Hate speech and the harm in Indonesian judicial decisions

Abstract

Indonesia’s hate speech provision under Article 28(2) of the Electronic Information and Transaction Law lacks an objective threshold. This article presents a case study of twenty-seven judicial decisions to investigate whether such lack had consequently limited or instead broadened the judiciary’s discretion in determining hate speech. The results of the study show that the construction of Article 28(2) has led to a broad determination of hate speech, as it reveals an inconsistency in considering the harm of hate speech. This paper discusses where the inconsistencies are found and why they are problematic when assessing harm. Furthermore, in light of the case studies, the paper took into perspective the recent development of the Criminal Code to suggest the way forward for Indonesia’s law on hate speech.

Keywords:

• hate speech
• electronic information and transaction law
• judicial decisions
• harm
• Indonesia

1. Introduction

In 2020, a man in Indonesia was convicted for hate speech after publicly declaring to be an atheist and questioning the existence of God on Facebook.¹ Another Indonesian was declared guilty of hate speech after posting a warning against food poisoning on social media.² Many other cases were charged and convicted of hate speech—some directly incited violence, some used derogatory language toward others, and some were unclear as to how the contested speech constituted hate. It was cases such as these that prompted this study to investigate how hate speech is determined by the judiciary.

In theory, hate speech can be defined as hostile speech aimed at individuals or groups because of their identity or perceived characteristics.³ The term possesses a broad spectrum of harmful speech from abusive expression, extreme forms of prejudice, bias, and hatred, to incitement of hatred.⁴ Hate speech commonly possesses three characteristics. Firstly, a speech is directed against an individual or a group of individuals because of their distinctive features; secondly, it stigmatises the target by implicitly or explicitly ascribing to it undesirable qualities; and thirdly, it casts the target as an undesirable presence and a legitimate object of hostility.⁵ From a legal standpoint, hate speech is not protected under international human rights law.⁶

Hate speech can be made in various ways, including in writing, and illustration. Since the increased use of social media, more hate speech is committed online. This is especially the case in Indonesia. In 2017, 3.325 hate speech cases were reported and handled by the Indonesian police force, which signalled a 44,99% increase from 2016 with 1.829 cases.⁷ Moreover, from the beginning of the COVID-19 pandemic until June 2020, the regional police of Jakarta have received and investigated around 480 hate speech cases.⁸

In Indonesia’s legal context, online hate speech is criminalised under Article 28(2) of Law No. 11 of 2008 as amended by Law No. 19 of 2016 concerning Electronic Information and Transactions (“EIT Law”), which reads “Anyone who, intentionally and without the right, distribute information⁹ intended to create a sense of hate or enmity towards individuals and/or groups of certain communities based on ethnicity, religion, race, and intergroup.”¹⁰ [not the official translation]

The article does not define hate and enmity nor does it elaborate on the intent to create a sense of hate or enmity. None of these concerns are addressed in the elucidation of EIT law or elsewhere.¹¹ Consequently, the construction of the Article is left vulnerable to multi-interpretations. For example, the scope of “hate” under Article 28(2) is not defined. This is problematic recalling that hate speech possesses a broad spectrum of harmful expression. In the legal sense, this should render the need to distinguish between speech that would justify legal restrictions and those that would not.¹² Subsequently, this also means that the regulatory framework to counter hate speech needs to be holistic and allow multi approaches aside from criminal law¹³ for different forms and levels of hate speech. In this context, this paper argues that criminal sanction should only apply to severe forms of hate speech.

This is in line with the suggestion by the United Nations High Commissioner for Human Rights (UNHCHR) that a distinction should be made between speech that constitutes a criminal offense, speech that may justify a civil lawsuit or administrative sanction, and speech that would not give rise to either criminal, civil, or administrative sanction,¹⁴ yet, could be addressed through other policies including education and prevention in form of stricter censorship to detect hate speech on social media. The fact that Article 28(2) does not provide an objective threshold of prosecutable hate speech implies that the Article could be used to criminalise any speech regardless of form or level of harm. This is prone to abuse since it would leave the assessment of hate as well as enmity, purely subjective.

Given the lack of clarity, the Author conducted a study of twenty-seven judicial decisions on hate speech that was charged under Article 28(2) of EIT Law. Particularly, the study sought to investigate whether the construction of the Article had consequently limited or rather broadened the judiciary’s discretion in determining hate speech by examining how far the judiciary considered the harm arising from the contested speeches.

In defining harm, some scholars argue that immoral conduct is in and of itself harmful and thus justifies criminalisation.¹⁵ Other scholars advocate for a more concrete threshold of harm. Peršak for example, argues that the harm that qualifies under the harm principle is “harm to others”, that is, harm against an individual may that be against their life, body, liberty, property, health, or personal integrity.¹⁶ In this sense, the harm principle posits that in situations where harm occurs upon others, the maker of such harm may be rightfully subject to coercion or control.¹⁷ This means that personal conduct that does not result in harm to others would not warrant such control, indicating that causing harm to others matters as it explains a person’s moral blameworthiness to justify punishment.¹⁸

In criminal law, it is the harm that justifies a state’s intervention through means of criminal punishment. Ultimately, criminal procedure is reserved for conducts that are socially harmful,¹⁹ taking into consideration the gravity, harmfulness repulsion, and seriousness of the conduct in question.²⁰ In this sense, the greater the harm, the greater the necessity to apply criminal procedure. The lack of a proper assessment of harm thereof could result in the criminalisation of conduct that violates civil liberty.²¹

In the context of hate speech, Waldron essentially argues that the harm of hate speech is that it undermines the sense of assurance that people will not be discriminated against or subject to violence because they belong to a certain (vulnerable) group. As such, hate speech is considered an offense against other people’s dignity.²² Waldron also highlighted that the protection of dignity is different from the prevention of mere offense. Meaning that, while freedom of speech allows the liberty to present offensive opinions, it does not extend to allowing offenses against the dignity of other people.²³ However, Waldron’s idea of harm in hate speech was criticized for its unclear position: whether hate speech causes harm (consequential harm) or it constitutes harm (constitutive harm) in itself.²⁴

Constitutive harm posits that hate speech is harmful in itself against the dignity of other people. Meanwhile, consequential harm advocates for substantial effects as a result of hate speech, which requires proof of the relationship between the speech and the effect.²⁵ In this context, consequential harm can occur in four ways. First, if it persuades the audience of hate speech to believe in the negative stereotypes that cause them to engage in harmful conduct. Second, if it persuades them to believe in negative stereotypes in a way that it shapes their preferences. Third, creating an environment where expressing negative stereotypes and committing discrimination becomes normal. Fourth, leading the audience to adopt the same behavior.²⁶ This paper will discuss whether the judicial decisions adopt either of the approach and how it was used to decide on hate speech cases.

To assist in assessing harm in hate speech cases, the Author referred to the components presented in the Rabat Action Plan as issued by the UNHCHR in (2013). The Action Plan is not a binding document. Rather it contains the conclusions and recommendations from expert workshops that resulted in the adoption of the Action Plan. However, it was initiated to provide insight into the application of Article 20(2)²⁷ of the International Covenant on Civil and Political Rights (ICCPR) “because limitation of freedom of expression must remain an exception.” Indonesia itself has ratified the ICCPR through Law No. 12 of 2005. According to the Action Plan, the threshold of Article 20(2) of ICCPR consists of six components that need to be fulfilled in order for a speech to constitute a criminal offense: context; speaker; intent; content and form; extent of the speech act; and the likelihood. This paper argues that these components, when operated, may identify the extent of harm in each case because each component describes what is expected to label a speech as harmful in a way that constitutes a criminal offense.

According to the Action Plan, the element of context requires the assessment of speech within the social and political context prevailing at the time the speech was delivered. Meanwhile, the element of the speaker refers to the actor’s position in society where the person’s standing is capable of mobilizing the audience. With regard to intent, Article 20 of the ICCPR excludes negligence, recklessness, and especially the mere distribution of speech. Rather, it requires an active relationship between the object and subject of speech as well as the audience.

Furthermore, according to the Rabat Action Plan, the content and form of speech constitute as one of the critical aspects of incitement. The content can be analysed based on the degree to which the speech is provocative and direct. Whereas the form refers to the style or nature of arguments present within the contested speech. The fifth element is the extent of the speech act. Extent includes consideration of the reach of the speech, its public nature, magnitude, and size of the audience. In this sense, the element of the public can be measured by the means of distribution of speech, the frequency of broadcast, and the quantity and extent of communication. It is also worth considering whether the audience had the means to act on the speech. Lastly, the likelihood requires a degree of risk of harm. Consequently, the judiciary has to determine that there was a reasonable probability that the speech would incite a form of action against the targeted group.

