Introduction to Combating Hate Speech


This paper seeks to offer an introduction to the different concepts related to hate speech, introducing as much balanced definition of hate speech as possible, through exploring the heated debate around if restrictions on the right of free expression can be justified on account of combating hate speech and if so, where and how to draw the limits of such restrictions. The purpose is to offer an idea of the meaning of hate speech, the objective evaluation of its true harms, and how over legislation might threaten the right to free speech with consequences endangering Internet users beyond limiting their enjoyment of this right and going as far as exposing at least some of them to persecution and wrongful indictment.

Grounded in human rights discourses, this paper draws on International Human Rights Law, including such fundamental documents as the Universal Declaration of Human Rights (UDHR), the International Covenant on Civil and Political Rights (ICCPR), and other international human rights treaties. Complementary documents for the above-mentioned ones are also consulted. These include Human Rights Committee comments, UN Human Rights Council (HRC) resolutions, reports of UN Special Rapporteurs, and other documents issued by the Office of the High Commissioner for Human Rights (OHCHR), especially the Rabat Plan of Action Outcomes, and the Threshold Test on Hate Speech extracted from it.

Hate Speech and the Internet

Incitement to genocide is the term for what might be considered the extremest form of hate speech. Incitement to genocide has played an essential role in some of the largest and worst genocides in modern history: the Holocaust against European Jews by the German Nazis during World War II, the Armenian Genocide perpetrated by the Ottoman Empire during World War I, and the Rwandan Genocide during the Rwandan Civil War in 1994.

The International Criminal Tribunal for Rwanda has indicted several persons with the charge of “direct and public incitement to the commission of genocide” which it defined as:

directly provoking the perpetrator(s) to commit genocide, whether through speeches, shouting or threats uttered in public places or at public gatherings, or through the sale or dissemination, offer for sale or display of written material or printed matter in public places or at public gatherings, or through the public display or placards or posters, or through any other means of audiovisual communication.

It should be noted that the latest of the above-mentioned examples of genocides incited by widely disseminated hate speech took place in 1994. This was before the advent of the Internet that revolutionized communications among people within the same country and across the whole world. To publicize a message of any kind so it reaches as large an audience as never imagined before, has become much easier and more effective. Examples of today’s online hate speech fueled atrocities include the Rohingya genocide in Myanmar, the anti-Muslim riots in Sri Lanka in 2018, and the Pittsburgh synagogue shooting in the United States in the same year.

A study conducted by a group of Indian researchers has confirmed that “hateful content diffuse farther, wider and faster and have a greater outreach … .” Another study “which analysed 263 million conversations in the UK and US, between 2019 and mid-2021.” has found that online hate speech rose by 20% during the Covid-19 pandemic. It also found a considerable increase in online hate speech in the time of events like the Black Lives Matter protests in the United States in June 2020.

The horrific potential consequences of hate speech are undeniable, however the term hate speech in itself is very loose and vague especially in public use by the media and public audience. The tendency of governments around the world, alarmed by the threats, both real and imagined, posed by online hate speech, to rush into pushing the legislation of new or modified laws to restrict what they define as hate speech. This trend is in itself a great threat to the right of free expression. Besides the great importance of the right to free expression as one of the foundational human rights, without which many other rights can’t be fulfilled, restricting legitimate speech for the sake of avoiding the consequences of hate speech is counter productive as in the end of the day rooting out the real and deep social and cultural origins of hate speech can only be accomplished by encouraging even more free speech.

Combating the truly threatening hate speech while protecting the right to free speech requires striking a highly sensitive balance. This starts with reaching as clear and comprehensive definition of hate speech as possible. Such definition should make it easier to distinguish between hate speech that constitutes incitement to hatred, discrimination, violence and even genocide, and hate speech that still should be tolerated under the protection of free speech.

Defining Hate Speech: 1 – Existing Definitions

Most people tend to think of hate speech from an emotional point of view. This means first that they think of it as an expression of involuntary emotions, and second that they judge a speech to be a hate speech based on the subjective emotions it steers in them. While it is true that everybody is entitled to their feelings with no interference allowed whatsoever, dealing with social phenomena requires drawing on abstract concepts and objective rather than subjective definitions.

