sign in sign up
<<
Home , Fake news website, Sedition, Seven News

FIGHTING FAKE

A STUDY OF ONLINE REGULATION IN THE ASIA PACIFIC

Prepared and Written by ANDREA CARSON ASSOCIATE PROFESSOR () DEPARTMENT OF POLITICS, MEDIA & PHILOSOPHY, LA TROBE UNIVERSITY with LIAM FALLON

Prepared for , A FACEBOOK RESEARCH PROJECT latrobe.edu.au JANUARY 2021

Fighting : A Study of Online Misinformation Regulation in the Asia Pacific A First Published 2021 By La Trobe University Crn of Plenty Rd and Kingsbury Drive, Bundoora, , Australia 3086

© 2021 selection and editorial matter, Andrea Carson

This work was funded by Facebook.

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers.

Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

B La Trobe University Table of contents

Executive summary...... iv Key findings and conclusions...... v Background...... vi Thematic findings...... vi Theme 1: What is online misinformation – and why is it a problem?...... vi Theme 2: Tackling misinformation online and its challenges...... vii Theme 3: Potential remedies...... ix Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific. x Overview...... xi Aim and report objectives ...... xi Background and Australian context...... xii Method, scope and terminology...... xiii Chapter contents...... xiii About the Authors...... xvi Background chapters...... xvi Chapter 1 ’s approach to online misinformation...... 1 Introduction...... 2 Background and context...... 3 The Government’s approach to tackling misinformation...... 4 Laws and regulations...... 5 Government bodies...... 11 Timeline of Indonesia’s “fake news” laws and policies...... 12 Conclusion...... 13 Chapter 2 Singapore’s approach to online misinformation...... 14 Introduction...... 15 Background and context...... 15 The Government’s approach to tackling misinformation...... 16 Laws and regulations...... 16 The use of POFMA ...... 19 Reactions to POFMA...... 21 Timeline of Singapore’s “fake news” laws and policies ...... 24 Conclusion...... 25 Chapter 3 The : Approaches to online misinformation and ...... 26 Introduction...... 27 Background and context...... 27 The EU Code of Practice...... 28 Action Plan against Disinformation...... 28 Assessments of the Code...... 29 Reactions to the EU approach ...... 32 Other EU measures to tackle false information online...... 33 Member States’ initiatives...... 34 Timeline of ’s “fake news” regulations and policies...... 35 Conclusion...... 36

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific i Chapter 4 Australia’s approach to online misinformation...... 37 Introduction...... 38 Background ACCC Digital Platforms Inquiry...... 38 ACMA position paper...... 39 DIGI draft code: Australian Code of Practice on Disinformation...... 41 Comparing ACMA and DIGI positions on a new code ...... 44 Conclusion...... 45 Key Findings: Themes 1-3...... 46 Chapter 5 What is online misinformation – and why is it a problem?...... 46 Introduction...... 47 Definitions: Misinformation, disinformation and malinformation...... 47 Identifying the scale and effects of online misinformation...... 54 Negative consequences of online misinformation...... 55 Country-specific problems...... 58 Conclusion...... 61 Chapter 6 Tackling misinformation online and its challenges...... 62 Introduction...... 63 Section 1 Measures to tackle misinformation...... 64 Digital platform measures to deal with misinformation...... 64 Government measures to deal with misinformation...... 68 Section 2 Ongoing challenges...... 71 Challenges for platforms...... 71 Challenges for governments...... 72 Conclusion...... 76 Chapter 7 Potential remedies...... 77 Introduction...... 78 Common definitions...... 78 Digital literacy...... 79 Transparency...... 80 Inclusive process...... 81 Working with civil society and academics...... 81 Public interest journalism...... 82 Collaboration between technology platforms...... 83 Conclusion...... 83 Chapter 8 Concluding analysis...... 84 Concluding analysis...... 85 Acknowledgements...... 87 Reference list...... 88 Appendices...... 99 Appendix A: Tables of laws and regulations, Indonesia...... 99 Appendix B: Tables of laws and regulations, Singapore...... 103 Appendix C: Tables of laws and regulations, European Commission...... 107 Appendix D: Respondent profile and interview schedule...... 116

ii La Trobe University Tables, Figures and Boxes

Tables Table 1.1: Timeline of various laws and regulations implemented in Indonesia...... 12 Table 2.1: Timeline of various laws and regulations implemented in Singapore...... 24 Table 3.1: Timeline of various laws and regulations implemented in Europe...... 35 Table 4.1: Expected dates for draft and implementation plan of Australia’s voluntary Disinformation Code ...... 44 Figures Box 2.1 Alex Tan Case Study...... 19 Box 5.1 Indonesia: combining political and health misinformation...... 54 Box 5.2 Ahok Case Study...... 56 Box 5.3 Singapore: “plastic rice”...... 57 Boxes Figure 5.1: Seven types of “information disorder” ranging from low to high risk of harm...... 48 Figure 5.2: ’s contested Tweets...... 52 Figure 5.3: Facebook provides context to news posts using the info button...... 53

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific iii Executive summary Key findings and conclusions

The global spread of online misinformation has New laws in Indonesia and Singapore, ostensibly the potential to erode foundational elements of introduced to fight online misinformation and modern civilisation across much of the developed disinformation, are also alleged to have been misused and developing world. Social cohesion, public health by governments to crack down on political and safety, political stability and democracy are all and suppress freedom of expression and the media, under threat by the rapid and sometimes malicious further undermining already compromised democratic dissemination of false information within and across structures and institutions in Southeast Asia. national borders.

Advanced liberal democracies face a difficult challenge Online misinformation is an extremely complex and trying to balance the sometimes competing aims of multi-layered problem that defies simple, one-size- pluralism, free expression and individual liberty against fits all solutions; effective mitigation can only be the need to protect citizens from online harms and to achieved through multi-pronged strategies involving uphold the integrity of the electoral system. collaboration and cooperation between governments, digital platforms and community-based organisations. Liberal democracies, including Australia, should be alert to the potential for any new anti-misinformation to date by governments and global laws and regulations to be misused by governments to technology companies to tackle online misinformation undermine and the media. and disinformation have been constrained by a multitude of challenges and policy shortcomings, not least being a lack of co-ordination and co-operation The European Union’s 2018 Code of Practice on between key players. Disinformation, despite some shortcomings and ongoing issues, has gained attention as a benchmark for self-regulation of online content, and is influencing Traditional media outlets can also contribute to the debate in Australia on the development of its proposed online misinformation problem through amplification of co-regulation framework. the existence of misinformation campaigns and should be included in strategies to mitigate it. Despite the uniquely difficult challenges faced in culturally-diverse jurisdictions like Indonesia, extensive Popular terms such as misinformation, disinformation, work has been undertaken at and higher fake news, false news and are used loosely levels to mitigate online misinformation, with some and interchangeably, leading to widespread confusion success. However, the need for new approaches and about their meanings, and adding to the already initiatives is also apparent. formidable challenge of trying to tackle the issue.

The global spread of online misinformation surrounding The extent of online misinformation is extremely COVID-19 has exposed the broader potential of difficult to gauge and to accurately measure, in part “information disorder’’ on the to jeopardise due to contested definitions about what is being public health and safety – particularly, but not measured. exclusively, in less developed countries.

Governments, global technology companies and other interested parties need to work together towards achieving a broadly agreed set of definitions of key terms surrounding online misinformation to enable coordinated and more effectively targeted strategies to be developed at local, national and global levels.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific v Background Trump. Of course there is nothing new the research aims to provide insights about fake news; it has been around for and points of comparison to assist The internet has revolutionised the way centuries. What has changed in recent Government, relevant we live and connect with each other, decades is how the internet has enabled authorities and digital platforms in bringing transformational economic and the mass transmission of falsehoods to their deliberations on managing the lifestyle benefits to much of the planet. any part of at a rapid pace. complex and fraught issue of online But at what cost? misinformation and its spread. The In the past five years, almost 20 findings contain both valuable insights Increasingly, the rewards of the internet countries have adopted new laws and and warnings for policy makers in age are being shadowed by unintended regulations in attempts to tackle online Australia and elsewhere. consequences – not least the insidious misinformation and disinformation. spread of false and misleading Two of those countries, Indonesia and Led by La Trobe University Associate information in the online sphere, with Singapore, are the focus of detailed Professor Andrea Carson, the research sometimes harmful consequences. This examination in this report. Australia, is based primarily on in-depth interviews global problem can threaten essential meanwhile, is pursuing a voluntary conducted in the second half of 2020 elements of our modern civilisation, co-regulation scheme involving the with online experts and practitioners including social harmony, health and major digital platforms, scheduled to living in Indonesia and Singapore. safety, the rule of law and democracy. commence in 2021. Project participants included academics, technology industry experts, media The focus of governments and law This independent research project, executives, online journalists, editors makers on the issue has been heightened funded by Facebook Inc (based in Menlo and community-based activists in both since 2016, when emerged Park), reflects the views of the authors countries. The report also incorporates about Russian operatives attempting to and should not be taken to reflect the other recent research into how the use online disinformation to interfere in views of Facebook. Through the detailed spread of online misinformation is the US presidential election campaign; examination of the experiences of being tackled in Europe and Australia. and, also when the term “fake news” was Australia’s near neighbours Singapore popularised by then candidate Donald and Indonesia, and the European Union,

Thematic findings

This study offers new insights summarised below and Our typology is not standardised. For example, First Draft, a global coalition organised around three broad themes identified out of the of newsrooms, universities, online in-depth interviews with experts in Singapore and Indonesia. platforms and civil society groups, definesmisinformation more narrowly: as verifiably false content that is spread The themes are: For the purposes of this report, we without the of causing harm. have adopted the following definitions y What is online misinformation – The First Draft typology, while embraced for online misinformation and and why is it a problem? (chapter 5) by many, has its problems. Social disinformation: y Tackling misinformation online and media and technology platforms argue its challenges (chapter 6) y Online misinformation is the that a person’s intention when posting y Potential remedies (chapter 7) spread of inaccurate or misleading information is too difficult to divine. content online. They prefer to identify misinformation y Online disinformation is the spread based on the veracity or accuracy Theme 1: of inaccurate or misleading content of the content, and disinformation What is online online with conscious intent to with reference both to the actors and mislead, deceive or otherwise behaviours of those who spread it, and to misinformation – cause harm. the content’s propensity to cause harm. and why is it a problem? In this way, this report characterises Other complications emerge. The profusion of false content surrounding Contested definitions online disinformation as a substantial subset of the broad, overarching problem COVID-19, including bogus remedies, The lack of universally agreed definitions of misinformation. By our definition, shows how misinformation, as defined of terms such as online misinformation, misinformation can be spread with or by First Draft, can be spread without disinformation and fake news presents without ill-intent, while disinformation intention to cause harm and yet it also significant obstacles to achieving necessarily involves ill-intent. We can engender real-word harm. Further, consensus on how to tackle the problem. have adopted these definitions – and some online content that starts out Even among experts who contributed in particular a broad interpretation of being spread with ill-intent can be to this project, significant diversity of misinformation – in part because it spread innocently by others and still opinion emerged over the meanings closely aligns with the usage of experts cause harm. of misinformation and disinformation. who participated in this project.

vi La Trobe University Another complication surrounds Consequences of misinformation one or more different measures, with the term fake news, which has the choice of response depending on evolved from a descriptor of factually The negative consequences of online the type and severity of the suspected incorrect news content to a phrase misinformation can be many and varied, breach. Low-level breaches may lead to increasingly weaponised by some – ranging from emotional and physical no more than the publication of terms most conspicuously populist-styled harm to individuals, to wider damage and conditions of use with reference political leaders such as Trump – to to communities through adverse health to a platform’s published community try to delegitimise factual content in outcomes (such as from anti-vaccination standards. At the other end of the scale, the mainstream . Loosely- campaigns) and undermining of flagrant and serious breaches can result defined terms such as fake news can democracy, particularly during election in users being banned from the platform. also result in the conflation of very campaigns. Experts who participated in Facebook employs a broad three- different and distinct problems – such this project also identified a number of pronged approach involving shutting as threats to electoral integrity, hate marginalised groups in Indonesia and down offending accounts; limiting speech and cyber threats – which, in Singapore they considered at high risk distribution of problematic content; and turn, can potentially lead to inappropriate from online misinformation campaigns, providing cues to online users about one-size-fits-all remedies. Conversely, including women (particularly in Indonesia), the quality of the information they are when governments and platforms people identifying as LGBTI, journalists, consuming, with a view to boosting the adopt separate, or “siloed” approaches human rights and anti-corruption activists, media literacy of users. Within these to misinformation and disinformation, and minority ethnic groups. They also broad categories, Facebook: problems can emerge when falsehoods cited wider social consequences of fall into both categories, as can happen online misinformation, including: y Provides tools and campaigns to on issues such as COVID-19. y Reduced public trust in news media improve digital literacy and politics y Commissions third-party fact Scale and measurement of y Low quality discourse in the online checking of suspect content the problem environment y Attaches false information warning labels to problematic content It is generally agreed that online y Reduced capacity of the general y At times, require proof of user identity misinformation is a greater problem in public to distinguish fact from fiction. to improve account transparency Indonesia than in Singapore. Indonesia’s y Takes down harmful content, or geographical, cultural, linguistic and Theme 2: technically limits its algorithmic spread religious diversity provides fertile y Reduces services to users found ground for political polarisation, and Tackling to have breached the platform’s considerable scope and motivation for misinformation online community standards spreading false information. Its varied y Provides resources and financial literacy levels in a mass population also and its challenges support to news media organisations make it a challenging environment for in some countries. online education against the proliferation Industry measures of misinformation. Multinational technology companies Facebook and other platforms have also: such as Facebook, and The extent of online misinformation y Used artificial intelligence (AI) to have implemented a range of measures, is extremely difficult to gauge and to detect and identify problematic at both local and global levels, to try accurately measure, in part due to online content to reduce harmful and “inauthentic” contested definitions about what is being y Forged partnerships with community- content on platforms. Given the essential measured. Moreover, misinformation based organisations that can detect differences between misinformation transcends geographical borders as well and forewarn platforms of emerging and disinformation, the platforms have as online and traditional platforms, and misinformation campaigns. This has tended to assign separate strategies, and is often shared on encrypted platforms been particularly useful in Indonesia, in some cases separate teams, to each. such as WhatsApp. This report finds where demographic diversity presents some cause for optimism that greater Disinformation is typically defined as unique challenges levels of misinformation are being the spread of false information with the y Exchanged information with each detected and acted on. But whether the intention of causing harm. However, other and with governments about gap has narrowed between the volumes given the inherent difficulty of discerning known online misinformation threats of what is being spread and what is being the intent of individuals who post false y Funded academic research (including acted on remains an open question. information, the platforms focus on other this project) to expand their knowledge criteria – in particular the online actors about online misinformation and Common forms of misinformation and their behaviour patterns – when possible remedies The report finds the most commonly trying to identify and deal with suspected y Provided mechanisms for the public detected forms of online misinformation disinformation. By contrast, platform to report online misinformation, in Indonesia and Singapore relate to strategies to identify and manage and to alert platforms to other politics, public health and misinformation (defined for the purposes content that deviates from their theories. Online hate speech is a major of this report simply as the spread of community standards. global problem, but it is outside the inaccurate content) tend to focus on the scope of this study as its spread is veracity or otherwise of the content itself, most often regarded as disinformation. rather than the behaviour of the poster. Observed motivations for spreading However, when problematic content falls misinformation ranged from political into both categories – typically at varying and financial, to mischief-making. times – platform responses may include

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific vii Government measures y Government pressure to remove as they consider new approaches to the content at odds with the platforms’ issue domestically. Not least among The governments of Singapore and own assessments these challenges is how to balance Indonesia have strongly focused on y Content on encrypted platforms, the need for social responsibility and legislative measures in response to which adds to the difficulty of harm minimisation against freedom the quantum of online misinformation detecting misinformation and of expression and other fundamental in the past decade, utilising existing alerting users to its spread. values and rights held dear in advanced laws and enacting new ones – with liberal democracies. sometimes controversial outcomes. A number of remedies have been It is a fast-moving policy landscape suggested to counter the sharing of false Challenges for governments with new laws passed even as this content between different platforms. One report was going to print. approach would be to follow the example Globalisation – and the massive flow of of the European Code of Practice on internet content over national boundaries The major legislative initiatives have Disinformation, which prescribes regular – places major practical limitations been the 2019 Protection from Online meetings of a range of stakeholders. on the effectiveness of national laws Falsehoods and Manipulation Act For Asia-Pacific countries, this would to fight misinformation, particularly (POFMA) in Singapore, and 2016 mean establishing and formalising closer when there is disagreement between revisions to the Information and cooperation between rival platforms, countries on what constitutes legitimate Electronic Transactions Law (ITE) in and forging ongoing relationships content. The international reach of Indonesia. Singapore has also utilised with community-based organisations online misinformation can render one existing broadcasting and to develop more systematic early country’s legal jurisdiction relatively laws, while Indonesia has invoked its alert systems for the emergence of impotent in the absence of cooperation and anti-pornography laws, misinformation campaigns. from other countries. And when and proposed controversial changes to disagreements emerge across national its Criminal Code, which critics argue Indonesia’s demographic diversity borders about content appropriateness, will give the Government too much and geographic spread present major diplomatic issues can arise not just power to suppress freedom of speech obstacles on various fronts. For example, for governments, but for the platforms and expression. In December 2020, detection measures involving the use tasked with resolving them. Indonesia passed its highly controversial of artificial intelligence struggle against Ministerial Regulation 5/2020, which the multitude of languages in use across The tendency of the Indonesian and amongst other things requires the Indonesian archipelago. And while Singapore governments to prioritise platforms to remove content “against media and digital literacy campaigns law enforcement ahead of less punitive the public order” within 4 hours, or face have some utility in countering the mechanisms to deal with online heavy penalties, including blocking of spread of misinformation, their success misinformation has also created major platforms. Indonesia’s Communications is constrained by disparate levels of controversy and public distrust towards and Informatics (KOMINFO) Ministry education in very large populations. governments – particularly in light of accusations that anti-misinformation has also been active in the fight against Dealing with government demands and alleged hoaxes and disinformation laws have been misused for expectations relating to contested online political purposes. with its own fact checking exposés. content is more difficult to address. The governments of both Indonesia and Platform experts who contributed to Political misuse Singapore have often clashed with civil this project argued that: Experts say the governments of society groups and major technology y Deadlines set by governments to Indonesia and Singapore have used platforms over how to define and assess and remove misinformation misinformation laws to censor or silence respond to contested online content. tended to be too tight a wide spectrum of critics. They say Platforms have at times refused requests y Governments were hopelessly this has had detrimental implications or demands from governments to conflicted when they purported for freedom of speech, media freedom respond to disputed content in particular to be both the arbiter of online and political pluralism, and democratic ways – notably when governments misinformation and the enforcer representation in both countries. The have sought the prompt removal of of laws against misconduct governments have been accused of content before the platform says it has y The governments of Indonesia using the laws to target their political had time to assess it adequately. Such and Singapore had taken a binary opponents, as well as journalists, disagreements have occurred most approach to defining and identifying religious groups, political dissidents, frequently during election campaigns. misinformation, putting them at odds human rights campaigners and In this environment, the newly passed with the platforms’ global policies other activists. Indonesian Ministerial Regulation 5/2020 and Western values of trying to will likely be a closely watched law by balance concerns about freedom of In Singapore, most of the controversy defenders of civil liberties. expression against the need for social surrounds the 2019 Protection from responsibility and harm mitigation. Online Falsehoods and Manipulation Challenges for platforms Act (POFMA), which bestows extensive Despite their apparent progress Critics of the legislative approaches of powers on the Government and its in tackling some forms of online Indonesia and Singapore, which include ministers to declare information to misinformation, technology platforms significant jail terms and large fines for be false or misleading, and to force face formidable ongoing challenges. convicted offenders, say the laws have publishers to apply a correction notice Participants in this project identified at been used to suppress free speech and or remove content. Some MPs have least four major persisting impediments intimidate government opponents into accused the ruling PAP of creating for the platforms: self-. and using POFMA to serve its own political ends and, in the words of one, y The spread and detection of false While different polities, the problems tighten a “dictatorial government’s… content between different platforms with legislative responses to online hold onto absolute power”. Some have y The geographic spread, and the misinformation in Asian countries also questioned how the Government ethnic, religious and linguistic highlight some potential challenges for could make fair judgements in cases diversity of Indonesia Australian legislators and policymakers

viii La Trobe University where it was the subject of alleged community of internet users. Further, for which there is no agreed definition falsehoods, while others have highlighted the experts identified that existing and varying degrees of associated the potential for POFMA to be used measures and programs implemented by harm. Facebook undertakes twice yearly to intimidate independent media and governments and technology platforms reporting of its global measures, and government critics. need to be supplemented on a number quarterly reporting of country-specific of important fronts. These include: measures that were geoblocked because The Singapore Government has also of local laws. This data is available on been accused of using pre-existing laws, Common definitions the platform’s . However, critics including the Defamation Act, to deter argue greater efforts are required to dissent under the guise of prosecuting Agreement among key stakeholders communicate to the public what is being online misinformation. According to the on definitions of key terms – most done to tackle online misinformation. international organisation Human Rights importantly misinformation and Watch, the Defamation Act has been disinformation – will be essential to ensuring that authorities dealing Supporting public interest the Government’s “most powerful non- journalism criminal weapon” against its opponents. with online falsehoods are acting in Its large penalties are claimed to have concert and addressing the same Among other factors, the rise of the served as a government tool against problems. To date this has not occurred. major technology platforms – in speaking ill of the Government, with self- Currently, the terms misinformation particular Google and Facebook – has censoring widely observed and practised, and disinformation can mean different impacted the business model and government-initiated defamation things within and beyond national of companies and led actions often succeeding. borders – including in the labelling of to newsroom cutbacks and closures, as national legal codes. Even some experts the ACCC reported.4 At the same time, In Indonesia, the Government’s use of the who participated in this research project the platforms have provided financial Information and Electronic Transactions used the terms interchangeably. support for quality journalism for many Law (ITE) to prosecute online reasons, including because of its power misinformation has been characterised Education to help internet users distinguish fact by critics as a threat to freedom of from falsehood. So ironically, while speech and media freedom. Hundreds Digital education programs have been journalism has been a casualty of of Indonesians have been prosecuted rolled out in Indonesia and Singapore. the mega-success of the technology under the ITE law over online comments They include “train the trainer” programs platforms, Google and Facebook critical of the Government and President for digital literacy educators, and mass have recognised the importance of it Jokowi in recent years. During Jokowi’s advertising campaigns against false with targeted initiatives such as the first term (2014-2019), 241 individuals news designed to educate internet Journalism Project (Facebook) and were “criminalised for criticising authority users to identify false information and the Google News Initiative, which figures of the Jokowi administration,” avoid sharing it. But experts argue more provides journalists with data skills according to Usman Hamid relying needs to be done – and emphasise that and online tools. Google’s News Lab on figures.1 education programs cannot succeed in also collaborates with journalists and Meanwhile, academic Ross Tapsell found the absence of efforts on other fronts to entrepreneurs to drive innovation in no evidence of anyone being prosecuted curb the spread of false and potentially news. Facebook is piloting Facebook by the state for spreading falsehoods harmful content. News in the USA to enable quality about Jokowi’s opponent, Prabowo, Engagement and collaboration news outlet stories to reach bigger during the 2019 presidential election with all sectors audiences. Elsewhere, Twitter donates campaign.2 The Government also used to specific journalism projects such as the law against critics in West Papua, The project participants unanimously the Committee to Protect Journalists and the scene of a long-running struggle agreed that the involvement and the International Women’s Media for independence. In 2019, human cooperation of all interested parties – Foundation. Most major platforms also rights lawyer and West Papua defender governments, policymakers, technology provide journalism research funding to Veronica Koman was charged under the companies, academics, media groups, the academic community and support ITE law for allegedly spreading fake news journalists and grass roots community- for digital media start-ups. While these that caused unrest in the region. And based organisations – was an essential initiatives will not restore employment in August of that year, the Government precondition to effective action against of journalists to the levels of the pre- used Article 40 of the ITE law to shut a problem as complex and multi-layered Facebook and Google era, they are down internet access in West Papua.3 as online misinformation. an acknowledgement of the essential Greater transparency role of public interest journalism in Theme 3: functioning liberal democracies – and, Platforms have been criticised for by extension, in the fight against online Potential remedies allegedly inadequate reporting of the misinformation and disinformation. scale of the problem of misinformation, Experts agree that effective mitigation and for not being more transparent of online misinformation requires a about what they are doing to manage it. multi-pronged approach, with the active In their defence, platform experts point involvement of all interested groups – out the extreme difficulty of measuring the digital platforms, governments and a phenomenon that transcends national policymakers, academics, journalists, borders and often goes undetected, and community organisations and the wider

1 Usman Hamid, “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop,” The Conversation, November 25, 2019, https://theconversation.com/indonesias-information-law-has-threatened-free-speech-for-more-than-a-decade-this-must-stop-127446 2 Ross Tapsell, “Indonesia’s policing of news increasingly politicised,” ISEAS, no. 75 (2019), https://www.iseas.edu.sg/images/pdf/ISEAS_Perspective_2019_75.pdf 3 , “Rights group, journalists press Indonesian govt to lift internet curb in restive Papua,” Thomson Reuters Foundation, published August 23, 2019, https://news.trust.org//item/20190823055418-5iy2g/ 4 Australian Competition and Consumer Commission, Digital Platforms Inquiry: Final report (Canberra: ACCC, 2019), 34, https://www.accc.gov.au/system/files/Digital%20 platforms%20inquiry%20-%20final%20report.pdf

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific ix FIGHTING FAKE NEWS A Study of Online Misinformation Regulation in the Asia Pacific Overview

The internet is without question among the most Aim and report transformational human inventions of recent centuries. objectives In the space of just two decades, it has revolutionised the This independent research project, way we live and connect with each other, delivering previously funded by Facebook Inc. (based unimaginable economic and lifestyle benefits. But as with in Menlo Park, US), aims to inform many other monumental inventions, the internet has come contemporary deliberations in Australia about tackling online misinformation with some harmful unintended side-effects. And among the and disinformation, including the most concerning of these has been its enabling of the spread proposed development of a voluntary of false information, or “fake news”. code of technology industry practice in 2021. The research examines in depth how Singapore and Indonesia – There is nothing new about fake news; Elsewhere, external state actors are two of Australia’s closest Asia-Pacific it has been around for centuries. What increasingly using false online content neighbours – have tackled the problem, has changed is that near-universal to try to influence foreign election and compares and contrasts their access to online technology has enabled outcomes. Others are targeting responses to supra-national initiatives the spread of false information on an vulnerable groups and individuals with in the European Union. It is hoped the unprecedented scale. hate speech. Online misinformation is findings will help to guide the Australian also undermining global public health Government, its media regulator and Governments, technology platforms – most recently with the spread of digital technology companies as they and other stakeholders have become falsehoods and conspiracy theories consider various possible measures increasingly alert to the dangers of online surrounding COVID-19. Fake news can to try to contain the spread of false misinformation and disinformation over also make money, an example being the information on digital platforms in the past decade – particularly in the Macedonian youth entrepreneurs who the Australian context. To this end, time since Donald Trump popularised made millions in advertising revenue the report: the term fake news during his 2016 US by using made-up political headlines presidential election campaign. On many y Provides an overview of how as .1 fronts, the spread of online falsehoods governments and digital platforms appears to be undermining foundational Surveys in Australia find that most in Singapore, Indonesia and the elements of modern civilisation – most people are worried about fake news.2 European Commission have notably social harmony, public health Some scholars fear an epistemic crisis responded to the perceived threats and safety, and political democracy – with citizens unable to confidently of falsehoods on and in many parts of the world. separate fact from fiction.3 Other studies other digital platforms show that political leaders who seek y Provides the examples of Singapore In Indonesia and Singapore – two to delegitimise journalism as “fake and Indonesia to examine the dynamic Southeast Asian nations that news” could be undermining public potential implications of legislative are the primary focus of this report – trust in the mainstream news media, and other policy responses to online misinformation and disinformation have and leading people to rely on less misinformation for political discourse flooded the online space during recent trustworthy sources. and freedom of expression national elections, adding to serious y Documents detailed insights from key concerns about their degrees of liberal It is in this context that governments, actors (journalists, academics and governance. The concerns have been technology platforms and other interested technology platform experts) about compounded by accusations that the parties in all jurisdictions, including the major challenges arising from the national governments of both countries Australia, have been redoubling efforts to spread of false information online, and have used anti-misinformation laws tackle this serious and challenging public the consequences for public debate to target their political opponents and policy issue. y Consolidates the findings into a suppress free speech and freedom of timely publication that highlights the the media. public policy and political challenges and responses to the spread of false information on digital platforms in the selected countries y Identifies key considerations and challenges relevant to online platform governance in Australia.

1 https://www.wired.com/2017/02/veles-macedonia-fake-news/ 2 Caroline Fisher, “Australia,” in Reuters Institute Digital News Report 2019, eds. Nic Newman, Richard Fletcher, Antonis Kalogeropoulos and Rasmus Kleis Nielsen (Oxford: Reuters Institute for the Study of Journalism, 2019), 132-133, https://reutersinstitute.politics.ox.ac.uk/sites/default/files/inline-files/DNR_2019_FINAL.pdf; Michael Barthel, Amy Mitchell and Jesse Holcomb, “Many Americans believe fake news is sowing confusion,” Pew Research Centre, published December 15, 2016, https://www.journalism. org/2016/12/15/many-americans-believe-fake-news-is-sowing-confusion/ 3 Paul Dahlgren, “Media, knowledge and trust: The deepening epistemic crisis of democracy,” Journal of the European Institute for Communication and Culture 25, no. 1-2 (2018): 20-27, DOI: 10.1080/13183222.2018.1418819

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific xi Background and How governments have “monitoring efforts of digital platforms responded to implement credibility signalling”12, and Australian context for a voluntary digital platforms “code” In an overview of regulatory and other 13 In advanced liberal democracies like to counter disinformation. The report responses to online misinformation stated that the code should apply to Australia, the reach and spread of online across the globe, academics James networks presents a conundrum for complaints about disinformation that Meese and Edward Hurcombe identify involved “serious public detriment”14, legislators and policymakers. On one three distinct approaches: hand, the state must try to protect the and that it should be overseen by the public against harm from false and y Voluntary co-regulation initiatives “that Australian Media and Communications potentially dangerous information. On the do not involve mandated regulation or Authority (ACMA). 8 other hand, as a custodian of democracy, state oversight”. In these instances, ACMA’s subsequent position paper the state must seek to protect the rights government bodies encourage digital – Misinformation and news quality on of the people to freedom of expression platforms to work with stakeholders digital platforms in Australia: A position and speech. Governments therefore face to develop and implement a paper to guide code development – a difficult challenge trying to balance the broad set of aims to tackle online was released in 2020. It said digital sometimes competing aims of pluralism, misinformation, as is the case with the platforms to be covered by the code free expression and individual liberty EU Commission. Australia is currently should include online search engines, against the need to protect citizens following this model. social media platforms and other “digital y from online harms and, in the case Direct legislative measures, such content aggregation services with at of democracy, to uphold the integrity as those adopted in Singapore and least one million monthly active users of the electoral system. Indonesia. Meese and Hurcombe in Australia”.15 Individual news media say that generally governments The dilemma is highlighted by some outlets were excluded from consideration adopting this approach have tended for coverage by the code because they worrying global trends. Freedom House to circumvent consultation. Others and other global political and media were “not considered key distributors claim such laws have been misused 16 indices paint an alarming picture of of misinformation in Australia.” This for political purposes, and tend to was despite subsequent findings that how democracy and pluralism have operate in jurisdictions already known come “under assault”4 in the online age, traditional media outlets also contribute for low levels of media and political to the online misinformation problem with the world experiencing 14 years of 9 freedoms. The European liberal in Australia.17 decline in political rights and civil liberties democracies of France and 5 up to 2019. Similarly, media freedom have also opted for legislation, but with In response to the ACMA and ACCC has reportedly been in decline for more more checks and balances against initiatives, the platform providers, than a decade – under both illiberal and misuse than Indonesia and Singapore. represented by the Digital Industry Group 6 liberal governments, with some using Even so, the German Government has Inc. (DIGI), produced the Australian the spectre of fake news to restrict online been criticised for over-reach.10 code of practice on disinformation for news content. y Non-regulatory activities such as the public consultation in October 2020 But there is also cause for optimism. government funding of digital literacy (see chapter four for details). The use Global technology platforms are campaigns (Indonesia and the EU) and of the word “disinformation” in the title investing heavily in efforts to verify online factchecking operations (Indonesia).11 is noteworthy in light of the previous information, with measures ranging use of “misinformation” in the title of from sophisticated artificial intelligence Australian response ACMA’s position paper. As discussed elsewhere in this report, ambiguity over detection, to manual fact checking on a In 2017, the Australian Government the meanings of key terms including large scale across both developed and directed the Australian Competition misinformation and disinformation developing jurisdictions. Yet it remains and Consumer Commission (ACCC) has to some extent compounded the to be seen how far these measures go to oversee a Digital Platforms Inquiry. challenge of tackling the broad problem towards restoring public trust in public The final report of the inquiry in 2019 of online falsehoods. information and in established news contained detailed proposals to address organisations in an era blighted by the disinformation and misinformation on This research report is intended to spread of misinformation.7 online platforms operating in Australia complement the public consultation (see chapter four). The report called for process in the lead-up to the planned enactment of the voluntary code in 2021.

4 Sarah Repucci, “A Leaderless Struggle for Democracy: Democracy and Pluralism are under assault”, Freedom House, 2020, https://freedomhouse.org/report/freedom- world/2020/leaderless-struggle-democracy 5 Ibid. 6 Freedom House, Special Report 2018: Attacks on the Record - The State of Global Press Freedom, 2017-2018 (Freedom House, 2018), https://freedomhouse.org/report/ special-report/2018/attacks-record 7 Andrea Carson, Investigative Journalism, Democracy and the Digital Age (New York: Routledge, 2020), 232. 8 James Meese, J and Edward Hurcombe, Regulating Misinformation: Policy Brief (Melbourne, Analysis & Policy Observatory, RMIT University), https://apo.org.au/node/309357 p.3 9 Ric Neo “The Securitisation of Fake News in Singapore.” International Politics, October 4, 2019. https://doi.org/10.1057/s41311-019-00198-4. 10 James Meese, J and Edward Hurcombe, Regulating Misinformation: Policy Brief (Melbourne, Analysis & Policy Observatory, RMIT University), https://apo.org.au/node/309357 11 Ibid. 12 ACMA defines credibility signalling as “flagging trusted sources and providing users with access to information on publishers and other sources” (Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia: A position paper to guide code development (Canberra: ACMA, 2020), 29, https://www. acma.gov.au/sites/default/files/2020-06/Misinformation%20and%20news%20quality%20position%20paper.pdf 13 Australian Competition and Consumer Commission, Digital Platforms Inquiry: Final report (Canberra: ACCC, 2019), 34, https://www.accc.gov.au/system/files/Digital%20 platforms%20inquiry%20-%20final%20report.pdf 14 Australian Competition and Consumer Commission, Digital Platforms Inquiry, 34. 15 Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia: A position paper to guide code development, 3. 16 Ibid., 22. 17 Axel Bruns, Stephen Harrington, and Edward Hurcombe, “‘Corona? 5G? Or Both?’: The dynamics of COVID-19/5G conspiracy theories on Facebook,” Media International Australia 177, no. 1 (November 2020): 12–29, https://doi.org/10.1177/1329878X20946113

xii La Trobe University Method, scope and Scope and terminologies The narrower definition of misinformation can be problematic The original empirical findings of this terminology because in some instances, as the study detailed in chapter five to seven COVID-19 pandemic has shown, false This research project centres on two focus on the legislative responses to content spread without ill-intent can case countries in the Asia Pacific region online misinformation and disinformation also cause harm.19 And while the idea of – Indonesia and Singapore – that have in Indonesia and Singapore. The findings disinformation as false content spread adopted a legislative approach to tackling are intended to provide insights and a with ill-intent is conceptually straight online misinformation. The project also point of comparison for those involved forward, in practice it can be difficult to examines the alternative of voluntary in deliberations over future measures to establish or to prove the intentions of co-regulation between governments and tackle online misinformation in Australia. platforms adopted in European Union people posting content. For this reason, nations, on which Australian authorities A limitation that should already be technology platforms focus on other have heavily drawn in their deliberations apparent from this introduction is criteria such as the actors who are ahead of the proposed adoption of an the lack of broad consensus on the spreading false information and patterns Australian voluntary code. meanings and applications of terms of behaviour when trying to identify such as misinformation, disinformation, and mitigate disinformation, and on the Literature and policy review malinformation and fake news. Even veracity of the content when identifying experts who contributed to this other types of misinformation. The first step of the research process project were not completely aligned or involved a detailed review of previous In summary, when this report uses the consistent in their use of these terms. research examining countries that have word disinformation, it should be taken This lack of consensus on definitions is taken a legislative approach to online to mean spreading false or misleading acknowledged as a significant barrier misinformation, and others that have content with conscious intent to mislead, to developing effective policies and adopted voluntary co-regulation. The deceive or otherwise cause harm. When measures to tackle the problem. extensive literature review identified we use the broader term misinformation, key themes that helped to inform To try to mitigate confusion for readers or quote others using that word, it refers the in-depth interviews conducted of this report, we have adopted a to the spread of inaccurate or misleading with 14 experts in the online sectors simple and broad definition of online content online that may or may not cause in Indonesia and Singapore. The misinformation: the spread of inaccurate harm – depending on the context. or misleading content online. interviewees included journalists, editors, The report further acknowledges that media regulators, academics, digital We define disinformation, by contrast, both misinformation and disinformation platform experts, human rights activists as the spread of inaccurate or misleading are a part of what academic Claire and members of non-government content online with conscious intent Wardle and her colleagues call the organisations with practical knowledge to mislead, deceive or otherwise cause broad problem of online “information of how false information online – and harm. In this way, we consider online disorder”.20 We also acknowledge and policy responses to it – impact citizens’ disinformation to be a substantial subset adopt the British Government’s broad daily lives and public discourse. The of the broad, overarching problem of definition of the term fake news “as views of government representatives misinformation. This is a similar position shorthand for both disinformation and have been recorded in official documents to the ACMA. misinformation”.21 and cited in background chapters. It must be emphasised that this Interviews typology is not an accepted standard; Chapter contents others use slightly or even widely Plans for face-to-face meetings with differing definitions. We have adopted Chapters one and two examine how experts in Indonesia and Singapore this typology for the sake of clarity the spread of online misinformation and had to be abandoned in favour of online and consistency. It also reflects broad disinformation has been approached interviews after the COVID-19 pandemic usage of the term misinformation – to by governments in Indonesia and precluded international travel. Each mean all types of false or misleading Singapore, two of Southeast Asia’s most expert was interviewed for approximately online content – by a majority of expert prominent and dynamic nations. In one hour on the general nature of online participants in this project. each case we consider the unique local information disorder in the digital age, historical and cultural contexts in which Our broad definition of misinformation is and how it impacts on work and society, this most global of 21st century issues in contrast to some popular definitions and government responses to it (see is playing out. We examine the various which hold that misinformation is Appendix 4 for detail). Some interviewees laws (pre-existing and new), regulations confined to false and/or misleading participated on condition of anonymity and other tools they have deployed in content created without the intention of due to their concerns about possible the fight against misinformation – and causing harm.18 This provides a simple adverse repercussions. how these measures have at times and clear counterpoint to disinformation been as controversial as the problems when it is defined as false or misleading they purport to tackle. The report gives content spread with the intention of particular attention to claims that causing harm. However, as we show in governments in both countries have this study, there are practical problems used anti-misinformation measures with both these definitions. as weapons to silence their critics and suppress freedom of speech and the news media. By extension, we look at the

18 Claire Wardle, “The Need for Smarter Definitions and Practical, Timely Empirical Research on Information Disorder.” Digital Journalism, 6, no. 8: 953. 19 Ibid. 20 Claire Wardle, “The Need for Smarter Definitions and Practical, Timely Empirical Research on Information Disorder.” 21 DDCMS, The Cairncross Review: A Sustainable Future for Journalism, ( : Department for Digital, Culture, Media & Sport,12 February 2019), 103, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/779882/021919_DCMS_Cairncross_Review_.pdf

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific xiii vexed questions of who decides what is and what is not misinformation, and what it all might mean for the governance of Indonesia and Singapore. Chapter three provides a brief history and overview of the measures in place in the EU. It focuses on the development and performance of the EU’s main measure, the EU-wide Code of Practice on Disinformation, established in 2018. It also examines other important initiatives and actions taken in Europe to manage the spread of misinformation and disinformation, including laws passed by individual EU Member States. Chapter four looks in detail at Australia’s experience of online misinformation, and how governments have responded to it. We trace the recent history of debates and development of proposed co-regulation, and draw comparisons, where relevant, with measures undertaken in other countries. Chapters five, six and seven are the results chapters derived from the interviews. Each chapter reflects one of the three broad themes outlined in the Executive Summary. Together, they examine the challenges and responses to misinformation and disinformation in the Asia-Pacific region, drawing on expert local knowledge within Indonesia and Singapore. Chapter five draws on interviewees’ experiences and knowledge to explore the nature and scale of online misinformation in Indonesia and Singapore. Chapter six examines existing measures in Indonesia and Singapore being used to address online misinformation, including legislative responses as well as initiatives of digital platform providers and non- government and media organisations. It also examines accusations of political misuse of the laws by governments in both countries, and the implications for their democratic status. Chapter seven considers potential remedies to online misinformation from the experts’ perspectives. High on the list is the need for common definitions, digital education programs, involvement of trusted partners from the non- government sector, support for quality journalism and academic research, and greater reporting transparency from the platforms. Chapter eight brings together the earlier findings and salient lessons observed from different countries’ responses to online misinformation. It provides final observations about how the Australian Government, policy makers, regulators and the digital platforms and their users could work cooperatively to tackle this insidious problem.

xiv La Trobe University Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific xv About the Authors

Dr Andrea Carson

is a political scientist and an Associate Professor in journalism in the Department of Politics, Media and Philosophy at La Trobe University in Melbourne, Australia. She has authored numerous articles on political communication, election campaigns and digital media and fake news. Her research focus is on quality information in the public sphere (fake news, media trust), political representation (women) and political communications. Her 2020 book examines the role of investigative journalism in democracies: Investigative journalism, democracy and the digital age. She researches the media’s role in public policy agenda-setting using big data and was awarded Australian Research Council (2018-2021) funding as part of a team; and an ARC Linkage (2020-2024) studying female political representation in local government. She previously worked as a newspaper journalist and section editor at in Melbourne from 1997-2001 and as a radio broadcaster (RRR) and radio and television producer (Australian Broadcasting Corporation) from 2001‑2010.

Liam Fallon

is a researcher whose work at La Trobe University on research projects with Associate Professor Andrea Carson has covered political science, journalism, democracy and gender in politics. Liam is also a research assistant at Deakin University, where he has recently been working on a Victorian Government Department of Justice and Community Safety funded study on masculinity and violent extremism with Dr Joshua Roose. Liam’s research interests include polarisation, truth and citizenship in democratic societies, public service media and cultural studies. He graduated from the University of Melbourne with a Bachelor of Arts (Honours) in 2016. He has previously worked in community broadcasting and the NGO sector.

This has been an independent research project, funded by Facebook Inc (based in Menlo Park, US). It reflects the views of the authors and should not be taken to reflect the views of Facebook. Ethics approval to undertake the research interviews was granted through La Trobe University, Australia.

xvi La Trobe University CHAPTER 1 2 Singapore’sIndonesia’s approach to online misinformation Introduction

In this chapter and the following we examine how the purport to tackle. We give particular attention to claims that governments spread of online misinformation and disinformation has in both countries have used anti- been approached by governments in Indonesia and misinformation measures as weapons Singapore, two of Southeast Asia’s most prominent to silence their critics and suppress and dynamic nations. freedom of speech and the news media. By extension, we look at the vexed questions of who decides what is and what is not misinformation, and what it In each case we consider the unique and new), regulations and other all might mean for the future of political local historical and cultural contexts tools they have deployed in the fight rights and civil liberties in Indonesia in which this most global of twenty- against misinformation – and how and Singapore. first century issues is playing out. We these measures have at times been examine the various laws (pre-existing as controversial as the problems they

2 La Trobe University Background and context

Mass staged in Indonesia in September 2019 signalled History and freedom deep and widespread public discontent with proposed changes to Historical narratives also influence a law about the Corruption Eradication Commission; the changes contemporary Indonesian politics and its to Indonesia’s Criminal Code (KUHP) was one of several other approach to tackling online information. grievances protestors expressed at the rallies.1 Of particular concern The authoritarian, anti-democratic rule of are the KUHP Bill’s ostensible attempts to curb the spread of online President Suharto from the 1960s to the misinformation with provisions targeting “hoaxes,” known locally as late 1990s – founded on anti-communist ideology and the national philosophy of hoaks, Indonesia’s term for fake news. “Pancasila,”8 and marked by institutional corruption – continues to shape modern Indonesia. While direct presidential It also clamps down on defamation of of , including limits elections staged every five years since the Government, which has raised fears on free speech and the media, in their 2004 (indirect through parliament since of further limits on freedom of speech attempts to rule and hold together the the 1999 election) have been deemed and the media. While the protests highly diverse ethnic, religious and free and fair by external adjudicators, prompted President Joko Widodo to cultural groups that make up Indonesia’s fabricated claims of ballot fixing in the postpone a vote on the Bill, fears about 270 million people. Javanese account 2019 election9 were a reminder that its illiberal potential persist. for 40.1 per cent of the population, democracy in Indonesia remains fragile. followed by Sundanese (15.5 per cent), In this chapter, we track the history of Malay, Batak, Madurese and Betawi Feeding into this narrative of democratic Indonesia’s approach to tackling online (each around 3 per cent) and dozens vulnerability is evidence of declining misinformation over the past decade, more.2 Ethnic Chinese, though only civil freedoms. Indonesia scored just leading up to the proposed new Criminal 1.2 per cent of the population, have a 61/100 (down from 65/100) in the latest Code. We look at how existing laws and disproportionately large presence and Freedom House report, making it “partly regulations have been adapted and used role in Indonesia’s society and economy, free.”10 It was marked down on issues by the Government to take on internet which sometimes leads to tensions with including rights for minorities, corruption, abusers – and, according to critics, to majority groups.3 transparency, religious freedom suppress dissent and free speech. The (atheism is banned), judicial process, chapter also examines the unique cultural, The disparate ethnic and language state violence, and legal, political and ethnic and historical characteristics groups struggled collectively for social equality.11 of Indonesia that help to explain the independence from Dutch colonial evolution of its approach to the issue. A rule in the 20th century.4 In recent Most established media timeline of the implementation of “fake decades, however, in the absence of news” laws and policies is tabled at the a shared colonial enemy, ethnic and groups, though technically conclusion of the chapter (Table 1.1). regional differences have become more independent of the pronounced, keeping governments alert Government, have strong 5 Ethnic and religious to potential and actual divisions. partisan links to parties The existence of significant religious and politicians.12 diversity minorities also presents further Suspicion about the Government’s challenges. While a large majority of Independent publications, such as the motives for wanting to amend the Indonesians identify as Muslim, there are weekly magazine Tempo, are in the Criminal Code is informed partly by also significant numbers of Christians minority. And despite the proliferation Indonesia’s chequered history on (9 per cent), Hindus (1.7 per cent), and of technologies enabling citizen-led 6 freedom of speech. Successive regimes others. And within Islam, tensions exist journalism, much of this activity has have employed varying levels between the large Sunni majority and been co-opted into the systems of smaller, mainly Shi’a communities.7 established media conglomerates, ensuring continued concentration of media ownership.13

1 Eve Warburton, “Indonesia’s pro-democracy protests cut across deep political cleavages”, New Mandala, October 3, 2019, https://www.newmandala.org/indonesias-pro- democracy-protests/ 2 Central Intelligence Agency, “The world factbook: Indonesia,” accessed August 19, 2020, https://www.cia.gov/library/publications/the-world-factbook/geos/id.html 3 Sarah Turner and Pamela Allen, “Chinese Indonesians in a rapidly changing nation: Pressures of ethnicity and identity,” Asia Pacific Viewpoint 48, no. 1 (2007): 112-127. 4 Christine Drake, National Integration in Indonesia: Patterns and Policies (Honolulu: University of Hawaii Press, 1989), 45. 5 Jacques Bertrand, Nationalism and Ethnic Conflict in Indonesia (Cambridge: Cambridge University Press, 2003) doi:10.1017/CBO978051155934; Richard Chauvel, and Ikrar Nusa Bhakti, The Papua Conflict: Jakarta’s Perceptions and Policies, (Washington, D.C: East-West Center Washington, Policy Studies 5, 2004) 6 Oliver Wolter et al., “Indonesia: Religions,” Britannica, last modified November 11, 2020, https://www.britannica.com/place/Indonesia/Services 7 Chiara Formichi. “Violence, sectarianism, and the politics or religion: Articulations of anti-Shi’a discourses in Indonesia,” Indonesia, no.98, (2014): pp. 1-27. 8 Pancasila, literally meaning ‘five principles’, is the nation’s core philosophy. It entails ‘the belief in one God, just and civilized humanity, Indonesian unity, democracy under the wise guidance of representative consultations, and social justice for all the peoples of Indonesia’ (“Pancasila,”Britannica. accessed August 20, 2020, https://www.britannica. com/topic/Pancasila). 9 After Joko Widodo was declared the winner of the 2019 Presidential election, his opponent, Prabowo Subianto, stated to the media that he and his team would not accept the result on grounds of alleged vote-rigging by Widodo. 10 “Freedom in the world 2020: Indonesia,” Freedom House, accessed November 13, 2020, https://freedomhouse.org/country/indonesia/freedom-world/2020 11 Ibid. 12 Ross Tapsell, “Platform convergence in Indonesia: Challenges and opportunities for media freedom,” Convergence 21, no. 2 (2014): 7-8. 13 Ibid., 13-14.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 3 Indonesia is a major online player, transparency of internet restrictions, According to MAFINDO, a prominent accounting for the fourth and third judicial protection, criminalisation of community organisation dedicated to largest number of internet and Facebook online acts, state surveillance, and extra- fighting online misinformation,16 political users globally.14 But as online activity has legal intimidation and violence. content accounts for around 70 per soared in the past decade, Indonesia’s cent of online hoaxes17 – making efforts Official attempts by the Indonesian internet freedom ratings, as measured to combat it a contested and at times Government to tackle online hoaxes by Freedom House, have declined15 over complicated endeavour. have fuelled many of these concerns. issues including equality of access,

