Offensive Speech, Religion, and the Limits of the Law
NICHOLAS HATZIS
Great Clarendon Street, Oxford, OX2 6DP, United Kingdom
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above
You must not circulate this work in any other form and you must impose this same condition on any acquirer
Crown copyright material is reproduced under Class Licence Number C01P0000148 with the permission of OPSI and the Queen’s Printer for Scotland
Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2020948243
ISBN 978–0–19–875844–0
DOI: 10.1093/oso/9780198758440.001.0001
Printed and bound in the UK by TJ Books Limited
Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.
Acknowledgements
I received invaluable advice from colleagues and friends who discussed with me the issues explored here or commented on draft chapters. I am indebted to Eric Barendt, Lior Barshack, Vincent Blasi, Sionaidh Douglas-Scott, Leslie Green, Kent Greenawalt, George Letsas, Ian Loveland, Virginia Mantouvalou, Yoram Shachar, the late Stavros Tsakyrakis, and the anonymous reviewers for Oxford University Press. For discussions on religion in constitutional law, and for unfailing support over many years, I am grateful to Nicos Alivizatos. My greatest debt in writing this book is to Gerald Neuman, who encouraged my initial interest in the relation between speech and religion and gave detailed comments on the manuscript.
I presented some of the ideas developed in the book in a series of lectures on free speech at the Radzyner Law School, The Interdisciplinary Center Herzliya in Israel. I thank Dean Amnon Lehavi, Lior Barshack, and the members of the faculty for the invitation and hospitality, and my students at the IDC for their lively interest and their questions and contributions in the lectures.
At the Press, Alex Flach and Jamie Berezin were generous with their help during the time it took to complete the work, and Brianne Bellio was instrumental in turning the manuscript into a book.
Chapter 6 incorporates some material first published in my article ‘Lying, Speech and Impersonal Harm’ (2019) 38 Law and Philosophy 517, and I thank the publisher and editors for permission to republish here.
R v Keegstra [1990] 3 SCR 697
Table of Cases
Suszkin v Israel [1998] IsrSC 52(3) 289
CANADA
ISRAEL
UNITED KINGDOM
Barnett v Kensington and Chelsea Hospital Management Committee (1969) 1 QB 428
Bowman v Secular Society [1917] AC 406
Brutus v Cozens [1973] AC 854
Casamitjana Costa v The League Against Cruel Sports, case number 3331129/2018, consent judgment (2 March 2020)
Grainger plc v Nicholson [2010] ICR 360
Overseas Tankship (UK) Ltd v Morts Dock and Engineering Co Ltd (The Wagon Mound) (No 1) [1961] AC 388
R v Chief Metropolitan Stipendiary Magistrate ex p Choudhury [1991] 1 QB
R v Lemon [1979] AC 617
R v Ramsay and Foote (1883) 15 Cox CC 231
Redmond-Bate v Director of Public Prosecutions [2000] HRLR 249 150
Robinson v Kilvert (1889) 41 Ch D 88
St Helen’s Smelting Co v Tipping (1865) 11 HL Cas 642
Thomas and Others v National Union of Mineworkers [1986] Ch 20
Taylor’s Case (1676) 1 Vent 293
UNITED STATES
Abrams v United States 250 US 616 (1919) .
111n.98
.115–16
92–93, 161n.20
Baggett v Bullitt 377 US 360 (1964) 125
Brandenburg v Ohio 395 US 444 (1969) 163–64, 165–66
Burstyn v Wilson 343 US 495 (1952) 136, 138
Cantwell v Connecticut 310 US 296 (1940)
Chaplinsky v New Hampshire 315 US 568 (1942) .
Clark v Community for Creative Non-Violence 468 US 288 (1984) .
102–3, 106–7, 137, 138
69n.44, 70
. . 14n.26, 85n.28
Cohen v California 403 US 15 (1971) 70, 73–74, 76–77, 101, 106–7, 111, 138
Connally v General Construction Co 269 US 385(1926) 124n.49
Cox v Louisiana 379 US 536 (1965) 149
Debs v United States 249 US 211 (1919) .
Dennis v United States 341 US 494 (1951).
Edwards v South Carolina 372 US 229 (1963) .
81–82, 159, 160
161, 164
149
Employment Division Department of Human Resources of Oregon v Smith 494 US 872 (1990) 128
Epperson v Arkansas 393 US 97(1968)
Feiner v New York 340 US 315 (1951)
Frisby v Schultz 487 US 474 (1988)
Frohwerk v United States 249 US 204 (1919) .
Gitlow v New York 268 US 652 (1925) .
Grayned v City of Rockford 408 US 104 (1972)
Gregory v Chicago 394 US 111 (1969)
Hill v Colorado 530 US 703 (2000)
Holder v Humanitarian Law Project 561 US 1 (2010)
Hustler Magazine v Falwell 485 US 46 (1988)
Jordan v De George 341 US 223 (1951) .
Kolender v Lawson 461 US 352 (1983) .
Masses Publishing Co v Patten 244 F 535 (SDNY), rev’d, 246 F 24 (2nd Cir 1917)
New York Times v Sullivan 376 US 254 (1964)
Noto v United States 367 US 290 (1961) .
Papachristou v City of Jacksonville 405 US 156 (1972)
People v Ruggles 8 Johns 290 (NY 1811) .
163, 164–65
.117–18
Robertson v Baldwin 165 US 275(1897) 119n.31
Scales v United States 367 US 203 (1961) 163
Schacht v United States 398 US 58 (1970) .
Schenck v United States 249 US 47 (1919) .
Sherbert v Verner 374 US 398 (1963) .
Smith v Goguen 415 US 566 (1974) .
