we say that they
are among the ways in which real-world societies, in Europe for ex-
ample, try to make themselves more well ordered (better ordered) than
they would otherwise be?
My question is not about Rawls the man.108 I want to make use of
a Rawlsian idea, but run with it in a direction that may be quite dif-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
108 I am not asking this Rawlsian question in order to get at John Rawls’s own views in the free
speech-hate speech debate. What Rawls says about free speech, set out mainly in a lecture en-
titled The Basic Liberties and Their Priority, RAWLS, supra note 105, at 289–371, is not particu-
larly interesting for our purposes. It does not address the specific issue of hate speech or group
libel at all. And it does not follow up on the implications of Rawls’s own characterization of pub-
lic knowledge and assurance in a well-ordered society in the way that I want to. Also, it is a bit
ferent from the direction in which Rawls would have run. Nor am I
asking whether those who enact hate speech legislation are appealing
to Rawlsian ideas. I am asking whether hate speech regulation
amounts in effect to an embrace of one element of Rawls’s conception
of a well-ordered society — an element on which I am going to focus
continually in this Part. That element is the idea of assurance — the
assurance of support and dignitarian vindication that a well-ordered
society offers to all of its members, especially the most vulnerable.
One of the most important things that Rawls says about a well-
ordered society is that “everyone accepts, and knows that everyone else
accepts, the very same principles of justice.”109 “[T]his knowledge,” as
he puts it, “is publicly recognized.”110 That is what I want to concen-
trate on: the assurance of a generalized commitment to the fundamen-
tal elements of justice and dignity that a well-ordered society is sup-
posed to furnish to its citizens as part of “the public culture of a
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
confusing because unlike almost everything else in Political Liberalism, the chapter on “The
Basic Liberties and Their Priority” is focused on real-world constitutions, with all their flaws and
messiness, rather than on the more utopian abstraction of a well-ordered society. Rawls says that
one possible method for developing a list of basic liberties is to “survey the constitutions of demo-
cratic states and put together a list of liberties normally protected, and . . . examine the role of
these liberties in those constitutions which have worked well.” Id. at 292–93. In the context of
free speech, then, his analysis is grounded in the actual role of this concept in constitutional doc-
trine. See id. at 340–63. Rawls draws mainly upon the American experience, though he has ac-
knowledged elsewhere that, as things stand, the United States certainly cannot be regarded as a
well-ordered society. See JOHN RAWLS, Kantian Constructivism in Moral Theory, in COL-
LECTED PAPERS 303, 355 (Samuel Freeman ed., 1999).
There is some speculation in Rawlsian literature on what Rawls’s view on hate speech
might have been, or what the implications are for this issue or other more abstract views that he
did hold. See Richard Delgado & Jean Stefancic, Four Observations About Hate Speech, 44
WAKE FOREST L. REV. 353, 368 (2009); Richard H. Fallon, Jr., Individual Rights and the Powers
of Government, 27 GA. L. REV. 343, 351–52 (1993); T.M. Scanlon, Adjusting Rights and Balancing
Values, 72 FORDHAM L. REV. 1477, 1484–86 (2004); R. George Wright, Dignity and Conflicts of
Constitutional Values: The Case of Free Speech and Equal Protection, 43 SAN DIEGO L. REV.
527, 553–54 (2006). But that discussion is mostly inconclusive.
The closest Rawls gets to the issue of hate speech is in a discussion of seditious libel, where
he insists — in line with American free speech orthodoxy — that a well-ordered society will be
one in which anything and everything may be published, even things which tend to question the
basic principles of a given society. Subversive advocacy, he says, must be permitted. But I am
not sure whether he thinks this should extend even to advocacy against the fundamentals of jus-
tice — for example, to public advocacy for the exclusion or subordination of a given group, or
their disenfranchisement, segregation, enslavement, concentration, or deportation. Rawls does not
consider the status of speech that in its content and tone runs counter to the assurances that citi-
zens are supposed to have of one another’s commitment to equality. I suspect he would not have
dissented from First Amendment orthodoxy in this regard; certainly that is what his expressed
admiration of the work of Harry Kalven, see RAWLS, supra note 105, at 342–44, intimates
(though it has to be said that Kalven’s own discussion of group libel in KALVEN, supra note 18, at
7–64, is nuanced, thoughtful, and complicated).
109 RAWLS, supra note 105, at 35.
110 Id. at 66 (“[C]itizens accept and know that others likewise accept those principles, and this
knowledge in turn is publicly recognized.”).
democratic society.”111 I want to take the measure of this assurance

and, to the extent that it is important, consider how comfortable we
should be with public and semi-permanent manifestations of racial
and ethnic hatred as visible aspects of the civic environment.
B. Hatred and Law in a Well-Ordered Society
Will hate speech be tolerated by law in a well-ordered society? We
have already considered one response: yes, it will be tolerated as part
of the energizing diversity of a free market in ideas. Another response
might go as follows: a society cannot be well ordered if people are ad-
vocating racial and religious hatred. The idea of a well-ordered socie-
ty is the idea of a society fully and effectively governed by a concep-
tion of justice. In technical terms, it is strict compliance theory rather
than partial compliance theory.112 On this account, a society with suf-
ficient rancor and division to generate hate speech cannot be a well-
ordered society.
Compare what Rawls says about illiberal religions. Intolerant reli-
gions, Rawls says, “will cease to exist in the well-ordered society of po-
litical liberalism.”113 A society cannot be well ordered unless religions
that demand the suppression of other religions, religions that insist on
establishment, or religions that demand the adoption of comprehensive
conceptions of the good by the whole society have, so to speak, died
out. So the question of what to do about such religions in a well-
ordered society will not arise. And similarly, a society cannot have be-
come well-ordered unless bigots and racists have given up their mis-
sion and accepted the basic principles of justice and equal respect.
Thus, the question of what to do about hate speech and group defama-
tion in a well-ordered society does not arise. A well-ordered society
will definitely not look racist; it will not present or exhibit the offen-
sive manifestations that I spoke about earlier. But this will not be on
account of prohibitory laws. It will be because citizens have no wish
to express themselves in these terms.
Taking this response one step further, our well-ordered respondent
may say: so, even if it is true that Rawls’s ideal society would not be
festooned with racial signage, Islamophobic leaflets, and ethnically
prejudiced billboards, nothing of interest follows for the debate we are
conducting. A well-ordered society would not need such laws. Maybe
the lesson for us, in our much-less-than-well-ordered society, is that we
must hope that hate speech just withers away, not because of coercive
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
111 RAWLS, Kantian Constructivism in Moral Theory, supra note 108, at 355.
112 For these terms, see JOHN RAWLS, A THEORY OF JUSTICE 8 (rev. ed. 1999). For the idea
of well-ordered society as part of strict compliance theory, see RAWLS, Kantian Constructivism in
Moral Theory, supra note 108, at 355.
laws limiting freedom of speech, but because of changes of attitude

and changes of heart, including — not least — changes brought about
by effective answers to hate speech in the free marketplace of ideas.
I think that this response — I am not blaming it on anyone in par-
ticular — is misconceived at a number of levels. Consider again the
case of intolerant religions. They do not feature in a well-ordered so-
ciety. Why? Presumably because they have died out. But Rawls says
a little more than that; he says that the basic institutions of a just so-
ciety “inevitably encourage some ways of life and discourage others, or
even exclude them altogether.”114 That is an ambiguous formulation:
what does “discourage” mean here, in terms of the operation of institu-
tional arrangements? And what does it mean to exclude certain ways
of life altogether?
One thing is for sure. We should not think of a well-ordered socie-
ty as a utopian fantasy, in which laws are unnecessary because every-
one’s attitudes are now utterly just. No one supposes that law can be
eliminated from a well-ordered society or that we can drop the laws
about murder or burglary because, by definition, no one in such a so-
ciety would ever be motivated to engage in those crimes. Rawls’s so-
ciety is not utopian in that fantasy sense; it is steadfastly located in the
circumstances of justice, which include among other things the subjec-
tive circumstances of anxiety and limited strength of will among its
citizens.115 Rawls himself gives a fine account in A Theory of Justice
of the role of law in a well-ordered society:
[E]ven in a well-ordered society the coercive powers of government are to
some degree necessary for the stability of social cooperation. For although
men know that they share a common sense of justice and that each wants
to adhere to the existing arrangements, they may nevertheless lack full
confidence in one another. . . . [T]he existence of effective penal machinery
serves as men’s security to one another.116
Maybe in a well-ordered society “sanctions . . . never need to be
imposed.”117 But that does not mean that their existence or the laws
providing for them are unnecessary or redundant. Apart from any-
thing else, laws have an important expressive as well as coercive func-
tion; and one would expect that expressive function to be at the fore in
a well-ordered society, particularly in connection with the public and
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
114 Id. at 195.
115 For a discussion of the circumstances of justice, see RAWLS, A THEORY OF JUSTICE, supra
note 112, at 109–112.
116 Id. at 211.
117 Id.
visible assurance of just treatment that a society is supposed to pro-

vide to all of its members.118
In any case, even if a well-ordered society could dispense with laws
prohibiting group defamation, it would be spectacularly dumb to infer
from this that the societies we know must be prepared to dispense
with those laws as a necessary way of becoming well-ordered. Socie-
ties do not become well ordered by magic. The expressive and dis-
ciplinary work of law may be a necessary ingredient for the change of
heart on the part of its racist citizens that a well-ordered society pre-
supposes. And anyway, as with all issues of justice, the necessity of
such laws is a matter of the goods to be secured and the likelihood that
they can be secured in the absence of legal intervention. If, as I am
going to argue, the good to be secured is a public good, a general and
diffuse assurance to all the inhabitants of a society concerning the
most basic elements of justice, then it is natural to think that law
would be involved — both in its ability to underpin the provision of
public goods and in its Durkheimian ability to express and communi-
cate common commitments. This is particularly likely to be true in the
case of societies, like European societies (and I think also the United
States), which have not yet entirely shaken off histories of murderous
racist oppression.
C. Political Aesthetics
What should a well-ordered society look like? I suppose we could
ask, with equal sense: what should a well-ordered society sound like?
We might have our positive impression: say, the flat steady drone of an
interminable but well-ordered exercise of what Rawls calls public rea-
son — respectful and mutually comprehensible speech in matters of
common concern, with every third word being “autonomy,” “equality,”
or “liberty.” And we might contrast that impression with darker
sounds: the tread of marching feet and ominous chants, or perhaps the
genocidal radio broadcasts of Radio Télévision Libre des Mille Col-
lines (RTLM) in Rwanda that my NYU colleague, Professor Ted Me-
ron, American representative on the International Criminal Tribunal
for Rwanda, sought to privilege as free speech in his dissenting opinion
in the Nahimana case.119
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
118 See ROGER COTTERRELL, ÉMILE DURKHEIM: LAW IN A MORAL DOMAIN 152 (1999).
For an application of Durkheim’s conception in relation to hate speech, see THOMAS DAVID
JONES, HUMAN RIGHTS: GROUP DEFAMATION, FREEDOM OF EXPRESSION AND THE LAW
OF NATIONS 88–89 (1998).
119 See Nahimana v. Prosecutor, Case No. ICTR-99-52-A, Partly Dissenting Opinion of Judge
Meron ¶¶ 4–5, 9–21 (Int’l Crim. Trib. for Rwanda Trial Chamber I Nov. 28, 2007). But cf. Susan
Benesch, Vile Crime or Inalienable Right: Defining Incitement to Genocide, 48 VA. J. INT’L L.
485 (2008) (articulating a more complex view).
I said in Part I that an emphasis on speech is an emphasis on the