To investigate how the judiciary decides on hate speech cases with a particular focus on the assessment of harm, this paper will first introduce how hate speech is regulated under Indonesia’s legal framework. Then, the discussion will proceed to present the pattern from the judicial decisions to show how hate speech was decided. Lastly, the paper will argue that the findings indicate the need to clarify the harm of hate speech in Indonesian law.

2. Methodology

This paper is based on a study of twenty-seven judicial decisions that were collected from the Indonesian Supreme Court Online Directory. All of the decisions were regarding cases of hate speech charged under Article 28(2) of EIT Law. All twenty-seven decisions were decided at the first instance court in 2020, with all the defendants declared guilty. Furthermore, in all of the cases, the alleged hate speech was committed on social media.

After collecting the cases, the Author studied the facts of each case and the analysis of the judges. To organize the analysis, the Author then made four categories reflecting the elements of hate speech based on Article 28(2) of EIT Law, namely: the scope of protected groups, the spread of speech, the content of speech, and the nexus between speech and harm. The analysis of the judges on how the facts satisfy the elements of hate speech and the threshold that they used were then extracted and clustered into these four categories.

Once the categorization was completed, the Author sought to see whether there was a pattern. These patterns are then used to identify and address particular issues within the judges’ assessment of the elements of hate speech. A similar method was used in previous works reviewing the judicial reasoning of the Indonesian Supreme Court,²⁸ and a review of Indonesia’s religious court decisions,²⁹ where the reviews were based and structured on the patterns found within the judicial reasonings.

3. Hate speech in Indonesian legal framework

Under Indonesian law, hate speech is regulated by three different laws: Article 156 and Article 157(1) of the current Indonesian Criminal Code, Article 4(b) of Law No. 40 of 2008 on the Eradication of Discrimination against Race and Ethnic Groups, and Article 28(2) of Law No. 11 of 2008 as amended by Law No. 19 of 2016 concerning Electronic Information and Transactions (“EIT Law”). The difference between hate speech within these laws is that the current Indonesian Criminal Code applies generally to hate speech committed in the public sphere, whereas hate speech under Law No. 40 of 2008 is limited to those committed against racial and ethnic groups, and the EIT Law applies to hate speech committed online. Furthermore, since the hate speech provision under EIT Law can be seen as lex specialis to the one under the current Indonesian Criminal Code, therefore, interpretation of the provision could be referred back to the Code in case of ambiguity.

The judicial decisions discussed in this paper are limited to hate speech charged under the EIT Law. Understandably, hate speech is constructed differently in each of these laws but what is highlighted here is that the threshold of hate speech under the EIT Law is lower compared to the others.

Article 156 of the current Criminal Code criminalises anyone who “publicly gives expression of hostility, hatred, or contempt against one or more groups of the population of Indonesia … ” According to this article, the term “group” shall be understood as each part of the population of Indonesia that distinguishes itself from one or more other parts of that population by race, country of origin, religion, origin, descent, nationality, or constitutional standing. Meanwhile, Article 157(1) of the current Criminal Code criminalises anyone who “broadcasts, publicize or display writings or paintings in public that depicts statements of enmity, hate, insults between certain groups in Indonesia with the intent to make such content known to the public … ”

As a comparison, Article 4(b) of Law No. 40 of 2008 prohibits the demonstration of “hate or a sense of hate toward others due to their racial or ethnic differences,” that may be committed through written or drawn content that is placed, displayed, or spread in public space or other places that may be seen or read by others; or through speeches, statements, or other forms of spoken words in public space. Finally, as already mentioned above, Article 28(2) of EIT Law criminalises “anyone who, intentionally and without the right, distribute information intended to create a sense of hate or enmity towards individuals and/or groups of certain communities based on ethnicity, religion, race, and intergroup.”

The threshold of hate speech under the EIT Law is lower compared to the other laws when seen from how the element of “public” is regulated. Article 156 of the current Indonesian Criminal Code explicitly requires hate speech to be made in public. Similarly, Article 4(b)(1) of Law No. 40 of 2008 defines the spread of speech as “writing texts or drawing illustrations to be placed, pasted, or disseminated in public places or places where they could be seen or read by others.” On the other hand, Article 28(2) of EIT Law requires speech to be distributed to many individuals.³⁰ The implication is that this element could then be interpreted to include speech sent to individuals on private chat for example, which is inherently different from making a speech publicly known.

Furthermore, the scope of hate speech is made broader by Circular Letter of the Chief of the Police Force No. 6/X/2015 which seems to acknowledge different forms of hate speech to include insult, defamation, blasphemy, unlawful act, provocation, incitement, and spread of hoax—despite the fact that each of these conducts has their own definition, provisions, and elements of crime in the Criminal Code. Nonetheless, according to the Circular Letter, these conducts are considered to be forms of hate speech as they may potentially cause discrimination, violence, casualties, and/or social conflict. In addition, The Circular Letter of the Chief of the Police Force No. 6/X/2015 also broadens the scope of protected groups as it listed 11 vulnerable groups based on the following category: tribe, religion, religious sect, beliefs, race, intergroup, skin color, ethnicity, gender, disability, and sexual orientation.

The Circular Letter also provides that the police are obligated to conduct a set of measures when indications of hate speech are found or reported. Those measures are to monitor whether a subsequent act occurs; approach the person who committed hate speech; facilitate a meeting between the offender and the target of speech; seek a peaceful solution between the parties, and explain to the offender the impact that might arise from the speech. Only when these measures fail, criminal procedure is invoked.

4. Hate speech in Indonesian judicial decisions

4.1. The scope of groups protected against the harm of hate speech

This part addresses the scope of groups that are protected against the harm of hate speech. Article 28(2) of EIT Law mentions four groups: ethnic, religious, racial, and intergroup (antargolongan). The judicial decisions show that the government can be considered as intergroup, and thus, is afforded the same protection as the other three groups. The Author argues that “intergroup” cannot include organs of government. Rather, it should be meant to protect socially constructed identities that are not explicitly mentioned in Article 28(2) of EIT Law. In delivering the discussion, this part will start by explaining how intergroup is interpreted. Several cases will be discussed to show how District Courts have adopted the idea that the government constitutes as intergroup for the purpose of Article 28(2). Then, the Author will argue why intergroup should be limited to socially constructed identities.

To begin with, Article 28(2) of EIT Law recognises four types of protected groups: ethnic, religious, racial, and intergroup. As previously mentioned, since the hate speech provision under EIT Law is the lex specialis to the one under the Indonesian Criminal Code, interpretation of the provision could be referred back to the Code in case of ambiguity. That being said, “intergroup” under Article 28(2) can be interpreted based on Article 156 of the Code. According to which, “intergroup” is every group of the Indonesian people that is distinct from one another because of race, country of origin, religion, residence, origin, descent, nationality, or legal standing according to constitutional law.

However, in 2017, Article 28(2) was brought to the Constitutional Court for its lack of clarity on the term “intergroup” that was perceived to be contradictory to the constitutional right to speech. According to the claimant, the term is still vulnerable to multi-interpretation as it was unclear whether a distinction could also be made on the basis of other factors such as social status, political party, sexual orientation, or organization³¹ which could then hinder the exercise of free speech due to fear of being prosecuted.

In responding to the challenge, the Constitutional Court first argued on the basis of Bhinneka Tunggal Ika (unity in diversity) which serves as the identity of Indonesia. This identity entails the acknowledgment that as a state, Indonesia is built upon diversity in many aspects including economy, politics, society, and culture. As a consequence, such diversity needs to be protected. The Court then went on to reiterate that the right to speech under Article 28 of the Constitution is not without limitation. Rather, the exercise of such a right needs to be in respect of other rights. In this sense, if the term “intergroup” is excluded from Article 28(2) of EIT Law, it will leave other groups vulnerable to hate speech, which would then create a lack of legal protection as guaranteed by the Constitution under Article 27(1) and 28(D).³² Following this Constitutional Court decision, District Courts have interpreted “intergroup” in the context of hate speech to also include communities, organisations,³³ the government including the police force,³⁴ as well as “between the government and the society.”³⁵

The implication that follows the Constitutional Court’s ruling misaligns with a previous decision back in 2007 when the Constitutional Court declared hate speech against the government under Articles 154³⁶ and 155³⁷ of the Criminal Code as unconstitutional because they are vulnerable to abuse which may hamper the citizens’ constitutional right to convey their criticisms.³⁸ This finding was supported by the lack of clear criteria in the formulation of both Articles 154 and 155, which leaves “feelings of hostility, hatred, or contempt” without any clarity.

