Article by Amal Clooney & Philippa Webb
Amal Clooney is a barrister at Doughty Street Chambers and a Visiting Professor at Columbia Law School. Philippa Webb is a barrister at 20 Essex Street Chambers and Reader (Associate Professor) in Public International Law at King’s College London. We thank Matthew Nelson, Anna Bonini, Katarzyna Lasinska, Raphaelle Rafin, Tiffany Chan, Deborah Tang, Ollie Persey, and Mirka Fries for excellent research assistance. We are grateful to Professor Guglielmo Verdirame, Professor Michael Posner, Professor Vince Blasi, and Nani Jansen for comments.
This Article originally appeared in Volume 48.2 of the Columbia Human Rights Law Review. A PDF of this Article may be found here or by clicking on the title above.
States all over the world are enacting new laws that criminalize insults, and using existing insult laws with renewed vigour. In this article, we examine state practice, treaty provisions, and case law on insulting speech. We conclude that insulting speech is currently insufficiently protected under international law and regulated by confused case law and commentary. We explain that the three principal international treaties that regulate speech provide conflicting guidance on the right to insult in international law, and the treaty provisions have been interpreted in inconsistent ways by international courts and United Nations bodies. We conclude by recommending that international law should recognize a “right to insult” and, drawing on U.S. practice under the First Amendment, we propose eight recommendations to guide consideration of insulting speech in international law. These recommendations would promote coherence in international legal standards and offer greater protection to freedom of speech.
Freedom of speech is under attack. States all over the world are enacting new laws that criminalize insults and are using existing insult laws with renewed vigour. Today, in many states, it is a criminal offence to insult royalty, rulers, or religion. 1 And prosecutions for such insults are on the rise. The number of journalists who are being imprisoned across the world is, today, at its highest point in over twenty-five years. 2
Under international law, speech is primarily regulated by three treaties: the International Covenant on Civil and Political Rights, the Convention on the Elimination of Racial Discrimination, and the Convention on the Prevention and Punishment of the Crime of Genocide. What emerges from the wording of these legal provisions is that, at one end of the spectrum, speech that intentionally incites genocide must be criminalized, and speech inciting violence may be criminalized. At the other end of the spectrum, speech that is merely disturbing or shocking should not be a crime. 3
In between these two poles is insulting speech that incites hatred, hostility or discrimination, but not violence. This falls into a more controversial grey zone, but international human rights bodies currently allow states to imprison individuals for such speech. 4 This Article argues that international standards should be interpreted or amended to provide that such speech should not be criminalized. Instead, imprisonment for speech should be reserved for the most dangerous forms of insulting speech that intentionally and directly incite imminent or concretely-identified violence or criminal offences.
I. State Practice on Insulting Speech
States all around the world are imprisoning journalists and others for insults. Insults may result from a lack of awareness or indifference as to how the words will be received, or they may be deliberately provocative and attention-seeking. Insulting speech may stigmatise minorities and be harmful to relations between communities. But it can also challenge those in power or blow the whistle on unlawful conduct. In many jurisdictions such speech is punished under the guise of national security or blasphemy laws, simply to silence dissent. Indeed, it is currently a crime in many countries around the world to insult one of the three “Rs”—rulers, religion, or royals—and people are being prosecuted for such insults in criminal and military courts.
A. Insulting Rulers
- Defamation Laws
In most countries, a person who harms another person’s reputation through an oral or written insult can be sued for defamation in the civil courts and may have to pay damages to the offended party. But to protect free expression, many laws on defamation have exceptions for certain types of speech. This includes speech that is true, speech that reflects an opinion, and speech about a matter of public interest. These exceptions are intended to ensure that political speech remains protected from the civil penalties that would otherwise apply.
But in many countries, defamation can also be a criminal offence, without exception for political speech. Criminal defamation laws are widespread around the world, still exist in a number of U.S. states, 5 and are prevalent in Europe. Twenty-three of the twenty-eight member states of the European Union (E.U.) criminalize defamation, 6 and in twenty countries imprisonment—as opposed to fines—is the penalty. 7
Prosecutions under criminal defamation laws are routine even in some European countries. 8 In the last five years alone, journalists in fifteen E.U. countries have been convicted of defamation in criminal trials. 9
In addition to criminal defamation laws that apply to all types of insults, many countries have criminal sedition laws that make it a crime to insult the government specifically. Throughout history, sedition laws have been used to silence minority views and have thus been called “the hallmark of an unfree society.” 10 Gandhi, Galileo, and Nelson Mandela were all at one time prosecuted for sedition. 11
Indian law, even today, criminalizes a speaker who “brings or attempts to bring into hatred or contempt, or excites or attempts to excite disaffection towards the Government”. 12 Other countries have laws with the same elements but a different name. For example, most Arab countries have laws that make it a crime to “instigat[e] hatred and disrespect against the ruling regime”. 13 Similarly, China criminalizes “incitement of subversion of State power”.14 And in Europe, many states have laws that make it a crime to insult the head of state or other political figures. For instance, Cyprus makes it a crime to insult the army or a foreign head of State. 15 In Denmark, attacking a public official “with insults, abusive language or other offensive words or gestures” is punishable by up to six months’ imprisonment. 16 In Germany, defaming the president can result in a prison term of up to five years. 17
These laws do not simply exist on paper: prosecutions for sedition are taking place all over the world. In Malaysia, a cartoonist named Zunar faces up to forty-three years for sedition for posting nine cartoons on Twitter that mocked the government and its sham trial of opposition leader Anwar Ibrahim. 18 In India, an actress was charged with sedition after she stated in a political speech that “Pakistan is not hell, people there are just like us.” 19 In Botswana, an editor faces three years in prison for sedition because a story in his newspaper claiming that President Ian Khama failed to report a road traffic accident 20 was deemed “likely to cause disaffection or hostility towards the head of state or the government.” 21 And in Turkey, over two thousand people have been prosecuted in a two-year period since Recep Tayyip Erdoğan became President for the crime of insulting him. 22
- Anti-Discrimination Laws
Anti-discrimination laws that criminalize “hate speech” targeting minorities are also widespread, and can be applied to a broad range of political speech. Such laws exist all over the world, including in Europe, and they often carry criminal penalties. South Africa, which already allows civil penalties for expressions that “promote or propagate hatred” is also considering a new law that would criminalize hate speech. 23 The speech would be defined broadly as a direct or electronic communication that “advocates hatred” and incites others to harm or causes contempt or ridicule. 24 The law would punish an offender with up to ten years in prison.25
Such laws are routinely used to prosecute speakers around the world. In the Netherlands, a far-right politician was convicted for hate speech after he called for “fewer Moroccans” in the country. 26 In Germany, the government threatened to prosecute Facebook if it did not remove hate speech targeting Jews from the website swiftly enough. 27 And in Turkey, a former Minister and Member of Parliament who took part in a meeting about human rights in which he criticized governmental policies was convicted of “incitement to hatred and hostility.” 28
- Terrorism Laws
Many countries’ terrorism laws are worded in such vague terms that they capture speech that is merely insulting to the state’s officials or even private actors. For instance, in Saudi Arabia,