The Government’s approach to tackling misinformation

Multi-faceted strategy groups like MAFINDO to help identify Fighting religious and hoaxes, and sought cooperation from The Government has employed a variety social media companies to share their ethnic discrimination of tools, some of them direct and blunt, knowledge of hoaxes and ultimately Preserving religious and ethnic stability is 23 in its fight against alleged hoaxes. For have them removed. a core element of Indonesia’s approach example, it shut down internet access to tackling online hoaxes. This is after riots over protests against racial Conflicts of interest reflected in the criminalisation of discrimination in Papua that grew into information sharing that aims to incite broader calls for an independence Many critics have questioned the hostility based on ethnic or religious 18 referendum, and in Jakarta during post- Government’s motives and impartiality divisions,25 and blasphemy allegations.26 19 election riots in May 2019. It has also when it comes to deciding what is false Concerns about religious radicalism empowered the National Police to information. Given that much of the and political and ethnic tensions in people alleged to have disseminated content being scrutinised is political some instances have led the Indonesian misinformation, while encouraging in nature and often anti-government Government to pressure social media members of the public to report suspect in sentiment, the potential conflicts of platforms to remove content.27 People 20 online behaviour to authorities. These interest are substantial – and readily deemed to have contradicted “the beliefs efforts are aided by new technologies, apparent in the operations of the National or political ambitions’’ of the Muslim such as artificial intelligence, that can Cyber and Encryption (BSSN) and State majority online have also been targeted identify inauthentic behaviour such Intelligence Agencies, which report directly under Indonesian laws.28 In one case a as automated bot accounts on social to the President. While these bodies man who posted “God does not exist” 21 media. The Government has also may be necessary to tackle the problem, and other comments deemed offensive implemented public education programs it raises the spectre of government to Muslims on Facebook was charged 22 to increase media literacy, engaged manipulation to protect itself.24 under the Criminal Code29 and the

14 “Leading countries based on Facebook audience size as of July 2020,” Statista, published July 24, 2020, https://www.statista.com/statistics/268136/top-15-countries-based- on-number-of-facebook-use 15 Usman Hamid, “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop,” The Conversation, November 25, 2019, https://theconversation.com/indonesias-information-law-has-threatened-free-speech-for-more-than-a-decade-this-must-stop-127446; “Freedom on the net 2019: Indonesia,” Freedom House, accessed November 13, 2020, https://freedomhouse.org/country/indonesia/freedom-net/2019 16 “About,” MAFINDO, accessed August 10, 2020, https://www.mafindo.or.id/about 17 Santi Indra Astuti quoted in Cristina Tardaguila, “Fact-checkers are teaching Indonesian housewives how to stop misinformation,” Poynter, published December 12, 2019, https://www.poynter.org/ifcn/2019/fact-checkers-are-teaching-indonesian-women-how-to-stop-misinformation/ 18 Reuters, “Rights group, journalists press Indonesian govt to lift internet curb in restive Papua,” Thomson Reuters Foundation, published August 23, 2019, https://news.trust.org//item/20190823055418-5iy2g/ 19 Karina M. Tehusijarana, and Jessica Valentina, “Jakarta riot: Government temporarily limits access to social media, messaging apps,” The Jakarta Post, May, 22 2019, https://www.thejakartapost.com/life/2019/05/22/jakarta-riot-government-temporarily-limits-access-to-social-media-messaging-apps.html 20 Ross Tapsell, “Indonesia’s policing of hoax news increasingly politicised,” ISEAS, no. 75 (2019), https://www.iseas.edu.sg/images/pdf/ISEAS_Perspective_2019_75.pdf 21 Kate Lamb, “Muslim Cyber Army: a 'fake news' operation designed to derail Indonesia's leader,” , March 13, 2018, https://www.theguardian.com/world/2018/ mar/13/muslim-cyber-army-a-fake-news-operation-designed-to-bring-down-indonesias-leader 22 See ‘About Siberkreasi’, Siberkreasi, accessed November, 13, 2020, https://siberkreasi.id/tentang-siberkreasi/ 23 Safrin La Batu, “Nine social media, messaging apps agree to combat fake news,” The Jakarta Post, January 31, 2018, https://www.thejakartapost.com/news/2018/01/31/ nine-social-media-messaging-apps-agree-to-combat-fake-news.html 24 Tapsell, “Indonesia’s policing of hoax news increasingly politicised,” 3. 25 “Law No. 11 of 2008 on Information and Electronic Transactions [UU ITE] (Indonesia), 2008 [revised 2016]” HUKUM Online, accessed November 13, 2020, Article 28(2), https://www.hukumonline.com/pusatdata/detail/lt584a7363785c8/node/lt56b97e5c627c5/uu-no-19-tahun-2016-perubahan-atas-undang-undang-nomor-11-tahun-2008- tentang-informasi-dan-transaksi-elektronik 26 “Draft Revised Criminal Code [RKUHP] (Indonesia), 2019,” accessed November 13, 2020, Article 304, https://www.hukumonline.com/pusatdata/detail/17797/rancangan- undang-undang-2019# 27 Interview 7, interviewed by Andrea Carson, 19 August 2020. 28 Ismail Hasani, “The decreasing space for non-religious expression in Indonesia,” in Religion, Law and Intolerance in Indonesia, ed. Tim Lindsey and Helen Pausacker (London and New York: Routledge, 2016), 202. 29 “Criminal Code [KUHP] (Indonesia), 1946,” accessed November 13, 2020, Article 156, https://www.unodc.org/res/cld/document/idn/indonesian_penal_code_html/ I.1_Criminal_Code.pdf

4 La Trobe University Information and Electronic Transactions be ethnically Chinese, and to be leading disinformation through fake accounts.”39 law (ITE), the latter leading to 30 months’ a government involved in Chinese A Prabowo spokesperson later decried jail and a US$10,000 fine.30 conspiracies to influence Indonesian the behaviour, saying it creates “a affairs.35 A doctored image was once climate where voters become ever-more Numerous groups and individuals, posted purporting to show Jokowi at polarised and find it difficult to ascertain including journalists, have used a communist rally in 1955 – six years what is true and what is fiction – which, misinformation to play on Indonesians’ before he was born.36 However blatant, in the end, can only damage our ability to fears of Islamophobia.31 Groups found such falsehoods have been known to hold rational discourse.”40 Patterns in the to have intentionally inflamed religious gain traction, and are commonly used timing and targeting of hoax crackdowns tensions for such purposes include by political players to try to discredit have also prompted some to accuse Saracen and the Muslim Cyber Army.32 opponents and incite public unrest.37 the state of attempting to “strengthen During post-election protests controls on oppositional information” in the lead-up to elections.41 in 2019, many protestors Misinformation as said they were motivated a political weapon by online news and widely Prominent political figures have been shared private messages among those employing hoaxes to that falsely reported attacks advance their political objectives. An on Mosques and other forms investigation by The Guardian in 2018 33 found links between the Muslim Cyber of religious discrimination. Army and Indonesian opposition parties, as well as the Indonesian military.38 Content that seeks to frame individuals During the 2019 presidential election, the as sympathetic to communism (it is campaign teams behind Jokowi and his illegal under Indonesia’s Criminal Code rival Prabowo Subianto were found to to communicate communist ideology34) be “funding sophisticated social media has also been prominent. President operations to spread and Jokowi has been variously claimed to

Laws and regulations

Indonesia has a number of laws that it exercises in its fight against to rewrite the code have prompted misinformation. The key laws are detailed below, including plans to widespread consternation within Indonesia. update and amend the 1945 Criminal Code, and the new Ministerial Regulation no. 5/2020 launched as this report was going to print. In its current form, under Article 14(1) of the KUHP, people charged with spreading false information or news that intentionally causes public disorder 43 Criminal Code independence from the Netherlands was can face ten years in prison. Under declared. The code is a modified version Article 14(2), publishing false information Indonesia’s Criminal Code, or Kitab of the previous Dutch-imposed code,42 that may reasonably be considered to Undang-Undang Hukum Pidana (KUHP), and has remained largely unchanged for cause public unrest can result in up to has been in operation since 1945, when more than 70 years. Recent attempts three years’ jail.44 And under Article 15,

30 Ismail Hasani, “The decreasing space for non-religious expression in Indonesia.” 31 Andreas Harsono, “Indonesia’s journalists grapple with Islamism,” Human Rights Watch, published August 21, 2019, https://www.hrw.org/news/2019/08/21/indonesias- journalists-grapple-islamism 32 Lamb, “Muslim Cyber Army: a 'fake news' operation designed to derail Indonesia's leader.” 33 Quinton Temby, “Disinformation, violence, and anti-Chinese sentiment in Indonesia’s 2019 elections,” ISEAS, no. 67 (2019), https://www.iseas.edu.sg/images/pdf/ISEAS_ Perspective_2019_67.pdf 34 Ibid. 35 Amy Chew, “Chinese army uniforms” in Indonesia: Another hoax to discredit Beijing, Widodo,” South Morning Post, July 29, 2020, https://www.scmp.com/week-asia/ politics/article/3095217/chinese-army-uniforms-indonesia-another-hoax-discredit-beijing 36 Kieren Aris and Rachel Fielden, “Tackling online disinformation in Indonesia through media literacy: The redirect method,” (Paper presented at Fake News and Elections in Asia Conference, Bangkok, Thailand, July 10-12, 2019), 307. Accessed November 13, 2020, https://asiacentre.org/wp-content/uploads/2019/09/Fake-News-Conference- Proceeding.pdf 37 Temby, “Disinformation, violence, and anti-Chinese sentiment in Indonesia’s 2019 elections.” 38 Lamb, “Muslim Cyber Army: a 'fake news' operation designed to derail Indonesia's leader.” 39 Fanny Potkin and Augustinus Beo Da Costa, “Fact-checkers vs. hoax peddlers: A fake news battle ahead of Indonesia’s election,” Reuters, April 11, 2019, https://www.reuters.com/article/us-indonesia-election-fakenews-insight/fact-checkers-vs-hoax-peddlers-a-fake-news-battle-ahead-of-indonesias-election-idUSKCN1RM2ZE 40 Irawan Ronodipuro quoted in Tse Yin Lee, “Analysis: How effective was Indonesia's war on fake news in its election?,” BBC Monitoring, published May 16, 2019, https://monitoring.bbc.co.uk/product/c200tf1l 41 Tapsell, “Indonesia’s policing of hoax news increasingly politicised,” 4. 42 “RKUHP Explainer: All the controversial articles in Indonesia’s criminal code overhaul,” Coconuts Jakarta, September 19, 2019, https://coconuts.co/bali/features/rkuhp- explainer-all-the-controversial-articles-in-indonesias-criminal-code-overhaul-2/. 43 Agung Nugroho, “Comparative study on hoax handling policies in Indonesia and Singapore,” Advances in Social Science, Education and Humanities Research 343, (2019): 360. 44 Tapsell, “Indonesia’s policing of hoax news increasingly politicised,” 6.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 5 information or news that is uncertain, The Criminal Code has also been used disseminating news or information that exaggerated or incomplete, and that may to crack down on the use of the internet is uncertain, exaggerated or incomplete, cause public unrest, may result in up to by separatists in West Papua. However, and that may reasonably be believed to two years’ jail.45 in these cases the activists have been cause public unrest, could result in two accused of rather than spreading years in prison or a fine of Rp50 million Apart from controversy surrounding misinformation.52 ($AUD4,500 approx).57 proposed amendments, the use of the Criminal Code in its current form as a The Government has invoked articles Other provisions with the potential to weapon against online misinformation 14 and 15 of KUHP against the spread limit public expression cover criticism has attracted considerable commentary of misinformation about the COVID-19 of the President or Vice President (up and criticism. Institute of Criminal pandemic.53 Efforts to police COVID- to 4.5 years’ jail, or a Rp200 million fine) Justice Reform chief Anggara Suwahju related posts have also extended and of the Government (four years and says “commotion among netizens to crackdowns on criticism of the Rp500 million). There are also provisions (citizens of the internet)” shouldn’t be Government’s handling of the pandemic, covering blasphemy “against the religion enough to trigger the “chaos” article including “anyone caught insulting the adopted in Indonesia”58 (five years’ jail of the law.46 Critics also say the code President and other top officials in or a Rp500 million fine), and up to 18 is a “rubber article,” with ambiguous relation to COVID-19 policies,” which months’ jail, or a Rp10 million fine, for definitions allowing the Government can result in 18 months’ prison.54 defaming someone in public writing freedom to interpret and apply the law or imagery. However, it would not be to suit its own interests.47 Despite these Draft revisions of the criminal if doing so was in the public concerns, there is strong support in Criminal Code interest or in the accused’s defence.59 Indonesia for the proposition that the The Government’s declaration in 2018 Political debate Bill is due for revision to be relevant to of its plans to amend the KUHP was 48 modern realities. met with enthusiasm by some observers Indonesia’s opposition politicians have KUHP has been used during periods who felt it was overdue – and serious been relatively circumspect on the of crisis and potential public disorder in apprehension by others who feared proposed amendments, leading to the recent years to prosecute alleged cases it could lead to further tightening of suggestion by the Jakarta Post of “a of online misinformation. For example, restrictions on civil freedoms. The holy alliance between the ruling and the National Police charged more than first public outline of the proposed opposition camps in the House to get a dozen people under Article 14(2) with revisions confirmed these fears, leading the amended Criminal Code enacted into spreading fake news about a series of to widespread public protests and the law.”60 While several high-profile political purported child , and about consequent postponing of a vote on figures, including the deputy head of the Lion Air flight that crashed in October the Bill in September 2019. However, Indonesia’s House of Representatives, 2018.49 But efforts to quell the spread of renewed parliamentary deliberation have raised concerns that some articles fear in the community have sometimes has been flagged.55 in the Bill could be used against civil liberties, the main political parties have led to punishment of less educated Under proposed Article 262(1), anyone stated publicly that they stand behind Indonesians who unwittingly shared false who knowingly disseminates false the new draft.61 information while genuinely trying to information, resulting in public unrest, 50 build awareness about serious events. would be liable for a maximum prison The draft Bill has attracted criticism from The Jokowi Government has shown term of six years, or a fine of Rp500 politicians in neighbouring Southeast a willingness to use the Criminal million ($AUD47,000 approx). Under Asian countries. But these critics have Code against political opponents. 262(2), someone who spreads news they focused mainly on proposals affecting After the 2019 presidential elections a may suspect is a hoax, and that they may private and same-sex relationships.62 Prabowo campaign spokesman and reasonably believe could incite public an Opposition party politician were unrest, faces up to four years jail or a fine charged “on suspicion of of Rp200 million ($AUD19,000 approx).56 and spreading hoaxes.”51 Article 263 states that

45 Nugroho, “Comparative study on hoax handling policies in Indonesia and Singapore,” 360. 46 Tapsell, “Indonesia’s policing of hoax news increasingly politicised,” 6. 47 Article 19 and Aliansi Jurnalis Independen [AJI], Freedom of Expression and the Media in Indonesia, (London and Jakarta: Article 19 and Aliansi Jurnalis Independen, 2005), https://www.refworld.org/pdfid/4754185f0.pdf. 48 “RKUHP Explainer: All the controversial articles in Indonesia’s criminal code overhaul.” 49 “Beware what you share: Police step up war on hoaxes,” The Jakarta Post, November 12, 2018, https://www.thejakartapost.com/news/2018/11/12/beware-what-you-share- police-step-war-hoaxes-fake-news.html 50 Tapsell, “Indonesia’s policing of hoax news increasingly politicised,” 3. 51 “Prabowo campaign spokesman arrested on subversion charges,” The Jakarta Post, May 21, 2019, https://www.thejakartapost.com/news/2019/05/21/prabowo-campaign- spokesman-arrested-on-subversioncharges.html. 52 “Open letter on the increasing use of makar charges against Papuan activists to stifle freedom of expression,” Amnesty International, published October 2, 2019, https://www.amnesty.id/open-letter-on-the-increasing-use-of-makar-charges-against-papuan-activists-to-stifle-freedom-of-expression/. 53 Tri Indah Oktavianti, “National Police detail tightened measures to combat COVID-19, support government policy,” The Jakarta Post, April 8, 2020, https://www.thejakartapost.com/news/2020/04/08/national-police-detail-tightened-measures-to-combat-covid-19-support-government-policy.html. 54 Ibid. 55 Irfan Fathurohman, “ICJR urges parliament and government to not need to hurry to pass the Criminal Code bill,” IDNTimes, published July 11, 2020, https://www.idntimes.com/news/indonesia/irfanfathurohman/icjr-desak-pemerintah-dpr-tidak-buru-buru-sahkan-ruu-kuhp/4 56 RKUHP, Article 262. 57 Ibid., Article 263. 58 Ibid., Article 304. 59 Ibid., Article 440. 60 “Criminalisation code,” The Jakarta Post, July 18, 2019, https://www.thejakartapost.com/academia/2019/07/18/criminalization-code.html. 61 Aisyah Llewellyn, “Punitive measures: Indonesia’s new draft criminal code” Lowy Institute, published February 9, 2018, https://www.lowyinstitute.org/the-interpreter/punitive- measures-indonesias-new-draft-criminal-code 62 Saifulbahri Ismail, “ASEAN lawmakers urge Indonesia to reject changes to criminal code,” Channel News Asia, February 7, 2018, https://www.channelnewsasia.com/news/ asia/asean-lawmakers-urge-indonesia-to-reject-changes-to-criminal-9935950

6 La Trobe University Public reactions of Expression Network (SAFEnet) said to make it “seem to be authentic.”79 that by punishing both the producers and Breaches carry up to 12 years’ jail and Initial public reaction in Indonesia to the sharers of hoaxes, the Bill had the scope fine of Rp12 billion ($AUD1.14 million).80 proposal to amend the law was often to charge almost anyone in Indonesia, positive. Human Rights Watch’s Andreas ITE also criminalises online defamation,81 particularly given the country’s low Harsono had hoped the changes could with up to four years’ jail and a fine of media literacy.73 establish a “rights-respecting” and Rp750 million ($AUD 70,000).82 63 “modernising and open” Indonesia. When the law was revised in 2016, it Others hoped it was an opportunity Information and included a controversial “right to be to make the law less discriminatory forgotten” clause (Article 26(3)), allowing 64 Electronic Transactions and toxic. However, goodwill soon applications to internet providers and started to recede as fears grew that Law platforms to have personal information the amendments would be “used as a The Information and Electronic removed. Critics have complained political tool.”65 Harsono said “the current Transactions Law (ITE) has been that this provision can be misused draft is much more draconian” than the by politicians to censor unwanted 66 the primary law used against online existing law. Because the current draft is 83 misinformation and hoaxes in Indonesia. commentary. Another revision, intended to please religious conservatives, Essentially, it extends the Criminal Code Article 40, enables the Government, it is likely to be more discriminatory for offences including hate speech, slander, either directly or through an intermediary, against religious minorities, LGBT groups 84 defamation and blasphemy where they to disable access to “prohibited” content. and women, he feared.67 are deemed to have occurred on the This sentiment was highlighted during internet.74 And unlike the corresponding Political debate the September 2019 protests.68 As law in Singapore, Indonesian citizens can The 2016 revisions to the law have leading Indonesian law firm SSEK stated: report perceived breaches of the law to not been universally supported by “Whatever the intention, the reality is authorities to prosecute.75 the political establishment. Before that the KUHP would criminalise more Article 28 is commonly used to the amendments were implemented, conduct than the current Criminal Code.”69 prosecute online hoaxes. It covers the House of Representatives deputy Concerns about freedom of speech and anyone who “knowingly… disseminates chairman called on free speech the media have dominated criticisms of false and misleading information advocates to generate momentum the draft Bill, particularly in the media resulting in consumer loss in Electronic to force changes to the defamation sector. One critic argued it would allow Transactions.”76 In practice, it has been clause, which he feared would be 85 the Government to “imprison anyone it used much more broadly. Breaches used to silence the public. In the end, dislikes,” including journalists.70 Others carry up to six years’ jail and a fine of the maximum jail sentence under its 86 claim deliberate ambiguity in the law Rp1 billion ($AUD95,000 approx).77 defamation clause was increased. about who is being targeted will enable Article 28 also bans the spread of The vice chair of the lower house suppression of media freedom.71 information that aims to incite hatred said the ITE law should be revised as it contained double standards,87 While there is sympathy among media or hostility based on ethnicity, religion, race and other social divisions (SARA).78 while 2019 vice presidential candidate supporters for the notion that hoaxes are Sandiaga Uno promised to revise the an “enemy” of freedom of expression, the Article 35 targets “any person who law “to protect us from ourselves” and problem cannot be tackled “by arresting knowingly and without authority… because “I don’t want us to be using 72 the people who share” hoaxes. Damar manipulates, creates, alters, deletes, [or] those laws to hit our opponents.”88 Juniarto of the Southeast Asia Freedom tampers with electronic information”

63 Andreas Harsono, “Indonesia’s president urges delay on Criminal Code vote,” Human Rights Watch, published September 23, 2019, https://www.hrw.org/news/2019/09/23/ indonesias-president-urges-delay-criminal-code-vote 64 Ibid. 65 Veronica Koman quoted in Llewellyn, “Punitive measures: Indonesia’s new draft criminal code.” 66 Interview 2, interviewed by Andrea Cars, August 20, 2020. 67 Ibid. 68 Human Rights Watch, World report 2020 (Human Rights Watch, 2020), 276, https://www.hrw.org/sites/default/files/world_report_download/hrw_world_report_2020_0.pdf 69 Dewi Savitri Reni and Syarifah Reihana Fakhry, “RKUHP: What it means for criminal justice in Indonesia,” SSEK, published March 3, 2020, https://www.ssek.com/blog/rkuhp- what-it-means-for-criminal-justice-in-indonesia 70 “Criminalisation code.” 71 Ade Wahyudin (public defender at LBH Press) quoted in Arzia Tivany Wargadiredja, “Indonesia’s answer to fake news is a threat to free speech,” Vice News, published February 6, 2018, https://www.vice.com/en_asia/article/j5bq54/indonesias-answer-to-fake-news-might-be-a-threat-to-free-speech 72 Ibid. 73 Damar Juniarto (regional coordinator of SAFEnet) quoted in Wargadiredja, “Indonesia’s answer to fake news is a threat to free speech.” 74 Andreas Harsono, interviewed by Andrea Carson, August 20, 2020. 75 Ibid. 76 UU ITE, Article 28(1). 77 Ibid., Article 45(2). 78 SARA is an “Indonesian abbreviation for ethnicity, religion, race, and other social divisions” (Nathaneal Sumaktoyo, “Ethnicity and Jakarta’s election,” New Mandala, published February 13, 2017, https://www.newmandala.org/ethnicity-jakartas-election/.) 79 UU ITE, Article 35. 80 Ibid., Article 51(1). 81 Ibid., Article 27(3). 82 “ITE law aims to protect, manage electronic information, transactions,” The Jakarta Post, October 27, 2016, https://www.thejakartapost.com/news/2016/10/27/ite-law-aims- to-protect-manage-electronic-information-transactions.html. 83 Nadine Freischlad, “Controversial “right to be forgotten” finds its way into Indonesian law,” Tech In Asia, published December 1, 2016, https://www.techinasia.com/indonesia- recognizes-right-to-be-forgotten.; Krithika Varagur, “Indonesia poised to pass Asia’s first “right to be forgotten” law,” Voa News, November 7, 2016, https://www.voanews.com/ east-asia/indonesia-poised-pass-asias-first-right-be-forgotten-law 84 UU ITE, Article 40. 85 “Lawmakers push for amendment of electronic law,” Tempo, September 13, 2013, https://en.tempo.co/read/512919/lawmakers-push-for-amendment-of-electronic-law 86 “ITE law aims to protect, manage electronic information, transactions.” 87 S. Sudhakar, “Ahmad Dhani imprisoned, Fahri Hamzah encourages revised LAW,” Suara, published January 29, 2019, https://www.suara.com/news/2019/01/29/144219/ ahmad-dhani-dipenjara-fahri-hamzah-dorong-uu-ite-direvisi 88 Vice-Presidential candidate Sandiaga Uno quoted in “Q&A: Sandiaga Uno on economic policy,” New Mandala, published March 4, 2019, https://www.newmandala.org/qa- sandiaga-uno-on-economic-policy/

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 7 Popular reactions should be a “last resort,” and that the Dr Ross Tapsell of Australian National focus should be on “mediation in the University says ITE provisions were Community organisations and sections name of restorative justice.”101 used during the most recent general of the media have been vocal critics of election campaign against ordinary the ITE law, claiming it threatens freedom Similar arguments have been mounted Indonesians who were unknowingly of speech and media freedom. Asep by international observers. Human sharing fake news about serious events Komaruddin of the Legal Aid Institute Rights Watch and Amnesty International such as natural disasters.108 Tapsell for the Press said the requirement for have both warned that the Jokowi believes the extent of the Government’s “electronic system organisers” to delete administration’s growing use of the ITE intervention in this space is in part due information at the direction of courts89 Law threatens free speech.102 Freedom to the absence of effective controls was a potential threat to press freedom.90 House took particular issue with the from global tech platforms.109 He says defamation clause, which often lands SAFEnet argued that the defamation the Government’s overreach against alleged perpetrators in extended pre-trial components of the Bill did not distinguish citizens could fuel increased distrust detention.103 While some defamation between civil and criminal defamation, towards the state and the police, and charges never actually make it to court, it and were generally ambiguous,91 allowing a diminishing of “Indonesia’s generally said the threat of pre-trial detention was people to press charges against anyone open and vibrant public sphere.”110 This used by the Government to intimidate that insults or provokes them.92 SAFEnet has been confirmed in public opinion others. Human Rights Watch and identified problems with deletion of polls.111 Tapsell has proposed an SAFEnet agree that the Government’s information,93 defamation,94 termination alternative strategy of policing to focus focus should be on providing correct of online access,95 and the threat of on “syndicates” that spread fake news for information to citizens, rather than criminal punishment.96,97 It argued that the political or monetary purposes. criminalising what citizens say because law had multiple potential interpretations, they lack correct information.104 resulting in legal uncertainty and creating Uses and consequences a “revenge” dynamic, enabling people to Academic views SAFEnet, which monitors the use of the use the law against their enemies. It also ITE law, says the law was used 263 times Academics have also been critical of had the potential to be used as “shock from its inception in 2008 to the end of the content and application of the ITE therapy” against the public, causing a 2018,112 overwhelmingly against members 98 law. Usman Hamid, a lecturer in law at chilling effect. of the public rather than members of Indonesia’s Jentera School, argued that particular groups. In 2017 and 2018, Amnesty International Indonesia’s while the law was introduced ostensibly journalists and media professionals were Usman Hamid highlighted issues with to provide protections for consumers targeted more than any other groups the wording of the law, saying “electronic online, “in practice, the Government and (apart from the public). The number of information” could reasonably cover law enforcement officials have abused court cases relating to ITE has risen emails and text messages, and thus the law to silence political dissidents.”105 dramatically, from 20 in 2015 to 292 in threaten gross invasions of personal Others said a lack of clarity in the law’s 99 2018. According to SAFEnet, 7 per cent privacy. Moreover, he said the law defamation provisions could result in of cases brought under the ITE law have did not clearly differentiate between silencing of some people and groups, related specifically to hoaxes, 10 per cent insults and defamation as defined in including women who speak out against 100 to information manipulation, 22 per cent the country’s Criminal Code. And gender-based violence.106 Provisions to hate speech, and 24 per cent to MAFINDO’s Septiaji Eko Nugroho said such as article 40, which the Government defamation. the criminalisation of online speech has used to disable access to prohibited online content, were also seen to “carry huge risks of power abuses.”107

89 UU ITE, Article 26. 90 “Right contained in ITE Law to delete negative news threatens press freedom: Activist,” The Jakarta Post, October 28, 2016, https://www.thejakartapost.com/ news/2016/10/28/right-contained-in-ite-law-to-delete-negative-news-threatens-press-freedom-activist.html 91 Enricko Lukman, “How Indonesia’s problematic internet law will impede freedom of expression,” SafeNet, published September 10, 2014, https://safenet.or.id/2014/09/how- indonesias-problematic-internet-law-will-impede-freedom-of-expression/ 92 Ibid. 93 UU ITE, Article 26. 94 Ibid., Article 36. 95 Ibid., Article 40. 96 Ibid., Article 45. 97 “Persoalan UU ITE dan Praktik Pelanggaran Hak Digital di Indonesia,” SAFEnet, published November 14, 2019, https://id.safenet.or.id/wp-content/uploads/2019/11/Persoalan- UU-ITE-dan-Pelanggaran-Hak-Digital-SAFEnet-2019.pdf 98 Ibid. 99 Hamid, “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop.” 100 Ibid. 101 Septiaji Eko Nugroho (head of Mafindo) quoted in “Indonesian police intensify crackdown on fake news,” Voa News, November 21, 2018, https://www.voanews.com/east-asia- pacific/indonesian-police-intensify-crackdown-fake-news 102 Lee, “Analysis: How effective was Indonesia's war on fake news in its election?” 103 Freedom House, Freedom on the Net 2016 (Freedom House, 2016), 424, https://freedomhouse.org/sites/default/files/2020-02/Freedom_on_the_Net_2016_complete_book. pdf 104 “Hundreds Arrested for ‘Fake News’ in Asia,” Courthouse News Service, April 10, 2020, https://www.courthousenews.com/asia-arrests-hundreds-for-fake-news-on-virus/ 105 Hamid, “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop.” 106 Fitri Bintang Timur, “Unequal power makes women brunt of cyber law,” SafeNet, published March 11, 2017, https://safenet.or.id/2017/03/unequal-power-makes-women-brunt- of-cyber-law/ 107 Wahyudi Djafar quoted in “Revised ITE Law could hamper freedom of expression: Researcher,” The Jakarta Post, October 31, 2016, https://www.thejakartapost.com/ news/2016/10/31/revised-ite-law-could-hamper-freedom-of-expression-researcher.html 108 Tapsell, “Indonesia’s policing of hoax news increasingly politicised,” 3. 109 Ibid., 7. 110 Ibid. 111 Tribunnews.com, “Publik Lebih Takut Berbicara Politik di Era Presiden Jokowi? Berikut Hasil Survei LSI”, Tribunnews.com, 5 November 2019, https://www.tribunnews.com/ nasional/2019/11/05/publik-lebih-takut-berbicara-politik-di-era-presiden-jokowi-berikut-hasil-survei-lsi?page=all. 112 “Persoalan UU ITE dan Praktik Pelanggaran Hak Digital di Indonesia.”

8 La Trobe University Defamation a third of those involved instances of attack on freedom of speech.127 But the purported “hate speech” and “insults” Government said the situation in West While public officials can use the towards Jokowi, of which a majority were Papua demanded strong measures to Criminal Code’s defamation clause to communicated through social media. reduce misinformation and restore order quash criticism, the number of legal Conversely, Tapsell found no evidence – even though protests at which the action cases involving personal defamatory of anyone being prosecuted by the state was targeted had already stopped.128 statements has increased since the for spreading falsehoods about Jokowi’s introduction of the ITE law.113 The Also in 2019, Facebook removed hundreds opponent, Prabowo, during the 2019 defamation clause is also “periodically of accounts that regularly posted about election campaign.121 The ITE law was used to prosecute alleged religious West Papua on the grounds that they also used to block access to WhatsApp, defamation,”114 and has more recently demonstrated “coordinated inauthentic Facebook and Instagram features during been used against critical messages on behaviour.”129 Indonesia’s Minister for the 2019 post-election riots in Jakarta WhatsApp.115 Communication and Information denied “to avoid ” from the spread 122 any government role in this intervention. Ethnic and religious protections of misinformation. COVID-19 The ITE law has been used in conjunction Critics of the Government’s high-profile with the law on the Elimination of Race prosecutions of political dissidents The ITE law has been the Government’s and Ethnic Discrimination116 to prosecute claim this has encouraged attacks on principal weapon-of-choice against hoaxes.117 A recent case saw a member dissidents by government supporters, alleged COVID-19 hoaxes. Among the of the infamous online syndicate Saracen who in some cases have responded targets have been people spreading 123 sentenced to 32 months’ prison for with physical intimidation. Attempts false information about the pandemic, intentionally spreading information to by dissidents to challenge their perpetrators of alleged “health incite hate on social media.118 In another prosecutions have repeatedly failed; a equipment ”130 and critics of the case, a man was sentenced to five years total of seven applications for judicial Government’s handling of the crisis131 124 jail after being found guilty of spreading review have all been denied. and Jokowi’s leadership.132 By April 2020, the Government had made more than 80 hate speech in a Facebook post that West Papua “questioned the faith of other Muslims.”119 over alleged COVID-related fake The Government also used the law news.133 The actions drew claims that Crackdown on dissent against critics in West Papua, the COVID-19 was being used by authorities Hundreds of Indonesians have been scene of a long-running struggle for to justify crackdowns on journalists 134 prosecuted under the ITE law over independence. In 2019 human rights critical of Jokowi and the Government.’ lawyer and West Papua defender comments critical of the Government Other misinformation laws and President Jokowi in recent years. Veronica Koman was charged under the During Jokowi’s first term (2014-2019), ITE law for allegedly spreading fake news Several other laws that have potential 125 241 individuals were “criminalised that caused unrest in the region. In applications in the fight against online for criticising authority figures of the August of that year the Government used hoaxes in Indonesia have been used Jokowi administration,” according to Article 40 of the ITE law to shut down more sparingly. 126 Usman Hamid relying on unpublished internet access in West Papua. Legal Amnesty International figures.120 About experts argued the move was an illegal

113 Freedom House, Freedom on the Net 2016, 437 114 Ibid. 115 “[Press release] SAFEnet sends Amicus Curiae for the Saiful Mahdi case,” SAFEnet, published February 11, 2020, https://id.safenet.or.id/2020/02/rilis-pers-safenet-kirim- amicus-curiae-untuk-kasus-saiful-mahdi/ 116 “Law No. 40 of 2008 on Elimination of Racial and Ethnic Discrimination (Jakarta): 2008” HUKUM Online, accessed November 13, 2020, https://www.hukumonline.com/ pusatdata/detail/28539/node/1011/uu-no-40-tahun-2008-penghapusan-diskriminasi-ras-dan-etnis 117 Muhammad Hatta, “The spread of hoaxes and its legal consequences,” International Journal of Psychosocial Rehabilitation 24, no. 3 (2020): 1756-1757. 118 “Indonesia court sentences administrator of 'fake news factory' Saracen to jail,” The Straits Times, January 12, 2018, https://www.straitstimes.com/asia/se-asia/indonesia- court-sentences-administrator-of-fake-news-factory-saracen-to-jail 119 “Indonesian man arrested for Facebook post allegedly insulting Islam,” Mid-Day, May 1, 2018, https://www.mid-day.com/articles/indonesian-man-arrested-for-facebook-post- allegedly-insulting-islam/19374167. 120 Hamid, “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop.” 121 Tapsell, “Indonesia’s policing of hoax news increasingly politicised.” 4. 122 Karina M. Tehusijarana and Jessicha Valentina, “Jakarta riot: Government temporarily limits access to social media, messaging apps,” The Jakarta Post, May 22, 2019, https://www.thejakartapost.com/life/2019/05/22/jakarta-riot-government-temporarily-limits-access-to-social-media-messaging-apps.html 123 Hamid, “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop.” 124 Ibid. 125 Kate Lamb, “Outcry as Indonesia seeks to arrest renowned West Papua rights lawyer,” The Guardian, September 5, 2019, https://www.theguardian.com/world/2019/sep/05/ outcry-as-indonesia-seeks-to-arrest-renowned-west-papua-rights-lawyer 126 Reuters, “Rights group, journalists press Indonesian govt to lift internet curb in restive Papua.” 127 Oce Madril and Herlambang P Wiratraman cited in “The Experts: The Indonesia Ministry of Communication and Informatics and the President Violates Human Rights and State Administrative Law,” SAFEnet, published March 17, 2020, https://safenet.or.id/2020/03/experts-the-ministry-of-communication-and-information-and-the-president- violates-human-rights-and-state-administrative-law/ 128 “Indonesian gov’t justifies maintaining internet block in Papua despite officials declaring security restored,” Coconuts Jakarta, August 28, 2019, https://coconuts.co/jakarta/ news/indonesian-govt-justifies-maintaining-internet-block-in-papua-despite-officials-declaring-security-restored/. 129 Adi Renaldi, “Facebook Removed Hundreds of Accounts for Allegedly Spreading Propaganda in Indonesia,” VICE News, October 10, 2019, https://www.vice.com/en_asia/ article/ne8qbq/facebook-removed-hundreds-accounts-spreading-propaganda-indonesia 130 Petir Garda Bhwana, “Coronavirus-crackdown aimed at anti-President smears, hoaxes,” Tempo, April 6, 2020, https://en.tempo.co/read/1328272/coronavirus-crackdown- aimed-at-anti-president-smears-hoaxes 131 Kanis Dursin, “Hoax stories hamper Indonesia’s COVID-19 fight,”Bangkok Post, July 6, 2020, https://www.bangkokpost.com/opinion/opinion/1946480/hoax-stories-hamper- indonesias-covid-19-fight 132 Oktavianti, “National Police detail tightened measures to combat COVID-19, support government policy.” 133 “Hundreds Arrested for ‘Fake News’ in Asia.” 134 “Indonesia used Covid-19 to silence criticism of government,” Reporters Without Borders, April 16, 2020, https://rsf.org/en/news/indonesia-used-covid-19-silence-criticism- government

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 9 Law No. 17/2011 on State tasked with managing conflicts and members to stop them turning up at Intelligence intelligence during elections.142 This election polling stations, which he said was apparent during the 2019 protests, would be a form of terrorism.149 Article Articles 30-34 give the State Intelligence when BIN worked alongside the National 1.4 defines the threat of violence as Agency (BIN) the power to conduct Police and National Defence Forces to “speech, writing, picture, symbol or surveillance, information extraction and “safeguard the country.”143 Under a new body language… which could create other tasks relating to “activities that regulation introduced in July 2020, BIN widespread fear.”150 135 threaten national interests and security” operates directly under the control of such as terrorism, political violence and the President, raising concerns that it Reactions “cyber warfare.”136 To the extent that the may become another instrument used Some members of Parliament have law is applied to hoaxes, BIN is tasked to quell political dissent.144 with investigating and combatting the rejected the idea of using the Terrorism “potential dangers they pose,”137 and The COVID-19 pandemic has brought Act against hoaxes, arguing that the ITE 151 assisting with any prosecution under BIN’s activities further into the public law is sufficient to prosecute them. the other relevant laws. spotlight. As the national overseer Human rights activists have complained of intelligence, BIN has been given that the 2018 Terrorism Act revisions had Reactions responsibilities that have “extended to tightened what were already restrictive everything related to the pandemic,” and punitive blasphemy and defamation Efforts to enact new laws to combat including even using its resources to laws.152 As of March 2019, no person had terrorism and other security threats, conduct contact tracing to monitor been charged with spreading hoaxes particularly after the 2002 Bali bombings, its spread.145 BIN has also provided under the Terrorism Law.153 initially failed amid public opposition to public information about the pandemic, giving BIN “Stasi-like secret police” powers announcing in July “the discovery of five Ministerial Regulation no.5/2020 and status.138 A revised bill was eventually combinations of drugs to fight COVID-19.”146 passed in 2011, but has remained the In December 2020, the Indonesian Embarrassingly for an institution tasked subject of controversy since a failed government launched its new Ministerial with stamping out misinformation, BIN 2012 bid by a coalition of civil society Regulation no.5/2020 to further regulate recently disproved its own “99 per cent organisations to have judicial review of how private Electronic Service Operators accurate” predictions on COVID case 16 of its clauses. Concerns included the (ESO) such as cloud computing service numbers and had to retract them.147 law’s “vague language and broad reach.”139 providers, social media platforms and other applications serving Indonesians Human Rights Watch, while acknowledging Terrorism Act 2018 should operate. In essence, the ministerial the need to protect Indonesians from Article 1 of the Terrorism Act can be used regulation provides more authority for threats such as terrorism, has called for against “perpetrators of the spread of the government to obtain data for law the law’s repeal, saying “loose language… hoaxes… if there is an element of threat enforcement purposes; and to hold 140 invites dangerous misuse.” of violence (that) creates an atmosphere platforms accountable to remove misinformation and other content deemed Uses and consequences of terror and widespread fear,” according to a senior police official.148 A senior as public disorder with a strict turnaround Since around 2015, BIN has been government official said this provision time of four hours. Failure to comply can empowered to increase its focus could apply in the cases of hoaxes result in heavy financial penalties and on online threats,141 and specifically threatening violence against community even internet blocking of platforms.

135 “Law No. 17 of 2011 on State Intelligence (Indonesia), 2011,” accessed November 13, 2020, Article 31(a), https://www.icj.org/wp-content/uploads/2011/01/Indonesia- intelligence-law-draft-2011-eng.pdf 136 Ibid., General I. 137 Budi Gunawan and Barito Mulyo Ratmono, “Social media, cyberhoaxes and national security: Threats and protection in Indonesian cyberspace,” International Journal of Network Security 22, no.1 (2020): 99-101. 138 Bhatara Ibnu Reza, “Why strengthening anti-terror laws is not the answer” Indonesia at Melbourne, published February 1, 2016, https://indonesiaatmelbourne.unimelb.edu.au/ why-strengthening-anti-terror-laws-not-the-answer/ 139 Ibid. 140 Elaine Pearson, deputy Asia director of Human Rights Watch, quoted in “Indonesia: Repeal new intelligence law,” Human Rights Watch, published October 26, 2011, https://www.hrw.org/news/2011/10/26/indonesia-repeal-new-intelligence-law. 141 Ina Parlina, “Jokowi declares fight against disseminators of fake news,” The Jakarta Post, December 30, 2016, https://www.thejakartapost.com/news/2016/12/30/jokowi- declares-fight-against-disseminators-of-fake-news.html 142 “BIN budget cuts to jeopardise regional election security, says spy chief,” Jakarta Globe, September 10, 2015, https://jakartaglobe.id/context/bin-budget-cuts-jeopardize- regional-election-security-says-spy-chief/ 143 Minister for Home Affairs Tjahjo Kumolo quoted in Office of Assistant to Deputy Cabinet Secretary for State Documents & Translation, “National political stability under control: Home ministry,” Cabinet Secretariet of the Republic of Indonesia, published September 26, 2019, https://setkab.go.id/en/national-political-stability-under-control- home-ministry/ 144 “New regulation to have intelligence agency work directly under President Jokowi,” The Jakarta Post, July 20, 2020, https://www.thejakartapost.com/news/2020/07/20/new- regulation-to-have-intelligence-agency-work-directly-under-president-jokowi.html 145 Tangguh Chairil, “Indonesia’s intelligence service is coming out to counter COVID-19” The Diplomat, June 19, 2020, https://thediplomat.com/2020/06/indonesias-intelligence- service-is-coming-out-to-counter-covid-19/ 146 Wahyudi Soeriaatmadja, “Eucalyptus necklace among “cures” raising alarm in Indonesia,” The Straits Times, July 10, 2020, https://www.straitstimes.com/asia/se-asia/ eucalyptus-necklace-among-cures-raising-alarm-in-indonesia 147 Chairil, “Indonesia’s intelligence service is coming out to counter COVID-19.” 148 Dedi Prasetyo quoted in Lis Yuliawati and Bayu Nugraha, “Penyebar Hoax Dijerat UU Terorisme, Polri: Tergantung Fakta Hukum,” Viva, March 21, 2019, https://www.viva.co.id/berita/nasional/1132456-penyebar-hoax-dijerat-uu-terorisme-polri-tergantung-fakta-hukum? 149 Coordinating Minister for Political, Legal and Security Affairs Wiranto quoted in “Senior Indonesian minister suggests anti-terrorism law be used against election hoax spreaders,” Coconuts Jakarta, March 2, 2019, https://coconuts.co/jakarta/news/senior-indonesian-minister-suggest-anti-terrorism-law-used-election-hoax-spreaders/ 150 Krithika Varagur, “Activists: Indonesian counterterrorism law threatens civil liberties,” Voa News, May 30, 2018, https://www.voanews.com/east-asia-pacific/activists- indonesian-counterterrorism-law-threatens-civil-liberties 151 House of Representatives member Nasir Diamil quoted in “Senior Indonesian minister suggests anti-terrorism law be used against election hoax spreaders.” 152 Varagur, “Activists: Indonesian counterterrorism law threatens civil liberties.” 153 “Senior Indonesian minister suggests anti-terrorism law be used against election hoax spreaders.”

10 La Trobe University Government bodies

Ministry of in conjunction with MAFINDO and would arrest those suspected of initiating the Google News Initiative seeks to them.164 More recently, it blocked access Communication educate Indonesians about how to spot to some online locations to fight the and Information fake information.157 In a similar vein, spread of misinformation about the Technology KOMINFO established the “Siberkreasi” COVID-19 pandemic.165 program in 2017 to educate and involve In addition to using the law in the the public in overcoming “the spread of National Cyber and Encryption fight against online misinformation, negative content through the internet Agency Indonesia’s Ministry of Communication such as hoaxes, cyberbullying and online Like KOMINFO, the National Cyber and Information Technology (KOMINFO) 158 radicalism.” The program also seeks and Encryption Agency (BSSN) has has actively engaged with the online to engage with the private sector, been empowered by the Indonesian industry and the community through academics, civil society organisations Government to assist with the fight various programs and operations. 159 and the media. Despite these against online hoaxes. Unlike KOMINFO, The ministry has been particularly initiatives, digital literacy in Indonesia however, its operations appear to be focused on seeking cooperation from has been found to be “incidental confined largely to monitoring online 160 social media platforms to monitor and and sporadic.” activity, rather than prosecuting remove false online information. In KOMINFO has also been involved in at breaches.166 Established in 2017, it has a January 2018, for example, it persuaded times controversial initiatives targeting Head of Agency who reports directly to nine social media and messaging alleged political hoaxes. Before the 2019 President Widodo. platforms to sign an agreement to presidential election, a “war room” of 100 In the lead-up to the 2019 presidential tackle hoaxes and hate speech on staff was tasked with identifying online election, the agency was tasked with their sites – a move widely seen as an 161 hoaxes 24 hours a day. KOMINFO helping intelligence agencies and law to shore up controls ahead was also a key player in the curbing of enforcement efforts to combat online of that year’s local elections.154 Under social media access during post-election misinformation and hoaxes.167 the agreement, platforms were obliged protests, including the blocking of video to remove accounts designated as and photo uploads and downloads Reactions prohibited by Indonesia’s election on several social media platforms.162 Politically, BSSN has come under watchdog, the Bawaslu. KOMINFO has However, these efforts had limited scrutiny over the limits of its authority. also established an artificial intelligence impact due to misinformation continuing One senior member of the House of system, Cyber Drone 9, that seeks out “apace” on the encrypted Telegram.163 Representatives warned that such limits “targeted content” on .155 Once need to be enforced so that BSSN does the content is identified, internet service Uses and consequences not become “a censorship agency.”168 providers are told to remove it. KOMINFO has used its powers to quell Civil society groups have argued that KOMINFO has launched multiple the spread of misinformation during BSSN could be more proactive in initiatives to engage with the community non-political crises. For example, the providing guidelines to individuals on on online misinformation. In 2018 2018 Sulawesi earthquake prompted the how to protect themselves against it announced it would begin regular spread of false reports that authorities , particularly during the briefings to detail and correct hoaxes believed had the potential to incite COVID-19 pandemic.169 it had identified.156 It has launched a panic. KOMINFO issued press releases dedicated website, “Stophoax,” which “debunking” the hoaxes, and said they

154 La Batu, “Nine social media, messaging apps agree to combat fake news.” 155 Ibid. 156 Kate Lamb, “Indonesian government to hold weekly “fake news” briefings,” The Guardian, September 27, 2018. https://www.theguardian.com/world/2018/sep/27/ indonesian-government-to-hold-weekly-fake-news-briefings 157 “About Stophoax.id at a glance,” Stophoax, accessed November 13, 2020, URL: https://stophoax.id/tentang-kami/ 158 “About Siberkreasi.” 159 Ibid. 160 Novi Kurnia and Santi Indra Astuti, “Researchers find Indonesia needs more digital literacy education,”The Conversation, September 26, 2017, https://theconversation.com/researchers-find-indonesia-needs-more-digital-literacy-education-84570 161 Jack Board, “Inside Indonesia’s ‘fake news’ war room, fighting political hoaxes in election season,” Channel News Asia, April 13, 2019, https://www.channelnewsasia.com/news/asia/indonesia-election-fake-news-war-room-fighting-political-hoaxes-11439398 162 Quinton Temby, “Indonesia’s election riots offer a lesson on the perils of fake news,” , September 9, 2019, https://www.scmp.com/week-asia/politics/article/3026057/indonesias-election-riots-offer-lesson-perils-fake-news 163 Ibid. 164 Kate Lamb, “Sulawesi tsunami: Indonesia battles fake news as hoaxers spread panic,” The Guardian, October 4, 2018, https://www.theguardian.com/world/2018/oct/04/ sulawesi-tsunami-indonesia-battles-fake-news-as-hoaxers-spread-panic 165 “Indonesia wages war against coronavirus misinformation as hoaxes spread online,” The Jakarta Post, February 3, 2020, https://www.thejakartapost.com/news/2020/02/03/ indonesia-wages-war-against-coronavirus-misinformation-as-hoaxes-spread-online.html 166 Vience Mutiara Rumata and Ashwin Sasongko Sastrosubroto, “Net-attack 2.0: Digital post-truth and its regulatory challenges in Indonesia” Advances in Social Science, Education and Humanities Research 165, (2018): 119 167 Daniel Funke and Daniela Flamini, “A guide to anti-misinformation actions around the world,” Poynter, accessed November 14, 2020, https://www.poynter.org/ifcn/anti- misinformation-actions/#indonesia 168 Dewi Nurita, “Fadli Zon urge National Cyber Agency hit the brake,” Tempo, January 8, 2018, https://en.tempo.co/read/914656/fadli-zon-urge-national-cyber-agency-hit-the- brake 169 eltsazar A. Krisetya, “COVID-19 exposes vulnerabilities in our cyberspace,” The Jakarta Post, April 1, 2020, https://www.thejakartapost.com/academia/2020/04/01/covid-19- exposes-vulnerabilities-in-our-cyberspace.html.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 11 Uses and consequences helped Facebook to identify accounts leading BSSN head Siburian to credit the linked to the Saracen hate group.170 It agency with Indonesia’s rise from 70 to In addition to providing information that has also launched programs to raise 41 in the Global Cybersecurity Index in enables other agencies to prosecute cyber-security literacy and awareness,171 2017.172 alleged hoax spreaders, BSSN has

Timeline of Indonesia’s “fake news” laws and policies

Table 1.1: Timeline of various laws and regulations implemented in Indonesia

Year Legislation/Policy Development 1946 Indonesia adopts the minimally-revised Criminal Code (KUHP) that had been enforced by the Dutch colonial authorities.