In reality, there have been claims that Article 28(2) has been used to criminalise political opponents or members of society who express political dissatisfaction through sarcastic criticisms.³⁹ These people were mostly reported by the government apparatus or their supporters, as they perceive such expressions as either hate speech, hoax, or defamation.⁴⁰

An example can be seen in a case that involves criticism and discontent towards the government. The defendant in this case made numerous posts on Facebook. In one of the posts, the defendant explicitly expressed his disappointment with how the government addressed the conflict in Poso which discriminates between Muslims and Christians. According to him, the government was favouring those of the Christian faith over Muslims. In another post, the defendant stated “ … the police conducted a raid this evening in several areas of Jakarta, but the Chinese enter Indonesia without being checked” (referring to police raids in crowded places during the Covid pandemic and comparing it to the loose international incoming travel policy). On another occasion, the defendant uploaded a picture of President Joko Widodo with the following caption: “you guys still believe in the deceptions by this weasel? … only a fool who falls in the same sewer (twice).” In its decision, the court acknowledged that the defendant’s speeches were an expression of discontent toward government policy. However, the court ruled nonetheless that the statements spawn hatred or enmity against the government⁴¹ without reasoning. Other examples, which will be discussed later on, include cases of expressing obscene language towards the police force,⁴² making a mockery of the military,⁴³ and using vulgar language in resistance against the government’s social distancing policy during Covid.⁴⁴

In this regard, the Author argues that to include the government as part of “intergroup” would fall outside the protective purpose for which hate speech is criminalised. Referring to the international law regime, the protected groups against genocide, hate speech, and discrimination are limited to four categories namely race, ethnicity, national, and religious group.⁴⁵ The question of whether or not it is possible to expand from such categories had been addressed before by the International Criminal Tribunal for Rwanda in the case of Akayesu⁴⁶ in which the tribunal decided that any protected group would have to be similar to the four groups in terms of stability and permanence.⁴⁷ The Author opines that expanding from the categories cannot be based on stability and permanence alone. The reason is that only race can have stability and permanence in terms of genetics,⁴⁸ while people can change nationality or religion. With regard to ethnicity, this group is characterised by a common language and culture.⁴⁹ Therefore, people can identify themselves to belong in an ethnic group so long as they speak the group’s language and practice the culture.

Instead, the Author opines that it may be helpful to consider other commonalities between these protected groups. That being said, it could be worth considering the fact that race, ethnicity, national, and religious groups are categories of socially-constructed identities. Historically speaking, these groups are often vulnerable to a degree of prejudice in cases where they categorize as a minority in a given environment, thus providing the basis for minority claims that require the protection of the state.⁵⁰ Waldron also pointed out the necessity of hate speech bans to protect the dignity of vulnerable minorities.⁵¹ From this perspective, criminalising hate speech is meant to be a form of protection for vulnerable groups. The Circular Letter of the Chief of the Police Force No. 6/X/2015 seems to have already taken this approach when it listed 11 vulnerable groups mentioned above.

With this in mind, the Author argues that a government is not a socially-constructed identity. In addition, it would be difficult to imagine a government being similarly vulnerable as other groups considering their position of power that allows them to navigate both the political and legal discourse of a state’s affairs. However, it is equally important to emphasize that the Author does not suggest that organizations, organs of government, or the likes should be left without legal protection. The fact of the matter is, the Indonesian Criminal Code already criminalises a wide range of speech crimes in the current Criminal Code. For example, Articles 310 (defamation), 311 (slander), 315 (insult), and 317 (false report). The difference between defamation and hate speech is that the former can be defined as false statements of a factual nature that insults and damages a person’s reputation.⁵² Whereas the latter is an expression of hate targeting a specific group because of their identity. For defamation, the Code does not limit the subjects to which the speech is targeted at. Meaning that legal protection is provided to every person regardless of their identity, including the government.

4.2. The extent or spread of speech conducive to causing harm?

As previously mentioned, the Rabat Action Plan suggests that the extent of speech must be considered in identifying hate speech. Extent refers to the reach of the speech, its public nature, its magnitude, and the size of the audience. This means considering the distribution of speech, the frequency of broadcast, and the quantity and extent of communication, as well as the reach of audience.

Considering these elements would be helpful to assess whether the spread of speech is conducive enough to cause harm. However, the elucidation provided in the EIT Law provides little guidance to measure the spread of hate speech. The elucidation explains that “distributing” speech refers to the act of sending and/or spreading speech to “many” individuals using electronics. The elucidation is used by the judiciary as the threshold to assess the spread of speech. However, “sending or spreading speech to many individuals” is interpreted differently.

In some cases, evidence was presented to show that a speech was not only spread but also read and responded to by an audience. An example can be seen in a case adjudicated by the District Court of Ternate, involving a defendant who was charged under Article 28(2) EIT Law after posting a false narrative of Jesus’s sexual relations. The post was a reaction to another Facebook user who posted an insult against the prophet Muhammad. The defendant’s post was shared 91 times and attracted 340 comments which rendered enough for the court to satisfy the element of distribution of speech.⁵³ In a different case, the defendant posted a video expressing threats against non-Aceh people living in Aceh on his Facebook account that was set to “public”. The video received over 100 disapproving comments and was liked by 225 users. These statistics were considered enough by the Court to satisfy the element of distribution of speech.⁵⁴ Similarly, in another case, the defendant mocked a religion with vulgar terms affiliated with genitals on his open Facebook account. The Facebook post was commented on by 410 users and shared 18 times. This was also enough to prove the spread of speech.⁵⁵ In addition, the District Court of Pangkalan Bun, when adjudicating a case of mockery against the police force on social media, decided that “reposting” is another evidence of the spread of speech because it indicates the increase of viewers who view a post from accounts other than the original one.⁵⁶

In some cases, the spread of speech was assumed to have reached an audience. For example, speeches posted on Facebook accounts whose privacy setting is set as public⁵⁷ and accounts with over 1.000 “friends”⁵⁸ satisfy the element of “spreading speech.” However, the spread of speech is also perceived to be satisfactory if it was sent to a group chat.⁵⁹ An example can be found in a case adjudicated by the North Jakarta District Court, where the defendant was proven to have forwarded a hoax video to two group chats. In the video, the defendant claimed that Chinese military uniforms were found at a local laundromat and said that the Chinese were ready for war. The groups to which the video was sent consisted of 94 and 185 members. In this case, the court decided that the spread element was satisfied on account of the group members who accessed the video.⁶⁰

This paper argues that there should be a careful consideration of the type of space where the contested speech is delivered. After all, there is a legal difference between speeches made in public and those in private. The latter falls within the right to privacy that is guaranteed under Article 32 of Law No. 39 of 1999 concerning Human Rights which includes the freedom and privacy to communicate through electronic means. The problem is that Article 28(2) of EIT Law only requires a speech to be sent to “many individuals” regardless of the space in which the text is sent. Consequently, sending a text message to more than one individual, for example, a group chat, would fulfill the element of “many individuals” even if the text was meant for private consumption within a group. The Author argues that sending a text message to “many individuals” cannot be equated with making a speech in public.

Instead, Article 28(2) of EIT Law should include the “public” element. Given that Article 28(2) applies to speeches made online, the element could be proven if contested content is posted or published under “public” setting, allowing other people who may not necessarily be followers to read the content. Alternatively, if the content is shared by followers. The Circular Letter of the Police Force No. 6/X/2015 mentions a list of means through which hate speech could be delivered. The list did include social media without further clarification. However, an explanation was found in a Joint Decree between the Ministry for Communication and Information, General Attorney, and Chief of the Police Force⁶¹ which states that “distributing” should be understood as making a speech publicly known through, among others, social media accounts that are set in public or open chat groups.

In any case, it should be noted that hate speech is not targeted at the victim themselves, but rather at the people the actor means to stir or influence to display hateful behaviour. This is also required by the Rabat Action Plan which suggests that there must be an active relationship between the perpetrator and the audience. Therefore, the spread of speech should not only be assessed numerically by, for example, the number of likes, viewers, or followers. Rather, the general reception of the audience should be taken into account. For example, reverting to the video that expresses threats against non-Aceh people living in Aceh. The content was met with over 100 disapproving comments, which suggests that even though the content was well “distributed,” there is an indication that the perpetrator was not successful in influencing the audience to display hateful behavior. To analyse whether a speech might have the power of influence, it is important to consider the content of the speech itself which is discussed below.