anti-terror legislation criminalizes “insult to the reputation of the state”. 29 And in Kyrgyzstan, Article 11 of the Law on Countering Extremist Activity prohibits the dissemination of extremist materials that call for a “breach of national dignity”. 30
In Tunisia, a journalist was charged with terrorism for publishing a photograph of Seifeddine Rezgui, the perpetrator of a terrorist attack that killed thirty-eight foreigners, emerging from a car before he began shooting. 31 In Russia, a blogger received a two-year sentence for conveying support for the ideology and practices of terrorism over an online post entitled “Who Did Putin’s Falcons Actually Bomb?” that criticized Russia’s military actions in Syria. 32 And in Spain, two puppeteers were arrested on charges of “glorifying terrorism” because their Madrid Carnival puppet show depicted a policeman trying to catch a witch who was holding up a sign linking herself to Al Qaeda and the secessionist group ETA. 33 The puppeteers responded that their arrest proved their show’s point—that the country’s counterterrorism laws are used for witch-hunts. 34
- Fake News Laws
Many countries make it a crime to publish “fake” or “false” news, which can in turn be interpreted by prosecutors and judges as being any information that the government considers to be insulting. Many countries in the Arab region include this offence in their penal codes, 35 and some countries in South Asia provide similar penalties. 36
More of these laws are being adopted around the world. In 2013, The Gambia introduced a new offence of “spreading of false news against the government or public officials” punishable by up to fifteen years in prison or a fine of sixty-four thousand euros. 37 In China, a 2013 law established that it should be a crime to be a rumour-monger, defined as an individual who intentionally posts a false rumour that is reposted five hundred times or more, or viewed five thousand times or more. 38 And in Qatar, a 2014 law makes it a “cybercrime” to create a website that spreads false news in order to jeopardise the safety or “general order” of the state. 39 Egyptian prosecutors used a similar law in 2014 to prosecute the entire senior staff of Al Jazeera television for broadcasting “false news . . . abroad regarding the internal situation in the country” even though there was no evidence that any of the broadcasts contained material that was unproven or falsified. 40
- Public Order Laws
Many countries criminalize disturbances of public order, but then use such laws to criminalize speech that is insulting to the ruling authorities. For instance, in Russia the all-female band Pussy Riot was sentenced to two years’ imprisonment for hooliganism for singing a punk song which opened with the line: “Virgin Mary, Mother of God, banish Putin, banish Putin, banish Putin!” 41 And in China, a well-known free speech activist named Yang Maodong was sentenced to six years’ imprisonment for “gathering a crowd to disrupt order in a public place” and “picking quarrels and provoking trouble” after he held up banners calling on public officials to disclose their assets and for China to ratify the International Covenant on Civil and Political Rights. 42
- Other Laws Ostensibly Unrelated to Speech
In many politically-motivated cases, individuals are being pursued because of speech that is insulting to the ruling regime, but they are charged with crimes that are ostensibly unrelated to speech. For instance, in Azerbaijan journalists have routinely been charged with crimes such as tax evasion, illegal entrepreneurship, and abuse of power in cases where the European Court of Human Rights has later found that there was no reasonable suspicion of a criminal offence. 43
B. Insulting Religion
Insulting religion, like insulting rulers, is widely criminalized around the world. Although blasphemy has been de-criminalized in many Western countries, it remains an offence in many others and it is often expressed in vague terms that cast a wide net over insulting speech. It is, for example, a crime to “incit[e] religious unrest” in Turkmenistan, and to “promote division between religious believers and non-believers” in Vietnam. 44 Other countries criminalize “contempt of heavenly religions”, “outraging religious feelings”, “promoting one’s own individual opinion on issues that are in disagreement among Islamic scholars”, “inciting people to disputes”, and “talking about religions other than Islam”. 45
Prosecutions for blasphemy are routine in some countries and carry the harshest penalties. In Saudi Arabia, a blogger named Raif Badawi published atheist views online and was sentenced to ten years in prison and one thousand lashes for insulting Islam. 46 In Iran, members of the heavy metal band Confess face execution for supposed “satanic” lyrics. 47 In Myanmar, a café owner from New Zealand was sentenced to two-and-a-half years’ imprisonment and hard labour for insulting Buddhism by using an image of Buddha wearing headphones in a post on Facebook. 48 In Singapore, a teenager was convicted of “wounding religious feelings” and disseminating obscene images after he posted a video containing “remarks against Christianity”. 49 And in Greece, a man who created a satirical Facebook page comparing a deceased Orthodox monk with a pasta dish was charged with “malicious blasphemy and insulting religion” and sentenced to ten months in prison. 50
C. Insulting Royalty
In some countries, it is not only criminal to insult rulers or religion; it is also a crime to insult royals. Thailand considers it a crime—known as lèse-majesté—to insult the monarchy, and it is punishable by up to fifteen years in prison. 51 Malaysia and Morocco have similar laws. 52 Even in Europe, insulting the King or Queen is punishable by up to five years in prison in countries such as the Netherlands 53 and Norway. 54
Such laws have been used to impose severe penalties in a number of recent cases, particularly where the trials are administered by military courts. For instance, a factory worker in Thailand who posted a sarcastic message about the King’s dog on the internet currently faces up to thirty-seven years in prison for lèse-majesté. 55 And in Morocco, an eighteen-year-old was convicted of this crime after he changed the national slogan on a school blackboard from “God, the Nation, the King” to “God, the Nation, Barca”—a tribute to his favourite football club. 56
II. Legal Provisions Governing Insulting Speech in International Law
A. The International Legal Framework
The international law framework governing freedom of expression is contained in treaty law: Articles 19 and 20 of the International Covenant on Civil and Political Rights (ICCPR),57 Article 4 of Convention on the Elimination of Racial Discrimination (CERD), 58 and the specific prohibition on “direct and public incitement to commit genocide” in Article III(c) of the Genocide Convention. 59 There are also relevant but broadly-worded provisions on freedom of expression in the Universal Declaration of Human Rights (UDHR). 60
Implementation of the ICCPR and CERD is monitored by U.N. treaty bodies, 61 and disputes arising under the Genocide Convention can be adjudicated by the International Court of Justice.62 These treaties exist alongside regional conventions that apply in many countries, including the European, African, and Inter-American human rights treaties. States’ obligations under these treaties are supervised by the respective regional courts.
- Legal Protections for Speech in International Law
i. International Treaties
The principal international human rights treaties, including the ICCPR and CERD, were drafted by diplomats from around the world as part of the inter-state negotiations that led to the creation of the United Nations and the adoption of the UDHR in the years following World War II. The first drafters were a small committee led by Eleanor Roosevelt, and over two decades various drafting committees contributed to the negotiations until the text was submitted to the U.N. General Assembly for final consideration and adoption. 63
The drafting history of the provisions on free speech shows that the delegates negotiating the language of these treaties had in mind the recent experience with Nazi Germany and were keen to prevent the dissemination of totalitarian and racist theories by propaganda.64 In many of the discussions, free speech advocates pushed for less sweeping language that would ensure that speech could only be criminalized if it incited violence. 65 But the majority of diplomats who drafted these provisions believed that a prohibition of incitement to violence alone was not sufficient to prevent the recurrence of Nazi-style crimes, and that speech inciting “hatred” and “discrimination” should also explicitly be prohibited. 66 The treaties took years to negotiate, but eventually, in 1965, CERD was adopted by the U.N. General Assembly, and the ICCPR was adopted in the same way the following year.
