Thailand’s Law on Criminal Online Falsehoods: A Critical Discussion

Lasse Schuldt * and Pudit Ovattananavakhun †

Abstract

This article critically discusses the Thai criminal law applicable to online falsehoods, namely Section 14 para. 1(1) and (2) of the Act on Computer-Related Offences. Linking developments in Thailand to global and Southeast Asian fake news discourses, the article’s main part sheds light on several interpretational and constitutional complexities. Conflicting concepts of falsity, structural inconsistencies and an uncertain ambit of protected interests are found to persist, despite legislative amendments. It is argued that criminal punishment should depend on proof of actual rather than likely damage. In the light of recent constitutional jurisprudence, the level of punishment is found to constitute a disproportionate restriction of the right to freedom of expression. The article provides an in-depth analysis that contributes to the evolving scholarship on the challenges of regulatory responses to fake news. It concludes that education in media literacy and critical reflection are the approaches best suited to enhance society’s resilience against manipulated information.

Keywords: Fake news — Computer crimes — Free speech

I. INTRODUCTION: CRIMINAL ONLINE
FALSEHOODS AND THE FAKE NEWS PARADIGM

In 2007, the Thai National Legislative Assembly enacted the Act on Computer-Related Offences 2007. [1] It was Thailand’s first law criminalising the distribution of false information through computer systems. The relevant provisions, namely Section 14 para. 1(1) and (2) constitute Thailand’s anti-fake news law. This article discusses its development, enforcement, interpretation and constitutionality. Despite Thailand’s ostensible frontrunner position among countries criminalising fake news, it is argued that multiple legal uncertainties have remained. Meanwhile, however, online falsehoods have become a ubiquitous phenomenon, with governments around the world vowing to eradicate them. The newly created laws focus on the protection of public interests, as opposed to reputational concerns as under traditional defamation laws.

Since the 2016 Presidential Election campaign in the United States, the discourse on the threat from fake news has been spreading globally. The former U.S. President Donald Trump, who himself allegedly benefited from false news stories on social media before the election, [2] popularised the term, frequently labelling traditional mainstream media as “fake.” [3] The global public as well as governments on all continents rapidly adopted the term. [4] Google Trends shows that the interest in “fake news” has indeed risen—quite specifically since September 2016. [5]

The fake news paradigm has also attracted significant academic attention. Initial reservations against adopting the allegedly imprecise term [6] have largely given way to pragmatic acceptance. [7] While academic publications had previously referred to a variety of different phenomena as “fake news”—for instance, news satire and parody, the fabrication or manipulation of reports, advertising techniques, propaganda, and the discrediting of traditional news or of dissenters [8] —the majority now appear to understand the term broadly to mean intentional misinformation, particularly perpetrated via social media. As for the effects of online falsehoods, it is frequently asserted that they may lead to a loss of trust in institutions and modes of governance, contribute to an increase in social polarisation, spread hate, or incite to violence; [9] effects that may be accelerated by selective information exposure and so-called “filter bubbles” on social media. [10] Due to the alleged agenda-setting power of deliberately false information, [11] the truth is described as being increasingly subject to contest. As a result, terms such as “alternative facts” or “post-truth” have found their way into everyday language. [12]

At the same time, research shows that the fake news label has also become a discursive tool to discredit unwanted information, media outlets and politicians. [13] It has been argued that “fake news” is a Laclaudian floating signifier “lodged in-between different hegemonic projects seeking to provide an image of how society is and ought to be structured.” [14] Thus, while online falsehoods have become a legitimate concern for societies and governments around the world, the fight against them has been frequently abused in order to justify unwarranted or excessive restrictions of constitutionally protected free speech. This Janus-faced character of the fake news discourse calls to mind the “war on terror” and its two-decade-long global utilisation to extend governmental powers under the banner of national security. [15]

The growing body of literature on regulatory responses to fake news discusses the promise and the challenges of state-led approaches, self-regulation, and models where fake news is simply “swamped” by truth. [16] While some publications focus more on the question of “how” rather than “if” the phenomenon should be regulated, [17] others have called for careful consideration as to whether governments and technology corporations should become the arbiters of truth. [18] The European Commission has recommended increasing the pressure on social media companies and improving media literacy through enhanced curricula at schools and universities. [19] In Southeast Asia, ASEAN member states favour a stronger role for governments in terms of monitoring and clarification, and the creation of laws, norms and guidelines. [20]

II. SOUTHEAST ASIA’S LEGAL WAR ON FAKE NEWS

In hindsight, Thailand’s criminalisation of online falsehoods foreshadowed a global legislative trend in which Southeast Asian countries quickly took a leading role, making the region the world’s foremost laboratory for anti-fake news laws. [21] This part outlines the major legislative steps across the region, before addressing the Thai developments in a more detailed manner.