4.3. The content of ‘hateful’ speech

As stipulated under Article 28(2) of EIT Law, a speech must be “intended to create a sense of hatred or enmity” towards a group. In other words, creating hatred or enmity against an individual or group must be the purpose of the speech.⁶² The judicial decisions describe hate speech as causing tension towards a group or conflict between groups;⁶³ a speech of ill intent, showcasing a strong emotional drive to cause harm against the target of speech, in form of provocation, incitement, insult,⁶⁴ or defamation;⁶⁵ and a speech that expresses dislike, antipathy, or even hostility that is underlined by the feeling of wanting a group to be destroyed.⁶⁶ One particular case described hate speech as promoting discrimination, by referring to Article 4 of the International Convention on the Elimination of All Forms of Racial Discrimination 1965.⁶⁷

From these decisions, it seems that the judiciary adopts the consequential approach to the harm of hate speech, since it requires the speech to have caused harm. However, if the consequential approach is indeed the case, the judicial decisions were unclear in terms of explaining how the content of speech is hateful enough to cause harm. As suggested by the Rabat Action Plan, the content of a speech must be taken into account and can be measured based on the degree to which the speech is provocative and direct.

Based on the case studies, some cases show direct incitement to cause harm. One example can be seen in a case adjudicated by the District Court of Probolinggo, where the defendant expressed discontent towards the Indonesian police force on his Facebook account which was set to public, to the extent that he made a general call to kill the police force: “let’s kill the police, [I] hate the police. Better be protected by the military forces than those bastards.”⁶⁸ Another example that promotes violence is a case adjudicated by the District Court of Kutai Barat. In this case, the defendant made a public post on Facebook saying that if the “law enforcers do not act swiftly” he will share the location where Madura people reside and “we will massacre them as we did 20 years ago because Madura people have no honour … ”⁶⁹

Other cases were decided to be hate speech because they contained derogatory language against a group. An example is a case adjudicated by the District Court of Makassar, where the defendant commented on a Twitter post by Al Jazeera News about the ongoing demonstrations in Papua. The defendant wrote “drive all of those (university) students and Papuan monkeys back to Papua, we, the people of the Republic of Indonesia stand ready to destroy […] all the student monkeys and young monkeys in Java, Sumatra, Kalimantan, Sulawesi…”⁷⁰

Other decisions were unclear in presenting how the contested speech is hateful enough to cause harm against a protected group. For example, a case in 2020 concerned a post on Facebook about warnings of food poisoning. To be precise, the post reads the following: “Food Poisoning. An advice … !! For my brothers and sisters who feels they have dark skin and curly hair, be careful of consuming food including instant (ones) that are sold by irresponsible and vague sellers … ”⁷¹ This case was seen as using derogatory language against people with certain physical characteristics.

Another example can be seen in a case adjudicated by the District Court of Rengat. In this case, the defendant posted a video on Facebook showing a woman being possessed, along with a caption expressing his disbelief in God. In his caption, the defendant claimed that he did not understand why his friends would take offense at his opinion towards God, arguing that he “was not born from God”, “God does not control” his life, “God could not even stop Corona virus”, and that he was “fed up hearing the name of God everywhere.” He went on to question “where is your God now?”, “how God creates miracles”, and argued that “all God does is misled people.”⁷² This case was successfully prosecuted as hate speech as it was perceived by the court to have projected hatred towards people of faith and was even labeled as blasphemy.

Another case can be found in Medan, where the defendant posted a video of the military theme song, with the caption that says the military went from “tigers to cats”. The defendant also seemingly mocked the military with the written depiction of cat sounds: “meong meong meong”. Aside from this, the post did not contain any incitement or derogatory language. Nonetheless, the court decided that the caption is an “express of hatred or dislike towards military forces” or its high-ranking officials, without elaborating further.⁷³

Furthermore, a decision by the North Jakarta District Court recites a case about the defendant’s action of uploading a video to two WhatsApp groups consisting of 94 and 185 members respectively. In the video, the defendant stated “Chinese military uniforms are being washed at a laundromat in Kelapa Gading … this is bad, there are a lot of Chinese militaries, (they’re) ready to go to war I guess”. During the trial, the defendant admitted that his claim was in fact a lie. However, the court decided that the narrative constitutes hate speech without elaborating how and why.⁷⁴

As disturbing as these messages may be, the Author argues that without a form of incitement or provocation, it would be challenging to take a speech as being hateful enough to cause harm. The reason is that the wording of Article 28(2) of EIT Law requires the contested speech to be “intended to cause a sense of hate or enmity.” In essence, this implies that the speech must promote a certain behaviour to influence the public. Thus, adding the element of incitement or provocation would ease the judiciary to measure whether a speech was indeed intended to cause hate or enmity while considering the context in which it was made and whether it promotes violence, hatred, or intolerant behaviour. Adding the element of incitement or provocation would also be in line with Article 20(2) of the ICCPR. Under this Article, any speech or advocacy of national, racial, or religious hatred can be prohibited by law if it constitutes incitement to discrimination, hostility, or violence.

4.4. The inconsistent assessment of the nexus between speech and hate or enmity

The previous section discussed how on the one hand, the judicial decisions indicated that Article 28(2) conceptually adopts the consequential approach to harm. But on the other hand, the judicial decisions were unclear in terms of explaining how the content of speech is hateful enough to cause harm.

This section takes the discussion further. In the assumption that Article 28(2) conceptually adopts the consequential approach to harm, the judicial decisions were also unclear in deciding the nexus between speech and harm. In other words, how the speech was connected with the perceived harm. In this context, as shown in the table 1 below, the judicial decisions showed three different patterns: speech that directly led to subsequent acts of violence, hatred, or discrimination; speech that was decided to potentially cause harm; and speech that is assumed capable of causing harm.

Table 1. The pattern of nexus between speech and harm according to the 27 judicial decisions

Speech
1 Directly causing consequences	2 Calculated to cause harm	3 Assuming capable of causing harm
Signalled by how people reacted to the speech	Based on pre-existing conditions that are assessed by the following factors: Prevailing circumstances; The social/historical context under which the speech was given.	Based on the subjective perception of the judges

With regards to the first pattern namely speeches that directly cause consequence, an example could be seen in a case adjudicated by the District Court of Surabaya that involves pre-existing tension between two groups called the PSHT Nusantara and Pagar Nusa. In this case, the defendant had sent a text message to a WhatsApp group chat that reads: “seeing how our female friends are beaten up by those mongoose bastards … it’s about time we finish them.” In their consideration, the court acknowledged the provocative nature of the text message, as it called for violence. In establishing the nexus between the speech and the element of enmity, the court emphasised the fact that following the text message, a ruction occurred between the two groups⁷⁵ showing a direct cause-and-effect relationship between the speech and enmity.

Another example can be found in a case concerning a post on Facebook that presented blasphemous claims against the Prophet Muhammad. The post was seen and read by a witness who was part of the Societal Alliance for Anti-Blasphemy against the Islamic Religion. In deciding the nexus between the speech and enmity against a group, the District Court of Simalungun considered how the post offended the Muslim locals in Simalungun, in which 300 of them came to the town hall demanding that the owner of the Facebook account be apprehended.⁷⁶

With regards to the second pattern, namely, speech that is predicted to cause harm. In determining the probability of a speech causing such consequence, the judicial decisions showed that the courts considered the circumstances under which the speech was delivered,⁷⁷ and the social⁷⁸ or historical context.⁷⁹ An example can be seen from a case adjudicated by the District Court of Makassar, when a person commented on a news feed on Facebook about how the regional police force in South Sulawesi was in pursuit of those who provokes anti-rapid test for Covid. In his comment, the defendant wished the head of the regional police “gets contracted with Corona along with his children” because he is a “conman who extorts the whole nation.” In its consideration, the court argued that the comment was posted in the midst of the government’s attempt to enforce a social distancing policy through the police force during the Coronavirus pandemic. Under this circumstance, the court deemed that such a comment could potentially elevate existing tensions between the public and the police force, as there had been numerous occasions where the police force engaged in violent resistance from those who refused to abide by the social distancing policy. In the end, the court decided that given the circumstances under which the speech was presented, it could potentially lead to a subsequent hatred or enmity against the police force.⁸⁰