.138–39
81–82, 159, 164
69n.41, 124–25, 129
Snyder v Phelps 562 US 443 (2011) 111n.100
Spence v Washington 418 US 405 (1974) 16n.33
State of Maryland v West 9 Md App 270 (1970) .
Street v New York 394 US 576 (1969) . . .
Terminiello v City of Chicago 337 US 1 (1949) .
119
69
Texas v Johnson 491 US 397 (1989) 68–69, 73–74, 76–77, 102, 106–7, 139
Thomas v Review Board of the Indiana Employment Security Division 450 US 707 (1981) 140n.107
United States v Alvarez 132 S Ct 2537 (2012)
United States v Eichman 496 US 310 (1990) .
United States v National Dairy Products Corp 372 US 29 (1963) . .
158, 165
69n.40
125n.54
Virginia v Black 538 US 343 (2003) 102n.74
Walker v Texas Division, Sons of Confederate Veterans 576 US 200 (2015) 87n.34
West Virginia State Board of Education v Barnette 319 US 624 (1943) 69n.41
Whitney v California 274 US 357 (1927)
Yates v United States 354 US 298 (1957) .
EUROPEAN COURT OF HUMAN RIGHTS
Belkacem v Belgium App no 34367/14 (20 July 2017)
Bergens Tidende and Others v Norway (2000) 31 EHRR 430 .
Bladet Tromso and Stensaas v Norway (1999) 29 EHRR 125 83n.21
Castells v Spain (1992) 14 EHRR 445 83n.20
EB v France (2008) 47 EHRR 21 116n.13
ES v Austria (2019) 69 EHRR 4 .
Garaudy v France App no 65831/01 (24 June 2003) .
Giniewski v France (2007) 45 EHRR 23 .
60, 64, 73–74, 89, 151, 156, 171
95n.59
88, 89, 109
Glimmerveen and Hagenbeek v Netherlands (1982) 4 EHRR 260 95n.59
Gunduz v Turkey (2005) 41 EHRR 5 24–25n.54
Handyside v United Kingdom (1976) 1 EHRR 737 107n.91
Hertel v Switzerland (1998) 28 EHRR 534 83n.21
IA v Turkey (2007) 45 EHRR 30 .
Incal v Turkey (1998) 29 EHRR 449 .
Jersild v Denmark (1995) 19 EHRR 1 .
. . . 60, 63, 64, 73–74, 84, 89, 156
83n.20
107n.92
Jerusalem v Austria (2003) 37 EHRR 25 83n.20
Karácsony v Hungary (2016) 42 BHRC 1 83n.20
Kokkinakis v Greece (1994) 17 EHRR 397 57, 127–28
Kuliś and Różycki v Poland App no 27209/03 (6 October 2009)
Lehideux and Isorni v France (2000) 30 EHRR 665
Lingens v Austria (1986) 8 EHRR 407
83n.21
94, 97
M’Bala v France App no 25293/13 (20 October 2015) 95n.59
Malone v United Kingdom (1984) 7 EHRR 14 126
Members of the Gdani Congregation of Jehovah’s Witnesses v Georgia (2008) 46 EHRR 30
Müller v Switzerland (1991) 13 EHRR 212
Murphy v Ireland (2004) 38 EHRR 13
113n.104
75n.69
.103–4, 108
Norwood v United Kingdom (2005) 40 EHRR SE11 95n.59
Otto-Preminger-Institut v Austria (1995) 19 EHRR 34 22, 56–57, 63, 64, 73–74, 80, 83–84, 89, 104–5, 108, 156, 166–67, 170
Pavel Ivanov v Russia App no 35222/04 (20 February 2007)
Redfearn v United Kingdom (2013) 57 EHRR 2
Selisto v Finland (2004) 42 EHRR 144
95n.59
96, 97
83n.21
Silver v United Kingdom (1983) 5 EHRR 347 126n.59
Sunday Times v United Kingdom (1979) 2 EHRR 245 126
Tagiyev and Huseynov v Azerbaijan App no 13274/08 (5 December 2019)
Tierfabriken v Switzerland (2001) 34 EHRR 159
Thorgeir Thorgeirson v Iceland (1992) 14 EHRR 843
60–61
83n.21, 105
83n.21
Wingrove v United Kingdom (1996) 24 EHRR 1 59, 64, 73–74, 84, 89, 108, 115–16, 156
EUROPEAN COMMISSION ON HUMAN RIGHTS
Choudhury v United Kingdom App no 17439/90 (5 March 1991) 116n.14
Table of Legislation
CANADA
Criminal Code s 319(2) 27
GERMANY
Criminal Code s 130 26
Criminal Code section 166 141–42
IRELAND
Blasphemy (Abolition of Offences and Related Matters) Act 2019 . . . . . 122–23
UNITED STATES
Espionage Act 1917 81–82, 159
Smith Act 1940
Stolen Valor Act 2005
UNITED KINGDOM
Criminal Justice and Immigration Act 2008, s 79(1) .
Public Order Act 1936 s 5
Public Order Act 1986
s29B(1) .
S29J
116–17
9n.6
28–29
Racial and Religious Hatred Act 2006 29–30 s1 28–29
List of Abbreviations
ACHR American Convention on Human Rights
ECHR European Convention on Human Rights
ECtHR European Court of Human Rights
ICCPR International Covenant on Civil and Political Rights
Introduction
I learn belatedly, thanks to the seaside indolence of August on a Dalmatian island that piles up stacks of outdated newspapers and weeklies, that in Denmark they have purged—I imagine from the schools—a tale by Hans Christian Andersen with a Christian ending, or Christian elements in any case, so as not to offend the faithful of other churches. In its respectful stupidity, this is a decisive step in the universal history of censorship. In this case, it is a well-intentioned censorship, moved by a concern not to upset cultural or religious minorities. But censorship, after all, is always well-intentioned: it seeks to protect morality, one’s country, the family, institutions, order, society, progress, the people, the children, health.