Dictionaries are were we find the most basic definitions of a term, so let’s start with the definition offered by the Cambridge Dictionary for hate speech which is:

public speech that expresses hate or encourages violence towards a  person or group based on something such as race, religion, sex, or sexual orientation (= the fact of being gay, etc.).

As basic as this definition might be it is still much more specific than defining hate speech by mere expression of hate. First of all such an expression must be public. More than expressing hatred, hate speech can also encourage violence against its target. Coming to target/s of hate speech there we find what most characterizes its specificity compared to mere expression of hate. Targets of hate speech can be an individual or a group of people, but it is not just any expression of hate against this individual or group, this expression mus be based on either the individual’s or the group’s race, religion, sex, or sexual orientation. In other words hate speech is always based on some group attribute whether its target was an individual or a sub-group of people.

Repeating this to be as clear as possible. A speech is judged to be a hate speech by the most basic criteria of being public, of its potential of encouraging violence, and most importantly by being based on the target belonging to some group distinguished per race (e.g., being black), religion (e.g., being Muslim, or Shia Muslim), sex (e.g. being a female), or sexual orientation (e.g., being gay or lesbian).

Many states already have the criminalization of some forms of hate speech legislated into their laws. Based on some common characteristics of the definitions stated in these laws the Wikipedia page on hate speech comes up with the following definition:

speech, gestures, conduct, writing, or displays that incite violence or prejudicial actions against a group or individuals on the basis of their membership in the group, or that disparage or intimidate a group or individuals on the basis of their membership in the group.

Additionally, some laws set criteria for groups protected by criminalizing hate speech against them.

This definition is more specific as should be expected as here it describe criminalized acts which expose their perpetrators to criminal trial and time in prison if convicted. It also details the acts constituting the general term “speech” to include beside speech as the simple act of verbal pronouncement of some ineligible words, gestures, conduct, writing or displays, the whole set is more accurately referred to by the term “expression”. Incitement of prejudicial actions is added to violence here, and either one or the other is required, and not only a potential outcome. Also added as alternatives the acts of disparagement, and intimidation. One or more of these is required for a speech to be criminalized as hate speech. As per the previous definition the basis of speech to be considered a hate speech is that the target should be a group or individuals on account of being members of the group. No enumeration of the categories characterizing the group is provided opening it to any possible basis for categorization.

The last note is quite important. Some laws do indeed set some criteria for the groups protected by them. This means that not any say ethnic, religious or sex group can be target for hate speech. Such laws condition hate speech by causing actual or potential harm. Some groups are not vulnerable for acts of violence, prejudicial actions, disparagement or otherwise intimidation by hate speech. Ethnic and religious majorities in any country are such groups. The sex group of males in any country is also not vulnerable for incitement to any of the mentioned acts. Only groups already vulnerable on account of being a minority, or being the subordinate party in a domination relationship as it is the case with women, are endangered by the materialization of hate speech into actual harm.

This exploration of existing definitions of hate speech ends with one by the United Nations Strategy and Plan of Action on Hate Speech. Before presenting it however it should be noted that there is no definition of hate speech in any binding treaty or covenant, so the following definition is more of a recommended starting point for states (and companies) to take into consideration when building their strategies for dealing with hate speech. So, hate speech is defined as:

any kind of communication in speech, writing or behavior, that attacks or uses pejorative or discriminatory language with reference to a person or a group on the basis of who they are, in other words, based on their religion, ethnicity, nationality, race, colour, descent, gender or other identity factor.

– OHCHR, What is Hate Speech?
    (Emphasis from the original text.)

On first reading this definition seems to be rephrasing the previous ones. A careful reading however will reveal that it is considerably different. This definition however and the whole framework of the Strategy and Plan of Action on Hate Speech can only be fully understood and analyzed within the context of the current International Human Rights System, which is the subject of the next section.