The result was that the three principal legal provisions governing speech at the international level—Articles 19 and 20 of the ICCPR and Article 4 of CERD—allow insulting speech to be silenced if it constitutes discriminatory “hate speech” targeting minorities on the basis of race, nationality or religion even if there is no criminal intent or risk that it will lead to violence. Similar provisions are found in the regional human rights instruments applied by the human rights courts that operate in Europe, Africa, and the Americas.
a. Article 19 of the ICCPR
Article 19 of the ICCPR sets out the general “right to freedom of expression.” But this is not an absolute right. Article 19(3) provides for permissible restrictions on speech in the following terms:
The exercise of the [right to freedom of expression] carries with it special duties and responsibilities. It may therefore be subject to certain restrictions, but these shall only be such as are provided by law and are necessary:
(a) For respect of the rights or reputations of others;
(b) For the protection of national security or of public order (ordre public), or of public health or morals. 67
The drafting history of Article 19(3) shows that many states considered incitement to discrimination, hatred, and violence as limitations to free speech that could have been included in this provision, but ultimately these were considered redundant given that they were expressly provided for in Article 20. 68
b. Article 20 of the ICCPR
Article 20 of the ICCPR imposes further restrictions on the right to freedom of expression in the following terms:
1. Any propaganda for war shall be prohibited by law.
2. Any advocacy of national, racial or religious hatred that constitutes incitement to discrimination, hostility or violence shall be prohibited by law.
Article 20 establishes a positive obligation on states to prohibit speech that constitutes “propaganda for war” or “advocacy of” certain types of hatred through “adopt[ing] the necessary legislative measures prohibiting the actions referred to therein.” 69
This provision was first proposed by a Soviet diplomat during the ICCPR drafting negotiations in 1947. In his view, it would be “a powerful weapon . . . to restrict the dissemination of Nazi-Fascist propaganda.” 70 During the drafting process, various delegations opposed this on free speech grounds. The U.S. delegation, for instance, warned that “any criticism of public or religious authorities might all too easily be described as incitement to hatred and consequently prohibited.” 71 Similarly, the United Kingdom objected to vague terms such as “hatred” that are “not easy to define as a penal offence” and bore too great a risk of abuse. 72 By the time Article 20 came before the U.N. General Assembly, most nations were in favour of it, though some objected on the basis that it should be restricted to incitement to violence. It was ultimately adopted with fifty-two votes in favour, nineteen votes against, and twelve abstentions. 73
The U.N. Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression has clarified that Article 20 does not require states to criminalize speech, although according to the Rapporteur criminalization would be appropriate in “serious and extreme instances of incitement to hatred” determined according to a seven-part test. 74
c. Article 4 of CERD
The third speech-related provision in the international legal framework—Article 4 of CERD—requires states to outlaw racially-motivated hate speech:
States Parties condemn all propaganda and all organizations which are based on ideas or theories of superiority of one race or group of persons of one colour or ethnic origin . . . and, to this end, with due regard to the principles embodied in the Universal Declaration of Human Rights and the rights expressly set forth in article 5 of this Convention [which protects “[t]he right to freedom of opinion and expression”] . . . :
(a) Shall declare an offence punishable by law all dissemination of ideas based on racial superiority or hatred, incitement to racial discrimination, as well as all acts of violence or incitement to such acts . . . .75
Unlike Articles 19(3) and 20 of the ICCPR, the plain language of Article 4 of CERD (“shall declare an offence punishable by law”) requires the criminalization of speech. 76 The “due regard” 77 clause in Article 4 has been interpreted in different ways by states and scholars, possibly to try to alleviate or modify this requirement. 78 But the Committee on the Elimination of Racial Discrimination, the U.N. body that monitors implementation of CERD, avoids the balancing exercise that a “due regard” clause would usually entail. The Committee in its General Recommendation No. 35 instead emphasizes that Article 4 levies on states an obligation to “effectively sanction as offences punishable by law” any “dissemination of ideas based on racial or ethnic superiority or hatred, by whatever means” and even all “expression of insults, ridicule or slander of persons or groups or justification of hatred, contempt or discrimination” on the grounds of their race, colour, descent, or national or ethnic origin. 79 Such insults must, pursuant to Article 4, be qualified as a criminal offence. 80
d. Summary of Free Speech Provisions in International Treaties
There is a tension among the three key provisions in international human rights treaties governing speech. First, Article 20(2) of the ICCPR goes further than Article 19(3). Whereas Article 19 allows for a restriction of speech in the interest of “respect of the rights or reputations of others”, Article 20 requires the restriction of any speech that constitutes advocacy of national, racial or religious hatred—as long as it incites “discrimination”. 81 Article 4 of CERD then goes further than Article 20 of the ICCPR by requiring the criminalization of certain “hate speech” of a racist nature. In addition, whereas Article 20 of the ICCPR requires that advocacy of hatred must lead to “incitement” of discrimination, hostility or violence, under CERD no incitement is needed—any racist “ideas” must be subject to a criminal ban.
Each of these human rights treaties has been widely ratified 82 and many of their provisions now reflect customary international law. However, a large number of states have entered reservations to the articles on freedom of expression in the ICCPR and CERD. 83 Seven states have entered declarations or reservations to Article 19 of the ICCPR and seventeen states have entered declarations or reservations to Article 20. Most of these reservations are based on a desire to expand protections for speech or to preserve existing broadcast licensing arrangements, so no other states have objected to these reservations. 84 For instance, the U.S. entered a reservation to Article 4 of CERD, stating that it “does not accept any obligation under this Convention, in particular under Articles 4 and 7, to restrict [the right to free expression and other rights] . . . protected by the Constitution and laws of the United States.” 85 Similarly, Denmark, Finland, and Iceland have objected to Article 20(1) on the basis that “a prohibition of propaganda for war could limit the freedom of expression.” 86
As Judge Meron stated in a case at the International Criminal Tribunal for Rwanda involving speech that was alleged to have incited genocide:
The number and extent of the reservations [to the speech-related provisions in international treaties] reveal that profound disagreement persists in the international community as to whether mere hate speech is or should be prohibited, indicating that Article 4 of the CERD and Article 20 of the ICCPR do not reflect a settled principle [on insulting speech]. Since a consensus among states has not crystallized, there is clearly no norm under customary international law criminalizing mere hate speech. 87
This means that the requirements of international human rights law for each state will be determined on a case-by-case basis, with regard to the precise treaty obligations in place for each country.
ii. Regional Human Rights Treaties governing Free Expression
The equivalent of Article 19 of the ICCPR in regional human rights treaties is Article 10 of the European Convention on Human Rights (ECHR), 88 Article 13 of the Inter-American Convention, 89 and Article 9(2) of the African Charter. 90 Article 10 of the ECHR has a longer list of what may constitute a permissible restriction on speech than Article 19(3) of the ICCPR, but there is no equivalent in the ECHR to the speech-restrictive provisions in Article 20 of the ICCPR nor Article 4 of CERD.