This is the beginning of a short essay by Claudio Magris on censorship.1 In a few sentences, he captures some of the main aspects of the problem of offensive speech: that offence has become a major preoccupation in Western, liberal democracies; that religion is very often at its root; that sometimes the mere expression of a different viewpoint is thought of as offensive; that offence to beliefs is linked to issues about respect for the person holding the beliefs; that purging from public space what is considered offensive aims at protecting people’s feelings; and that those who argue for restrictions are motivated by a particular view of the role of speech in society.
In this book, I explore how these themes matter for the way we think about free speech and the legal right to freedom of expression. While most of what I will say applies to offensive speech more generally, the focus is on speech which offends religious beliefs and sensibilities. Religion has been at the heart of global controversies about free expression, sparked by, among others, the Danish cartoons and the Charlie Hebdo publications. On several occasions, the
1 Claudio Magris, ‘A New Writer: The Censor’ in Snapshots (translated by Anne Milano Appel) (Yale University Press 2019) 69.
result was violence and death. But even discussions about offence to religion are often very intense and agitated, conducted with a passion which is unusual when talking about other matters of public interest. Coping with what we experience as profoundly upsetting is always complicated, but it appears that we find it easier to put up with insults to our beliefs about, say, politics or society; offence to religious beliefs activates a different kind of visceral reaction on the part of the offended. This particular difficulty with religious insults is reflected in judicial practice, the prime example being the European Court of Human Rights. As we will see in Chapters 3 and 4, the Court consistently shies away from any meaningful scrutiny of measures which restrict religiously offensive speech, showing wholesale deference to the views of national authorities. By contrast, it scrutinizes restrictions on political speech much more strictly, even when the speaker’s message is profoundly offensive, and has ruled that the offensiveness of a political view is not sufficient reason for banning its public expression.
The question underlying this book is when, if ever, is the government justified in silencing those who offend the religious beliefs of their listeners? Starting from the fact that speech is usually considered a privileged activity compared with other forms of human conduct, I explore the normative reasons which may support or negate offensive speech bans. In order to show how those reasons operate in constitutional practice I use cases from various jurisdictions but it is not the purpose of the book to offer a complete description of the law under any particular system.
Often, I draw on US case law and First Amendment theory. A common objection raised in debates on freedom of expression is that the American model, with its very strong protection of speech, is so exceptional that it is of limited value when we consider speech regulation in other liberal democracies. I think this concern is exaggerated. The First Amendment model developed, to a considerable degree, as a response to the reality that American society was increasingly plural—in politics, religion, and ways of life; enhanced protection for the public expression of diverse views was seen as a way of negotiating the inevitable tensions brought about by the experience of difference. Nowadays, most Western societies are similarly plural and diverse and face the same dilemmas which have led to some of the landmark judgments of the US Supreme Court. I am not suggesting that American law should be transplanted in other legal systems but it seems to me impossible to discuss freedom of expression without looking closely at the theory and practice of the First Amendment, which, with subtlety and sophistication developed over the course of more than a century, offers valuable insights regardless of jurisdiction. Besides, all liberal systems,
while casting the interests involved in free speech disputes in different ways, accord speech a privileged position in the sense that they acknowledge that there is something particularly important in being able to speak our minds.
Further, I am not aiming at constructing a theory of offensive speech, the sort of system of principles or comprehensive methodology which can be applied whenever a communication causes offence. I doubt that this is either desirable or possible. Rather, I will attempt to flesh out the moral content of the claims for offensive speech prohibitions and explore how the values justifying free expression affect how we understand the conflict between speech and the protection of religious sensibilities, taking into account the institutional arrangements which guarantee, and impose limits, on speech. While my suggestion is that the arguments for proscribing offensive expression are unpersuasive (with some very limited exceptions), I hope that the discussion will be interesting for readers with diverse views on the role of speech for the political life in our plural societies.
The first two chapters set the scene. In Chapter 1, I explore the nature of offence as a constellation of unpleasant feelings stirred up when our expectations to be treated in a particular way are disappointed. In the offended state of mind, psychic pain is caused by the clash between internal reality and the external world. Legal systems usually proscribe several types of offensive conduct but offence caused by speech poses particular challenges because of the importance of free expression for individuals and society. I propose that for the purpose of assessing a political community’s response through its institutions to communications which offend some of its citizens, the category of offensive speech should be construed narrowly and distinguished from speech which promotes hatred. The two are morally different, and it is a mistake to borrow arguments from the hate speech debate to support the restrictions on expression which offends.
Then, in Chapter 2, I consider the structure of the claim that the government ought to proscribe religiously offensive speech, and the sort of reasons that can be invoked to support it. The claim, a form of practical reasoning, implies that respect for religious beliefs is a value and that by banning religious insults the government realizes that value. However, coercive restrictions on liberty ought to be based on reasons which all reasonable citizens can be expected to find acceptable, regardless of their personal views on what constitutes an ethical life. This is usually described as a public reason requirement. My suggestion is that even public reason theories that allow for reliance on religious grounds
on some occasions, cannot accommodate the use of coercion against speakers who offend religious sensibilities. The fact that an act or opinion is sinful or otherwise wrong from a religious perspective is not a proper ground for its prohibition.
Is there a non-religious basis for regulating sacrilege and profanity?
A common strategy is to say that protection from offence to religious beliefs is part of the more general right to freedom of religion. When the government proscribes offensive speech is giving effect to that right, and this is a reason which can justify the exercise of its coercive force over speakers and be accepted by everyone as legitimate, independently of their views about God and the sacred, because we all have an interest in living under a regime which protects religious freedom. This is the issue addressed in Chapter 3. I argue that although freedom of religion is an important right, believers have no right-based claim to be protected from unpleasant feelings caused by religious insults. It is, therefore, erroneous to interpret, as the European Court of Human Rights does, the legal right to religious freedom as including a right not to suffer offence to one’s religious beliefs. In turn, this means that cases on religious offence do not involve a conflict between two fundamental rights, namely speech and religion.