ephemeral. There I had in mind the occasional angry and politically
incorrect use of one or another racial epithet, and I contrasted that —
using the figure of slander versus libel — with the relatively enduring
expression of public signage or the published word. But it is true, on
the other hand, that the accepted vocabulary of a culture can become
part of its established environment. And certainly the broadcast word
can be as much a matter of enduring concern as the printed word, es-
pecially when it insistently and repeatedly demonizes a minority as
cockroaches and vermin.
So there is the visible and the audible. We might round out the
picture with the emphasis by Professors Richard Delgado and Jean
Stefancic on tangible aspects of a society’s self-presentation. Their
book Understanding Words that Wound has a chapter entitled “When
Hate Goes Tangible: Logos, Mascots, Confederate Flags, and Monu-
ments.” And the authors say this:
[S]tatues, monuments, and the like . . . perhaps because they are intended
to be seen by a large audience, . . . contribute to a climate of opinion that
is injurious to members of the group singled out. . . . [T]angible symbols
have a quality that words — at least of the spoken variety — do not: They
are enduring. Words disappear as soon as they are spoken. They may
resonate in the mind of the victim, causing him or her to recall them over
and over again. But a flag [or a] monument . . . is always there to remind
members of the group it spotlights of its unsolicited message.120
Delgado and Stefancic do not advance the discussion much beyond
this in their short book; but I shall try to proceed in the spirit of the
concerns that they raise.
A general consideration of what a well-ordered society looks like,
sounds like, smells like, and feels like to the touch, might be an exer-
cise in political aesthetics. It is the sort of thing we find in Edmund
Burke’s observation that “[t]o make us love our country, our country
ought to be lovely”121 and in his talk about “the pleasing illusions,
which [make] power gentle, and obedience liberal, . . . the superadded
ideas, furnished from the wardrobe of a moral imagination, which the
heart owns, and the understanding ratifies, as necessary to cover the
defects of our naked shivering nature, and . . . raise it to dignity in our
own estimation.”122 Political aesthetics invite us to think about such
things as monuments, cenotaphs, public statues, public architecture,
flags and banners, and the ceremonies (coronations, inaugurations, ar-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
120 RICHARD DELGADO & JEAN STEFANCIC, UNDERSTANDING WORDS THAT WOUND
142 (2004).
121 EDMUND BURKE, REFLECTIONS ON THE REVOLUTION IN FRANCE 78 (L.G. Mitchell
ed., Oxford Univ. Press 1999) (1790).
122 Id. at 77.
mistice days, and the like) for public or political events and their set-
tings, choreography, and costumes. That is political aesthetics — and I
think we need to pay more attention to this topic in political philoso-
phy than we do.123
Notice that the examples I have mentioned are mostly a matter of
official or publicly sponsored appearances. But in the case of hate
speech, it is appearances sponsored by private persons, not the state,
that we are concerned about. There is of course sometimes a messy
interface between public and private, which shows up (for example) in
the United States in the First Amendment jurisprudence of church and
state: I mean issues about the presence of religious symbols — crosses,
crèches, menorahs, depictions of the Ten Commandments — in the
town square on public property. We know that it is possible for a so-
ciety to look religious, without in any official or governmental sense
being religious.124 There may be temples, steeples, churches, mosques,
and synagogues as far as the eye can see, and many of us think this
can be so without any message being conveyed that the society as a
whole is committed to any particular religion. All this may be compat-
ible with a society being well ordered in the sense of being religiously
neutral.
Balancing private religious expression with society’s perceived en-
dorsement is not a simple matter. Some liberals express concern about
the appearance in public of private individuals dressed in certain
ways; think of the controversy in countries like France about Muslim
women appearing in public with headscarves or veiled or with the full
covering of the burqa.125 President Nicholas Sarkozy and others who
call for a ban on the burqa are also interested in our theme: what
should a well-ordered society look like? For them, the interest is the
appearance that people present to one another. It is as though they are
using the old idea of sumptuary laws: people must dress in a way that
connotes the dignity of a free person, not in a way that intimates their
subordination. (The burqa might be compared to a portable private
realm carted around in public — like an Edwardian bathing machine
— as though women may appear in public only by remaining in effect
in the private realm.) I am not a supporter of the proposal to ban the
burqa. But the arguments that are used on the other side are not a
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
123 For a fine account of the presence and importance of monuments in modern society, see
generally AJUME H. WINGO, VEIL POLITICS IN LIBERAL DEMOCRATIC STATES (2003).
124 This paradox was noted in Karl Marx, ‘On the Jewish Question’ (1844), reprinted in NON-
SENSE UPON STILTS: BENTHAM, BURKE, AND MARX ON THE RIGHTS OF MAN 137, 137–38
(Jeremy Waldron ed., 1987).
125 Doreen Carvajal quotes President Nicholas Sarkozy as saying, “The burqa . . . is a sign of
the subjugation, of the submission, of women. . . . I want to say solemnly that it will not be wel-
come on our territory.” Doreen Carvajal, Sarkozy Backs Drive To Eliminate the Burqa, N.Y.
TIMES, June 23, 2009, at A4.
million miles from the argument that I am pursuing in this Article.

What individuals do, how they present themselves, can add up to an
impression that matters: maybe not the burqa, but the appearance of
large numbers of masked men in white sheets and pointy hats
is a problem, protestations about purely private dress codes
notwithstanding.
D. Assurance and Security
Why does it matter what a well-ordered society looks like? Why do
appearances count? I believe the answer has to do with assurance,
with conveying to people a sense of security in the enjoyment of their
most fundamental rights. And this is what I want to draw from
Rawls’s account. Rawls insists throughout his work that a well-
ordered society is one “in which everyone accepts, and knows that
everyone else accepts, the very same principles of justice.”126 That
shared knowledge plays an important role in everyone’s life.
The content of the relevant assurances may vary. In Rawls’s philo-
sophical ideal, a well-ordered society is defined by reference to the
whole detailed array of principles that characterize his conception of
justice as fairness: what people know and assure each other of,
people’s joint allegiance to the difference principle, the details of the
first principle, the exact formulation of the second principle, the bal-
ance between the difference principle and the equal opportunity prin-
ciple, and the various priority rules. Rawls is of course right to note
that one of the reasons why we cannot describe the United States as a
well-ordered society is that there is nothing approaching a consensus
about justice at this level of detail. But in the real world, when people
call for the sort of assurance to which hate speech laws might make a
contribution, it is not on the controversial details of justice. Instead, it
is on some of the most elementary fundamentals — that all are equally
human and have the dignity of humanity, that all have an elementary
entitlement to justice, and that all deserve protection from the most
egregious forms of violence, exclusion, and subordination.127 Hate
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
127 My reference here to the fundamentals of justice is similar to, but not quite the same as,
Rawls’s idea of “constitutional essentials.” Id. at 214, 227. The idea is that some claims of justice
are based on or presuppose others; some represent controversial developments of or extrapolations
from others. The fundamentals of justice are the claims that lie at the foundations of these deri-
vations and controversies. They include propositions establishing everyone’s right to justice and
elementary security, everyone’s claim to have their welfare counted along with everyone else’s
welfare in the determination of social policy, and everyone’s legal status as a rights-bearing mem-
ber of society. They also include repudiations of particular claims of racial, sexual, and religious
inequality that have historically provided grounds for denying these rights. See Jeremy Waldron,
Basic Equality (N.Y. Univ. Sch. of Law Pub. Law & Legal Theory Research Paper Series, Work-
ing Paper No. 08-61, 2008), available at http://ssrn.com/abstract=1311816 (discussing basic equali-
speech or group defamation involves the express denial of these fun-

damentals so far as some group in society is concerned. It seems to me
that if we are imagining a society on the way to becoming well or-
dered, these fundamentals must be secured in people’s minds by public
guarantees, even if we are not yet in a position to secure a more de-
tailed consensus on justice.
I referred earlier to political aesthetics, the “decent drapery” cele-
brated by Edmund Burke.128 But when we talk about the public
manifestation of a society’s commitment to fundamental principles of
justice, we are not just talking about justice displayed for the sake of
an impressive or pretty show (in the way that a society might display
the might of its military resources, the splendor of its culture, or the
pride of its athletes). We are talking about a display that matters prac-
tically to individuals. It matters to them in their reliance on the prin-
ciples of justice in the ordinary course of their lives, and in the security
with which they enjoy that reliance. In a well-ordered society, where
people are visibly impressed by signs of one another’s commitment to
justice, everyone can enjoy a certain assurance as they go about their
business. People know that when they leave home in the morning they
can reasonably count on not being discriminated against, humiliated,
or terrorized. They feel secure in the basic rights that justice defines;
they can face social interactions without the elemental risks that inter-
action would involve if one could not count on others to act justly.
We can put the same points negatively. When a society is defaced
with anti-Semitic signage, burning crosses, or defamatory racial leaf-
lets, that sort of assurance evaporates. A vigilant police force and Jus-
tice Department may still keep people from being attacked or ex-
cluded, but people no longer have the benefit of a general public
assurance to this effect, provided and enjoyed as a public good, fur-
nished to all by all. There is security in such public knowledge for the
proper pride — holding one’s head upright — that we associate with
human dignity. President Lyndon Johnson once gave this reason for
the moral necessity of the Civil Rights Act: “A man has a right not to
be insulted in front of his children.”129 The security that people look
for is security against the soul-shriveling humiliation that accompanies
the manifestation of injustice in society. In the landmark Canadian
case, R. v. Keegstra,130 Chief Justice Brian Dickson said this about the
effect that public expressions of hatred may have on people’s lives:
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
ty, as distinct from various egalitarian policies). For further discussion of this idea of the funda-
mentals of justice, see infra pp. 1646–47.
128 See supra p. 1624.
129 DAVID BROMWICH, POLITICS BY OTHER MEANS: HIGHER EDUCATION AND GROUP
THINKING 157 (1992) (internal quotation marks omitted).
130 [1990] 3 S.C.R. 697 (Can.).
The derision, hostility and abuse encouraged by hate propaganda . . . have

a severely negative impact on the individual’s sense of self-worth and ac-
ceptance. This impact may cause target group members to take drastic
measures in reaction, perhaps avoiding activities which bring them into
contact with non-group members or adopting attitudes and postures di-
rected towards blending in with the majority. Such consequences bear
heavily in a nation that prides itself on tolerance and . . . respect for the
many racial, religious and cultural groups in our society.131
The point of the visible self-presentation of a well-ordered society,
then, is not just aesthetic; it is the conveying of an assurance to all citi-
zens that they can count on being treated justly.
This assurance is connected in important ways to the values of
dignity and reputation that were discussed in Part I of this Article. A
person’s dignity is not just a decorative fact about him or her. It is a
matter of status, and as such, it is in large part normative: it is some-
thing about a person that commands respect from others and from the
state. Moreover, one holds a certain status not just when one happens
to have a given set of rights or entitlements, but also when the recogni-
tion of those rights or entitlements is basic to how one is treated. So it
is with the fundamentals of social reputation. We accord people digni-
ty on account of the sorts of beings human individuals are, and we are
gravely concerned when it is said publicly that some people, by virtue
of their membership in a racial, ethnic, or religious group, are not real-
ly beings of that kind and so are not entitled to that dignity in one way
or another. Such hateful claims are not just anthropological specula-
tion: they intimate that people should expect to be treated in a degrad-
ing manner if the person making the hateful claim (and the fellow
travelers that he is appealing to) have their way.
Does this mean that individuals are required to accord equal re-
spect to all their fellow citizens, that they may not esteem some and
despise others? That proposition seems counterintuitive. Much of our
moral and political life involves differentiation of respect. People re-
spect those who obey the law and do good while withholding their re-
spect from those they regard as wrongdoers. Democrats respect Presi-
dent Barack Obama, while some conservatives despise him; most
Republicans have a great deal of respect for former President George
W. Bush, while some of his political opponents want him tried as a
war criminal. Many people despise bankers after the recent financial
crisis. Are we now saying that these distinctions of respect are imper-
missible and that everyone has a duty to respect everybody else? Are
we saying that no one is permitted to express his or her contempt for
anyone else? Not quite. It is important to distinguish between two
senses of respect that might be in play here: what Stephen Darwall has
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
131 Id. at 746.
called “appraisal respect” (in which one’s estimation of people varies