The European Court has, however, used Article 17 of the ECHR to restrict speech. This is a general provision that provides that:
Nothing in this Convention may be interpreted as implying for any State, group or person any right to engage in any activity or perform any act aimed at the destruction of any of the rights and freedoms set forth herein or at their limitation to a greater extent than is provided for in the Convention.
Article 17 has been interpreted to mean that speech which is so odious that it could not possibly be protected under Article 10 of the Convention can be dealt with under Article 17, which allows a case to be struck out without examination of the merits. This is a drastic “guillotine” 91 provision because it does not involve any balancing of the right to free expression against the other values protected in Article 10. 92
iii. Other International Law Sources relating to Free Expression
In addition to international and regional treaties, the meaning of the treaty law on freedom of expression has been elaborated upon by various non-treaty sources which add some detail to the applicable legal standards for speech. These include U.N. mechanisms, 93 regional courts and commissions, and NGOs. 94
Some soft law sources have sought to limit the permissible restrictions on insults under these treaty standards. For instance, the United Nations, through soft law, 95 has reached a high degree of consensus in rejecting the concept of “defamation of religions” and finding blasphemy laws incompatible with freedom of speech. 96 Other soft law guidance is also more protective of speech than the treaty framework, 97 and has helped put some flesh on the bones of the terms “hatred,” “discrimination,” “hostility,” “violence,” and “incitement.” 98 The U.N. High Commissioner has counselled that States’ legislation should include “robust definitions” of key terms like “hatred,” “discrimination,” “violence,” and “hostility.” 99 And NGOs like Article 19 have sought to define these terms. 100 But such guidance is piecemeal and does not go far enough. Much of it is buried in long U.N. reports or on NGO websites that are not widely accessed, and in some cases their authority is uncertain. In addition, such guidance is not binding, nor is it universally applied.
III. When Can Insults be Criminalized under International Human Rights Law?
Legal provisions in international treaties governing speech are applied in individual cases by regional courts and U.N. human rights bodies, and are subject to commentary from various U.N. sources. According to this international case law, the issue of whether an insult can be criminalized is highly dependent on context. 101 An insult made at a private dinner may have a different impact from the same insult made by a presidential nominee at a public rally or by a religious leader in a neighbourhood where sectarian tensions are running high. Speech, in other words, has to be looked at in context.
The context of an insult can be distilled into seven factors: (i) what was said, (ii) who said it and to whom, (iii) how was it said, (iv) when was it said, (v) where it was said, (vi) what intent the speaker had, and (vii) what impact the statement had. 102 Each of these factors will be addressed in turn.
A. What was said?
Under international human rights law, the content, nature, and tone of a statement affects its level of protection. Political speech, or commentary on public figures, for instance, generally receives more protection than other speech. 103 So public officials, including heads of state, are expected to withstand more extensive public criticism than ordinary citizens. 104
Satirical speech is also given special protection as it is a form of artistic expression and social commentary. 105 Any interference with a speaker’s right to use these forms of expression must therefore be assessed with particular care. 106 The European Court upheld this principle in Eon v. France when it reviewed the conviction of a man who had been charged with the crime of “insulting the President” for waving a placard reading “Casse toi, pauv’ con” 107 at then-President Sarkozy at a public event. This phrase was chosen because Sarkozy said the same words to a farmer who refused to shake his hand at an agricultural show. 108 The European Court held that the conviction violated the man’s right to free speech under the ECHR because the intention of the comment was satirical and its criminalization would have a deterrent effect on free debate of questions of general interest. Additionally, the imposition of a criminal sanction was held to be disproportionate and unnecessary in a democratic society in a case involving satirical speech of this kind. 109
The fact that speech is political or satirical speech will not always mean it is protected, however. In many cases courts and human rights bodies have considered the nature of the speech but ultimately found that convictions for political speech, satirical speech, and even cartoons that are insulting or critical were appropriate under international or regional human rights law standards. 110
When it comes to the content of speech, the question of truthfulness is also relevant. The right to freedom of expression covers speech that is both true and false, 111 and as a result, under international law, states should not criminalize a statement simply because it is false, unless it also meets additional requirements by, for example, inciting hatred, hostility or discrimination. In addition, the U.N. Human Rights Committee has stated that truth should be a defence to defamation, “in particular penal defamation laws.” 112 It has also found that statements of opinion, as opposed to fact, should not be criminalized even if they are insulting because, unlike facts, which can be shown to be true, opinions “are not, of their nature, subject to verification.” 113
B. Who said it and to Whom?
According to cases applying international human rights law, the identity of the speaker also matters. His or her relationship with—and influence over—the audience will inevitably affect the power of his or her words. Issues include whether the speaker is a public or private figure, a politician or a journalist.
As a general rule, the more privileged and powerful the speaker is relative to the targeted person or group, the greater the potential harm. 114 Journalists and politicians have a special status because their speech contributes to public dialogue and awareness. Attempts to criminalize their speech will therefore be subject to stricter scrutiny 115 because such speech is in need of special protection. 116 But in other cases, this factor makes the court lean the other way: politicians 117 and journalists 118 have been said to have a special responsibility not to promote hate speech, and on that basis convictions for racist speech by such speakers may be more likely to stand.
If a politician is not the speaker but the target of the speech, this will not, however, protect him or her from criticism. Indeed, insulting political speech receives a higher degree of protection than other insults. As the U.N. Human Rights Committee, interpreting the ICCPR in its General Comment No. 34, has said:
[A]ll public figures, including those exercising the highest political authority such as heads of state and government, are legitimately subject to criticism and political opposition. Accordingly, the Committee expresses concern regarding laws on such matters [as] defamation of the head of state and the protection of the honour of public officials . . . . States parties should [also] not prohibit criticism of institutions, such as the army or the administration. 119
An illustrative case is Konaté v. Burkina Faso, 120 the first African Court on Human and Peoples’ Rights case to address freedom of expression. A journalist named Lohé Konaté published articles in a weekly publication calling a senior prosecutor “a rogue officer” and alleging that he was involved in money laundering. 121 A criminal court sentenced Konaté to twelve months’ imprisonment for the crimes of defamation, public insult, and contempt. 122 But the African Court found that this violated Article 9 of the African Charter (the equivalent of Article 19 of the ICCPR and Article 10 of the ECHR). 123 The Court held that although the interference with Konaté’s speech was “provided by law” and served the “legitimate objective” of protecting the honour and reputation of magistrates, 124 it was neither necessary nor proportionate as the target was a public figure, 125 and the author was a journalist performing his duties, meaning that a higher degree of tolerance was expected and there was no justification for criminal sanctions. 126
Other human rights courts and bodies have reached a similar result. In upholding the conviction of a public school teacher in Canada for allegedly anti-Semitic speech, the Human Rights Committee highlighted not only his leadership role but also the vulnerability of the target group, school children. 127 Similarly, in Handyside v. United Kingdom, the European Court showed deference to the United Kingdom’s ban on an allegedly scandalous book, taking into account that the intended audience of the speech were children. 128
But this is an area of slippery definitions. The development of technology means that anyone with a smartphone is potentially a “journalist” who can instantaneously spread their opinion on social media. 129 And the notion of a “public figure” in our digital age has become more diffuse. The European Court has, for example, included within the status of “public figure” members of a royal family 130 and a television actor on a long-running show. 131