In Malaysia, where the former Prime Minister Najib Razak called fake news the “new plague,” [22] an Anti-Fake News Act was adopted in April 2018, just a month before UMNO’s historic electoral defeat. At the time, evidence of Najib Razak’s alleged involvement in the 1MDB corruption scandal was mounting and was being shared on social media, which might have been a reason for the law’s quick passage. Section 4 of the Act provides that “any person who, by any means, maliciously creates, offers, publishes, prints, distributes, circulates or disseminates any fake news or publication containing fake news” commits a crime. Fake news was defined as “any news, information, data and reports, which is or are wholly or partly false” (Section 2). Thus, the law criminalised the spreading of falsehoods, regardless of any probable or actual damage. After the May 2018 election, the Mahathir Mohamad administration pursued the goal of repealing the Act—a campaign promise that was eventually delivered in December 2019. Mahathir himself, however, had previously described the law as “good.” [23] His successor in the office of Prime Minister, Muhyiddin Yassin, has said that he would make use of various laws to combat misinformation on the Internet. [24]

Another prominent case is Singapore, where the Protection from Online Falsehoods and Manipulation Act 2019 (POFMA) was passed by Parliament in May 2019. It entered into force in October of the same year. POFMA’s Section 7 makes it a crime for anybody to make available a false statement of fact to end-users in Singapore that is likely to have negative effects on enumerated public interests. These include national security, public health, safety, tranquillity, finances, international relations, presidential or parliamentary elections or referenda, peaceful relations between different groups, and public confidence in the performance of state authorities. The type of statements the Act considers to be false is defined in Section 2(2)(b) which stipulates that “a statement is false if it is false or misleading, whether wholly or in part, and whether on its own or in the context in which it appears.” The Singaporean government has made it clear that it considers it “not conventional warfare, but a battle within all our societies” to fight against hate speech and fake news that “spread like wildfire.” [25]

In the Philippines, the COVID-19 pandemic triggered the enactment of the Bayanihan to Heal As One Act (Republic Act No. 11469). The Act gave the president emergency powers and criminalised the creation, perpetration and spreading of “false information regarding the COVID-19 crisis on social media and other platforms, such information having no valid or beneficial effect on the population, and . . . clearly geared to promote chaos, panic, anarchy, fear, or confusion” (Section 6(f)). While further anti-falsehood legislation was still pending in the lower house and the Senate, this provision effectively became the Philippines’ first anti-fake news law. At the same time, President Rodrigo Duterte had repeatedly labelled accurate news as fake, [26] and the government itself was found spreading falsehoods. [27]

In Indonesia, the pending draft revision of the Penal Code aims to introduce an anti-falsehood provision in Article 309(1). The draft provision stipulates that “any person who broadcasts fake news or hoaxes resulting in a riot or disturbance shall be punished.” [28] The Indonesian government had already launched a new cybersecurity agency in 2018, which monitors the internet for fake news and holds regular briefings. [29] Due to a perceived COVID-19-related infodemic, [30] the first weeks of the pandemic saw dozens of people arrested for spreading false information online. Similar arrests took place in Vietnam, where in early 2020 the government stipulated administrative fines for people who provide and share fake, untruthful, distorted and slanderous information on social media. COVID-19 also triggered the Cambodian government to enact the Law on National Administration in the State of Emergency. It restricts the publication of information that could cause panic or chaos.

Academic literature with a focus on free speech on the Internet in Southeast Asia has so far largely focused on the effects of improved internet access on political participation, [31] and issues of cyber-repression and defamation. [32] Publications on the Southeast Asian fight against fake news and the corresponding legislation, however, are still scarce. Nonetheless, the techniques Singaporeans use to authenticate information they encounter on social media have been analysed. [33] An edited volume on fake news and elections in Southeast Asia is currently forthcoming. [34] And in Thailand, to which we will now return, a study has investigated social media literacy and the impact of fake news on public opinion in Bangkok. [35]

III. DEVELOPMENT AND ENFORCEMENT OF
THAILAND’S CRIMINAL LAW ON ONLINE FALSEHOODS

In comparison to neighbouring Malaysia and Singapore, the enactment of Thailand’s Act on Computer-Related Offences 2007 was an early legislative response to what would later become the fake news threat. It was part of a broader legislative project going back to 1998, aimed at developing Thai law with a view to keeping pace with advancements in information technology. [36] In 2017, the Act was changed and amended in various respects. [37] This part outlines the development and enforcement of the Thai criminal law on online falsehoods, and the accompanying government discourse.