Furthermore, an example where historical context is given a significant consideration can be found in a case adjudicated by the District Court of Banda Aceh. In this case, the defendant was a member of the Pembebasan Kemerdekaan Atjech Darussalam/Atjeh Merdeka (PKAD/AM), which is an organisation that operates to free Aceh as an independent state. The defendant uploaded several videos that conveyed the group’s intention to resume their attempt for independence. In the videos, the defendant constantly warned viewers that “non-nationals of Aceh” and “less-interested parties” should leave the area before 4 December 2019 and threatened to “take action” against those who do not abide. The defendant also conveyed his plans to kidnap and “perhaps murder” 21 people whom he deemed disloyal to the commitments of PKAD/AM. He also reiterated that “firm actions” will be taken against “non-nationals of Aceh” who choose to remain in the area “either by force or in good faith.” The court perceives these messages as creating a “we v. them” distinction between the PKAD/AM and “non-nationals of Aceh” therefore, fulfilling the element of “intergroup.” Furthermore, the court viewed that the messages could potentially lead to consequence given two pre-existing conditions. First, the history of Aceh related to violent attempts to separate from the Republic of Indonesia. Second, the comments posted in response to the videos that saw the tension between those in favour and those against the idea of the separatist movement. Based on these considerations, the court decided that the video could potentially “provoke followers to join the separatist movement and possibly resort to violence that would give rise to hatred and/or enmity amongst the groups in the society.”⁸¹

Lastly, with regards to the third pattern, some case decisions show that the court assumed the negative impact of a given speech. However, the courts taking this approach did not elaborate further on how they measured the probability of impact, thus giving no substantive indication of harm. Reverting to a previous example about the food poisoning post, without any elaboration, the court perceived that the post displayed provocative nature that may disturb public order and security in the area of Mimika. Reverting to another previous example concerning the atheistic post, the court refrained from explaining how such a post would “anger the audience” where no such evidence was presented in the case.

In another instance that concerned blasphemous remarks against a certain religion, the remarks were perceived to have insulted people belonging to the religion without elaborating how, when no evidence was given to support such findings.⁸² Another example can be seen in a case that concerns a post expressing disbelief against religion and God. The court considered that such a post “might cause hatred against an individual or group based on religion” because “it demeans and insult God” without further elaborating on which group and the actual impact.⁸³ Considering that Article 28(2) explicitly focuses on hate speech made against an individual or a group, this practice implies that blasphemous remarks against religion is automatically hate speech against the people holding such religion.

Another example is a case adjudicated by the East Jakarta District Court. The case concerned a man who posted a video of himself cursing and expressing obscene language towards the police force because the police had come to his home in the middle of the night.⁸⁴ In the video, the man was seen to equate the police to dogs, calling them “bastards” and even went on to threaten to kill them. In its decision, the court decided that the defendant’s words “disturbed the public and potentially causes hatred against the Indonesian Police Force.”⁸⁵ The conclusion was premature, considering the lack of reasoning and evidence to support the public disturbance findings except for the fact that it was a policeman who reported the defendant after randomly conducting cyber patrol on social media.

Furthermore, in a case adjudicated by the District Court of Limboto, the defendant uploaded a photo on social media of a third person with tattoos edited on both his cheeks. In the caption, the defendant wrote that the person was the champion of the area who could take on just about anyone and that neither the police nor the military was a match for him—claiming that he had once put a military personnel on a neck lock with his foot. The defendant also claimed that the person sends his regard to the city police station in Gorontalo and that he did not fear them.⁸⁶ Without elaboration, the court decided that the statements “will cause hatred or enmity against the military.”

One particular case actually acknowledged that no direct effect occurs, however, the defendant was still punished. The case was adjudicated by the District Court of Pangkalan Bun. The defendant had posted an Instagram story showing himself to be in a crowded area despite the social distancing policy that had come into effect following the Coronavirus pandemic. The caption of the Instagram story contained derogatory remarks aimed at the police who were patrolling local public spaces at the time. The defendant called the police with vulgar language, “hey policemen you d***s and dogs, I’m outside, come at me. F**k Covid”). Seeing the comments responding to the Instagram story, the court acknowledged that no direct effect occur following the post. Rather, the comments on the Instagram story seem to disapprove the defendant’s action of violating the social distancing policy—meaning that the Instagram story provoked no one to display the same behaviour. Despite so, the court reasoned that a conviction would be appropriate to educate the general public that using social media comes with the obligation to act responsibly. In addition, the court considered that similar regard towards the police might reoccur creating a pattern conducive to hatred against the police force.⁸⁷ From these cases above, none of the speeches resulted in any of those consequences except disapproving comments from the readers themselves.

5. The unclear scope of harm

Based on the cases above, the Author found that there is an unclear scope of harm. In cases of hate speech, there is more urgency to include such element. Aside from the idea that harm justifies the state’s intervention through means of criminal punishment, hate speech is a negative product of the freedom of expression that serves as a pillar of democracy.⁸⁸ Particularly, if seen from the perspective of political participation, speeches about political matters, in particular, are the paradigmatic kind of expression that such freedom serves to protect⁸⁹ since they are a way to prevent abuses of power by public officials.⁹⁰ That being said, the state must provide a justification to punish people who are exercising their fundamental liberties⁹¹ because limiting ideas that citizens are entitled to discuss over law and policy “encroaches upon the elements that make the state a democracy.”⁹² In other words, to criminalise people for expressing a degree of dissatisfaction against the status quo, for example, would deprive the essence of democracy itself. As Feinberg noted, the harm principle is the only valid principle to legitimize the invasion of liberty.⁹³

With regard to the criteria of harm, the Author refers to the “clear and present danger” test that was introduced by Justice Holmes in Schenck v. United States.⁹⁴ The test requires assessing whether the contested speech is used in a way that creates a clear and present danger that will bring about substantive evils.⁹⁵ However, the test changed radically in Abrams v. United States⁹⁶ when Holmes and Brandeis opined that the nature of harm must be very serious and certain in a way that requires an “immediate check” in order “to save the country.”⁹⁷ In essence, his notion is that only imminent harm can justify punishment for the expression of speech because it could be measured objectively.⁹⁸

That being said, the Author argues that only the first (directly causing harm) and second (calculated to cause harm) patterns would suffice as harm for the purpose of prosecuting hate speech since clear and present danger could be assessed in both.

The presence of an actual harm would deliberately render criminal liability for the maker of hate speech. This is because in criminal law, the liability of an offender depends on the legal proximate cause of harm, referring to the conduct upon which the law determines to be the primary cause of injury.⁹⁹ This is known as a “cause in fact”, which requires the offender’s action to have been necessary to the victim’s harm.¹⁰⁰ In establishing “cause in fact” the judiciary would have to establish whether the same result or harm would have happened anyway without the defendant’s action. In other words, the offender’s action must produce a foreseeable consequence without any external intervention. A good example of how hate speech causes actual harm is the case in Surabaya which concerned a conflict between the groups “PSHT Nusantara” and “Pagar Nusa”.

On the other hand, imminent harm concerns the possibility of harm or indication of danger, risk, or remote harm.¹⁰¹ In line with the Rabat Action Plan, this would require an assessment of the degree of likelihood that harm will occur. Of course, such assessment works on a case-by-case basis but it involves consideration of the gravity of risk taking into account the context that prevails at the time when the speech is delivered, including political, historical, and social context. The presence of harmful side consequences can also be an indication of high likelihood. Therefore, courts should assess the probability of the impact itself or in other words, the likelihood that a speech might actually lead to the risk of others acquiring the same negative impression or perception towards a certain subject.¹⁰² Without such a clear assessment, it would mean that individuals could be prosecuted without probable cause. A good example of a case where imminent harm was assessed is the PKAD/AM case.

As such, the Author argues that assuming harm would occur (the third pattern found) should not be a threshold for the judiciary to determine criminal liability. The reason is that this would rely solely, if not too much, on subjective assessment. In this sense, whatever effect that a speech might inflict depends on how it is received by the audience, intellectually, emotionally, and in terms of prior doxastic attitudes.¹⁰³ As seen from the cases above, if enough group members are offended, the court would accept that as a sign of hate against the group. However, this threshold is prone to abuse by simply gathering individuals that share the same identity.¹⁰⁴ Furthermore, assuming harm would occur would largely be based on general patterns of influence as opposed to anything specific related to the speech itself and how it could actually cause harm.¹⁰⁵ The danger of the third pattern is further amplified by the subjective nature of hate and enmity. Without the presence of actual harm or imminent harm, any individual can claim to be victimised by hate speech, especially considering no threshold is available to objectively assess hate and enmity.¹⁰⁶

6. Moving forward

There should be a clear distinction between hate speech that renders criminal liability and those that do not. For this purpose, Assimakopoulos et.al. presented two categories of hate speech: hard hate speech, which includes prosecutable forms of hate speech that are prohibited and criminalised by law; and soft hate speech, which remains lawful, despite showcasing indications of intolerance and discrimination.¹⁰⁷

Such kind of distinction is indicated in a recent development in Indonesia when the Ministry for Communication and Information, the General Attorney, and the Chief of the Police Force issued a joint decree that is meant to provide a guideline to implement hate speech provision in EIT Law.¹⁰⁸ The Decree was issued pursuant to President Jokowi’s instruction to formulate a guideline on the interpretation of certain articles in the EIT Law including Article 28(2). According to this decree, hate speech under Article 28(2) EIT Law requires the legal enforcers to prove a “motive to create hatred or enmity on the basis of ethnicity, religion, race, and intergroup.” The decree went further as to state that the mere expression of opinions and statements of discontent towards individuals or groups do not amount to the definition of hate speech unless the above indications are present. If we compare the wording of Article 28(2) of EIT Law and the Joint Decree, the difference lies in how the latter clarifies the character of hate speech that contains a motive to cause hate or enmity as can be indicated by incitement or other forms to such effect.