Hate Speech within the Framework of the Human Rights International System

As mentioned before, there is no definition of hate speech in the covenants and treaties making up the current binding legal framework of the International Human Rights System. This however does not mean that hate speech is non-existent in this framework. It is present there by virtue of its glaring absence. In fact we can find hate speech in the tiny void separating Articles 19 and 20 of the International Covenant on Civil and Political Rights. It is right there filling this void with tension. To be accurate however hate speech has not always resided in this rather uncomfortable place. At the time of passing the covenant, at the end of one of the most tumultuous processes in human rights history, hate speech resided in Article 20, understood to be implied by it. To see this let’s reread the relevant text of this article:

Any advocacy of national, racial or religious hatred that constitutes incitement to discrimination, hostility or violence shall be prohibited by law.

ICCPR, Article 20 para. 2

Though quite brief this text should recall the definition of hate speech by some states laws. Indeed, many if not the great majority of what we call hate speech laws were legislated in conformance with the obligation set in this article for the states party to the covenant. It should also be obvious why for quite a long time criminalized hate speech was adequately implied by this paragraph of the article. How and why then the definition of hate speech started to move from its comfortable place in this article gravitating towards Article 19? To answer this question let’s also reread the relevant paragraph of Article 19:

Everyone shall have the right to freedom of expression; this right shall include freedom to seek, receive and impart information and ideas of all kinds, regardless of frontiers, either orally, in writing or in print, in the form of art, or through any other media of his choice.

– ICCPR, Article 19, para. 2

This paragraph is provided here for reference as the following paragraph which is of concern here refers to it.

The exercise of the rights provided for in paragraph 2 of this article 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:

  1. For respect of the rights or reputations of others;

  2. For the protection of national security or of public order (ordre public), or of public health or morals.

– ICCPR, Article 19, para. 3

The definition of hate speech gravitates specifically to the to sub-paragraphs (a), and (b), enumerating (rather loosely) the cases where restrictions may be applied to the rights stipulated by the second paragraph of the article. In most cases where rights groups sought to expand the rather limited definition of hate speech implied in Article 20 in order to protect vulnerable groups, they tended to cite sub-paragraph (a)’s “rights of others” clause. Governments on the other hand tend to cite sub-paragraph (b)’s clauses “national security”, and “public order”. (Two of the most favorite terms for all governments).

The move of hate speech definition from Article 20 toward the licences for imposing restrictions on the right to free expression in Article 19 created much tension that continues today. The definition by the United Nations Strategy and Plan of Action on Hate Speech seems to seek resolving this tension by providing a humans rights’ experts modified point of view countering the rush of governments towards imposing harsh restrictions on the right to free expression depending on their perceived threats posed by hate speech to public order and national security.

That definition actually reflects the developments the human rights movements has gone through during the last two decades:

Firstly, it takes the threats posed by hate speech seriously.

Secondly, it is obviously influenced by feminist movement developments through its 3rd wave and into what many consider it to be its 4th wave.

Thirdly, replacing speech with communication and emphasizing that it is of any kind, includes communications made in private, a channel of much online hate speech targeting women mainly but other vulnerable groups as well.

Fourthly, diverging sharply from other definitions, this one makes the communication i.e., the hate speech, an active factor that directly does the actions of attack and use of pejorative or discriminatory language. Absent here is the incitement to hatred, discrimination or violence. Symbolic attacks, and demeaning language are considered in themselves the harm incurred by victims. All these are highly contextual with tight cultural underpinnings. This reflects the strong trend in current human rights theory of shifting from universalism to cultural specificity.

Finally, the reference to “who they are” evidently reflects the rise to prominence of identity politics. If anything, this definition is sort of a template that may be filled with culturally specific and identitarian content, seeking to balance the rights of communities and those of the individual, though in practice the cultural specificity and the identitarian trends are almost always mutually exclusive, with the first stumping on the second most of the time.

Surprisingly, there is not much to say about this definition in the way of balancing the protection of the right to free expression and the restrictions imposed on it to combat hate speech. While the action plan does confirm that the international law only requires state to criminalize in their laws that hate speech that crosses the threshold of incitement to “discrimination, hostility or violence” as per Article 20 of the covenant, it points out that “It is important to underline that even when nit prohibited, hate speech may be harmful.”