The Thai Criminal Code has long criminalised false statements of fact that amount to defamation due to their negative effect on the reputation of another person (Sections 326 to 333). The criminal provisions are accompanied in private law by a specific defamation tort that provides a claim for damages (Section 423 of the Civil and Commercial Code). Thai mass media laws do not explicitly address the distribution of false information. They do, however, prohibit the broadcasting of information or the import of printed material which induces to overthrow the administration under the democratic form of government with the King as Head of State; or which is likely to negatively affect national security, public order or good morals; or which is obscene, or may cause a serious deterioration of the mind or health of the people. [38] Thai consumer protection law prohibits false or exaggerated advertisements. [39] In addition, the Criminal Code prescribes punishment for the malicious dissemination of false information that causes people to panic (Section 384). The latter provision’s goal to protect public order conforms with some of the aims pursued by the relevant provisions of the Act on Computer-Related Offences.

Interestingly, initial drafts of the Act which were drawn up between 2002 and 2006 and which used the title “Computer Crime Act” [40] did not contain any content-related provisions. These were added in later versions which were approved by the Legislative Assembly further down in the legislative process, after the military coup of September 2006. [41] The vast majority of the Act’s substantive criminal provisions are genuine computer-related crimes such as unauthorised access to computer systems, deletion of data, and similar acts (Sections 5 to 13). The scope of these sections is largely comparable to the requirements to signatories made by the “Budapest Convention”—the Council of Europe’s Convention on Cybercrime of 2001. [42]

As for the content-related provisions, Section 14 para. 1(1) and (2), which deal with the dissemination of false information through computer systems, form part of a Section that also criminalises the entry into a computer system of computer data which constitutes an offence against the security of the Kingdom, or which constitutes terrorism, or which is obscene (Section 14 paras. 1(3), (4)). Another content-related Section deals with the dissemination of images that affect another person’s reputation (Section 16). The latter part of the Act (Chapter 2, or Sections 18 to 31) stipulates the powers of executive officials and courts in relation to investigations and other interventions. The power to suppress or remove information from computer systems (Section 20) is particularly noteworthy.

The initial wording [43] of Section 14 was as follows:

Section 14. Any person who perpetrates the following offenses shall be subject to imprisonment not exceeding five years or a fine not exceeding one hundred thousand baht, or both:

(1) put into a computer system forged computer data, partially or entirely, or false computer data, in a manner that is likely to cause damage to another person or the public;

(2) put into a computer system false computer data in a manner that is likely to damage national security, or cause panic in the public;

(3) put into a computer system any computer data which is an offense about the security of the Kingdom or is an offense about terrorism, according to the Criminal Code;

(4) put into a computer system any computer data which is obscene, and that computer data is accessible by the public;

(5) disseminate or forward any computer data when being aware that it is computer data as described in (1), (2), (3) or (4).

During the first years of its existence, the Act was enforced with varying intensity. While the restriction of website access figured most prominently, [44] criminal investigations and prosecutions for the distribution of illicit content also formed part of the enforcement by police and judicial authorities. Between July 2007 and December 2011, at least 215 criminal proceedings were launched that focused on content-related crimes under the Act (Sections 14 to 16). About half of these cases dealt with defamation, others with royal defamation, fraud and obscenity, and six of them with national security. [45]

In the following years until the Act’s amendment, charges under Section 14(1) mostly focused on defamation. [46] Among the allegedly defamed persons were private individuals, juristic persons, politicians, state officials and institutions of the state. During this period, defamation charges were frequently based on the Act on Computer-Related Offences—rather than on Sections 326 to 333 of the Criminal Code—if the relevant acts were committed online. In a notable Supreme Court decision, Matichon Public Company Limited was found criminally liable under Sections 15 and 14(1) for publishing false information on its news website “Matichon Online.” The Court decided in the final instance that the story falsely named the plaintiff as the core leader of a rubber farmers’ protest in Nakhon Sri Thammarat. [47]

In several cases, charges under Section 14(1) were brought in combination with charges under Section 112 (royal defamation) or 116 (sedition) of the Criminal Code. [48] At the same time, the less frequent cases under Section 14(2) dealt with the online distribution of false information likely to impact public interests—namely having a negative effect on national security or causing panic to the public. [49]

In 2017, after ten years of enforcement, the Act on Computer-Related Offences was changed by the National Legislative Assembly, which had been in charge of legislation since the military coup of May 2014. Additional criminal provisions were inserted among the computer-related crimes (Sections 5 to 13) as well as the content-related crimes (Sections 14 to 16/2). The competences of officials (Sections 18 to 31) were expanded, in particular regarding powers to block access or remove information from computer systems (Section 20). Section 14 received a second paragraph, while numbers 1 and 2 of the first paragraph were also subject to important modifications: [50]

Section 14. [Changes marked]

Any person who perpetrates the following offenses shall be subject to imprisonment not exceeding five years or a fine not exceeding one hundred thousand baht, or both:

(1) with ill or fraudulent intent, put into a computer system distorted or forged computer data, partially or entirely, or false computer data, in a manner that is likely to cause damage to another person or the public, in which the perpetration is not a defamation offense under the Criminal Code ;

(2) put into a computer system false computer data in a manner that is likely to damage the maintenance of national security , public safety, national economic security or public infrastructure serving national public interest, or cause panic in the public;

(3) – (5) . . . [unchanged]

If the offense under Paragraph One (1) has not been perpetrated against the public but against a particular individual, the perpetrator or a person who distributes or transfers the computer data shall be subject to imprisonment not exceeding three years and a fine not exceeding sixty thousand baht, or both, and it is a compoundable offense.