In the most recent development, the new Law No. 1 of 2023 regarding the Criminal Code¹⁰⁹ replaced Article 28(2) of EIT Law with Article 243(1) of the Code which reads:

Anyone who broadcasts, shows, or attaches writing or pictures so that they are visible to the public or plays recording so that they are heard, or disseminates by means of information technology, which contains statements of hostility with the intention to make the content known by the public, against one or several groups of the Indonesian population based on race, nationality, ethnicity, skin color, religion, belief, gender, mental or physical disability that results in violence against people or property shall be punished by a maximum of 4(four) years imprisonment or a maximum fine based on Category IV

There are several differences between Article 28(2) of EIT Law and Article 243(1) of the Code. First, the element of distribution of speech is further clarified with the requirement that such distribution is visible to or heard by the public. Second, the statement of hostility is delivered with the intention to make the content known to the public. These two elements solidify the character of hate speech as being “public” and thus exclude statements made through private communication. Third, hate speech under Article 243(1) specifies the types of socially constructed groups, thus excluding the multi-interpreted term “intergroup.” Fourth, the speech results in violence against people or property.

In light of this new development, the law now makes it clear that the harm in hate speech is defined by the consequential approach. As stated in the article, the statement of hostility is expected to result in violence in order to qualify as hate speech. However, this approach would also require proving an active relationship between the speech and the harm. The challenge is that neither the provision nor the elucidation of the newly adopted Criminal Code explains the form of a statement that is capable of causing harm.

Such a threshold must be present since a consequential approach to harm would mean that a speech is capable of persuading the audience to believe in the negative stereotypes that lead them to engage in harmful conduct; or capable of shaping the preferences of the audience so that they are persuaded of negative stereotypes; or capable of conditioning the environment so that expressing negative stereotypes and carrying out further discrimination becomes normalized; or causing the audience to display the similar sentiment or behaviour.¹¹⁰ This implies that a statement must carry a power of influence in order to render a hate speech.

Without having a threshold in place, a similar problem of inconsistency in the judicial practice concerning hate speech cases might reoccur and there would be no certainty as to the form of statements that may render hate speech. Having such a threshold becomes more pressing, considering that the elaborative guidelines provided in the Circular Letter and the Joint Ministerial Decree are applicable to Article 28(2) of EIT Law and do not extend to the newly adopted Criminal Code—leaving Article 243(1) open to multi-interpretations.

As such, the Author opines that perhaps another guideline is necessary to complement Article 243(1) of the Criminal Code, to clarify that the form of the statement must be incitement or carry an effect conducive to provoking others to engage in negative stereotypes or behave discriminately or violently instead of being framed as a mere comment. The forms of such statements could be imagined by referencing the provisions of some relevant international conventions. For example, “direct and public incitement” of genocide under the Convention on the Prevention and Punishment of the Crime of Genocide. Aside from incitement, the nature of the speech could also be an advocacy. For example, “any advocacy of national, racial, or religious hatred that constitutes incitement to discrimination, hostility or violence … ” under Article 20(2) of ICCPR. This formulation falls short of incitement of genocide because it excludes any dolus specialis intent to destroy a protected group. Rather, Article 20(2) only requires incitement to discrimination, hostility, or violence towards a certain group without the intent to destroy the group. Furthermore, Article 4 of the International Convention on the Elimination of all Forms of Racial Discrimination requires states to “condemn all propaganda and all organisations which are based on ideas of theories of superiority of one race or group of persons of one colour or ethnic origin, or which attempt to justify or promote racial hatred and discrimination in any form…”

Meanwhile, speeches that use profanity to express dislike, intolerance, dissatisfaction, or discontent should not constitute prosecutable hate speech without incitement; or provocation to discriminate, commit violence, or threaten to use violence; advocacy, or propaganda. These speeches, depending on each case, could fall into other categories of speech crimes for example insult.

Other speeches of profanity that render a lower gravity could be classified as “soft hate speech” in line with the suggestion of the UNHCHR above. For example, the use of profanity in expressing sentiment towards perceived injustice would, at least, be understood.¹¹¹ In this regard, soft hate speech could be addressed through extra-legal measures including counter-speech and education.¹¹² According to Brown, in order to make sense of extra-legal measures, we need to acknowledge that hate speech is not only a legal concept that would naturally limit the responses to the contours of law. Rather, hate speech is also an ordinary concept that is also found in social, cultural, and political domains.¹¹³ Therefore, by acknowledging this, we may open ourselves to other forms of responses outside of criminal for example civil litigation.¹¹⁴

7. Conclusion

The lack of an objective threshold of Article 28(2) of EIT Law had led to the broad determination of hate speech as seen in how harm is considered in hate speech cases. The twenty-seven judicial decisions indicate that the judiciary adopts the consequential approach to the harm of hate speech. However, the judicial assessment of the scope of protected groups, the spread of speech, the content of speech, as well as the nexus between speech and consequence is inconsistent in terms of showing how the contested speech is harmful enough to have caused any consequence. Recently, the newly adopted Criminal Code replaced Article 28(2) of EIT Law with a provision that provides better clarity on the nuance of hate speech. Even so, the content of the speech is merely characterized as a statement of hostility, which leaves the provision open to interpretation pertaining to the form of such a statement. But if the statement is expected to result in violence, therefore, the statement must be a form of incitement or it must carry an effect conducive to provoking others to engage in negative stereotypes or behave discriminately or violently.

Competing interests

The author has no competing interests to declare.

Correction

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Acknowledgments

The Author would like to thank the Research and Publication Unit at the Faculty of Law, Universitas Gadjah Mada for the funding that made this research article possible. Her gratitude also extends to the Ministry of Education, Culture, Research, and Technology of the Republic of Indonesia and the Indonesia Endowment Funds for Education for the scholarship provided to the Author to support this publication. Special gratitude is also conveyed to the Author’s research assistants: Ferdinandus Credo R.A. and Indy Putri Riyadi, the anonymous reviewers, as well as the colleagues who offered constructive feedback on the manuscript (alphabetically ordered): Diantika Rindam Floranti, Elaine Mak, Kees van den Bos, and Mohammad Ibrahim.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research article is funded by the Faculty of Law, Universitas Gadjah Mada, the Ministry of Education, Culture, Research and Technology of the Republic of Indonesia (Balai Pembiayaan Pendidikan Tinggi), and the Indonesia Endowment Funds for Education (LPDP).

Notes on contributors

Devita Kartika Putri

Devita is an Assistant Professor at the Criminal Law Department of the Faculty of Law, Universitas Gadjah Mada, Indonesia. Currently, she is undergoing a Ph.D. study at Utrecht University where her research focuses on Indonesia’s deradicalization program and how it affects the convicted terrorists’ perceptions of law and democracy. Her research interest includes violent radicalization and terrorism, human rights and criminal law, and transnational criminal law.

Notes

1. 293/Pid.Sus/2020/PN.Rgt, District Court of Rengat, (2020).

2. 206/Pid.Sus/2019/PN Tim, District Court of Timika, (2020).

3. Raphael Cohen-Almagor, “Freedom of Expression v. Social Responsibility: Holocaust Denial in Canada,” Journal of Mass Media Ethics 28, no. 1 (Cohen-Almagor, 2013): 15.

4. James B Jacobs and Kimberly Potter, Hate Crimes: Criminal Law and Identity Politics (New York: Oxford University Press (Jacobs & Potter, 1998)).

5. B. Parekh, “Is There a Case for Banning Hate Speech?,” in The Content and Context of Hate Speech: Rethinking Regulation and Responses, ed. M. Herz and P. Molnar (Cambridge: Cambridge University Press (Parekh, 2012)), 37–56; J.W. Howard, “Free Speech and Hate Speech,” Annual Review of Political Science 22 (Howard, 2019): 93–109, https://doi.org/10.1146/annurev-polisci-051517–012343.