by their merits, their virtues and vices, their crimes, their views, and so
on) and “recognition respect” (which is fundamental to the dignity of
persons and invariant in the face of differential merit, even command-
ing how people are to be treated when they are guilty of terrible
crimes).132 It is recognition respect that we are talking about here; and
the fact that in private we might subscribe to varying estimations of
different persons as a matter of appraisal respect — something the
government is also entitled to do — does not show that we may not
reasonably be required to play our part in society by according the ba-
sics of recognition respect to each person.
How exactly is this assurance that we seek to provide for one
another conveyed? I do not think Rawls imagines that there will be
billboards proclaiming the difference principle or the list of basic liber-
ties. The creepy totalitarian flavor of that makes us uneasy, and
rightly so. There may be some affirmative efforts: I think of the public
proclamation of a new constitution, like the South African Constitu-
tion, seeking to focus everyone’s attention on the fact that they now
have certain rights; or just the mundane business of pamphlets and
advertisements ensuring that people know their rights and how to
claim them. I saw a sign recently on the New York subway, in English
and Spanish, telling people that they do not have to put up with un-
wanted sexual touching in a crowded subway car.
Mostly, however, the assurance is implicit, as though the underlying
status of each person as a citizen in good standing goes without saying.
Various forums of social, political, and commercial interaction are just
open to all as a matter of course; no one has to say “Muslims Wel-
come” or “African-Americans Allowed.” It is tremendously important
that the assurance be conveyed in this implicit and ubiquitous way so
that it can be effortlessly accepted and that people who might other-
wise feel insecure, unwanted, or despised can put that insecurity out of
their minds and concentrate on what matters to them in social interac-
tion — its pleasures and opportunities.
At the same time, the implicitness of this assurance makes it tre-
mendously vulnerable. For example, suppose a spate of discriminatory
signs appear — maybe they intimate a real intention to discriminate or
maybe they do not — but suddenly the stakes have changed for those
to whom they are directed. This helps us see what hate speech is
about. The point of the hateful displays that we want to regulate is
not just autonomous self-expression on the part of the bigots. The dis-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
132 See STEPHEN DARWALL, THE SECOND-PERSON STANDPOINT: MORALITY, RESPECT,
AND ACCOUNTABILITY 122–23 (2006). See generally Stephen L. Darwall, Two Kinds of Respect,
88 ETHICS 36 (1977).
plays target the assurance on which members of vulnerable minorities

are supposed to be able to rely. Their point is to negate the implicit
assurance that a society offers to the members of vulnerable groups —
to undermine it, call it in question, and taint it with visible expressions
of hatred, exclusion, and contempt. And so it begins: what was implic-
itly assured is now visibly challenged, and now there is a whole new
set of calculations for a minority member setting out to take a walk
with his family.
E. Public Goods
Provision of the assurance that I am talking about is like a public
good, albeit a silent one. It is implicit rather than explicit, but none-
theless real — a pervasive, diffuse, ubiquitous, general, sustained, and
reliable underpinning of people’s basic dignity and social standing,
provided by all for all. A well-ordered society, it seems to me, has an
interest in the provision of this public good, that is, in the furnishing of
this assurance and in the recognition and upholding of the basic digni-
ty on which it is predicated. It is a public good and part of the public
order, but it is not ultimately a communal good, enjoyed collectively.133
Instead, like street lighting, it is a public good whose benefit redounds
ultimately to individuals — those whose dignity is affirmed when its
social underpinnings might be otherwise in question; those who rely
implicitly on a sense that there does not have to be anything specific
on which to rely. But unlike street lighting, which can be provided by
a central utility company, the public good of assurance depends on and
arises out of what thousands or millions of ordinary citizens do singly
and together. It is, as Rawls puts it, a product of “citizens’ joint activi-
ty in mutual dependence on the appropriate actions being taken by
others.”134 It may not affirmatively require a great deal of the ordinary
citizen; that is part of what it means that this is an implicit good. But
just because assurance is a low-key background thing, the prime re-
sponsibility for its provision that falls upon the ordinary citizen is to
refrain from doing anything to undermine it or to make the furnishing
of this assurance more laborious or more difficult. And that is the ob-
ligation that hate speech laws or group defamation laws are enforcing.
Those who publish or post expressions of contempt and hatred of
their fellow citizens, those who burn crosses, and those who scrawl
swastikas are doing what they can to undermine this assurance. Their
actions may not seem all that significant in themselves; an isolated in-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
133 For the distinction between public goods whose ultimate payoff is collective and public
goods that benefit individuals, see Jeremy Waldron, Can Communal Goods Be Human Rights?, in
LIBERAL RIGHTS, supra note 82, at 339, 354–59. Some public goods may have both aspects. See
Jeremy Waldron, Safety and Security, 85 NEB. L. REV. 454, 500–02 (2006).
cident here, a forlorn Nazi procession there, some ratty little racist
leaflet. But precisely because the public good that is under attack is
provided in a general, diffuse, and implicit way, the flare-up of a few
particular incidents can have a disproportionate effect on the quality
of assurance in society. I will say a little more at the conclusion of this
Part about the social and historical context. But consider this observa-
tion by William Peirce Randel, a historian of the Ku Klux Klan, about
isolated instances of cross-burning: “Such is the symbolic power of the
fiery cross that people in many parts of the country still talk in sub-
dued voices about the cross that was burned one night years ago in the
field across the road or on a local hilltop.”135 Randel also added this
about the isolated instances of the burning cross:
What [a cross-burning] is commonly taken to mean is that neighbors one
sees every day include some who are Klan members, and that Klaverns
supposedly extinct are only dormant, ready to regroup for action when the
Klan senses that action is needed. It casts a shadow on many a neighbor-
hood to know that it harbors a potentially hostile element which at any
moment may disrupt the illusion of peace.136
Hate speech does not just seek to undermine the public good of
implicit assurance. It also seeks to establish a rival public good as the
wolves call to one another across the peace of a decent society. The
publication of hate speech, the appearance of these symbols and
scrawls in places for all to see, is a way of providing a focal point for
the proliferation and coordination of the attitudes that these actions
express, a public manifestation of hatred by some people to indicate to
others that they are not alone in their racism or bigotry. Frank Collin,
the leader of the Nazis who sought to march through Skokie, said, “We
want to reach the good people — get the fierce anti-Semites who have
to live among the Jews to come out of the woodwork and stand up for
themselves.”137 Accordingly, hate speech regulation aims not only to
protect the public good of dignity-based assurance, but also to block
the construction of this rival public good.
Some object that such laws simply drive hate underground. But
that is the whole point — to convey the sense that the bigots are iso-
lated, embittered individuals, rather than to permit them to contact
and coordinate with one another in the enterprise of undermining so-
ciety’s most fundamental principles. True, there is a cost to this: such
laws may drive racist sentiment into spaces where it cannot easily be
engaged. As I shall suggest in Part III, however, hate speech laws op-
erate best in an environment when the time for active debate on an is-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
135WILLIAM PEIRCE RANDEL, THE KU KLUX KLAN: A CENTURY OF INFAMY 224 (1965).
136Id.
137PHILIPPA STRUM, WHEN THE NAZIS CAME TO SKOKIE: FREEDOM FOR SPEECH WE
HATE 15 (1999) (internal quotation marks omitted).
sue is over and so the benefits of engaging with the racists (except for
redemptive purposes) are low.138 Thus, while isolating racist sentiment
from public discourse is a potential problem, introducing hate speech
legislation at the proper time can help to mitigate this concern.
F. The Role of Law and the Role of Individuals
I have said that the good of dignity-based assurance is a public
good provided to all by all and that unlike the benefit of street lighting
it cannot be provided by a central utility. I am sure some readers will
balk at this and say that it is a mistake for me to saddle private citi-
zens with what is surely a responsibility of government. Is not the
manifestation of commitment by government much more important
than the manifestation of the attitudes of citizens to one another? If
laws against discrimination are upheld and if people are confident that
they will be upheld, what does it matter what signs citizens display?
If laws protecting people from being driven out of their neighborhoods
are upheld and if people are confident that they will be upheld, what
does it matter whether the odd cross is burned on somebody’s lawn?
If the laws protecting people against violence and mass murder are
upheld and if people are sure that they will be upheld, what does it
matter what it says on the placards that neo-Nazis carry through Jew-
ish neighborhoods in an Illinois suburb? It is law enforcement that
matters, not the cardboard signs: that is the objection.
But this objection is based on a false contrast. In no society is the
state able to offer these guarantees on its own without a complementa-
ry assurance that ordinary citizens will play their part in the self-
application of the laws.139 Think of the administration of antidiscri-
mination laws. The law does not have the resources to provide an
armed escort for every minority member who wants to approach and
enter a school, or university, or other public accommodation without
fear of being turned away and humiliated on racial grounds. The
spectacle of the National Guard being turned out to desegregate a
school in Little Rock, Arkansas, in 1957 showed us exactly what such
an escort would look like. States do not have the coercive resources to
do this in any but a very few cases, and anyway, to proceed under
armed escort is hardly a satisfactory indication of assurance. Military
enforcement cannot be the basis of the effortless, implicit, and perva-
sive assurance that people need for the conduct of their ordinary lives.
Even routine enforcement efforts by the Department of Justice against
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
138See section III.E, pp. 1646–49.
139For a discussion of the idea of self-application, see HENRY M. HART, JR. & ALBERT M.
SACKS, THE LEGAL PROCESS: BASIC PROBLEMS IN THE MAKING AND APPLICATION OF
LAW 120–22 (William N. Eskridge, Jr. & Philip P. Frickey eds., 1994).
routine discrimination can handle only a handful of cases. By and

large, the law has to rely in this area — as in almost every area — on
self-application by ordinary citizens. And that means that any citizen
who relies upon the law is relying indirectly on the voluntary coopera-
tion of fellow citizens.
Conversely, part of the concern about the public expression of rac-
ist attitudes is that these expressions are intimations that certain mem-
bers of the public (and those they are trying to influence) will not play
their necessary part in the administration of the laws, if they can get
away with it. What is more, as I said in the previous section, they are
playing a competing assurance game, using public and semi-permanent
displays to assure those who are inclined not to play their part in
upholding laws against violence and discrimination that they are not
alone, that there are plenty of others like them.
Ronald Dworkin takes the view that all this is a matter for the
government to handle. The government is the entity that is required
to display equal concern and respect for all its citizens. But, he says,
the citizens themselves do not share an identical obligation: on most
occasions they are permitted to show respect for some and concern for
others — respect for their parents and concern for their children that
differ from the concern and respect that they manifest to strangers.140
There may be something in this division of private and public respon-
sibility. But as stated it is too simple. Government is not an entity
separate from the people, not in the formation of its policies or in the
enactment of its laws, and certainly not in the discharge of its distinc-
tive responsibilities. The discharge of some governmental responsibili-
ties is impossible without the whole-hearted cooperation of members of
the public, and the discharge of other public responsibilities is vulner-
able to what private people do in public. The responsibility of provid-
ing the dignity-based assurance that Rawls tells us will be a feature of
a well-ordered society is of the latter sort. We must not be misled into
treating hate speech and group defamation as essentially private acts
with which governments are perversely trying to interfere in the spirit
of mind-control. Hate speech and group defamation are actions per-
formed in public, with a public orientation, aimed at undermining
public goods. We may or may not be opposed to their regulation, but
we need at least to recognize them for what they are.
G. Transition and Assurance
In this Part of the Article, I have taken up Rawls’s suggestion that
the members of a well-ordered society ought to be able to rely upon
public assurance of one another’s commitments to justice and that this
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
140 See RONALD DWORKIN, LAW’S EMPIRE 295–301 (1986).
reliance should be public knowledge, publicly conveyed. I have ar-