If anything, the definition by the UN Action Plan opens wider holes into the protective shield of the right to free expression by using terms culturally defined by nature, and by tilting toward the subjective on the expense of the objective, both are pitfalls that the tradition of protecting free expression fought against fiercely, including the ICCPR treaty body, the Human Rights Committee, which is in charge of interpreting the articles of the covenant among other responsibilities. In its general comment No. 34 concerning Article 19, the committee cities its previous general comment No. 22 concerning restrictions of rights on account of protecting morals, a highly culturally specific category:

“the concept of morals derives from many social, philosophical and religious traditions; consequently, limitations… for the purpose of protecting morals must be based on principles not deriving exclusively from a single tradition”. Any such limitations must be understood in the light of universality of human rights and the principle of non-discrimination

– HRC, GC 34, para. 32

This text actually amounts to a universalization of the category of morals canceling its inherent cultural specificity by obliging states party to the covenant to draw upon the whole range of world “social, philosophical and religious traditions.” The text also does not fail to reconfirm the “universality of human rights.”

When it comes to subjectivity, we may seek an example provided by the United States’ judicial tradition one of the most fierce defenders of the right to free expression. In a very famous decision the Illinois Supreme Court ruled against ordinances issued by the village council of Skokie which sought to prevent holding a march of the Neo-Nazi group the National Socialist Party of America in the village that had a large number of Holocaust survivors Jews. in its ruling the court stated clearly that:

public expression of ideas may not be prohibited merely because the ideas are themselves offensive to some of their hearers.

– Supreme Court of Illinois,
The United State Village of Skokie
v. National Socialist Party of America,
373 N. E. 2d 21 (Ill. 1978)

The court specifically ruled against the applicability of the Fighting Words doctrine to this case. This doctrine was first set by the Supreme Court of the United States (SCOTUS) in the case of Chaplinsky v. New Hampshire in 1942. Interestingly the doctrine in a way matches the meaning of “attack” in the discussed definition of hate speech. Fighting words are insulting ones that are injurious in their own and deemed by the court as not protected by the right to free expression. Specifically, a speech that has no content, i.e. that does not communicate any ideas, other than fighting words can be prevented and punished without violating the right to free expression. In the case of Skokie, the court set the rule that the injury of fighting words can’t be subjective. Later on the SCOTUS further narrowed the fighting words doctrine so that a speech is considered fighting words only if it would “produce a clear and present danger of a serious intolerable evil that rises above mere inconvenience or annoyance.” This practically restored the doctrine to the limited space demarcated by Article 20 of the ICCPR twenty years later.

The Six Points Threshold Test of Hate Speech

The six points threshold test of hate speech were extracted from the Rabat Plan of Action. It should be noted that this meant for hat speech that warrant prevention and punishment. The points are unbinding guidelines for concerned parties (mainly states and their legislature bodies) willing to implement a legal system that both punish hate speech that truly threaten inflection of harm on vulnerable groups and their members while properly protecting the right to free speech. The points are provided in a form suitable for use as a test applicable to each individual case. A case that fulfills all six points justifies criminalization and legal action against the individual or group accused of dissemination of hate speech.

The six points are: context, speaker, intent, content and form, extent of the speech act, and likelihood including imminence. – Context, requires analyzing the speech act within the social and political context. Does the context allows, or helps incitement to produce its intended results? – Speaker, requires taking into consideration the position or status of the speech source, whether it was an individual or an organization, relevant to the audience. How receptive and willing to comply with the speakers incitement the audience might be. – Intent, requires making sure the potential harm was intended. How evident it is that the speech act was not a result of negligence or recklessness? – Content and form, requires analyzing the content of the speech. How provocative and direct was the speech? – Extent of the speech act, requires analyzing the reach of the speech act. How much accessible the speech is? How far it might go? How widespread it might be? – Likelihood, including imminence, while the speech act itself is punishable whether it produced its intended result or not, it is still required to be analyzed as for the likelihood of producing this result. What is the probability that this speech act may produce actual harm to its target(s)?

First of all, there is no denying that these points are quite comprehensive. Taken together they are enough to draw clear threshold separating hate speech, the state are required to punish by law as per Article 20 of the ICCPR, and hate speech that the state are likely to be violating the right to free expression if they punished it by law. However, being comprehensive is one thing and being practically applicable is another. The points simply place a heavy burden on the judiciary required to undertake these tests for each case. Provided that the judiciary in question is independent, impartial, and has the required resources including knowledge and time, it is still not guaranteed to come up with objective measurements for each test. Considering the possibility of the judiciary is not independent, or impartial, the absence of objective criteria makes judgments based on their use difficult to appeal in higher degree courts. Additionally the nominal use of these points will legitimate what might otherwise be biased or guided judgement, with no easy way to prove due process in using the tests was not observed. In many cases this would make the task of proving that a trial was not fair much more difficult, and may result in failure to expose the use of law as a tool of oppression under totalitarian and authoritarian regimes.