The National Legislative Assembly considered the changes in Section 14 necessary in order to render the provision better suited to respond to current threats of advanced technologies, and to enhance its clarity. As regards Section 14 para. 1(1), only actions committed with ill or fraudulent intent should be criminalised. In addition, the drafters explicitly excluded defamation from the Section’s scope, as they considered that the Act’s purpose was to protect public interests rather than personal reputation. Accordingly, Section 14 para. 2 was added to render Section 14 para. 1(1) a compoundable offence for cases where public interests would not be infringed. [51]

The amendment stopped defamation charges from being brought or pursued under the Act. The Supreme Court, for instance, dismissed charges brought by King Power Suvarnabhumi Company Limited against administrators of the website Manager Online, based on Section 14 para. 1(1) of the Act on Computer-Related Offences, while confirming a concurrent conviction based on Section 328 of the Criminal Code. The case dealt with defamatory statements made at a rally of the People’s Alliance for Democracy that were reproduced on the website. [52] Another Supreme Court decision equally rejected defamation charges due to the subsequent change of Section 14 para. 1(1). [53] A further case against Manager Online website administrators ended with an acquittal by the Supreme Court, as the special intent required by the new Section 14 para. 1(1) could not be proven. [54] In yet another case, however, the Supreme Court dismissed charges based only on concurrences, though the facts of the case would also have allowed a rejection with reference to the change in law. [55]

Overall, the number of cases based on Section 14 para. 1(1) apparently dropped, rendering Section 14 para. 1(2) the more frequently used provision. False information that would likely affect national security or cause public panic were the most important legal grounds in a practical sense, despite the introduction of additional protected interests. [56]

At the same time, the global fake news discourse was reaching Thailand. Government and military officials began to frequently refer to the threat from online falsehoods, speaking of “cyber and hybrid warfare,” hidden enemies, and political parties trying to educate young people with falsities. [57] Fake news was described as being “embedded within every aspect of our society.” [58] Ministerial spokespersons announced legal action against fake news spreaders. [59] Besides the Act on Computer-Related Offences, charges could also be brought based on two post-coup announcements (ประกาศ; prakat) from the National Council for Peace and Order (NCPO), which prohibited the publication of false information by the media and any unreasonable criticism of the NCPO based on false information. [60] The orders were revoked shortly before the elected government assumed office in July 2019.

In November 2019, the government introduced a new mechanism in the fight against fake news when it inaugurated its “Anti-Fake News Center.” [61] The Center operates a website as well as Facebook and Line accounts where allegedly false news is corrected, and allegedly true news is confirmed. In addition, it has the power to refer cases to the police for further investigation. At the launch of the Center, the responsible Minister referred to fake news as “one of the critical threats that could harmfully affect people's lives and the economy.” [62]

Another temporary anti-falsehood law was enacted in late March 2020, two months into the global COVID-19 pandemic. After the Thai government had declared an emergency, a related stipulation (ข้อกำหนด; khokamnot) prohibited anyone, under threat of criminal punishment, from spreading false information about the COVID-19 situation in Thailand in a manner creating public fear, and from distorting information so that it could create misunderstanding leading to disturbances in public order or good morals. [63] Due to the merely temporary character of this stipulation, however, we maintain our focus on the permanent anti-falsehood provisions of the Act on Computer-Related Offences, and proceed to their critical evaluation.

IV. INTERPRETATIONAL PITFALLS

Despite the amendments of 2017, the interpretation of Section 14 para. 1(1) and (2) has remained doubtful in several respects. The uncertainties laid out in this part raise fundamental questions about the purposes pursued by the Act. Two main contentious issues can be distinguished: The meaning of “false” (1.) and the scope of the protected interests (2.).