gued that one way of thinking of the purpose of group defamation
laws is that they protect this assurance against egregious forms of den-
igration and coordinated defiance. People need this assurance, and
they need to be protected against displays and manifestations whose
point is to undermine the assurance and to begin constructing an as-
surance of exactly the opposite — an assurance that, whatever the
Constitution and the laws say, those who discriminate or those who try
to drive minorities out of majority neighborhoods will be in good com-
pany. I argue that people have a responsibility to participate in the
provision of these goods, at least to the extent of not participating in
undermining them, and that certainly society is permitted to enforce
that responsibility to protect this vital good.
I suspect that one could make a case along these lines, in the ab-
stract, for any society. In any society, people are likely to stand in need
of the assurances which, on Rawls’s account, it is the task of a well-
ordered society to provide. But the case becomes particularly pressing
when we think not just of the abstractions of political philosophy, but
also of the prospects of real-world societies becoming anything like
well ordered. For us, the issue is not only the abstract need for assur-
ance that people might have even in the best of social circumstances,
but also a need for assurance in relation to the history of a society that
has been far from well ordered — indeed hideously ill ordered — so
far as the basic elements of justice and dignity are concerned.
It is often said that there is a historical reason why European coun-
tries are more receptive to laws prohibiting group defamation than the
United States has been. This is half true: European peoples do have to
think about these matters against the background of Nazism and the
Holocaust (which is still within living memory). But it is false if this
rationale is supposed to suggest that Americans have no such burden.
Quite the contrary: the United States has historic memory within the
last two centuries of one of the most vicious regimes of chattel slavery
the world has ever known, which was upheld by the very Constitution
that purported then and still purports to guarantee individual rights;
living memory of institutionalized racism, segregation, and denial of
civil rights in many of its states; living experience — here and now —
of shameful patterns of discrimination and racial disadvantage; and
above all living memory of racial terrorism — lynching, whipping,
church bombing, cross-burning, and all the paraphernalia of Klan
symbolism — from the end of the Civil War to the present.
This history is the background against which members of formerly
subordinated minorities have to situate public manifestations of race
hate, group libel, and the like. It is not merely that the tone and con-
tent of such manifestations are at odds with the guarantees supposedly
afforded as public goods by the members of a well-ordered society to
each other. The worry is that these manifestations intimate a return to
the all-too-familiar circumstances of murderous injustice that people or

their parents or grandparents experienced. Such intimations are di-
rectly at odds with the assurances that a well-ordered society is sup-
posed to provide. Those assurances are sought not only in the ab-
stract, but also in relation to precisely the history that these displays
nightmarishly summon up.
III. LIBEL AND LEGITIMACY

It is time now to consider some objections. For all their good in-
tentions, for all the fine work that they may do in protecting the assur-
ances that a good society will provide to its most vulnerable citizens,
statutory provisions of the kind I am talking about are designed to
stop people from printing, publishing, distributing, and posting things
that they would like to say, things that they would like others to read
or hear. There is no getting around this fact. Such regulations make
the public expression of ideas less free — in the straightforward nega-
tive sense of more constrained141 — than it would otherwise be. Some
defenders of hate speech restrictions toy with the idea that, since hate
speech tends to silence minorities or to exclude them from the political
process, the net effect of censoring it may be to empower more in the
way of expression than it denies.142 I do not want to rule out that pos-
sibility; but I believe many countries would uphold their hate speech
laws even if that were not the case — that is, even if the harm done to
minorities were not primarily their exclusion from public discourse.
A. The Objection from Autonomy
I have never been convinced by an objection that bases itself sim-
ply on autonomy. It is true that laws of the kind we are discussing
make the public expression of ideas less free in a way that matters to
individuals.143 Often what racists or Islamophobes are punished for
expressing in public is the very thing, out of all the messages a person
could convey, that matters most to them. For them, other aspects of
political expression pale into insignificance compared with their leaf-
lets libeling Muslims as terrorists or their public portrayals of people of
other races as apes or gibbons. It is not exactly true that they them-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
141 For the notion of negative freedom, see ISAIAH BERLIN, FOUR ESSAYS ON LIBERTY 122–
31 (1969).
142 See, e.g., Tsesis, supra note 76, at 499–501 (“Hate speakers seek to intimidate targeted
groups from participating in the deliberative process. Diminished political participation because
of safety concerns, in turn, stymies policy and legislative debates.” Id. at 499.). For a more
nuanced discussion of this point, see generally Frank Michelman, Universities, Racist Speech and
Democracy in America: An Essay for the ACLU, 27 HARV. C.R.-C.L. L. REV. 339 (1992).
143 See C. Edwin Baker, Autonomy and Hate Speech, in EXTREME SPEECH AND DEMOC-
RACY, supra note 10, at 139, 143.
selves are silenced — they can say what they like as they like on in-
numerable other topics of public concern. But it seems to matter to
them that they be free to express racist ideas in a hateful form; and the
question is whether we should say, on this ground, that hate speech
regulations compromise their autonomy.144
The fact, however, that hateful expression lies at the valued core of
free speech so far as the racist is concerned does not by itself show that
it should not be restricted, any more than this position is furthered by
the high value someone may place on posting child pornography on
the internet. There are all sorts of exceptions to the free speech prin-
ciple. And there are all sorts of other expressions of autonomy, central
to people’s idiosyncratic values (ranging from the use of narcotics to
cruelty to animals), that are also legitimately restricted. Or think of
the various ways in which we regulate religious activity,145 which is at
least as central to worshippers’ autonomy as hateful expressions of rac-
ism are to the autonomy of bigots.
The question is whether the particular mode of autonomous self-
expression that is at stake here has a special significance that goes
beyond the simple point that the law — for what I think are very good
reasons — is stopping people from doing what they want. It does not
help the autonomy objection that even those who make the objection
acknowledge the wrongness and the social undesirability of the speech
they are protecting, and the suffering (if not the harm) that it inflicts
upon others. In other circumstances, they might be tempted to use the
old distinction between liberty and license to characterize hate
speech.146 Ronald Dworkin, for example, makes extensive use of that
distinction elsewhere in his political philosophy when he suggests that
we distinguish between a flat sense of liberty that “carries, in itself, no
suggestion of endorsement,” and a normative sense of the word that
identifies a value, virtue, or ideal that we endorse and are concerned
about.147 In the former sense, he says, someone may say that liberty is
reduced by laws prohibiting murder;148 but that would be a preposter-
ous objection to the laws against murder. But our use of “liberty” in
the latter sense indicates that we have already calculated the impor-
tance of people doing what they please in a certain area and found
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
144 See C. Edwin Baker, Autonomy and Informational Privacy, or Gossip: The Central Meaning
of the First Amendment, SOC. PHIL. & POL’Y, July 2004, at 215, 225–26; Charles Fried, The New
First Amendment Jurisprudence: A Threat to Liberty, 59 U. CHI. L. REV. 225, 233 (1992).
145 See, e.g., Employment Div. v. Smith, 494 U.S. 872 (1990).
146 See, e.g., JOHN LOCKE, TWO TREATISES OF GOVERNMENT 270 (Peter Laslett ed., Cam-
bridge Univ. Press 1988) (1690) (observing, concerning the state of nature, that “though this be a
State of Liberty, yet it is not a State of Licence”).
147 RONALD DWORKIN, SOVEREIGN VIRTUE: THE THEORY AND PRACTICE OF EQUALI-
TY 125 (2000).
148 Id.
that this benefit outweighs or trumps considerations of the dangers or

harms that might accrue from such behavior. Dworkin uses this dis-
tinction to show, for example, that there is no real objection on the
grounds of liberty to the pursuit of economic equality.149 The question
I find interesting is why Dworkin and other opponents of hate speech
regulation do not take that approach to the liberty claimed by the
hatemongers; why they do not say that whether this is liberty worth
fighting for depends precisely on its potential for harm to individuals
and the social order.
Alternatively, what we need is an account of why the liberty at
stake in the controversy about hate speech deserves our special con-
cern, in a way that entitles us to forgo calculations of harmfulness.
Professor Ed Baker took the view that the element of self-disclosure in
hate speech is what makes it special. The key value, according to
Baker, is individual autonomy. For autonomy to be respected, said
Baker, a person “must have a general right over the value-expressive
uses of herself — her own body.”150 In some circumstances a person
may not be permitted to act on her values (if they are thought to be
harmful to others or destructive of the social order). But she must be
permitted at least to disclose and express her values, if only so that she
can present herself to others and in public as the value-bearing person
she wants to be (or wants to be regarded as). Baker argued:
[T]he law must not aim at eliminating or suppressing people’s freedom to
make decisions about behavior or values. These requirements have clear
implications for speech, namely, that a person should be able to decide for
herself what to say. . . . This view centrally identifies the person with
agency, with action, and with the possibility of choice. In a sense, this is
an activity view of personhood: it favors a person’s activity of speech over
the status of being unknown.151
That view offers a plausible account of what might make speech spe-
cial, an account that might perhaps explain an inclination to protect
hate speech against regulation.
But I am not sure whether the argument works. Baker’s argument
depends, as I said, on a distinction between actions that disclose a per-
son’s values and words that disclose a person’s values. Regulating the
former is permissible, on Baker’s account, but regulating the latter is
not. However, that distinction must assume that what it categorizes as
mere words cannot be destructive of individually or socially important
values in the way that action can; a reference to autonomy cannot by
itself explain why we want to immunize self-disclosing words from
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
149 See id. at 128–39.
150 Baker, supra note 143, at 142. Baker further argues that “[a] person is not treated as formal-
ly autonomous if the law denies her the right to use her own expression to embody her views.” Id.
151 Baker, supra note 144, at 225–26.
regulation, because actions might be equally important for self-