C. How was it said?
When international courts look into the contextual factor of “how” insulting speech was delivered, this can encompass various sub-factors: whether the insult was made in public or private, or made in private and then leaked; whether it was oral or written; whether it was made online; and whether it was spontaneous or
In the European Court case of Gündüz v. Turkey, for instance, the applicant made insulting comments on a live television show and was convicted for inciting hatred and hostility on the basis of religion. 132 He was fined and sentenced to two years’ imprisonment. 133 The Court found that his conviction breached the ECHR partly on the basis that the insults were made on a live television program. 134 In its words, “the Court cannot overlook the fact that . . . the applicant’s statements were made orally during a live television broadcast, so that he had no possibility of reformulating, refining or retracting them before they were made public.” 135
D. When and Where was it said?
Insults in the immediate aftermath of an event, when emotions are raw, may be judged differently by a court than an insult made at a less volatile time. The passage of time may mean that such insults come to be viewed as part of a historical debate, and protected from criminalization on this basis.
In Perinçek v. Switzerland, for instance, the European Court considered whether the denial of the Armenian genocide could be criminalized on the basis that such denial was an insult to memory of the victims. 136 The applicant, a Turkish politician, said at a press conference in Switzerland in 2005 that “the allegations of the Armenian genocide are an international lie” and then denied the Armenian genocide on two further occasions. 137 The Swiss courts convicted him of the crime of “racial discrimination” and imposed a suspended sentence and a fine. 138 In determining that Mr Perinçek’s freedom of expression should not have been restricted, the Grand Chamber of the European Court highlighted that the events he spoke of occurred one hundred years ago and that the speech took place in Switzerland, which does not have a large Armenian or Turkish community. These factors were relevant to the Court’s finding that his speech was not likely to incite violence or racial hatred. 139
In contrast, in the Leroy case, the “when” and “where” factors swung the other way. 140 Leroy had published a cartoon in a French newspaper showing the burning twin towers in New York with a caption, “We have all dreamt of it . . . Hamas did it.” 141 The European Court approved Leroy’s conviction for condoning terrorism, and in doing so highlighted that the speech was made in the politically-sensitive Basque region of France where such extreme statements may very well foster instability. The Court also emphasised that the cartoon was published on 13 September 2001, just two days after the twin towers had been attacked. 142
E. What intent did the speaker have?
Cases applying international human rights law sometimes, but not always, require intent in the “advocacy of . . . hatred” 143 in Article 20 of the ICCPR and in the “incitement” to discrimination, hostility or violence under that article as well as Article 4 of CERD. 144 However, this requirement is not applied consistently or rigorously. 145 This means that insulting speech can be prohibited and even criminalized on very broad grounds, even where the speaker did not intend to bring about hatred, discrimination, hostility or violence.
The Human Rights Committee has on occasion suggested that an intent to incite discrimination, hostility or violence should be required if speech is to be criminalized under international standards. 146 And in its 2013 General Recommendation on racial hate speech the CERD Committee published new guidance that State parties should recognize as “important elements” of any offence of incitement “the intention of the speaker and the imminent risk or likelihood that the conduct desired or intended by the speaker will result from the speech in question.” 147 Nevertheless, both the CERD Committee and the Human Rights Committee have allowed convictions for hate speech to stand even where there was no analysis of intent, 148 including in recent years. 149
The European Court has been similarly inconsistent in its treatment of intent. In some cases, intent seems to be dispositive of whether insulting speech can be criminalized. For instance, in Jersild v. Denmark, the Grand Chamber of the European Court dealt with the conviction of a television presenter for taking part in a programme that included offensive racist speech, such as “a nigger is not a human being, it’s an animal, [and] that goes for all the other foreign workers as well, Turks, Yugoslavs and whatever they are called.” 150 The Court held that the conviction violated Article 10 of the ECHR because this speech was presented as part of a documentary about the Greenjackets, a group of racist youths based in Copenhagen who Jersild interviewed for a news program. 151 Mr. Jersild did not intend to “aid and abet” the dissemination of the Greenjackets’ racist views, and his conviction therefore violated the ECHR. 152 Similarly, in other cases where the Court finds that the speaker’s intent was to contribute to a genuine debate, this is deemed protected speech which should not be criminalized. 153
In some cases the European Court seems to require a specific intent to incite violence in order for speech to be criminalized; but in other cases an intent to incite lesser harms will do. 154 And in yet other cases the Court seems to lower the bar to a standard of recklessness 155 or neglects to analyse the intent of the speaker entirely. 156 For instance, the Court concluded that a far right politician’s conviction and six-month prison sentence in Romania for “nationalist chauvinist propaganda” could stand without any discussion of his intent. 157 The police had found posters with slogans like “Stop Romania becoming a country of Roma . . . Romania needs children not gays” at the politician’s home. 158 According to the Court, these slogans “by their content” aimed to instigate hatred against these minorities and “by their nature” could create “tensions” in the population. 159 This was considered sufficient for the Court to find that the speech was “contrary to the fundamental values of the Convention and a democratic society”, so the case was thrown out under Article 17 of the Convention without even considering the balancing exercise required under Article 10 that protects free speech. 160
F. What impact did the statement have?
Under international human rights law, the final contextual factor that is relevant to the question of whether insulting speech can be criminalized is the likely or foreseeable impact of an insult and whether such an impact is imminent or not. It seems clear from the case law of both the Human Rights Committee and the regional human rights courts that there is no need to show that the harm that was incited—the discrimination, hostility or violence—actually occurred. But some cases suggest that likely imminent harm may be necessary instead. 161
In an early draft of its General Comment No. 34, the Committee stated that, in order to come within the prohibition on speech in Article 20, an insulting statement must be “likely to trigger imminent acts”; but this language did not appear in the final draft. 162 Some of the case law of the Committee hints that there is a requirement to show that such harm was foreseeable 163 but this standard is not applied clearly or consistently. In its 2013 General Recommendation on racial hate speech, the CERD Committee published new guidance that State parties should recognize as “important elements” of any offence of incitement “the imminent risk or likelihood that the conduct desired or intended by the speaker will result from the speech in question.” 164 But in many cases both the Human Rights Committee and the CERD Committee have sanctioned criminal convictions for speech without explicitly taking into account whether a prohibited impact was foreseen, likely or imminent. 165
Confusion also reigns at the European Court when it comes to the relevance of the impact of insulting speech. Some cases seem to require that the speaker intends to incite violence and that violence might objectively ensue in order for the speech to lose the protection of Article 10 of the ECHR. For instance, in Gündüz, the Court suggested that merely calling for shariah law, without calling for violence to introduce it, could not be criminalized in Turkey without violating the ECHR’s free speech protections. 166 But in many other cases, there is no analysis of the intended or actual impact at all, and yet the Court finds that a conviction was permissible. 167
IV. Analysis and Criticism of the Current Position in International Law
A review of state practice shows that laws outlawing insulting speech which give effect to states’ obligations under Article 20 of the ICCPR and Article 4 of CERD 168 are prevalent around the world including in Europe, Canada, and Australia. 169 And prosecutions for such speech have led to an alarming number of journalists and others being imprisoned for it. 170 International courts and human rights bodies reviewing such speech analyze various contextual factors in determining whether such speech can be restricted, but in many cases such bodies have approved the imposition of civil or even criminal penalties being imposed at the national level. In the authors’ view, the right to insult is not sufficiently protected under international law. Laws outlawing insulting speech are inherently capable of abuse, as the analysis of state practice above has shown. Yet international standards have proved to be confusing and ultimately inadequate in reining in national laws that restrict free speech.