2008 Information and Electronic Transactions Law (UU ITE) is adopted. First judicial review of UU ITE fails.173 2009 Second judicial review of UU ITE fails. 2010 Third judicial review of UU ITE fails. 2011 Law No. 17 of 2011 on State Intelligence is adopted. 2012 Susilo Bambang Yudhoyono government plans to revise ITE law, but it does not proceed 2013 The government, through the Ministry of Law and Human Rights, tables the revised Criminal Code Bill (RKUHP) and Criminal Procedural Code Bill (RKUHAP) into the People’s Representative Council (DPR). However progress of the Bill stops following public opposition and the timing of upcoming 2014 election.174 Fourth judicial review of UU ITE fails. 2015 Recently elected President Joko Widodo signals that the revised Criminal Code is a priority bill.175 Discussions between the Government and legislature over the drafting of a new Criminal Code begin.176 2016 Fifth and sixth judicial reviews of UU ITE fail. UU ITE revisions are adopted, which include an expanded definition of defamation under Article 27(3), increased maximum jail sentences for defamation from four years to six years, and a reduction of the fine from Rp1 billion ($AUD94,800 approx) to Rp750 million ($AUD70,500 approx). 2017 National Cyber and Encryption Agency (BSSN) is established in law, effectively amalgamating the National Encryption Agency (“Lemsaneg”) and the National Cyber Agency. 2018 [January] National Cyber and Encryption Agency (BSSN) Head of Agency is named, beginning its operations in earnest. [January] After 50 years of failed attempts, the government reveals its plans to revise the Criminal Code, causing concern from civil society groups in Indonesia. [May] Terrorism Act is revised only nine days after the deadliest terror attack since 2002 Bali Bombings. Seventh judicial review of UU ITE fails. 2019 [September] The scheduled vote on the revised Criminal Code inspires mass demonstrations in Indonesia that last for over a month, mainly led by student protestors. They were the largest student protests since those that were central to the fall of Suharto in 1998. President Widodo postpones vote on RKUHP. 2020 [December] Indonesia passes the controversial Ministerial Regulation 5/2020, which, amongst other things, requires digital platforms to remove content “against the public order” within 4 hours, under the threat of severe penalties (including blocking of platforms). Source: Authors

170 “Facebook takes down hundreds of Indonesian accounts linked to fake news syndicate,” Reuters, February 1, 2019, https://www.reuters.com/article/us-facebook-indonesia/ facebook-takes-down-hundreds-of-indonesian-accounts-linked-to-fake-news-syndicate-idUSKCN1PQ3JS 171 General Hinsa Siburian, head of BSSN, quoted in Febi Trihermanto and Helen Brown, “Indonesia focused on cyber resilience, open to foreign cooperation,” Australia Indonesia Centre, published November 12, 2019, https://digital.australiaindonesiacentre.org/news/insights/indonesia-focused-on-cyber-resilience-open-to-foreign-cooperation/ 172 Ibid. 173 SAFEnet, ‘Persoalan UU ITE dan Praktik Pelanggaran Hak Digital di Indonesia’, SAFEnet, 14 November 2019, URL: https://id.safenet.or.id/wp-content/uploads/2019/11/ Persoalan-UU-ITE-dan-Pelanggaran-Hak-Digital-SAFEnet-2019.pdf 174 Eddyono, Supriyadi Widodo, Abidin, Zainal, Arsil and Yuntho, Emerson, ‘Brief notes on the proposed discussion of 2015 Criminal Code Bill’, National Alliance for Criminal Code Reform, 26 March 2015, URL: https://elsam.or.id/wp-content/uploads/2015/08/Brief-Notes-on-the-Proposed-Discussion-of-2015-Criminal-Code-Bill.pdf 175 Ibid. 176 Indonesia at Melbourne, ‘Indonesians better get ready for jail, as flawed new criminal code looks set to pass’, University of Melbourne [website], 3 September 2019, URL:https://indonesiaatmelbourne.unimelb.edu.au/indonesians-better-get-ready-for-jail-as-flawed-new-criminal-code-looks-set-to-pass/

12 La Trobe University Conclusion

Misinformation poses a serious and destabilising threat in Finally, the revised draft of the Criminal Code, which has partisan support, places Indonesia due to a combination of historical, demographic further restrictions on civil liberties. and political factors. Indonesia’s status as one of the largest While Indonesia has fair elections, there internet-using populations and its low levels of educational appears to be a ruling elite, existing attainment also present significant challenges for actors across political lines, that has sought to consolidate its power at the expense seeking to defend the citizenry against misinformation. of the civil liberties achieved since the The nation’s conservative religiosity, combined with fall of Suharto.177 According to some national attitudes of anti-communism and anti-atheism, commentators, this continued erosion of civil liberties suggests the country may ensure the list of prohibited discourse is considerably long. be in the midst of protracted “democratic Lingering memories of autocratic repression, and more regression.”178 External statistics support recent state interference in public discourse, feed fears this view. While press freedom has improved over the past decade (albeit of government overreach. still ranking in the bottom quartile globally),179 internet freedom180 and political and civil rights181 have declined – These factors have helped create an Secondly, and to that point, the punitive particularly in the last couple of years. often unruly information environment, criminalisation of misinformation is where incendiary attacks on highly problematic. The Indonesian There is an attendant complexity, religious, ethnic, gender, sexual, and Government has shown readiness to however, to Indonesia’s handling of political grounds are unfortunately use the police to arrest those suspected misinformation. While questions are not uncommon. of spreading misinformation, which being asked of the Government’s use appears to be largely unchecked by of its anti-misinformation laws, the link The 2019 post-election riots other mechanisms of the state. There between misinformation and disruptive encapsulated the highly charged is evidence to suggest that these arrests public unrest in Indonesia is hard to experience of misinformation in are intended to have an intimidatory ignore. As the existence of online Indonesia. While rioters were initially effect, since many cases do not make shadow groups such as Saracen and prompted by Prabowo’s repudiation of it to trial. the Muslim Cyber Army show, actors the “facts” of the election result, further in Indonesia have exploited this reality unrest was stoked by false reports of Thirdly, there is a clear conflict of to stoke the country’s religious and government attacks on sacred mosques. interest in the definition and use of anti- political sensitivities in pursuit of their misinformation laws. For example, the Several key aspects of the Indonesian own goals. As such, any evaluation prohibition of defamatory or insulting Government’s response to of the Government’s measures to statements against the President has misinformation are worth restating. combat misinformation requires been used to prosecute those critical Firstly, its multilateral programs and nuanced consideration of Indonesia’s of the Government. These laws can be civil society coalition building are complex cultural, demographic used to repudiate opposition and deny encouraging. With low digital literacy and political realities. the validity of unwanted commentary. levels, public education is essential. This was seen during the post-election However, commitment to this approach riots, in West Papua and against is undermined as the Government relies opposition politicians. upon more blunt and efficient measures to police misinformation.

177 Eve Warburton and Edward Aspinall, “Explaining Indonesia’s democratic regression: Structure, agency and popular opinion,” Contemporary Southeast Asia 41, no, 2 (2019): 268. 178 Ibid. 179 “Indonesia,” Reporters Without Borders, 2020, https://rsf.org/en/indonesia. 180 “Freedom on the net 2019: Indonesia,” Freedom House, 2019, https://freedomhouse.org/country/indonesia/freedom-net/2019 181 “Freedom in the world 2020: Indonesia,” Freedom House, 2020, https://freedomhouse.org/country/indonesia/freedom-world/2020

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 13 CHAPTER 2 Singapore’s approach to online misinformation Introduction

In January 2018, Singapore’s ruling People’s Action Party of falsehoods and to prosecute those responsible – including the technology (PAP) appointed a parliamentary committee to investigate companies that host alleged breaches. the spread of misinformation and deliberate falsehoods POFMA quickly superseded other laws online. In its report later that year, the committee found as the Government’s primary tool in that proliferation of online falsehoods posed a growing the war against online falsehoods. At the same time, detractors claim it also 1 threat to social cohesion, public in government became a new and highly effective institutions, democracy and national sovereignty. The report weapon to silence critics, suppress proposed a new multi-pronged offensive, focusing on the freedom of speech, and further entrench government power. roles of journalists, fact checkers, technology companies, In this chapter, we look at the evolution the Government and citizens. Among its proposals was of the Government’s approach to public education to help Singaporeans identify misinformation; online misinformation over the past training of journalists to enhance quality and accuracy; and decade, culminating in the enactment further regulation of technology companies. of POFMA, and how the Government’s approach fits in with Singapore’s unique historical, demographic and cultural characteristics. A timeline of But arguably the most consequential problem.2 This led to the introduction the implementation of “fake news” laws proposal was for new mechanisms to in April 2019 of the Protection from and policies is tabled at the conclusion enable swift government intervention Online Falsehoods and Manipulation of the chapter (Table 2.1). to provide the necessary “scope, Act (POFMA), providing sweeping new speed and adaptability” to combat the powers to ministers to call out instances

Background and context

As a city-state located in the heart of Southeast Asia, Singapore has The PAP’s continuous hold on power for a relatively small population of around 5.7 million, comprising diverse more than six decades has been marked 3 by a combination of authoritarian and cultures and ethnicities. Its multi-racial and multi-faith demography paternalistic themes, with frequent is central to Singapore’s identity – and to the policy approaches of the allusions to “the public interest” being ruling party, which has been in power continuously since 1959. used to justify government determinations – both in public statements and the text of legislation. Hence, restrictive anti- The importance of maintaining especially those that establish more misinformation measures, including social cohesion amid Singapore’s intrusive governmental powers.4 This POFMA, have been accompanied by “fragile racial and religious harmony” theme has been continued when it references to “right things”5 and “what is often invoked by politicians when comes to measures to tackle online is in the public interest,” as determined introducing and justifying new laws – misinformation. by the Government.6

1 Ministry of Communications and Information and Ministry of Law, Deliberate and Online Falsehoods: Challenges and Implications (Singapore: Parliament of Singapore, 2018), https://www.nas.gov.sg/archivesonline/government_records/Flipviewer/grid_publish/6/6797717d-f25b-11e7-bafc-001a4a5ba61b 06012018Misc.10of2018/web/html5/ index.html?launchlogo=tablet/GovernmentRecords_brandingLogo_.png&pn=1 2 Parliament of Singapore, Report of the Select Committee on Deliberate Online Falsehoods – Causes, Consequences and Countermeasures, (Singapore: Parliament of Singapore, 2018), 164. https://sprs.parl.gov.sg/selectcommittee/selectcommittee/download?id=1&type=subReport 3 74% of Singaporeans identify as Chinese, while Malays (13%), Indians (9%) and those nominating ‘Other’ (3%) are the other notable ethnic groups that live in Singapore (Department of Statistics Singapore 2020). Religion provides even greater diversity as a marker, with Buddhists (34%), Christians (18%), Muslims (14%), Taoists (11%), Hindus (5%) and non-religious people (16%) representing significant proportions of the population. SBS, “Cultural Atlas”, SBS (2016), https://culturalatlas.sbs.com.au/ singaporean-culture/singaporean-culture-greetings#singaporean-culture-greetings 4 Howard Lee and Terence Lee, “From contempt of court to fake news: public legitimisation and governance in mediated Singapore,” Media International Australia 173, no. 1 (2019): 83-84; Alex Au Waipang, “Singapore bloggers wary of news site licence scheme,” Committee to Protect Journalists, published June 4, 2013, https://cpj.org/2013/06/ singapores-news-site-license-plan-raises-questions/ 5 Au Waipang, “Singapore bloggers wary of news site licence scheme.” 6 Parliament of Singapore, Protection from Online Falsehoods and Manipulation Act Parliament of Singapore, 2019, https://sso.agc.gov.sg/Acts-Supp/18-2019/ Published/20190625?DocDate=20190625

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 15 The Government’s approach to tackling misinformation

Since its introduction in 2019, the Protection from Online Falsehoods as justification for a heavy-handed and Manipulation Act (POFMA) has been the primary tool employed approach to online misinformation. And in Singapore, as in Indonesia, critics by Singapore’s Government in the fight against online misinformation. have highlighted the inherent conflicts of The Government has used the law’s extensive coercive powers to interest in laws that give the Government force the removal or amendment of online content it deems to be extensive powers and discretion to false or misleading – including multiple instances of false content decide what constitutes misinformation.8 about COVID-19. Another notable element of the Singapore Government’s approach to online misinformation, including Controversially, the Government has In this respect its approach has parallels since the advent of POFMA, has been also frequently targeted online political with that of the Indonesian Government, an emphasis on the need for speed in content, drawing accusations from rival which has also been accused of using responding to and prosecuting cases. politicians, journalists, academics and anti-misinformation laws to silence its The imperative was described by one technology companies that it is using political opponents in the lead-up to parliamentarian in these terms: “A lie can the Act as a weapon not just against national elections. As in Indonesia, the travel halfway around the world while misinformation but to silence political Government of Singapore has also cited the truth is still putting on its shoes.”9 opponents and further restrict freedom the need to maintain harmony between A potential casualty of the high-speed of speech and the media.7 the nation’s diverse ethnic groups approach, however, can be fairness – and the judiciary’s ability to adequately adjudicate the merits of each case.

Laws and regulations

Controversy surrounding the Singapore Government’s approach to country, “Demon-cratic Singapore.” tackling online falsehoods and misinformation long pre-dates the The site makes clear that all characters, parties, etc. are fictional. He was charged introduction of POFMA. Several other laws and regulations have been with sedition because of two cartoons, utilised over the past decade or so – and continue to be used – to one about race, claiming that Singapore’s clamp down on alleged online misinformation and other activities “Malay population was declining” – deemed by the Government to be against the public interest. And something he asserted was “a fact.”12 controversy has often followed. In another recent case, the Government pursued charges against The Real Singapore, an independent online news site that came under scrutiny for to be accused of inciting contempt of publishing alleged falsehoods relating the Government or ill-will among the to ethnic groups in Singapore. The case The has been used citizenry under the Sedition Act, even centred on an article about an alleged a number of times to prosecute the when they have claimed a defence dispute between a Filipino family and authors of online statements to which of truth.11 An example of this was the members of the Tamil community. The the Government has objected. Critics say charging of cartoonist Leslie Chew with news site was judged to have published the absence of a definition of sedition in sedition in 2013 for cartoons published “deliberately fabricated articles”13 that the law10 leaves wide scope for people on a Facebook page about a fictitious

7 Howard Lee and Terence Lee, “From contempt of court to fake news: public legitimisation and governance in mediated Singapore,; Alex Au Waipang, “Singapore bloggers wary of news site licence scheme,” Committee to Protect Journalists. 8 Karishma Vaswani, “Concern over Singapore’s anti-fake news law,” BBC News, 4 April 2019, https://www.bbc.com/news/business-47782470 9 “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Parliament of Singapore, 8 May 2019, 94, Part 2, https://sprs.parl.gov.sg/search/ sprs3topic?reportid=bill-367 10 Human Rights Watch, Kill the Chicken to Scare the Monkeys: Suppression of Free Expression and Assembly in Singapore Human Rights Watch, 2017, 43, https://www.hrw.org/sites/default/files/report_pdf/singapore1217_web.pdf 11 Ibid., 45. 12 Human Rights Watch, Kill the Chicken to Scare the Monkeys, 47-48. 13 “Singaporean authorities shut down The Real Singapore news website,” Committee to Protect Journalists, published May 5, 2015, https://cpj.org/2015/05/singaporean- authorities-shut-down-the-real-singapo/.

16 La Trobe University undermined national harmony14 – a said the new rules aim to “to protect the with self-censoring widely observed common justification for Sedition Act interest of the ordinary Singaporean” and practised,32 and Government- prosecutions.15 The site was shut down, and “to make sure they read the right initiated defamation actions invariably and its editors jailed for 8-10 months.16 things.”23 One of its stated purposes is to succeeding.33 Prime Minister Lee Hsien prohibit “content that undermines racial Loong has invoked the Act several times, or religious harmony.”24 This aim was including a defamation suit against The Online News Licensing invoked against The Real Singapore for Online Citizen (TOC) editor Terry Xu over Scheme an alleged breach of content standards, a story detailing a family feud between leading to its shutdown order,25 him and his siblings, which was before The 2013 Online News Licencing concurrently with its prosecution under the court at the time of writing.34 PM Scheme, which extends content the Sedition Act (see above). Lee has also filed a lawsuit against a standards established under the financial blogger for sharing a story Critics have claimed that the new Broadcasting Act and Internet Code of about his alleged involvement in money 17 licencing framework was established Practice, has also been used against laundering.35 In a separate case, PM in response to the PAP’s historically perceived online falsehoods. Under Lee also unsuccessfully demanded that poor election result in 2011, which the framework, news sites meeting Facebook remove an Australian-based was attributed in part to the increasing certain criteria – such as having at least blogger’s “false and malicious” post.36 prominence of online news and .26 50,000 unique Singaporean visitors While the law’s criminal provisions are each month – are required to pay used rarely,37 one writer for The Online $S50,000 annually for a licence, and Defamation Act Citizen is currently facing trial over an can be ordered to remove prohibited article alleging government corruption.38 content18 within 24 hours, or be shut The Defamation Act,27 covering slander, down.19 Licenced websites are banned libel and other falsehoods, has been from receiving financial support from “the Government’s most powerful Protection from foreign sources.20 And in response to non-criminal weapon” against its Harassment Act concerns the framework would constrain opponents, according to the international individuals’ freedom of speech online,21 organisation Human Rights Watch.28 It The Protection from Harassment Act the Government stated that personal claims the law has been used to sue, (POHA), introduced in 2014, outlaws blogs would not be covered “so long as bankrupt and intimidate government and provides remedies against various they do not morph into news sites.”22 critics. Damages owed by those found actions, including “civil remedies related… to .”39 Like other laws and measures, the guilty of defaming the Government 29 licensing framework does not provide have been as high as S$150,000, and Section 15 of the Act allows individuals specific definitions of fake news criminal defamation charges can lead to to apply for court orders against false 30 or misinformation. However, the jail time of up to two years. statements published about them.40 Government has drawn on familiar The large penalties under the Defamation An attempt by the Government to themes to justify its use. At the time Act are claimed to have served as use the Act against critical reporting of its introduction, the Minister for a highly effective deterrent against of a ministry41 was disallowed by the Communications and Information speaking ill of the Government,31 High Court on the grounds that the

14 Pearl Lee, “TRS co-founder Yang Kaiheng jailed 8 months for sedition,” The Straits Times, , 2016. https://www.straitstimes.com/singapore/courts-crime/trs-co- founder-yang-kaiheng-jailed-8-months-for-sedition 15 Human Rights Watch, Kill the Chicken to Scare the Monkeys, 46. 16 Lee, “TRS co-founder Yang Kaiheng jailed 8 months for sedition.” 17 Infocomm Media Development Authority, Internet code of practice (Infocomm Media Development Authority, 1997), https://www.imda.gov.sg/-/media/Imda/Files/ Regulations-and-Licensing/Regulations/Codes-of-Practice/Codes-of-Practice-Media/PoliciesandContentGuidelinesInternetInterneCodeOfPractice.pdf 18 As outlined in the Internet Code of Practice 1997, which covers all licenced ‘Internet Service Providers and Internet Content Providers’ – ‘Prohibited material is material that is objectionable on the grounds of public interest, public morality, public order, public security, national harmony, or is otherwise prohibited by applicable Singapore laws’ (pp. 1-2) 19 Human Rights Watch, Kill the Chicken to Scare the Monkeys, p. 49. 20 Paul Meyer, “Singapore’s First Election Under the Fake News Law,” The Diplomat, July 7, 2020. https://thediplomat.com/2020/07/singapores-first-election-under-the-fake- news-law/ 21 Au Waipang, “Singapore bloggers wary of news site licence scheme.” 22 “What is the licensing framework for online news sites all about?,” Singapore Government, published June 18, 2013, https://www.gov.sg/article/what-is-the-licensing- framework-for-online-news-sites-all-about 23 Au Waipang, “Singapore bloggers wary of news site licence scheme.” 24 Infocomm Media Development Authority. Fact Sheet - Online news sites to be placed on a more consistent licensing framework as traditional news platforms (Infocomm Media Development Authority, 2019), https://www.imda.gov.sg/news-and-events/Media-Room/archived/mda/Media-Releases/2013/fact-sheet--online-news-sites-to-be-placed-on- a-more-consistent-licensing-framework-as-traditional-news-platforms 25 Valerie Koh, “Govt orders shutdown of The Real Singapore,” Today Online, May 4, 2015, https://www.todayonline.com/singapore/mda-suspends-licence-socio-political-website- real-singapore 26 Meyer, “Singapore’s First Election Under the Fake News Law.” 27 Parliament of Singapore, Defamation Act (Singapore) (Parliament of Singapore, 1965), https://sso.agc.gov.sg/Act/DA1957 28 Human Rights Watch, Kill the Chicken to Scare the Monkeys, 96. 29 Lydia Lam, “PM Lee sues blogger Leong Sze Hian for defamation over sharing of article,” Channel News Asia, December 7, 2018. https://www.channelnewsasia.com/news/ singapore/pm-lee-sues-blogger-leong-sze-hian-defamation-libellous-article-10999710 30 Parliament of Singapore, Penal Code (Singapore) (Parliament of Singapore, 1972), Section 499, https://sso.agc.gov.sg/Act/PC1871?ProvIds=pr499- 31 Ibid. 32 Ibid., 6 33 Ibid., 96. 34 Amy Gunia, “Singapore’s prime minister threatens website editor with libel,” Time, September 2, 2019, https://time.com/5666835/singapore-prime-minister-lee-hsien-loong- libel/; https://www.straitstimes.com/singapore/courts-crime/pm-lee-hsien-loongs-defamation-suit-against-tocs-terry-xu-begins-monday 35 Fathin Ungku, “Singapore PM files defamation suit against blogger who shared article on Facebook,” Reuters, December 5, 2018, https://www.reuters.com/article/us- singapore-politics-malaysia-scandal/singapore-pm-files-defamation-suit-against-blogger-who-shared-article-on-facebook-idUSKBN1O414L 36 Belmont Lay. “Facebook declines S’pore govt request to take down States Times Review post linking PM Lee with 1MDB,” Mothership, November 10, 2018, https://mothership. sg/2018/11/facebook-turns-down-singapore-request/ 37 Human Rights Watch, Kill the Chicken to Scare the Monkeys, 118. 38 Lam, “PM Lee sues blogger Leong Sze Hian for defamation over sharing of article.” 39 Parliament of Singapore, Protection from Online Harassment Act (Singapore) (Parliament of Singapore, 2014), https://sso.agc.gov.sg/Act/PHA2014. 40 Ibid., section 15. 41 Human Rights Watch, Kill the Chicken to Scare the Monkeys, 100

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 17 Government is not a person, and that the upon some previous legislative it would consider to be a representation reporting was not false.42 This helped measures, and is the first law created of fact.”51 A statement is deemed to be set the scene for POFMA a few years specifically to target the issue. false “if it is false or misleading, whether later, with the Ministry of Law stating wholly or in part, and whether on its own It bestows extensive powers on the in response to the High Court’s POHA or in the context in which it appears.”52 Government and its ministers. Under Part ruling: “The Government will study the Controversially, however, the Act does 3, any minister can declare information judgment, and consider what further not seek to define “misleading.” to be “false or misleading” and force its steps it should take to correct the publisher to apply a correction notice Penalties for breaches of the Act are deliberate spreading of falsehoods.”43 or remove the material.46 Ministers significant. Individuals face fines of Under amendments passed in 2020, can use this power if they believe up to S$50,000 and/or jail terms of up entities including companies can now publication of the statement is against to five years. Non-individuals (such as use POHA to obtain remedies against the “public interest.”47 Both criteria, the internet companies) face up to S$1 online falsehoods including stop, charge of “false and misleading” and million in fines plus $100,000 per day correction and disabling orders,44 similar “against the public interest,” must be of non-compliance per section 27 of to remedies available under POFMA. met. Under Part 4, ministers can order POFMA. These penalties also apply to The Government and its agencies – to internet intermediaries (such as social fake online accounts or bots being used the dismay of critics – remain exempt media platforms and internet service to spread online falsehoods.53 Internet from prosecution under the law.45 providers) to remove a false article of companies that decline to disable access information from their web locations.48 to locations of false statements face And Part 5 allows ministers to instruct fines of up to S$20,000 per day up to a Protection from internet intermediaries to block access total of $500,000.54 to declared online locations if they have Online Falsehoods Legal experts have suggested that received at least three POFMA orders.49 POFMA allows the Government to and Manipulation Act This can be three pieces of content effectively force internet intermediaries that are subject to the same order; it (POFMA) to track their users’ viewing habits does not necessarily need to be three POFMA is the primary legislative and keep records of this information,55 separate orders.50 instrument now used by the Singapore adding to concerns about privacy on Government to tackle online falsehoods. The Act defines a “statement of fact” as the internet.56 The Act, which became law in 2019, “a statement which a reasonable person incorporates and significantly expands seeing, hearing or otherwise perceiving

42 Ibid., 101 43 Ibid., 102-103. 44 Parliament of Singapore, Protection from Online Harassment Act (Singapore), section 15. 45 Jalelah Abu Baker, “Exemption of Government from being sued under POHA amendments a “glaring ”: Pritam Singh,” Channel News Asia, May 7, 2019, https://www.channelnewsasia.com/news/singapore/exemption-government-sued-under-poha-amendments-pritam-singh-11511118. 46 Parliament of Singapore, Protection from Online Falsehoods and Manipulation Act , section 7. 47 See Protection from Online Falsehoods and Manipulation Act, section 4, for definition and section 7; Section 7 defines statements against the public interest once they are ‘prejudicial to Singapore’s security’, ‘public health, safety, tranquillity or finances’, its relations with other countries, they might influence the outcome of elections, ‘Incite feelings of enmity, hatred or ill-will between different groups of persons’, or ‘Diminish public confidence in the government.’ 48 Parliament of Singapore, Protection from Online Falsehoods and Manipulation Act, Part 4. 49 Ibid., section 32. 50 Ibid., section 40 51 Ibid., section 2. 52 Ibid. 53 “Singapore Fake News Laws: Guide to POFMA (Protection from Online Falsehoods and Manipulation Act),” Singapore Legal Advice, published February 3, 2020, https://singaporelegaladvice.com/law-articles/singapore-fake-news-protection-online-falsehoods-manipulation/ 54 Shibani Mahtani, “Singapore introduced tough laws against fake news. Coronavirus has put them to the test,” Washington Post, March 16, 2020, https://www.washingtonpost. com/world/asia_pacific/exploiting-fake-news-laws-singapore-targets-tech-firms-over-coronavirus-falsehoods/2020/03/16/a49d6aa0-5f8f-11ea-ac50-18701e14e06d_story. html. 55 Jennifer Daskal, “This “fake news” law threatens free speech. But it doesn’t stop there,” New York Times, May 30, 2019, https://www.nytimes.com/2019/05/30/opinion/hate- speech-law-singapore.html. 56 Ibid.

18 La Trobe University The use of POFMA

Critics of the Box 2.1 Alex Tan Case Study fact checking – they clearly make up stuff.”67 The PAP argue that Tan has Alex Tan is a Singapore-Australian Government “repeatedly conveyed falsehoods and blogger and former opposition not complied with any of the POFMA The first use of POFMA came in candidate who is an outspoken critic directions”.68 November 2019 in the form of a of the Singaporean government. He correction order against Progress has lived in Australia in recent years The POFMA orders targeting Tan Singapore Party member Brad Bowyer, yet maintained a consistent online include correction direction orders, in response to a Facebook post implying presence in Singapore, writing at one orders for Facebook to disable access government control of investments by time or another for several websites to his various pages, and deeming his government-linked companies.57 The including The Real Singapore, which Facebook pages a ‘Declared Online Minister of Finance said the post was was forced to shut down by the Location’. Tan is the first to receive a false and “undermines public trust Singapore Media Development DOL. Under a DOL, Tan’s Facebook in government.”58 Since then, the law Authority.64 pages are required to declare has been used multiple times against themselves as such, notifying visitors Central to why he has written for outspoken critics of the Government, of their “history of communicating so many online publications is that including New Naratif59 and Lawyers for falsehoods.”69 Tan is also not able the Singaporean government has Liberty (LFL – a Malaysian NGO), which to derive financial benefit from been active in closing them – and posted allegations that the Singapore these pages. their associated Facebook pages. In Prisons Service committed human rights response, Tan has repeatedly started As a result of these disabling abuses during executions.60 LFL refused a “new page from zero”, as he puts it.65 measures, Tan’s once substantial to obey a POFMA order “a correction Facebook followership of almost direction” to juxtapose the alleged As of July 2020, he has been 60,000 has dwindled to about 3000.70 falsehoods with “the facts” according to implicated in more POFMA orders the government. This refusal prompted than anyone else with 31 of 71 Tan is based in Australia and his the Government to block access to total orders, 14 of which have capacity to ignore POFMA orders its website.61 been issued to him directly. These arguably demonstrates POFMA’s include orders for false information inability to counter claims of foreign Khairulanwar Zaini found that “most relating to misquotes of government disinformation. of these orders have been directed ministers, the availability of face at opposition politicians or anti- Nonetheless, Tan’s case shows masks, COVID-19 outbreaks and establishment critics.”62 According to what tools the government has arrests and POFMA criticism. While the online website POFMA’ed,63 which at its disposal to combat alleged he contends that he has “achieved regularly reports about uses of POFMA misinformation: First, to demand that what I intended which is exposing orders, 31 out of 71 POFMA orders relate citizens adhere to its correction and the hypocrisy of the Singaporean to ruling party critic Alex Tan and his takedown orders and, failing that, to government”, others observe that various webpages. See Box 2.1. demand that the online platforms his websites are “known more for respond. Facebook was compelled the vehemence rather than the to comply with the Government’s accuracy of [their] frequent diatribes demand to revoke local access to 66 against the PAP government.” His Tan’s website, despite voicing concern websites have been labelled “very for the precedent POFMA sets about anti-government in rhetoric”, with “no limiting freedom of expression in Singapore.71

57 “Fake news law invoked for the first time over Facebook post,”The Straits Times, November 26, 2019, https://www.straitstimes.com/politics/fake-news-law-invoked-for-the- first-time-over-facebook-post. 58 Ibid. 59 “Thum Ping Tjin, New Naratif to challenge POFMA notice despite compliance.” 60 “Singapore: Use of Anti-Fake News Law reveals institutional framework that undermines the promotion and protection of human rights,” FORUM-ASIA, published February 11, 2020, https://www.forum-asia.org/?p=30931. 61 Khairulanwar Zaini, “Singapore in 2019: In holding pattern,” Southeast Asian Affairs, no. 1 (2020): 305 62 Ibid. 63 Teo Kai Xiang, “About,” POFMA’ed, accessed November 13, 2020, http://pofmaed.com/. 64 Other sites Tan has written for were: Temasek Review News, Straits Times Review (later changing the name to States Times Review), Singapore States Times, National Times Singapore, among others. 65 Alex Tan quoted in Wilson, Cameron, ‘This man’s Facebook page was blocked for spreading false information about the coronavirus’, BuzzFeed News, 19 February 2020, URL: https://www.buzzfeed.com/cameronwilson/singapore-facebook-fake-news-law-alex-tan-coronavirus 66 Zaini, Khairulanwar, ‘Singapore in 2019: In Holding Pattern’, Southeast Asian Affairs, 2020(1), p. 305. 67 Kirsten Han, Interview #6 68 Minister for Communications and Information S. Irawan quoted in ‘This man’s Facebook page was blocked for spreading false information about the coronavirus’. 69 Ministry of Communications and Information, ‘The “Singapore States Times” and Alex Tan’s Facebook pages declared as declared online locations under POFMA’ [Media release], 6 May 2020, URL: https://www.pofmaoffice.gov.sg/documents/media-releases/2020/May/pofma-pr-mci-06may2020-01.pdf 70 Xiang, Teo Kai, ‘About’, POFMA’ed [website], 2020. 71 Facebook spokesperson quoted in Tham, Yuen-C, ‘Facebook blocks access in Singapore to States Times Review page for breaching Pofma’, The Straits Times, 18 February 2020, URL: https://www.straitstimes.com/politics/facebook-blocks-access-to-states-times-review-page

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 19 Independent media have been the which the SDP did not challenge.75 In his can theoretically be blanketed under target of 33 out of 71 POFMA orders. ruling, the judge said the burden of proof a POFMA order for the entirety of the Apart from Tan, who is included in this was on the Government to prove the election period. category, other independent media falsity of information. Data on the use of POFMA during the operators to have faced POFMA orders In a separate appeal in February 2020, July 2020 election campaign shows include Kirsten Han and PJ Thum as The Online Citizen (TOC) was charged an unusually high incidence of orders well as The Online Citizen, AB-TC City with communicating false statements against political opponents of the News and Singapore. from a press release by Lawyers For Government. Of 18 POFMA orders issued Singapore Press Holdings (SPH), which Liberty alleging abuses during executions during the campaign period, 12 (67 owns The Straits Times, is a publicly by the Singapore Prisons Service. TOC’s per cent) targeted opposition political listed company and it has also been appeal failed on the grounds that the statements, including five against issued a POFMA order, however this initial statement was not held to be true.76 websites that had posted claims by the related to a post made by an anonymous In this case, however, in direct contrast leader of the SDP. In 55 previous uses of user on its “HardwareZone” Forum to the SDP case, the judge determined POFMA, only about 20 per cent fitted the and not the company itself. Channel that the burden of proof fell on TOC.77 same criteria. NewsAsia, belonging to the state-owned The Court of Appeal is set to resolve this Temasek Holdings’ Mediacorp, also POFMA has also had a controversial matter shortly. received a POFMA order during the impact on online political advertising, 2020 election. Despite claims of its efficiency, the with a provision requiring internet appeals process has also been criticised intermediaries to keep records of POFMA’ed data also reveals that 34 for its combative nature, and for the all online political advertisements, orders have targeted political groups high cost to litigants.78 Contested prompting Google to remove such or figures (including Alex Tan). Of the POFMA orders remain in effect during content altogether in Singapore.82 One 27 cases giving rise to the 71 POFMA the appeals process, confirming initial critic claimed this showed the provision orders,72 20 involved statements about concerns that the law “effectively may “worsen the asymmetric playing the Government, one of its agencies reverses the traditional presumption field between the incumbent party and or a government-linked company. The that an accused is innocent until the Opposition.”83 three cases that did not involve criticism proven guilty.”79 of the Government all related to false COVID-19 reports. One digital platform COVID-19 expert contrasted the figures with the Elections Government’s assurances that “they A high proportion of POFMA orders have would stay away from opinions and One of POFMA’s most controversial involved allegedly false or misleading political speech.”73 aspects is that, through built-in delays statements about COVID-19 (35 out in the appeals process, the Government of 71 orders up to 11 July 2020). Data has effective power to silence critics from the watchdog website POFMA’ed Appeals process during election campaigns – which shows many of these cases concern last for a minimum of nine days. During claims critical of Singapore’s handling of Concerns over the effectiveness of elections, powers to exercise POFMA the crisis, as well as alleged falsehoods POFMA’s appeal mechanisms have been passes to the civil service. The first point relating to certain groups in society, underlined by two recent cases (detailed of appeal against a POFMA order is the including migrant workers.84 below) in which judges disagreed over relevant Minister and then, if that fails, interpretations of the law.74 The potentially life-threatening spread the High Court. But the best appellants of misinformation during the pandemic In January 2020, the Singapore can hope for is to have their case heard has allowed the Government to claim Democratic Party was charged with as early as nine days after a challenge vindication for its introduction of communicating false statements about is first brought.80 In other words, barring stringent misinformation laws.85 growing local retrenchments. A High a change of heart from the Minister, Court judge dismissed the SDP’s appeal a POFMA order applies for at least against the charge on the grounds that nine days – the same period for which the statements were false “in the face conventionally politicians campaign in of statistical evidence against them” – general elections.81 Hence, information

72 While there were 72 total uses of POFMA (as of July 11), many of these related to the same case; for example, different publishers of the same material were handed notices, or the same publisher was handed separate types of notices. 73 Interview 1, interviewed by Andrea Carson, 7 August 2020 74 N. Tang, “POFMA: A song of fact and opinion,” The Online Citizen, published February 4, 2020, https://www.onlinecitizenasia.com/2020/02/04/pofma-a-song-of-fact-and- opinion/ 75 Justice Ang Cheng Hock quoted in Lydia Lam, “Judge dismisses SDP’s POFMA challenge, says statements were false in face of statistical evidence,” Channel News Asia, February 5, 2020, https://www.channelnewsasia.com/news/singapore/judge-dismisses-sdp-s-pofma-challenge-says-statements-were-false-12394932?cid=h3_referral_ inarticlelinks_24082018_cna. 76 Nicole Chang, “Judge dismisses The Online Citizen’s POFMA challenge,” Channel News Asia, February 19, 2020, https://www.channelnewsasia.com/news/singapore/the- online-citizen-toc-pofma-challenge-appeal-dismiss-12449826 77 Ibid. 78 Han, “Big Brother’s regional ripple effect,” 68; Foreign Correspondents Association of Singapore, “The Foreign Correspondents Association of Singapore is deeply…” 79 Kaye, Mandate of the Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression, 6 80 Yuen-C Tham, “Fake news Act: Speed, cost of appeals process set out,” The Straits Times, October 2, 2019, https://www.straitstimes.com/politics/fake-news-act-speed-cost- of-appeals-process-set-out 81 “Parliamentary elections,” Elections Department Singapore, accessed November 13, 2020, https://www.eld.gov.sg/elections_parliamentary.html 82 Meyer, “Singapore’s First Election Under the Fake News Law.” 83 Ibid. 84 Ministry of Health and Ministry of Manpower, “Joint MOM-MOH statement on the issuance of correction directions under the Protection from Online Falsehoods and Manipulation Act (POFMA) against National University of Singapore Society (NUSS), The Online Citizen Asia (TOC), CNA and New Naratif,” POFMA Office, published July 5, 2020, https://www.pofmaoffice.gov.sg/documents/media-releases/2020/July/pofma-pr-mom-05jul2020-01.pdf 85 Yuen-C Tham, “Falsehoods on coronavirus show why Pofma is necessary,” The Straits Times, February 4, 2020, https://www.straitstimes.com/politics/falsehoods-on- coronavirus-show-why-pofma-is-necessary

20 La Trobe University Reactions to POFMA

Political Despite its vocal opposition to POFMA, potential conflation of public interest the Workers’ Party has not yet been with political gain,100 and the lack Singapore’s main opposition party, subjected to it. It is not entirely clear why of a definition for “misleading.” The the Workers’ Party (WP), has been the the WP has escaped sanction but it may ambiguities have caused uncertainty most prominent parliamentary critic of be due to its leader Pritam Singh’s public among Singaporeans about what POFMA, with some MPs accusing the commentary about open society when they can and can’t post, and fear PAP of creating and using POFMA to he said: “we believe that the national that the door is wide open for the crush dissent and strengthen its hold discourse should be conducted on the abuse of anti-misinformation laws for on power.86 One former MP, Low Thia basis of objective facts.” While there is political purposes.101 Khiang, said POFMA had tightened a still “a lot of room for opinion,” he said Several Nominated MPs – non-partisan “dictatorial government’s…hold onto public commentators risked the charge members appointed by the President absolute power.”87 He feared the ruling of falsity under POFMA as it currently – offered less strident assessments, party would “manipulate opinions exists.95 Another possible explanation for neither fully opposing nor fully and spread falsehoods in order to win the lack of POFMA orders against the WP supporting POFMA. Three Nominated elections.”88 Another MP raised the is that the PAP has other, softer means MPs102 proposed amendments to spectre of a “chilling” of free speech, of censure. This was evident prior to the ensure that opinion, satire and other particularly “expressions of views 2020 election when prominent members forms of non-factual commentary that run counter to the government of the PAP questioned Singh and his were excluded.103 narrative.”89 party’s loyalty to Singapore.96 During the Law and Home Affairs Minister The Workers’ Party highlighted the election campaign, a police investigation Kasiviswanathan Shanmugam Act’s inherent conflicts of interest,90 and sanctioned by the Attorney-General’s known as K Shanmugam, who led expressed doubt about whether the office was opened into two Facebook the Government’s side of the POFMA Government could make fair judgements posts published by a WP candidate, said to 97 debate, dismissed the criticisms, arguing in cases where it was the subject of promote racial and religious enmity. One the law narrowed rather than widened the alleged falsehood.91 It said the of the posts was more than two years old. government powers, and that it provided Government could exploit the Act’s Why the PAP has not used POFMA “greater judicial oversight.”104 He also lack of a definition of “misleading” to against the WP (but does so against questioned the notion of universal free defend its policies and public image.92 smaller opposition parties) may be to speech before POFMA was enacted, It also criticised the “onerous” appeals avoid giving too much attention to its stating: “There’s been no engagement on process against POFMA orders, leading opposition, whose popularity the key issue on why there should be free particularly as the burden of proof lay reached new heights during the speech in this area. What speech are you with the appellant,93 and as contested 2020 election.98 protecting? Why are these things entitled online statements could be corrected Critics have focused on ambiguities in the to free speech?”105 or removed even before the appeals Act including its lack of differentiation process had run its course.94 between facts and opinions,99 the

86 Low Thia Khiang quoted in Vernon Lee, “Fake news law is ploy by government to hold on to “absolute power”: Workers’ Party MP Low Thia Khiang,” Yahoo News Singapore, May 8, 2019, https://sg.news.yahoo.com/fake-news-law-is-ploy-by-government-to-hold-absolute-power-low-thia-khiang-155910029.html. 87 Ibid. 88 Ibid. 89 MP Perera quoted in “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Parliament of Singapore, published May 8, 2019, Vol. 94 Part 1, https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-366. 90 Pritam Singh quoted in Fann Sim, “Workers’ Party opposes online falsehoods Bill, says Pritam Singh,” Channel News Asia, May 7, 2019, https://www.channelnewsasia.com/ news/singapore/online-falsehoods-workers-party-opposes-bill-pritam-singh-11511450. 91 Muhamad Faisal Bin Abdul Manap quoted in “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill’, Parliament of Singapore, published May 8, 2019, Vol. 94, Part 1, https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-366. 92 Ibid. 93 Sylvia Lim quoted in “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Parliament of Singapore, published May 8, 2019, Vol. 94 Part 1, https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-366. 94 Ibid, 95 Nicholas Yong, “GE2020: Workers’ Party is not ‘pulling its punches’ in spite of POFMA, says Pritam Singh,” Yahoo News Singapore, July 7, 2020, https:// sg.news.yahoo.com/ge-2020-workers-party-is-not-pulling-its-punches-in-spite-of-pofma-says-pritam-singh-063326754.html?guccounter=1&guce_ referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAAIrOu4fGibE2qg0_wGysQZWxGlrMs_D4ufdDdO6OzEkNrRDGDD45KSSR9ldVhmsr66-Mvrh8dfWWI KbYwHBHci5KarXgRua1iR2bMVxuLWxI9cNhQYQxFivabsbiKXpcqH0JFTv89ncOn32AUssYPlQuGdUBNVyXjXRcmXF-A-5H 96 Jalelah Abu Baker, “Tan Wu Meng’s opinion piece on Pritam Singh was ‘serious’ and ‘thoughtful’: Shanmugam,” Channel News Asia, June 21, 2020, https://www. channelnewsasia.com/news/singapore/tan-wu-meng-pritam-singh-aflian-shanmugam-12856406 97 Rei Kurohi, “Singapore GE2020: WP’s Raeesah Khan apologises for posts which allegedly promoted enmity between different groups,” The Straits Times, updated September 1, 2020, https://www.straitstimes.com/politics/singapore-ge2020-wps-raeesah-khan-apologises-for-posts-which-allegedly-promoted-enmity 98 “Singapore ruling PAP party wins elections, but support falls,” BBC, July 10, 2020, https://www.bbc.com/news/world-asia-53358650 99 MP Khiang quoted in “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Parliament of Singapore, published May 8, 2019, Vol. 94 , Part 2, https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-367. 100 NMP Prof. Walter Theseira quoted in “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill’, Parliament of Singapore, published May 8, 2019, Vol. 94, Part 1, https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-366. 101 For political discussion of this lack of definition, see Low Thia Khiang quoted in Singapore “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Parliament of Singapore, published May 8, 2019, Vol. 94 , Part 2, https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-367; Human rights groups have criticised POFMA’s ambiguity, as outlined in ‘Reactions’ below. 102 ‘NMPs contribute independent and non-partisan views in Parliament’ (see “Members of parliament,” Parliament of Singapore, accessed November 13, 2020, https://www.parliament.gov.sg/about-us/structure/members-of-parliament.). 103 Benjamin Joshua Ong, “Symposium on POFMA: Parliamentary debates about POFMA – Hansard beyond statutory interpretation?,” Research Collection School of Law (November 2019): 3, https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=4926&context=sol_research 104 “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Vol.94, Part 2. 105 Ibid.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 21 Media, academic, digital Others have attacked POFMA more about POFMA’s impact on freedom broadly as inherently anti-democratic of speech and innovation. Facebook platforms and civic and a threat to freedom of speech cited the broadening of Singapore’s groups and freedom of the media. Media powers to “compel us to remove content professionals and human rights they deem to be false.”121 Google POFMA has attracted widespread organisations expressed concern over expressed broader fears for “innovation commentary and criticism in Singapore, the “chilling effect” that stringent policing and growth of the digital information much of it in line with the main of speech could elicit.112 POFMA orders, ecosystem,”122 while Twitter raised objections raised by opposition MPs. they argued, would discourage public concerns about freedom of expression Journalists and civil rights groups fear discourse,113 drive news organisations and regulatory overreach.123 the Government’s increased powers to withhold important stories,114 and 106 Human rights groups and NGOs have may go unchecked. Some have cited ultimately result in self-censorship.115 the broad and opaque definitions of argued for more media literacy, rather breaches, and the potential for any form Some have predicted that, given than more restrictions. Some have of government criticism to be judged Singapore’s international reputation as an questioned the Government’s motives false or misleading in service to the “efficient, highly-respected government,” with POFMA, and advocated for greater “public interest,”107 with intimidation of its regulatory approach to fake news may powers of arbitration to be conferred on independent media and government legitimise the introduction of similar laws the courts. 116 108 in neighbouring countries. critics resulting. Academics fear that POFMA will have The exemption of the Government and While there is general acknowledgment a detrimental impact on academic its agencies from prosecution under of the need for action against online discourse and research in Singapore. POFMA109 has, according to some misinformation, alternative strategies The advocacy group Academics 117 commentators, created a “two-tiered” are proposed. Journalist Kirsten Han Against Disinformation stated that the society in which the Government argued the Government’s successful new restrictions may “compromise and its allies “can spread falsehoods response to COVID-19 “demonstrated Singapore’s notable efforts to develop with impunity” while its critics are that there are many ways we can itself into an internationally-recognised 124 heavily policed.110 This highlights the communicate, interact, and build trust hub for excellence in higher education.” Government’s conflicted position as an in the system without resorting to They also feared the law would be 118 arbiter of the Bill. POFMA.” Others suggested improving emulated in other countries. The group the public’s media literacy, which would also questioned the implicit notion in Critics pointed out ambiguities in the allow citizens to discern falsehoods the law that statements could always law, with distinctions between facts themselves,119 and having independent be objectively determined as false or and opinions, satire and other forms of bodies to adjudicate POFMA cases.120 misleading,125 saying interpretations of 111 commentary blurred. This could leave even generally agreed upon “facts” may Global digital technology companies with a journalist with a different interpretation vary greatly.126 of facts to the Government vulnerable to Asia-Pacific headquarters in Singapore prosecution under POFMA. have also expressed concern

106 Kirsten Han, “Big Brother’s regional ripple effect,” Index on Censorship 48, no. 2 (2019): 67–69.; “Thum Ping Tjin, New Naratif to challenge POFMA notice despite compliance,” Yahoo News Singapore, May 15, 2020. https://sg.news.yahoo.com/thum-ping-tjin-new-naratif-to-challenge-pofma-notice-despite-compliance-071938288.html 107 Han, “Big Brother’s regional ripple effect,” 67, 69 108 “Thum Ping Tjin, New Naratif to challenge POFMA notice despite compliance.” 109 POFMA, section 61, reads: ‘The Minister may, by order in the Gazette, exempt any person or class of persons from any provision of this Act.’ 110 ‘Joint statement regarding the Protection from Online Falsehoods and Manipulation Bill’ [Joint statement from arts and civil society organisations], https://theindependent.sg/ nearly-30-civil-society-arts-and-community-groups-express-concerns-over-draft-fake-news-law/ 111 Journalists, ‘Journalists call for withdrawal of Singapore’s “fake news” Bill’ [letter to Minister of Communications and Information S Iswaran], 24 April 2019. 112 Tessa Wong, “Singapore fake news law polices chats and online platforms,” BBC, May 9, 2019. https://www.bbc.com/news/world-asia-48196985. 113 Charles Santiago quoted in “Singapore: Amend the “fake news” bill,” ASEAN Parliamentarians for Human Rights, published May 7, 2019, https://aseanmp.org/2019/05/07/ singapore-fake-news-bill/?fbclid=IwAR114nescHaE8VQ5NDYlV8YxxTcKi_6ogBLXROZFR-8C5Uf5OnwCVjePQZ8 114 Jewel Stolarchuk, “Sarawak Report founder joins other prominent journalists in calling for the withdrawal of POFMA,” The Independent News and Media,, 25 April 2019, https://theindependent.sg/sarawak-report-founder-joins-other-prominent-journalists-in-calling-for-the-withdrawal-of-pofma-bill/ 115 Foreign Correspondents Association of Singapore, “The Foreign Correspondents Association of Singapore is deeply concerned about the Singapore government’s passing of a bill…”, Facebook, 9 May 2019, https://www.facebook.com/FCASG.org/posts/2358487210882553. 116 Han, “Big Brother’s regional ripple effect.” 117 Function 8, “Submission to the Parliamentary Committee on Deliberate Online Falsehoods – Causes, Consequences and Countermeasures,” Facebook, February 28, 2018, https://m.facebook.com/story.php?story_fbid=911516705691971&id=350013055175675.; Terry Xu, “TOC’s submission to Select Committee on Deliberate Online Falsehoods,” The Online Citizen, March 7, 2018, https://www.onlinecitizenasia.com/2018/03/07/tocs-submission-to-select-committee-on-deliberate-online-falsehoods/. 118 Kirsten Han, “Coronavirus: How effective is Singapore’s anti-fake news law?,” The News Lens, April 8, 2020, https://international.thenewslens.com/article/133521. 119 Xu, “TOC’s submission to Select Committee on Deliberate Online Falsehoods.”; Association of Women for Action and Research. No new restrictions needed: AWARE submission to Select Committee on “Deliberate Online Falsehoods (Association of Women for Action and Research, 2018), 1, http://d2t1lspzrjtif2.cloudfront.net/wp-content/ uploads/AWARE-submission-to-Select-Committee-on-Deliberate-Online-Falsehoods-1.pdf 120 Stephanie Neubronner, “Bill to protect from online falsehoods: Refinements needed,” RSIS Commentary 85, (April 30, 2019): n.p. 121 Simon Milner quoted in Fathin Ungku, “Facebook, rights groups hit out at Singapore’s fake news bill,” Reuters, April 1, 2019, https://www.reuters.com/article/us-singapore- politics-fakenews/facebook-rights-groups-hit-out-at-singapores-fake-news-bill-idUSKCN1RD279. 122 Aradhana Aravindan, “Google says Singapore’s fake news law could hamper innovation,” Reuters, May 9, 2019, https://www.reuters.com/article/us-singapore-politics-google- idUSKCN1SF08X. 123 Ibid. 124 “POFMA: Media release,” Academics Against Disinformation, published April 14, 2019, http://www.academia.sg/pofma-media-release/. 125 “POFMA: Letter to Education Minister,” Academics Against Misinformation, published April 11, 2019, http://www.academia.sg/pofma-letter/. 126 Ibid.