A. The Meaning of “False”

Up until the present, drafters, scholars and courts apparently have not agreed on the question of what is “false” in the sense of Section 14 para. 1(1). Early drafts of Section 14 covered only the act of introducing forged (ปลอม; plom) computer data [64] into a computer system. The provision was meant to close a gap in the law: provisions dealing with the forgery of documents (Sections 264 to 269 of the Criminal Code) were only applicable to paper documents. [65] Thus, the new law was necessary to protect the trustworthiness [66] of computer data. During the legislative process, however, the Council of State proposed the inclusion of additional paragraphs covering false computer data (ข้อมูลคอมพิวเตอร์อันเป็นเท็จ; khomun khomphiotoe an pen thet), computer data that constitute an offence against the security of the Kingdom, as well as obscene computer data. These proposals were later included in Section 14 as Nos. 2 to 4. [67] At that point in the drafting process, therefore, No. 1 dealt with forged computer data and No. 2 with false computer data. Had this version been enacted, the apparent difference between the two provisions would have constituted the following: Forged computer data, analogous to forged documents, would have been understood as data that has been changed by a person who pretends to be the original data creator; false computer data, by contrast, would have been untruthful data as regards content—in other words, false statements of fact.

Significant interpretational uncertainty ensued, however, when the National Legislative Assembly (NLA) decided to include false, besides forged, computer data under No.1, as well as the element of likely damage to another person. [68] In addition, the NLA further decreased the provision’s resemblance to Section 264 of the Criminal Code by deleting the phrase “to cause another person to believe . . . .” (เพื่อให้บุคคลอื่นเชื่อว่า; phuea hai bukkhon uen chuea wa). [69] No. 1 was thus not meant to be a mere gap-filler for data forgery but to serve the additional [70] purpose of protecting the truth. [71] With one pen stroke, No. 1 now also addressed false statements of fact—an area already covered by No. 2. [72]

The fact that police, prosecutors and courts mainly relied on Section 14(1) of the Act’s 2007 version to bring charges of defamation [73] demonstrates that the provision was widely interpreted as protecting the truthfulness of computer data. Though defamation was subsequently explicitly excluded from the Section’s scope, the NLA in 2017 confirmed its intention to protect the content-accuracy of information, not only through Section 14 para. 1(2), but also through the addition into No. 1 of another alternative: the introduction of distorted (บิดเบือน; bitbuean) computer data. Distortion of computer data apparently occurs where the perpetrator “does not tell 100% of the truth” or “does not tell the truth that should be told.” [74] As a result, Section 14 para. 1(1) maintained its twofold character as a provision that protects both the trustworthiness and the truthfulness of computer data. Earlier scholarly interpretations that focused exclusively on trustworthiness have therefore been superseded. [75] A recent Supreme Court decision apparently also considers that the Section covers computer data that is false as regards content. [76] This overlap between No. 1 and No. 2 is not only systematically unfortunate, but raises important questions as to the protected interests of these provisions. Below, we will further discuss this issue.

As far as Section 14 para. 1(1) and (2) protect the truthfulness of computer data, the provisions address statements of fact that can be proved true or false. They are not applicable to opinions reflecting personal standpoints. This is in line with the tort of defamation (Section 423 of the Civil and Commercial Code), which also distinguishes between facts and opinions. [77] It differs, however, from criminal defamation, which covers both statements of fact and opinions. This can be seen most explicitly in Section 329 of the Criminal Code, which provides a justifying ground for opinions (ความคิดเห็น; khwam khit hen) and statements (ข้อความ; kho khwam) if made in good faith. [78] Another distinguishing feature is the fact that a person can be found criminally liable of defamation even if the relevant statement of fact was true, [79] which again distinguishes criminal from civil defamation. [80] It also distinguishes criminal defamation from the Act on Computer-Related Offences: Sections 14 para. 1(1) and (2) demand of the plaintiff to prove that the statements in question were false. The general rules of criminal procedure regarding the burden of proof [81] apply.

For a statement to be subjectively false, the perpetrator must have acted with intention regarding all elements of the offence (Section 59 of the Criminal Code), including the falsity of the facts in question. [82] Besides the question whether the statement was objectively and subjectively false, it also must have had the likely effect of impairing one of the protected interests. This raises further interpretational uncertainties—which we will deal with in the following part.

B. The Scope of Protected Interests

The preceding part revealed that Section 14 para. 1(1) of the Act on Computer-Related Offences protects both the trustworthiness and the truthfulness of computer data, while No. 2 focuses exclusively on truthfulness. However, the provisions do not protect trustworthiness and truthfulness for their own sake, but stipulate that the action in question must be likely to cause damage to the public (No. 1) or to damage the maintenance of national security, public safety, national economic security, or public infrastructure serving national public interest, or to cause panic to the public (No. 2). [83] Thus, both provisions aim to protect public interests. Indeed, the Act’s 2017 amendment effectively removed individual interests from No. 1 with the deletion of “likely damage to another person,” and the clarification that the offence must not be defamation. No. 2 had from the outset protected only public interests. However, Section 14 para. 2, also introduced in 2017, at least partly conserves the protection of individual interests by extending the scope of Paragraph 1(1) to cases where the offence “has not been perpetrated against the public but against a particular individual.” Such offences are compoundable, and the punishment must be reduced. [84]