disclosure. Now, the whole tenor of my argument in Part II is to show
that words, especially when they are expressed in a publicly visible
and enduring form, can seriously undermine social values that are im-
portant for sustaining the dignity and reputation of vulnerable indi-
viduals. That effect does not evaporate as soon as the word “autono-
my” is mentioned,152 and unless Baker and others who make this
argument want to suggest that anything that discloses a person’s val-
ues must be immunized from regulation no matter how dangerous it is,
then we cannot regard autonomy as a conclusive objection in the
present context.
B. Viewpoint Discrimination
I am also not convinced by arguments that complain that hate
speech legislation restricts free expression on the grounds of the
content or the viewpoint being expressed.153 I know this idea is one of
the central pillars of American free speech doctrine.154 Indeed, not
only does my argument in favor of the regulation of hate speech
confront the doctrine that prohibits viewpoint discrimination, it also
confronts what seem to be the most compelling reasons behind the
doctrine. In the view of Professor Geoffrey Stone, “[b]y definition,
content-based restrictions distort public debate in a content-differen-
tial manner. . . . Such a law mutilates ‘the thinking process of the
community.’”155
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
152 Of course, those who make the autonomy argument often dispute the impact that hate
speech is said to have on others, and no doubt they will dispute the claims about the importance
of assurance and the claims about the impact of hate speech upon assurance that I have made
in Part II of this Article. They may or may not be right about this, but it is odd to think that
this position — which is essentially an empirical claim — should be nourished by the claim about
autonomy. No doubt those who make the autonomy claim want it to be true that hate speech
does not have a deleterious impact, but certainly the claim about autonomy does not give any
reason for thinking that that is true. On the contrary, where an exercise of freedom in Dworkin’s
“flat sense,” see supra p. 1636, is challenged on grounds of the social harm that it produces,
that challenge must be evaluated before we decide that the exercise of freedom is entitled to an
elevated designation such as autonomy. The autonomy designation cannot be used to brush aside
the challenge.
153 I shall treat these two ideas — content-based regulation and viewpoint-based regulation —
as synonymous.
154 American free speech doctrine rests on the principle that an imposition on the freedom of
speech may not be based on the viewpoint of the speakers, the particular content of what is said,
or the distance between what is said and some official orthodoxy to which everyone in society is
supposed to subscribe in public. For a helpful discussion of the analytic difficulties surrounding
this distinction, see generally R. George Wright, Content-Based and Content-Neutral Regulation
of Speech: The Limitations of a Common Distinction, 60 U. MIAMI L. REV. 333 (2006).
155 Geoffrey R. Stone, Content-Neutral Restrictions, 54 U. CHI. L. REV. 46, 55 (1987) (quoting
ALEXANDER MEIKLEJOHN, POLITICAL FREEDOM: THE CONSTITUTIONAL POWERS OF
THE PEOPLE 27 (1960)).
Now, words like “distort” and “mutilate” beg the question, privileg-
ing what public debate would be like without intervention. It is worth
asking why we should privilege the unregulated process or its output.
Defenders of free speech sometimes draw an analogy to free markets in
the economic sphere. Left to themselves, free markets may generate
efficient outcomes by processes that economists say they understand.
And analogously, it is said (though without any analogous explana-
tion), in the long run the free marketplace of ideas will generate truth
or the acceptance of truth if it is left to its own devices. Actually this
is more of a superstition than analogy. Economists understand why
economic markets are capable of producing some good things and not
others; they may produce efficiency, but they may not produce dis-
tributive justice or they may undermine distributive justice. In the
case of the marketplace of ideas, is truth the analogue of efficiency or
is it the analogue of justice?
Stone is surely right to point out that restrictions on group defama-
tion or hate speech are intended to modify the character of public de-
bate. That is the whole point. They are designed to have such an im-
pact in circumstances where it is reasonable to believe that without
regulation public debate will have effects that the government has rea-
son to be concerned about. We enact and enforce restrictions on the
economic market for this reason all the time, prohibiting certain trans-
actions and regulating others. We do this in the marketplace of ideas,
too, as with the restriction of child pornography.
C. Ronald Dworkin’s Argument About Legitimacy
The objections I seek to dismiss may gain traction by being asso-
ciated with arguments about the relationship between free expression
and the political system. A democracy is supposed not only to respect
the autonomy of its voters, but also to be responsive to the views that
they hold. And the political process is supposed to be governed by the
expression of views of the citizens, not rigged in advance to privilege
some viewpoints over others.
There are a number of arguments in the literature that link the
protection of free speech to the flourishing of self-government in a de-
mocracy. Some say little more than that, though they say it sonorously
and at great length.156 In a few of these arguments, however, the posi-
tion is advanced beyond a general concern for the democratic process.
It is sometimes said that a free and unrestricted public discourse is a
sine qua non for political legitimacy in a democracy.157
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
156See generally, e.g., MEIKLEJOHN, supra note 155.
157See, e.g., James Weinstein, Extreme Speech, Public Order, and Democracy, in EXTREME
SPEECH AND DEMOCRACY, supra note 10, at 23, 28, 38. Professor Robert Post also makes this
Some make the point even more strongly than that and suggest that
the political legitimacy of certain specific legal provisions and institu-
tional arrangements may be imperiled by the enactment and enforce-
ment of hate speech laws. The most powerful argument of this kind is
presented by Ronald Dworkin.158 According to Professor Dworkin,
freedom to engage in hate speech or group defamation is the price we
pay for enforcing the laws that the haters and defamers oppose (for
example, laws forbidding discrimination). Here’s how the argument
goes:
Dworkin agrees that it is important for the law to protect people,
particularly vulnerable minorities, from, for example, discrimination or
“unfairness and inequality in employment or education or housing.”159
He is as committed to these laws as any proponent of racial equality.
Now, everyone knows that if we adopt such laws, often it will have to
be over the opposition of people who favor discrimination. In the face
of such opposition, we usually say that it is sufficient that such laws be
supported by a majority of voters or elected representatives, provided
of course that the opponents of the bills are not disenfranchised from
that process. They must have a chance to vote against it or vote for
candidates who oppose it, just as we have the chance to vote in its fa-
vor. But actually, says Dworkin, this is not all that is required:
Fair democracy requires . . . that each citizen have not just a vote but
a voice: a majority decision is not fair unless everyone has had a fair
opportunity to express his or her attitudes or opinions or fears or
tastes or presuppositions or prejudices or ideals, not just in the hope of
influencing others (though that hope is crucially important), but also
just to confirm his or her standing as a responsible agent in, rather
than a passive victim of, collective action.160
Opponents of antidiscrimination laws must have the opportunity to
voice their opposition, and the ground of their opposition, in public;
otherwise the political process by which their opposition is defeated is
unfair. Free expression, says Dworkin, is a necessary condition of po-
litical legitimacy: “The majority has no right to impose its will on
someone who is forbidden to raise a voice in protest . . . before the de-
cision is taken.”161 If we want legitimate laws against violence or dis-
crimination, we must let their opponents speak. And then we can legi-
timize those laws by voting.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
argument. See Robert C. Post, Racist Speech, Democracy, and the First Amendment, 32 WM. &
MARY L. REV. 267, 279–83 (1991).
158 Dworkin, supra note 10, at v–ix.
159 Id. at viii.
160 Id. at vii.
161 Id.
Now, some opponents of antidiscrimination laws will have no de-

sire to express their opposition hatefully. But some may; for them, de-
faming the groups that these laws are supposed to protect is the es-
sence of their opposition. Dworkin’s position is that it does not matter
how foul and vicious the hatemonger’s contribution to political debate
is. He must be allowed his say. Otherwise no legitimacy will attach to
the laws that are enacted over his opposition. It does not even matter
that the hatemonger’s speech is not couched as a contribution to politi-
cal debate at all. “A community’s legislation and policy,” says Dwor-
kin, “are determined more by its moral and cultural environment, the
mix of its people’s opinions, prejudices, tastes, and attitudes than by
editorial columns or party political broadcasts or stump political
speeches.”162 Whether the message is scrawled on the walls, smeared
on a leaflet, painted up on a banner, spat out onto the internet, or il-
luminated by the glare of a burning cross, it must be allowed to make
its presence felt in the maelstrom of messages that populate the mar-
ketplace of ideas.
And so, Dworkin’s legitimacy argument boils down to this: we
want to protect people with laws against discrimination and violence,
and it is natural to want to legislate also against the causes of discrim-
ination and violence.163 But, for Dworkin, there is only so much we
can do about those causes without forfeiting legitimacy for the laws we
most care about. Perhaps we can legislate against incitement. But,
Dworkin would say:
[W]e must not try to intervene further upstream, by forbidding any
expression of the attitudes or prejudices that we think nour-
ish . . . inequality, because if we intervene too soon in the process
through which collective opinion is formed, we spoil the only demo-
cratic justification we have for insisting that everyone obey these laws,
even those who hate and resent them.164
The structure of the position is interesting. Dworkin recognizes
that arguments about hate speech often involve two sorts of laws, not
one. On the one hand, there are the hate speech laws themselves (or
the proposals that people would like to see enacted): regulations re-
stricting expressions of racial or religious hatred, group defamation,
and so on. On the other hand, there are other laws in place protecting
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
162 Id. at viii.
163 In addition, Dworkin has doubts about some of the causal claims made by defenders of
hate speech laws. Id. at vi. That is a separate argument, though, as I observed, see supra note
152, it is perhaps not surprising that opponents of hate speech legislation hope that the causal
claims are false. See RONALD DWORKIN, FREEDOM’S LAW: THE MORAL READING OF THE
AMERICAN CONSTITUTION 219 (1996) (commenting on the causal claims of anti-pornography
campaigners).
164 Dworkin, supra note 10, at viii.
the people who are supposedly also protected by hate speech laws —
laws against discrimination, hate crimes, and so on. Following Dwor-
kin’s metaphor, I am going to call these upstream laws and down-
stream laws.165
Those who support the upstream laws often say that they are nec-
essary in order to address the causes of violations of the downstream
laws. If we leave hate speech alone, they say, then we are leaving
alone the poison that leads to violence and discrimination. Dworkin
turns the tables on this argument by saying that if you interfere coer-
cively upstream then you undermine political legitimacy downstream
— and that, he thinks, is a cost that even the defenders of hate speech
legislation should not be willing to incur.
D. Legitimacy: A Difference of Degree
The first thing to say in response to Dworkin’s argument is that
there is a question about what “spoiling the legitimacy” of these down-
stream laws amounts to.166 In social science, legitimacy often means
little more than popular support. Dworkin means it, however, as a
normative property: either the existence of a political obligation to
obey the laws or the appropriateness of using force to uphold them.167
Whichever of these he means, there is a question about how literally
we should take the claim that legitimacy is spoiled by the enforcement
of hate speech laws. I know that Dworkin does not mean that racists
are entitled to rise up in revolt against a society that enforces hate
speech regulation: it is not a loss of legitimacy in that drastic sense.168
But then what sort of loss of legitimacy are we talking about?
At worst it is supposed to be a loss of legitimacy in relation to these
particular downstream laws, rather than a catastrophic loss of legiti-
macy for the legal and political system generally. But even with that
limitation, the position seems counterintuitive if it is taken literally. In
Britain, there are laws forbidding the expression of racial hatred.169
There are also laws forbidding racial discrimination, not to mention
laws forbidding racial and ethnic violence and intimidation,170 and
laws against criminal damage protecting mosques and synagogues
from desecration; these are the downstream laws, the laws whose legit-
imacy Dworkin believes is hostage to the enforcement of hate speech
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
165 Id.
166 It will be obvious in what follows that I am greatly indebted to Professor Dworkin for dis-
cussing with me the ideas in this and the following sections.
167 See DWORKIN, supra note 140, at 190–92.
168 Email from Ronald Dworkin to author (Oct. 4, 2009, 21:34 EST) (on file with the Harvard
Law School Library).
169 Public Order Act, 1986, c. 64, §§ 3, 3A.
170 Race Relations Act, 1976, c. 74, § 70.
regulation. Should we really believe that in Britain citizens have no

obligation to obey these downstream laws? Should we really believe
that the enforcement of these downstream laws is morally wrong and
that the use of force to uphold them is just like any other illegitimate
use of force? A landlord discriminates against English families of
South Asian descent in a way that is prohibited by the Race Relations
Act. Do we really want to say that he has no obligation to obey the
antidiscrimination law and that no action should be taken against him,
at least so long as the statute book also contains provisions banning
him from publishing virulent anti-Pakistani views? Some skinheads
beat up a Muslim minicab driver in the wake of the 7/7 atrocities. Is it
wrong for the police to pursue, arrest, and indict these assailants be-
cause Britain has religious hate speech laws, which deprive down-
stream laws forbidding this sort of assault of their legitimacy? Must
the police stand by and not intervene, because any intervention would
be wrong? On a literal account, that is what “deprived of legitimacy”
means. It is not just Britain. Almost every advanced democracy has
hate speech laws, which, according to Dworkin, spoil the legitimacy of
any antidiscrimination laws that they have.171 It would seem that the
only advanced democracy entitled to have and enforce such laws is the
United States. Can that be right? That is American exceptionalism
with a vengeance.
Dworkin believes that hate speech laws may “spoil the only demo-
cratic justification we have for insisting that everyone obey [the down-
stream] laws.”172 But I do not think he really means us to take this
phrase in the literal sense considered in the previous paragraph. Any
argument will look silly if “it is pushed to an extreme.”173 So let us
consider some more moderate possibilities. One possibility is that the
enforcement of hate speech laws undermines the legitimacy of some
downstream laws and not others: perhaps it undermines the legitimacy
of laws forbidding discrimination but not the legitimacy of laws for-
bidding violence. After all, laws of the latter type have independent
reasons in their favor quite apart from the debate over race. But this
position will be hard for Dworkin to maintain in light of his more ho-
listic observations about the importance for legitimacy of speech that is
just part of the cultural environment, even when it is not intended as a
contribution to formal discussion of any particular law.174
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
171 See Dworkin, supra note 10, at viii.
172 Id.
173 Cf. JOHN STUART MILL, ON LIBERTY 26 (Currin V. Shields ed., 1956) (1859) (“Strange it
is that men should admit the validity of the arguments for free discussion, but object to their be-
ing ‘pushed to an extreme,’ not seeing that unless the reasons are good for an extreme case, they
are not good for any case.”).
A second possibility (compatible with the first) is that the legitima-