A. Insufficient Protection of Speech
Two treaty provisions in particular allow speech to be silenced on very broad grounds: Article 20(2) of the ICCPR and Article 4 of CERD. 171 Article 4 is even more restrictive of speech than Article 20 since it explicitly requires the outlawing of speech as a crime, rather than as a civil offence. 172 And it casts an even broader net over speech since it requires the criminalization of “all dissemination of ideas” that are racist, even if they don’t involve incitement to any specified harm. 173 This creates criminal liability which is very broad indeed and would capture a significant amount of speech in any society.
Although Article 4 is limited to speech that is discriminatory on racial grounds, this simply highlights the arbitrary nature of the standard. Why is racial discrimination more damaging than discrimination based on religious, ethnic or sectarian grounds, or political grounds or grounds of sexual orientation? Article 4 is not only oppressive in a free society—it also fails to protect the interests of victims of discrimination on an equal basis. 174
Such vague and inconsistent bases for restricting speech are inherently open to abuse by undemocratic authorities, and even for well-meaning bodies they lead to decision making that is unpredictable and muddled—leaving citizens confused as to what speech may result in a prison term being meted out. Such vague standards also make it possible for the jurisprudence of international bodies to be content-driven—meaning that a court can, consciously or not, tend to defend speech that it can itself tolerate while condemning that which it finds abhorrent to become a crime.
An analysis of the international case law on insults shows that these concerns are not merely hypothetical. Although the Human Rights Committee, the CERD Committee, the European Court of Human Rights, and other bodies take into account the context of speech in determining whether it can be criminalized, none of these bodies has consistently imposed strict requirements that would be protective of speech, such as an insistence that insulting speech must be accompanied by a requisite criminal intent 175 or a strict requirement that harm, or more specifically violence, will be its direct result. 176 Although these bodies have sometimes alluded to such requirements, they have not applied them strictly or consistently to protect free speech. The courts will also very rarely point to evidence of intent, or allude to empirical, sociological or historical data to back up its conclusions on what impact was objectively foreseeable. 177
Nor have international bodies shied away from commenting on the inherent “value” of the speech and whether the criticism at issue is “warranted”. 178 Indeed, the European Court seems to consider certain topics such as Holocaust denial to be unprotected speech per se, 179 even if no evidence of any specific intent has been shown.180 This has resulted in jurisprudence that tolerates a high level of criminalization of speech, and a suspicion that restrictions are more likely to be allowed where the court itself does not agree with the content or the viewpoint being expressed. This is particularly true in cases where the European Court throws out a case under Article 17 of the ECHR, without even weighing the value of free-speech against other interests protected under Article 10 of the Convention. 181
Such international jurisprudence casts a wide criminal net over speech that happens to be critical or insulting. Let us take as an example statements made by Donald Trump during the U.S. presidential election campaign, in which he referred to Mexicans as “rapists” 182 and suggested that all Muslims should be prevented from entering the U.S. 183 This can be said to advocate hatred on national and religious grounds and would certainly inspire discrimination, if not also hostility. Such speech is offensive and insulting, but that should not mean that the speaker should be subjected to legal penalties for saying it, let alone a prison term.
An approach that is more protective of speech—that recognizes that insults should not be met with prison terms—would not mean that there is no action taken against the scourge of racism and other forms of discrimination. Nor would such an approach preclude the criminalization of speech by violent terrorist groups. Where the very purpose of the group is violence, as with ISIS, the recruitment-speech could easily pass the test that requires an intent to incite imminent or concretely-identified violence and a likelihood that the speech will produce such violence. 184 But where the speech is merely insulting, criminal penalties should not apply.
Instead, proportionate civil penalties may be imposed by legislation, but there is also a myriad of non-legal tools available. First and foremost, such speech should be heard, debated, rebutted, and vigorously challenged. 185 This way, “bigoted speech is exposed by more speech that decries bigotry”. 186 Politicians and other public officials can play a crucial role in countering hate speech by publicly and formally condemning instances of hate speech. 187 Civil society organizations are also critical in responding to hate speech, by identifying the speech and providing online resources to educate users and expose the falsehoods that underlie hateful messages. For instance, the Southern Poverty Law Centre’s Intelligence Project in the U.S. publishes information and details of hate speech to not only aid potential disciplinary action, but to publicly name and shame groups and individuals who are responsible for such speech. 188
Technology can also be used to expose and counter hate speech effectively. 189 The growing amount of communication conducted online can be used for harassment, threats or insults, but it can also be used to create communities for constructive dialogue and exchange. A free and fearless press can communicate information and ideas to the public, including reporting on racism and intolerance. 190 Training and community initiatives can increase awareness and improve society’s understanding of the effects of racism and discrimination in the community. Even the CERD Committee has recently underlined the importance of an educational approach to eliminating racist hate speech, noting that it is often the product of “indoctrination or inadequate education”. The Committee therefore recommends “education for tolerance, and counter-speech . . . [as] effective antidotes to racist hate speech”. 191 This can include educating children about other religions to tackle religious discrimination. 192
The current approach, devaluing speech, is not only wrong in principle: it is not working. There is no compelling evidence that the criminalization of racially-inspired insults has led to a reduction in racially-inspired violent crime. On the contrary, a study on the effectiveness of criminal laws for hate speech prepared for the U.N. Office of the High Commissioner for Human Rights observed that “the criminal model is not an efficient tool when it comes to addressing the real causes of discrimination”, and that “[t]he massive presence of criminal regulations” does “not seem to have made a meaningful contribution to reducing racism or . . . discriminatory conduct”. 193 And despite robust criminal laws in many E.U. member states targeting hate speech, the European Parliament has recently conceded that hate speech and hate crimes are on the rise. 194 Yet international law allows such speech to be criminalized, and indeed, Article 4 of CERD requires that this be done.