22 La Trobe University Others criticised the lack of academic Howard and Terence Lee observed consultation before the law was enacted, how the appearance of public debate and the Government “surreptitiously in the lead-up to the enactment of setting norms on what is appropriate and POFMA established its legal and inappropriate academic work.”127 public legitimacy.135 This careful and limited “incorporation of democratic However, some academics have elements” into the law-making process defended POFMA. Wei Yao and Kenny also legitimised the PAP’s right to Chng128 echoed Minister Shanmugam’s govern.136 By establishing an “us versus doubts about the inherent sanctity them” dynamic during the law-making of free speech. Others defended the process,137 they believed the Government Government’s central role in tackling had sought to portray POFMA’s falsehoods, saying courts were too detractors as being against the interests slow, and that relying on internet of the country. companies would be less democratic and less accountable than relying on the Government.129 International reactions 130 Ong defended the lack of definition of International human rights groups have “misleading” in POFMA as necessary, criticised POFMA as a threat to free saying it prevented the law “being evaded speech. The International Commission through selective statements of facts of Jurists assailed the law’s judicial that are individually true but collectively review mechanism,138 while Amnesty 131 paint a false picture.” Ong argued that International argued the PAP’s legislated the judiciary had historically acted as role as the arbiter of falsehoods would an effective check on the Government’s stifle the voices of its critics.139 The power. Similarly, Howe concluded United Nations Special Rapporteur on that judicial oversight of the Act was the promotion and protection of the right 132 “indeed, adequate.” to freedom of opinion and expression140 Professor of Journalism Cherian George also took issue with POFMA’s legal questioned the courts’ capacity under appeals process, saying it reversed the existing structure of the law to keep the presumption of innocence until the Government accountable, saying the proven guilty.141 interference of courts in “the executive’s However, POFMA is also being assessment of what the public considered as a model for tackling interest requires” was frowned upon online misinformation in other countries. 133 in Singapore. He proposed explicit Nigeria is considering an almost amendments ensuring judicial review identically named “Protection from 134 powers. Internet Falsehood and Manipulation Bill 2019.” One of its key Senate promoters named Singapore as an inspiration for the Nigerian legislation.142

127 Howard Lee and Terence Lee, “Polarising dissent: The constructed narrative of Singapore’s new “fake news” law,” The Asia Dialogue, June 10, 2019, https://theasiadialogue. com/2019/06/10/polarising-dissent-the-constructed-narrative-of-singapores-new-fake-news-law/. 128 Wei Yao and Kenny Chng, “Symposium on POFMA: Reflections on thinking about the POFMA,” Singapore Public Law, (November 2019): n.p. https://ink.library.smu.edu.sg/cgi/ viewcontent.cgi?article=4944&context=sol_research. 129 Siyuan Chen and Chen Wei Chia, “Singapore’s latest efforts at regulating online hate speech,” Research Collection School of Law, (2019): n.p. https://ink.library.smu.edu.sg/cgi/ viewcontent.cgi?article=4879&context=sol_research 130 Benjamin Joshua Ong, “Looking beyond the vague terms in Singapore’s fake news laws,” TodayOnline, May 14, 2019, https://www.todayonline.com/commentary/looking- beyond-vague-terms-fake-news-laws. 131 Ibid. 132 Lee Kay Howe, “True or false or misleading: [A]dequate judicial oversight over Part 3 directions under the Protection from Online Falsehoods and Manipulation Act,” Singapore Comparative Law Review, (2019): 239. 133 Cherian George, “Online falsehoods bill: Will words in legislation mean whatever S’pore govt chooses them to mean?,” Mothership, published April 6, 2019, https://mothership. sg/2019/04/online-falsehoods-cherian-george/ 134 Ibid. 135 Howard Lee and Terence Lee, “From contempt of court to fake news,” 82 136 Ibid. 137 Ibid., 86 138 “Singapore: Parliament must reject internet regulation bill that threatens freedom of expression,” International Commission of Jurists, published April 4, 2019, https://www.icj.org/wp-content/uploads/2019/04/Singapore-fake-news-bill-News-web-story-2019-ENG.pdf. 139 Wong, “Singapore fake news law polices chats and online platforms.” 140 David Kaye, Mandate of the Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression (United Nations, 2019), https://www.ohchr.org/Documents/Issues/Opinion/Legislation/OL_SGP_3_2019.pdf. 141 Ibid., 6 142 Ellie Bothwell, “Fake news laws may ‘catch on’ during coronavirus,” Times Higher Education, April 6, 2020, https://www.timeshighereducation.com/news/ fake-news-laws-may-catch-during-coronavirus

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 23 Timeline of Singapore’s “fake news” laws and policies

Table 2.1: Timeline of various laws and regulations implemented in Singapore

Year Legislation/Policy Development/Initiative 1963 Malaysia’s Sedition Ordinance 1948 is extended to Singapore upon joining the Federation of Malaysia. 1965 Singapore maintains sedition legislation upon gaining independence from Malaysia. Defamation Act is enacted. 1985 Current version of Sedition Act is enacted. 1997 Within the Broadcasting Act, the Internet Code of Practice comes into effect. It outlines responsibilities for internet service and content providers to ensure content standards are in line with the public interest, decency, and national harmony.143 2013 Online News Licencing Scheme is established, under the Broadcasting Act. 2014 Defamation Act is revised. Protection from Harassment Act commences.

2018 [January] Singaporean government tables its green paper, Deliberate Online Falsehoods: Challenges and Implications, which establishes a Select Committee to investigate online falsehoods. A call for public submissions to the committee also goes out.144 [March] Singapore government conducts public hearings over eight days, inviting advocacy groups, tech companies such as Facebook and Twitter, among others to air their views about deliberate online falsehoods145 [September] Select Committee presents its report on online falsehoods to parliament.

2019 [January] Ministry of Culture, Community and Youth launches seminar module, Combatting Fake News.146 It aims to educate religious and community groups on fake news and how to detect it. [April] POHA amendments, which include the establishment of a specialist court, protections against “doxing,” and strengthened protections for victims of harassment and falsehoods, are tabled to parliament.147 [May] Protection from Online Falsehoods and Manipulation Act (POFMA) is passed in parliament, 72 in favour to 9 against. [October] POFMA comes into force in Singapore. [November] First use of POFMA. It is against Progress Singapore Party member Brad Bowyer. 2020 POHA amendments come into force, extending access to the law's use from individuals to companies.

Source: Authors

143 Infocomm Media Development Authority, Internet Code of Practice (Singapore: IMDA, 1997), https://www.imda.gov.sg/-/media/Imda/Files/Regulations-and-Licensing/ Regulations/Codes-of-Practice/Codes-of-Practice-Media/PoliciesandContentGuidelinesInternetInterneCodeOfPractice.pdf 144 Ministry of Communications and Information and Ministry of Law, Deliberate and Online Falsehoods: Challenges and Implications (Singapore: Parliament of Singapore, 2018), https://www.nas.gov.sg/archivesonline/government_records/Flipviewer/grid_publish/6/6797717d-f25b-11e7-bafc-001a4a5ba61b 06012018Misc.10of2018/web/html5/ index.html?launchlogo=tablet/GovernmentRecords_brandingLogo_.png&pn=1. 145 ,Fathin Ungku, “Singapore invites rights group to fake news hearing as dispute grows,” Reuters, March 28, 2018, https://www.reuters.com/article/us-singapore-politics- malaysia-scandal/singapore-pm-files-defamation-suit-against-blogger-who-shared-article-on-facebook-idUSKBN1O414L. 146 “Combating fake news,” National Library of Singapore, last updated October 29, 2020, https://sure.nlb.gov.sg/resources/audience/adults/combating-fake-news 147 Ministry of Law, Enhancements to the Protection from Online Harassment Act (‘POHA’) (Singapore: Ministry of Law, April 1, 2019), https://www.mlaw.gov.sg/news/press- releases/enhancements-to-the-protection-from-harassment-act-poha

24 La Trobe University Conclusion

As noted in the introduction, Singapore’s unique demographic and political structure has strongly influenced its approach to tackling online misinformation, as with so many other issues in the past. A culturally and ethnically diverse population, while part of the national identity, is also viewed by policymakers as a potential point of instability, prompting strict regulatory and legislative measures such as POFMA.

Evidence that the Government has used And any notion that a Government might POFMA as an effective tool to silence be deterred from abusing its power for political opponents and suppress media fear of an electoral backlash149 would freedom should not come as a surprise not appear to apply in Singapore, rated to students of Singaporean history who as “partly free” with a democracy score can recall how former leader Lee Kuan of 50 out of 100,150 enabling the PAP Yew once held that freedom of the press Government to reign over the island must be “subordinate to the primacy of state since 1959. purpose of an elected government.”148

148 Quoted in Human Rights Watch, Kill the Chicken to Scare the Monkeys, 22 149 Yao and Chng, “Symposium on POFMA: Reflections on thinking about the POFMA.”; Minister K Shanmugam quoted in “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Parliament of Singapore, published May 8, 2019, vol. 94, Part 2, https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-367. 150 Sarah Repucci, “A Leaderless Struggle for Democracy: Democracy and Pluralism are under assault”, Freedom House, 2020.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 25 CHAPTER 3 The European Union: Approaches to online misinformation and disinformation Introduction

This chapter examines the European Union’s approach to over its first two years of operation, demonstrating that while considerable tackling online misinformation and disinformation – including progress has been achieved, much the development and performance of the EU’s Code of Practice more is required to sufficiently combat on Disinformation, established in September 2018. the problem. Actions taken against misinformation and disinformation in individual EU member states are also As a first-of-its-kind, voluntary agreement and is considered by many as the current documented. A timeline summarising between the European Commission, benchmark for action on misinformation Europe’s major laws and policies digital platforms and advertising groups, and disinformation. addressing this pernicious problem is the Code has proved to be an influential provided at the end of the chapter. Full This chapter investigates EU and example of self-regulation internationally, details of the different regulations and independent assessments of the Code laws are available in Appendix C.

Background and context

In March 2015, in one of the first major European initiatives to commitment by the Commission to counter online falsehoods, the European Council, citing ’s support the development of a voluntary Code of Practice.8 It called on digital sustained disinformation campaigns, proposed an Action Plan on platforms to participate in the Code 1 Strategic Communication. This led to the creation of EUvsDisinfo, and forecast its establishment by 2018. a website set up to identify and counter Russia’s ongoing The EU outlined fundamental 2 disinformation campaigns. requirements be met in its war on disinformation. Foremost was the need to balance the fundamental right to In April 2018 the European Commission “EU citizens’ health, the environment or freedom of expression with the right released Tackling Online Disinformation: security.” Excluded from the definition of the public to be properly informed.9 A European Approach, which became the were “reporting errors, satire and parody, Disinformation impaired freedom of central document for further actions in or clearly identified partisan news and expression, yet state actors had an 3 6 the field. It stated that the main threat commentary.” obligation to refrain from censorship, was disinformation, not ‘fake news’ The report enshrined four overarching since legal content even when harmful or misinformation.4 principles for online communication: was generally protected under freedom The report defined disinformation transparency, credibility of information, of expression principles.10 A focus on as “verifiably false or misleading diversity of information and inclusive responsible behaviour in conveying information that is created, presented cooperation.7 Its focus was on online information to end users, a pluralistic and disseminated for economic gain or platforms; fact checking; online media policy that increased exposure of to intentionally deceive the public, and accountability; election integrity; media diverse content, and the empowerment may cause public harm.”5 Public harm literacy; quality journalism; and strategic of online consumers through media may include “threats to democratic communication. However, the most literacy initiatives was prioritised.11 political and policymaking processes” or important aspect of the report was a

1 European Council, “European Council meeting (19 and 20 March 2015) – Conclusions,” published March 20, 2015, EUCO 11/15, 5, https://www.consilium.europa.eu/ media/21888/european-council-conclusions-19-20-march-2015-en.pdf 2 “About,” EUvsDisInfo, accessed November 17, 2020, https://euvsdisinfo.eu/about/ 3 European Commission, Tackling Online Disinformation: A European Approach (Brussels: European Commission, April 26, 2018), https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=CELEX:52018DC0236&qid=1583456906698&from=EN 4 The HLEG characterised misinformation as “misleading or inaccurate information shared by people who do not recognise it as such,” in Commission High Level Expert Group on Fake News and Online Disinformation, A Multi-Dimensional Approach to Disinformation: Report of the Independent High-Level Expert Group on Fake News and Online Disinformation (Brussels: European Commission, 2018), 10, http://ec.europa.eu/newsroom/dae/document.cfm?doc_id=50271 5 European Commission, Tackling Online Disinformation: A European Approach, 3-4. 6 Ibid., 3. 7 Ibid., 6. 8 Ibid., 8. 9 , Charter of Fundamental Human Rights of the European Union (2000/c 364/01) (European Parliament, December 18, 2000), Article 11, https://www.europarl.europa.eu/charter/pdf/text_en.pdf; European Court of Human Rights, European Convention on Human Rights (European Court of Human Rights, 2010), Article 10, https://www.echr.coe.int/Documents/Convention_ENG.pdf 10 European Commission, Tackling Online Disinformation, 1. 11 Andrea Renda, The Legal Framework to Address “Fake News”: Possible Policy Actions at the EU Level (Brussels: Policy Department for Economic, Scientific and Quality of Life Policies, June 2018), 20-21, https://www.europarl.europa.eu/RegData/etudes/IDAN/2018/619013/IPOL_IDA(2018)619013_EN.pdf

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 27 The EU Code of Practice

In September 2018 the Code of Practice was voluntarily agreed to EU agencies urged the Commission by 13 signatories, including Facebook, Twitter, Google and Mozilla. to take further regulatory action if 12 measures against disinformation Microsoft signed up to the Code in 2019 and TikTok in June 2020. and other elements of the Code were not implemented quickly or satisfactorily.14 Appraisals of the Code The Code is based on five pillars setting y Integrity of services (which by the Commission itself and relevant out actions relating to: largely concerns inauthentic and stakeholders are discussed shortly. manipulative activity) y Scrutiny of ad placements y Empowering consumers y Political advertising and issue- y Empowering the research community.13 based advertising

Action Plan against Disinformation

Soon after the establishment of the Code, the Commission released cross-border cooperation among media 19 its Action Plan against Disinformation in December 2018.15 It outlined literacy practitioners. the Commission’s commitments to monitoring the Code and also It also committed to establishing a earmarked the European Regulators Group for Audiovisual Media European network of independent 16 fact checkers with investment in new Services (ERGA) to provide additional monitoring services. technologies for content verification.20 The Social Observatory for Disinformation and Media Analysis The Action Plan focused on improving y Improve societal awareness (SOMA) was launched to enable the capabilities of EU institutions to: and resilience.17 networking and knowledge exchange y Detect and analyse disinformation The Commission would seek to train between independent fact checkers.21 y Coordinate joint responses journalists in better quality media to disinformation practices18 and, as part of Media y Mobilise the private sector to tackle Literacy Week (March 2019), support disinformation

12 Natasha Lomas, “TikTok joins the EU’s Code of Practice on Disinformation,” TechCrunch, June 22, 2020, https://techcrunch.com/2020/06/22/tiktok-joins-the-eus-code-of- practice-on-disinformation/ 13 European Commission, EU Code of Practice on Disinformation (Brussels: European Commission, September 26, 2018), https://ec.europa.eu/newsroom/dae/document. cfm?doc_id=54454 14 European Economic and Social Committee, Opinion on Action Plan against Disinformation (Brussels: European Commission, March 20, 2019), https://webapi2016.eesc. europa.eu/v1/documents/EESC-2018-06302-00-00-AC-TRA-EN.docx/content; European Committee of the Regions, Opinion of the European Committee of the Regions on “Tackling online disinformation: a European Approach” (Brussels: Official Journal of the European Union, May 16, 2019), comment 24, https://eur-lex.europa.eu/legal-content/ EN/TXT/PDF/?uri=CELEX:52018IR3908&qid=1583456906698&from=EN 15 European Commission, Action Plan against Disinformation (Brussels: European Commission, December 5, 2018), https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52018JC0036&qid=1583456906698&from=EN 16 European Commission, Assessment of the Code of Practice on Disinformation – Achievements and Areas for Further Improvement (Brussels: European Commission, September 10, 2020), https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=69212 17 European Commission, Tackling COVID-19 Disinformation – Getting the Facts Right (Brussels: European Commission, June 10, 2020), 2, https://eur-lex.europa.eu/legal- content/EN/TXT/PDF/?uri=CELEX:52020JC0008&from=EN 18 European Commission, Action Plan against Disinformation, 15. 19 Ibid., 11. 20 Ibid., 9. 21 “About us,” Social Observatory for Disinformation and Media Analysis, accessed November 19, 2020, https://www.disinfobservatory.org/about-us/; European Commission, Report on the Implementation of the Action Plan against Disinformation (Brussels: European Commission, June 14, 2019), 6-7, https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=CELEX:52019JC0012&from=EN

28 La Trobe University Assessments of the Code

Since its implementation in September 2018, the code’s Other issues noted in the Commission’s performance has been assessed by the Commission and analysis concerned the accuracy of signatories’ political ad labelling31 and various independent consultants. the modest interest in the Code from the digital and advertising industries.32

Post-2019 EU elections with its monitoring responsibilities, the ERGA: Assessment of Commission focused on areas requiring The European parliamentary elections immediate improvement. the Implementation of in May 2019 were identified in the Action The Commission surmised that efforts the Code of Practice Plan against Disinformation as a critical to empower users and the research test of the code’s capacity to protect and ERGA’s May 2020 assessment of the community lagged behind measures ensure transparency around Europe’s Code shared many of the Commission’s to improve scrutiny of ad placement, democratic processes.22 To this end, concerns. It found signatories’ self- political and issue-based advertising the Commission observed that the assessment reports lacked sufficient and integrity of services. It called on Code helped “increase transparency of data, which was a particular problem platforms to facilitate better relationships political communications”, limit online in regard to strengthening political with relevant stakeholders such as information manipulation and enabled advertising, ad placement and integrity researchers, fact checkers and the 33 citizens to discern the source of political of services measures. Limited access media. In particular, platforms’ reticent .23 High ranking EU officials said in a to and transparency of data hindered and “episodic” data sharing was said to joint statement that Facebook, Google independent investigations of the code’s hamper independent research and fact and Twitter had “made some progress performance and prevented adequate checking efforts.27 This underdeveloped under the Code of Practice.”24 Despite evaluation of the code’s impact by cooperation meant the potential 34 this acknowledgement, the Commission the EU and member states. Like to identify “persistent or egregious said there was still much to be done the Commission, ERGA lamented the purveyors of disinformation” was not to combat the evolving threat of “limited” number of signatories, arguing being fulfilled.28 The implementation of disinformation campaigns. that “significant” actors such as private ad scrutiny and safety measures messaging platforms WhatsApp and was also hampered by signatories’ lack Messenger were “missing.”35 of coordination with these stakeholders. First annual ERGA raised some other issues with self‑assessment reports While user empowerment initiatives, the code’s performance over its first such as media literacy campaigns and year. It found the measures of the Code In October 2019, Code signatories training, had improved before the Code’s “too general in terms of content and tabled their first annual self-assessment inception, the Commission experienced structure”, with “a lack of uniformity in the reports, from which the Commission difficulty in measuring their impact and procedures (and the definitions) adopted provided a summary and analysis. efficacy. Again, platforms’ inadequate by the different platforms”.36 As a result, The Commission championed the reporting was identified as a central ERGA encouraged signatories to strive Code for increasing transparency of 29 cause. The absence of specificity in the for greater consistency and specificity in platforms’ policies on disinformation, information shared by platforms made their definitions and measures, including and establishing “a framework for it hard for the Commission to measure in the way they respond to fact-checked structured dialogue to monitor, “malicious behaviour specifically news.37 ERGA also found media literacy improve and effectively implement targeting the EU and the progress initiatives had been “conducted in a those policies.”25 Signatories’ reports achieved by the platforms to counter scattered manner”.38 demonstrated their “comprehensive such behaviour”.30 efforts” to meet the commitments of the Code, such as granting insights into their actions against coordinated inauthentic behaviour.26 Nonetheless, in keeping

22 European Commission, Action Plan against Disinformation, 2. 23 “A Europe that protects: EU reports on progress in fighting disinformation ahead of European Council,” European Commission, Published June 14, 2019, https://ec.europa.eu/Commission/presscorner/detail/en/IP_19_2914 24 Ibid. 25 European Commission, Code of Practice on Disinformation: First Annual Reports (Brussels: European Commission, October 2019), 1, https://ec.europa.eu/ newsroom/dae/document.cfm?doc_id=62698. 26 Ibid., 8-9. 27 Ibid., 2. 28 Ibid., 5. 29 Ibid., 10-11. 30 Ibid., 9. 31 Ibid., 7. 32 Ibid., 2. 33 European Regulators Groups for Audiovisual Media Services, ERGA Report on Disinformation: Assessment of the Implementation of the Code of Practice (Brussels: European Commission, May 2020), 47-48, https://erga-online.eu/wp-content/uploads/2020/05/ERGA-2019-report-published-2020-LQ.pdf 34 Ibid., 3,49 35 Ibid., 3. 36 Ibid., 3. 37 Ibid., 52. 38 Ibid., 48.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 29 ERGA’s primary recommendation was violations may have contributed to this authoritative content, their actions to to shift the code’s approach towards reticence.46 Effort was needed from both improve users’ awareness, manipulative co-regulation, since the current self- sides to improve the effectiveness of behaviour on their platforms, and regulatory model exhibited various this relationship. advertising of COVID-19 disinformation compliance issues.39 While the specifics on their platforms and third-party sites. The report also suggested refinement of of such an approach were not outlined, the use of the terms disinformation and In September 2020, the Commission ERGA supports the implementation of misinformation. It was argued a better released Code signatories’ first baseline more stringent reporting obligations and understanding of intent may resolve the reports on these COVID-19-related asks. “the introduction of a formal backstop ambiguity of terminology and “improve The reports demonstrated signatories mechanism to deliver the required the effectiveness of specific actions to had increased access to and visibility monitoring and enforcement elements”.40 combat specific behaviours”.47 of authoritative information.52 Notably, Hence, ERGA is in favour of empowering however, the application of these the Commission and national regulatory measures was not uniform across the EU. authorities to more effectively hold Tackling COVID-19 platforms accountable to deliver on In particular, platforms “did not detect their commitments.41 disinformation coordinated disinformation operations In a June 2020 report, the Commission with specific focus on COVID-19 run on 53 reviewed the COVID-19 “infodemic” their services”. This finding suggests Independent and the various instruments working that misinformation and uncoordinated consultation to combat coronavirus-related disinformation were more common misinformation and disinformation. The manifestations of harmful content about The Commission appointed consultancy Commission found that platforms had COVID-19. Yet the Commission argued firm Valdani, Vicari and Associates to reported “adjustments to their policies that misinformation did not necessarily conduct an independent review of the to address the COVID-19 disinformation require government intervention, since policies and performance of signatories threat”, including promoting “accurate public literacy and fact checking could under the Code.42 Released in May and authoritative information” from intervene. Disinformation, on the other 2020, the review stated that the code’s trusted sources such as the World Health hand, needed “to be addressed through regular monitoring framework fostered Organisation and professional media other means, including actions taken greater transparency of social media 54 outlets.48 Platforms demoted false and by governments”. platforms during elections and other misleading content, limited suspicious political campaigns. In line with previous In the second phase of the process, ads and removed content deemed to assessments, it said pillars 1-3 had the Commission committed to have the potential to cause health and “produced a more positive change” provide monthly assessments of the public harm.49 than pillars 4 and 5.43 effectiveness of these measures from However, the Commission observed August to December 2020. For pillar 4 (empowering users), while that platforms had failed to empower it was reported that the Code had civil society actors, fact checkers and contributed to improving consumers’ researchers – due largely to limited European Commission’s awareness of disinformation, better information sharing. The Commission assessment of the Code reporting on the results of platforms’ said the crisis had further highlighted tools and actions was needed.44 of Practice the need for greater transparency and For pillar 5 (empowering the research accountability from platforms.50 The European Commission’s community), while social media assessment of the Code in September The report outlined safeguards which platforms were said to have increased 2020 to examine its initial 12 months the Commission expected platforms collaboration with researchers and of operation praised the establishment to implement to combat the COVID-19 fact checkers, and improved their of a framework for dialogue between infodemic.51 The Commission vowed to access to relevant data, the research relevant stakeholders, monitoring, and monitor how platforms were tracking community reported being unsatisfied greater transparency and accountability in this space on a monthly basis. with the responses from platforms in of platform operations.55 Improvements Signatories have been asked to provide relation to data requests.45 Platforms’ were noted under each “pillar”. Greater information regarding the promotion of concerns about user data privacy scrutiny of ad placements had

39 Ibid., 3. 40 Ibid., 4. 41 Ibid., 3. 42 Valdani, Vicari and Associates, Study for the Assessment of the Implementation of the Code of Practice on Disinformation (Brussels: European Commission, May 2020). https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=66649. 43 Ibid., 96-97. 44 Ibid., 97. 45 Ibid. 46 Ibid., 58. 47 Ibid., 97. 48 European Commission, Tackling COVID-19 Disinformation – Getting the Facts Right, 8. 49 Ibid. 50 Ibid. 51 Ibid., 9. 52 “First baseline reports – Fighting COVID-19 disinformation monitoring program,” European Commission, Published September 10, 2020, https://ec.europa.eu/digital-single- market/en/news/first-baseline-reports-fighting-covid-19-disinformation-monitoring-programme 53 Ibid. 54 European Commission, Tackling COVID-19 Disinformation – Getting the Facts Right, 4. 55 European Commission, Assessment of the Code of Practice on Disinformation – Achievements and Areas for Further Improvement, 4.

30 La Trobe University “contributed to reducing monetisation Lack of uniform definitions Issues with the voluntary model incentives for actors that disseminate The lack of clarity and uniformity The Commission acknowledged disinformation online for economic around definitions of the problem multiple inherent problems with gain”56 Labelling measures increased was identified as a significant issue in the code’s voluntary model. First, it transparency of political and issue-based the Code. The COVID-19 “infodemic” created “regulatory asymmetry”, where advertising and provided information emphasised the need to differentiate signatories were at a disadvantage on the source, audiences and impact of between different forms of false and to non-signatories since they must these ads.57 Action against inauthentic misleading information and manipulative implement sometimes costly measures accounts led to their disclosure.58 behaviour.63 Such that the Commission to fulfil their commitments.72 Further, Technological tools that promote reliable argued for greater incorporation malicious actors had the option of using information, improve access to a broad of misinformation and “influence non-signatory platforms to spread diversity of information sources and new operations”64 in understandings of disinformation, which the Commission disinformation flagging tools empowered the wider problem.65 observed throughout the pandemic. users, while cooperation with fact Second, key actors in the disinformation checkers also enhanced users’ capacity Gaps in the coverage of Code space outside the Code, such as private to critically assess information.59 Some commitments messaging platforms and advertising data sharing initiatives with researchers The Commission argued that “the groups, were under no obligation to and fact checkers were also noted. vagueness of the Code’s commitments… become signatories.73 Yet the central message of the creates serious risks of incomplete Third, a voluntary model meant there Commission’s review was that action” against forms of inauthentic were no mechanisms to ensure adequate signatories were still falling short of the activity that might fall outside its scope, independent oversight, cooperation code’s expectations. The Commission such as micro-targeting of political with member state authorities, and outlined four main issue areas: ads and fairness in political advertising compliance with Code commitments.74 online.66 Moreover, the absence of y Inconsistent and incomplete Consequently, the sole sanction for non- clear and measurable key performance application of the Code across compliance was expulsion from the Code indicators in the Code stymied objective platforms and Member States; – a penalty that somewhat optimistically measurements and comparisons y Lack of uniform definitions; depended on signatories’ aversion to between platforms.67 y Gaps in the coverage of Code reputational damage.75 commitments; Meanwhile collaboration with Finally, the Code did not have procedures y Other limitations intrinsic to the self- independent fact checking initiatives to ensure that “fundamental rights”, such regulatory nature of the Code.60 fell short of the Commission’s as freedom of speech, were upheld by expectations.68 Sufficient access to signatories.76 Incomplete and inconsistent data and resources for fact checking application of the Code and research activities as outlined in In sum, despite recognising the User empowerment initiatives, while the Code had “not been achieved” and achievements of the Code, the 69 increasing during the COVID-19 crisis, remained “episodic and arbitrary,” the Commission also acknowledged were not deployed across all European Commission found. glaring deficiencies in its design, scope, terminology and practice. countries and languages. No data was Regarding actions targeting inauthentic Signatories were strongly encouraged provided to measure their effectiveness behaviour and content, signatories did to escalate their efforts in addressing in increasing user engagement with not provide accurate, relevant, location- misinformation and disinformation as the credible information, critical thinking and specific data, meaning the Commission EU reviewed its strategy. civic participation.61 User complaints could not evaluate their effectiveness in mechanisms were found to be lacking the EU and its Member States.70 These uniformity and ease of accessibility measures were also found to focus too across platforms – should they exist closely on foreign interference, with EU- at all. The Commission found they based actors escaping due attention.71 were unable to conduct comparisons between platforms’ different fact checking measures, since measurable performance indicators were not provided.62

56 Ibid. 57 Ibid., 5. 58 Ibid. 59 Ibid., 6. 60 Ibid., 7. 61 Ibid. 62 Ibid. 63 European Commission, Assessment of the Code of Practice on Disinformation – Achievements and Areas for Further Improvement, 12. 64 The Commission defines influence operations as “information campaigns by third-country actors that employ false or misleading information in combination with manipulative online techniques to interfere in EU or Member State electoral or policy-making processes, where the intention to deceive the public may be presumed” (Ibid.). 65 Ibid. 66 Ibid., 14. 67 Ibid., 15. 68 Ibid., 11. 69 Ibid. 70 Ibid., 9. 71 Ibid., 10. 72 Ibid., 17. 73 Ibid., 18. 74 Ibid. 75 Ibid. 76 Ibid., 19.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 31 Reactions to the EU approach

Much of the public commentary surrounding the EU’s approach Recently, some member states have echoes the Commission’s evaluations. tabled their official judgement of the Code. In June 2020, Estonia, Lithuania, and Latvia released a position Tasked by the Commission to assess “indicates a lack of consensus paper which argued that the self- the Code, a ‘Sounding Board’ comprising among key stakeholders regarding regulatory framework “is insufficient representatives from the media, civil the scope of the issue and therefore and unsuitable” to address online 86 society, fact checkers and academia its potential solutions”.80 What disinformation. In September 2020, the argued the Code contained: actions had been undertaken so far heads of state of France, Lithuania and were also characterised by unclear Latvia published a joint statement that No common approach, legal foundations and an unreliable implored the Commission to push for political mandate.81 more stringent measures, particularly no clear and meaningful concerning the protection of democratic commitments, no Some academics have argued for a elections.87 They called for more measurable objectives or mandatory co-regulatory model, with one effective mechanisms enabling member calling it the “best imperfect” approach to states to identify and counter disruptive KPIs, hence no possibility 82 tackling disinformation. Co-regulation activity, and to ensure transparency to monitor process, and no would see the EU legislate the objectives and enforcement of Code measures. compliance or enforcement of actions against disinformation, which Consistent in both these critiques is tool: it is by no means self- digital platforms would then fulfil under the view that the Code has not gone the auspices of a Code developed in far enough. regulation, and therefore conjunction with civil society and the the Platforms, despite their Commission.83 A similar model was The digital platform experts interviewed efforts, have not delivered a endorsed elsewhere.84 for this study were largely ambivalent on suggestions that the EU approach 77 Code of Practice. Other notable critiques were that the EU’s might be the global standard for tacking approach lacked baseline reporting, had misinformation and disinformation. One Misinformation scholars found the budgetary and institutional restraints expert stated that the platform they scope and terminology of the Code and a lack of effective cooperation with worked for “would prefer” a European- 78 lacked clarity. Leaving platforms to member states, posing a significant style cross-platform and self-regulatory resolve these ambiguities presented a problem for the integrity of elections.85 approach, rather than “more prescriptive” 79 possible risk for freedom of expression. regulation as seen in countries such The code’s ill-defined terminology as Singapore.88

77 Sounding Board, The Sounding Board’s Unanimous Final Opinion on the So-Called Code of Practice (European Commission, September 24, 2018), https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=54456. 78 James Pamment, The EU Code of Practice on Disinformation: Briefing Note for the New EU Commission (Washington: Carnegie Endowment for International Peace, March 2020), 4, https://carnegieendowment.org/2020/03/03/eu-code-of-practice-on-disinformation-briefing-note-for-new-european-Commission-pub-81187.; Nenadic, “Unpacking the ‘European approach’ to tackling challenges of disinformation and political manipulation,” 13. 79 Ibid. 80 Pamment, The EU Code of Practice on Disinformation: Briefing Note for the New EU Commission, 5. 81 James Pamment, The EU’s Role in Fighting Disinformation: Taking Back the Initiative (Washington: Carnegie Endowment for International Peace, June 2020), 5, https://carnegieendowment.org/files/Pamment_-_Future_Threats.pdf. 82 Paul-Jasper Dittrich, Tackling the Spread of Disinformation: Why a Co-regulatory Approach is the Right Way Forward for the EU (: Jacques Delors Centre, December 12, 2019), 10, https://www.bertelsmann-stiftung.de/fileadmin/files/BSt/Publikationen/GrauePublikationen/EZ_JDI_BST_Policy_Paper_Disinformation_Dittrich_2019_ENG.pdf 83 Ibid., ii. 84 Chris Marsden, Trisha Meyer, and Ian Brown, “Platform values and democratic elections: How can the law regulate digital disinformation?,” Computer Law and Security Review 36 (2020): 1, DOI: 10.1016/j.clsr.2019.105373. 85 Kanzanira Thorington, “Europe’s elections: The fight against disinformation,”Council on Foreign Relations, Published May 23, 2019, https://www.cfr.org/blog/europes-elections- fight-against-disinformation. 86 Samuel Stolton, “EU code of practice on disinformation ‘insufficient and unsuitable,’ member states say,” EURACTIV.com, June 5, 2020, https://www.euractiv.com/section/ digital/news/eu-code-of-practice-on-disinformation-insufficient-and-unsuitable-member-states-say/. 87 Emmanuel Macron, Gitanas Nauseda and Arturs Krisjanis Karins, Joint Statement By The Presidents Of France And Lithuania, And Prime Minister Of Latvia On Protection Of Democracies (Paris: Palais de l’Elysee, September 28, 2020), https://static.lsm.lv/documents/z5.pdf 88 Interview 1, interviewed by Andrea Carson, August 7, 2020.

32 La Trobe University Other EU measures to tackle false information online

While the Code of Practice is the central initiative aimed at tackling New Laws misinformation and disinformation, there are other programs the EU There are also several important has established. parliamentary initiatives still to be released in 2020 by the EU. The European Democracy Action Plan closed public consultation in September. Rapid Alert System Media Literacy It aims to “ensure that citizens are able to participate in the democratic Outlined in the Action Plan Against Initiatives system through informed decision- Disinformation, the Rapid Alert System Other initiatives focus more directly upon making free from unlawful interference was established in March 2019 and is a public awareness. A November 2018 and manipulation.”93 In addition to data and knowledge sharing network. directive of the European Parliament disinformation, it will include measures It is used to connect disinformation and Council stated that Member States on electoral integrity, media freedom and experts across Europe and facilitate are required to promote and oversee media pluralism. It has been said that it communication and strategic responses the development of their citizens’ media will build on the progress of the Action between governments.89 Despite its literacy skills.91 Additionally, media Plan and Code of Practice and help intentions, in practice the system is not service providers and video-sharing Europe determine next steps in its fight often used by Member States. James platforms are required to provide “for against disinformation.94 Pamment ascribed this to “major effective media literacy measures and Further, a Digital Education Act Plan differences in how Member States view tools and [to] raise users’ awareness of was scheduled to be tabled in the EU in the threat of disinformation” and “a lack those measures and tools.”92 of trust” between them.90 As such, issues late 2020 that aims to facilitate digital around inadequate information sharing literacy.95 A stated aim is to help people appear to afflict not only signatories’ counter disinformation. efforts but also those of governments. The Digital Services Act was also slated to be enacted late 2020. It aims to modernise the legal framework of Europe’s digital market. It is expected to provide new regulatory obligations on digital service providers so that platforms clearly outline their responsibilities to users.96 It is anticipated that new powers will address platform responses to online disinformation.97

89 European Commission, Tackling COVID-19 Disinformation – Getting the Facts Right, 2. 90 Pamment, The EU’s Role in Fighting Disinformation: Taking Back the Initiative, 9. 91 European Parliament and the Council, Directive (EU) 2018/1808 of the European Parliament and of the Council (Brussels: European Parliament, November 14, 2018), Article 33a, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32018L1808&from=EN 92 Ibid., section 3j. 93 “Protecting European democracy from interference and manipulation – European Democracy Action Plan,” European Commission, accessed November 19, 2020, https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12506-European-Democracy-Action-Plan 94 Pamment, The EU’s Role in Fighting Disinformation: Taking Back the Initiative, 10. 95 https://ec.europa.eu/education/education-in-the-eu/digital-education-action-plan_en 96 “The Digital Services Act package,” European Commission, last updated June 22, 2020, https://ec.europa.eu/digital-single-market/en/digital-services-act-package 97 Pamment, The EU’s Role in Fighting Disinformation: Taking Back the Initiative, 10.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 33 Member States’ initiatives

Among Europe’s Member States, only Germany, France and Hungary have passed laws aimed at tackling disinformation and misinformation. In response to the Coronavirus Some states have implemented other non-legislative measures. ‘infodemic’, Hungary passed new measures that enabled President Orbán to “rule by decree”, as some critics claimed.106 It allowed the Government to Germany France issue prison sentences to those caught intentionally spreading false information Germany introduced a 2018 law that France passed its law against the that could hamper the government’s bans online hate speech and requires manipulation of information in response to the pandemic. The Bill’s social media platforms to remove or November 2018, which addresses false opponents feared it could be misused block “manifestly unlawful content” information disseminated on social to censor criticism of the government.107 within 24 hours and “unlawful content” media and media outlets influenced by While the Government rescinded within seven days.98 The Network foreign states.101 The law is specifically its extraordinary powers in June,108 Enforcement Act, known commonly as aimed at curbing misinformation and critics fear the legislation allows the ‘NetzDG’, covers social media platforms disinformation around elections. The law Government to reinstate another state with over two million users. It mandates allows judges “to order the immediate of emergency and its “rule by decree” an accessible user complaint process removal of online articles” containing as it sees fit under the pretence of with platforms to report biannually disinformation during the election fighting COVID-19.109 on their handling of user complaints campaign.102 Other measures enable regarding unlawful content. the national broadcasting agency to suspend foreign television channels NetzDG has had significant influence on Others’ measures which “deliberately disseminate false other countries’ approach to regulation. information likely to affect the sincerity Other Member States have implemented A report published by Danish think tank of the ballot,” while online political non-legislative measures to mitigate the Justitia found that among European campaigns are required to disclose spread of fake news. The Netherlands, in countries, France, the , their spend and financial backers.103 early 2019, launched an online campaign Russia and Belarus had all cited NetzDG Breaches of the law could result in a around fake news in the lead up to in policy discussions.99 Beyond Europe, one-year jail term or fine of €75,000. its elections.110 Finland has deployed other nations had considered Germany’s The law was considered an affront fact checking organisation Faktabaari approach, including Singapore and to liberty by critics,104 and was twice to teach professional fact checking Australia. The law has come under rejected in the Senate before the bill was techniques in Finnish schools.111 Spain criticism for perceived breaches of passed. French opposition parties and employed members of the International freedom of speech, and social media journalist associations have criticized Fact-Checking Network to monitor its platforms have been observed pre- the law for curtailing press freedom and 2019 elections.112 Since 2017, Italy has emptively deleting “more content than jeopardising pluralist debate necessary introduced media literacy into the school necessary out of fear of incurring in a healthy democracy.105 curriculum.113 In the lead up to the 2018 large fines.”100

98 Act to Improve Enforcement of the Law in Social Networks (Network Enforcement Act), Bundestag, (July 12, 2017), 2-3, https://www.bmjv.de/SharedDocs/ Gesetzgebungsverfahren/Dokumente/NetzDG_engl.pdf?__blob=publicationFile&v=2 99 Jacob Mchangama and Joelle Fiss, The Digital Berlin Wall: How Germany (Accidentally) Created a Prototype for Global Online Citizenship (Copenhagen: Justitia, November 2019), https://globalfreedomofexpression.columbia.edu/wp-content/uploads/2019/11/Analyse_The-Digital-Berlin-Wall-How-Germany-Accidentally-Created-a-Prototype-for- Global-Online-Censorship.pdf 100 Florian Saurwein and Charlotte Spencer-Smith, “Combating disinformation on social media: Multilevel governance and distributed accountability in Europe,” Digital Journalism 8, no. 6 (2020): 830. 101 Marine Guillaume, Combating the Manipulation of Information – A French Case (: Hybrid CoE, 2019), https://www.hybridcoe.fi/wp-content/uploads/2019/05/ HybridCoE_SA_Combating-the-manipulation-of-information.pdf 102 Sarwein and Spencer-Smith, “Combating disinformation on social media: Multilevel governance and distributed accountability in Europe,” 828. 103 Michael-Ross Fiorentino, “France passes controversial ‘fake news’ law,” Euronews, May 9, 2019, https://www.euronews.com/2018/11/22/france-passes-controversial-fake- news-law 104 Marc Watin-Augouard, “Law no. 2018-1202 of 22 December 2018 on combating the manipulation of information,” FIC-Observatory, published November 1, 2019, https://observatoire-fic.com/en/law-no-2018-1202-of-22-december-2018-on-combating-the-manipulation-of-information-by-general-marc-watin-augouard-founder-of-the-fic/ 105 Fiorentino, “France passes controversial ‘fake news’ law.” 106 Benjamin Novak, “Hungary moves to end rule by decree, but Orban’s powers may stay,” New York Times, June 16, 2020, https://www.nytimes.com/2020/06/16/world/europe/ hungary-coronavirus-orban.html 107 Shaun Walker and Jennifer Rankin, “Hungary passes law that will let Orban rule by decree,” The Guardian, March 31, 2020, https://www.theguardian.com/world/2020/mar/30/ hungary-jail-for-coronavirus-misinformation-viktor-orban 108 “Coronavirus update: Bill on rescinding the government’s emergency powers submitted to parliament today,” About Hungary, published May 26, 2020, http://abouthungary.hu/ news-in-brief/coronavirus-update-bill-on-rescinding-the-governments-emergency-powers-submitted-to-parliament-today/ ; Anyone who violates the law could face one year in prison and a fine of €75,000. 109 Novak, “Hungary moves to end rule by decree, but Orban’s powers may stay.” 110 “Dutch Government to launch anti-fake news campaign,” NL Times, December 13, 2018, https://nltimes.nl/2018/12/13/dutch-government-launch-anti-fake-news-campaign 111 Emma Charlton, “How Finland is fighting fake news – in the classroom,” World Economic Forum, published May 21, 2019, https://www.weforum.org/agenda/2019/05/how- finland-is-fighting-fake-news-in-the-classroom/ 112 Cristina Tardaguila, “Spain has a new government and its fact-checkers had an impact on the campaign,” Poynter, published May 3, 2019, https://www.poynter.org/fact- checking/2019/spain-has-a-new-government-and-its-fact-checkers-had-an-impact-on-the-campaign/ 113 Yasmeen Serhan, “Italy scrambles to fight misinformation ahead of its elections,” The Atlantic, February 24, 2018, https://www.theatlantic.com/international/archive/2018/02/ europe-fake-news/551972/

34 La Trobe University elections, Italians were also able to report news since 2013.115 In 2017, the Czech internet platforms and less reliance on fake news directly to the police, leading government established a monitoring protections traditionally provided by to concerns about the undue interference unit that debunks misinformation and the European Union.117 Table 3.1 recaps of the state.114 Belgium, Finland, Latvia, maintains a Twitter account informing the major initiatives to tackle fake news Cyprus, Luxembourg, and the public of its actions.116 in Europe since 2015 when Russian France have also enacted national disinformation campaigns in Europe Some argue that these national-level media literacy campaigns, while Greece triggered the first policy response to efforts lead to stronger regulation of has an official website debunking fake online disinformation.

Timeline of Europe’s “fake news” regulations and policies

Table 3.1: Timeline of various laws and regulations implemented in Europe

Year Legislation/Policy Development/Initiative Jurisdiction

2015 EEAS East StratCom Task Force’s Action Plan on Strategic Communication EU

2016 Joint Framework on Countering Hybrid Threats is released EU Finland media literacy schooling curriculum introduced Finland

2017 [June] Network Enforcement Act is passed by Bundestag Germany

2018 [April] Tackling Online Disinformation: A European Approach EU [September] Code of Practice against Disinformation EU Observatory for Social Media Analysis (SOMA) is established EU [November] Law against the Manipulation of Information approved France [December] Action Plan against Disinformation EU

2019 Netherlands online fake news awareness campaign Netherlands [March] European Media Literacy Week EU [June] European Commission reports on the performance of the Code concerning the May EU 2019 EU elections [October] First self-assessment reports tabled by signatories to EU Code of Practice EU

2020 [March] Act XII of 2020 on the Containment of the Coronavirus enables the Government to Hungary imprison anyone found disseminating untrue or misrepresented facts [May] ERGA release assessment of first year of EU Code EU [May] Independent consultants VAA release assessment of first year of EU Code EU [June] Hungarian Government rescinds extraordinary powers policing COVID-19 Hungary misinformation [June] Tackling COVID-19 Disinformation – Getting the Facts Right is released, outlining EU the Commission’s expectations for signatories to counter COVID-19 misinformation [September] First baseline reports on signatories’ measures to counter COVID-19 misinformation EU [September] European Commission releases its assessment of the first year of the Code EU

Source: Authors

114 Daniel Funke, “Italians can now report fake news to the police. Here’s why that’s problematic,” Poynter, published January 19, 2018, https://www.poynter.org/fact- checking/2018/italians-can-now-report-fake-news-to-the-police-heres-why-thats-problematic/; La Cour, Christina, “Governments countering disinformation: The case of Italy,” Disninfo Portal, accessed November 19, 2020 https://disinfoportal.org/governments-countering-disinformation-the-case-of-italy/ 115 European Commission, Assessment of the Implementation of the Code of Practice on Disinformation. 75-77; Saurwein and Spencer-Smith, “Combating disinformation on social media: Multilevel governance and distributed accountability in Europe,” 827-830. 116 Emily Schultheis, “The ’s fake news problem,” The Atlantic, October 21, 2017, https://www.theatlantic.com/international/archive/2017/10/fake-news-in-the- czech-republic/543591/ 117 Saurwein and Spencer-Smith, “Combating disinformation on social media: Multilevel governance and distributed accountability in Europe,” 834.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 35 Conclusion

Europe’s actions to facilitate digital platforms’ self-regulation consideration. That said, the EU – alongside independent commentators to tackle misinformation and disinformation reveal not – have suggested tweaks, not drastic only the challenges of the endeavour, but outcomes that transformation. A shift toward mandated can be achieved. The assessments of the EU’s framework co-regulation has been flagged. Yet conveyed in this chapter show that such an approach can despite the EU’s rhetoric, such a move is sure to raise concerns about be effective in parts. government over-reach as platforms look to protect freedom of speech and their own autonomy. The impending It has mitigated some of the impacts of vague and restrictive. Initially concerned European Democracy Action Plan and false and misleading information during primarily with “disinformation” at the Digital Services Act may suggest the elections and crisis events, such as the commencement of the Code, the future direction of the EU’s framework for COVID-19 “infodemic”; promoted reliable, Commission has recently recognised managing the spread of misinformation trustworthy information to online users; the necessity of incorporating and disinformation online. opened dialogue between platforms misinformation into the scope of its and governments; and helped restrict regulatory remit. The experience of the While many of the commentators in inauthentic and manipulative activity. COVID-19 infodemic, in particular, has this chapter and interviewed for this demonstrated the urgent threat posed study were reluctant to label the EU Yet, as the Commission acknowledges, by the spread of misinformation. This approach as the global “gold standard”, while progress has been made toward shift indicates that attempts to curb several admit it may be the best model improving the digital industry’s response the spread of false information cannot developed so far. As we see with the to the problem – an indication that its be reduced to a study of user “intent”. development of a Code of Practice expectations are not merely aspirational A more nuanced understanding of the in Australia, Europe has had a strong – more actions by all stakeholders complex and often organic exchange influence on regulation development are needed. Overall, the measures of false and misleading information in other jurisdictions. Despite its signatories implemented under the is needed. self-admitted shortcomings, the EU Code were viewed as incomplete and framework remains an instructive inconsistent. The monitoring of the Consistent in the Commission’s model for legislators around the world. code’s effectiveness proved difficult messaging about the Code is an – a product of platforms’ reluctance increasing impatience for greater to fully embrace the transparency that accountability from platforms. The comes with open information sharing Code is not enforceable as it currently and stakeholder cooperation. And exists, with the Commission indicating definitions of the problem’s scope a graduation from self-regulation and terminology were found to be too to a more stringent model is under

36 La Trobe University CHAPTER 4 Australia’s approach to online misinformation Introduction

This chapter examines recent discussions and developments addressing misinformation on digital platforms and improving news quality. around voluntary regulatory approaches to misinformation This initiative was born out of the 2019 and disinformation in Australia. With the onset of the Australian Competition and Consumer COVID-19 pandemic and heightened need for quality health Commission’s (ACCC) Digital Platforms information on and offline, 2020 was a watershed year for Inquiry final report. In response to the ACMA and ACCC initiatives, the news and media organisations around the world. It was platform providers, represented by also a critical year for digital technology companies, with the Digital Industry Group Inc. (DIGI), the Australian Government setting an end-of-year deadline produced an Australian Code of Practice on Disinformation for public for the digital platforms to establish a voluntary framework consultation in October 2020. DIGI to counter low-quality and harmful information online. The also commissioned a discussion paper voluntary code also requires the companies to establish produced by academics at the University of Technology Centre for Media ways for the public to make informed decisions in relation to Transition as a companion document to content quality. the draft code. While there are commonalities and overlap between each of the documents, This chapter outlines the background to digital platforms in Australia: A position this chapter discusses the substance the formation of these self-regulatory paper to guide code development. of each and points of differences initiatives, beginning with an overview of The ACMA position paper, released between ACMA’s and DIGI’s positions the Australian Media and Communications in June 2020, sets out the Australian that may require further attention before Authority’s (ACMA) position paper, Government’s guidelines for an industry- a consensus approach to tackling the Misinformation and news quality on wide voluntary code of practice for problem can be developed.

Background ACCC Digital Platforms Inquiry

In December 2017, then Treasurer Scott Morrison directed the The ACCC advised that a voluntary Australian Competition and Consumer Commission (ACCC) to code should apply to “complaints about disinformation that meet a undertake an inquiry into the digital platforms amid concerns about ‘serious public detriment’ threshold their influence on competition, consumers and society. The inquiry as defined in the code.”4 The ACCC received submissions from Facebook, Fairfax, Google, Netflix and defines disinformation as “inaccurate Twitter, among others.1 information created and spread with the intent to cause harm in relation to news and journalism, or content The final report, released in July 2019, called for the implementation of a presented as news and journalism, on included two recommendations dealing “digital platforms code to counter their services”. As is discussed in more specifically with online misinformation disinformation”.3 detail in the next chapter of this report, and disinformation. Recommendation the ACCC definition of disinformation The report recommended that a code to 14 covered “monitoring efforts of digital diverges from that of platforms and address both disinformation and “malign” platforms to implement credibility some commentators. This is not information be established, overseen by signalling,”2 while recommendation 15 unexpected, as there is no universally an independent regulator such as ACMA. agreed definition for disinformation.