cy of any given law is itself a matter of degree and that, on the mod-
erate version of Dworkin’s argument, the enforcement of hate speech
laws diminishes the legitimacy of downstream laws without destroying
it altogether. I will address this point in detail in a moment.
A third possibility (also compatible with the other two) is that legi-
timacy is relative to persons. Professor Robert Post has suggested a
version of this: “If the state were to forbid the expression of a particu-
lar idea, the government would become, with respect to individuals
holding that idea, heteronomous and nondemocratic.”175 In Dworkin’s
argument one might say the downstream law becomes legitimately un-
enforceable as against the person silenced by the upstream law even
though it may be legitimately enforceable against others. But this
third possibility gets tangled up in issues about generality. Hate
speech laws are presented in quite general terms: they forbid anyone
from hateful defamation of racial, ethnic, and religious groups. Even
if they only have to be enforced against a few isolated extremists, they
have (and are intended to have) a chilling effect on everyone’s speech.
The second moderate position seems the most plausible: legitimacy
is not an all-or-nothing matter; the existence of hate speech laws di-
minishes the legitimacy of downstream laws but does not eliminate it
altogether. Dworkin puts it this way:
[T]here is something morally to regret when we enforce general non-
discrimination laws against racists who were not allowed to influence the
formal and informal political culture as they wished to do. On balance
Britain is entitled to enforce such laws, I think, but we are left with a def-
icit in legitimacy — something to regret under that title — because of the
censorship.176
So it is all a matter of degree:177 the “something to regret” might be
more or less considerable; the “deficit in legitimacy” might be larger or
smaller.
However, if we are going to recognize differences of degree, we
should recognize them on the other side of the equation as well. Let
me explain. On a given issue — say the desirability of an antidis-
crimination law — an individual, X, may have a range of views:
(1) X may oppose it because he thinks he will be worse off under
the law;
(2) X may oppose it because he thinks it will generate perverse eco-
nomic incentives, undermining economic efficiency;
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
175 Post, supra note 157, at 290 (emphasis added).
176 Email from Ronald Dworkin to author, supra note 168.
177 For a general acknowledgement that legitimacy is a matter of degree, see RONALD DWOR-
KIN, IS DEMOCRACY POSSIBLE HERE? 97 (2006).
(3) X may oppose it because he distrusts the bureaucracy necessary

to administer it; and
(4) X may oppose it because he denies that the intended beneficia-
ries of the law are worthy of the protection that it offers them.
Let us focus particularly on (4). It may be expressed in various
ways:
(4a) X may simply express his dissent from the broad abstract prin-
ciple that governments must show equal concern and respect to all
members of the community;
(4b) X may expound some racial theory that he thinks shows the in-
feriority of certain racial or ethnic groups;
(4c) X may express the view that the citizens who are intended to
be protected by the antidiscrimination law are no better than animals;
and
(4d) X may say in a leaflet or on the radio that these citizens are no
better than the sort of animals we would normally seek to exterminate
(like rats or cockroaches).
Out of all these various views and expressions, laws regulating hate
speech or group defamation are almost certain to restrict (4d), quite
likely to restrict (4c), and maybe they will restrict some versions of (4b)
depending on how hatefully these views are expressed.
On the other hand, most such laws bend over backwards to ensure
that there is a lawful way of expressing something like the proposi-
tional content of views that become objectionable when expressed pub-
licly as vituperation. They often insist on certain adverbial character-
istics as a condition of restriction: in Britain, for example, racist hate
speech that is not expressed in a way that is “threatening, abusive, or
insulting” is not to be restricted.178 Some such laws also try affirma-
tively to define a legitimate mode of roughly equivalent expression, a
sort of “safe haven” for the moderate expression of the gist of the view
whose hateful or hate-inciting expression is prohibited. The most gen-
erous such provision I have seen is in the Australian Racial Discrimi-
nation Act, which says that its basic ban on actions that insult, humil-
iate, or intimidate a group of people done because of their race, color,
or national or ethnic origin “does not render unlawful anything said or
done reasonably and in good faith: . . . in the course of any statement,
publication, discussion or debate made or held for any genuine aca-
demic, artistic or scientific purpose or any other genuine purpose in the
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
178 Britain prohibits the display of “any written material which is threatening, abusive or in-
sulting” if its display is associated with an intention “to stir up racial hatred,” but says that no of-
fense is committed if the person concerned “did not intend . . . the written material, to be, and
was not aware that it might be, threatening, abusive or insulting.” Public Order Act, 1986, c. 64,
§ 18(1)(a).
public interest.”179 The purpose of these qualifications is precisely to

limit the application of the restriction to the bottom end of something
like a (4a)–(4d) spectrum.
Now if we adopt the basic framework of Dworkin’s position and if
we accept that legitimacy is a matter of degree, we may want to say
that a law that prohibited the expression of (4a) and (4b) as well as (4c)
and (4d) would have a worse effect on downstream legitimacy than a
law that merely forbade something like (4d). And if we had a law that
was specifically tailored to prohibit only expression at the viciously vi-
tuperative end of this spectrum, it might be an open question whether
it would have anything more than a minimal effect on legitimacy.
Part of our estimation of the effect on legitimacy would surely also
depend upon the reasonableness and importance of the objectives of
the restrictive upstream laws. We see something similar with regard to
noncontent-based restrictions on speech (laws restricting time, place,
and manner of political demonstrations, for example). If such restric-
tions are arbitrary or motivated by only very minor considerations of
public order, we might say that they gravely impair the legitimacy of
collective decisions on the matters that the demonstrators want to ad-
dress. But if the motivation is based on serious considerations of secu-
rity, we might be more understanding. Something similar may be true
in the case of hate speech laws. A motivation oriented purely to pro-
tect people’s feelings against offense is one thing. But a restriction on
hate speech oriented to protecting the basic social standing — the ele-
mentary dignity, as I have put it — of members of vulnerable groups,
and to maintaining the assurance they need in order to go about their
lives in a secure and dignified manner, may seem like a much more
compelling objective. And the complaint that attempting to secure
this dignity damages the legitimacy of other laws may be much less
credible as a result.
E. Time and Settlement
We would not do justice to Dworkin’s argument without discussing
one much more difficult and challenging response to it. If the proposal
were to ban people from expressing contemptible views about welfare
recipients or about democratic socialism, then I think there would be a
case to be made along the lines of Dworkin’s argument — a case that
such suppression would put in question the legitimacy of our pursuit
of policies based on premises that people were being fined or put in
jail for denying. But, as I said in Part II, we are talking about the
fundamentals of justice, not its contestable elements.180 By the fun-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
179 Racial Discrimination Act, 1975, § 18D.
180 See supra note 111 and accompanying text.
damentals of justice, I mean things like elementary racial equality, the

basic equality of the sexes, the dignity of the human person, freedom
from violence and intimidation, and so on. These matters are founda-
tional in two senses. On the one hand, they represent things that
people rely on comprehensively and diffusely in almost every aspect of
their dealings with others. If one cannot exact respect for one’s basic
status as a rights-bearing individual, then almost everything is thrown
into question. On the other hand, they are also fundamental in the
sense that they represent relatively settled points or premises of mod-
ern social and legal organization.181 I do not mean that there is literal
unanimity about them — the hatemongers show that. Still, these mat-
ters are more or less settled in our laws and Constitution and so it is
not just morally necessary, but it is also quite reasonable now for us to
treat them as foundations for an awful lot else that we do. A debate
can be over in the sense that intelligent opinion has settled the matter
and it is inconceivable that public policy could proceed on any other
basis, and yet there can still be dangerous enclaves of politically pow-
erless but socially destructive outliers who do what they can to un-
dermine the furnishing of assurances based on these settlements.
So banning hate speech should not be understood as a way of in-
fluencing the great national debate about racial or sexual equality or
religious tolerance, nor should it be seen as a way of contributing to
the ending of that debate (as though without the intervention of these
laws the racists might win). The argument about assurance that I
made in Part II presupposes that the debate is more or less over. The
evil that it seeks to remedy is not the evil of the racists’ thinking or be-
lieving certain things. Rather, it is the evil of these racists trying to
create the impression that the equal position of members of vulnerable
minorities in a rights-respecting society is less secure than implied by
the society’s actual foundational commitments.
Maybe there was a time when we had to have a great national de-
bate about race — about whether there were different kinds of human
beings, inferior and superior lines of human descent, ranked hierar-
chies of capability, responsibility, and authority.182 But I think it is fat-
uous to suggest that we are in the throes of such a debate now — a vi-
tal and ongoing debate of a sort that requires us to endure the ugly
invective of racial defamation as a contribution in our continuing de-
liberation on a more or less open question. There is a sense in which
the debate about race is over, won, finished. As I said, there are outly-
ing dissenters; but we are moving forward as a society as though this
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
181 For a helpful characterization of this settlement — in relation to racial equality — see David
Kretzmer, Freedom of Speech and Racism, 8 CARDOZO L. REV. 445, 447 (1987).
182 See, e.g., IVAN HANNAFORD, RACE: THE HISTORY OF AN IDEA IN THE WEST 277–368
(1996).
were no longer a matter of serious or considerable contestation. And

the basis on which we move forward is that the settling of this debate
is fundamental to almost every aspect of the wellbeing, dignity, and se-
curity of formerly vulnerable minorities. On issues like affirmative ac-
tion, we continue to debate ways of moving forward on the basis of
the settled conviction that racism is wrong, but we are no longer in
need of a continuing debate about the premises of that argument.
If anything like this is true, then there is something odd about
Dworkin’s legitimacy argument. The impression he gives is that the
discourse to which racist hatemongers offer their “contributions” is a
living element of public debate, on which we divide temporarily into
majorities and minorities, but in respect to which no majoritarian laws
can be legitimate unless there is some provision for this important de-
bate to continue, so that the losers (the racists and the bigots) have a
fair opportunity to persuade the majority of their position on these
fundamentals the next time around. Dworkin’s position seems to as-
sume that debates are timeless and that considerations of political le-
gitimacy relative to public debate must be understood in a way that is
impervious to progress and settlement.
I understand the delicacy of any claim that a debate is over and
finished and that any further attempt to challenge the winning position
should now be suppressed.183 To clarify: I am not putting this claim
forward as an all-purpose license for the suppression of dissent or un-
popular views or even as an all-purpose response to objections to the
regulation of hate speech. I am using the idea of a debate being over
only with reference to this question of how seriously we should regard
the Dworkinian alarm about legitimacy, particularly in light of Dwor-
kin’s own acknowledgment that the effect of hate speech laws on legit-
imacy is a matter of degree. The legitimacy-impact of restricting de-
bate about an issue that is live and open seems to me quite a different
matter from the legitimacy-impact of restricting continued debate
about a foundational issue that was settled effectively decades ago.
I am mindful of John Stuart Mill’s point about the importance of
sustaining a “living apprehension” of the truths on which our social
system is organized, even when certain debates are for all intents and
purposes settled. Mill worried in his essay On Liberty that the settled
foundations of our social philosophy might become stale unless we
kept alive a sense of what needed to be said to their most determined
and persistent opponents.184 Most of us, however, part company with
Mill when he seems to suggest that it might be appropriate to cultivate
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
183 I would not like to see such a claim in the hands of someone who thought, for example, that
the debate about socialism was over.
184 See MILL, supra note 173, at 48–55.
opposition in order to enliven our deepest convictions.185 We might

agree with Mill when he says:
As mankind improve[s], the number of doctrines which are no longer dis-
puted . . . will be constantly on the increase; and the well-being of man-
kind may almost be measured by the number and gravity of the truths
which have reached the point of being uncontested. The cessation, on one
question after another, of serious controversy is . . . as salutary in the case
of true opinions as it is dangerous and noxious when the opinions are er-
roneous.186
Yet we need not accept his claim that this cessation of controversy
brings with it a certain cost — namely, “[t]he loss of so important an
aid to the . . . living apprehension of a truth as is afforded by the ne-
cessity of . . . defending it against[] opponents.”187 Mill concedes that
the loss of lively debate is not sufficient to outweigh the benefit of the
universal recognition of some truth, but he says it is “no trifling draw-
back.”188 He even suggests that if we did not have local racists to keep
our egalitarianism alive and jumping, we might have to invent
them.189 Most people, I think, are very chary of that rather daft sug-
gestion, and rightly so when manufacturing or empowering a “dissen-
tient champion” 190 has an effect not only on the liveliness of the debate
but also (and destructively) on the dignity-based assurance of vulnera-
ble members of society.
Let me emphasize again that the argument of this section is devel-
oped, not as a freestanding position, but as a response to Professor
Dworkin’s argument about legitimacy. I think we are now past the
stage where we are in need of such a robust debate about matters like
race that we ought to bear the costs of what amount to attacks on the
dignity and reputation of minority groups — or, more importantly, to
require individuals and families within those groups to bear the costs
of such humiliating attacks on their dignity and social standing — in
the interests of public discourse and political legitimacy. I believe we
are well past the point where we would sacrifice the legitimacy of our
antidiscrimination laws or the laws prohibiting racial violence by not
permitting people to defame one another in these terms and with those
effects.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
185 See id. at 53–55.
186 Id. at 53.
187 Id.
188 Id.
189 See id. at 54.
190 Id.
F. The Owens Case in Saskatoon