A problematic issue already alluded to concerns the relationship between the interests protected in No. 1 and No. 2. Damage to “the public” (ประชาชน; prachachon) in No. 1 and damage to the more specific public interests in No. 2 resemble the relationship between a lex generalis and a lex specialis. In other words, if national security, public safety, national economic security or public infrastructure serving national public interest are likely to be affected, or if public panic is likely to be caused, one would assume that this amounts to a likely damage to the public. The objective elements of No. 1 would be fulfilled as well. Since 2017, however, No. 1 requires a specific (ill or fraudulent) intent which is not required by No. 2. These systematics seem to indicate that the likely damage to the interests protected by No. 2 was considered particularly severe by the drafters. As far as false computer data is concerned, No. 1 could therefore be classified as a “fall-back” provision for less severe cases which, in compensation, requires an elevated subjective threshold. With regard to “misleading” computer data, however, No. 1 is the only applicable provision as No. 2 only covers false computer data. Thus, again, the systematics of the provision seem not entirely consistent.

Finally, significant practical problems are caused by the statutory requirement to prove that damage or panic was “likely” (โดยประการที่น่าจะเกิด . . . .; doi prakan thi na cha koet . . . .). Judges need to determine in hindsight whether and how third persons were to react to the news. The anticipated reaction of the audience becomes, retrospectively, the threshold for criminal liability. This reflects the character of Section 14 para. 1(1) and (2) as a discursive crime that needs an audience, just like defamation. [85] Although, would a judge need to have ex-post facto “prophetic powers” to decide the case? [86]

Some examples may help to illustrate this issue. In one instance, the police and the prosecution considered that the message of a social media user, according to which the Prime Minister and his wife had transferred a substantial amount of money to Singapore, was likely to damage national security. Both the first instance and the appeals court dismissed the case, however. [87] In another case, the authors of a news article that referred to a Reuters report claiming that Thai naval forces earned money from omitting to perform their duties regarding the trafficking of Rohingya people were brought to criminal trial. The court, however, considered that damage to national security was not likely. It also rejected the likelihood of a public panic. [88] A further example could be the case of a person who falsely claimed on social media that a provincial authority was to close a dam, risking floods. An additional message called on the public to attend a mock cremation ceremony for the provincial governor. The court found that the posts were likely to cause public panic and pronounced a 4-year prison sentence, suspended for two years, and a fine of 110,000 baht. [89] Public panic-related charges were also brought in relation to claims that the Deputy Prime Minister had bought expensive coffee cups; in response to allegations that the Prime Minister fled rebellion charges; and in reaction to criticism of the Thai Raksa Chart Party dissolution. [90]

None of these cases seem to indicate a concrete threat to national security or a risk of immediate outbreak of public panic. This provokes the question of which level of probability needs to be applied. Scholarly commentary argues that the perspective of an average person (วิญญูชนคนธรรมดาทั่วไป; winyuchon khon thammada thua pai) should be adopted. [91] This would be in congruence with the standard applied to the tort of defamation. [92] As a consequence, proof of an “abstract panic” [93] rather than concrete evidence would be required. Section 14 para. 1(1) and (2) therefore provide significant interpretational leeway to authorities and, in the final instance, courts, which are neither able nor willing to conduct opinion polls to test the public’s reaction to certain messages. At the same time, it goes without saying that individual assumptions and preferences are not suitable guidance for judicial decision-making.

An approach to limiting the interpretational uncertainty could be to require proof of an “ill or fraudulent intent” not only concerning No. 1, but also No. 2. A stronger focus on the subjective elements of the crime would indeed approximate online falsehoods to defamation, where Thai scholars have considered that intention forms the core of the crime. [94] However, it also appears undesirable to render the provision a crime of conscience, where bad intentions turn the balance. A more precise objective threshold would be preferable. It appears that this could only be achieved by requiring proof of actual—rather than likely—threats to the protected interests. Such a solution might not only suffice from the perspective of penal policy, but also better safeguard the constitutional right of freedom of expression—to which we turn now.

V. CONSTITUTIONAL REQUIREMENTS

This part discusses the criminalisation of online falsehoods from the perspective of constitutional law. The contentious issues are related to the statutory restriction of free speech, and they include: The general constitutional protection of falsehoods (1.) and the proportionality of criminal punishment (2.).

A. The General Constitutional Protection of Falsehoods

Thai constitutional law has protected individual rights to differing extents over the course of the past ninety years. While the temporary Constitution Law of June 1932 did not contain any rules on rights or liberties, the permanent Constitution of December 1932 broadly stipulated in Section 14 the protection of several individual rights, including the rights to speak, write and publish. [95] Beginning with the Constitution of 1949, a pattern of constitutional provisions emerged that distinguished the scope of a right from enumerated grounds for restrictions. [96] The pattern reflects the general constitutional doctrine related to individual rights that defines a right’s general scope ( Schutzbereich) that might then, in a second step, be subject to restrictions for constitutionally defined purposes.