1 Full list of submissions - https://www.accc.gov.au/focus-areas/inquiries-ongoing/digital-platforms-inquiry/submissions 2 ACMA defines credibility signalling as “flagging trusted sources and providing users with access to information on publishers and other sources” (Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia: A position paper to guide code development (Canberra: ACMA, 2020), 29, https://www.acma.gov.au/sites/default/files/2020-06/Misinformation%20and%20news%20quality%20position%20paper.pdf 3 Australian Competition and Consumer Commission, Digital Platforms Inquiry: Final report (Canberra: ACCC, 2019), 34, https://www.accc.gov.au/system/files/Digital%20 platforms%20inquiry%20-%20final%20report.pdf 4 Australian Competition and Consumer Commission, Digital Platforms Inquiry, 34.

38 La Trobe University Like the definition in the EU Code of person posting suspect content, while voluntary code. If the code was Practice on Disinformation, the ACCC pertinent, is for practical purposes too deemed to be ineffective, it said a more definition is notable for its emphasis difficult to divine with confidence. So, direct regulatory approach should be on intent. According to the ACCC they focus more on the behaviour of considered.5 In response to the inquiry, definition, disinformation involves a online users to detect “inauthentic” use the Government asked major digital deliberate intention to mislead, while of their platforms. They also point out platforms with an online presence in misinformation does not. The ACCC also that not all disinformation is necessarily Australia to “develop a voluntary code holds that content must be inaccurate false information (see chapter five). to cover the recommendations”,6 with to be classified as disinformation. ACMA overseeing its development and The ACCC recommended that ACMA Some technology companies, such as performance in addressing “the state of conduct regular monitoring of the Facebook, argue that the intent of the disinformation in Australia”.7

ACMA position paper

ACMA’s June 2020 position paper outlined its expectations and Consistent with the European Union’s recommendations for the code’s development. The paper came approach (see chapter three), a single industry-wide code was favoured.15 at a time of significant concern over misinformation in Australia. ACMA argued that a single code would Widespread online misinformation surrounding the 2019-20 bushfires engender greater efficiency, more and the COVID-19 pandemic highlighted the potential for the problem collaboration between platforms and to cause significant community harm.8 ACMA’s position paper was standardisation – leading to more informed by discussions with digital platforms including Facebook, streamlined complaints handling, reporting and measurement processes.16 Google, Twitter, Microsoft, TikTok, Apple and Amazon, as well as DIGI. It also suggested an “outcome-focused” Australian Government agencies such as the ACCC and international rather than process-focused approach, media regulators from the UK, Ireland and also contributed with signatories to the code having the insights and expertise. freedom to determine what measures to adopt to meet the stated objectives.17 In doing so they would publicly ACMA acknowledged that most digital million monthly active users in Australia,” “demonstrate commitment to addressing platforms had already implemented were considered to fall within the scope misinformation”18 and a willingness to measures to combat misinformation and of a future code.11 According to ACMA, participate in the code. disinformation, as well as to deal with this would cover Facebook, YouTube, issues around news quality in general. Twitter, Google Search and Google Nonetheless, the position paper argued News, Instagram, TikTok, LinkedIn, Definitions that greater codification, stringency and Apple News and Snapchat.12 However, As noted earlier, there is no general transparency were needed.9 Codification it was proposed that other private consensus within Australia (or elsewhere) was recommended on the grounds messaging platforms such as WhatsApp, on definitions of terms relating to false that it would guide new entrants into Weibo and WeChat be excluded from and harmful information online. DIGI’s the industry.10 the code – raising questions over why approach, which shares some aspects of Snapchat, which has similarities to the European approach, focuses on the those platforms, was included. Major term “disinformation” – which it defines Scope news outlets with an online presence broadly as “inauthentic” behaviour. ACMA argued for a voluntary code that were also excluded from the proposed According to DIGI, inauthentic behaviour platforms could individually implement. code because they were “not considered “includes spam and other forms of Digital platforms across the industry, key distributors of misinformation in deceptive behaviours (including via 13 comprising “online search engines, social Australia”. ACMA expected platforms automated systems) which encourages media platforms and other digital content to engage in public consultation during users of Digital Platforms to propagate 14 aggregation services with at least one the code’s development. content which may cause harm”.

5 Australian Competition and Consumer Commission, Digital Platforms Inquiry, 34. The report also has recommendations (12 & 13) around digital media literacy in the community and in schools. 6 Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia, 1. 7 Ibid., 4. 8 Ibid., 2. 9 Ibid., 9; 2 10 Ibid., 19. 11 Ibid., 3. 12 Ibid. 13 Ibid., 22. 14 Ibid., 25. 15 Ibid., 20. 16 Ibid. 17 Ibid., 23. 18 Ibid., 3.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 39 By comparison, ACMA uses pandemic.25 Academics James Meese Monitoring and quality- misinformation “as an umbrella and Edward Hurcombe, in their policy term to cover all kinds of potentially brief Regulating Misinformation, support control activities harmful, false, misleading or ACMA’s typology, with misinformation as As ACMA supports an outcome-based deceptive information, with deliberate an overarching umbrella term to capture approach, platforms are encouraged 26 disinformation campaigns considered grey areas between definitions. to adopt measures to contain 19 a subset of misinformation”. ACMA’s 34 ACMA divided harm caused by misinformation as they see fit. Some of position paper adopts similar definitions misinformation into two categories – the suggested measures included fact for individual terms to Claire Wardle’s “acute harms”, which it said, “have an checking, labelling of advertisements, First Draft, and to Hossein Derakhshan’s immediate impact on people, property flagging suspicious content, and typology often used in Europe (see or society”, and “chronic harms”, which providing media literacy tools. introduction and chapters three and “result from the cumulative effect of five). Thus, disinformation is identified as The ACMA position paper recognised misinformation and may only become “false or inaccurate information that is the importance of data sharing by apparent over time”.27 deliberately created and spread to harm platforms. It also recognised that the a person, social group, organisation or It also argued that the “concept of EU code had not yet fully achieved country.” 20 Misinformation is defined as news quality can be subjective,” further its goal of empowering research “false or inaccurate information that is complicating the task of measuring communities by overseeing data sharing not created with the intention of causing information credibility.28 ACMA also between platforms and independent 35 harm.”21 And malinformation is defined alluded to the wider societal value researchers. ACMA’s approach aimed 36 as “accurate information inappropriately of access to reliable news content, to address this gap, urging platforms spread by bad-faith actors with the saying quality news and information “is commit to “facilitate research, share intent to cause harm, particularly to the accurate, reliable and timely, providing relevant data and undertake activities operation of democratic processes”.22 people with the knowledge they need to improve the understanding of 37 ACMA acknowledged that it was to make informed choices and to misinformation in Australia”. 29 sometimes hard to distinguish between participate in public life”. To address a perceived lack of these types of malign information. ACMA proposed a graduated and transparency in the industry, ACMA ACMA also pointed out that potential risk-based approach, where measures recommended platforms publish harm was not always intended. Intention used in response to misinformation community guidelines outlining the itself “can be difficult to determine”, and disinformation were proportionate responsibilities of platforms and users, making “the line between bad actors to the risk the content posed. Adopting and promoting this information widely and innocent users… not always clear.”23 this approach “may also alleviate some through emails, landing pages and 38 ACMA found the 2019-2020 Australian concerns around freedom of speech”, tutorials. In doing so, ACMA highlighted bushfires and the COVID-19 pandemic since censorship and other strong a lack of transparency in the EU’s code 39 had highlighted how misinformation, measures would only apply to the most of practice. Facebook already had though not intentionally harmful, “has harmful content.30 Platforms would thus some examples of this kind of practice the potential to cause significant harm”.24 not be expected to serve as arbiters with its published Community Standards 40 Hence, ACMA said discussions limited of truth.31 Criteria to assess the risk and QandA sections on its webpages. to disinformation failed to account of harm would include content and ACMA also proposed a clear complaint for the threat posed by forms of false context; agent and purpose; breadth and handling process to enable users to and misleading information where speed of dissemination; and distribution easily “submit complaints about the ill-intent was not apparent. It noted channel.32 DIGI has also referenced the performance of the platform against the that the European code’s focus on need for a proportional response, and code and its application of published 41 disinformation would have failed to lists 11 different factors to consider policies and procedures”. This proposal address the types of misinformation before applying one or more measures to comes amid criticism of the EU code associated with the bushfires and counter disinformation.33 for failing to have clear performance indicators and meaningful objectives and commitments, which is said to have hampered monitoring and compliance.42

19 Ibid., 1. 20 Ibid., 9. 21 Ibid., 10. 22 Ibid., 10. 23 Ibid. 24 Ibid. 25 Ibid., 15. 26 James Meese and Edward Hurcombe, Regulating Misinformation: Policy Brief (Melbourne, Analysis & Policy Observatory, RMIT University), https://apo.org.au/node/309357 27 Ibid., 11-12. 28 Ibid., 9. 29 Ibid. 30 Ibid., 12-13 31 Ibid., 12. 32 Ibid., 15. 33 Digital Industry Group Inc., Australian code of practice on disinformation: Draft for public consultation (DIGI, October 16, 2020), 9-10, https://digi.org.au/wp-content/ uploads/2020/10/Disinformation-Code-draft-for-public-consultaion-.pdf. 34 Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia, 29. 35 Ibid., 7, 17. 36 Ibid., 17. 37 Ibid., 24. 38 Ibid., 30. 39 Ibid. 40 Guy Rosen, Katie Harbath, Nathaniel Gleicher, and Rob Leathern, Helping to protect the 2020 US elections, (Facebook, October 21, 2019). 41 Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia, 30 42 J. Pamment, The EU’s Role in Fighting Disinformation: Taking Back the Initiative (Washington D.C.: Carnegie Endowment for International Peace, 2020), 5; Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia, 16.

40 La Trobe University ACMA also proposed a monitoring and code” and evaluate “the overall impact” are guided by the self-regulatory and reporting framework to be developed of platforms’ contributions to combatting voluntary Media Entertainment and Arts with input from academia, digital misinformation in Australia.46 As part Alliance (MEAA) Journalist Code of platforms and government agencies.43 of its monitoring duties, ACMA would Ethics; the ABC Act 1983 (for the public Signatories would be required to advise the Government on whether broadcaster), and ACMA’s oversight of create an action plan and then further regulatory action was needed if the standards and codes for TV and report annually on their performance code objectives were not being met.47 radio broadcasters that contain specific against the plan. ACMA and/or other provisions stipulating “accuracy and independent organisations would also fairness” in content. Journalist David conduct rigorous assessments of Local reactions Swan wrote that, “the code would make 44 the signatories. Australian mainstream journalists and social media platforms responsible for misinformation in the same way Further, ACMA proposed a commentators were largely supportive that Australian media organisations “misinformation and news quality of the ACMA proposals, arguing that are bound by regulatory requirements monitoring and reporting framework”.45 technology companies should be bound for journalistic ethics and accuracy in This would guide industry on how ACMA by the same rules as newspapers, TV news reporting”.48 would “measure the effectiveness of the and radio. In Australia, journalism ethics

DIGI draft code: Australian Code of Practice on Disinformation

In October 2020, DIGI released a draft of the proposed Australian Scope Code of Practice on Disinformation for public consultation. Developed DIGI’s draft code embraces voluntary with assistance from the University of Technology Sydney’s Centre elements of ACMA proposals, with for Media Transition (CMT) and First Draft, an international project digital platforms to “opt-in” to whichever fighting disinformation and misinformation, the DIGI draft code measures they prefer depending on their responded to the Australian Government’s Regulating in the Digital business model and operational needs. Age49 – which was the Government’s official response to the ACCC’s Since “not all objectives and outcomes will be applicable” to all signatories, DIGI Digital Platforms Inquiry – and ACMA’s position paper. said each “may adopt one or more of the measures set out… in a manner that is relevant and proportionate to their DIGI also commissioned the CMT and approach – primarily the EU code and, different services and products”.52 The its affiliate First Draft to assist in the to a lesser degree, the approaches of opt-in provision would mean platforms 50 production of a discussion paper to Taiwan and . While acknowledging are only bound to those commitments accompany the draft code and provide the need for more action on the issue, that they have nominated.53 In theory, background context. The discussion DIGI also highlighted measures already this aims to accommodate the paper acknowledges the influence of implemented by digital platforms “to multiplicity of platforms and their international examples on DIGI’s tackle the propagation of disinformation” varying characteristics, functions and on their services.51

43 Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia, 38 44 Ibid., 24 45 Ibid., 36 46 Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia, 36. 47 Ibid., 4. 48 David Swan, “Code to rein in spread of fake data,” The Australian, June 26, 2020. 49 Australian Government. Regulating in the digital age: Government response and implementation roadmap for the Digital Platforms Inquiry (Canberra: Commonwealth of Australia 2019), https://treasury.gov.au/sites/default/files/2019-12/Government-Response-p2019-41708.pdf 50 UTS Centre for Media Transition, Discussion paper on an Australian voluntary code of practice for disinformation (DIGI, October 16, 2020), 39, https://digi.org.au/wp-content/ uploads/2020/10/Discussion-Paper-Final.pdf 51 Digital Industry Group Inc., Australian code of practice on disinformation, Section 1.3. 52 Ibid., Section 5.2. 53 Ibid., Section 7.1.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 41 uses. As stated in the discussion paper, on digital platforms following Referring to definitional issues, the “no single technological fix, labelling the summer bushfires ACMA spokesperson stated: system or filter” will suffice.54 Instead, the diversity across digital platforms and early stages of the “As a voluntary industry requires “an equally diverse approach to COVID-19 pandemic, we felt process, it is ultimately up to resolving the impact of misinformation this terminology provided 55 interested platforms to reach and disinformation”. While not directly platforms with the necessary opposing ACMA’s industry-wide a consensus on this issue. approach, DIGI’s proposal seeks greater flexibility to address a range We are continuing to actively accommodation of the industry’s of harmful behaviours and work with DIGI and platforms diversity and complexity. content, while encouraging to better understand each The DIGI proposal covers search engines graduated, proportionate other’s views on scope and platforms that accommodate user- measures that appropriately generated content. However, it excludes prior to the finalisation of 59 63 private messaging services, media texts reflect the level of harm. the code.” for entertainment or educational purposes, The ACMA spokesperson acknowledged email services and content authorised An affiliate at ’s the narrower scope of DIGI’s draft by the Australian Government.56 Hence, Berkman Klein Centre and fellow in Australian Code of Practice for the code excludes WhatsApp, Weibo and New America’s Cybersecurity Initiative, Disinformation when it comes to false other popular messaging platforms, as Camille François, uses a “disinformation news online: suggested by ACMA. ABC” framework – focusing on actors, behaviour and content – to understand DIGI has proposed a much harmful online content. Francois, Definitions narrower scope – addressing whose framework was cited in the DIGI discussion paper, identified “manipulative The discussion paper outlined several only inauthentic behaviour actors” (people or groups with the clear influences on DIGI’s conceptualisation from users of digital intention to disrupt the information of misinformation and disinformation. platforms that results in the ecosystem), “deceptive behaviors” Primary influences were the EU code propagation of potentially (tactics and techniques used by the and First Draft, while Britain’s Full Facts’ actors) and “harmful content” (used definition of levels of harm was also harmful digital content for to hurt, undermine or influence) as cited.57 Wardle and Derakhshan’s umbrella the purposes of economic “three key vectors characteristic of viral term of “information disorder”, which gain or to mislead the .”64 The DIGI-commissioned covers misinformation, disinformation public. Harm in this context report agreed that it was a useful and malinformation, was highlighted in rubric to understand the facets of the discussion paper (see also chapter is also defined narrowly disinformation. five).58 ACMA’s approach to definitions as imminent and serious differs from DIGI’s draft insofar as it threats to democratic While DIGI’s discussion paper canvassed defines misinformation as the overarching academic and practical definitions of term, under which disinformation and political and policymaking the problem, it resisted committing to malinformation sit as subsets. An ACMA processes or public goods a single approach – perhaps reflecting spokesperson said this was intended to such as the protection diversity of views among platforms that capture a range of harms: of citizens’ health, the comprise DIGI’s membership. environment or security.60 Disinformation is defined in DIGI’s draft We adopted ‘misinformation’ code as inauthentic behaviour, and as a catch-all umbrella term Like the ACMA position paper on covers content that is promulgated – reflecting any potentially misinformation, DIGI’s draft code “for the purpose of economic gain harmful false, misleading acknowledges the overlap between or to mislead or deceive the public, disinformation, misinformation and that may cause harm and is not or deceptive information similar terms.61 It accepts that there is otherwise unlawful”.65 This focus on distributed online, with or no universal definition of the problem, harm and intention is in step with other without malicious intent. In and that in the course of implementing approaches, including ACMA’s. The DIGI light of the types of material measures to combat it, such terms code also adopts a similar definition are likely to be used interchangeably of disinformation as the EU Code, we were seeing circulating by signatories.62 including “an imminent and serious threat” to “democratic political and policymaking processes”.

54 UTS Centre for Media Transition, Discussion paper on an Australian voluntary code of practice for disinformation, 25. 55 Ibid. 56 Digital Industry Group Inc., Australian code of practice on disinformation, Section 4.1; Section 4.2. 57 UTS Centre for Media Transition, Discussion paper on an Australian voluntary code of practice for disinformation, 10-11. 58 Clair Wardle and Hossein Derakhshan, Information disorder: Toward an interdisciplinary framework for research and policy making (Council of Europe Report, September 27, 2017), https://rm.coe.int/information-disorder-toward-an-interdisciplinary-framework-for-researc/168076277c 59 Interview 14. Correspondence with Andrea Carson, 18 November 2020. 60 Ibid. 61 Digital Industry Group Inc., Australian code of practice on disinformation, Section 5.4. 62 Ibid. 63 Interview 14. Correspondence with Andrea Carson, 18 November 2020. 64 Camille Francois, Actors, behaviours, content: A disinformation ABC: Highlighting three vectors of viral deception to guide industry & regulatory responses (The Transatlantic High Level Working Group on Content Moderation Online and Freedom of Expression, September 20, 2019), https://science.house.gov/imo/media/doc/Francois%20 Addendum%20to%20Testimony%20-%20ABC_Framework_2019_Sept_2019.pdf 65 Digital Industry Group Inc., Australian code of practice on disinformation, Section 3.2.

42 La Trobe University Consistent with broader discussions y Strengthen public understanding of arbiters of truth.75 Instead, they state that about fake news definitions, the UTS disinformation through support of their responsibility is to protect users discussion paper questioned the strategic research from potentially harmful content while inclusion of intent as a key component y Publicise the measures (signatories) upholding their freedom of expression.76 when defining disinformation, take to combat disinformation, DIGI’s proposal states that, in addition arguing that, “the question of whether including regular reports to to digital platforms, “a range of disinformation requires an element of government and/or the public.69 relevant stakeholders have roles intention needs careful consideration”. These objectives intersect with several and responsibilities in dealing with It said “the difficulties in establishing areas that ACMA wanted to improve, disinformation including public intention may make it more appropriate including data sharing, transparency, and authorities, academia, civil society, and to make conduct, rather than content, monitoring and compliance processes. news organisations”.77 It also stresses the focus of regulatory attention”.66 the need for collaboration between all The UTS paper observed that “not all DIGI proposed 16 further “measures” stakeholders, and endorses regulation manipulated content is malicious”,67 to which platforms could “opt-in” via a encouraging media literacy and digital which complicated the use of intent “nomination form”. While little description literacy as a way to ensure further as a marker of disinformation. was provided of what these measures sharing of responsibility for combating would specifically entail, the draft code the problem by news organisations and provides examples of specific actions the public.78 Guiding principles platforms might take, including content and objectives removal, partnerships with third-party groups and fact checking.70 The DIGI draft code outlines a set of Monitoring and “guiding principles” underpinning the In line with ACMA’s recommendations, quality‑control approach of digital platforms. These DIGI’s draft code suggests a graduated measures included the protection of freedom of risk-based approach to evaluating expression and user privacy, scrutiny disinformation. Under this model, Under monitoring mechanisms outlined of advertising placement, user measures would be “proportionate and in the code, DIGI would serve as the code 71 empowerment, integrity and security relevant to their specific context”. Risk administrator, and oversee a committee of services and products, supporting factors that platforms might consider that would meet every six months to independent researchers, and acting include the: evaluate signatories’ performances in 79 “without prejudice” towards other anti- y Types of actors involved meeting their nominated commitments. disinformation initiatives.68 y Breadth and speed of the content’s Signatories would report progress annually to DIGI, the Government and/ In step with ACMA’s recommendations, dissemination y or the public. Signatories would also DIGI’s draft code adopts an outcome- Level of harm y commit to a complaints-handling based approach. It’s six objectives are to: Behaviours and intent behind the online activity mechanism within six months of the y Improve “safeguards against y Credibility of the content code’s commencement, and submit to disinformation”, including preventing y Freedom of speech considerations.72 regular monitoring from DIGI to evaluate the spread of and exposure to code compliance. The code would be disinformation, and introducing better As with the EU Code, the importance reviewed after two years, with input from reporting mechanisms for users of freedom of speech is emphasised in all relevant stakeholders. y Disrupt advertising and monetisation DIGI’s proposed code. For example, the incentives for disinformation document states that signatories should y Work to ensure the public benefit of not be required to “remove content services and products delivered by solely on the basis of perceived falsity digital platforms if the content would not otherwise be 73 y Empower consumers to make better unlawful”. Requirements to “signal informed choices of digital content, the veracity” of user content are also 74 particularly around matters that may rejected. In this way, platforms uphold cause harm their “strong desire” to avoid being

66 UTS Centre for Media Transition, Discussion paper on an Australian voluntary code of practice for disinformation, 10. 67 Ibid., 36. 68 Digital Industry Group Inc., Australian code of practice on disinformation, Section 2.8. 69 Digital Industry Group Inc., Australian code of practice on disinformation, Section 5. 70 Ibid., Section 6.2. 71 Ibid., Section 6.1. 72 Ibid. 73 Ibid., Section 2.2. 74 Ibid., Section 5.6. 75 UTS Centre for Media Transition, Discussion paper on an Australian voluntary code of practice for disinformation, 25. 76 Digital Industry Group Inc., Australian code of practice on disinformation, Section 2.2. 77 Ibid., Section 1.6. 78 UTS Centre for Media Transition, Discussion paper on an Australian voluntary code of practice for disinformation, 22. 79 Digital Industry Group Inc., Australian code of practice on disinformation, Section 7.3.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 43 Comparing ACMA and DIGI positions on a new code

DIGI’s proposed code of practice responds to some key elements ACMA responded critically to DIGI’s in ACMA’s position paper. It supports a requirement on platforms draft code in comments reported in The Sydney Morning Herald in late October to share data with one another and with other stakeholders, 2020. Citing the absence of detail about including independent researchers, to address the spread of how DIGI would achieve the code’s online disinformation.80 The code also addresses concerns about objectives, ACMA chair Nerida O’Loughlin transparency, outlining commitments for financial support and non- was quoted saying: “The draft code interference in independent research. released by DIGI for public consultation is a long way from the model that we proposed to address these important 85 However, the draft code also combat disinformation. The ACCC has issues.” The Sydney Morning Herald demonstrates DIGI’s concerns about criticised as “opaque” existing measures article pointed out that DIGI’s proposed privacy and the “need for transparency and algorithms that platforms employ criteria for evaluating and acting on to be balanced against disclosure to demote misleading content. It argued contested content could allow material risks”.81 It states that, “in implementing this lack of transparency limited a from anti-vaccination activists or commitments to promote the public regulator’s capacity to measure their conspiracy theorists “not published with transparency of measures taken under efficacy in achieving the code’s stated malice” to escape action under the code. this code,”82 the release of information outcomes.84 ACMA has warned that if differences to the public opens digital platforms to DIGI’s proposals do not detail intended in perspectives between the parties on exploitation by malicious actors, “which consequences for platforms that fail to voluntary co-regulation are too great, a can result in more deceptive behaviour”.83 follow through on their commitments. government regulatory framework may In addition to maintaining user privacy, 86 This may be due to the assumption that be the inevitable next step. platforms are also concerned about under the monitoring and compliance protecting the integrity of their services procedures set out by the code, ACMA and existing measures that aim to will oversee such penalties.

Table 4.1: Expected dates for draft and implementation plan of Australia’s voluntary Disinformation Code

Estimated Date Expected Action June 2020 ACMA position paper

July-August 2020 Workshops to discuss ACMA monitoring and assessment

October 2020 Monitoring framework

October 2020 Industry draft code released for public review and consultation

24 November 2020 Public submissions to DIGI’s Draft Code close

December 2020 Industry code delivery deadline

January 2021 Action plan published by signatories

June 2021 Assessment report to government

Source: Author

80 Ibid., Section 5.19. 81 Ibid., Section 5.7. 82 Ibid. 83 UTS Centre for Media Transition, Discussion paper on an Australian voluntary code of practice for disinformation,19. 84 Australian Competition and Consumer Commission, Digital Platforms Inquiry, 1. 85 Zoe Samios and Lisa Visentin, “ACMA: Tech giants’ code to handle fake news fails to meet expectations,” Sydney Morning Herald, October 26, 2020. https://www.smh.com.au/politics/federal/acma-tech-giants-code-to-handle-fake-news-fails-to-meet-expectations-20201026-p568oq.html 86 Australian Communications and Media Authority, Misinformation and news quality on digital platforms in Australia, 37.

44 La Trobe University Conclusion

Australia’s approach to tackling the spread of misinformation on digital platforms so far is based upon voluntary cooperation between the major digital platforms (represented by DIGI) and Australia’s media regulator (ACMA). This is consistent with the European model of voluntary co-regulation to tackle online disinformation. As confirmed in this chapter, ACMA and DIGI share a stated commitment to addressing online misinformation and disinformation, and share some common ground about what is required to achieve it.

Their proposals are informed by key build consensus towards the voluntarily aspects of the European approach regulation of the quality of information, – in particular the shared preference and to adhere to a common standard for voluntary participation, and for an of conduct in the digital sphere, may outcome-focused approach. However, be jeopardised. the parties’ differing approaches The next three chapters report the to defining misinformation and original findings of this study. The disinformation – ACMA has a broad three chapters (chapter 5-7) focus conceptualisation of misinformation exclusively on the user experiences of capturing all forms of deceptive online misinformation and governments' behaviour, information, and conduct, and digital platforms' responses to it in while DIGI’s proposes narrower Singapore and Indonesia. These findings definitions – leaves them at odds on key provide rich lessons for Australia as it issues. Without a shared understanding considers its next steps to manage this of these key concepts, attempts to urgent, yet insidious problem.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 45 CHAPTER 5 What is online misinformation – and why is it a problem? Introduction

The lack of a clear, universally agreed definition of online In this chapter we explore in detail the views of participating experts – misinformation is perhaps the biggest obstacle to achieving technology company experts, journalists, consensus on how to tackle the problem in countries like editors, academics and community- Indonesia and Singapore. based activists – about the nature and scale of online misinformation, and their individual online experiences that Among those who participated in this there was general consensus on the helped shape their views. This provides research project, significant diversity of notion that it is a compelling problem in essential background to the next views emerged on: both Indonesia and Singapore, requiring chapter, which examines ways in which major remedial action. Participants governments and digital platforms in y What defines online misinformation broadly agreed that unchecked online Indonesia and Singapore can (and do) y The prevalence and scale of misinformation has the potential to manage perceived online misinformation, the problem seriously disrupt citizens’ access to and identifies particular areas that y The consequences for digital reliable information, and that this can contributors to this project believe platform users. have dire implications for democratic require the most attention. Yet, amid the contrasting viewpoints participation, freedom of speech and, in on the nature of online misinformation, some instances, public health and safety.

Definitions: Misinformation, disinformation and malinformation

As suggested in the introduction, online misinformation is a complex We accept that these definitions are not phenomenon that cannot be simply or universally defined. National standardised. For example, First Draft, a non-profit global coalition of newsrooms, governments, technology platforms, academics and other experts universities, online platforms and civil have so far failed to achieve consensus on precise definitions of society groups, led by academic Claire popular terms such as misinformation, disinformation and fake news. Wardle, recognises seven types of potentially problematic online content – or what it calls “information disorder” – in To try to mitigate the confusion and Thus, we characterise online three broad categories: misinformation, achieve consistency and clarity, this disinformation as a major subset of disinformation and malinformation1 report has adopted the following the broad, overarching problem of (see Figure 5.1 next page). definitions for online misinformation misinformation. By our definition, and disinformation: misinformation can be spread with or without ill-intent, while disinformation Online misinformation is the spread of necessarily involves ill-intent. We have inaccurate or misleading content online. adopted these definitions – and in Online disinformation is the spread of particular a broad interpretation of inaccurate or misleading content online misinformation – in part because they with conscious intent to mislead, deceive closely align with the usage of experts or otherwise cause harm. who participated in this project.

1 Claire Wardle, “The Need for Smarter Definitions and Practical, Timely Empirical Research on Information Disorder.” Digital Journalism, 6,8 (2018): 951-963.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 47 Figure 5.1: Seven types of “information disorder” ranging from low to high risk of harm

Source: First Draft, 20202

According to this schema, Facebook’s approach to the issue on misinformation, we also examine “misinformation” refers to verifiably starts from the premise that a person’s disinformation – in part because of false content that is spread without intention when posting information the inherent fluidity of these concepts the intention of causing harm. online is difficult to divine. The company and definitions. The variety of views 'Disinformation', by contrast, is false therefore defines misinformation based among participants in this project on information spread with the intention of on the veracity of the content, rather how to define the issue and its essential causing harm. The British Government than on the perceived intent of the components is reflective of a lack of clear is more expansive when defining user. And it defines disinformation with consensus in wider community debates. disinformation – referring to false or reference both to the actions of those Amid the debates, however, there is at manipulated information that has been who spread it, and to the content’s least some agreement among experts on produced or distributed for political ends propensity to cause harm. However, how to distinguish misinformation from that may be damaging to democratic it labels such activity “coordinated disinformation. According to one digital processes, or that may lead to personal inauthentic behaviour” rather than the platform expert, a useful way to define or financial harms.3 Wardle identifies more popular disinformation.4 misinformation is by what it is not: a further sub-category of problematic Wardle acknowledges that information online content – malinformation – which disorder is a “fluid spectrum”. This is Misinformation is false and she defines as truthful information readily observable in the ways false intended to cause harm, such as misleading information that content can be variously classified as malicious gossip. is shared not necessarily misinformation and disinformation, Wardle’s fake news typology, like this sometimes depending on its timing, and out of bad intent. It is shared report’s terminology, is not universally how it can have varying consequences, when people may not be accepted, as the extensive debates and depending on who circulates it, how fully aware that it’s false. disagreements outlined in this chapter it is spread, and why. So while the And that’s very different reveal. overwhelming focus of this chapter is from disinformation, which is false information that is shared by a person who knows that it’s false, and they do so with an intention to mislead other people.5

2 Claire Wardle, “The Need for Smarter Definitions and Practical, Timely Empirical Research on Information Disorder.”, 953; Figure used with permission from First Draft, email correspondence with author, 17 November 2020. 3 James Meese and Edward Hurcombe, Regulating Misinformation: Policy Brief (Melbourne: RMIT University, 2020), 1. 4 Nathanial Gleicher, Coordinated Inauthentic Behavior Explained. (Facebook, 2018, 6 December), https://about.fb.com/news/2018/12/inside-feed-coordinated-inauthentic- behavior/ 5 Interview 1, interview with Andrea Carson, 7 August 2020

48 La Trobe University The Editor-in-Chief of Tempo magazine When misinformation is defined by Thus, according to the typology of in Indonesia, Wahyu Dhyatmika, stressed the veracity of the content, a logical Wardle and others, a false story can the importance of intention when response to the problem is fact be either, or both, misinformation and separating the two concepts and referred checking.8 By contrast, managing disinformation depending on who is to the First Draft concepts: disinformation tends to involve tracking spreading it and why, and its impacts and and curbing behaviours and, if harm is potential for harm. occurring, removing the content and/ Misinformation is when Another key consideration is or the user.9 However, distinguishing false information is transparency. Historian and Managing between misinformation and published or shared widely, Director of the New Naratif in disinformation can be more complicated Singapore PJ Thum believes the lack of not with ill‑intent, but mostly than just focusing on the content and transparency about who’s behind a fake behaviour, as one academic explained: because of ignorance. news story and what their intent is, are Disinformation is more It gets muddled up when the main problems with misinformation dangerous because there’s online. He argued that unlike real news, an individual picks up an intention to spread where reporting agendas are more disinformation and transparently subjective, those behind confusion or anxiety.6 spreads it unknowingly. fake news disguise their subjectivity: But some observers, including False information may With fake news, the problem participants in this project, find this be deliberately planted, emphasis problematic because, they is that it tries to present itself argue, it is not possible to know an designed to be picked up very often in a way that is individual’s intentions when sharing false by individuals, and then objective in order to mislead, information online. Hence, some experts spread by individuals on are more focused on the behaviours to influence politics and their closed social media influence public opinion. But of online users and the veracity of the 10 content rather than intention. networks. it’s not transparent about its Technology companies have become This potential for disinformation to be intention. So I think those particularly focused on behaviours of unwittingly distributed by others who are the two main things – individual users when seeking to identify believe it to be true adds another layer designed to mislead, and online disinformation. For example, a of difficulty to defining, and therefore “super poster” who generates the same mitigating the problem. And the lack of transparency about content hundreds of times can attract emergence of video and pictorial content who’s behind it, where it’s scrutiny in this context, as can people such as “deep fakes” – doctored images coming from and what are setting up multiple accounts with fake of individuals, often celebrities – further its intentions.12 identification and other apparently complicates these issues: “inauthentic” actions. By contrast, the Wahyu Dhyatmika argued that platforms tend to identify misinformation A deep fake is a special anonymous and fake automated not by behaviour but by the factuality of classification where (bot) accounts made it easier for the content. One digital platform expert it’s a hybrid between organisations and individuals to attack, described the approach thus: bully and spread misinformation. He said misinformation and his outlet often received anonymous Misinformation typically disinformation. It’s online threats after publishing critical deals with content, false misinformation, because stories about powerful individuals and information, being shared, it’s usually synthetic and organisations, including the Indonesian being spread. We believe that Government. “Most of the threats come fake. It’s also disinformation from bot accounts, so it’s orchestrated. disinformation is not about because someone has to be But we never can pinpoint these content, because someone purposely trying to create cyber‑attacks.” proposing disinformation it to deceive. So there is an The case for banning and removing may be doing it based adversarial motive behind anonymous and bot accounts to reduce on truth – they may be the scenes.11 misinformation online might seem compelling. However, such an approach promoting true things – but could have unintended consequences in misrepresenting it in certain countries where some activists rely on ways. And therefore we look anonymity to circumvent government at users’ behaviours.7 restrictions on freedom of expression.

6 Author interview with Wahyu Dhyatmika, Editor-in-Chief of Tempo magazine, 24 August 2020 7 Interview 4, interview with Andrea Carson, 7 August 2020 8 ibid 9 Interviews 4; Interview 9, interview with Andrea Carson, 8 September 2020 10 Interview 3, Interview with Andrea Carson, 18 August 2020 11 Interview 4 12 Author interview with PJ Thum, Historian and Managing Director of the New Naratif, 19 August 2020

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 49 As one platform expert stated: Hoax is relatively now, I think the difference similar to the concept of is that for, especially It’s a complicated issue. We disinformation, meaning online, it is being spread have very clear policies, that the information is not by companies, but by like the “real name policy” not accurate, and it is individuals. So your friends so that we can remove fake intentionally produced or are now spreading such accounts. But then there are distributed to harm people. information, not just the certain jurisdictions where Misinformation is actually mainstream media.18 it’s harder to enforce because different in terms of its people in opposition under As discussed in the next section, intention.17 authoritarian regimes rely ambiguity around the definitions of misinformation and disinformation, and on those fake accounts to be Misinformation and disinformation the overlap between them, can present have been around for centuries. What’s able to express themselves serious challenges for authorities trying different now is that the internet has without persecution.13 to deal with false information online – enabled and facilitated the easy and rapid in particular for persuading technology spread of false information by almost Further, not all fake content is platforms and citizens, including any individual, and on a scale previously potentially harmful. A fake account set activists, to comply with state measures unimaginable – in volume and audience up in a pet’s name would be in most to remove it. cases an example of harmless online reach. The mass penetration of online misinformation. An online parody relying misinformation into the everyday lives of on obviously false content might also citizens in countries like Indonesia and Government definitions be deemed harmless. But where should Singapore has led some observers to Although government representatives the line be drawn? One interviewee said: suggest people may now be more alert did not participate in this project, several “When it comes to fake accounts, are we to falsehoods and crude propaganda interviewees shared their experiences talking about malicious fake accounts? than in the past. PJ Thum said that in and insights into how legislators in What is it that we’re trying to solve here?” the first decades of advertising, citizens Singapore and Indonesia view and define Social media and digital platform experts were more susceptible to claims that, misinformation and disinformation. “A argued that more clarity and agreement by today’s standards, would seem lot of countries look at misinformation on definitions of misinformation and “obviously invented” or “exaggerated”. and disinformation as a content problem, disinformation would enable all platforms At the same time, despite increasing and that makes it really hard for us,” to operate more consistently and develop public awareness of propaganda and said a digital platform expert.19 The consistent standards.14 This included false claims, the “people who run tendency of governments to conflate avoiding the broad term “fake news”. misinformation campaigns are getting more sophisticated at it,” said PJ Thum. misinformation and disinformation can have significant consequences We don’t really use the term One academic argued there should when it comes to prosecution of cases. fake news, because we be no distinction between online and For example, a true believer in a false understand that it means offline when it comes to regulating may spread the misinformation: different things to different information with a virtuous intention to “enlighten” others.20 But if governments people… So the term that I think intention is key. define this content as disinformation, we use a lot is false news, We’ve always had human they “risk criminalising people who because then it’s very errors as a result of being themselves have been duped.” This neutral and it doesn’t make a rushed with deadlines punitive approach, arising out of a catch- 15 all definition and a legislated response, judgement about intent. to get to the printers and may make it more difficult for Facebook so forth. I think the issue and other social media companies to One academic disliked the term fake combat disinformation: news for another reason – it was an with misinformation is oxymoron. “I don’t agree with the the deliberate attempt Targeting or criminalising concept of fake news because news to influence using false is not fake.”16 people sharing this type of information. We’ve had misinformation may not be Indonesians commonly describe fake or [propaganda] like in the false news stories as “hoaxes” – though the best solution to solving usually in reference to disinformation Hearst newspapers in the disinformation because rather than misinformation, according US [during the Spanish- it simply drives them to participants in this project: American war]. So I don’t see underground and then they a difference there. It’s just try to hide their sources.21

13 Interview 13, Interview with Andrea Carson, 16 September 2020 14 Interview 13 15 Interview 1 16 Interview 7, interview with Andrea Carson, 19 August 2020 17 Interview 7 18 Interview 3 19 Interview 4 20 Interview 4 21 Interview 4

50 La Trobe University Such situations may encourage Hence, the company was primarily At the time of interviews for this conspiracy theorists to create fake online focused on fact checking to ensure the project in August and September 2020, accounts. And if there are many fake accuracy of content. And when it came platforms maintained an exemption for accounts, “that creates lots of noise,” to government requests to have content political communications from their said the digital platform expert. “How corrected or removed, the company’s “malicious actors” category of content do we locate the really bad ones in a response was “based on… whether the to be removed if it involved politicians.30 timely manner?”22 correction is correct. So if the content is One rationale for this exemption was accurate, and the government is trying that politicians were not considered Fake Facebook accounts – which to promote false information, we would malicious actors.31 the company says its reports show refuse to comply.”26 constitute about five per cent of worldwide monthly active users – are While Facebook sees advantages in Malicious to us mean now officially banned under company having separate divisions dealing with you’re trying to lie, policy. In the second quarter of 2020 misinformation and disinformation, misrepresent yourself to alone, Facebook said it blocked this approach can lead to problems. someone. Politicians don’t and deleted about 1.5 billion fake As previously mentioned, the same accounts.23 However, as mentioned false story can be deemed both misrepresent themselves. previously, this policy has potentially misinformation and disinformation, They may misrepresent negative implications for activists depending on who is spreading it and their positions. That’s not vulnerable to political persecution in less why. Dealing with misinformation a problem (for us). We don’t liberal countries. separately to disinformation can also potentially delay an organisation’s think that’s inauthentic. Digital platform companies also response times. That’s authentic. If they are have misgivings about the binary approach of governments to defining A disinformation specialist at lying politicians, they’re misinformation – saying they tend to one platform said the in-house authentically lying.32 divide content simply between real and disinformation team was mostly false. This approach failed to address concerned with various issues including The company position meant the need for a balance between free account security (accounts being that “really blatant” falsehoods or and “responsible” expression. It was misused) and transparency to prevent misleading content in political speeches used by some governments to favour malign actors deploying propaganda were exempt from the platform’s fact censorship over freedom of speech. on the site, including propaganda from checking policy.33 As one digital platform expert said, the state‑controlled media. The platform policy expert said the Indonesian Government “considers that platform did not allow politicians to all misinformation are of equal weight The primary thing that spread misinformation that could and should not be tiered based on topics we (disinformation team lead to harm, such as harmful COVID 24 such as prevalence and virality.” members) do is protect our misinformation, or misinformation that platform and our users could lead to violence. “We do not allow Social media politicians to share such misinfo and against influence operations. have removed a number of politicians’ companies’ definitions (But) we don’t look at content, content under the policy.”34 27 and policies we look at behaviour. Since the interviews were conducted (and at the time of writing), both With their policy of explicit differentiation Amid the spread of false information between misinformation and Facebook and Twitter have adopted and bogus remedies surrounding proactive stances on alleged false disinformation, some platforms have COVID-19, digital platforms have recently separate divisions and staff assigned content surrounding the 2020 US acknowledged that well-meaning Presidential election, with both to deal with cases of both as they misinformation can also cause great arise.25 A digital platform expert said platforms placing warnings on some harm. Facebook has consequently political content. that under their platform’s approach to developed a “misinformation and misinformation, content was assessed harm” policy, allowing it to remove The most notable examples were labels based on “what the content says, as misinformation with the potential to added to tweets by President Donald opposed to the user’s intent,” which he cause harm to users.28 The platform has Trump claiming voter . Twitter said was “very difficult to judge.” another harm-related policy surrounding added the words, “This claim about hate speech, which allows for removal election fraud is disputed,” and provided of any discriminatory content based on a link to information discounting the personal characteristics including race President’s claim (see Figure 5.2 on 35 and religion.29 the next page).

22 Interview 4 23 Facebook. Fake Accounts. (Facebook, 2020). https://transparency.facebook.com/community-standards-enforcement#fake-accounts 24 Interview 9, interview with Andrea Carson, 8 September 2020 25 Interview 9 26 Interview 1 27 Interview 4 28 Interview 9 29 Interview 1 30 Interview 1; Interview 4 31 Interview 4 32 Interview 4 33 Interview 1 34 Interview 4 35 Twitter. Voter fraud of any kind is exceedingly rare in the US, election experts confirm (Twitter, 2020) https://twitter.com/i/events/1308626736066617344

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 51 Figure 5.2: Donald Trump’s contested Tweets

Source: Twitter, Donald Trump account 12 November 2020

Civil society Community-based actors such as tackles online hoaxes in Indonesia, non-government organisations and Septiaji Eko Nugroho, said his group organisations and volunteers tend to adopt definitions embraced First Draft’s three-category academic perspectives of misinformation used by academics schema, entailing “misinformation, and/or multinational organisations disinformation and malinformation”. on definitions such as the United Nations. The Chair This was also embraced by UNESCO. of MAFINDO, a civic organisation that

Identifying the scale and effects of online misinformation

Interviewees for this project were equivocal on whether the problem Authority as breaching laws on regulated of online misinformation in Singapore and Indonesia had been curbed, goods. Restricted access was applied to one item relating to alleged defamation.38 or was getting worse. Part of the problem was that the quantum of misinformation was so difficult to measure and compare over time. It While Facebook reports twice a year on removals or restrictions on was also dependent on country context. misinformation and disinformation based on local laws, misinformation is generally not reported for individual nations. As a A digital platform expert acknowledged In Indonesia, Facebook removed 170 company employee explained: the difficulty: items between July and December 2019 in response to requests by the We’re not doing it because That is actually hard Communications and Informatics it is actually quite difficult. to answer because our Ministry (KOMINFO) and the Indonesian Content that is in English capability has been National Police over alleged violations of laws related to treason and separatism, language is used in various expanding. So we’re unlawful assembly, manipulated images, countries. One item could detecting more. But that promotion of regulated wildlife products, 37 be posted in one country, might be because we’re and extremism. consumed in other countries now more capable, or it During the same period in Singapore, and reported from another may be because there’s 16 items were removed, including 10 allegedly containing material inciting country. 39 more misinformation and racial hatred, strife and intolerance. disinformation. We don’t Restrictions were also placed on five have a whole picture.36 items reported by the Health Sciences

36 Interview 4 37 Facebook. Restrictions by Product: Indonesia. (Facebook. 2020), https://transparency.facebook.com/content-restrictions/country/ID 38 Facebook. Restrictions by Product: Singapore. (Facebook. 2020), https://transparency.facebook.com/content-restrictions/country/SG 39 Interview 1 52 La Trobe University However, for languages confined largely Figure 5.3: Facebook provides context to news posts using the info button to one nation: “I do think there could be a way to release more information or to be more transparent … But that’s a work in progress.”40 One platform expert in Singapore believed online misinformation seemed less prevalent in the city-state in late 2020 compared to other times: I think it’s been a bit quieter lately. I would like to think it’s at least partly because of our work with the fact checkers… We [are] able to get to at least the most viral, the most potentially harmful Information misinformation.41 button

Facebook’s strategy to deal with “false news” is three-fold42 involving removing, reducing and checking quality of information: y Remove accounts and content that violate its Community Standards or Source: Author ad policies y Reduce the distribution of false news If a video is watched by COVID-19 is a new topic that and other inauthentic content like 20 million people and is bringing a lot of this out. clickbait by altering algorithmic flows it contains harmful That doesn’t mean that the y Extensive fact checking y Adding context to news posts. misinformation… The fact other kind of disinformation that we were able to remove networks have stopped. In some jurisdictions, Facebook has added a circular information or context it after six hours, to me, They’re still going on, and we button with an “i” in it on the bottom right shows that we’re getting need to prioritise COVID-19 corner of shared news stories to enable better at these things but work on them too.44 users to make judgements about the because before it might have veracity of news stories (see Figure 5.3). He said the problem was particularly When clicked the button provides taken us a lot longer. But if acute in Asia. “The way this information information about what a publisher has it was already watched by works in APAC is slightly different than previously posted, its page 20 million people after six the way it works out in Europe or the and where the post has been shared. It hours, clearly we have to do a USA. There [in Europe and US] the actors also provides the user an opportunity to 43 tend to be much more organised.” follow that news outlet to add “quality” lot better. The Chair of MAFINDO, Septiaji Eko news to their feed. The sheer volume and rapid spread Nugroho, hoped that fact checking None of the digital platform experts of false information about COVID-19 and media literacy initiatives spreading interviewed for this project believed that underlined the broader potential of across Indonesia would help mitigate the problem of online misinformation had online misinformation for global harm. online misinformation. However, there been adequately tamed – yet. One A digital platform expert said the was still much progress to be made platform expert cited the technical and pandemic presented new and difficult on media literacy, which would enable temporal difficulties in addressing false challenges for technology and social Indonesians to more readily identify false content quickly: media companies, with global conspiracy news and avoid sharing it: theories, misinformation and sometimes state propaganda “all mixed in,” while the Even some people with platforms were still trying to deal with high education still share other sources of false information as well: misinformation, so it is quite complex. It’s not only related to the level of education; people still don’t understand how to read online information correctly.45

40 Interview 1 41 Interview 1 42 Facebook. Hard Questions: What’s Facebook’s Strategy for Stopping False News? (Facebook, 2018, 23 May) https://about.fb.com/news/2018/05/hard-questions-false-news/ 43 Interview 1 44 Interview 4 45 Septiaji Eko Nugroho, interview with Andrea Carson, 26 August 2020 Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 53 Common forms of Box 5.1 Indonesia: of the ruling party, and sometimes 53 misinformation combining political and journalists and digital platforms. health misinformation From its enactment in May 2019 up Political misinformation and hate Pro-Prabowo netizens usually until October 2020, POFMA was used speech are considered among the most 71 times to prosecute cases of alleged prevalent and problematic forms of produce a lot of hate speech on how Indonesian President Joko Widodo online misinformation. Most orders online misinformation in both Singapore (known as Jokowi) handled COVID. under POFMA have been “correction and Indonesia. One platform expert Like they accuse Jokowi of creating directions” requiring a correction notice said that if misinformation from “hate the conspiracy theory on COVID, that and a link to the facts alongside a groups” was placed under the banner of COVID is not real, COVID is about statement about why the content has political misinformation, “then politics [a] money arrangement between been deemed false. Other orders require would be the most common form of Indonesian government [and the] the removal of content and blocking misinformation.”46 He highlighted the World Health Organisation.51 websites. Facebook had received more potentially dire consequences of hate than 20 of these orders, including speech mixed with politics: orders to post correction notices and While Facebook has been widely used in to disable access to web pages at the I think there’s more both Indonesia and Singapore to spread time of writing. Twenty orders were misinformation in this APAC misinformation, the Facebook-owned posted during Indonesia’s general region that leads to real and encrypted WhatsApp has also been election period from 23 June and 10 world harm in the sense of a popular platform for misinformation July 2020. Many of these related to spreaders in Indonesia. questions around government policy,54 physical violence, as we’ve raising concerns that POFMA was being seen in and Sri It’s particularly bad because used by the ruling party to shut down Lanka, and Indonesia to a of the number of WhatsApp political debate and dissent. As one certain extent.47 users in Indonesia. One academic observed:

Damar Juniarto observed that political person tends to have more I think the fear now is misinformation tended to focus on than one (account) and this is a slippery slope individuals rather than parties’ policies. so the approach to tackle kind of situation where Politicians and candidates were misinformation has to POFMA will be used common targets for misinformation be different because it’s “hoaxes” and rumour.48 as a tool for politically end‑to‑end encrypted. The next most common type of false motivated censorship. news was the “mischievous” variety, Facebook’s misinformation strategy During the elections, there such as conspiracy theories and on WhatsApp has included limiting the were instances where it related hate group content.49 False number of times that messages can bordered on politically and misleading content created to be forwarded: “You can only forward motivated censorship.55 make money was also becoming more a message to up to five people – this common, according to a technology is now a global feature, not just in platform expert. These types of false Singapore and Indonesia,”52 a Facebook content align with Wardle’s typology in expert confirmed. Figure 5.1. In Singapore, much of the content Misinformation about health issues such targeted under the Government’s as COVID-19 was also prominent, and Protection from Online Falsehoods and in Indonesia was sometimes blended Manipulation Act (POFMA) has focused with political conspiracy theories (see on statements by political opponents example below).50

46 Interview 4 47 Interview 1 48 Author interview with Damar Juniarto, Executive Director SAFEnet, Southeast Asia Freedom of Expression, 31 August 2020 49 Interview 4 50 Interview 11, Interview with Andrea Carson, 20 August 2020 51 Interview 7, Interview with Andrea Carson, 19 August 2020 52 Interview 1 53 Interview 11, Interview 3, Interview PJ Thum 54 Aqill Haziq Mahmud “IN FOCUS: Has POFMA been effective? A look at the fake news law, 1 year since it kicked in,” Channel NewsAsia, 2020, 3 October, https://www.channelnewsasia.com/news/singapore/singapore-pofma-fake-news-law-1-year-kicked-in-13163404 55 Interview 11

54 La Trobe University Negative consequences of online misinformation

Participants in this project identified a wide spectrum of negative At the time of writing, the policy’s consequences attributed to the spread of online misinformation in activation was confined to countries in Asia Pacific, Africa and the Middle East Indonesia and Singapore. These included emotional harm to members because of their specific requirements of targeted minority and vulnerable groups such as the LGBTI and limited digital literacy. community and women (in the case of misogyny); physical harm arising out of misinformation that incites violence; racial and religious Misinformation in this based vilification; and public confusion and distrust resulting from region [APAC] is unique citizens being unable to distinguish misinformation from fact. because, while obviously there are a lot of really bad things going on in the US Emotional and Impacts on election related to misinformation, over here when people physical harm campaigns, political hear some rumours or a Real world harm is more often vilification completely false story that associated with online disinformation The potential of misinformation to has been debunked in many rather than misinformation. But interfere with election campaigns misinformation can also cause distress – cases, people have lower and political communication was a for example, through online harassment significant concern among participants digital literacy. Immediately, and bullying – and physical harm through in this project. While decades of research I think it could result in misinformation that incites violence, or findings are mixed on the impact that the 57 adverse health outcomes due to false violence. media, including social media, can have information about COVID-19 treatments on voters’ decision making, respondents In Indonesia, the prevalence of political or other diseases. Septiaji Eko Nugroho expressed strong views on the subject. misinformation online tends to be cited a video clip online featuring a highest around election times, with Although Facebook has historically celebrity with 3 million followers who the risks and problems particularly removed posts deemed to be interviewed a purported professor acute in regional and local contexts. disinformation, it was only relatively offering medicine to treat COVID-19. “But Technology platforms have found that recently that it adopted a similar option actually he’s not a professor, and it is not working with trusted community-based for misinformation – specifically for really medicine. It’s like herbal remedies.” organisations can help them detect cases where there is the potential for Andreas Harsono of Human Rights early signs of trouble or the potential for harm. A platform expert explained: Watch for Indonesia agreed that misinformation campaigns to circulate. misinformation could be dangerous We saw that misinformation However, this can be difficult to achieve and deadly. He cited the example of in remote and rural areas: controversy and tension surrounding around the Rohingya crisis the re-election bid of former Jakarta in Myanmar, and the Easter governor Basuki Tjahaja Purnama bombing crisis in Sri Lanka (known as Ahok). He believed the public really led to people attacking enmity toward Ahok – fed by the false claim that Ahok had blasphemed Islam other people. And we didn’t – could have escalated: “To some extent, have the policy at the time we were lucky the Islamists won (the to remove misinformation. election). If not, things might have been And we don’t want to just very ugly.” remove content, you know, A digital platform policymaker also saw first-hand how misinformation without a principled policy, surrounding the 2019 general election so we created that policy, in Indonesia could trigger violence56 misinformation harm. (see case study 5.2 on the next page).