This distinction between debates that are over and debates that are
not is illustrated by a recent Canadian decision. In 1997, a corrections
officer in Saskatchewan named Hugh Owens offered for sale in the
Saskatoon Star Phoenix newspaper a bumper sticker designed to pro-
claim what he believed to be the Christian message concerning gay
marriage and perhaps homosexual relations in general. He said that
the advertisement was a Christian response to Gay Pride Week.191 Af-
ter a complaint by three gay men, who felt that the advertisement be-
littled them and subjected them to public hatred, Owens was hauled
before a one-person board of inquiry, set up by the Saskatchewan
Human Rights Commission in Saskatoon.192 He and the newspaper
were ordered to pay $1500 to the complainants.193 A court in Saska-
toon upheld the decision,194 but when Owens appealed to the Saskat-
chewan Court of Appeals, it reversed the decision.195 The court of ap-
peals recognized that “[p]art of the context which must inform the
meaning of Mr. Owens’ advertisement is the long history of discrimi-
nation against gay, lesbian, bisexual and trans-identified people in this
country and elsewhere.”196 But it also said this:
[I]t is significant that the advertisement . . . was published . . . in the mid-
dle of an ongoing national debate about how Canadian legal and constitu-
tional regimes should or should not accommodate sexual identities. . . .
Parliament would not pass legislation to make government programs and
benefits available on an equal basis to gay and lesbian couples until three
years after the advertisement appeared. When Mr. Owens’ message was
published the judicial sanctioning of same-sex marriage in Saskatchewan
was still seven years in the future and its sanctioning by the Supreme
Court of Canada was eight years in the future. This does not mean that a
newly won right to be free from discrimination should be accorded less
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
191 Anti-gays Ad Draws Protests, THE RECORD (Kitchener-Waterloo, Ont.), July 4, 1997, at A6
(“Regina resident Hugh Owens said he placed the ad in response to an ad announcing Gay Pride
Week in Saskatoon. He said he was not condoning gay-bashing but ‘as a Christian, I felt I had to
respond in some way.’”).
192 I should explain that Canada employs what is in my view a very silly arrangement utilizing
human rights commissions, which seem to have the power to summon citizens to appear before
them and issue injunctions and penalties, and initiate some hate speech proceedings in response to
private complaints. Under the Canadian Human Rights Act, 2010 S.C., ch. H-6, the Human
Rights Commission administers the hate speech and other provisions of the statute. In many oth-
er countries, similar laws are administered much more carefully, often with a requirement that
prosecutions not proceed without the specific authorization of the Attorney General in his or her
nonpartisan capacity. See, e.g., Public Order Act, 1986, c. 64, § 27(1) (stating that “[n]o proceed-
ings for an offence under this Part may be instituted in England and Wales except by or with the
consent of the Attorney General”).
193 Hellquist v. Owens, [2001] 40 C.H.R.R. D/197, para. 34 (Sask. Bd. Inq.) (Can.).
194 Owens v. Human Rights Comm’n, [2002] 228 Sask. R. 148 (Sask. Q.B.) (Can.).
195 Owens v. Human Rights Comm’n, [2006] 279 Sask. R. 161 (Sask. Ct. App.) (Can.).
196 Id. at 177.
vigorous protection than similar rights based on more historically estab-

lished grounds . . . . But, for purposes of applying a provision like [this], it
is important to consider Mr. Owens’ advertisement in the context of the
time and circumstances in which it was published. That environment fea-
tured an active debate and discussion about the place of sexual identity in
Canadian society. . . . Seen in this broader context, Mr. Owens’ advertise-
ment tends to take on the character of a position advanced in a continuing
public policy debate rather than . . . a message of hatred or ill will . . . .
Both the Board of Inquiry and the Chambers judge erred by failing to
give any consideration to this wider context.197
Many conservative Christians in Canada seem to have thought that
Owens’s speech should have been protected because all he did was cite
Bible passages.198 But that was not the argument of the court of ap-
peal, and rightly so. If someone had set up an equivalent bumper
sticker with a citation of Genesis 9:18–29 and an equals sign and a de-
piction of slavery, the fact that it was a Bible quotation would not
help. Equally, had Owens produced a bumper sticker citing the same
passages with the message conveyed in a bumper sticker in Queens-
land, Australia, “Under God’s law the only ‘rights’ gays have is the
right to die,”199 he might well have been liable to penalty. Our com-
mitment to the principle of human dignity has advanced beyond the
point where that sort of vituperation is tolerable. For, whatever the
state of the ongoing debate about gay marriage and the accommoda-
tion of sexual identities, Canada has committed itself as a society to
the proposition that the basic dignity and social standing of individuals
— their basic entitlement to recognition and respect in the sense de-
fined by Keegstra — is unimpaired by whatever we think about their
sexual activity and about civil recognition or nonrecognition of various
types of relationships.
I am conscious that this may seem an inadequate position to those
who are impatient with prolonging the debate about gay marriage. It
may seem like a grudging sort of toleration: we respect the person,
even while we disagree — and are permitted to express disagreement
— about the legal accommodation of their sexuality and relationships.
But I am not presenting this position as a general theory of toleration
or of the civil rights of gays and lesbians. It is presented only as an
account of when prosecution for hate speech may or may not be ap-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
197 Id. at 177–78 (citations omitted).
198 See ReligiousTolerance.org, Freedom of Canadians to Religiously Discriminate: A Canadian
Civil Rights Ruling Dealing with Anti-Gay Hate Verses in the Bible, http://www.religious
tolerance.org/bibl_hate3.htm (last visited Mar. 27, 2010).
199 See Somebody Think of the Children, Anti-Gay Bumper Sticker: Free Speech, Hate
Speech? (Sept. 24, 2008), http://www.somebodythinkofthechildren.com/anti-gay-bumper-sticker-
free-speech-hate-speech/ (discussing a Queensland bumper sticker that led to a conviction under
hate speech legislation).
propriate, and — like everything I said in the previous section — it is

developed as a response to the Dworkin position that everything must
be left free and completely up for grabs. It illustrates Dworkin’s posi-
tion that there may be a serious loss of political legitimacy if real de-
bates are closed down too quickly; but I believe my example also illu-
minates by contrast the point I have made, that some such debates —
about the basis of individual dignity, for example — must be treated as
essentially over, at least so far as the implications of the legitimacy ar-
gument are concerned.
G. Islamophobia
Everything I have said in these lectures is conditioned by a number
of important distinctions: (1) a distinction between the basics and the
contestable details of social justice and individual identity, so far as the
restriction of hate speech is concerned; (2) a distinction between hate-
ful and moderate modes of expressing essentially the same message,
which most hate speech statutes admit; (3) a distinction between the
legitimacy implications of regulating speech (for the sake of dignity
and assurance) when the speech “contributes” to a debate about fun-
damentals that is essentially over and the legitimacy implications of
regulating speech that contributes to a debate that is alive and ongo-
ing; (4) a distinction between speech that undermines the social enter-
prise of securing dignity and assurance and speech that merely offends;
and finally (5) a distinction between attacks on a person and attacks
on a position that they hold or the content of a set of beliefs they iden-
tify with or a lifestyle that they are wedded to.
Let me add one word about that last distinction. In many people’s
minds, there is a connection between Islam, as a religion, and jihadist
terrorism. Indeed there is a robust debate going on inside the Islamic
community about how substantial or inevitable this connection is.
And there is a similar debate going on in the world at large. Like the
debate about gay marriage, this debate too is not settled. To the de-
bate about the Islam-terrorist connection, I suspect that Mark Steyn’s
infamous piece in Macleans magazine in Canada, The New World Or-
der,200 and maybe even the Danish cartoons (portraying the prophet
Mohammed as a bomb-throwing terrorist)201 make some sort of
twisted contribution; and I believe they should be tolerated as such.
I do not mean that these contributions are admirable. In my view
there is something foul in the self-righteousness with which Western
liberals have clamored for the publication and republication of the
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
200 Mark Steyn, The New World Order, MACLEANS, Oct. 23, 2006, at 30.
201 The cartoons were originally published in the Danish newspaper JYLLANDS-POSTEN,
Sept. 30, 2005, at 3.
Danish cartoons in country after country and forum after forum. Of-
ten the best they could say for this was that they were upholding their
right to publish the cartoons. But a right does not give the right-
bearer a reason to exercise the right one way or another, nor should it
insulate him against moral criticism.202 My view is that the exercise of
this right was unnecessary and offensive; but as I have now said
several times, offensiveness by itself is not a good reason for legal
regulation.
However, where we are concerned with law and prosecutions, it is
also important to distinguish an attack on religious tenets and even an
attack on the founder of a religion from an attack on the dignity of the
believers. It is important not to let one’s critique of a religious or ec-
clesiastical or clerical position roll over into the denigration of the be-
lievers’ basic social standing, committed as they are to a given faith,
church, and religious practice in their ordinary lives. They are not to
be defamed, even if their religious beliefs are fair game. We find this
distinction embodied in statutes prohibiting religious hatred, for exam-
ple in the Public Order Act203 in the United Kingdom. Section 29C(1)
of that statute says that “[a] person who publishes or distributes writ-
ten material which is threatening is guilty of an offence if he intends
thereby to stir up religious hatred.”204 But section 29J insists that this
shall not “be read . . . in a way which prohibits or restricts . . . ex-
pressions of antipathy, dislike, ridicule, insult or abuse of particular re-
ligions or the beliefs or practices of their adherents.”205 No doubt this
provision fails to give Muslim communities the legal protections that
they want — namely, protection for Islam or punishment for defaming
its founder — just as the decision in the Owens case in Canada fails to
give the gay community what it wants. I have used these examples
nevertheless to illustrate distinctions that I think necessarily accompa-
ny any regulation based on the considerations that I have emphasized.
Some will say that these are hard lines to draw. So they are. But I
do not infer from this that we should therefore give up the position.
Legislative policy is often complicated and requires nuanced drafting
and careful administration, and outside the United States the world
has accumulated some experience of how to draft these regulations and
how to administer these distinctions. Some people believe that no po-
sition can be valid in these matters of constitutional concern unless it
is presented with rule-like clarity, is uncontroversially administrable,
and requires nothing in the way of further moral judgment or careful
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
202 See generally Jeremy Waldron, A Right To Do Wrong, 92 ETHICS 21 (1981), reprinted in
WALDRON, supra note 82, at 63.
203 Public Order Act, 1986, c. 64, amended by Racial and Religious Hatred Act, 2006, c. 1.
204 Id. § 29C(1).
205 Id. § 29J.
thought and discretion. I do not belong to that school. I belong to a