The position that religious offence is a proper target for state coercion also entails a broader view about the nature and functions of public discourse and the place of religious speech in it. That view considers public discourse as a highly regulated space for orderly discussion, with the role of the regulator reserved for the state, which can set the terms for what can be said publicly. One of its central themes is the distinction between content and style: the government cannot generally proscribe unpopular, anti-religious speech which is moderate and respectful but can legitimately intervene when the speaker has exceeded the limits imposed by norms of civility. Chapter 4 offers an argument against this approach. It suggests that it is better to think of public discourse as the locus of an open deliberative process where all citizens have an equal claim to speak even if what they say is abusive or offensive. This way it performs a legitimizing function for the exercise of democratic authority by the state and allows individuals to discern, even in decisions with which they disagree, a trace of their moral agency. Such conditions make possible for society a process of self-reflection, which is necessary for change and progress.
Two important limitations on governmental power to regulate speech follow from this understanding of public discourse. On the one hand, freedom of speech covers both the substance and the manner of expression. Speakers can address the important and difficult issues around the role of the sacred in human life in a civil and moderate manner but it should also be possible to use
humour, satire, mockery, sarcasm, and abuse. On the other hand, the agenda and rules of public discussion are not value-neutral issues situated outside public discourse to be determined by the government. They are part of it— substantive norms subject to constant debate and reinterpretation.
The themes of religious speech in public debate and the civility norms relied on to restrict religious offence are pursued further in relation to blasphemy, and its most recent reincarnation as defamation of religion, in Chapter 5. Blasphemy laws have been the usual response to the problem of speech which offends religion and remain rather common even in liberal, democratic legal systems. The first part of the chapter traces the transformation of blasphemy from an offence aimed at the protection of Christianity to a much broader, inclusive norm which protects all religions (and, in some cases, non-religious conscientious beliefs) from vilification and insult. I argue that the even this modernized version should be rejected both because it raises serious concerns about vagueness and unequal enforcement and because it is incompatible with free expression values.
Subsequently, I address the question whether modern blasphemy laws can be justified as a means for safeguarding public order. The argument in their favour rests on the idea that those whose beliefs are offended may become indignant and resort to violence, so the government ought to prohibit expression which has this effect. This position starts from the correct premise that the government has a legitimate interest in maintaining peace and order but concludes, wrongly, that this should be achieved by silencing non-violent speakers. Instead, the morally sound course of action is to arrest those who cannot control their rage and threaten to become violent when their beliefs are insulted.
In Chapter 6 I discuss, and reject, two further arguments for restricting offensive speech which targets the beliefs of religious minorities. The first is that it insidiously reinforces negative stereotypes about them, eventually leading to discrimination against their members. I explore what exactly is meant by the claim that speech leads to discriminatory acts and what causal link between expression and discrimination ought to be required for the government to suppress religious insults. The second argument is that offending religious beliefs is a form of intolerance towards those who hold them, so when the government is restricting communications which offend, it gives effect to the value of toleration. My discussion focuses on the kind of moral obligations which flow from toleration. I defend the view that a commitment to toleration should always leave some room for the expression of a negative attitude towards that which is being tolerated and that, in principle, speech is the appropriate means for conveying such an attitude.
Finally, some clarifications. First, the arguments in this book do not depend on any particular view about the appropriate relation between religion and the state. Secularists who support a strict separation between the two will probably also think that it is impermissible for the government to silence speakers who offend the religious sensibilities of listeners. But it is entirely possible to think that religious viewpoints have their place in our common life as a political community and that it is appropriate for the state publicly to acknowledge the importance religion has in the lives of many of its citizens and still disagree with prohibitions of religiously offensive speech. Put differently, there is no necessary link between support for strict secularism and the rejection of censorship of religious insults.
Secondly, the book does not discuss all the cases where offence is morally relevant reason for determining an actor’s conduct. My concern is exclusively with whether the state can coerce speakers into silence when listeners find their messages offensive. Clearly, there can be other circumstances where the fact that someone will be offended can count as reason against speaking in a particular manner. If I find my friend’s religious beliefs entirely absurd or misguided but know that by ridiculing them I will make him feel deeply insulted, I may decide to temper my language, and this may be what is required of me as a friend. The fact that one has a right to speak does not necessarily mean that speaking is the right thing to do. The fact of offence, then, is reason guiding my conduct.
Thirdly, the discussion in this book proceeds on the assumption that religion is on the receiving end of the insults. The offence-giver is the speaker who attacks religious beliefs and the offence-taker is the devout person whose beliefs are insulted. The position I argue for is that the speaker has a right to express his or her views in a manner which can be respectful or insulting, and that the believer has no claim to be protected from the unpleasant feelings caused by the insults. But the same holds true if the beliefs insulted are secular, with the offensive speech stemming from religious speakers. There are many instances where religious views, and the way they are expressed, can be experienced by non-religious listeners as very offensive. The arguments against prohibitions of speech apply to such cases with equal force. If the law denies its protection to believers whose religious sensibilities are insulted, it ought to do the same when the devout attack the sensibilities of the secular causing them offence.