56 Interview 9 57 Interview 1

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 55 Box 5.2 Ahok Case Study Indonesia today. In legal terms, that Ahok When it comes to was ultimately convicted for blasphemy, After giving a speech in September sub‑regional elections despite prosecutors opting to pursue 2016, Jakarta governor Basuki Tjahaja the lesser charge of hate speech, raises we don’t have partners Purnama, known as Ahok, faced doubts about the fairness of the judiciary there on the ground to allegations of blasphemy against the and one’s capacity to access even basic Qur’an. Ahok, a Christian and ethnic flag misinformation that legal protections.63 Chinese, was claimed to have stated: could potentially turn into “You don’t have to vote for me and you The video supporting his conviction was real world harm. So these probably won’t, if you’ve been misled by initially not found to have been doctored conversations need to be those using Surat Al Maidah 51 (Verse by the court. The video was used as 68 51) of the Qur’an in a misleading way.”58 the primary evidence against Ahok to initiated. This verse holds that Muslims should convict him of Blasphemy. Later in the Recognising this issue, Facebook has not let Christians or Jews be their auliya, year, however, the video uploader was in recent years worked to develop often interpreted as “leader.”59 found to have incorrectly transcribed the video of Ahok, misleading the public, partnerships with community groups While initial responses to the speech and incriminating Ahok. The man was in conflict-prone areas to cement its were muted, footage of the speech was convicted under the ITE law for the act presence in lesser-developed provinces, uploaded to YouTube where it became of tampering with the footage.64 The strengthen its election integrity activities viral. Crucially, the man who uploaded example illustrates the possibility that and to overcome the challenges of the video was eventually found to have manipulated content can be used to operating in a country with hundreds incorrectly transcribed what Ahok had prosecute and convict. of languages, and where artificial said, making it appear as though his intelligence technologies may not pick remarks were critical of the Qur’an itself While it is difficult to confirm whether up early signs of disinformation and and not his political opponents who Ahok's conviction was politically misinformation: were allegedly using the Quranic verse to motivated, as some have argued,65 marginalise non-Muslim leaders.60 the vehemence of public outrage We select conflict-prone against him is seen to have caused Alok’s comments led to massive areas that intersect with Jokowi, his former gubernatorial and occasionally violent public running mate, to grow wary of the regional elections in places demonstrations in Jakarta in the political threat posed by Islamist like Papua or Ambon Island, following months, where hundreds groups, leading Jokowi to increasingly of thousands of Muslims called for which have a history of quash opposition through the ITE law his arrest for insulting Islam.61 In May conflict, and we set up a and further align with mainstream 2017, Ahok was sentenced to two years conservative Islamic organisations.66 smaller team of trusted jail and thus could not complete his As such, a “political mainstreaming” of partners and academics term as governor. Since his release in conservative Islamism has occurred, 2019 Ahok has declared his career in that we could potentially furthering Indonesia’s perceived shift government over.62 toward illiberalism.67 work with that could In several ways this case encapsulates potentially flag to us early the political, legal, social and on misinformation that misinformation challenges that exist in could surface. When we go to different parts, the misinformation is very much localised – it’s very thick on local context. That would not normally surface for someone that sits in Jakarta.69

58 Butt, Simon, “Why is Ahok in prison? A legal analysis of the decision”, Indonesia at Melbourne, 6 June 2017, https://indonesiaatmelbourne.unimelb.edu.au/why-is-ahok-in- prison-a-legal-analysis-of-the-decision/ 59 “Why is Ahok in prison?. 60 Soeriaatmadja, Wahyudi, “Man who uploaded controversial video of ex-Jakarta governor Ahok sentenced to jail”, The Straits Times, 14 November 2017, https://www.straitstimes.com/asia/se-asia/man-who-uploaded-controversial-ahok-video-sentenced-to-jail 61 Agence France-Presse, “Jakarta protests: Muslims turn out in force against Christian governor Ahok”, The Guardian, 2 December 2016, https://www.theguardian.com/ world/2016/dec/02/jakarta-protests-muslims-against-christian-governor-ahok 62 Bhwana, Petir Garda, “Ahok reflects on never becoming a government official again”, Tempo.co, 22 October 2019, https://en.tempo.co/read/1262677/ahok-reflects-on-never- becoming-a-government-official-again 63 Butt, Simon, “Why is Ahok in prison? A legal analysis of the decision”, Indonesia at Melbourne, 6 June 2017. 64 “Man who uploaded controversial video of ex-Jakarta governor Ahok sentenced to jail”. 65 Butt, Simon, “Why is Ahok in prison?” 66 Warburton, Eve and Aspinall, Edward, “Explaining Indonesia’s democratic regression: Structure, agency and popular opinion”, Contemporary Southeast Asia 41(2), 2019, p. 271. 67 Ibid., p. 260. 68 Interview 9 69 Interview 9

56 La Trobe University A particular concern among interviewees Tackling misinformation shared on Misleading the public was the tendency for politically- WhatsApp is difficult because of motivated misinformation to be end-to-end encryption of the shared It is self-evident that online weaponised against individual opposition content. Facebook has tried to misinformation can mislead the public. candidates and their supporters, overcome the problem by limiting the But how much harm can it cause, and particularly in Indonesia. number of times individuals can share how effective are measures to mitigate the same information. But this has it? As Pew Research Centre studies have They are attacking the had the unintended consequence of confirmed, misinformation can make candidate, not attacking the disadvantaging small business owners it difficult for citizens to determine real news from false information, affecting program… even attacking who have used WhatsApp to promote their goods and services. As a Facebook their capacity to make informed choices, the belief of the people who expert explained: particularly at the ballot box.74 support the candidate, like We want to support the But as explained in the example below, saying that this person is correcting even harmless fake news can not Islam enough to be our Indonesian Government to create confusion by reiterating the false leader or, as in the Ahok case address misinformation. But information. This is sometimes called the “backfire effect”. in 2017, this person is ‘kafir’. one community that shouted So that means if you are like the loudest when we actually him, you are not following made those adjustments Box 5.3 Singapore: the rules of Islam… The use are the small and medium “plastic rice” of political identity as a way businesses that thrive on In Singapore there is kind of a famous of propaganda is something WhatsApp. You can win fake news, and it happens every now that’s more effectively used in one area and lose in the and then, of plastic rice, and the news other.72 will go around that, “hey, this rice is than criticising the political plastic, looks like real rice but plastic. 70 program. In Singapore, the prevalence of alleged It’s being sold in the supermarket. online misinformation during election Don’t buy the rice from the supermarket Damar Juniarto observed this campaigns has raised different issues. because they are selling this plastic phenomenon in the actions of the “cyber Some observers have accused the rice.” So if you look at that, then the armies” and “militias” who engaged in Singapore Government of overreach in harm is not so serious. But, the thing trolling and “doxing” (posting private its attempts to combat misinformation about trying to correct information is information with malicious intent) of in the lead-up to national elections in that you have to tell people what the political opponents. July 2020. While acknowledging that false information was. And I think the research in this space has shown that There’s so many people misinformation relating to elections is problematic, one academic said when you do that, some people believe 75 being targeted. So we are the Government was conflicted in the false information. working together with other its self-appointed role of overseeing organisations, we cannot do and regulating election-related this by ourselves. That’s why misinformation: Racial and religious we are forming our national There should be an vilification coalition against persecution independent group set up to The case of former Jakarta governor to protect the people and ask evaluate false information Basuki Tjahaja Purnama (Ahok) the police to investigate and in that election period, and provides a high-profile example of how online misinformation can be used to to (find the) organisations then it should be disbanded. vilify individuals and groups on racial behind the attacks. And that’s it.73 and religious grounds, sometimes for political purposes. Ahok’s Christian Political misinformation during election Similarly, for other forms of faith made him a target of hostile online campaigns is frequently shared through misinformation with the potential to misinformation during his last election informal and personal networks such cause harm (such as COVID-19 related campaign – and Andreas Harsono as WhatsApp – the Facebook-owned false content), he recommended believes this was central to his defeat. platform that is very popular among following the model spawned by More broadly, Septiaji Eko Nugroho says Indonesians, and particularly prone to Britain’s Leveson inquiry – appointing Indonesia’s ethnic and religious diversity, misuse. “In the context of the presidential a committee of experts from and related polarisation in political election in Indonesia last year, the different sectors (such as medicine affiliations, is a significant contributing friendship groups (on WhatsApp) and academia) at arm’s length from factor to the proliferation of online became a key source of political government to regulate the online space, misinformation. hoaxes,” said one researcher.71 and with the authority to order the removal of misinformation after it has been assessed.

70 Author interview with Damar Juniarto 71 Interview 7 72 Interview 9 73 Interview 3 74 Michael Barthel, Amy Mitchell and Jesse Holcomb. Many Americans believe Fake News is sowing confusion. (Pew Research Centre, 15 December 2016. https://www.journalism.org/2016/12/15/many-americans-believe-fake-news-is-sowing-confusion/ 75 Interview 3

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 57 Targets: Civil activists, SAFEnet recorded almost 40 reports Online hoaxes in Indonesia are frequently of online attacks on members of not just political, but disinformation and women and minority these groups over a recent 12-month misinformation often is targeted against groups period. Juniarto said it was critical women. According to one expert, sexual for civic groups to have direct lines of harassment of females is prevalent on Community-based activists and communication to digital platforms like WhatsApp.76 To mitigate gender-based marginalised groups – including women Twitter, Facebook and YouTube to report online harm, academic researchers have and people who identify as LGBTI – online abuses, and to find timely ways to been working with women across four have been frequent targets for online strengthen and protect targeted groups. cities to identify and try to improve their attacks in Indonesia and Singapore. media and digital literacy. The program Damar Juniarto identified a number Direct reporting helps a lot. aimed to train women to teach other of “high risk groups” in Indonesia for women about how to recognise and deal online misinformation with the potential It is also very important if you know how things work with online misinformation, as well as for harm. Most prominent among hate speech and disinformation. Due these were: inside the company. They to the pandemic and postponement y Journalists have a long, long process to of local elections in 2020, the face-to- y Human rights defenders deal with misinformation, face “train the trainer” sessions were y Anti-corruption activists but they are willing to help to conducted online: “We choose women y Ethnic group defenders who are already active, not only in the y Women escalate our report. WhatsApp groups, but also active in their y LGBTI community members. communities. So they become an engine of change,” the researcher said.

Country-specific problems

Despite their geographic proximity, Indonesia and Singapore are In Singapore, the key demographic vastly different countries in so many ways – including in how online characteristic in the context of online misinformation is not religion, but misinformation manifests online. Journalist Kirsten Han observed the nation’s relatively large migrant that while “big level disinformation campaigns” were common in population. This has manifested most Indonesia, the bigger problem in Singapore concerned freedom of recently in online campaigns that political expression. exaggerated the role of migrants in COVID-19 outbreaks.78

Democratic deficits: Geographic, cultural and tensions between local and international community standards about acceptable Restrictions on freedom religious diversity content on issues such as pornography. of speech With its sprawling archipelago spread across five time zones, and its enormous There is a tension between Identification and management of cultural, ethnic, linguistic and religious government and the undesirable content are key areas of diversity, Indonesia is an uncommonly international social media tension between government authorities and social media companies. Demands challenging environment for technology platforms because social platforms and governments seeking to by the Indonesian Government for deal with online misinformation. media has their standards or content removal have sometimes created community guidelines, but friction with social media companies, Islam, the dominant religion, also has it’s produced for international which have a range of alternative an important place in the nation’s wider options for dealing with misinformation, cultural, social and legal fabric. This needs. In some cases depending on the perceived level of extends to the Government’s upholding Indonesians have specific potential harm.79 of blasphemy laws and other Islamic needs that have to be tenets such as kafir, which effectively In some situations – often involving impose limitations on some types contextualised with the local political tension and/or threats of of online content and freedom of conditions. In Indonesia the violence – the Indonesian Government has acted to slow down or selectively expression. As one Indonesian academic definition of pornography is explained, this creates inevitable shut down the internet on the grounds very different.77 that certain content may “ruin the general

76 Interview 7 77 Ibid 78 Interview 3 79 Interview 7

58 La Trobe University purpose of the society.”80 An example of seeing the re-emergence of military Younger Singaporeans are this was when the Government blocked and police roles in civilian life.” In this much better at this because the internet in Papua in 2019 following environment, religious and political days of protests against Indonesian rule issues were very sensitive, said Septiaji they are digital natives and of the province.81 Eko Nugroho. One academic noted the they grew up with this. prevalence of hate speech before, during SAFEnet’s Damar Juniarto said They seem to ask more and after elections.82 his organisation closely watched questions about what they’re governments in Southeast Asia that reading, (which) the older sought to interfere with citizens’ internet Media and digital access, or to use misinformation as a Singaporeans don’t do. reason to shut down or slow down the literacy Han noted, however, that due to internet. However, he said SAFEnet’s Media training and teaching digital government controls over the media, vision for freedom of expression online literacy are considered important tools high levels of general literacy did not to improve democratic participation to help mitigate the harm caused by necessarily equate with media literacy. needed to be balanced against the need misinformation. However, given low “Local mainstream media is seen as for people to feel safe online. To help levels of general literacy in Indonesia, controlled by the Government,” she said. achieve this, SAFEnet had worked with it is difficult to increase media literacy This limited the media’s “fourth estate” Facebook and Google to monitor and to help citizens detect fake news, said function of holding the Government to report hate speech, with a particular Septiaji Eko Nugroho, Chair of fact- account. Indeed, Han said that since focus on gender-based harassment checking group MAFINDO. Prime Minister Lee Kuan Yew’s time and violence. Singapore, by contrast, has relatively high (1959-1990) there was an explicit Andreas Harsono was also concerned rates of both general and digital literacy, rejection of the idea of the media being about curbs on freedom of expression though less so among older citizens, said a “fourth estate”, because he said that in Indonesia. “We are seeing the rise of journalist Kirsten Han: an unelected media can’t have that sort Islamism in Indonesia, and we are also of power over the government.

80 Interview 7 81 Helen Davidson, “Indonesia arrests dozens of West Papuans over claim flag was thrown in sewer,” The Gaurdian, August 18, 2019, https://www.theguardian.com/world/2019/ aug/18/indonesia-arrests-dozens-of-west-papuans-over-claim-flag-was-thrown-in-sewer 82 Interview 7

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 59 She said it was also “hard to teach media to Andreas Harsono. Indonesian users Parliament, because even the politicians literacy when the country has only one tended to have more than one account, rely on Facebook: main newspaper. And you’re supposed to and WhatsApp groups could be as large take that as truth.”83 as 250 people,87 presenting additional It’s quite common to see a challenges for those trying to deal with Government employees are “beholden” to government agency put their online misinformation, Nugroho said. the Official Secrets Act, which precludes press statement on Facebook them from sharing information without before they even put it on authorisation, Han said. Hence, the tools The difficulty for us is needed to establish the truth were often that it is like a private their website. Because that’s held exclusively by the state, she said.84 communication. So we don’t how it gets seen. And that’s really exactly understand how it gets shared, and Internet use and the numbers. For Facebook, ministers will openly rebut preferences most of the contents are in people on Facebook. The a public space. So Facebook, local press pick it up and just In Indonesia, somewhere between quote him off his Facebook 50 and 60 per cent of citizens have Twitter and Instagram, I internet access, with about 95 per cent think are much easier for us post rather than in a press of the connected population having to get some knowledge about conference or interview. And 85 Facebook accounts. However, younger their use.88 so everyone uses Facebook. and urban Indonesians tend to prefer Instagram over Facebook. Other popular It is an important way to Similar platform usage trends were kind of reach people and see platforms include WhatsApp, Twitter, observed in Singapore. One academic Netflix, YouTube and TikTok. COVID-19 said about 40 per cent of recent survey what they are talking about. prompted an increase of about 40 respondents said that they used Han said Telegram was a favoured per cent in internet and social media WhatsApp to get news, just short of the platform for political mobilisation, use in Indonesia, according to one 50 per cent who relied on Facebook.89 86 given government restrictions on academic. And with more use comes While Chinese-speaking Singaporeans public assemblies and other offline more misinformation, said Septiaji Eko favoured WeChat and Weibo,90 younger political activities. Nugroho. “We see misinformation on Singaporeans were more drawn to Instagram and also on TikTok.” Instagram, Twitter and Telegram. Apart from issues with encryption, some According to Kirsten Han, it is said unique ways in which private messaging (in jest) that you get more politics on platforms are used in Indonesia make Facebook than in Singapore’s it more difficult to control, according

83 Author interview with Kirsten Han 84 Kirsten Han, “WTC Long Read: A mother’s concerns in a time of POFMA,” We The Citizens, 9 November 2020, https://wethecitizens.substack.com/p/wtc-long-read-a-mothers- concerns 85 Interview 1 86 Interview 1 87 Interview 1 88 Author interview with Septiaji Eko Nugroho, Chairman of MAFINDO, 26 August 2020 89 Interview 11 90 Interview 11

60 La Trobe University Conclusion

The definition of misinformation is contested. While it is often while removing fake accounts might seem desirable, it can have the understood in opposition to disinformation, this distinction is questionable effect of suppressing not always clear. Some have posited a simple distinction – dissident voices in countries under non- misinformation being a falsehood spread with innocent intent, democratic rule. Fact checking to identify and disinformation involving deliberate ill-intent. Yet social false content, while objectively desirable, is not only costly and time consuming, media platforms say it is impossible to divine the intent of the but needs to be better customised to person posting content. So, when managing and categorising deal with false content on encrypted problematic content, their focus tends to be more on its sites such as WhatsApp. factual veracity, and its relative potential to cause harm. Notwithstanding the lack of consensus on how to define, measure and mitigate misinformation online, clear unanimity Governments in Indonesia and These are but a few examples of exists about the urgent need to curb Singapore, on the other hand, while the lack of clarity and consensus it in both online and offline spheres to also strongly focused on potential harm on fundamental defining aspects of improve the quality of information vital to (particularly in relation to issues like online misinformation. The ambiguity the health and prosperity of communities COVID-19) have been widely accused matters because it signifies a lack of and democracies everywhere. of applying self-serving political criteria agreement and clarity about how best when defining and identifying alleged to address the problem. This can result online misinformation. This has put them in fractured or “siloed” approaches by in conflict with major platforms and civil governments and digital platforms to society actors over what is and is not misinformation and disinformation. misinformation, and what should and While governments have tended towards should not be allowed online, particularly overreach (sometimes criminalising in the lead-up to political elections. citizens’ inadvertent breaches of laws, and suppressing freedom of speech and Other ambiguities and complications dissent), platforms may have historically abound. For example, as the COVID-19 underestimated the harm of online epidemic has revealed, misinformation misinformation and not addressed about medical treatments can be spread falsehoods or removed harmful content without the intention to cause harm – quickly enough. meeting one definition of misinformation – but false cures may inflict serious Although Facebook and other platforms physical harm. Likewise, true information have refined their misinformation can be harmful and used for political strategies, some changes have brought purposes, such as spreading innuendo additional complications. For example, and gossip.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 61 CHAPTER 6 Tackling misinformation online and its challenges Introduction

This chapter explores recent attempts by digital platforms digital platforms face ongoing challenges in navigating this complex regulatory and lawmakers in Singapore and Indonesia to tackle the space. The chapter is divided into two problem of misinformation online. It considers the various sections. The first outlines what social issues – technological, political, demographic, geographic media companies and governments are and legal – that surround this 21st century problem, and doing to manage online misinformation, according to interviewees. The second focuses on the effects, intended or otherwise, that measures section examines ongoing challenges taken to mitigate misinformation have had on civic society, that both platforms and governments democracy and the technology platforms. face tackling this endemic problem. We specifically examine the stated problems with the laws that deal with We identified widespread fears that potential to undermine media freedom, misinformation, including how these attempts to date by lawmakers to free speech and democracy in both laws are perceived to be misused combat misinformation also hold the countries, and that governments and for purposes other than reducing online misinformation.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 63 Section 1 Measures to tackle misinformation

Digital platform Community Standards (see below), or a platform] for violating community who engage (or claim to participate) in standards, no one actually knows why,” measures to deal with behaviours involving fake accounts or she said. “So, I think the platforms should misinformation foreign or government interference.3 be more transparent and work with local civil society more about these policies.”6 Social media platforms have introduced The measures discussed in more detail a variety of measures in recent years below, in alphabetical order, fit within Artificial intelligence to try to tackle online misinformation. Facebook’s three-pronged approach Artificial intelligence (AI) is considered an In interviews for this project, digital and were the most visible or known indispensable tool in platforms’ efforts platform experts involved in addressing measures that digital platforms were to detect misleading content that has misinformation detailed various using in Singapore and Indonesia to the potential to be both harmful (such as approaches ranging from top-down to tackle misinformation as identified false information about COVID-19) and to community-led, technical and social, by the expert respondents. be disseminated rapidly and widely. AI is restrictive and corrective, and human and Community standards valued primarily for its ability to identify non-human strategies. For Facebook, misinformation quickly. According to these community standards revolve “Community standards” adopted one digital platform expert, for any around the areas of “Authenticity”, by digital platforms, and agreed tool to be able to match the virality of ensuring content on Facebook is to by users, have been a primary misinformation, it “needs to be” AI.7 While authentic; “Safety”, ensuring users can resource to help platforms combat effective AI systems are already in use8 participate in a safe online environment misinformation. By declaring types there was still a “need to be a lot faster” free of threats and intimidation; “Privacy”, of content and behaviours that are in tracking and tackling misinformation.9 the platform’s commitment to protecting admissible on platforms, community personal information and privacy; and standards help people determine what The chairman and founder of the “Dignity”.1 constitutes misinformation, and how Indonesian civil society group tackling to distinguish it from illegal information online misinformation MAFINDO, Septiaji There are at least 26 areas of concern and material that is simply contrary Eko Nugroho, said a limitation of AI was that fit beneath these broad objectives to the agenda of individuals or groups local context and culture: of which false news is one and the issue that oppose it.4 Facebook works with most closely aligned with misinformation community service organisations (CSO) Sometimes what we see is – the subject of this study. Under this to convey its community standards: section heading, Facebook outlines its the machine doesn’t really “We continue to train CSOs on how three-pronged strategy to: “Remove” know the local context, we look at our community standards accounts that violate its policies; and how our community standards especially for Indonesia “Reduce” inauthentic content and the look at misinformation,” but conceded where we have more spread of false news, and to “Inform” that more “outreach” was needed, or help online users make informed than 700 languages, local “especially for Indonesia,” to ensure decisions about “what to read, trust and languages, and some word the public and government understand share.”2 To paraphrase, Facebook defines how misinformation is regarded by meanings could be very “inauthentic” as the use of its platforms the organisation.5 different. A word could be to mislead others about the identity, purpose, popularity or origin of their However, Kirsten Han, a journalist based very polite in one area, but online presence. It includes users who in Singapore, said community standards very insulting in another.10 may engage in coordinated activities and on platforms sometimes lacked clarity. who do not comply with the platforms’ “When you get taken down [from using

1 Facebook, “Community Standards,” Facebook, 2020, https://www.facebook.com/communitystandards/ 2 Facebook, “Hard Questions: What’s Facebook’s Strategy for Stopping False News,” Facebook, 23 May 2020, https://about.fb.com/news/2018/05/hard-questions-false-news/ 3 Facebook, “Inauthentic Behaviour,” Facebook, 2020, https://m.facebook.com/communitystandards/inauthentic_behavior/ 4 Interview 9, interview with Andrea Carson, 8 September 2020; Interview 4, interviewed with Andrea Carson, 7 August 2020 5 Interview 9 6 Interview with Kirsten Han, interview with Andrea Carson, 25 August 2020 7 Interview 1, interview with Andrea Carson, 7 August 2020 8 Interview 9 9 Interview 1 10 Author interview with Septiaji Eko Nugroho, Chairman of MAFINDO, 26 August 2020

64 La Trobe University For this reason, human analysis It’s a difficult conversation to prevent people abusing these features.24 of misinformation and inauthentic have with the Government… But such measures can have unintended content remains essential, argued consequences for other users and uses, an Indonesia academic.11 they get a lot of pressure such as small business (see [about the platform] not Chapter 5 for more details regarding Content removal, platform adhering to the local norms WhatsApp forwarding features).25 bans and feature limits 19 and traditions. Digital literacy Academics and the technology platforms cited various tools that can be used to Editor-in-Chief of Tempo magazine, Several interviewees identified tackle online misinformation. These Wahyu Dhyatmika, said it was not teaching digital literacy to the general include the removal of content,12 fact uncommon for the Indonesian community as essential to combating checks (along with statements about Government to demand hoaxes be taken misinformation in Singapore and false claims) and, in the most extreme down from the platform or otherwise Indonesia. One platform expert said: cases, banning users who have threaten measures such as revoking repeatedly spread misinformation from a their licence to operate in Indonesia For a country to be platform.13 However, only misinformation or legal action. This made the issue of successful in clamping that risked causing “real world harm” “take downs” and user bans a complex down on misinformation, tended to be removed quickly.14 For one. He explained: example, in reference to recent riots in they need to put digital Indonesia, one platform expert said: The tech platforms also get literacy and ways to pushback from civil society recognise misinformation The basis of what happened if Government asked them to in the curriculum. I think with those riots was actually take down content that’s not the answer is in people’s misinformation… that led to necessarily misinformation awareness… it’s like wearing real world harm. That’s why but criticism towards the a seatbelt so that everyone we could actually instantly government. So they are in the country will grow remove the content… We really working on a thin line up knowing that… there’s don’t (usually) remove on if they want to keep being an a risk that you will find the go… we would normally open platform for all voices. misinformation and be consult with our trusted misled… if you use the 15 But, I suppose, they need to partners on the ground. also obey the law and follow internet.26 Misinformation relating to serious the government’s line. So it’s Recent campaigns aimed at educating public health issues such as COVID-19 20 a very complex issue. Indonesian citizens on how to detect and vaccinations tends to be deemed online misinformation were “very harmful, as does racial and religious Added to this complexity is the mixed successful,” said another platform hate speech.16 One academic said that quality of reports of misinformation expert.27 Other measures included while platforms preferred less severe sent to platforms by the Government. campaigns to “teach people to measures than content removal, the According to one digital platform expert think critically when they use the internet Government sometimes saw removal it was hard to act on misinformation before they share anything,” and “train as “the best solution.”17 A digital platform when the reporting is poor or the the trainer” programs, where the platform expert said platforms reserved user bans misinformation not well defined.21 provides training to organisations, for “the worst actors.”18 Another issue is disagreements between which in turn train other people in their platforms and governments over what Tension between local norms and communities.28 Another platform policy constitutes misinformation.22 A possible customs and broader considerations expert said the key was to teach people solution to this was more independent of human rights can also cause conflict that not everything they see online is fact checking.23 between platforms and governments authentic. One way to do this was a about what is misinformation and Platforms can also restrict features that general public education campaign in how to respond to it. A digital platform have been used to spread information, or Indonesia: expert said: just “turn off” certain functions to

11 Interview 7, interview with Andrea Carson, 19 August 2020 12 Interview 1; Interview 7; Interview 9 13 Interview 4, interview with Andrea Carson, 7 August 2020 14 Interview 9 15 Interview 9 16 Interview 1 17 Interview 7 18 Interview 4 19 Interview 9 20 Author interview with Wahyu Dhyatmika, Editor-in-Chief of Tempo magazine, 24 August 2020 21 Interview 9 22 Interview 9 23 Interview 9 24 Interview 4 25 Interview 9 26 Interview 1 27 Interview 9 28 Interview 1

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 65 We tried to educate the from technology platforms (seven of False content warning labels which are signatories to the International public on how they need Platforms have also taken to using Fact-Checking Network (IFCN) – a warning labels to alert users to content to look at misinformation. unit of the dedicated deemed to be false or unreliable.38 One We built partnerships with to bringing together fact checkers social media expert said labelling of worldwide), and government agencies.32 Siberkreasi, a consortium misinformation and the “promotion” of Platforms have also performed fact of CSOs, academics and correct information had been particularly checking in Singapore, with a particular private sector that care 33 focused on serious public health issues focus on election campaigns. 39 about digital literacy, and such as COVID-19 and vaccination. One advantage of fact checking With COVID-19, algorithms are used delivered programs such compared to other anti-misinformation to highlight and prioritise credible as “Asah Digital”, aimed to measures is that it allows platforms to information from bodies like the World educate students, parents correct false content while respecting Health Organisation and the Government. users’ rights to publication. As one and teachers to better platform expert said, a correction allows Investing in quality journalism identify misinformation the continued right to share content while Platform initiatives to support journalism using critical thinking “everyone has the right to know also that communities such as Facebook’s 34 and empathy. In the lead it’s been fact checked as false.” Journalism Project and Google News up to election we also ran However, fact checking is resource- Initiative and Twitter grants to media campaigns on our platforms intensive and expensive. Wahyu outlets were welcomed by respondents Dhyatmika, editor of Indonesian who thought the platforms had a role to to tackle misinformation media company Tempo, described play in supporting quality news online. and had stickers around the the challenges: One academic said: “These projects city buses, all over Jakarta. where companies invest in helping Reporters can do five to And I hope [we] can continue journalism to survive are the kinds of seven news items per day, these public education models that should be replicated in and our editors can get this misinformation arena.”40 campaigns.29 through 15 articles per day. Others valued the contributions and Anti-misinformation activist Septiaji Eko But fact checkers sometimes commitment platform companies had Nugroho endorsed a “multi-pronged only produce one fact made to third-party fact checking. These approach” to digital literacy involving not outsourced roles also provided another just platforms, but governmental and checked article per day… it’s stream of revenue for media outlets in community organisations involved in very resource intensive.35 difficult times for the business model training the public.30 He said MAFINDO of legacy media. collaborated with more than 100 Another digital platform expert, noting organisations including civil society the large volumes of misinformation Public reporting measures and groups and government agencies to online, said of third-party fact checkers: proof of accounts “They need to be very strategic in the improve the public’s digital literacy Asking social media users to provide type of misinformation that they want through various targeted campaigns. proof of identification aims to ensure to debunk, because it does take a lot transparency and accountability, said of work.”36 Fact checking one digital platform expert.41 Others said fact checking, though Technology platforms have initiated Providing features and avenues for valuable, needed to be used in extensive third-party fact checking users to report misinformation to conjunction with other measures. operations, which in Indonesia include platform regulators can also help with Singapore-based journalist Kirsten the Indonesian civil society organisation the detection and management of Han said fact checking alone would (CSO) and fact-checking outlet, online misinformation.42 This can be not stamp out misinformation. But she MAFINDO, and media organisations particularly useful when dealing with 31 believed more fact checking would such as Tempo. In some cases, private messaging applications such help to shift accepted norms and the platforms use other external as WhatsApp, over which content remind people that they need to look experts, such as the World Health regulation is made difficult by privacy out for misinformation.37 In this way, Organisation and other health experts and encryption controls. for consultations regarding harmful fact checking may also serve as an COVID-19 misinformation. educational tool for social media users and should be used in conjunction with Fact checking in Indonesia is dynamic, digital literacy education. with independent media and CSOs collaborating with and receiving support

29 Interview 9 30 Author interview with Septiaji Eko Nugroho, Author interview with Andrea Carson, 26 August 2020 31 Interview 1 32 Author interview with Septiaji Eko Nugroho 33 Interview 11, interview with Andrea Carson, 20 August 2020 34 Interview 4 35 Author interview with Wahyu Dhyatmika 36 Interview 9 37 Author interview with Kirsten Han 38 Interview 1 39 Interview 1 40 Interview 11 41 Interview 4 42 Interview 1

66 La Trobe University Relationships between Another digital platform expert stressed Despite these positives, one digital technology platforms the importance of collaboration platform expert lamented the lack of between platforms: consultation between the Indonesian Technology platforms acknowledge and Government and CSOs, compared embrace their central role in combatting I think collaboration and to the “quite regular” consultation by online misinformation. However, one trust needs to be developed, the Ministry of Communications and social media expert expressed a desire Informatics with platforms. for more transparency by platforms and there needs to be proper about their interventions and policies.43 information sharing across To ensure that the And while there was general agreement platforms because this on the need for co-operation between conversation continues to platforms, there were mixed views on information is not just on develop in the right direction, the extent to which it occurs. any (one) network.49 maybe we should start One platform expert said: Partner organisations also advocated holding public consultation… a proactive role for platforms. The with the CSO communities If we investigated and found Indonesia-based head of the Southeast as well.52 certain bad actors that are Asia Freedom of Expression Network pushing propaganda across (SAFEnet) Damar Juniarto said civil The Indonesian Government our platform, we would share society groups could not tackle the problem of misinformation alone and has also been accused of some of what we found with that they had “asked the social media asking platforms to remove our partners, so that they platforms to take responsibility as well.”50 content simply because they can then further their own Juniarto highlighted the potential for civil society organisations (CSOs) and “don’t like” it, said Damar investigation and try to take 53 platforms to work together to hold the Juniarto. down the network on their government accountable for perceived site as well.44 regulatory overreach that could threaten The major platform companies also fund freedom of expression and political academic research to further knowledge Community organisations strongly discourse. This highlights how the about the nature of the problem and support co-operation and information agendas of platforms and CSOs (which ways to mitigate it. Global funding sharing between platforms. Damar are otherwise called non-government rounds call for academics to submit Juniarto of SAFEnet recalled how organisations, or NGOs, in Australia and proposals on particular themes. The he had shared a model of reporting some other countries) may intersect. companies also commission academics misinformation introduced by one and research units for specific projects, In Indonesia, social media platforms platform with other platforms. “It including this report. provide resources to media helps a lot,” he said.45 organisations for measures such as fact Trusted partners and checking operations.51An example of this collaboration is the fact checking coalition CekFakta, comprising 22 media companies, Digital platform experts stressed numerous CSOs such as MAFINDO, and the importance of collaboration with is financially supported by the Google community organisations, local media, News Initiative, anti-misinformation academics and governments to tackle NGOs Internews and First Draft. The misinformation. One said they would coalition aims to create almost 100 normally consult with “trusted partners” “master trainers” among journalists, and to validate what type of information is a curriculum to train another 4000 fact worth removing immediately: “We don’t checkers, said MAFINDO head Septiaji 46 really want to be the arbiter of truth.” Eko Nugroho. Trusted partners could reduce platforms’ Tempo, one of the media companies burden of responsibility in the task of involved in CekFakta, through a separate deciding which content to manage partnership, had also established a 47 through fact checking activities. team of health experts to help tackle One platform interviewee said it was misinformation related to COVID-19, important that fact checking partners according to editor-in-chief Wahyu operated independently: “Ultimately it’s Dhyatmika. up to them… to validate whether that specific information is true or not.”48

43 Interview 1 44 Interview 4 45 Author interview with Damar Juniarto, SAFEnet, Southeast Asia Freedom of Expression, Executive Director, 8 September 2020 46 Interview 9 47 Interview 9 48 Interview 9 49 Interview 4 50 Author interview with Damar Juniarto 51 Author interview with Wahyu Dhyatmika, Author interview with Andrea Carson, 24 August 2020. 52 Interview 9 53 Author interview with Damar Juniarto, Author interview with Andrea Carson, 31 August 2020.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 67 Government measures are worried in the current SAFEnet’s Damar Juniarto said the climate. If Indonesia is ITE law is “misused” during election to deal with campaigns, especially to undermine rival misinformation to revise the Criminal political candidates by reporting them Code, what is the current to police for allegedly breaching the law. The governments of Indonesia and situation? We are already This was hurting freedom of expression Singapore have both enacted laws seeing the rise of Islamism and democratic participation, he argued: specifically to combat misinformation “The misuse of internet laws is always online (see chapters 1 and 2 for more in Indonesia. We are also increasing during the election time… details). Pre-existing laws are also used seeing the re-emergence of freedom of expression in Indonesia to tackle the problem. While there is is decreasing.” broad agreement about the need for military and police role in 55 government regulation of harmful online civilian life. A digital platform policy expert said the content, critics in both countries question ITE law was problematic because: the effectiveness of the existing laws Within the existing Criminal Code is an and the ways in which governments article pertaining to treason with the It contains a vague article have applied them. provision to jail alleged offenders for up that says something that to 20 years.56 Harsono said the article is used unjustly as a tool of the state could disrupt the harmony or Indonesia to arrest and jail Papuan separatists. stability in the community is Amnesty International has recorded illegal essentially. And so the There are several laws and proposed at least 77 prisoners of conscience laws that attempt to deal with Government has used that detained on treason charges in Papua misinformation and disinformation and Indonesia, including at least 20 law to prosecute people for in Indonesia. Below are the laws people under city arrest.57 you know, all kinds of things that interviewees referenced in their responses. that they post on social Papuans want to be media.60 Revised Criminal Code independent of Indonesia Editor-in-Chief of Tempo magazine, Proposed revisions to Indonesia’s or ethnic Moluccan who Wahyu Dhyatmika, said the main near-century old Criminal Code – aimed also want to be independent problem with the ITE law was a lack of in part at curbing the spread of online from Indonesia. They were transparency about how it is used and misinformation – have prompted sentenced between three that it tended to be used in favour of widespread public outcry, including government officials, not civil society. street protests in September 2019 (see and 20 years in prison for He argued: chapter 1). Human rights activist Andreas raising their flags. Now this Harsono said the revisions would lead draconian Criminal Code There’s problem with to an expansion of government powers, article is extended into the and a “much more draconian” and transparency [of ITE] and discriminatory law.54 Plans for a vote in internet law (ITE), because, also about the effectiveness Parliament on the draft Bill have been of course, in 1918 the internet of its legal methods to stalled at the time of writing following did not exist. In 1918 it was deal with misinformation. the protests amid pressure on the Government to resubmit a revised plan. mostly print and radio If it’s misinformation – Harsono was particularly concerned broadcasts, with broadcast the distributors may not about what a revision would mean for TV also covered under knowingly distribute religiosity and politics in Indonesia: broadcast.58 contents with regards to the The blasphemy law in the impact or with its accuracy – Information and Electronic then using laws just creates existing criminal code Transactions Law has only one article. In the polarisation, it creates other Indonesia’s Information and Electronic problems in the society. draft law, the draft penal Transactions Law (ITE) was introduced code, it will be expanded in 2008 and revised in 2016. Harsono It doesn’t really deal with into six articles. You know, said the law is “basically an extension” the core issue that we are blasphemy is nonsense, that of the Criminal Code as it applies to dealing with, which is the hate speech, slander, defamation and low rate of media literacy. it is going to be expanded, blasphemy to include online material. slander and libel will also “Both of them are pretty ugly laws,” And, you know, low trust in be expanded with longer jail he said.59 the government.61 terms. That’s why we

54 Author interview with Andreas Harsono, Human Rights activist and Researcher at Amnesty International Indonesia, 20 August 2020 55 Ibid 56 Ibid 57 Epa, K. “Rights groups demand end to Indonesian treason laws,” UCAnews. 2020, 13 March, https://www.ucanews.com/news/rights-groups-demand-end-to-indonesian- treason-laws/87448# 58 Author interview with Andreas Harsono 59 Ibid 60 Interview 1 61 Author interview with Wahyu Dhyatmika

68 La Trobe University Further, defamation – alongside libel, A digital platform expert said this Observers say POFMA was enacted in slander and blasphemy – is also covered proposed regulation on electronic system response to platforms’ unwillingness by the Criminal Code and ITE law.62 As operators (ESO)65 if passed will force to remove some types of alleged part of the general attack against online platforms to meet short turnarounds to misinformation at the request of the content deemed harmful, lawmakers and remove content if deemed inappropriate Government.67 The Government had enforcers in Indonesia have also sought by the government. previously been unable to convince to limit and criminalise pornography. social media platforms to remove The Government has taken content they considered unharmful, such Personal Data Protection (PDP) the most conservative as opposition criticism.68 Draft Bill approach which is that if it’s One academic criticised the Singapore SAFEnet is also concerned about a misinformation that could Government’s treatment of POFMA as Cybersecurity Draft Bill before the “a panacea” to tackle misinformation Indonesian House of Representatives potentially be a threat to online.69 While there was “a place for at the time of writing that ostensibly the country, then it should legislation and regulation,” he argued “it aims to enforce data protection and be removed. I think what should be a multi-pronged approach.”70 to punish data breaches with fines up has been missing from this However, he conceded that POFMA has to $A20million for corporations and been useful in mitigating misinformation seven years jail for individuals. It is discussion, with regards relating to COVID-19 and hate speech. said to be modelled on aspects of the to content moderation Freelance journalist Kirsten Han said General Data Protection Regulation regulation, has been our POFMA had increased the vulnerability 2016/679 (GDPR) issued by the European friends from the CSO of freelance journalists. She said that Parliament and Council of the European getting “POFMA-ed” or getting sued Union. However, Juniarto said the community. Not only are for defamation could disrupt the work proposed laws (known as RUU KKS) they the ones who will be of freelancers like her who lacked the contain provisions that would enable the impacted, CSOs will be able resources to fight charges. “I wouldn’t government to suppress the spread of have the benefit of legal counsel that information by cutting data connections. to provide a substantial input a company will pay for,” she said.71 Her It also threatened to sanction public to the draft on trends and situation contrasted with journalists in opinion if deemed a cyber threat, without 66 common issues. mainstream Singapore media outlets, any judicial oversight. “I’m a little bit which were mostly government-aligned worried because they [the Indonesian and rarely sued for defamation. Government] are not involving enough Singapore stakeholders to think together about The law that respondents found most Other Singaporean laws used this law. It is one-sided for the sake of against misinformation the Government… not for the people,” problematic in Singapore was the new Juniarto said. 2019 law relating to misinformation Other laws used in Singapore in the and online falsehoods, best known name of fighting misinformation include The Indonesian Government’s as POFMA. sedition laws designed to prevent Communications and Informatics citizens or others displaying or enacting (KOMINFO) Ministry also counters Protection from Online hatred, contempt or disaffection against alleged hoaxes and misinformation on its Falsehoods and Manipulation Act the Government. Under this old 1938 law website. KOMINFO issued a statement (updated in 1985), the intention of the on 4 November 2019 that a new The Protection from Online Falsehoods offender is irrelevant. It has been used regulation GR71 would fill a regulatory and Manipulation Act (POFMA) has been in the past to criminalise actions that vacuum until the PDP was passed.63 This the subject of vocal opposition from were said to have a “seditious tendency” regulation provides measures to prevent many Singaporeans since it took effect and may attract fines or a jail term the circulation and use of prohibited in September 2019. Critics claim the law of up to three years. One critic said it content including what the government is ambiguous, has inadequate appeal added “another layer of problems in the deems as “negative content”, among mechanisms and can be used arbitrarily implementation of regulations” to deal other uses. This includes online by the government to censor or silence with misinformation.72 information that may “cause public its opponents. disquiet and disturb public order.”64

62 Author interview with Andreas Harsono 63 DataGuidance. “Indonesia: New GR 71 enters into force and ‘proves more legal certainty’. OneTrust DataGuidance, 2019, 7 November, https://corporate.dataguidance.com/ indonesia-new-gr-71-enters-into-force-and-provides-more-legal-certainty/ 64 ABNR, “Indonesia Issues Important New Regulation on Electronic (Network and Information Systems” ABRN. 2019, 30 October, https://www.lexology.com/library/detail. aspx?g=cd6e5251-6dd7-4b46-b6be-759c78c9bf7b 65 Draft of Ministerial Regulation for Private ESO (Registration and Content Moderation) 66 Interview 9 67 Interview 1 68 Interview 1 69 Author interview with Kirsten Han 70 Interview 11 71 Author interview with Kirsten Han 72 Author interview with Damar Juniarto

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 69 An academic said Singaporean problematic because election “It pushes you to find more evidence to politicians also still “resort to civil and campaigns in Singapore run for ten support the story,” she said. An academic criminal defamation” against certain days.76 So, a successful appeal against also said POFMA had been useful in commentary.73 Singapore’s Broadcasting a POFMA order about election-related managing COVID-19 misinformation Act has also been used against material usually took longer than that and hate speech.78 perceived misinformation. According and could not be resolved during an However, such positive angles about the to one academic, the Act is among the election campaign. various misinformation laws in Singapore laws used by the Government to censor Kirsten Han took a similar view on POFMA: and Indonesia were rare in interviews political dissent.74 conducted for this project. Penalties and legal appeals There’s value in appealing Observers in Indonesia questioned just for the sake of showing Critics of POFMA have raised alarm the usefulness of strong fines and about the large costs faced by that you don’t agree. But… other penalties as a weapon against people who defy orders to remove they use POFMA during the misinformation. One academic said the penalties had not had a significant misinformation, or who want to challenge election campaigns. None such orders. deterrent effect. Despite a high number of those appeals would have of cases in courts, there was still a Historian and activist PJ Thum said gone through fast enough to lot of misinformation in social media, mounting an appeal against a POFMA be in time for the election.77 suggesting law enforcement is “not charge of spreading misinformation easy to bring.”79 would cost around $1000 for a company. However, Han also acknowledged “The Government says it would be fast POFMA has had some unintended and cheap to appeal. A thousand dollars benefits. The ever-present threat of her 75 is not cheap to us.” reporting being challenged under POFMA One platform expert said the appeals (or “POFMA-ed,” as critics call it) has process has proven to take at least forced journalists to always keep in mind two weeks, which was particularly the need for evidence to support a story.