While Thai scholarly literature has devoted most of its attention to developing and clarifying constitutional standards for restrictions, [97] the scopes of rights have often remained undefined. Doctrinally speaking, however, this question might be considered upstream to the discussion on restrictions; if a right is not applicable to a given case in the first place, the issue of restrictions becomes a moot point. Thus, the question arises as to whether freedom of expression in Thai constitutional law generally also protects falsehoods. In a word: Are false statements of fact included in its scope? If that is not the case, Section 14 para. 1(1) and (2) of the Act on Computer-Related Offences would not amount to a restriction of that right. The constitutional mechanism that requires legislators to conform restrictions to the constitutionally permitted purposes would not be triggered. [98]

The question of whether falsehoods are protected speech has been discussed in foreign legal systems, including Germany and the United States of America. Based on its wording, Article 5(1), first sentence [99] of the German Basic Law of 1949 has been interpreted by the German Constitutional Court to protect primarily the expression of opinions. Consequently, statements of fact are protected only to the extent that they are a prerequisite for the formation of opinions. Thus, the protection of freedom of expression ends where the factual assertions can contribute nothing to this constitutional prerequisite of formation of opinion. Seen from this perspective, incorrect information is not an interest that merits protection. The Federal Constitutional Court starts with the supposition that a factual assertion which the asserting party knows is untrue, or which has been proven untrue, is not encompassed by the protection afforded by Art. 5(1), first sentence. As the distinction between value judgments and factual assertions can be difficult in a given case, [100] the Court would usually extend the protection to the entire statement. However:

the correctness of the factual portions can then play a role in the context of the balancing. If the expression of opinion contains factual assertions which the asserting party knows are untrue, or which have been proven untrue, then the basic right of freedom of opinion routinely will yield to the legal interest protected by the statute that limits the basic right. [101]

German constitutional law thus affords false statements of fact only limited constitutional protection, and may exclude them entirely from Article 5’s scope if they do not at all contribute to the formation of opinions. In contrast, the First Amendment of the Constitution of the United States of America does not provide a textual basis for distinguishing between opinions and factual statements. [102] Moreover, the U.S. Supreme Court has held that the liberal conception of free speech in U.S. constitutional law also generally prohibits the government from criminalising falsehoods:

The remedy for speech that is false is speech that is true. This is the ordinary course in a free society. The response to the unreasoned is the rational; to the uninformed, the enlightened; to the straight-out lie, the simple truth. . . . And suppression of speech by the government can make exposure of falsity more difficult, not less so. Society has the right and civic duty to engage in open, dynamic, rational discourse. These ends are not well served when the government seeks to orchestrate public discussion through content-based mandates. [103]

The Thai Constitutional Court has so far not addressed this question. Constitutional scholarship, however, has touched upon the issue. Notably, Khonthai Kuanhin has argued for the inclusion of statements of facts (การแสดงข้อเท็จจริง; kan sadaeng kho thet ching) into the scope of freedom of expression. While acknowledging that the Thai term for freedom of expression (การแสดงความคิดเห็น; kan sadaeng khwam khit hen) literally refers to the expression of opinions, he adopts a functional approach and plausibly considers that facts contribute to the formation of public opinion in an important manner. [104] Occasionally, Thai legal scholarship has also referred to Article 19(2) of the International Covenant on Civil and Political Rights (1966) [105] which stipulates that the right to freedom of expression includes the freedom to seek, receive and impart “information and ideas of all kinds.” [106]

The wording of the current Thai Constitution of 2017 [107] may not directly allow a definitive conclusion. Section 34(1), first sentence, provides that “a person shall enjoy the liberty to express opinions, make speeches, write, print, publicise and express by other means.” [108] Even if the provision’s first element, the liberty to express opinions, does not include the right to impart factual statements, the remaining elements all appear to be of equal rank rather than constituting sub-categories of the first. The provision thus apparently protects a more general freedom of “expression” (การสื่อความหมาย; kan sue khwam mai) in contrast to a narrow conception of freedom of “opinion.”

This relationship between the Section’s elements stands in an interesting contrast to the wording of Section 39 of the Thai Constitution of 1991 in the version of its fifth amendment of 1995, which introduced the term “liberty to express opinions” (การแสดงความคิดเห็น; kan sadaeng khwam khit hen) for the first time into the text of a Thai Constitution. [109] This provision differed from later wordings in the important respect that the liberty to express opinions was apparently meant to be exercised “by” (โดย; doi) making speeches, writing etc., [110] thereby indeed relegating these latter elements to sub-categories of the freedom to express opinions. However, the Constitutions of 1997, 2007 and 2017 removed the “by.” Thus, even considering that in these Constitutions the “liberty to express opinions” itself is confined to imparting opinions, the comparison with the text of 1995 appears to confirm that the remaining elements do not seem to be limited in that respect.