6. Various of international human rights treaties restrict speeches that are discriminatory in nature namely, Article III of the Convention on the Prevention and Punishment of the Crime of Genocide, Article 20(2) of the International Covenant on Civil and Political Rights, and Article 4 of the International Convention on the Elimination of All Forms of Racial Discrimination.

7. Yulida Medistiara, “Selama Polri Tangani 3.325 Kasus Ujaran Kebencian,” Detik News, December (Medistiara, 2017).

8. Igman Ibrahim, “Polda Metro Jaya Tangani 480 Kasus Ujaran Kebencian Dan Hoaks Selama Pandemi Corona,” Tribunnews, June (Ibrahim, 2020).

9. For this Article, I will use the term “speech” as interchangeable with “information”.

10. Article 28(2) is complemented by Article 45A(2) of EIT Law which reads.Anyone who, intentionally and without right, distribute information intended to create a sense of hatred or enmity towards individuals and/or groups of certain communities based on ethnicity, religion, race, and intergroup as regulated in Article 28(2) shall be punished by a maximum of 6 (six) years of imprisonment and/or a maximum fine of Rp. 1.000.000.000,00 (one billion Rupiah).

11. Budiman, Adhigama A , Maya, Genoveva Alicia K.S. , Rahmawati, Maidina , Abidin, Zainal, “Mengatur Ulang Kebijakan Pidana Di Ruang Siber” Institute for Criminal Justice Reform (Jakarta, 2021).

12. Tarlagh McGonagle, “The Council of Europe Against Online Hate Speech: Conundrums and Challenges” (McGonagle, 2013).,

13. Alexander Brown, “What Is Hate Speech Part 1: The Myth of Hate,” Law and Philosophy 36 (Brown, 2017): 419–68, https://doi.org/10.1007/s10982-017-9297-1.

14. United Nations High Commissioner for Human Rights, “Report of the United Nations High Commissioner for Human Rights on the Expert Workshops on the Prohibition of Incitement to National, Racial or Religious Hatred” (2013).,

15. Herbert L Packer, The Limits of the Criminal Sanction (California: Stanford University Press (Packer, 1968)).

16. Nina Peršak, Criminalising Harmful Conducts: The Harm Principle, Its Limits and Continental Counterparts (New York: Springer (Peršak, 2007)), 48.

17. L.W. Sumner, “Criminalizing Expression: Hate Speech and Obscenity,” in The Oxford Handbook of Philosophy of Criminal Law, ed. John Deigh and David Dolinko (Oxford: Oxford University Press (Sumner, 2011)), 3.

18. Michael Moore, “Causation in the Criminal Law,” in The Oxford Handbook of Philosophy of Criminal Law, ed. John Deigh and David Dolinko (Oxford: Oxford University Press (Moore, 2011)), 1.

19. Joxerramon Bengoetxea, “Ultima Ratio and the Judicial Application of Law,” Oñati Socio-Legal Series 3, no. 1 (Bengoetxea, 2013): 107.

20. Bengoetxea, “Ultima Ratio and the Judicial Application of Law”

21. Douglas Husak, “Retribution in Criminal Theory,” San Diego Law Review 37, no. 4 (Husak, 2000): 962.

22. Jeremy Waldron, The Harm in Hate Speech (United States: Harvard University Press (Waldron, 2012)).

23. Jeremy Waldron, The Harm in Hate Speech.

24. Eric Barendt, “What Is the Harm of Hate Speech?” Ethical Theory and Moral Practice 22, no 3 (Barendt, 2019)

25. Katherine Gelber and Luke McNamara, “Evidencing the Harms of Hate Speech,” Social Identities 22, no. 3 (Gelber & McNamara, 2016).

26. Ishani Maitra and Mary Kate McGowan, eds., Speech and Harm: Controversies over Free Speech (Oxford: Oxford University Press (Maitra & McGowan, 2012)).

27. Article 20 reads “Any propaganda for war shall be prohibited by law. Any advocacy of national, racial or religious hatred that constitutes incitement to discrimination, hostility or violence shall be prohibited by law.”

28. Simon Butt, “Judicial Reasoning and Review in the Indonesian Supreme Court,” Asian Journal of Law and Society 6 (Butt, 2019): 67–97.

29. Rohmawati, “Legal Reasonings of Religious Court Judges in Deciding the Origin of Children: A Study on the Protection of Biological Children’s Civil Rights,” Ijtihad: Jurnal Wacana Hukum Islam Dan Kemanusiaan 21, no. 1 (Rohmawati, 2021): 1–19.

30. 910/Pid.Sus/2020/PN Bdg, District Court of Bandung (2020).

31. Constitutional Court, Constitutional Court Decision No 76/PUU-XV/2017, (2017).

32. Constitutional Court Constitutional Court Decision No 76/PUU-XV/2017, (2017).

33. 1527/Pid.Sus/2020/PN Sby, District Court of Surabaya, (2020).

34. 135/Pid.Sus/2019/PN Pbl, District Court of Probolinggo, (2020).

35. 121/Pid.Sus/2020/PN Plk, District Court of Palangka Raya, (2020).

36. Any person who publicly gives expression to feelings of hostility, hatred or contempt against the Government of Indonesia, shall be punished by a maximum imprisonment of seven years or a maximum fine of three hundred Rupiahs.

37. Any person who disseminates, openly demonstrates or puts up a writing where feelings of hostility, hatred or contempt against the Government of Indonesia are expressed with intent to give publicity to the contents or to enhance the publicity thereof, shall be punished by a maximum imprisonment of four years and six months or a maximum fine of three hundred Rupiahs.

38. Constitutional Court Constitutional Court Decision No 6/PUU-V/2007, (2007).

39. Z. Aditya and S. Al-Fatih, “Indonesian Constitutional Rights: Expressing and Purposing Opinions on the Internet,” The International Journal of Human Rights (Aditya & Al-Fatih, 2020)22–23.

40. Aditya and Al-Fatih, 22–23.

41. 422/Pid.Sus/2020/Pn.Bks, District Court of Bekasi, (2020).

42. 886/Pid.B/2020/PN.JKT. TIM, District Court of Jakarta Timur, (2020).

43. 61/Pid.Sus/2020/PN Lbo, District Court of Limboto, (2020).

44. 205/Pid.Sus/2020/PN.Pbu, District Court of Pangkalan Bun, (2020).

45. For the protected groups against genocide, see the chapeau elements of Article 6 in the Rome Statute of the International Criminal Court. For protected groups against hate speech, see Article 20(2) of the International Covenant on Civil and Political Rights. For protected groups against discrimination, see Article 1 in the International Convention on the Elimination of All Forms of Racial Discrimination.

46. The case of Prosecutor v. Jean-Paul Akayesu in the International Criminal Tribunal for Rwanda (ICTR) concerned acts of genocide committed by the ethnic group of Hutu against the Tutsi. When adjudicating the case, the Chamber faced difficulties in deciding which of the four protected groups the Tutsi falls into. In making their determination, the Chamber considered the intent of the drafters of the Genocide Convention, which, according to the travaux préparatoires, was meant to protect any stable and permanent group. Even though Indonesia is not bound by the decisions of the ICTR, however, their decisions reflect the interpretation and application of international law, hence, making them an appropriate source of law.

47. Prosecutor v. Akayesu, lCTR Trial Chamber, Judgment, ICTR T. Ch. I 2.9. (1998).

48. William Schabas, “The Crime of Genocide in the Jurisprudence of the International Criminal Tribunals for the Former Yugoslavia and Rwanda,” in International and National Prosecution of Crimes under International Law (Berlin: Berlin Verlag (Schabas, 2001)), 451–52.

49. Prosecutor v. Akayesu, lCTR Trial Chamber, Judgment, ICTR T. Ch. I 2.9. (1998) para 513.

50. United Nations, “Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities A/RES/47/135,92nd Plenary Meeting (18 December (1992))” (1992).

51. Waldron, The Harm in Hate Speech.

52. C. Phiri, “Criminal Defamation Put to the Test: A Law and Economics Perspective,” University of Baltimore Journal of Media Law & Ethics 9, no. 1 (Phiri, 2021); Vidya Prahassacitta and Harkristuti Harkrisnowo, “Criminal Disinformation in Relation to the Freedom of Expression in Indonesia: A Critical Study,” Comparative Law Review 27 (Prahassacitta & Harkrisnowo, 2021): 135–65.