1
The Nature of Offence
Feeling offended is a universal experience. Even the most thick-skinned of us, those who can routinely brush off whatever others say or do and move on with their lives, have, at times, felt offended. The emotional experience of having suffered offence varies from mere irritation and annoyance to profound and intensely felt shock; the more serious the offence, the deeper and more longlasting the wound inflicted by it will be. The Iliad begins with an offended Achilles sulking in his tent—Agamemnon, the leader of the Greeks, has taken away his slave girl, Briseis—and refusing to fight, thus bringing the campaign to conquer Troy to a standstill. Achilles’s insult is not caused merely by the loss of a possession but by the symbolism entailed in Agamemnon’s action. When the two of them quarrelled in the presence of the other kings, Agamemnon boasted: ‘I will then lead off the fair-cheeked daughter of Briseus, going myself to your cabin—your own prize—so that you know well how much greater than you I am’.1 The effect of these words, Homer tells us, was that ‘in Peleus’ son pain grew; in his shaggy breast was the spirit divided and two ways pondered the matter, whether to draw from the sheath of his thigh his keen-whetted sword blade, make the assembly disperse, then cut down Atreus’ scion, or to desist from his anger and put a strong curb on his spirit’.2 For both the offencegiver and the offence-taker, it is the symbolic significance of the offensive conduct which determines their experience of it.
Insults strike at the view we have of ourselves and our expectations to be treated by others in a certain way—with respect, consideration, deference. They imply that the offence-giver occupies a superior position, looking down on the victim of the insult, who is made to feel belittled. ‘Insult is about humiliation and the assertion of superiority, the assertion or assumption of dominance’.3 This explains why offence causes psychic pain. Our understanding of who we are and our sense of self-worth are inevitably influenced by how others see and treat us. When we receive insults instead of respect this
1 Homer, The Iliad (translated by Rodney Merrill) (The University of Michigan Press 2009) Book 1, 185–86.
2 ibid 188–92.
3 Jerome Neu, Sticks and Stones: The Philosophy of Insults (Oxford University Press 2008) 4.
understanding is shaken and, depending on the nature of the offence and our emotional capacity for working through it, we may be shattered altogether. This can make us feel that we are under attack, inferior, worthless. As Homer knew, in the offended state of mind insults eat us from the inside, arousing murderous rage.
In the example from The Iliad, offence was intentional and directed to a specific person. It is also possible to cause offence to others without referring to them directly, or even unintentionally. A derogatory comment made in the media about a group with a certain characteristic, such as religion, race, ethnic origin, or sexuality, can make someone who has that characteristic feel insulted, even though the comment was not addressed to her as an individual. But offence can also be caused without the offence-giver intending to insult.4 This can happen because he was careless, ignorant, or inconsiderate. He may have entirely failed to notice the risk of giving offence when he was reasonably expected to have known about it, or he was aware of the risk but he accepted it and did or said something which could be interpreted as an insult. Such conduct resembles the legal concept of negligence: the offence-giver has failed to exercise reasonable care not to insult others. In ethics and in law we usually think of injuries caused with intent as more blameworthy than inadvertent injuries. The latter may benefit from an excuse, as, for example, when it was not reasonably possible for the offence-giver to have known that his conduct could offend. Yet, even unintentional insults leave the victim with a bitter taste in the mouth.
It is not necessary for offence to be caused by acts of a specific individual, as it is possible for a person to experience unpleasantness by merely thinking of a state of affairs which he finds generally offensive. One may be offended at the thought of world poverty, or climate change, or the level of illiteracy in some countries. These are cases where offence is caused by a social phenomenon which the offended party thinks is incompatible with her values while being, at the same time, too complicated and widespread to be attributable to individual people. It may be argued that such offensive states of affairs, the impersonal forces of history or the economy notwithstanding, could not have come about unless a number of specific individuals had behaved in a reprehensible manner. For instance, it may be said that the ‘very rich’ are largely responsible for the unequal distribution of wealth and, therefore, world poverty or that the ‘environmentally indifferent citizens’ are largely responsible for
4 For more on unintentional insults see Neu (n 3) 18–24.
climate change. While there may be some truth in those statements, the pool of the people whose conduct is under scrutiny is too large and undifferentiated, and their relationship to the actual offended mental state is too remote to allow us to attribute in a meaningful way offence to conduct. Therefore, it is better to think of offence taken at such states of affairs as a type of impersonal offence whose cause cannot be traced to a specific person.
Finally, there are cases where, although it is possible to relate offence to an individual, it would be unreasonable to ascribe to him any kind of responsibility for the unpleasant mental state he has caused to others. The sight of a seriously wounded person covered in blood after a car accident may be experienced by a witness as an offence to the senses, but it would be absurd to say that he is in any way morally culpable for the observer’s discontent.5 In such cases, the issue of blameworthiness does not even arise.
The range of words and actions which can offend is limitless. Given that what counts as an insult is determined by several factors—social conventions, history, personal expectations, susceptibility to offence—different societies, and different people within the same society, will develop their own understanding of offensiveness. Faced with the ubiquitous nature of offence, a legal system needs to determine whether it is ever justified to use the coercive force of the state to prevent and punish it, and, if so, under what circumstances this should be done. In this chapter I look, first, at some general considerations about the law’s response to offence and then turn to the more specific issue of offence caused by speech.6
5 The example is from Joel Feinberg, The Moral Limits of the Criminal Law, vol 2: Offense to Others (Oxford University Press 1985) 2. Feinberg classifies it as an offended state of mind in the broad sense to distinguish it from what he calls offence in the ‘strict and narrow sense’ which, in addition to suffering a disliked mental state, includes attributing that state to the wrongful conduct of another and resenting the other for causing the offence.