If factual statements are thus generally covered by Section 34(1) of the 2017 Constitution, this logically applies to both true and false statements, in the absence of any evidence to the contrary. Consequently, Section 14 para. 1(1) and (2) of the Act on Computer-Related Offences constitutes a restriction of freedom of expression which needs to satisfy the constitutional requirements laid down in Section 34(1) and 26 of the 2017 Constitution, including proportionality—to which we now turn.

B. The Proportionality of Criminal Punishment

Besides the requirement that restrictions of rights must pursue the constitutionally permitted purposes, [111] every limitation must also be proportionate. This is explicitly stipulated in Section 26(1) of the Thai Constitution of 2017 and is considered a fundamental principle of Thai public law. [112] As this article cannot expound the genesis and scope of the principle of proportionality in Thai scholarship and jurisprudence, [113] it focuses on recent jurisprudence of the Constitutional Court of Thailand with respect to the proportionality of criminal punishment. [114]

In recent years, the Court has referred to the principle of proportionality, including the three-step test of suitability, necessity, and proportionality in the narrow sense, in several of its decisions. [115] In decision 30/2563, however, the Court for the first time invalidated a criminal law provision (also) due to the disproportionality of the prescribed punishment. The case dealt with two announcements by the National Council for Peace and Order (NCPO) which were issued in 2014 after the military coup. The announcements functioned as the legal basis for ordering persons to report to the NCPO, and stipulated a criminal penalty of up to two years' imprisonment and/or a fine of up to 40,000 baht in case of non-compliance. In late 2020, the Constitutional Court declared these announcements unconstitutional. Besides finding a retroactive character, the Court judged them disproportionate based on a comparison with Section 368(1) of the Criminal Code. This Section criminalises the refusal to comply with an official order, and stipulates a penalty of up to ten days' imprisonment and/or a fine of up to 5,000 baht. This stark discrepancy regarding the prescribed punishment led the Court to find the NCPO announcements disproportionate according to Section 26 of the 2017 Constitution. [116]

The Constitutional Court’s reasoning provokes the comparison of Section 14 para. 1(2) of the Act on Computer-Related Offences with Section 384 of the Criminal Code, which was mentioned earlier in this article. Section 384 criminalises the malicious dissemination of false information, thereby causing the people to panic. It stipulates a punishment of up to one month’s imprisonment and/or a fine of up to 10,000 baht. Section 14 para. 1(2) of the Act on Computer-Related Offences criminalises, among the other alternatives, the—not necessarily malicious—entry of false computer data into a computer system that is likely to cause the public to panic. The stipulated punishment is imprisonment for up to five years and/or a fine of up to 100,000 baht. Thus, despite the fact that Section 384 demands a specific intent in the form of malice, and the additional fact that Section 384 demands an actual, not only likely instance of panic, Section 14 para. 1 prescribes a punishment which is more than 60 times higher regarding imprisonment and ten times higher regarding the fine. With a view to the Constitutional Court’s decision 30/2563, it appears doubtful whether this discrepancy can be reconciled with the principle of proportionality as applied by the Court.

VI. CONCLUSION

Pre-empting the global trend to criminalise online falsehoods, Thailand adopted the Act on Computer-Related Offences before the dawn of the fake news era. In recent years, the political and legal awareness of the topic has sharply risen. The relevant provisions of the Act, however, reveal several interpretational and constitutional complexities that deserve critical evaluation. In particular, this article raised questions regarding the systematic structure of Section 14 para. 1(1) and (2) and the meaning of “false” computer data. More importantly, it was shown that the statutory requirement of damage or panic that is “likely” to occur may result in uneven and disparate law enforcement. Proof of actual damage was therefore suggested as an alternative that would also limit the impact on the constitutional right to freedom of expression. This right fully applies, as statements of fact are within its scope of protection. Finally, the prescribed criminal punishment seems hardly in accordance with recent jurisprudence of the Thai Constitutional Court.

The case of Thailand thus reflects the challenges in finding the right response to the apparent fake news threat. Developing a constitutionally sound balance between regulation and laissez-faire requires not only an understanding of social media’s technological, aggregational and psychological aspects, but also a consideration of chilling effects on the formation of public opinion. Speech without any measurable adverse effects on protected interests might not deserve to be criminalised. At the same time, society’s resilience against the manipulatory potential of fake news needs to be enhanced and reinforced. Education curricula will need to focus on media literacy and critical reflection skills to an even higher extent than is already the case. This continues to be the most promising approach to prevent fake news from falling on fertile ground, while limiting the collateral damage on constitutional liberties.

[Date of submission: 14 February 2021; Revision: 10 May 2021; Acceptance: 20 October 2021]