53. 223/Pid.Sus/2020/PN Tte, District Court of Ternate, (2020).

54. 40/Pid.Sus/2020/PN.Bna, District Court of Banda Aceh, (2020).

55. 121/Pid.Sus/2020/PN Bla, District Court of Blora, (2020).

56. 205/Pid.Sus/2020/PN.Pbu, District Court of Pangkalan Bun.

57. 690/Pid.Sus/2020/PN Mtr, District Court of Mataram (2020); 293/Pid.Sus/2020/PN.Rgt, District Court of Rengat.

58. 422/Pid.Sus/2020/Pn.Bks, District Court of Bekasi.

59. 1234/Pid.Sus/2020/PN Jkt.Utr., District Court of North Jakarta (2020); 39/Pid.Sus/2020/PN Skh, District Court of Sukoharjo (2020); 1527/Pid.Sus/2020/PN Sby, District Court of Surabaya.

60. 1234/Pid.Sus/2020/PN Jkt.Utr., District Court of North Jakarta.

61. Ministry for Communication and Information, General Attorney, and Chief of the Police Force, “Joint Decree between the Ministry for Communication and Information, the General Attorney, and the Chief of the Police Force, No. 229 of 2021, No. 154 of 2021, and No. KB/2/VI/2021 Concerning the Implementing Guideline for Certain Articles in Law No. 11 of 2008 regarding Electronic Information and Transaction as amended by Law No. 19 of 2016 on the Amendments of Law No. 11 of 2008 regarding Electronic Information and Transaction” ().

62. 181/Pid.Sus/2020/PN Stg, District Court of Sintang, (2020).

63. 690/Pid.Sus/2020/PN Mtr, District Court of Mataram.

64. 240/Pid.Sus/2020/PN Cbn, District Court of Cirebon (2020); 372/Pid.Sus/2020/PN Sim, District Court of Simalungun (2020).

65. 126/Pid.Sus/2019/PN Msb, District Court of Masamba, (2020).

66. 910/Pid.Sus/2020/PN Bdg, District Court of Bandung (2020); 181/Pid.Sus/2020/PN Stg, District Court of Sintang.

67. 121/Pid.Sus/2020/PN Plk, District Court of Palangka Raya, (2020).

68. 135/Pid.Sus/2019/PN Pbl, District Court of Probolinggo.

69. 154/Pid.Sus/2020/PN Sdw, District Court of Kutai Barat, (2020).

70. 1520/Pid.Sus/2019/PN Mks, District Court of Makassar, (2020).

71. 206/Pid.Sus/2019/PN Tim, District Court of Timika.

72. 293/Pid.Sus/2020/PN.Rgt, District Court of Rengat.

73. 3478/Pid.Sus/2019/PN Mdn, District Court of Medan, (2020).

74. 1234/Pid.Sus/2020/PN Jkt.Utr., District Court of North Jakarta.

75. 1527/Pid.Sus/2020/PN Sby, District Court of Surabaya.

76. 372/Pid.Sus/2020/PN Sim, District Court of Simalungun (2020).

77. 1226/Pid.Sus/2020/PN Mks, District Court of Makassar, (2020).

78. 90/Pid.Sus/2020/PN Mtw, District Court of Muara Tewe, (2020).

79. 40/Pid.Sus/2020/PN.Bna, District Court of Banda Aceh.

80. 1226/Pid.Sus/2020/PN Mks, District Court of Makassar.

81. 40/Pid.Sus/2020/PN.Bna, District Court of Banda Aceh.

82. 121/Pid.Sus/2020/PN Bla, District Court of Blora.

83. 155/Pid.Sus/2020/PN Pmk, District Court of Pamekasan, (2020).

84. 886/Pid.B/2020/PN.JKT. TIM, District Court of Jakarta Timur.

85. 886/Pid.B/2020/PN.JKT. TIM, District Court of Jakarta Timur.

86. 61/Pid.Sus/2020/PN Lbo, District Court of Limboto.

87. 205/Pid.Sus/2020/PN.Pbu, District Court of Pangkalan Bun.

88. Kevin W. Saunders, Free Expression and Democracy: A Comparative Analysis (United Kingdom: Cambridge University Press Saunders, (2017)).

89. Howard, “Free Speech and Hate Speech”

90. Vincent Blasi, “The Checking Value in First Amendment Theory,” American Bar Foundation Research Journal 2, no. 3 (Blasi, 1977): 521–649.

91. Douglas Husak, “The Criminal Law as Last Resort,” Oxford Journal of Legal Studies 24, no. 2 Husak, (2004): 210.

92. E. Heinze, Hate Speech and Democratic Citizenship (Oxford: Oxford University Press (Heinze, 2016)), 5.

93. Joel Feinberg, The Moral Limits of the Criminal Law: Harm to Others, Vol. 1 (Oxford: Oxford University Press (Feinberg, 1984)).

94. Schenck v. United States, No. 249 US 47 (Schenck, 1919).

95. Ronald J. Krotoszynski Jr., “The Clear and Present Dangers of the Clear and Present Danger Test: Schenck and Abrams Revisited,” SMU Law Review 72, no. 3 (Krotoszynski & Ronald, 2019).

96. Abrams v United States, No 250 US 616, 630, (1919).

97. Ronald J. Krotoszynski Jr., “The Clear and Present Dangers of the Clear and Present Danger Test: Schenck and Abrams Revisited.”

98. Vincent Blasi, “Holmes’s Understanding of His Clear-and-Present-Danger Test: Why Exactly Did He Require Imminence?,” Seton Hall Law Review 51 (Blasi, 2020): 176.

99. H.L.A. Hart, “Acts of Will and Responsibility,” in Punishment and Responsibility Essays in the Philosophy of Law (Oxford: Oxford Scholarship Online (Hart, 2009)).

100. Moore, “Causation in the Criminal Law”

101. Peršak, Criminalising Harmful Conducts: The Harm Principle, Its Limits and Continental Counterparts.

102. van Marloes Noorloos, “Criminalising Defamation of Religion and Belief,” European Journal of Crime, Criminal Law, and Criminal Justice 22 (Noorloos van, 2014): 355.

103. M. Badar and P. Florijančič, “Assessing Incitement to Hatred as a Crime against Humanity of Persecution,” The International Journal of Human Rights 24, no. 5 (Badar & Florijančič, 2020): 656–87.

104. Muhammad Ryandaru Danisworo and Agustian Budi Prasetya, “Implementing Prohibition of Hate Speech Under Article 28(2) of Indonesian Electronic Information and Transaction Law,” in Proceedings of the 2nd International Conference on Law and Human Rights 2021 (Advances in Social Science, Education and Humanities Research (Danisworo & Budi Prasetya, 2021)), https://doi.org/10.2991/assehr.k.211112.020.

105. Vincent Blasi, “Holmes’s Understanding of His Clear-and-Present-Danger Test: Why Exactly Did He Require Imminence?,” 186.

106. Danisworo and Prasetya, “Implementing Prohibition of Hate Speech Under Article 28(2) of Indonesian Electronic Information and Transaction Law”

107. Stavros Assimakopoulos, Fabienne H. Baider, and Sharon Millar, Online Hate Speech in the European Union: A Discourse-Analytic Perspective (Springer Briefs in Linguistics (Stavros et al., 2017)), 4–5, https://doi.org/10.1007/978-3-319-72604-5.

108. Ministry for Communication and Information, General Attorney, and Chief of the Police Force, “Joint Decree between the Ministry for Communication and Information, the General Attorney, and the Chief of the Police Force, No. 229 of 2021, No. 154 of 2021, and No. KB/2/VI/2021 Concerning the Implementing Guideline for Certain Articles in Law No. 11 of 2008 regarding Electronic Information and Transaction as amended by Law No. 19 of 2016 on the Amendments of Law No. 11 of 2008 regarding Electronic Information and Transaction” (2021).

109. The new Criminal Code was adopted in early 2023. However, it will be enforced in 2026 after three years of transition period.

110. Ishani Maitra and Mary Kate McGowan, eds., Speech and Harm: Controversies over Free Speech (Oxford: Oxford University Press (Maitra & McGowan, 2012)).

111. Robert Post, “Hate Speech,” in Extreme Speech and Democracy, ed. Ivan Hare, James Weinstein, and Eds (New York: Oxford University Press (Post, 2009)), 123.

112. Brown, “What Is Hate Speech Part 1: The Myth of Hate”

113. Brown.

114. Nils Jareborg, “Criminalization as Last Resort (Ultima Ratio),” Ohio State Journal of Criminal Law 2 (Jareborg, 2005): 521–34.