73 Interview 11 74 Interview 11 75 Author interview with PJ Thum 76 Interview 1 77 Author interview with Kirsten Han 78 Interview 11 79 Interview 7

70 La Trobe University Section 2 Ongoing challenges

Challenges for platforms Geography, culture and language Indonesia, where hundreds of indigenous languages are spoken across the The location of most of Indonesia’s Digital platforms face various technical, country. One academic said the use social media companies’ headquarters cultural, political and legal challenges in of local languages on social media in Jakarta presents problems when their efforts to mitigate misinformation. made it difficult for platforms to quickly misinformation comes from other Observers from within and outside the detect problems such as hate speech, regions across the vast Indonesian platforms provide insights into these misinformation, radicalism, pornography archipelago, according to Andreas challenges in the following section. and even indications of intent to Harsono. The “Java-centrism,” due in suicide.85 Another observer lamented the part to the Indonesian Government Agnostic and cross platform lack of fact checking of regional media.86 misinformation spread barring platforms from opening offices in places such as Papua, meant platforms Damar Juniarto commended the The relatively free movement of content were not well placed to deal with platforms for supporting translation of between platforms adds to the task of misinformation issues in other regions. SAFEnet’s fact checking work. He also trying to mitigate misinformation and noted that SAFEnet’s content was able One digital platform expert said that disinformation. One observer said the to be widely translated and read using because misinformation can be “very movement of content between platforms ClaimReview, the global tagging system. much localised,” it might not be readily meant legislation or voluntary regulation Fact checkers use it to identify their recognisable as misinformation to needed to be “platform agnostic,” rather articles for search engines and social “someone that sits in Jakarta.”82 He said than targeted at certain platforms or media platforms to use to promote and platforms needed to make greater efforts platform functions, and focused “more highlight fact checked articles. However, 80 to bridge the geography gap: on behaviours as opposed to content.” the continuing challenges presented by When it comes to language diversity highlighted the value When Governments talk of platforms having partnerships with about disinformation and sub‑regional elections… We local actors. don’t have CSO partners influence operation, they Legal repercussions will try to read into the there on the ground to flag misinformation that Platforms can face potential legal motive of the people doing repercussions resulting from their it. And we think that’s could potentially turn into measures against misinformation and really hard, right? Like as real world harm… So these disinformation. One digital platform a platform, users are users, conversations need to be expert said the company had been initiated.83 threatened with lawsuits and, in one we don’t know which case, actively pursued over alleged ones are out to get people and Differences between Indonesian cultural reputational harm after labelling a if governments just simply values and internationally-influenced group as a disinformation network.87 tell us, “we believe these “community standards” of platforms But such lawsuits were rare “because also create ongoing problems, according our process is very rigorous.” people are trying to interfere to one academic.84 For example, in our local discussion and pornographic images considered Social media influencers civic discussion” and tell us acceptable in Western cultures were Platforms face unique challenges to block these people. Then, regarded as contraband in Indonesia when high-profile users promote 81 due to its socially conservative values misinformation to their thousands or that’s not straightforward. – highlighting the potential tensions millions of followers. One digital platform between the values of digital platforms Collaboration, trust and information expert said attempts to crack down on and the local populations they serve. sharing between platforms were social media influencers can backfire also needed to deal with this cross- Language diversity – both between when the influencer “cries foul” and 88 platform problem. and within countries – is another major attracts more sympathy from followers. challenge for platforms – particularly in

80 Interview 4 81 Interview 4 82 Interview 9 83 Interview 9 84 Interview 7 85 Interview 7 86 Author interview with Wahyu Dhyatmika 87 Interview 4 88 Interview 1

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 71 Timeliness Challenges for Law enforcement Platforms say they face sometimes governments Interviewees criticised the tendency unrealistic demands from governments of governments to prioritise law seeking quick responses to cases of Laws being relied upon by the Indonesian enforcement ahead of alternative, less alleged online misinformation. One and Singaporean governments to tackle drastic mechanisms to deal with online platform expert says demands for quick online misinformation are beset with misinformation. Septiaji Eko Nugroho action conflict with the platform’s need practical issues and obstacles to their of MAFINDO suggested there were to ensure thorough and fair processes.89 enforcement – not least globalisation more effective measures that could versus national sovereignty and be tried first.95 He supported the use We have a very rigorous unresolved debates over what defines of “restorative justice,” including local process that usually takes misinformation. “social sanctions” and mediation as several days, if not weeks, a “more persuasive and educational” Globalisation approach than law enforcement, which and in some cases, months, Globalisation presents a major challenge he said should be a last resort. to the effective implementation of depending on the size, and A notable feature of the POFMA law national laws against misinformation – the nefariousness of these in Singapore is its preference for particularly when there is disagreement 90 corrections over content removals. networks. between countries on what constitutes This attracts mixed views. One digital legitimate content. The international The digital platform expert expressed platform expert, while generally reach of online misinformation limits frustration at the eagerness of concerned about POFMA, supported the the effectiveness of one country’s legal governments to “shut down those Government’s approach of informing jurisdiction acting alone. And when there accounts in 24 hours” without trying platform users that a post contains is disagreement across national borders alternative mitigation mechanisms first. misinformation, rather than removing it.96 about the nature of the content, this can Another platform expert agreed: raise potential diplomatic issues not only But an academic argued that the harmful for governments, but for the platforms potential of misinformation, particularly There’s still a high given the task of resolving them. on an issue such as COVID-19, requires more stringent measures.97 He said: expectation for us to Journalist Kirsten Han questioned remove (misinformation) whether Singapore’s POFMA law could It is not censorship if you very instantly. And that’s be used to prosecute foreign-sourced remove or delete false something that I feel like the misinformation and disinformation: information on health… regulator doesn’t understand. POFMA is very effective if The impact can be It can’t be as straightforward you’re using it against a local exponential… it should not as that.91 activist who has limited be left on. And that is why Media and community group resources, and who is in I’m against the use of the representatives sympathised with the Singapore and has to comply. POFMA.98 platforms on the issue of time pressure. But if you are actually While wishing for a “fast response” to talking about a foreign state Critics also said the claimed intent of his requests for action, Damar Juniarto POFMA was at odds with the way it had of SAFEnet acknowledged that “it launching a disinformation been used by the Government to date. PJ takes time.”92 MAFINDO’s Septiaji Eko campaign… like what they Thum said definition of what constitutes Nugroho said platforms had been “quick say Russia is doing in the US, a falsehood had been blurred: enough” in their responses to potentially how would you use POFMA harmful content, particularly during It became very clear from 94 the pandemic.93 However, he said this against that? almost the very first use of had been made possible in part due to the active role of other groups such as Han cites the case of political dissident POFMA that it was about MAFINDO in detecting and reporting Alex Tan, the subject of several orders interpretations of statements, online misinformation. under POFMA, to illustrate the limitations and that the Attorney- of laws that seek to prosecute alleged misinformation coming from outside the General’s chambers held the country (see chapter 2 for details). position that as long as any interpretation of a statement could be construed as false, the Government was entitled to use POFMA.99

89 Interview 4 90 Interview 4 91 Interview 9 92 Author interview with Damar Juniarto 93 Author interview with Septiaji Eko Nugroho 94 Author interview with Kirsten Han 95 Author interview with Septiaji Eko Nugroho 96 Interview 1 97 Interview 3 98 Interview 3 99 Author interview with PJ Thum

72 La Trobe University Political use of Targeting the media The ruling party has used a number of regulatory measures to rein in Singapore misinformation laws Journalists have been key targets media companies and journalists. of misinformation laws in Indonesia Online industry insiders and other Kirsten Han said the party’s presumption and Singapore. One interviewee said observers say the governments of both of control over what constitutes a journalists faced the risk of significant jail Indonesia and Singapore have used falsehood, combined with the lack of terms for their reporting in Indonesia.102 misinformation laws to censor or silence freedom of information, made it hard for While noting that the Singaporean a wide spectrum of critics, with major journalists to circumvent POFMA orders. Government had not jailed anyone under implications for freedom of speech, And the costs of appealing against misinformation laws, PJ Thum, who runs media freedom and political democracy. such orders could be prohibitive for the independent media company New In Singapore, interviewees for this project freelance journalists and independent Naratif, said: raised concerns about the targeting media companies. of opposition politicians, activists, The Government can Critics say the ruling party has also journalists and bloggers under the tried to censor media organisations 100 simply ban us. They can POFMA law. Activist PJ Thum said by “gazetting” them as political POFMA had enabled the arbitrary use of block access to our site. And organisations, which means they are government power against many groups because I’m a Singaporean, banned from accepting money from and individuals: “Because the definitions they can come after me foreigners and forced to declare all and the discretion is entirely controlled and haul me into court and donations. Thum said New Naratif had by ministers, it’s impossible to appeal sought to circumvent these issues by against this or push back against this,” charge me with whatever registering offshore. he said. they want under the law… Han has worked around the “gazetting” In Indonesia, the ITE law was of And then they can come problem because she works particular concern. Damar Juniarto after New Naratif for up independently and writes for international of SAFEnet said “misuse” of internet publications. This allows her “to write laws was particularly prevalent around to half a million dollars. So these are the things hanging a report about Singapore outside of election time. He also lamented a government control.”105 general rise in the influence of online over our heads.103 misinformation on Indonesia’s political In Indonesia, Wahyu Dhyatmika said process dating back to 2012. Suggesting independent media Tempo had been on the receiving end organisations had received particular of multiple complaints to the Press Septiaji Eko Nugroho of MAFINDO said scrutiny, PJ Thum cited an example from Councils and sometime lawsuits from that before the coronavirus pandemic, legal the recent Singapore election campaign government ministers.106 In another cases against misinformation were mainly involving a statement by Chairman of the instance, after publication of a critical about political issues, particularly around Singapore Democratic Party, Professor investigation into the Indonesian elections, and hate speech. However, he Paul Tambyah to which the ruling party Government, Wahyu Dhyatmika said his added that the laws were not just used (PAP) objected. site was attacked and “bullied by bots” against government critics: “Sometimes in an orchestrated cyber-attack. While even big supporters of the President go he couldn’t prove the Government was to jail because of… misbehaviour issues in (The PAP) were very careful about it… they did not hit involved, he said the timing of the attack social media,” he said. soon after publication was suspicious. Others noted recent examples to the Singapore Democratic illustrate the Indonesian Government’s Party or Paul Tambyah Censorship, self-censorship and use of the ITE law and Criminal Code himself with POFMA, even freedom of expression to limit political expression. Andreas though he was the source of Critics say the introduction of laws such Harsono said Papuans and ethnic the statement. They hit the as POFMA has had what one called a Moluccans were threatened with the “chilling effect” on public discourse in law over their struggle for sovereignty, outlets which reported his Singapore. One academic feared self- while an academic cited two instances statement, which means he censorship will increase as POFMA is in 2019 – the post-election riots and had no standing to appeal. deployed to censor political dissent and unrest in Papua – when the Government alternative views.107 either slowed down or blanketed internet Only the media outlets access for citizens as a way to mitigate could challenge it. And PJ Thum said the fear of POFMA orders the unrest.101 given the very short election was already affecting media practice. He cited the typical example of an editor period… by the time that wanting to print something in the public (media outlets) appealed, the interest, but the author resisting because election was pretty much they don’t want to get “POFMA’ed.”108 over. So I do feel that this Thum said regulations that pre-date did sway the results against POFMA in Singapore had already acted as a deterrent, particularly for Tambyah. 104

100 Interview 1 101 Interview 1 102 Author interview with Andreas Harsono 103 Author interview with PJ Thum 104 Author interview with PJ Thum 105 Author interview with Kirsten Han 106 Author interview with Wahyu Dhyatmika 107 Interview 11 108 Author interview with PJ Thum

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 73 independent media. “Even if POFMA had said about the government laws. Central to this issue is unequal disappeared, you’d see no new planning policies.113 access to information, favouring media start-ups for social-political governments over their populations. In Indonesia, activists fear proposed issues because of the fear (of) the Kirsten Han said: changes to the Criminal Code will regulations.”109 The main problem with increase the Government’s powers POFMA, he said, was its contribution to to suppress free speech. Another In Singapore… there’s no a culture of self-censorship: civic rights activist, Damar Juniarto freedom of information A lot of these laws create of SAFEnet, cited proposed limits on laws. So the Government criticising the Government, saying the holds pretty much all the fear so that people end Bill was designed to benefit the powerful 114 cards in terms of data. So up self-censoring. So the over the people. I can say something, and Government can say that But one academic believed the public also they didn’t do anything. need to appreciate their responsibilities they go, “that’s not actually They didn’t actually apply when it came to free speech: true, because our data the law.110 shows this,” and then I can’t There is a grey area between do anything.119 Kirsten Han said of her own practice: freedom of expression… (and) “There is definitely kind of more how to be an active and also One academic also questioned the apprehension and hesitation… I definitely ethical digital media user or transparency of the appeals process under POFMA, despite it being portrayed have found myself hesitating on digital citizen. Sometimes some stories.” by the Government as inexpensive people interpret freedom of and accessible.120 In Indonesia, Wahyu Dhyatmika observed more caution among people engaging in expression to mean freedom Similar concerns have been raised about online discussions: to say anything, but they Indonesia’s ITE law. Wahyu Dhyatmika of forget that it should be in an Tempo said, “the problem is sometimes… You can see it from chats, ethical manner.115 there’s no transparency about how response or comments that it’s being used.” While action is often taken when government officials allege we have for our content Declining quality of debate misinformation and disinformation, on social media. The Interviewees blamed the use of members of the public lack the clout to commenters, the users misinformation laws, and the resulting get such action taken on their behalf. restrictions on free speech, for a general will remind each other, The backfire effect “careful with your comment. deterioration in the quality of public and political debate in both countries. Some observers believe Singapore’s Don’t post anything that’s One Singapore academic sensed the ruling party has damaged its standing sensitive, because who Government wanted to use POFMA to with the public by appointing itself the knows what will happen.”111 “stop debate,” leading to the loss of a arbiter of truth under POFMA. PJ Thum democratic contest of ideas.116 points to the PAP’s poor election result in 2020 as evidence that the Singapore Interviewees in both countries cited anti- PJ Thum said the Singapore ruling public “are not going to put up with” the misinformation laws as a restraint on party “hates” foundational elements use of laws such as POFMA to suppress freedom of expression and democratic of a functioning democracy such as “a democratic rights. He hoped its relatively discourse more broadly. The use of critical citizenry (that) asks questions poor poll result for a party in power since POFMA during the 2020 Singapore of the Government…they prefer to write 1959 would encourage the Government elections was a case in point. One laws which just give them vast sweeping to restrain its actions in future.121 academic said: amounts of power, and just tell people to just blindly listen.”117 Critics believe the use of POFMA to The Government used combat misinformation on private Interviewees also lamented a decline laws, and sometimes messaging platforms may also have in the standard of public debate in backfired on the ruling party. Kirsten quite bluntly, to coerce the Indonesia, which they blamed in part Han said that to issue a POFMA order technology companies… to on the use of misinformation laws to against such content appeared not only remove posts… during the persecute journalists and to control 118 impractical, but counter-productive elections.112 the free flow of information online. Lack of transparency The Government had also invoked POFMA to pressure political parties Critics also cite a lack of transparency to remove or retract what they surrounding the use of misinformation

109 Ibid 110 Author interview with PJ Thum, Historian and Managing Director of the New Naratif, 19 August 2020 111 Author interview with Wahyu Dhyatmika 112 Interview 11 113 Interview 11 114 Author interview with Damar Juniarto 115 Interview 7 116 Interview 3 117 Author interview with PJ Thum 118 Author interview with Andreas Harsono; Interview 7 119 Author interview with Kirsten Han 120 Interview 11 121 Author interview with PJ Thum

74 La Trobe University “because that means you’re actually One Singapore academic said the going to be amplifying the falsehood continued use of other laws against that you’re supposed to fight.”122 misinformation alongside POFMA had bred distrust. It suggested that POFMA One digital platform expert said “the vast was “not strictly necessary” – and raised majority” of POFMA corrections during questions about the Government’s the election campaign were issued justification for its introduction.128 in response to criticism of either the 123 Government or government policy.  Religious vilification Kirsten Han said: The central role of religion in public and political life in Indonesia is It didn’t seem like POFMA claimed by critics to have added to the orders were taken that discriminatory nature of misinformation seriously… it feels like laws. Andreas Harsono cited the case people see POFMA not as a of a Christian politician who complained about discriminatory regulations fact checking tool but as a made “in the name of Islamic Sharia,” political tool… It’s definitely and who was subsequently reported not escaped notice that to the police by an Islamist group for 129 no PAP politicians have alleged blasphemy. been POFMA’ed. It’s always Another well-known case of blasphemy opposition politicians, and law being used against a religious minority involved the former Jakarta its mainly independent news governor Ahok (see chapter 5). Harsono sites.124 believes the use of the law against Ahok in 2017 caused his election defeat. PJ Thum agreed that this approach may In this context, Harsono held grave have backfired on the Government: concerns about changes proposed to Indonesia’s Criminal Code: “The current I think people have reached draft [of] the Criminal Code is made in a point where they realise compliance with the Islamic Sharia. It POFMA is really a tool by the will discriminate against minorities.” Government to silence its opponents and to make sure its version of narratives gets perpetuated.125

In both Singapore and Indonesia, a perceived lack of transparency around misinformation laws was seen to have contributed to decreasing trust in governments.126 As PJ Thum argued: If we don’t trust the Government to regulate appropriately, if we don’t… have trust in their motives, then any regulation is greeted with hostility and seen as… the Government trying to interfere with our freedom of speech.127

122 Author interview with Kirsten Han 123 Interview 1 124 Author interview with Kirsten Han 125 Author interview with PJ Thum 126 Author interview with Kirsten Han; Author interview with Wahyu Dhyatmika 127 Author interview with PJ Thum 128 Interview 3 129 Author interview with Andreas Harsono

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 75 Conclusion

This chapter examined recent measures that social media and cross-platform misinformation. This again raises important questions about and governments have taken to deal with misinformation the need for clear definitions, and how online. Notwithstanding some vexed questions about to deal with different types of false and definitions of misinformation and disinformation, as misleading content. explored throughout this document, interviewees identified Even harsh critics acknowledge a range of platform measures and laws that have been governments have an indispensable role to play in combatting online enacted to deal with this harmful social problem that misinformation. But the experts impacts on democratic health. interviewed expressed less confidence in governments’ ability to be both the arbiter and enforcer. This chapter has For platform companies, these measures These measures highlight the multi- shown that the legislated approach to range from removing harmful actors and pronged approach needed to tackle a tackling misinformation in Indonesia and their content and fake accounts; reducing complex problem and they underscore Singapore has prompted widespread repeat offenders’ access to platform the importance of working in partnership concern that the mandated “solutions” features and platform users’ exposure to with users and civil society. Clearly, there can produce similar effects to the low-quality content; and informing users is also a need to work cooperatively original problem by undermining media of unreliable content (not suggestive with governments and other platform and political trust and free speech in of causing real-world harm) by using companies if the endemic problem of both countries. As has been discussed, independent fact checkers to label it as misinformation is to be mitigated. As some of these consequences, such false and downgrade its spread. Also in this chapter has shown, there are still as suppressing political dissent, can the arsenal of tools was using non- mighty challenges for platforms to serve government interests; while other human (AI) and human partners, such work through to make progress in the adversities such as criminalising those as trusted civil society groups, to detect battle against misinformation. These who inadvertently spread misinformation, and forewarn the platforms of emerging include physical hurdles such as working may be unintended. In any case, the misinformation campaigns. The public across political jurisdictions, vast limitations of the legislated response to also plays an important role in reporting geographies and different languages. tackle online misinformation highlights online misinformation with mechanisms Different government definitions and important considerations for Australian to alert the platforms of content that expectations of timely responses to mis- law and policymakers to consider in their deviate from its policies and community and disinformation are often at odds with pursuit of regulating misinformation standards. These documented the platform companies’ capacities and online. Chief among these questions is community standards enable social understandings. One possible solution how to balance responsible speech with media companies to articulate their is working more closely with other free expression without overreaching and expectations of online behaviour and technology companies and different in- damaging a vibrant democracy. content to users. house divisions to deal with transnational

76 La Trobe University CHAPTER 7 Potential remedies Introduction

Policy makers seeking solutions to the rampant spread based organisations. The scope of suggested initiatives was extensive of online misinformation face formidable challenges. The – from active programs to enhance starting point for any plan to deal with online misinformation digital literacy and journalism standards, must be recognition of the need for a multi pronged approach to increased collaboration between – and for cooperation between key stakeholders. digital platforms, more transparent and inclusive government policy-making and, crucially, a quest for consensus on the vexed issue of how to define online The multiple layers of complexity knowledge of the online environments misinformation. Some of the suggestions that define this global problem defy of Indonesia and Singapore. While are already active or are being developed. simple answers. individual ideas for how to tackle the This chapter outlines some of the main issue vary significantly, broad consensus These are among the main conclusions proposed remedies from the perspective exists on the need for multi-faceted to be drawn from extensive interviews of those interviewed. strategies involving all interested parties, conducted for this project with digital including government, the technology industry experts, academics, media industry, the media, and community- professionals and others with specialist

Common definitions

Ambiguity about what defines misinformation, and how this trying to address before we relates to considerations of free speech, is a recurring theme can even start developing among those seeking and proposing solutions to the scourge any kind of regulation.3 of online misinformation. Facebook’s approach is to deal with disinformation by focusing on actors, Some observers see the resolution A social media expert said journalists, behaviours and content. Disinformation of this confusion as a necessary policy-makers and regulators, involves actors and behaviours pre-requisite to tackling the issue. including the EU Code, often conflate and what it also calls “coordinated Singapore-based journalist Kirsten Han misinformation with disinformation: inauthentic behaviours” (CIB). Tackling believed digital platforms needed to take misinformation focuses on content, more of a lead, saying they “should be When they start conflating to remove false claims online working clearer about their content standards,” all these different issues, it with third-party fact checkers and comparing how content sits alongside which in turn would clarify behavioural makes it really difficult for expectations on users.1 Han said the its community standards and policies. anyone to be able to then platforms could do this by adopting However, the company concedes that uniform international standards on decide on what the actual misinformation and disinformation can human rights and free speech rather issue is. So in order to come involve the same content. A problem than arriving at their own definitions. up with a solution, we try to with content-based approaches is that One Singaporean academic said the have very clear definitions it is not systemic but a “whack-a-mole” Singaporean government, like other approach. The social media expert said nations including those in the European about what we are talking the company was more effective when Union, were still unclear about what about. So when it comes to it examined actors and behaviours is “deemed as a falsehood or as talking about any kind of because then it could see breaches misinformation and disinformation.” misinformation regulation against platform policies and remove Part of this uncertainty, the academic pages, their contents, and networks. noted, was a function of the newness we really fight for clarity as of regulating online misinformation and to what problem we are disinformation, and ideas were evolving.2

1 Interview with Kirsten Han, interview with Andrea Carson, 25 August 2020. 2 Interview 11, interview with Andrea Carson, 20 August 2020. 3 Interview 13, interview with Andrea Carson, 16 September 2020.

78 La Trobe University When we’re taking down We’re trying to educate There is a tension between these pages and networks, policymakers on these government and the we feel that it’s a greater differences, because I think it international platforms impact. And it’s more does make a huge difference because social media scalable than just taking in terms of the types of platforms have community down one piece of content at regulation that needs to standard or community a time. This is why we look be put in place, the types guidelines, but it’s produced at these three individual of policies that companies for international needs. pillars, because it then allows put in place, and mitigation Indonesians have specific us to kind of look at the approaches that need to be needs that have to be problem more holistically, put in place.5 contextualised with the but then also carve out local conditions, especially However, definitions also needed to different solutions to address be mindful of local norms, as one in terms of for example different areas.4 Indonesian academic observed. pornography. In Indonesia, The use of the term hoax instead of the definition of pornography If policy-makers, platforms and disinformation was one example. is different in terms of journalists had the same understanding Another was what content was of the definitions of misinformation considered harmful: community standards from and disinformation there would be less international social media confusion about measures needed to platform guidelines.6 combat electoral interference and foreign interference, according to the social media policy expert:

Digital literacy

Improving citizens’ media and digital literacy was advocated by scope for governments to claim sole all groups and individuals involved in this project. Education was responsibility for regulating the problem, and therefore make it more difficult pinned as being at “the centre” of any multi-pronged approach, with for them to justify some of their more “incentives to educate the public and get the various stakeholders contentious measures. 7 involved about disinformation, misinformation and fake news.” Amid broad agreement about the value of digital literacy, different suggestions emerged on how it should Digital literacy was seen as a valuable which he said was still in development, be pursued. One academic suggested bulwark against misinformation would seek to “fight misinformation with all stakeholders, including technology because it encouraged people to “use information” and could be particularly platforms, should be given incentives critical thinking before they share applicable to situations such as the to invest resources in education of a specific type of content,” as one COVID-19 “infodemic.” “the masses.”11 Another academic with digital platform expert put it.8 In this Singapore-based academic and activist experience in conducting digital literacy way the problem could sometimes PJ Thum said that Singapore had programs in Indonesia proposed a be mitigated in advance. Indonesian always proceeded on the “impossible” nuanced approach. “I always say that activist and head of anti-misinformation premise that, “we can regulate anything, you have to be not only smart using the group MAFINDO Septiaji Eko Nugroho if we just write the laws well enough digital media, but you have to be wise likened education in this context to and if we are given enough power and and also ethical. And it’s not easy to ask “immunisation” and “inoculation.”9 if you trust enough in our benevolence people to be ethical.”12 For these reasons This approach would focus on the “the and our intelligence.”10 He believed she said the design and implementation psychological aspects” of preparing the development of more “critical of digital literacy programs needed to people in advance to avoid being “misled minds” among citizens through media consider a variety of contextual factors by misinformation.” The approach, literacy would reduce the need and beyond the education itself.

4 Interview 13. 5 Interview 13. 6 Interview 7, interview with Andrea Carson, 19 August 2020. 7 Interview 11, interview with Andrea Carson, 20 August 2020. 8 Interview 9, interview with Andrea Carson, 8 September 2020. 9 Author Interview with Septiaji Eko Nugroho, interview with Andrea Carson, 26 August 2020. 10 Author interview with PJ Thum, Historian and Managing Director of the New Naratif, 19 August 2020 11 Interview 11. 12 Interview 7, interview with Andrea Carson, 19 August 2020.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 79 A digital platform expert with experience teaching citizens “lateral reading” – a including within the one country such in digital literacy programs in Indonesia self-sufficient verification method as as in Indonesia. Notwithstanding the criticised approaches that had so far outlined in one of ’s challenges, Septiaji Eko Nugroho saw focused only on “niche markets.”13 What free online curriculum resources. merit in teaching digital literacy through was needed, he said, was “a massive Researchers worked with classroom the school system. campaign and to really educate the teachers to produce a curriculum A Singaporean academic argued masses.” While acknowledging the aimed at civic online reasoning skills that more government and platform success of initiatives in which he (COR). These skills enable students resources needed to be committed to had been involved, he said they were to search for, evaluate and verify public education about low-quality online expensive and largely deployed during information. The materials are free information and incentives put in place election campaigns. “I feel like we need a online at https://cor.stanford.edu/.15 to encourage the platforms to run these program that is much more sustainable, Similar approaches, including “train education campaigns: “I don’t think there that continues to surface whether there the trainer” programs that give adult is incentive enough for the platform is an election or not.”14 citizens the tools to educate others, were companies to invest in more tools in also supported.16 A difficulty with this Another observer, Septiaji Eko Nugroho, educating the masses. I would like to see approach is low levels of general literacy proposed a “curriculum” to guide future more resources devoted to education and dealing with different languages, educational efforts. This would include from all stakeholders.”17

Transparency

Interviewees advocated more transparency – not just by policy Damar Juniarto of SAFEnet, said digital and law makers but also news organisations, journalists and digital platforms’ transparency reports such as Facebook’s twice-yearly reports19 were platforms to assist the fight against online misinformation. One an important step for civil society groups Singapore-based academic argued that governments needed to more to check on government demands to “clearly and transparently” communicate their laws to the public.18 take down online information that “they don’t like.” He said, “This is something we need to look deeply and closely at.” He Others said it was also incumbent to yourself about your blind spots, your said the platforms’ transparency reports upon non-government actors to be subjectivity, but also whether you’re revealed when the government applied transparent. PJ Thum argued that since transparent about that with others.” a “forceful approach” to the technology “everyone who reports has an agenda,” A high level of transparency may help platforms to remove content, allowing his it was important that “you yourself are prevent citizens from being misled by organisation to scrutinise such actions. aware of that agenda, of your own biases misinformation disguised as objective as a human being, as an organisation, fact, he said. as a platform, whether you’re honest

13 Interview 9. 14 Interview 9. 15 The university’s history education group collaborated with the Poynter Institute and Local Media Association to create MediaWise, an initiative supported by Google. For details see: https://news.stanford.edu/thedish/2019/12/09/stanford-education-researchers-create-free-lessons-to-teach-digital-literacy/ 16 Interview 1, interview with Andrea Carson, 7 August 2020; Interview 7; Interview 9 17 Interview 11. 18 Interview 11. 19 These reports are published on the platform’s website at: https://transparency.facebook.com/content-restrictions

80 La Trobe University Inclusive process

Several contributors to this project recommended more involvement be the public educating themselves on of non-government actors and civil society in the policy-making regulatory challenges and taking on 20 greater responsibility for their resolution. process. PJ Thum argued that a lack of public consultation in This in turn would help governments, law making would inevitably result in “imperfect” laws that would as the public would not be “so quick to be open to exploitation by government for political purposes. get angry or point fingers when things go wrong… but instead say, ‘okay, we need to go back and participate in the Public input could not only help that encourages people to reformulation and rethinking of new remedy these flaws, but improve participate in their creation regulations’ and find ways to then public confidence in institutions: improve these laws.” (and) ensures that all these What you need is a One social media expert said ‘continuous different viewpoints are public consultation’ has often been the robust way of creating those heard, that itself… makes the missing link in Indonesia.22 However, laws and regulations which law as robust as possible… Septiaji Eko Nugroho said efforts to is open to scrutiny, which it gives people more faith in correct this deficit were underway. He endorsed collaborative partnerships seeks to encompass as many the intent of the law, in the between civil society organisations different voices as possible, intent of the people behind (CSOs), government agencies, media which has people who it.21 organisations and digital platforms, are accountable, which is such as those MAFINDO had established For Thum, the key to resolving the “crisis to improve fact checking, deliver representative, and which is of faith” in governments, their intentions digital literacy and monitor elections, transparent in how it’s made. and “the system itself” is “to make among other activities. “We need to And once you have a very sure that everyone feels like their voice tackle [misinformation] with what we robust way of creating these is heard.” A flow-on effect of a more call comprehensive efforts or a multi- inclusive law-making process would pronged approach,” Nugroho said. laws and regulations, and

Working with civil society and academics

Several participants advocated governments and digital platforms So there was this policy develop more collaborative working relationships with civil society that you would have to be groups. While recognising that “different contexts around the world” a locally registered group would complicate this task for digital platforms, Kirsten Han said this to boost political content should not stop them being more transparent with local communities — but this only works in about their content policies. This was particularly necessary when an environment where governments exploited platforms’ policies to sideline civil society registration is democratic, organisations (CSOs) from participating in public debates online. which it isn’t in Singapore because the government can She cited a case in which a social media exclusion of civil society groups. Han choose to disallow particular platform’s new “accountability policies explained that social media platforms did groups from registering.23 about who can share and boost political not quite take into consideration the local content” inadvertently led to the context when they rolled out their policy:

20 Author Interview with PJ Thum; Interview 11; Interview with Septiaji Eko Nugroho; Interview 9 21 Author Interview with PJ Thum 22 Interview 9. 23 Author interview with Kirsten Han

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 81 This policy was also problematic bad things can happen.”25 While the that “nobody” could have achieved this for media outlets and journalists. trusted partner ecosystem was still “a without the help of MAFINDO. Han explained: work in progress,” he felt his platform and Tempo editor-in-chief Wahyu Dhyatmika CSOs “we’re in a much better position expressed similar sentiments, saying now” than two to three years ago. I was an editor-in-chief “we need a road map that everyone of a Southeast Asian Septiaji Eko Nugroho believed civil can agree on.”27 He said efforts at platform that was founded society’s contribution could be collaboration to date had been disjointed, by Singaporeans, but we particularly valuable in a “hierarchical” with civil society creating its “own country like Indonesia, where authority ecosystem” comprised of “mostly media were not allowed to register figures such as religious leaders or and civil society and not connected to in Singapore because (the celebrities had “more power to spread the Government or law enforcement.” the message.” Civil society groups could Government) claimed Independent civil society groups facilitate networks between influential that we were foreign were also seen to have an important people such as local leaders, journalists role in helping to identify and remedy interference… And this and academics, “so that if there is unintended consequences of either happens a lot in Southeast possible disinformation then they can government or platform-generated help us.”26 Asia, that human rights misinformation policies.28 Kirsten Han and activist groups are He related his experience in 2018 during cited the example of Google’s recent not registered because the the West Kalimantan regional elections, decision in Singapore to ban political Government for political when MAFINDO set up a “hoax crisis advertising. She said the decision centre” and a related website, bringing disproportionately impacted on smaller reasons do not allow us together people from the Government, political parties that relied relatively to register.24 religious leaders, journalists and heavily on . In another police, “and then we educated them case of unintended consequences in Platforms have acknowledged the need (about what) we needed to do… when Indonesia, small and medium business to work more closely and productively there is misinformation coming before groups alerted WhatsApp to problems with civil society groups. A digital or after the elections.” The centre with its new limits on forwarding platform expert said he valued “having a provided clarifications on suspected messages, which had negatively affected trusted partner ecosystem, where I feel misinformation, including about religious the ability of businesses to market their confident enough that they can report and ethnic issues. Nugroho asserted products.29 things directly to me or my team… before

Public interest journalism

Improving journalism standards and providing the public with access their apparent willingness to tackle the 31 to credible news online was advocated as a potentially valuable tool in problem. Dhyatmika advocated the introduction of fact checking at more combatting online misinformation. media outlets, particularly local outlets where it was “not that widespread yet,” to enable more detection and countering of Wahyu Dhyatmika said Tempo’s you will be able to detect any misinformation “early on.” international reporting collaborations misinformation early and try had “strengthened our credibility Platforms’ investments in journalism and reputation,” which had improved to counter it early. And that’s projects such as Google News Initiative engagement with audiences and, in turn, better than waiting for it to and Facebook’s Journalism Project would lead to more people alerting them become viral.30 were commended by respondents. to suspected misinformation as audience One academic said these projects help size and audience trust increased: Dhyatmika also highlighted the value quality journalism survive and would like of fact checking operations in media to see these models replicated by other 32 I think it comes down companies as a part of the fight against technology companies. to how engaged your misinformation, while digital platforms audience (is) with you in the that supported fact checking were commended by one academic for newsroom. Because then

24 Author interview with Kirsten Han 25 Interview 9 26 Author interview with Septiaji Eko Nugroho 27 Interview with Dhyatmika, interview with the author, 24 August 2020 28 Interview 6, interview with Andrea Carson, 25 August 2020; Interview 9 29 Interview 9 30 Author interview with Wahyu Dhyatmika 31 Interview 11 32 Interview 11

82 La Trobe University Collaboration between technology platforms

Greater collaboration and information sharing between technology academia is really key to solving 36 platforms is seen by some industry figures as a necessity in the the problem. Less clear was how coordinated communications between fight against online misinformation, given the ubiquity of the problem. industry and other allies were in the A digital platform expert suggested more cooperation would require a Asia Pacific. As another digital platform shift in attitudes by some platforms towards others. While platforms expert in the region stated: might be competitors in business, in the information space they were “friends.”33 He proposed legislation to facilitate collaboration, It’s very seldom that we saying: “Legislation should focus on… how civic society, journalists partner with other social and platform and tech industries could work together to develop media platforms. It’s almost best practice.”34 as if we just have to own the problem ourselves and address it head on, whether However, a social media expert said intelligence authorities and between the cooperation was more common than technology platforms such as Google, it’s directly with the regulator what might be known by the public. Twitter, Facebook, Microsoft and Reddit. or in partnership with the When one digital platform unearths a This coordination included annual CSOs. But, I think the key disinformation campaign they share conferences to share intelligence about is not so much partnering it with the others. “There’s constant disinformation threats and to exchange engagement that takes place between information: with the folks in the industry us and all the other industry partners but partnering with the In this way we can then tackle the in this space of disinformation.”35 problem together because we know that CSO partners… that’s the The expert said Europe was the most collaboration with our industry, partners, approach that we have taken advanced in information sharing with governments, civil society, and even so far.37

Conclusion

Central to the recommendations to manage online those with specialist knowledge such as policy-makers and academics. Inherent misinformation is the need for agreement on definitions, so in the duplication of recommendations that those trying to find remedies are actually addressing was the need for the different parties to the same problem. It is not clear that this has been the case communicate to the public about what to date. From the interviews it was also clear that mitigating they are already doing about false news online and why. misinformation requires a multi-pronged approach.

Digital education programs cannot ranging from civil society organisations, be done in isolation without efforts academia, journalism, digital platforms, to reduce false information and government and policy-makers. Likewise, remove harmful content. The experts any regulatory initiative to address also agreed that such a complex misinformation online needed to include problem requires the involvement those with grass-roots experience about and cooperation of different actors, the forms in which it appears, as well as

33 Interview 4 34 Interview 4 35 Interview 14. Correspondence with Andrea Carson, 18 November 2020. 36 Interview 13. 37 Interview 4.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 83 CHAPTER 8 Concluding analysis Concluding analysis

The globalisation of digital communication technologies in Governments have choices in how to respond. This report highlights the mixed the 21st century has delivered unprecedented opportunities and sometimes negative experiences and challenges. It promised a more inclusive digital public of two Asia-Pacific countries that have sphere, with the internet playing a democratising role to chosen a legislative pathway. Others, enable citizens from around the globe to come together. It such as the European Commission, have so far opted for voluntary regulation. also changed the ways in which we think about news. As The benefits and drawbacks of Europe’s New York academic Jay Rosen famously wrote, “the people approach feature in chapter three. formerly known as the audience” are now both consumers These choices come at a critical time – and producers of content.1 when media and political freedoms once taken for granted appear to be under threat across the globe. Both illiberal and The rise of the digital technology giants processes and public health and safety, liberal governments have presided over has overshadowed the cultural power to inciting violence against minorities 14 years of what the democracy activist of legacy media. Facebook, Twitter, and other vulnerable groups and group Freedom House describes as an YouTube and others have provided individuals. In cases when it leads to “assault” on democracy and pluralism. the means for anyone with an internet harm, misinformation can overlap with The group warns that digital platforms connection and a keyboard, smartphone, disinformation, which for the purposes and social media are the new frontier 4 camera or voice-activated software to of this study is regarded as false or in this world-wide attack on freedoms. participate in creating and exchanging manipulated content that causes The majority of countries that have news and information. political, personal or financial harms.3 opted for fake news laws already restrict their citizens’ political rights and civil This report deliberately focuses largely Misinformation’s pernicious spread liberties, and are considered “partly free” on the challenges, not the opportunities, has made it a top-level policy concern or “not free” on a democracy scorecard. of these profound developments. for governments around the world. In Singapore (with a score of 50 out of Supported by detailed case studies Australia, the Morrison Government, 100) and Indonesia (61) are lowly ranked of the experiences of Indonesia and acting on the recommendations of compared with Australia’s score of 97.5 Singapore, we have examined how an inquiry, has instructed the digital However, Australia scores less well on global communication networks have technology platforms to implement a internet freedom (76 out of 100) due to facilitated the prolific spread of online voluntary Code of Disinformation by existing “limits on content, violations of falsehoods – primarily misinformation 2021. The findings of this study aim to user rights and obstacles to access”.6 and disinformation – to create what assist public debate surrounding the academic Claire Wardle describes as a development of the code. These indicators serve as a warning world-wide phenomenon of “information as countries pursue regulatory and In pursuing remedies to the spread of disorder”.2 We have explored the legislative pathways to tackle online online misinformation, governments in potentially far-reaching consequences misinformation. The key findings of liberal democracies like Australia face of the spread of online misinformation, this report highlight worrying effects the considerable challenge of trying and how governments and technology of government over-reach through to strike a balance between mitigating platforms have responded with various legislative responses to misinformation. harm, and preserving basic democratic measures to tackle the problem. The governments of both Singapore tenets such as pluralism, freedom of and Indonesia have been accused of The report finds that misinformation, expression and media freedom. misusing their anti-misinformation even when transmitted innocently, can laws to censor or silence a wide have detrimental consequences, ranging spectrum of critics, including journalists, from undermining democratic political dissidents and human rights

1 Jay Rosen, “The people formerly known as the audience,” in Participation and Media Production: Critical Reflections on Content Creation, eds. Nico Carpentier & Benjamin DeCleen (Newcastle upon Tyne, UK: Cambridge Scientific Publishers Ltd, 2008), 163-165. 2 Claire Wardle, “The Need for Smarter Definitions and Practical, Timely Empirical Research on Information Disorder.” Digital Journalism, 6, 2018 no. 8: 953. 3 James Meese, J and Edward Hurcombe, Regulating Misinformation: Policy Brief (Melbourne, Analysis & Policy Observatory, RMIT University, 2020), https://apo.org.au/node/309357 4 Sarah Repucci, “A Leaderless Struggle for Democracy: Democracy and Pluralism are under assault”, Freedom House, 2020, https://freedomhouse.org/report/freedom- world/2020/leaderless-struggle-democracy 5 Ibid 6 Freedom House, “Freedom of the Net 2020”, (Freedom House, 2020), https://freedomhouse.org/country/australia/freedom-net/2020#C

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 85 campaigners, as well as religious sense of shared responsibility by cooperation between platforms; leaders. Experts criticised the tendency governments, digital platforms and continuing to fund academic research; of these governments to prioritise users. Non-government organisations, and greater data transparency for law enforcement over alternative, academia, and traditional media also researchers to engender understanding less drastic mechanisms to deal with have vital roles to play in the promotion about the scale of the problem and online misinformation. Academics and development of digital literacy methods to mitigate it. observe that fake news laws addressing education and sustaining a culture of The findings also highlight the misinformation can only be successfully reliable and trustworthy information. As importance of a multi-pronged implemented without damaging political already happens in Europe, these key response to this complex problem, and civil rights in robust democracies like players could generate considerable with input needed from a multitude France, which have added checks and benefits by holding regular meetings of actors ranging from policy makers balances to their legislative measures.7 to communicate and share knowledge and regulators to on-the-ground of known online risks and strategies to Empirical data highlights some positive fact checkers, educators and the mitigate misinformation including being results for non-regulatory measures organisations that work with victims mindful not to inadvertently, or otherwise, already being employed by digital of online misinformation campaigns. amplify it. platforms and others to tackle online Traditional media also have a role to play misinformation. These include: The development of a successful as misinformation can be amplified in voluntary code requires input from news reporting and distributed on its y Tools and campaigns to improve diverse stakeholders. But first there digital networks. digital literacy needs to be consensus on the meaning y Third-party fact checking of content The case studies examining Indonesia of key terms such as misinformation, and use of false information and Singapore highlight the need for disinformation and fake news so that warning labels regulatory responses to be customised everyone knows they are addressing y Requiring proof of identity to improve to national contexts, but to also the same problems. The present lack of account transparency incorporate international principles definitional clarity across the globe is a y Removing harmful content or limiting such as universal human rights. Expert key barrier to progress. its algorithmic spread participants in this project agreed that y Reducing services to users who Misinformation on global platforms was there was, as yet, no gold standard breach community standards also difficult to measure at a national for regulating misinformation. While y Supporting journalism by providing level, as it is not confined to country the EU code is often referenced and resources and financial support to borders. Adding further complexity to seen as adaptable to the Australian news media organisations. detecting and measuring misinformation, context, experts warned it was not are country-specific issues such flawless, and cautioned against its However, public recognition of this as Indonesia’ hundreds of spoken wholesale adoption. work appears to be limited. Platforms languages, many with different meanings would serve themselves and the public Despite the lack of consensus on for similar words that confound artificial well to improve the visibility of existing how to define, measure and mitigate intelligence techniques. measures, and to more widely publicise misinformation online, clear unanimity the role the public can play in limiting the Potential remedies identified in the exists about the urgent need to curb it spread of misinformation. report include further developing to improve the quality of information trusted partnerships with civic leaders; in the public sphere, which is vital to a Certainly, more work is needed to expanding digital literacy education healthy democracy. curb the scourge of online falsehoods, campaigns; expanding resources perhaps starting with a greater for public interest journalism; more

7 James Meese, J and Edward Hurcombe, Regulating Misinformation: Policy Brief (Melbourne, Analysis & Policy Observatory, RMIT University), https://apo.org.au/node/309357

86 La Trobe University Acknowledgements

This research project was focused on Indonesia and also to regulators at the Australian Media and Communication Authority Singapore and designed to produce new insights into the for sharing their insights on this nature of the problem of online misinformation. It would not pernicious problem. have been possible were it not for the generous assistance My gratitude extends to my La Trobe and time of many people and organisations who I would like University colleagues. Thanks to the to sincerely thank here. Head of my Department of Politics, Media and Philosophy, Professor James Leibold, for affording me time I begin with Facebook who provided Chairman of MAFINDO (civil society to attend to this urgent project. Many funding for this independent research group tackling hoaxes), Septiaji Eko thanks also to Dylan Bird for taking and gave me access to key people in Nugroho; Author and researcher on over my teaching so that I could meet the Asia Pacific and Europe. Thank you media and human rights with Human project deadlines, and for his excellent to Mia Garlick, Director of Public Policy, Rights Watch, Andreas Harsono; and proofreading skills. My gratitude to La Australia, and the Pacific Southeast Asia Freedom of Expression, Trobe’s research ethics team and Min Islands, and to Josh Machin, Head of Executive Director (SAFEnet) Damar Chul in the research office for assistance Public Policy, Australia, for the research Juniarto, and academic Professor Ang in meeting University requirements to opportunity and their support. Others Peng Hwa at Nanyang Technological undertake this work. at Facebook in Indonesia, Singapore University, Singapore. I’d also like to In finalising the report, I am indebted to and Europe have also assisted me acknowledge everyone’s flexibility to the brilliant Tom Ormonde, whose eye with details about how they work and convert to online interviews due to the for detail, commitment to clarity and what they see as the key challenges of COVID-19 pandemic. I also add the usual, knowledge of journalism improved the tackling online misinformation. I thank but sincere disclaimer, that any errors are final version. I am also very thankful them for their time and patience with my own. to my research assistants Liam Fallon my questions. To my academic colleagues who and Sophie Squires for their roles in I also want to thank the journalists, have offered valuable guidance and the information gathering and writing editors, human rights activists, fact assistance along the way, I also extend processes. Further thanks to Liam for checkers and academics who agreed to my heartfelt thanks. These include spending many hours on this project be interviewed so that I could advance Associate Professor Dirk Tomsa (La and his unrelenting efforts. understandings about the problems Trobe University); Dr Andrew Gibbons To La Trobe’s publishing team, I express of misinformation online in Singapore (University of Texas in Austin), and Dr my appreciation for their professional and Indonesia, and the consequences James Meese (RMIT). Thank you also services, with a special thanks to the of the laws introduced to tackle it. to learned colleagues at the Centre Digital Creative Team and IVE Creative Of those whose permission I have to for Media Transition at the University Services, La Trobe's agency partner. name, I thank historian and managing of Technology Sydney and the Fact editor of the New Naratif, PJ Thum; Checking Research Group at RMIT for Andrea Carson journalist Kirsten Han; Editor-in-Chief including me in valuable forums about of Tempo magazine, Wahyu Dhyatmika; tackling false news online. Thank you

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 87 Reference list

About Hungary. “Coronavirus update: Bill on rescinding the government’s emergency powers submitted to parliament today.” Published May 26, 2020. http://abouthungary.hu/news-in-brief/coronavirus-update-bill-on-rescinding-the-governments- emergency-powers-submitted-to-parliament-today/ Abu Baker, Jalelah. “Exemption of Government from being sued under POHA amendments a ‘glaring omission’: Pritam Singh.” Channel News Asia, May 7, 2019. https://www.channelnewsasia.com/news/singapore/exemption-government-sued-under-poha- amendments-pritam-singh-11511118 Abu Baker, Jalelah. “Tan Wu Meng’s opinion piece on Pritam Singh was ‘serious’ and ‘thoughtful’: Shanmugam.” Channel News Asia, June 21, 2020. https://www.channelnewsasia.com/news/singapore/tan-wu-meng-pritam-singh-aflian-shanmugam-12856406 Academics Against Misinformation. “POFMA: Letter to Education Minister.” Published April 11, 2019. http://www.academia.sg/pofma-letter/ Academics Against Misinformation. “POFMA: Media release.” Published April 14, 2019. http://www.academia.sg/pofma-media-release/ Act to Improve Enforcement of the Law in Social Networks (Network Enforcement Act). Bundestag. July 12, 2017. https://www.bmjv.de/SharedDocs/Gesetzgebungsverfahren/Dokumente/NetzDG_engl.pdf?__blob=publicationFile&v=2 Al Jazeera. “Indonesia passes controversial anti-terror laws to fight ISIL.” May 26, 2018. https://www.aljazeera.com/news/2018/05/indonesia-passes-controversial-anti-terror-laws-fight-isil-180525055635674.html Amnesty International. “Open letter on the increasing use of makar charges against Papuan activists to stifle freedom of expression.” Published October 2, 2019. https://www.amnesty.id/open-letter-on-the-increasing-use-of-makar-charges-against- papuan-activists-to-stifle-freedom-of-expression/ Aravindan, Aradhana. “Google says Singapore’s fake news law could hamper innovation.” Reuters, May 9, 2019. https://www.reuters. com/article/us-singapore-politics-google/google-says-singapores-fake-news-law-could-hamper-innovation-idUSKCN1SF08X Aris, Kieren and Rachel Fielden. “Tackling online disinformation in Indonesia through media literacy: The redirect method.” Paper presented at Fake News and Elections in Asia Conference. Bangkok, Thailand, 10-12 July, 2019. Accessed November 13, 2020, https://asiacentre.org/wp-content/uploads/2019/09/Fake-News-Conference-Proceeding.pdf Article 19 and Aliansi Jurnalis Independen. Freedom of Expression and the Media in Indonesia. London and Jakarta: Article 19 and Aliansi Jurnalis Independen, 2005. https://www.refworld.org/pdfid/4754185f0.pdf ASEAN Parliamentarians for Human Rights. “Singapore: Amend the ‘fake news’ bill.” Published May 7, 2019. https://aseanmp.org/2019/05/07/singapore-fake-news-bill/?fbclid=IwAR114nescHaE8VQ5NDYlV8YxxTcKi_6ogBLXROZFR- 8C5Uf5OnwCVjePQZ8. Association of Women for Action and Research. No new restrictions needed: AWARE submission to Select Committee on Deliberate Online Falsehoods. Association of Women for Action and Research, 2018. http://d2t1lspzrjtif2.cloudfront.net/wp-content/uploads/ AWARE-submission-to-Select-Committee-on-Deliberate-Online-Falsehoods-1.pdf. Au Waipang, Alex. “Singapore bloggers wary of news site licence scheme.” Committee to Protect Journalists. Published June 4, 2013. https://cpj.org/2013/06/singapores-news-site-license-plan-raises-questions/ Australian Communications and Media Authority. Misinformation and News Quality on Digital Platforms in Australia: A Position Paper to Guide Code Development. Canberra: ACMA, 2020. https://www.acma.gov.au/sites/default/files/2020-06/Misinformation%20and%20news%20quality%20position%20paper.pdf Australian Competition and Consumer Commission. Digital Platforms Inquiry: Final Report. Canberra: ACCC, 2019. https://www.accc.gov.au/system/files/Digital%20platforms%20inquiry%20-%20final%20report.pdf Australian Government. Regulating in the Digital Age: Government Response and Implementation Roadmap for the Digital Platforms Inquiry. Canberra: Commonwealth of Australia, 2019. https://treasury.gov.au/sites/default/files/2019-12/Government-Response-p2019-41708.pdf Barthel, Michael, Amy Mitchell and Jesse Holcomb. “Many Americans believe fake news is sowing confusion.” Pew Research Centre. Published December 15, 2016. https://www.journalism.org/2016/12/15/many-americans-believe-fake-news-is-sowing-confusion/ BBC. “Singapore ruling PAP party wins elections, but support falls.” July 10, 2020. https://www.bbc.com/news/world-asia-53358650 Bertrand, Jacques. Nationalism and Ethnic Conflict in Indonesia. Cambridge: Cambridge University Press, 2003. DOI:10.1017/ CBO9780511559341 Bhwana, Petir Garda. “Ahok reflects on never becoming a government official again.” Tempo, October 22, 2019. https://en.tempo.co/read/1262677/ahok-reflects-on-never-becoming-a-government-official-again