Even from a policy perspective, criminalizing insults is counterproductive because hate speech prosecutions only serve to give a platform and increased publicity to the insulting speaker. This was certainly the case for far-right Dutch politician Geert Wilders, 195 and Armenian genocide-denier Doğu Perinçek, who appears to have traveled to places where he knew he would be arrested for the denial in order to get additional publicity for his cause. 196
It is time to rethink the current system, which is restricting too much speech unnecessarily. And this means that the treaty standards themselves—and their application in individual
cases—need to be clarified or amended.
B. Vague and Anachronistic Grounds for Restricting Speech
International law on free speech is also unduly vague and anachronistic. The general guidelines on protecting speech expressed in Article 19 of the ICCPR are similar to those in the European, African, and Inter-American treaties, setting out a general rule in favour of free speech set against specific categories that justify restrictions. 197 But many of these treaties were adopted more than fifty years ago, and were negotiated at a time in which the horrors of World War II were still fresh in the minds of their drafters and the negotiators. 198 They all allow restrictions on vague grounds such as “respect for the right of others” and anachronistic grounds such as restrictions “for the protection of morals” 199 and a prohibition on “propaganda for war.” 200 These terms are not well defined in the jurisprudence of the U.N. bodies, leaving far too much scope for abuse. The fears voiced by many states during the drafting of the ICCPR and CERD regarding free speech 201 have turned out to be
well-founded: the terms used are too vague and susceptible to abuse. 202
C. Conflicting Guidance among Human Rights Bodies
In addition to being vague and outdated, a close study of international legal sources reveals inconsistent guidance on the question of when international law permits insults.
International bodies have diverged about whether states’ defamation laws—criminalizing insults that harm a person’s reputation—are compatible with international law. At one end of the spectrum, the U.N. has spoken out strongly against criminal defamation, 203 finding that states should “consider” decriminalizing defamation 204 and that imprisonment should never be a permissible penalty. 205
In stark contrast to this position, the U.N. CERD Committee has stated that criminal defamation laws are compatible with international law. In a case involving Denmark, for instance, the Committee found that it did “not consider it contrary to . . . the Convention if . . . the provisions of [a country’s] criminal law [include] a general provision criminalizing defamatory statements” which are racist in nature. 206 The case came before the Committee after a construction worker in Denmark got into an argument about certain payments with his boss. The boss told the worker to “[p]ush off home, you Arab pig”, called him an “immigrant pig” and stated that “both you and all Arabs smell . . . [d]isappear from here, God damned idiots and psychopaths.” 207 The Committee concluded that a criminal defamation action was perfectly permissible and would have constituted an “effective remedy” for the construction worker. 208
Like the CERD Committee, but unlike the Human Rights Committee, the regional courts have been tolerant of penalizing insults through criminal defamation laws. For instance, the
Inter-American Court has held that criminal defamation is legitimate where it “meets the requirements of necessity and proportionality,” 209 including when it is necessary to protect the honour or reputation of another individual or guarantee the right to privacy. 210
In addition to inconsistent guidance on the question of when international law permits insults to living persons, international law also diverges on the issue of denial laws, which protect against insults to the memories of deceased victims of crime. For instance, states like Switzerland, Austria, and Germany make it a crime to deny the Holocaust on the basis that such denials insult the dignity of the victims of this genocide. While the Human Rights Committee has stated that “[l]aws that penalize the expression of opinions about historical facts are incompatible with the obligations that the Covenant imposes on States parties in relation to the respect for freedom of opinion and expression,” 211 the European Court has gone the other way and held that denial laws are compatible with free speech. So, for instance, when a well-known historian wrote a letter to the editor of a German weekly stating that the Nazis had not planned the Holocaust, that most German officers did not know about it, and that Hitler did not intend to murder the Jews, the European Court approved his conviction and three-month sentence for insulting “the dignity of the deceased.” 212
D. Contradictory Case Law within Human Rights Bodies
In addition to inconsistencies between courts and human rights bodies at the international level, an analysis of each body’s jurisprudence reveals an inconsistent approach.
As discussed more fully above, the Human Rights Committee and the European Court have in some cases insisted that, in order to be criminal, speech has to be intended to incite violence. 213 But at other times a lower level of intent or no specific intent is required. 214 Similarly, in some cases, the speech must foreseeably lead to imminent harm in the form of discrimination, hostility or violence; in others, only imminent violence will suffice; and in others still no showing of likely harm is required at all. 215
Given how prevalent the criminalization of insults is around the world, it would be helpful for international law to set a clear standard that is appropriately protective of speech. Yet this is not the position today. Not only is there a different answer being given at the regional levels and the international level, but also the principal U.N. bodies supervising compliance with the two main treaties governing insulting speech are at odds with each other. The case law within each body is also confused. This leads to questions of what reforms are needed and how they might be achieved.
E. What the Law Should Be
We must distil a clear and consistent standard for regulating the right to insult in international law. We recommend that there should be the narrowest possible basis for criminalizing insulting speech, inspired by the approach adopted by U.S. courts under the First Amendment to the U.S. Constitution.
Although First Amendment jurisprudence is beyond the scope of this article, the approach of the U.S. Supreme Court to criminalizing speech is instructive because it is far more protective of speech than the international and regional human rights sources that have been reviewed. First Amendment jurisprudence should consequently serve as a source of inspiration for international bodies that seek to be more protective of speech. Two main principles emerge from a review of U.S. case law that could be incorporated at the international level and by doing so would increase the protections for speech.
First, as a general rule, insulting speech should be protected unless it is intended to and likely to produce imminent violence or other unlawful behaviour. 216 Under U.S. law, the First Amendment has been interpreted by the Supreme Court to permit the restriction of speech in very few circumstances. Speech is presumed to be free, and government regulation of speech is justified only if the speech falls into a category that is unprotected by the First Amendment. When it comes to insulting speech, the three relevant categories are: fighting words, incitement to imminent lawless action, and true threats. 217 The doctrines are linked, and there is debate over whether one is the subset of the other. 218
“Fighting words” are words “which by their very utterance inflict injury or tend to incite an immediate breach of the peace.” 219 This category encompasses insults, but has not been heavily relied on in practice 220 and its contours are very narrow. Offensive words will not suffice; fighting words must “have a direct tendency to cause acts of violence by the person to whom, individually, the remark is addressed.” 221 For this reason, the Supreme Court did not consider the words ““Fuck the Draft” to be “fighting words” because this expression, found on a defendant’s jacket, was shown in public generally, not directed at any one person, and neither the defendant nor anyone else engaged in violence as a result of the expression. 222
The second category, the “incitement” exception to the First Amendment, was defined by the U.S. Supreme Court in Brandenburg v. Ohio. 223 The Court held that a prohibition on speech can only be upheld when (a) the advocacy of violence is directed to inciting imminent lawless action, and (b) is likely to produce such action. 224 Although “lawless” action can be read broadly, Brandenburg is often read narrowly to apply only to the advocacy of force. 225 In addition, the temporal requirement of imminence has strictly limited the application of the incitement doctrine. In NAACP v. Claiborne Hardware Co., the statement “if we catch any of you going in any of them racist stores, we’re gonna break your damn neck” was held to be “emotionally charged rhetoric” and an “impassioned plea” protected under the First Amendment. 226 The Court found that that the “mere advocacy of the use of force or violence does not remove speech from the protection of the First Amendment”—a more concrete connection between the speech and the violence was required. 227 In this case such a connection was not sufficiently strong because although acts of violence against boycott violators did occur after the speech was made,228 most occurred weeks or months later 229 and were not clearly caused by it. 230
The third category, “true threats,” concerns statements where “the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals.” 231 Although the speaker “need not actually intend to carry out the threat,” it must be sufficiently concrete. The prohibition is supposed to protect individuals from the fear of violence and the disruption caused by fear, as well as from the possibility that the threatened violence will occur. Publishing “Wanted” posters of physicians who performed abortions was on this basis found to be a “true threat,” 232 whereas a young man stating during a political rally that “if they ever make me carry a rifle the first man I want to get in my sights is [President Lyndon B. Johnson]” was found to be “political hyperbole” and thus permitted speech. 233
All three categories of unprotected insulting speech under U.S. law therefore require a concrete link to violence or breach of the peace if the speech is to be criminalized.