Defining Hate Speech: 2 – Proposed Definition

For reaching a satisfactory definition of hate speech this paper contends that the following guidelines should be observed:

  • Any definition of hate speech that should or may justifiably and without violating the right to free expression be punishable by law must stay within the obligation set by Article 20. No speech act that does not fulfill the requirements of this article should be punishable by law.
  • There is absolutely no need for looking to Article 19’s cases where the right to free expression can be restricted on their account as these cases are already implied by the requirements of Article 20.
  • It is true that hate speech that is not punishable pursuant to the obligation set by Article 20 of the ICCPR can still be harmful. But such harm, as it is guaranteed not to amount to incitement to discrimination, hostility or violence, would be still be tolerable compared to the harm inflicted on the society as a whole due to undermining the right to free expression that presupposes equal enjoyment of it by all.
  • While not punishable by criminal law, hate speech to which Article 20 obligations cannot apply is still a problem that has to be dealt with. Provisions for allowing suing perpetrators under civil law can be encouraged in an open and transparent manner allowing more speech about the issue, not less.
  • Measures addressing the social roots for hate speech are highly preferred to any other course of action.
  • For a speech to qualify as hate speech, its targets have to be either groups, or individuals on account of being members of some groups.
  • Targets of qualified hate speech must be objectively judged to be vulnerable groups or individual members of them. Objective criteria of vulnerability include, but not necessarily be limited to, being a racial, ethnic, or religious minority or being historically discriminated against (such as women, LGBT people, colored people, etc.).
  • Hate speech targets can only be people; institutions, states, religions, beliefs, ideologies, etc. should by no means be protected from criticism including harsh forms of it, which in case they targeted people would be considered hate speech.
  • To be qualified as hate speech a speech should has no content that may be rationally argued. In other words it is nothing but insulting, demeaning, ridiculing etc.

Based on these guidelines the following definition of hate speech is proposed:

Hate speech is any kind of public expression whether in speech, writing or behavior, that can objectively judged as an attack by use of pejorative or discriminatory language with reference to a vulnerable group or a person on account of being a member of a vulnerable group on the basis of identity characteristics, either true or wrongfully attributed, Vulnerability of a group should be objectively decided based on minority status, historical prejudice, or otherwise deviation from socially sanctioned norms.


This paper sought to provide an overview of the concept of hate speech. It analyzed the most important existing definitions of the term. It also explored the tension between criminalizing some forms of hate speech and the protection of the right to free expression within the international human rights system, including the efforts by United Nations human rights related bodies to resolve this tension, especially the Rabat Plan of Action and the Six Points Threshold Test of Hate Speech extracted from it. Along the way, the paper scrutinized both existing definitions of hate speech, and the mentioned efforts to resolve the tension and apparent contradiction between combating hate speech and protecting the right to free expression. The critique advanced by the paper is based on two presuppositions:

  1. The protection of the right to free expression should be prioritized, being one of the most fundamental human rights and a necessity for constituting independent individuality, and for building free and democratic societies capable of combating hate speech.
  1. Combating hate speech by punishment is nothing more than a temporary treatment for the symptoms of a disease deeply rooted in the social and cultural foundations of societies. True cure for this disease has to deal with its roots which is a long and difficult process dependent on large part on encouraging more respect for sanctity of the right to free expression rather than imposing restrictions on it.

Finally, it should be noted that true and strict adherence to the principles of the objectivity and universality of human rights is imperative for preserving its international system that is being currently subjected to fierce attacks both in theory and practice. Making room for cultural specificity and subjectivity, due to legitimate but misguided fears of the rise of hate speech, threatens to undermine the objectivity and universality of human rights rendering adherence to them as a subjective or cultural specific point of view which practically destroys the foundations of an International system for human rights in the first place.