6 I am using offence and insult interchangeably to describe a mental state. In Brutus v Cozens [1973] AC 854 Lord Reid attempted to distinguish, for the purpose of statutory interpretation, ‘insulting’ conduct from other conduct which people may find upsetting. The case concerned a member of the audience at Wimbledon who interrupted the match of a South African player to protest against apartheid. The spectators became angry and he was later prosecuted for violating s 5 of the Public Order Act 1936, which provided that ‘any person who in any public place or at any public meeting uses threatening, abusive or insulting words or behavior with intent to provoke a breach of the peace or whereby a breach of the peace is likely to be occasioned, shall be guilty of an offence’. The magistrates held that his conduct was not insulting to the spectators but the Divisional Court set aside the judgment, stating that conduct which affronts others and shows disrespect or contempt for their rights counts as insulting for the purposes of the Act. In the House of Lords, Lord Reid held that the meaning of ‘insulting’ in the statute was the ordinary meaning of the word and its determination was a matter of fact, not law. The term ‘affront’ used by the Divisional Court was too vague and contempt for the rights of the person as opposed to contempt for the person himself was not as such insulting. He noted that ‘there are many grounds other than insult for feeling resentment or protesting’ (at 862) and that spectators may have been angry but they were not insulted within the meaning of the 1936 Act.
Legal Constraint of Offence
The most comprehensive and influential exploration of how the law should respond to offence can be found in the work of Joel Feinberg, who develops what he describes as an ‘offence principle’: ‘it is always good reason in support of a proposed criminal prohibition that it would probably be an effective way of preventing serious offense (as opposed to injury or harm) to persons other that the actor, and that it is probably a necessary means to that end’.7 The offence principle supplements the ‘harm principle’8 and extends the reach of the state’s power, which can be exercised to prevent both harm and offence to others. Feinberg’s theory differs in this fundamental respect from the classic formulation of liberalism developed by John Stuart Mill, who insisted that the prevention of harm to others is the only aim for which state coercion may be used legitimately. Taken together, the harm and offence principles ‘duly clarified and qualified, between them exhaust the class of morally relevant reasons for criminal prohibitions’.9
Offence, Not Harm
The term offence could be used to describe a very broad range of unpleasant mental states, such as irritation, discomfort, anger, rage, disgust, aversion, shame, humiliation, disappointment, and so on; one way or another, they can all be described as offensive for the person experiencing them. Those states become normatively significant when ‘caused by the wrongful (right-violating) conduct of others’.10 It is with offence in this normative sense that the offence principle deals, seeking to determine when the government is justified in restricting individual liberty to prevent it.
An offence principle is necessary only if offence is morally different from harm. It is possible to think that the relationship between the two is that offence is a minor or less serious form of harm, with both of them occupying different places in the same scale of bad things that can happen to the victim.11
7 Feinberg, Offense to Others (n 5) 1.
8 Joel Feinberg, The Moral Limits of the Criminal Law, vol 1: Harm to Others (Oxford University Press 1984).
9 ibid 14.
10 Feinberg, Offense to Others (n 5) 1–2.
11 See eg Donald Vandeveer, ‘Coercive Restraint of Offensive Actions’ (1979) 8 Philosophy and Public Affairs 175, 178, who takes the view that offence constitutes harm, their difference being one of degree (‘When people are offended, they are harmed-typically in a mild way, but not always. “Harm” tends to range over cases where individuals are made seriously to mildly worse off; “offence” tends to range over cases where individuals are made less seriously to mildly worse off’).
Understood in this manner, offence does not require a separate set of rules but can be dealt with under the harm principle, the only question being how the requirements of the harm principle may need to be adapted to accommodate injury caused by offence. Feinberg argues persuasively that while offence is, indeed, a less bad thing than harm, it constitutes a distinct category, and the two should not be placed on the same scale. Offence can perhaps reach on occasion a level of seriousness which causes harm to the victim, so the harm principle will become applicable. But those cases are the exception, not the norm. Usually, the effect of offence on the victim is an unpleasant emotional experience and ‘the offended mental state in itself is not a condition of harm’.12
Another way of explaining the distinct nature of offence is that, unlike harm, it does not affect negatively the victim’s opportunities for pursuing her interests and goals.13 When one suffers physical injury or one’s property is damaged, one’s resources are diminished, making it more difficult or impossible to engage in activities and projects which give meaning to one’s life. Harm ‘is prospective rather than backward-looking: it involves a diminution of one’s opportunities to enjoy or pursue a good life’.14 Offence, by contrast, is about a specific experience, in the sense that it produces a disliked mental state without depriving us of future opportunities. The unpleasant feelings will usually disappear when the offensive conduct stops but even if they linger on they do not, as such, cause a ‘prospective loss of resources’.15
Conditions Legitimising Coercion
Feinberg divides offensive conduct in two broad categories. He uses the term ‘offensive nuisances’ to describe the unpleasant or annoyed mental states which are usually the result of an unavoidable attack on the observer’s senses; loud noises and foul smells are typical examples of such types of offence.16 As the term itself suggests, this category is modelled on the common law tort of nuisance, which protects an individual’s interests in her land. The aspect of the tort which is particularly relevant for the analogy with offensive conduct is that which protects against activities which do not cause physical damage to land but interfere with the landowner’s peaceful enjoyment of it. In the common
12 Feinberg, Offense to Others (n 5) 3.
13 A P Simester and Andreas von Hirsch, Crimes, Harms, and Wrongs (Hart Publishing 2014) 109.
14 ibid.
15 ibid.
16 Feinberg, Offense to Others (n 5) ch 7.
law, noise, smoke, smells, and vibration are typical instances of this type of nuisance, which cause ‘personal inconvenience’ and affect ‘the senses or the nerves’.17
However, offensive conduct may strike at one’s sensibilities, moral, religious or other. Then the experience of offence is not aroused by an attack on the senses but tends to take a more deeply felt character, and usually persists even after the act which initially caused it has ceased. For example, one may be insulted by the burning of the national flag, the desecration of a religious symbol, or certain forms of sexual activity. Often, the offence is caused by the mere thought that those things have taken place, without the offended party having witnessed them. It is the fact that one’s idea of patriotism or religious devotion, or sense of sexual propriety, have been disturbed that arouses the feeling of insult, and consequent injury. Those are instances of ‘profound offence’, in which the offence derives mainly not from the perceptual experience of the offending act but from its cognitively mediated content, which is what is really considered by the observer to be an affront to her sensibilities.18 Offensive speech belongs to this category.