The second principle that can be drawn from U.S. jurisprudence and applied at the international level is that content-based or viewpoint-based restrictions will rarely be compatible with freedom of speech. As U.S. jurisprudence currently stands, within the three categories of “fighting words,” “incitement,” or “true threats,” the line between permissible and impermissible speech should not be based on the content or viewpoint expressed in the speech. In R.A.V. v. City of St. Paul, the Supreme Court struck down a statute seeking to criminalize the display of a symbol which “arouses anger, alarm or resentment in others on the basis of race, colour, creed, religion or gender,” such as a burning cross or Nazi swastika. 234 The Court considered this to be related to “fighting words,” but was not prepared to consider it unprotected speech because it constituted content-based discrimination. 235 By prohibiting fighting words that preached intolerance on the basis of “race, color, creed, religion or gender”, the statute allowed fighting words that preached racial and religious tolerance, and this viewpoint-discrimination was unacceptable. 236 The state had “no such authority to license one side of a debate to fight freestyle, while requiring the other to follow Marquis of Queensberry rules.” 237
To survive a constitutional challenge, a criminal statute must therefore meet the neutrality requirements laid out in R.A.V. and either the imminence and intent requirements of Brandenburg, the “fighting words” requirement of Chaplinsky, or the “true threat” of Watts. 238 The combined effect of these doctrines is that insulting speech is well protected, and that a high bar—usually linked to imminent violence—is required for criminalization. In addition, the prohibition on content or viewpoint-based laws means that judges are not basing their decisions on their own preferences about what speech should or should not be heard. This is a very different position to the one that emerges from a review of international jurisprudence on insults and should serve as inspiration to courts that wish to provide more protection to speech.
A. Eight Recommendations for the Right to Insult in International Law
We believe that international law on insulting speech should be applied in a manner that is more coherent and more protective of speech, in line with the approach espoused by the U.S. Supreme Court under the First Amendment. The difficulty in identifying the relevant international law when it comes to insulting speech, and the conflicting guidance that exists increases the scope for governmental abuse and also limits the relevance of international law when it comes to private actors who may struggle to understand the current international legal rules if they seek to apply them.
As a result, we propose the following eight recommendations to guide consideration of insulting speech in international law. 239
At the national level:
(1) The law should recognize a “right to insult”. An insult should only be criminalized if it is intended to incite violence or criminal offences and it is likely to produce such violence or offences. 240 There is a subjective element (intent to incite violence or crime) and an objective element (likelihood that violence or crime would occur). The violence or crime should be imminent or otherwise sufficiently concretely identified. Intent to incite hatred, hostility or discrimination should be insufficient to justify criminal sanctions. Defamation or libel should only attract civil penalties, never imprisonment. There may be aggravated penalties for insults based on race, nationality, ethnicity, gender, and so on, but these should not be criminal and should always be proportionate to the harm done. In order to recognize a higher measure of protection for speech than what is necessarily provided for under international human rights law, national parliaments may need to enact and amend domestic legislation, or even domestic constitutions or bills of rights. 241
(2) Insulting the three R’s—royalty, rulers, or religion—should not be criminalized. Laws imposing criminal sanctions for lèse majesté, desacato, disrespect for authority, disrespect for flags and symbols, defamation of the head of state or public institutions (such as the army or judiciary), the protection of the honour of public officials, sedition, and blasphemy are not in compliance with international law and should be abolished. 242 The only exception should be where the person making the insult intends to incite violence or a criminal offence and where it is likely in the circumstances that such violence or crime would ensue.
(3) Insulting victims of a historical event through “revisionism” or genocide-denial should not be criminalized. The position of the Human Rights Committee is to be preferred to that of the European Court of Human Rights because it recognizes that laws penalizing the expression of opinions about historical facts are incompatible with respect for freedom of expression.
(4) Laws on counter-terrorism, public order or other offences should not be used to prohibit speech that is merely insulting. Such laws should have precise terms to avoid authorities using these laws to suppress freedom of expression.
At the international level:
(5) CERD Article 4 should be deleted by the agreement of States Parties or excluded through reservations. It is both too broad and too narrow. It is too broad because it makes it a criminal offence to disseminate any idea that incites racial discrimination. It is too narrow because it is limited to racial hate speech. These deficiencies cannot be cured by interpretation.
(6) States should enter reservations to ICCPR Article 20 to prohibit speech only where it intentionally incites violence or a criminal offence that is likely to follow imminently (or is otherwise concretely identified) as a result of the speech. 243
(7) Vague and anachronistic concepts should not be relied on to justify the criminalization of insults. Terms such as “public morals” in Article 19(3) of the ICCPR and “propaganda for war” in Article 20(1) of the ICCPR should be defined more precisely in line with principles (1)–(3), or states should enter reservations or declarations in respect of these terms.
(8) Article 17 of the ECHR and its equivalent in the ICCPR, Article 5, should not apply to freedom of expression cases, which should always be subject to a balancing test under Articles 10 and 19 of these conventions, respectively.
These eight recommendations should also guide the private sector when they are requested to remove online content because it is considered insulting by a person, institution or government. 244 The U.N. has stated that businesses should respect human rights 245 and the U.N. Global Compact recommends that companies align their policies with recognized international standards on human rights. 246 The decisions made by technology companies are increasingly important since platforms like Google and Facebook are used by billions of people all over the world—they are now the gatekeepers of the information that consumers can access. Governments therefore often seek to influence these companies’ policies towards insulting speech.
Although some prominent technology companies have alluded to international human rights law as a guiding principle, 247 companies in fact rely on their own “community standards,” as well as national laws, to determine requests to censor insulting content. 248 But where national laws themselves violate international treaties that apply to that country, technology companies should be entitled to take that into account. These companies should be able to draw inspiration from international law in these areas, and they should, in turn, be as transparent as possible in their decisions on restricting or removing content.
Ultimately, States, as well as private actors, would benefit from legal reform in this area—the tide around the world is turning against free speech, and international law should have a more positive role to play.