school of thought that accepts that the tasks assigned to courts and
administrators in matters of fundamental right (for example, rights to
free expression and to dignity) will often be delicate and challenging,
often involve balancing different goods and essaying difficult value-
judgments. I belong to that school, which in other contexts is asso-
ciated with the work of Ronald Dworkin: the moral reading of the
Constitution.206 And I don’t think people should defect from this
school of thought just because they perceive some advantage in doing
so for their position in the hate speech debate.
H. Distrust of Government
I am conscious that I have not even come close to addressing all the
arguments that there are against hate speech legislation. For example,
I have not said anything in these lectures to address the general mis-
trust of government that many people think underlies all First
Amendment concerns and that explains why many American legal
scholars are so opposed to hate speech laws.207 Let me say something
about that now.
As I understand it, the idea is that government interference is al-
ways likely to be motivated by officials’ lust for power, their vanity,
their misguided insecurity, or their undue responsiveness to majoritar-
ian prejudice, anger, or panic. Officials may not always get it wrong,
but there is a standing danger that they will. Why this danger is felt
in the particular area of speech (as opposed to government actions in
general) and indeed in the even more particularized area of content-
based restriction on speech, I am not quite sure. There is something to
it, I guess, when the best explanation of some of the prosecutions un-
der the 1798 Sedition Act is the wounded vanity of high officials or
when the best explanation of some of the twentieth-century prosecu-
tions — culminating in the 1951 decision about the application of the
Smith Act in Dennis v. United States208 — has more to do with the
unpopularity of a view held by a minority (members of the Communist
Party, for example) than with any real-world danger that it poses to
the state. But why would anyone think this was true of hate speech
legislation or laws prohibiting group defamation? Why is this an area
where we should be especially mistrustful of our lawmakers? The
worry about majoritarianism seems particularly strange. No doubt
there are cases where majorities legislate for their own interests to the
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
206 See DWORKIN, supra note 163, at 7–12; see also Ronald Dworkin, Comment, in ANTONIN
SCALIA, A MATTER OF INTERPRETATION 115, 120–24 (Amy Gutmann ed., 1997).
207 I am grateful to Geoffrey Stone for pressing this point.
208 341 U.S. 494 (1951).
disadvantage of vulnerable minorities: the legacy of segregation laws

and anti-immigration laws reminds us of that. But hate speech laws
represent almost exactly the opposite: a legislative majority bending
over backward to ensure that vulnerable minorities are protected
against hatred and discrimination that might otherwise be endemic in
society.
I have heard people say that it is not surprising that I oppose con-
stitutional restrictions on hate speech laws, since I am in general an
opponent of judicial review of legislation. However, it is not quite as
simple as that. Many countries that regulate hate speech also have
strong judicial review: Germany and Canada are examples. But in a
broader sense this speculation is right. I have long believed that
American constitutional jurisprudence exaggerates the likelihood that
majoritarian legislation will simply promote the interests of the majori-
ty at the expense of vulnerable minorities, who therefore need protec-
tion by the courts. And I have written about this incessantly, some
would say incorrigibly.209
But hate speech is an area where, against all odds, majorities prove
us wrong. In every advanced democracy where they are given the op-
portunity, majorities legislate to put this sort of protection in place be-
cause they care about the plight of minority communities. And by and
large these laws are administered responsibly. Certainly they do not
seem to have been transformed into vehicles for the promotion of ma-
jority interests in the way that a general distrust of government inter-
ference would suggest.
You may say, “Well that’s because you are focusing on the wrong
minority. The relevant minority here is not the African-American,
Muslim, or gay community. The real minorities disadvantaged by hate
speech prohibitions are the unpopular racists and bigots and virulent
Islamophobes whose beliefs are detested by those who make these
laws. Attacking those unpopular groups is just as much an instance of
the tyranny of the majority as an attack on Communists or atheists.” I
am afraid I have no patience at all for that recharacterization. It cer-
tainly does not affect the point that hate speech laws really are enacted
for the benefit of vulnerable racial, ethnic, and religious minorities, to
uphold their reputation and their dignity. At most, it just introduces
an additional minority into the picture. And it is a desperate maneu-
ver: one might as well say that DWI laws represent an attack on the
discrete minority of drunk drivers. In both cases, we have an account
of a serious social harm that certain activities, if they are left unregu-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
209 See JEREMY WALDRON, LAW AND DISAGREEMENT 211–312 (1999); Jeremy Waldron,
The Core of the Case Against Judicial Review, 115 YALE L.J. 1346 (2006); Jeremy Waldron, A
Right-Based Critique of Constitutional Rights, 13 OXFORD J. LEGAL STUD. 18 (1993).
lated, are likely to cause. In both cases we have a minority of poten-

tial victims of that harm to consider and a minority of potential of-
fenders. We can play word games with “majority” and “minority” un-
til the end of time, but the fact remains that hate speech laws do not
involve putting the interests of the majority above those of vulnerable
groups.
A more respectable concern is that even if hate speech laws
represent legislative majorities going out of their way to protect vul-
nerable minorities, there is no way of ensuring that an exception made
for this sort of legislation will not be a Trojan horse for other majori-
tarian speech restrictions that are less benign. And I agree: it is hard
to see how the exception could be defined or cabined in our constitu-
tional law. “Group libel” was one possible category, but as we saw in
Part I, many American constitutionalists have done their best to make
that category unusable. It is as though we have gone down a blind al-
ley in our First Amendment jurisprudence, committing ourselves to a
particular vision of what acceptable exceptions must be like — non-
content-based, oriented to clear and present danger of physical harm
or violence, and so on — and there is nowhere to turn and no way
back that would not unravel the whole scheme, making it “open sea-
son” on speech of every kind. It is as good an example of path depen-
dency as you could wish for.
Other societies are not in this predicament. Their experience has
not been that hate speech laws pave the way for more comprehensive
restrictions on speech.210 They began quite early on with the convic-
tion that speech of this sort — defaming vulnerable minorities and in-
citing hatred against them — was sui generis, and that it had to be
regulated if any speech was. They work with a much more sensible
and explicit rubric for developing limitations to rights than we have (I
mean the idea of restrictions imposed by law that are demonstrably
necessary in a free and democratic society211). They have been able to
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
210 What may be true is that other countries’ openness to other restrictions on speech (for ex-
ample, general restrictions on defamation or general restrictions on journalism that are likely to
prejudice a fair trial) has meant that they have not been blinded to the importance of hate speech
laws as American First Amendment ideologues have been. But that is not the Trojan Horse hy-
pothesis; the causality works in the other direction.
211 Article 10 of the European Convention on Human Rights says that the right of free expres-
sion “may be subject to such . . . conditions, restrictions or penalties as are prescribed by law and
are necessary in a democratic society, in the interests of national security, territorial integrity or
public safety, for the prevention of disorder or crime, for the protection of health or morals, for the
protection of the reputation or the rights of others.” Convention for the Protection of Human
Rights and Fundamental Freedoms, Nov. 4, 1950, 213 U.N.T.S. 221, art. 10, ¶2. The Canadian
Constitution says this in Article 1 about all the rights and freedoms set out in the Charter: they
may be subject “to such reasonable limits prescribed by law as can be demonstrably justified in a
free and democratic society.” Part I of the Constitution Act, 1982, being Schedule B to the Cana-
da Act 1982, ch. 11, § 1 (U.K.).
draw on each others’ experiences in drafting and formulating these

laws. And they have been bolstered in this enterprise by a sense of
their international obligations, which include an obligation to ensure
that “[a]ny advocacy of national, racial or religious hatred that consti-
tutes incitement to discrimination, hostility or violence shall be pro-
hibited by law.”212
I said at the beginning that it was not my intention to make a con-
stitutional argument.213 I can offer no way out of the First Amend-
ment labyrinth. I have taken some American arguments on this mat-
ter seriously, but mostly my purpose has been to present an
interpretation of the enactment and upholding of these laws in other
countries (and of the impulse to enact and uphold them here too, to the
extent that that exists). My method has been Dworkinian interpreta-
tion:214 let us make the impulse to enact them the best that it can be;
let us try to make sense of the reasons behind them and the limitations
and exceptions that such laws embody. I am not saying that we
should blindly imitate the forms of regulation that we find in other
advanced democracies.215 But in much of the discussion that I hear in
this country, the impression is given that if we were to enact hate
speech laws we would have to reinvent the wheel — and how on earth
would we do it? where would we start? how would we phrase it? what
groups would we privilege? and how would we control it? — and we
recoil from the assignment on that ground alone, quite apart from our
substantive reasons for opposition. While we have been adding new
culs-de-sac to our First Amendment jurisprudence, other countries
have been working away quietly on this issue and are doing quite well.
Even if we do not propose to follow them, we should have a better
understanding of what they are doing so that we can give an intelli-
gent account of our position that is responsive to the best that can be
said against it.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
212 International Covenant on Civil and Political Rights art. 20, ¶2, Mar. 23, 1976, 999
U.N.T.S. 171. See also International Convention on the Elimination of All Forms of Racial Dis-
crimination art. 4, Dec. 21, 1965, G.A. Res. 2106 A (XX), 660 U.N.T.S. 195 (1969). The United
States entered reservations against both provisions at the time of its ratification of these treaties.
214 See DWORKIN, supra note 140, at vii–ix.
215 See Dieter Grimm, Freedom of Speech in a Globalized World, in EXTREME SPEECH AND
DEMOCRACY, supra note 10, at 11, 19–22. But for the damaging externalities of American ex-
ceptionalism, particularly with regards to other countries’ attempts to respond to hate speech on
the internet, see Alexander Tsesis, Hate in Cyberspace: Regulating Hate Speech on the Internet,
38 SAN DIEGO L. REV. 817, 853–58 (2001).

Dignity and Defamation (Visibility of Hate)

Uploaded by

Copyright:

Available Formats

Dignity and Defamation (Visibility of Hate)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dignity and Defamation (Visibility of Hate)

Uploaded by

Copyright:

Available Formats

What is the main topic of discussion in the document?

What is the main topic of discussion in the document?

What arguments does the author present in support of hate speech laws?

What arguments does the author present in support of hate speech laws?

2009 OLIVER WENDELL HOLMES LECTURES

DIGNITY AND DEFAMATION: THE VISIBILITY OF HATE

INTRODUCTION ....................................................................................................................... 1597

DIGNITY AND DEFAMATION: THE VISIBILITY OF HATE

toleration. But I am going to argue that the most important aim of

I. WHY CALL HATE SPEECH “GROUP LIBEL”?

population.”16 In other countries, “group libel” and “group defama-

land.”23 Blasphemous libel was held to be a common law offense in

printing, effigy, picture, or other fixed representation to the eye.”31

scandalum magnatum,35 to protect nobles and great men from outra-

aimed at destroying a race by exposing it to contempt, derision, and

allegation descends to particulars. But — so the argument goes — this

That seems to me a viable, or at least arguable, position. But Justice

with posters depicting Britons of African ancestry as apes.64 After his

and equality.68 The Universal Declaration of Human Rights begins its

status as a member of society in good standing. It validates the legal

dignity of the members of a group stands separately from the status of

prosecuting atheists always sat uncomfortably with the American posi-

ciations as not worth the effort of suppression (though strong enough

II. WHAT DOES A WELL-ORDERED SOCIETY LOOK LIKE?

known to be regulated) by principles of justice and inhabited by

democratic society.”111 I want to take the measure of this assurance

laws limiting freedom of speech, but because of changes of attitude

visible assurance of just treatment that a society is supposed to pro-

I said in Part I that an emphasis on speech is an emphasis on the

million miles from the argument that I am pursuing in this Article.

speech or group defamation involves the express denial of these fun-

The derision, hostility and abuse encouraged by hate propaganda . . . have

called “appraisal respect” (in which one’s estimation of people varies

plays target the assurance on which members of vulnerable minorities

routine discrimination can handle only a handful of cases. By and

reliance should be public knowledge, publicly conveyed. I have ar-

the all-too-familiar circumstances of murderous injustice that people or

III. LIBEL AND LEGITIMACY

that this benefit outweighs or trumps considerations of the dangers or

regulation, because actions might be equally important for self-

Now, some opponents of antidiscrimination laws will have no de-

regulation. Should we really believe that in Britain citizens have no

A second possibility (compatible with the first) is that the legitima-

(3) X may oppose it because he distrusts the bureaucracy necessary

public interest.”179 The purpose of these qualifications is precisely to

damentals of justice, I mean things like elementary racial equality, the

were no longer a matter of serious or considerable contestation. And

opposition in order to enliven our deepest convictions.185 We might

F. The Owens Case in Saskatoon

vigorous protection than similar rights based on more historically estab-

propriate, and — like everything I said in the previous section — it is

thought and discretion. I do not belong to that school. I belong to a

disadvantage of vulnerable minorities: the legacy of segregation laws

lated, are likely to cause. In both cases we have a minority of poten-

draw on each others’ experiences in drafting and formulating these

You might also like