Feinberg notes that for the offended person both types of offence are ‘objectionable for roughly the same kind of reason . . . they are annoying distractions, unwelcome demands on one’s attention, a bother that must be coped with however inconvenient it may be at the time to do so’.19 While he is right about the similarity of the reasons which make them objectionable, the explanation seems to leave something out. I think that the similarity lies in the fact that both types of offence are underpinned by an encroachment on boundaries.20 Offensive nuisances breach a boundary around one’s person, which is perceived with one’s senses and defines a space where one expects to be undisturbed by unwanted physical stimuli. Offence to sensibilities violates an internal boundary, intruding into one’s mental space which is shaped by conscious and unconscious processes. Here, offensive conduct disturbs one’s peace of mind, stirring up thoughts and emotions which one would have liked to avoid.
There are three ways in which the law may respond to offensive behaviour: leave it completely unregulated; regulate, and usually prohibit, certain, but not all, offensive acts; or prohibit all types of offensive behaviour. The third way is not an option for a liberal political system as it leaves very little room
17 St Helen’s Smelting Co v Tipping (1865) 11 HL Cas 642, 650.
18 Feinberg, Offense to Others (n 5) 14, 51.
19 ibid 9.
20 On the significance of boundaries see Neu, Sticks and Stones (n 3) ch 1.
for individual liberty, while the first is not usually followed because even liberal systems do not subscribe to John Stuart Mill’s principle that only harm to other people can justify punishment. Had that view been adopted, the infliction of mere offence would never be punishable. The second model is the most common, with the law prohibiting some offensive activities on the basis of a set of criteria which vary from case to case but often include the seriousness of the offence, the number of people affected, and the specific issue to which offence relates.
Most instances of regulation of offensive nuisances will be uncontroversial and take place through private law or administrative regulations. The prevention and punishment of offence to sensibilities presents more complicated issues. An obvious objection is that it is a form of moralism since it allows the state to punish people not for harming others but for making them upset or distressed.21 Of course, all legal systems enforce morality regularly but we do not object to it because usually the immoral act also causes harm to others. Murder is morally wrong and harms the victim, so its punishment falls within the harm principle. When, however, the alleged immorality does not involve harm to another person the law’s intervention aims at the enforcement of morality as such. Put differently, if we allow offence to count as a legitimizing reason for state coercion we have abandoned the liberal premise that harmless activities should not be prohibited because some people consider them (rightly or wrongly) as immoral. Feinberg has responded to the objection by clarifying and qualifying the offence principle through two ‘mediating’ criteria: the seriousness of the offence and the reasonableness of the offending behaviour. I will not discuss his theory in detail here, but I will give a brief description, which, although it does not do justice to its subtlety, will help to clarify what is usually meant by the thesis that the prevention of profound offence to one’s sensibilities legitimizes the use of state coercion.
The seriousness of the offence, Feinberg says, is to be determined by four factors: its magnitude, whether it can be reasonably avoided, whether the offence (or the risk of being offended) was voluntarily taken, and whether the offended state is the result of one’s abnormal susceptibility to offence.22 The magnitude of the offence depends, in turn, on its intensity, duration, and extent: the more intense, extensive, and long-lasting the offence, the more serious it is considered to be. All aspects of the criterion of seriousness relate to the
21 Feinberg discusses legal moralism in detail in Harmless Wrongdoing (Oxford University Press 1990).
22 Feinberg, Offense to Others (n 5) 35.
person being offended and the way she has experienced the offence; some of them, such as the intensity and the possible abnormal susceptibility to offence are predominantly subjective in nature, while others such as the duration and reasonable avoidability of the offence are of a more objective character.
Once the seriousness of the offence has been assessed, the focus shifts to the offensive conduct and the person engaging in it in order to determine its reasonableness. Again, Feinberg gives a list of factors against which reasonableness is to be measured: the importance of the offensive conduct to the actor himself, its social utility, the existence of non-offensive alternative opportunities to engage in the conduct, the actor’s motives, and the nature of the locality where the conduct takes place. Those general factors are supplemented by a subject-specific one—freedom of expression—because of the great social value and the importance which speech has for individuals and society.23
What follows, then, is an exercise where the legislator or the judge is expected to place the seriousness of the offence and its reasonableness on the scales and ‘read the balance’.24 In some cases, the application of one of the criteria will be enough to tilt the scales one way or the other. The typical example is consent. If the offended person has accepted the offensive conduct or the risk of being offended, he cannot claim the protection of the law regardless of the relative weight the other factors possess: ‘the Volenti standard preempts all the rest when it clearly applies’. 25 In cases where consent does not apply, the other criteria must be considered. Freedom of speech also has increased weight, albeit not of the same preemptive force as consent. The presumption in favour of speech covers its content fully, in the sense that, no matter how offensive an opinion is for some people, the state is not morally justified in using its coercive force to protect the offended party. The manner of expression, however, does not share this privileged status, and can be properly balanced against the other criteria.26 Thus, there is an internal moral hierarchy of the various
23 Feinberg does not include a reasonableness requirement in his test for the determination of the seriousness of the offence.
24 Feinberg, Offense to Others (n 5) 44.
25 ibid 45.
26 ibid 44. The distinction between the content of the speech and the manner in which it is expressed is a familiar one from First Amendment law, and I will return to it in later chapters. Restrictions based on the content of speech are subject to strict scrutiny while content-neutral restrictions are subject to intermediate scrutiny, which leaves more room for state regulation of speech. Classic examples of content-neutral measures include time, place, and manner restrictions. See eg Clark v Community for Creative Non-Violence 468 US 288, 293 (1984): ‘Expression, whether oral or written or symbolized by conduct is subject to reasonable time, place and manner restrictions . . restrictions of this kind are valid provided that they are justified without reference to the content of the regulated speech, that they are narrowly tailored to serve a significant governmental interest, and that they leave open ample alternative channels for communication of the information.’
considerations which the decision-maker must weigh: consent has absolute preemptive force and where it applies trumps the other criteria and offers unqualified protection to the offensive conduct; free speech ranks below consent but above the other criteria as the content of the expression is privileged regardless of its offensiveness but the manner of the expression is not; finally, the remaining considerations carry relative weight depending on the circumstances of each case.
The use of the terms ‘balancing’ and ‘balance’, and the metaphor of the scales, are both familiar to lawyers, as the judicial process is often described as a balancing exercise where the judge has to weigh competing interests. In constitutional adjudication balancing has been used to explain how constitutional courts resolve claims involving conflicts between rights (for example speech vs. privacy) or rights and legitimate public interests (such as public order or the confidentiality of judicial proceedings).27 Balancing does not imply that there is an automatic mathematical way of coming to a clearly correct answer; what the model does is to identify all the morally relevant considerations and the way they should be applied, and place them before the decision-maker, who needs to exercise her judgment when dealing with specific cases. It is always possible that she may strike the balance erroneously, in which case we can say that her decision is morally wrong but not illegitimate in the sense of taking into account impermissible factors.28
Feinberg’s theory starts from a liberal point of view and his aim is to establish appropriate limits to the desire of governments to regulate offensive conduct. One may doubt whether the project—constructing an offence principle which legitimizes coercion against some offensive conduct without collapsing into moralism—can succeed. It may not be possible to remain true to the type of the liberal view of the law Feinberg defends while maintaining at the same time that the law should protect one’s moral sensibilities from offence.29 Thus, A P Simester and Andreas von Hirsch have recently criticized his offence principle
27 For discussions of balancing in constitutional and human rights law see generally Francisco Urbina, A Critique of Proportionality and Balancing (Cambridge University Press 2017); Jacco Bomhoff, Balancing Constitutional Rights: The Origins and Meaning of Postwar Legal Discourse (Cambridge University Press 2013); Stephen Gardbaum, ‘A Democratic Defence of Constitutional Balancing’ (2010) 4 Law and Ethics of Human Rights 79; Stavros Tsakyrakis, ‘Balancing: An Assault on Human Rights’ (2009) 7 International Journal of Constitutional Law 468; Robert Alexy, ‘Balancing, Constitutional Review, and Representation’ (2005) 4 International Journal of Constitutional Law 572; T Alexander Aleinikoff, ‘Constitutional Law in the Age of Balancing’ (1987) 96 Yale Law Journal 942; and Louis Henkin, ‘Infallibility under Law: Constitutional Balancing’ (1978) 78 Columbia Law Review 1024.
28 Feinberg, Offense to Others (n 5) 45–46.
29 See Larry Alexander, ‘Harm, Offence and Morality’ (1994) 7 Canadian Journal of Law & Jurisprudence 199; Anthony Ellis, ‘Offense and the Liberal Conception of the Law’ (1984) 13 Philosophy and Public Affairs 3.
as being too broad and relying excessively on the consequences of offending acts.30 They suggest that an offence principle can legitimize coercion through the criminal law only if two additional requirements are met: the offending conduct constitutes a wrong31 and involves, ultimately, some type of harm. The latter condition relates to situations where offensive conduct which is not harmful per se might cause indirect harms to others.32 This leads to an overlap of the offence and harm principles.
The fact that Simester and von Hirsch find it necessary to invoke harm within the offence principle shows how difficult it is to justify the prohibition of offensive conduct which insults a person’s sensibilities. In the case of offence caused by speech, the difficulty is greater because of the special position speech occupies in the range of human activities.
Offence Caused by Speech
My concern in this book is with offensive speech and, specifically, speech which offends religious sensibilities. I use the term ‘speech’ in a broad sense, encompassing not only spoken or written words but also expressive conduct in the form of acts undertaken, wholly or primarily, for their symbolic content. Many of the things we do in everyday life carry some sort of symbolism but they would not necessarily qualify as speech in this sense because the expressive element is not the dominant factor. The US Supreme Court has provided a useful test for expressive conduct: whether ‘an intent to convey a particularized message was present, and [whether] the likelihood was great that the message would be understood by those who viewed it’.33 A range of offensive behaviour with symbolic significance would not satisfy the test. For example, some people may be offended by the thought that others have extramarital or premarital or homosexual sex. Those sexual acts are not a form of speech because, although they have a symbolic meaning concerning sexual mores, they are not undertaken by the actors with the main aim of conveying a message. But sometimes
30 Simester and von Hirsch, Crimes, Harms, and Wrongs (n 13) pt III.
31 As we saw earlier in this chapter, Feinberg also takes wrong to be a necessary element of the actions covered by his offence principle, referring to offence ‘caused by the wrongful (right-violating) conduct of others’. Simester and von Hirsch criticize him for indirectly abandoning this requirement when he states that ‘there will always be a wrong whenever an offended state . . is produced in another without justification and excuse’. See Feinberg, Offense to Others (n 5) 2. They suggest that ‘this supposes a general, prima facie duty not to generate offended mental states in others: consequently, the general requirement for a wrong disappears’: Simester and von Hirsch, Crimes, Harms, and Wrongs (n 13) 96.
32 Simester and von Hirsch, Crimes, Harms and Wrongs (n 13) 112–17.
33 Spence v Washington 418 US 405, 410–411 (1974).