88 La Trobe University Bhwana, Petir Garda. “Coronavirus-crackdown aimed at anti-President smears, hoaxes.” Tempo, April 6, 2020. https://en.tempo.co/read/1328272/coronavirus-crackdown-aimed-at-anti-president-smears-hoaxes Board, Jack. “Inside Indonesia’s ‘fake news’ war room, fighting political hoaxes in election season.” Channel News Asia, April 13, 2019. https://www.channelnewsasia.com/news/asia/indonesia-election-fake-news-war-room-fighting-political-hoaxes-11439398 Bothwell, Ellie. “Fake news laws may ‘catch on’ during coronavirus.” Times Higher Education. Published April 6, 2020. https://www.timeshighereducation.com/news/fake-news-laws-may-catch-during-coronavirus Bowie, Nile. “The changing face of Singapore democracy.” Asia Times. September 10, 2020. https://asiatimes.com/2020/09/the-changing-face-of-singapore-democracy/ Britannica. “Pancasila.” Accessed August 20, 2020. https://www.britannica.com/topic/Pancasila. Bruns, Axel, Stephen Harrington, and Edward Hurcombe. “‘Corona? 5G? Or Both?’: The dynamics of COVID-19/5G conspiracy theories on Facebook.” Media International Australia 177, no. 1 (November 2020): 12–29. https://doi.org/10.1177/1329878X20946113. Butt, Simon. “Why is Ahok in prison? A legal analysis of the decision.” Indonesia at Melbourne, June 6, 2017. https://indonesiaatmelbourne.unimelb.edu.au/why-is-ahok-in-prison-a-legal-analysis-of-the-decision/ Carson, Andrea. Investigative Journalism, Democracy and the Digital Age. New York: Routledge, 2020. Central Intelligence Agency. “The world factbook: Indonesia.” Accessed August 19, 2020. https://www.cia.gov/library/publications/the-world-factbook/geos/id.html Chairil, Tangguh. “Indonesia’s intelligence service is coming out to counter COVID-19.” The Diplomat, June 19, 2020. https://thediplomat.com/2020/06/indonesias-intelligence-service-is-coming-out-to-counter-covid-19/ Chang, Nicole. “Judge dismisses The Online Citizen’s POFMA challenge.” Channel News Asia, February 19, 2020. https://www.channelnewsasia.com/news/singapore/the-online-citizen-toc-pofma-challenge-appeal-dismiss-12449826 Charlton, Emma. “How Finland is fighting fake news – in the classroom.” World Economic Forum. Published May 21, 2019. https://www.weforum.org/agenda/2019/05/how-finland-is-fighting-fake-news-in-the-classroom/ Chauvel, Richard and Ikrar Nusa Bhakti. The Papua Conflict: Jakarta’s Perceptions and Policies. Washington, D.C: East-West Center Washington, Policy Studies 5, 2004. Chen, Siyuan and Chia Chen Wei. “Singapore’s latest efforts at regulating online hate speech.” Research Collection School of Law (2019): n.p. https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=4879&context=sol_research Chew, Amy. “’Chinese army uniforms’ in Indonesia: Another hoax to discredit Beijing, Widodo.” South China Morning Post, July 29, 2020. https://www.scmp.com/week-asia/politics/article/3095217/chinese-army-uniforms-indonesia-another-hoax-discredit-beijing Coconuts Jakarta. “Indonesian gov’t justifies maintaining internet block in Papua despite officials declaring security restored.” August 28, 2019. https://coconuts.co/jakarta/news/indonesian-govt-justifies-maintaining-internet-block-in-papua-despite- officials-declaring-security-restored/ Coconuts Jakarta. “RKUHP Explainer: All the controversial articles in Indonesia’s criminal code overhaul.” September 19, 2019. https://coconuts.co/bali/features/rkuhp-explainer-all-the-controversial-articles-in-indonesias-criminal-code-overhaul-2/ Coconuts Jakarta. “Senior Indonesian minister suggests anti-terrorism law be used against election hoax spreaders.” March 21, 2019. https://coconuts.co/jakarta/news/senior-indonesian-minister-suggest-anti-terrorism-law-used-election-hoax-spreaders/ Commission High Level Expert Group on Fake News and Online Disinformation. A Multi-Dimensional Approach to Disinformation: Report of the Independent High-Level Expert Group on Fake News and Online Disinformation. Brussels: European Commission, 2018. http://ec.europa.eu/newsroom/dae/document.cfm?doc_id=50271 Committee to Protect Journalists. “Singaporean authorities shut down The Real Singapore news website.” Published May 5, 2015. https://cpj.org/2015/05/singaporean-authorities-shut-down-the-real-singapo/ Courthouse News Service. “Hundreds arrested for ‘fake news’ in Asia.” April 10, 2020. https://www.courthousenews.com/asia-arrests-hundreds-for-fake-news-on-virus/. Criminal Code [KUHP] (Indonesia). Government of Indonesia, 1946. https://www.unodc.org/res/cld/document/idn/indonesian_penal_code_html/I.1_Criminal_Code.pdf Crowther, Craig. “Viktor Orban and Hungarian Government poised to give up unrestrained powers.” Euronews, May 27, 2020. https://www.euronews.com/2020/05/26/viktor-orban-and-hungarian-government-poised-to-give-up-unrestrained-powers Dahlgren, Paul. “Media, knowledge and trust: The deepening epistemic crisis of democracy.” Journal of the European Institute for Communication and Culture 25, no. 1-2 (2018): 20 -27. DOI: 10.1080/13183222.2018.1418819 Daskal, Jennifer. “This ‘fake news’ law threatens free speech. But it doesn’t stop there.” New York Times,, May 30, 2019. https:// www.nytimes.com/2019/05/30/opinion/hate-speech-law-singapore.html DDCMS. The Cairncross Review: A Sustainable Future for Journalism. London: Department for Digital, Culture, Media & Sport, February 12, 2019. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/ file/779882/021919_DCMS_Cairncross_Review_.pdf Defamation Act (Singapore). Parliament of Singapore, 1965. https://sso.agc.gov.sg/Act/DA1957 Digital Industry Group Inc. Australian Code of Practice on Disinformation: Draft for Public Consultation. DIGI, October 16, 2020. https://digi.org.au/wp-content/uploads/2020/10/Disinformation-Code-draft-for-public-consultaion-.pdf

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 89 Dittrich, Paul-Jasper. Tackling the Spread of Disinformation: Why a Co-regulatory Approach is the Right Way Forward for the EU. Berlin: Jacques Delors Centre, December 12, 2019. https://www.bertelsmann-stiftung.de/fileadmin/files/BSt/Publikationen/ GrauePublikationen/EZ_JDI_BST_Policy_Paper_Disinformation_Dittrich_2019_ENG.pdf. Draft Revised Criminal Code [RKUHP] (Indonesia). Government of Indonesia, 2019. https://www.hukumonline.com/pusatdata/detail/17797/rancangan-undang-undang-2019# Drake, Christine. National Integration in Indonesia: Patterns and Policies. Honolulu: University of Hawaii Press, 2019. Dursin, Kanis. “Hoax stories hamper Indonesia’s COVID-19 fight.” Bangkok Post, July 6, 2020 https://www.bangkokpost.com/opinion/opinion/1946480/hoax-stories-hamper-indonesias-covid-19-fight East StratCom Task Force. Action Plan on Strategic Communication. Brussels: European Commission, June 22, 2015. http://archive.eap-csf.eu/assets/files/Action%20PLan.pdf Eddyono, Supriyadi Widodo, Zainal Abidin, Arsil and Emerson Yuntho. Brief Notes on the Proposed Discussion of 2015 Criminal Code Bill. National Alliance for Criminal Code Reform, March 26, 2015. https://elsam.or.id/wp-content/uploads/2015/08/Brief-Notes-on-the-Proposed-Discussion-of-2015-Criminal-Code-Bill.pdf Elections Department Singapore. “Parliamentary elections.” Accessed November 13, 2020. https://www.eld.gov.sg/elections_parliamentary.html European Commission. “A Europe that protects: EU reports on progress in fighting disinformation ahead of European Council.” Published June 14, 2019. https://ec.europa.eu/commission/presscorner/detail/en/IP_19_2914 European Commission. “First baseline reports – Fighting COVID-19 disinformation monitoring program.” Published September 10, 2020. https://ec.europa.eu/digital-single-market/en/news/first-baseline-reports-fighting-covid-19-disinformation-monitoring- programme. European Commission. “Protecting European democracy from interference and manipulation – European Democracy Action Plan.” Accessed November 19, 2020. https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12506-European-Democracy-Action-Plan European Commission. “The Digital Services Act package.” Last updated June 22, 2020. https://ec.europa.eu/digital-single-market/en/digital-services-act-package European Commission. Action Plan against Disinformation. Brussels: European Commission, December 5, 2018 https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018JC0036&qid=1583456906698&from=EN European Commission. Assessment of the Code of Practice on Disinformation – Achievements and Areas for Further Improvement. Brussels: European Commission, September 10, 2020. https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=69212 European Commission. Code of Practice on Disinformation: First Annual Reports. Brussels: European Commission, October 2019. https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=62698. European Commission. EU Code of Practice on Disinformation. Brussels: European Commission, September 26, 2018. ttps://ec.europa.eu/newsroom/dae/document.cfm?doc_id=54454 European Commission. Joint Framework on Countering Hybrid Threats. Brussels: European Commission, April 6, 2016. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52016JC0018&from=EN. European Commission. Joint Framework on Countering Hybrid Threats from July 2017 to June 2018. Brussels: European Union, June 13, 2018. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018JC0014&from=EN. European Commission. Report on the Implementation of the Action Plan against Disinformation. Brussels: European Commission, June 14, 2019. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52019JC0012&from=EN European Commission. Tackling COVID-19 Disinformation – Getting the Facts Right. Brussels: European Commission, June 10, 2020. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52020JC0008&from=EN European Commission. Tackling Online Disinformation: A European Approach. Brussels: European Commission, April 26, 2018. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018DC0236&qid=1583456906698&from=EN European Committee of the Regions. Opinion of the European Committee of the Regions on “Tackling Online Disinformation: A European Approach”. Brussels: Official Journal of the European Union, May 16, 2019. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018IR3908&qid=1583456906698&from=EN European Council. European Council Meeting (19 and 20 March 2015) – Conclusions. Brussels: European Council, March 20, 2015. https://www.consilium.europa.eu/media/21888/european-council-conclusions-19-20-march-2015-en.pdf European Court of Human Rights. European Convention on Human Rights. Brussels: European Court of Human Rights, 2010. https://www.echr.coe.int/Documents/Convention_ENG.pdf European Economic and Social Committee. Opinion on Action Plan against Disinformation. Brussels: European Commission, March 20, 2019. https://webapi2016.eesc.europa.eu/v1/documents/EESC-2018-06302-00-00-AC-TRA-EN.docx/content European Parliament and the Council. Directive (EU) 2018/1808 of the European Parliament and of the Council. Brussels: European Parliament, November 14, 2018. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32018L1808&from=EN European Parliament. Charter of Fundamental Human Rights of the European Union (2000/c 364/01). Brussels: European Parliament, December 18, 2000. https://www.europarl.europa.eu/charter/pdf/text_en.pdf European Regulators Group for Audiovisual Media Services. ERGA Report on Disinformation: Assessment of the Implementation of the Code of Practice. Brussels: European Commission, May 2020. https://erga-online.eu/wp-content/uploads/2020/05/ERGA-2019-report-published-2020-LQ.pdf

90 La Trobe University European Union External Action Service. “Questions and Answers about the East StratCom Task Force.” Published December 5, 2018. https://eeas.europa.eu/headquarters/headquarters-homepage/2116/-questions-and-answers-about-the-east-stratcom- task-force_en EUvsDisInfo. “About.” Accessed November 17, 2020. https://euvsdisinfo.eu/about/ EUvsDisInfo. “Removal of three cases further to complaints by Dutch media.” Published March 8, 2018. https://euvsdisinfo.eu/removal-of-three-cases-further-to-complaints-by-dutch-media/ Fathurohman, Irfan. “ICJR urges parliament and government to not need to hurry to pass the Criminal Code bill.” IDNTimes. Published July 11, 2020. https://www.idntimes.com/news/indonesia/irfanfathurohman/icjr-desak-pemerintah-dpr-tidak-buru- buru-sahkan-ruu-kuhp/4 Fiorentino, Michael-Ross. “France passes controversial ‘fake news’ law.” Euronews, May 9, 2019. https://www.euronews.com/2018/11/22/france-passes-controversial-fake-news-law Fisher, Caroline. “Australia.” In Reuters Institute Digital News Report 2019, edited by Nic Newman, Richard Fletcher, Antonis Kalogeropoulos and Rasmus Kleis Nielsen, 132-133. Oxford: Reuters Institute for the Study of Journalism, 2019. https://reutersinstitute.politics.ox.ac.uk/sites/default/files/inline-files/DNR_2019_FINAL.pdf Foreign Correspondents Association of Singapore. “The Foreign Correspondents Association of Singapore is deeply concerned about the Singapore government’s passing of a bill…” Facebook. May 9, 2019. https://www.facebook.com/FCASG.org/posts/2358487210882553 Formichi, Chiara. “Violence, sectarianism, and the politics or religion: Articulations of anti-Shi’a discourses in Indonesia.” Indonesia, no. 98 (2014): 1-27. FORUM-ASIA. “Singapore: Use of anti-fake news law reveals institutional framework that undermines the promotion and protection of human rights.” Published February 11, 2020. https://www.forum-asia.org/?p=30931 Francois, Camille. Actors, Behaviours, Content: A Disinformation ABC: Highlighting Three Vectors of Viral Deception to Guide Industry & Regulatory Responses. The Transatlantic High Level Working Group on Content Moderation Online and Freedom of Expression, September 20, 2019. https://science.house.gov/imo/media/doc/Francois%20Addendum%20to%20Testimony%20-%20ABC_ Framework_2019_Sept_2019.pdf Freedom House. “Freedom in the World 2020: Indonesia.” Accessed November 13, 2020. https://freedomhouse.org/country/indonesia/freedom-world/2020 Freedom House. “Freedom on the Net 2019: Indonesia.” Accessed November 13, 2020. https://freedomhouse.org/country/indonesia/freedom-net/2019 Freedom House. “Freedom on the Net 2020: Australia.” Accessed November 20, 2020. https://freedomhouse.org/country/australia/freedom-net/2020#C Freedom House. Freedom of the Press 2017. Freedom House, 2017. https://freedomhouse.org/report/freedom-press/freedom-press-2017 Freedom House. Freedom on the Net 2016. Freedom House, 2016. https://freedomhouse.org/sites/default/files/2020-02/Freedom_on_the_Net_2016_complete_book.pdf Freedom House. Freedom on the Net 2017: Indonesia. Freedom House, 2017. https://www.justice.gov/eoir/page/file/1026001/download Freedom House. Special Report 2018: Attacks on the Record – The State of Global Press Freedom, 2017-2018. Freedom House, 2018. https://freedomhouse.org/report/special-report/2018/attacks-record. Freischlad, Nadine. “Controversial ‘right to be forgotten’ finds its way into Indonesian law.” Tech in Asia. Published December 1, 2016. https://www.techinasia.com/indonesia-recognizes-right-to-be-forgotten Function 8. “Submission to the Parliamentary Committee on Deliberate Online Falsehoods – Causes, Consequences and Countermeasures.” Facebook. February 28, 2018. ttps://m.facebook.com/story.php?story_fbid=911516705691971&id=350013055175675 Funke, Daniel and Daniela Flamini. “A guide to anti-misinformation actions around the world.” Poynter. Accessed November 14, 2020. https://www.poynter.org/ifcn/anti-misinformation-actions/#indonesia Funke, Daniel. “Italians can now report fake news to the police. Here’s why that’s problematic.” Poynter. Published January 19, 2018. https://www.poynter.org/fact-checking/2018/italians-can-now-report-fake-news-to-the-police-heres-why-thats-problematic/ Funke, Daniel. “Three publications are suing the EU over fake news allegations.” Poynter. Published February 28, 2018 https://www.poynter.org/fact-checking/2018/three-publications-are-suing-the-eu-over-fake-news-allegations/ George, Cherian. “Online falsehoods bill: Will words in legislation mean whatever S’pore govt chooses them to mean?” Mothership. Published April 6, 2019. https://mothership.sg/2019/04/online-falsehoods-cherian-george/ Guillaume, Marine. Combating the Manipulation of Information – A French Case. Helsinki: Hybrid CoE, 2019. https://www.hybridcoe.fi/wp-content/uploads/2019/05/HybridCoE_SA_Combating-the-manipulation-of-information.pdf Gunawan, Budi and Barito Mulyo Ratmono. “Social media, cyberhoaxes and national security: Threats and protection in Indonesian cyberspace.” International Journal of Network Security 22, no 1. (2020): 93-101. Gunia, Amy. “Singapore’s prime minister threatens website editor with libel.” Time, September 2, 2019. https://time.com/5666835/singapore-prime-minister-lee-hsien-loong-libel/

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 91 Hamid, Usman. “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop.” The Conversation, November 25, 2019. https://theconversation.com/indonesias-information-law-has-threatened-free-speech-for- more-than-a-decade-this-must-stop-127446 Han, Kirsten. “Big Brother’s regional ripple effect.” Index on Censorship 48, no. 2 (2019): 67–69. Han, Kirsten. “Coronavirus: How effective is Singapore’s anti-fake news law?” The News Lens, April 8, 2020. https://international.thenewslens.com/article/133521. Han, Kirsten. “WTC Long Read: A mother’s concerns in a time of POFMA.” We The Citizens. Published November 9, 2020. https://wethecitizens.substack.com/p/wtc-long-read-a-mothers-concerns Harsono, Andreas. “Indonesia’s journalists grapple with Islamism.” Human Rights Watch. Published August 21, 2019 https://www.hrw.org/news/2019/08/21/indonesias-journalists-grapple-islamism Harsono, Andreas. “Indonesia’s president urges delay on Criminal Code vote.” Human Rights Watch. Published September 23, 2019. https://www.hrw.org/news/2019/09/23/indonesias-president-urges-delay-criminal-code-vote Hasani, Ismail. “The decreasing space for non-religious expression in Indonesia.” In Religion, Law and Intolerance in Indonesia, edited by Tim Lindsey and Helen Pausacker, Chapter 9. London and New York: Routledge, 2016. Hatta, Muhammad. “The spread of hoaxes and its legal consequences.” International Journal of Psychosocial Rehabilitation 24, no. 3, (2020): 1750-1760. Howe, Lee Kay. “True or false or misleading: [A]dequate judicial oversight over Part 3 directions under the Protection from Online Falsehoods and Manipulation Act.” Singapore Comparative Law Review, (2019): 239-247. Human Rights Watch. “Indonesia: Repeal new intelligence law.” Published October 26, 2011. https://www.hrw.org/news/2011/10/26/indonesia-repeal-new-intelligence-law Human Rights Watch. Kill the Chicken to Scare the Monkeys: Suppression of Free Expression and Assembly in Singapore. Human Rights Watch, 2017. https://www.hrw.org/sites/default/files/report_pdf/singapore1217_web.pdf Human Rights Watch. World Report 2020. Human Rights Watch, 2020. https://www.hrw.org/sites/default/files/world_report_download/hrw_world_report_2020_0.pdf Indonesia at Melbourne. “Indonesians better get ready for jail, as flawed new criminal code looks set to pass.” Published September 3, 2019. https://indonesiaatmelbourne.unimelb.edu.au/indonesians-better-get-ready-for-jail-as-flawed-new-criminal-code-looks- set-to-pass/ Infocomm Media Development Authority. “Fact sheet – Online news sites to be placed on a more consistent licensing framework as traditional news platforms.” Last updated June 10, 2019. https://www.imda.gov.sg/news-and-events/Media-Room/archived/mda/ Media-Releases/2013/fact-sheet--online-news-sites-to-be-placed-on-a-more-consistent-licensing-framework-as-traditional- news-platforms Infocomm Media Development Authority. Internet Code of Practice. Infocomm Media Development Authority, 1997. https://www.imda.gov.sg/-/media/Imda/Files/Regulations-and-Licensing/Regulations/Codes-of-Practice/Codes-of-Practice- Media/PoliciesandContentGuidelinesInternetInterneCodeOfPractice.pdf International Commission of Jurists. “Singapore: Parliament must reject internet regulation bill that threatens freedom of expression.” Published April 4, 2019. https://www.icj.org/wp-content/uploads/2019/04/Singapore-fake-news-bill-News-web-story-2019-ENG.pdf Ismail, Saifulbahri. “ASEAN lawmakers urge Indonesia to reject changes to criminal code.” Channel News Asia, February 7, 2018. https://www.channelnewsasia.com/news/asia/asean-lawmakers-urge-indonesia-to-reject-changes-to-criminal-9935950 Jakarta Globe. “BIN budget cuts to jeopardise regional election security, says spy chief.” September 10, 2015. https://jakartaglobe.id/context/bin-budget-cuts-jeopardize-regional-election-security-says-spy-chief/ Jankowicz, Nina. Avoiding the Band-Aid Effect in Institutional Responses to Disinformation and Hybrid Threats. Washington D.C.: The German Marshal Fund, August 12, 2019. https://www.gmfus.org/sites/default/files/publications/pdf/Jankowicz%20-%20Bandaid%20Effect%20paper.pdf Jayakumar, Shashi, Benjamin Ang and Nur Diyanah Anwar. Disinformation and Fake News. Singapore: Palgrave MacMillan, 2021. “Joint statement regarding the Protection from Online Falsehoods and Manipulation Bill [Joint statement from arts and civil society organisations].” Accessed November 20, 2020. https://docs.google.com/document/d/1yNCUHvJBOkZG_ WbNt1W_8BAxzHKIjIBK1EJbMLu9Rr8/edit?fbclid=IwAR1Vu0C-CVkZbZlM7FiRdOcJiWLqWfSAYjXkdscE_1Ou6k5QMvDlElXh7bQ. “Journalists Call for withdrawal of Singapore’s ‘fake news’ Bill [letter to Minister of Communications and Information S Iswaran].” Published April 24, 2019. URL:https://docs.google.com/document/d/16pVee1fGx9cU6qADARrgfTtcV60-tVFpm46UGF1-j0o/edit. Kaye, David. Mandate of the Special Rapporteur on the Promotion and Protection of the Right to Freedom of Opinion and Expression. United Nations, 2019. https://www.ohchr.org/Documents/Issues/Opinion/Legislation/OL_SGP_3_2019.pdf Koh, Valerie. “Govt orders shutdown of The Real Singapore.” Today Online, May 4, 2015. https://www.todayonline.com/singapore/mda-suspends-licence-socio-political-website-real-singapore Krisetya, Beltsazar A. “COVID-19 exposes vulnerabilities in our cyberspace.” The Jakarta Post, April 1, 2020. https://www.thejakartapost.com/academia/2020/04/01/covid-19-exposes-vulnerabilities-in-our-cyberspace.html Kurohi, Rei. “Singapore GE2020: WP’s Raeesah Khan apologises for posts which allegedly promoted enmity between different groups.” The Straits Times. Updated September 1, 2020. https://www.straitstimes.com/politics/singapore-ge2020-wps-raeesah- khan-apologises-for-posts-which-allegedly-promoted-enmity

92 La Trobe University Kurnia, Novi and Astuti. “Researchers find Indonesia needs more digital literacy education.” The Conversation, September 26, 2017. https://theconversation.com/researchers-find-indonesia-needs-more-digital-literacy-education-84570 La Batu, Safrin. “Nine social media, messaging apps agree to combat fake news.” The Jakarta Post, January 31, 2018. https://www.thejakartapost.com/news/2018/01/31/nine-social-media-messaging-apps-agree-to-combat-fake-news.html La Cour, Christina. “Governments countering disinformation: The case of Italy.” Disninfo Portal. Accessed November 19, 2020. https://disinfoportal.org/governments-countering-disinformation-the-case-of-italy/ Lam, Lydia. “Judge dismisses SDP’s POFMA challenge, says statements were false in face of statistical evidence.” Channel News Asia, February 5, 2020. https://www.channelnewsasia.com/news/singapore/judge-dismisses-sdp-s-pofma-challenge-says- statements-were-false-12394932?cid=h3_referral_inarticlelinks_24082018_cna Lam, Lydia. “PM Lee sues blogger Leong Sze Hian for defamation over sharing of article.” Channel News Asia, December 7, 2018. https://www.channelnewsasia.com/news/singapore/pm-lee-sues-blogger-leong-sze-hian-defamation-libellous- article-10999710 Lamb, Kate. “Indonesian government to hold weekly ‘fake news’ briefings.” The Guardian, September 27, 2018. https://www.theguardian.com/world/2018/sep/27/indonesian-government-to-hold-weekly-fake-news-briefings Lamb, Kate. “Muslim Cyber Army: A ‘fake news’ operation designed to derail Indonesia’s leader.” The Guardian, March 13, 2018. https://www.theguardian.com/world/2018/mar/13/muslim-cyber-army-a-fake-news-operation-designed-to-bring-down- indonesias-leader. Lamb, Kate. “Outcry as Indonesia seeks to arrest renowned West Papua rights lawyer.” The Guardian, September 5, 2019. https://www.theguardian.com/world/2019/sep/05/outcry-as-indonesia-seeks-to-arrest-renowned-west-papua-rights-lawyer Lamb, Kate. “Sulawesi tsunami: Indonesia battles fake news as hoaxers spread panic.” The Guardian, October 4, 2018. https://www.theguardian.com/world/2018/oct/04/sulawesi-tsunami-indonesia-battles-fake-news-as-hoaxers-spread-panic Law No. 11 of 2008 on Information and Electronic Transactions [UU ITE] (Indonesia), 2008 [revised 2016]. Accessed November 13, 2020. https://www.hukumonline.com/pusatdata/detail/lt584a7363785c8/node/lt56b97e5c627c5/uu-no-19-tahun-2016- perubahan-atas-undang-undang-nomor-11-tahun-2008-tentang-informasi-dan-transaksi-elektronik Law No. 17 of 2011 on State Intelligence (Indonesia). Accessed November 13, 2020 https://www.icj.org/wp-content/uploads/2011/01/Indonesia-intelligence-law-draft-2011-eng.pdf Law No. 40 of 2008 on Elimination of Racial and Ethnic Discrimination (Indonesia): 2008. Accessed November 13, 2020. https:// www.hukumonline.com/pusatdata/detail/28539/node/1011/uu-no-40-tahun-2008-penghapusan-diskriminasi-ras-dan-etnis Lee, Howard and Terence Lee. “From contempt of court to fake news: public legitimisation and governance in mediated Singapore.” Media International Australia 173, no. 1 (2019): 81-92. Lee, Howard and Terence Lee. “Polarising dissent: The constructed narrative of Singapore’s new ‘fake news’ law.” The Asia Dialogue, June 10, 2019. https://theasiadialogue.com/2019/06/10/polarising-dissent-the-constructed-narrative-of-singapores-new-fake-news-law/ Lee, Pearl. “TRS co-founder Yang Kaiheng jailed 8 months for sedition.” The Straits Times, June 28, 2016. https://www.straitstimes.com/singapore/courts-crime/trs-co-founder-yang-kaiheng-jailed-8-months-for-sedition. Lee, Tse Yin. “Analysis: How effective was Indonesia’s war on fake news in its election?” BBC Monitoring, May 16, 2019. https://monitoring.bbc.co.uk/product/c200tf1l Lee, Vernon. “Fake news law is ploy by government to hold on to ‘absolute power’: Workers’ Party MP Low Thia Khiang.” Yahoo News Singapore, May 8, 2019. https://sg.news.yahoo.com/fake-news-law-is-ploy-by-government-to-hold-absolute-power-low-thia-khiang-155910029.html Library of Congress. “Laws protecting journalists from online harassment: Singapore.” Accessed November 20, 2020, https://www.loc.gov/law/help/protecting-journalists/singapore.php Lim, Adrian. “Parliament: Fake news law covers closed platforms like chat groups and social media groups, says Edwin Tong.” The Straits Times. May 7, 2019. https://www.straitstimes.com/politics/parliament-fake-news-law-covers-closed-platforms-like-chat- groups-and-social-media-groups Llewellyn, Aisyah. “Punitive measures: Indonesia’s new draft criminal code.” Lowy Institute. Published February 9, 2018. https:// www.lowyinstitute.org/the-interpreter/punitive-measures-indonesias-new-draft-criminal-code Lomas, Natasha. “TikTok joins the EU’s Code of Practice on Disinformation.” TechCrunch, June 22, 2020. https://techcrunch. com/2020/06/22/tiktok-joins-the-eus-code-of-practice-on-disinformation/ Luger, Jason. “Planetary illiberalism and the cybercity-state: in and beyond territory.” Territory, Politics, Governance 8, no. 1 (2020): 1-17. DOI: 10.1080/21622671.2019.1627906 Lukman, Enricko. “How Indonesia’s problematic internet law will impede freedom of expression.” SAFEnet. Published September 10, 2014. https://safenet.or.id/2014/09/how-indonesias-problematic-internet-law-will-impede-freedom-of-expression/ Macron, Emmanuel, Gitanas Nauseda and Arturs Krisjanis Karins. Joint Statement by the Presidents of France And Lithuania, and Prime Minister of Latvia on Protection of Democracies. Paris: Palais de l’Elysee, September 28, 2020. https://static.lsm.lv/documents/z5.pdf MAFINDO. “About.” Accessed August 10, 2020. https://www.mafindo.or.id/about/ Mahtani, Shibani. “Singapore introduced tough laws against fake news. Coronavirus has put them to the test.” Washington Post, March 16, 2020. https://www.washingtonpost.com/world/asia_pacific/exploiting-fake-news-laws-singapore-targets-tech-firms- over-coronavirus-falsehoods/2020/03/16/a49d6aa0-5f8f-11ea-ac50-18701e14e06d_story.html

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 93 Marsden, Chris, Trisha Meyer, and Ian Brown. “Platform values and democratic elections: How can the law regulate digital disinformation?” Computer Law and Security Review 36, (2020): 1-18. Mchangama, Jacob and Joelle Fiss. The Digital Berlin Wall: How Germany (Accidentally) Created a Prototype for Global Online Citizenship. Copenhagen: Justitia, November 2019. https://globalfreedomofexpression.columbia.edu/wp-content/ uploads/2019/11/Analyse_The-Digital-Berlin-Wall-How-Germany-Accidentally-Created-a-Prototype-for-Global-Online- Censorship.pdf Meese, James and Edward Hurcombe. Regulating Misinformation: Policy Brief. Melbourne: RMIT University, November 12, 2020. https://apo.org.au/sites/default/files/resource-files/2020-11/apo-nid309357.pdf Meyer, Paul. “Singapore’s first election under the fake news law.” The Diplomat, July 7, 2020. https://thediplomat.com/2020/07/singapores-first-election-under-the-fake-news-law/ Mid-Day. “Indonesian man arrested for Facebook post allegedly insulting Islam.” Published May 1, 2018 https://www.mid-day.com/articles/indonesian-man-arrested-for-facebook-post-allegedly-insulting-islam/19374167 Ministry of Communications and Information and Ministry of Law. Deliberate and Online Falsehoods: Challenges and Implications. Singapore: Parliament of Singapore, 2018. https://www.nas.gov.sg/archivesonline/government_records/Flipviewer/grid_ publish/6/6797717d-f25b-11e7-bafc-001a4a5ba61b-06012018Misc.10of2018/web/html5/index.html?launchlogo=tablet/ GovernmentRecords_brandingLogo_.png&pn=1 Ministry of Communications and Information. “The ‘Singapore States Times’ and Alex Tan’s Facebook pages declared as declared online locations under POFMA.” Published May 6, 2020. https://www.pofmaoffice.gov.sg/documents/media-releases/2020/May/pofma-pr-mci-06may2020-01.pdf Ministry of Health and Ministry of Manpower. “Joint MOM-MOH statement on the issuance of correction directions under the Protection from Online Falsehoods and Manipulation Act (POFMA) against National University of Singapore Society (NUSS), The Online Citizen Asia (TOC), CNA and New Naratif.” POFMA Office. Published July 5, 2020 https://www.pofmaoffice.gov.sg/documents/media-releases/2020/July/pofma-pr-mom-05jul2020-01.pdf Ministry of Law. Enhancements to the Protection from Online Harassment Act (‘POHA’). Singapore: Ministry of Law, April 1, 2019. https://www.mlaw.gov.sg/news/press-releases/enhancements-to-the-protection-from-harassment-act-poha National Library of Singapore. “Combating fake news.” Last updated October 29, 2020. https://sure.nlb.gov.sg/resources/audience/adults/combating-fake-news Nenadic, Iva. “Unpacking the ‘European approach’ to tackling challenges of disinformation and political manipulation.” Internet Policy Review 8, no. 4 (2019): 1-22. Neo, Ric. “The securitisation of fake news in Singapore.” International Politics 57 (2019): 724-740. DOI:10.1057/s41311-019-00198-4. Neubronner, Stephanie. “Bill to protect from online falsehoods: Refinements needed.” RSIS Commentary 85, (April 30, 2019): n.p. New Mandala. “Q&A: Sandiaga Uno on economic policy.” Published March 4, 2019. https://www.newmandala.org/qa-sandiaga-uno-on-economic-policy/ Nijeboer, Arjen. “Why the EU must close EUvsDisinfo.” Euobserver, March 27, 2018. https://euobserver.com/opinion/141458 NL Times. “Dutch Government to launch anti-fake news campaign.” December 13, 2018. https://nltimes.nl/2018/12/13/dutch-government-launch-anti-fake-news-campaign Novak, Benjamin. “Hungary moves to end rule by decree, but Orban’s powers may stay.” New York Times, June 16, 2020. https://www.nytimes.com/2020/06/16/world/europe/hungary-coronavirus-orban.html Nugroho, Agung. “Comparative study on hoax handling policies in Indonesia and Singapore.” Advances in Social Science, Education and Humanities Research 343, (2019): 358-363. Nugroho, Fiska Maulidian. “Study of fake news dissemination articles on Criminal Code regulations, Law of Information and Electronic Technology, and also Law of Terrorism Criminal Act eradication.” Paper presented at Fake News and Elections in Asia Conference, Bangkok, Thailand, July 10-12, 2019. Accessed November 13, 2020. https://asiacentre.org/wp-content/uploads/2019/09/Fake-News-Conference-Proceeding.pdf Nurita, Dewi. “Fadli Zon urge National Cyber Agency hit the brake.” Tempo, January 8, 2018. https://en.tempo.co/read/914656/fadli-zon-urge-national-cyber-agency-hit-the-brake Office of Assistant to Deputy Cabinet Secretary for State Documents & Translation. “National political stability under control: Home ministry.” Cabinet Secretariat of the Republic of Indonesia. Published September 26, 2019. https://setkab.go.id/en/national-political-stability-under-control-home-ministry/ Oktavianti, Tri Indah. “National Police detail tightened measures to combat COVID-19, support government policy.” The Jakarta Post, April 8, 2020. https://www.thejakartapost.com/news/2020/04/08/national-police-detail-tightened-measures-to-combat-covid- 19-support-government-policy.html Ong, Benjamin Joshua. “Looking beyond the vague terms in Singapore’s fake news laws.” TodayOnline, May 14, 2019. https://www.todayonline.com/commentary/looking-beyond-vague-terms-fake-news-laws Ong, Benjamin Joshua. “Symposium on POFMA: Parliamentary debates about POFMA – Hansard beyond statutory interpretation?” Research Collection School of Law (November 2019): 1-5. https://ink.library.smu.edu.sg/sol_research/2968 Pamment, James. The EU Code of Practice on Disinformation: Briefing Note for the New EU Commission. Washington: Carnegie Endowment for International Peace, March 2020. https://carnegieendowment.org/2020/03/03/eu-code-of-practice-on- disinformation-briefing-note-for-new-european-commission-pub-81187.

94 La Trobe University Pamment, James. The EU’s Role in Fighting Disinformation: Taking Back the Initiative. Washington: Carnegie Endowment for International Peace, June 2020. https://carnegieendowment.org/files/Pamment_-_Future_Threats.pdf. Parliament of Singapore. “Members of parliament.” Accessed November 13, 2020. https://www.parliament.gov.sg/about-us/structure/members-of-parliament Parliament of Singapore. “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill.” Published May 8, 2019. Vol. 94, Part 1. https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-366 Parliament of Singapore. “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill.” Published May 8, 2019. Vol. 94, Part 2. https://sprs.parl.gov.sg/search/sprs3topic?reportid=bill-367 Parliament of Singapore. Report of the Select Committee on Deliberate Online Falsehoods – Causes, Consequences and Countermeasures. Singapore: Parliament of Singapore, 2018. https://sprs.parl.gov.sg/selectcommittee/selectcommittee/download?id=1&type=subReport Parlina, Ina. “Jokowi declares fight against disseminators of fake news.” The Jakarta Post, December 30, 2016. https://www.thejakartapost.com/news/2016/12/30/jokowi-declares-fight-against-disseminators-of-fake-news.html Pausacker, Helen. “Ahok: Loved and hated governor of Jakarta.” Indonesia at Melbourne, August 4, 2015. https://indonesiaatmelbourne.unimelb.edu.au/ahok-loved-and-hated-governor-of-jakarta/ Penal Code (Singapore). Parliament of Singapore. 1972. https://sso.agc.gov.sg/Act/PC1871?ProvIds=pr499- Potkin, Fanny and Augustinus Beo Da Costa. “Fact-checkers vs. hoax peddlers: A fake news battle ahead of Indonesia’s election.” Reuters, April 11, 2019. https://www.reuters.com/article/us-indonesia-election-fakenews-insight/fact-checkers-vs-hoax- peddlers-a-fake-news-battle-ahead-of-indonesias-election-idUSKCN1RM2ZE Protection from Online Falsehoods and Manipulation Act (Singapore). Parliament of Singapore. 2019. https://sso.agc.gov.sg/Acts-Supp/18-2019/Published/20190625?DocDate=20190625 Protection from Online Harassment Act (Singapore). Parliament of Singapore. 2014. https://sso.agc.gov.sg/Act/PHA2014 Renaldi, Adi. “Facebook Removed Hundreds of Accounts for Allegedly Spreading Propaganda in Indonesia.” VICE News, October 10, 2019. https://www.vice.com/en_asia/article/ne8qbq/facebook-removed-hundreds-accounts-spreading-propaganda-indonesia Renda, Andrea. The Legal Framework to Address “Fake News”: Possible Policy Actions at the EU Level. Brussels: Policy Department for Economic, Scientific and Quality of Life Policies, June 2018. https://www.europarl.europa.eu/RegData/etudes/IDAN/2018/619013/IPOL_IDA(2018)619013_EN.pdf Reni, Dewi Savitri and Syarifah Reihana Fakhry. “RKUHP: What it means for criminal justice in Indonesia.” SSEK. Published March 3, 2020. https://www.ssek.com/blog/rkuhp-what-it-means-for-criminal-justice-in-indonesia Reporters Without Borders. “Indonesia used Covid-19 to silence criticism of government.” April 16, 2020. https://rsf.org/en/news/indonesia-used-covid-19-silence-criticism-government Repucci, Sarah. “A leaderless struggle for democracy: Democracy and pluralism are under assault.” Freedom House. Accessed November 20, 2020. https://freedomhouse.org/report/freedom-world/2020/leaderless-struggle-democracy Reuters. “Facebook takes down hundreds of Indonesian accounts linked to fake news syndicate.” February 1, 2019. https://www.reuters.com/article/us-facebook-indonesia/facebook-takes-down-hundreds-of-indonesian-accounts-linked-to- fake-news-syndicate-idUSKCN1PQ3JS Reza, Bhatara Ibnu. “Why strengthening anti-terror laws is not the answer.” Indonesia at Melbourne. Published February 1, 2016. https://indonesiaatmelbourne.unimelb.edu.au/why-strengthening-anti-terror-laws-not-the-answer/ Rosen, Jay. “The people formerly known as the audience.” In Participation and Media Production: Critical Reflections on Content Creation, edited by Nico Carpentier and Benjamin DeCleen, 163-165. Newcastle upon Tyne, UK: Cambridge Scientific Publishers Ltd, 2008. Rumata, Vience Mutiara and Ashwin Sasongko Sastrosubroto. “Net-attack 2.0: Digital post-truth and its regulatory challenges in Indonesia.” Advances in Social Science, Education and Humanities Research 165, (2018): 116-120. SAFEnet. “[Press release] SAFEnet sends Amicus Curiae for the Saiful Mahdi case.” Published February 11, 2020. https://id.safenet.or.id/2020/02/rilis-pers-safenet-kirim-amicus-curiae-untuk-kasus-saiful-mahdi/ SAFEnet. “Persoalan UU ITE dan Praktik Pelanggaran Hak Digital di Indonesia.” Published November 14, 2019. https://id.safenet.or.id/wp-content/uploads/2019/11/Persoalan-UU-ITE-dan-Pelanggaran-Hak-Digital-SAFEnet-2019.pdf SAFEnet. “The experts: The Indonesia Ministry of Communication and Informatics and the President violates human rights and state administrative law.” Published March 17, 2020. https://safenet.or.id/2020/03/experts-the-ministry-of-communication-and- information-and-the-president-violates-human-rights-and-state-administrative-law/ Samios, Zoe and Lisa Visentin. “ACMA: Tech giants’ code to handle fake news fails to meet expectations.” Sydney Morning Herald, October 26, 2020. https://www.smh.com.au/politics/federal/acma-tech-giants-code-to-handle-fake-news-fails-to-meet- expectations-20201026-p568oq.html. Saurwein, Florian and Charlotte Spencer-Smith. “Combating disinformation on social media: Multilevel governance and distributed accountability in Europe.” Digital Journalism 8, no. 6 (2020): 820-841. Schultheis, Emily. “The Czech Republic’s fake news problem.” The Atlantic, October 21, 2017. https://www.theatlantic.com/international/archive/2017/10/fake-news-in-the-czech-republic/543591/ Sedition Act (Singapore). Parliament of Singapore. 1948. https://sso.agc.gov.sg/Act/SA1948?ProvIds=pr3-. Serhan, Yasmeen. “Italy scrambles to fight misinformation ahead of its elections.” The Atlantic, February 24, 2018. https://www.theatlantic.com/international/archive/2018/02/europe-fake-news/551972/

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 95 Siberkreasi. “About Siberkreasi.” Accessed November 13, 2020. https://siberkreasi.id/tentang-siberkreasi/ Sim, Fann. “Workers’ Party opposes online falsehoods Bill, says Pritam Singh.” Channel News Asia, May 7, 2019. https://www.channelnewsasia.com/news/singapore/online-falsehoods-workers-party-opposes-bill-pritam-singh-11511450 Singapore Government. “What is the licensing framework for online news sites all about?” Published June 18, 2013. https://www.gov.sg/article/what-is-the-licensing-framework-for-online-news-sites-all-about Singapore Legal Advice. “Singapore fake news laws: Guide to POFMA (Protection from Online Falsehoods and Manipulation Act).” Published February 3, 2020. https://singaporelegaladvice.com/law-articles/singapore-fake-news-protection-online-falsehoods-manipulation/ Social Observatory for Disinformation and Media Analysis. “About us.” Accessed November 19, 2020. https://www.disinfobservatory.org/about-us/. Soeriaatmadja, Wahyudi. “Eucalyptus necklace among ‘cures’ raising alarm in Indonesia.” The Straits Times, July 10, 2020. https://www.straitstimes.com/asia/se-asia/eucalyptus-necklace-among-cures-raising-alarm-in-indonesia Soeriaatmadja, Wahyudi. “Man who uploaded controversial video of ex-Jakarta governor Ahok sentenced to jail.” The Straits Times, November 14, 2017. https://www.straitstimes.com/asia/se-asia/man-who-uploaded-controversial-ahok-video-sentenced-to-jail Sounding Board. The Sounding Board’s Unanimous Final Opinion on the So-Called Code of Practice. European Commission, September 24, 2018. https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=54456. Statista. “Leading countries based on Facebook audience size as of July 2020.” Published July 24, 2020. https://www.statista.com/statistics/268136/top-15-countries-based-on-number-of-facebook-users/ Stolton, Samuel. “EU code of practice on disinformation ‘insufficient and unsuitable,’ member states say.” EURACTIV.com, June 5, 2020. https://www.euractiv.com/section/digital/news/eu-code-of-practice-on-disinformation-insufficient-and-unsuitable- member-states-say/. Stophoax. “About Stophoax.id at a glance.” Accessed November 13, 2020. https://stophoax.id/tentang-kami/ Subramanian, Samanth. “Inside the Macedonian fake-news complex.” Wired. February 15, 2017. https://www.wired.com/2017/02/veles-macedonia-fake-news/ Sudhakar, S. “Ahmad Dhani imprisoned, Fahri Hamzah encourages revised LAW.” Suara. Published January 29, 2019. https://www.suara.com/news/2019/01/29/144219/ahmad-dhani-dipenjara-fahri-hamzah-dorong-uu-ite-direvisi Sumaktoyo, Nathaneal. “Ethnicity and Jakarta’s election.” New Mandala. Published February 13, 2017 https://www.newmandala.org/ethnicity-jakartas-election/ Swan, David. “Code to rein in spread of fake data.” The Australian, June 26, 2020. Tambini, Damian. Fake News: Public Policy Responses. London: Media Policy Project, London School of Economics and Political Science, 2017. http://eprints.lse.ac.uk/73015/1/LSE%20MPP%20Policy%20Brief%2020%20-%20Fake%20news_final.pdf Tandoc Jr., Edson. C., Wei Lim Zheng and Richard Ling. “Defining ‘Fake News’.” Digital Journalism 6, no. 2 (2018): 137-153. DOI:10.1080/21670811.2017.1360143 Tang, N. “POFMA: A song of fact and opinion.” The Online Citizen. Published February 4, 2020. https://www.onlinecitizenasia.com/2020/02/04/pofma-a-song-of-fact-and-opinion/ Tapsell, Ross. “Indonesia’s policing of hoax news increasingly politicised.” ISEAS, no. 75 (2019): 1-10. https://www.iseas.edu.sg/images/pdf/ISEAS_Perspective_2019_75.pdf Tapsell, Ross. “Platform convergence in Indonesia: Challenges and opportunities for media freedom.” Convergence 21, no. 2 (2014): 182-197. Tardaguila, Cristina. “Fact-checkers are teaching Indonesian housewives how to stop misinformation.” Poynter. Published December 12, 2019. https://www.poynter.org/ifcn/2019/fact-checkers-are-teaching-indonesian-women-how-to-stop-misinformation/ Tardaguila, Cristina. “Spain has a new government and its fact-checkers had an impact on the campaign.” Poynter. Published May 3, 2019. https://www.poynter.org/fact-checking/2019/spain-has-a-new-government-and-its-fact-checkers-had-an-impact-on-the- campaign/ Tehusijarana, Karina M. and Jessicha Valentina. “Jakarta riot: Government temporarily limits access to social media, messaging apps.” The Jakarta Post, May 22, 2019. https://www.thejakartapost.com/life/2019/05/22/jakarta-riot-government-temporarily- limits-access-to-social-media-messaging-apps.html Temby, Quinton. “Disinformation, violence, and anti-Chinese sentiment in Indonesia’s 2019 elections.” ISEAS, no. 67 (2019): 1-8. https://www.iseas.edu.sg/images/pdf/ISEAS_Perspective_2019_67.pdf Temby, Quinton. “Indonesia’s election riots offer a lesson on the perils of fake news.” South China Morning Post, September 9, 2019. https://www.scmp.com/week-asia/politics/article/3026057/indonesias-election-riots-offer-lesson-perils-fake-news Tempo. “Lawmakers push for amendment of electronic law.” September 13, 2013. ttps://en.tempo.co/read/512919/lawmakers-push-for-amendment-of-electronic-law Tham, Yuen-C. “Fake news act: Speed, cost of appeals process set out.” The Straits Times, October 2, 2019 https://www.straitstimes.com/politics/fake-news-act-speed-cost-of-appeals-process-set-out Tham, Yuen-C. “Falsehoods on coronavirus show why Pofma is necessary.” The Straits Times, February 4, 2020. https://www.straitstimes.com/politics/falsehoods-on-coronavirus-show-why-pofma-is-necessary

96 La Trobe University The Guardian. “Jakarta protests: Muslims turn out in force against Christian governor Ahok.” December 2, 2016. https://www.theguardian.com/world/2016/dec/02/jakarta-protests-muslims-against-christian-governor-ahok The Jakarta Post. “Beware what you share: Police step up war on hoaxes.” November 12, 2018. https://www.thejakartapost.com/news/2018/11/12/beware-what-you-share-police-step-war-hoaxes-fake-news.html The Jakarta Post. “Criminalisation code.” July 18, 2019. https://www.thejakartapost.com/academia/2019/07/18/criminalization-code.html The Jakarta Post. “Indonesia wages war against coronavirus misinformation as hoaxes spread online.” February 3, 2020. https://www. thejakartapost.com/news/2020/02/03/indonesia-wages-war-against-coronavirus-misinformation-as-hoaxes-spread-online.html The Jakarta Post. “ITE law aims to protect, manage electronic information, transactions.” October 27, 2016. https://www.thejakartapost.com/news/2016/10/27/ite-law-aims-to-protect-manage-electronic-information-transactions.html The Jakarta Post. “New regulation to have intelligence agency work directly under President Jokowi.” July 20, 2020. https://www. thejakartapost.com/news/2020/07/20/new-regulation-to-have-intelligence-agency-work-directly-under-president-jokowi.html The Jakarta Post. “Prabowo campaign spokesman arrested on subversion charges.” May 21, 2019. https://www.thejakartapost. com/news/2019/05/21/prabowo-campaign-spokesman-arrested-on-subversioncharges.html The Jakarta Post. “Revised ITE Law could hamper freedom of expression: Researcher.” October 31, 2016. https://www.thejakartapost.com/news/2016/10/31/revised-ite-law-could-hamper-freedom-of-expression-researcher.html The Jakarta Post. “Right contained in ITE Law to delete negative news threatens press freedom: Activist.” October 28, 2016. https://www.thejakartapost.com/news/2016/10/28/right-contained-in-ite-law-to-delete-negative-news-threatens-press- freedom-activist.html The Online Citizen. “Five statements of facts by Dr Thum Ping Tjin spared from POFMA correction direction – Why?’ Published May 14, 2020. https://www.onlinecitizenasia.com/2020/05/14/five-statements-of-facts-by-dr-thum-ping-tjin-spared-from-pofma- correction-direction-why/ The Straits Times. “Fake news law invoked for the first time over Facebook post.” November 26, 2019. https://www.straitstimes.com/politics/fake-news-law-invoked-for-the-first-time-over-facebook-post The Straits Times. “Indonesia court sentences administrator of ‘fake news factory’ Saracen to jail.” January 12, 2018. https://www.straitstimes.com/asia/se-asia/indonesia-court-sentences-administrator-of-fake-news-factory-saracen-to-jail Thomson Reuters Foundation. “Rights group, journalists press Indonesian govt to lift internet curb in restive Papua.” Published August 23, 2019. https://news.trust.org//item/20190823055418-5iy2g/ Thorington, Kanzanira. “Europe’s elections: The fight against disinformation.” Council on Foreign Relations. Published May 23, 2019. https://www.cfr.org/blog/europes-elections-fight-against-disinformation. Timur, Fitri Bintang. “Unequal power makes women brunt of cyber law.” SAFEnet. Published March 11, 2017. https://safenet.or.id/2017/03/unequal-power-makes-women-brunt-of-cyber-law/ Tribunnews.com. “Publik Lebih Takut Berbicara Politik di Era Presiden Jokowi? Berikut Hasil Survei LSI.” November 5, 2019. https://www.tribunnews.com/nasional/2019/11/05/publik-lebih-takut-berbicara-politik-di-era-presiden-jokowi-berikut-hasil- survei-lsi?page=all. Trihermanto, Febi and Helen Brown. “Indonesia focused on cyber resilience, open to foreign cooperation.” Australia Indonesia Centre. Published November 12, 2019. https://digital.australiaindonesiacentre.org/news/insights/indonesia-focused-on-cyber- resilience-open-to-foreign-cooperation/ Turner, Sarah and Pamela Allen. “Chinese Indonesians in a rapidly changing nation: Pressures of ethnicity and identity.” Asia Pacific Viewpoint 48, no. 1 (2007): 112-127. Ungku, Fathin. “Facebook, rights groups hit out at Singapore’s fake news bill.” Reuters, April 1, 2019. https://www.reuters.com/ article/us-singapore-politics-fakenews/facebook-rights-groups-hit-out-at-singapores-fake-news-bill-idUSKCN1RD279 Ungku, Fathin. “Singapore invites rights group to fake news hearing as dispute grows.” Reuters, March 28, 2018. https://www.reuters.com/article/us-singapore-politics-malaysia-scandal/singapore-pm-files-defamation-suit-against-blogger- who-shared-article-on-facebook-idUSKBN1O414L. Ungku, Fathin. “Singapore PM files defamation suit against blogger who shared article on Facebook.” Reuters, December 5, 2018. https://www.reuters.com/article/us-singapore-politics-malaysia-scandal/singapore-pm-files-defamation-suit-against-blogger- who-shared-article-on-facebook-idUSKBN1O414L UTS Centre for Media Transition. Discussion Paper on an Australian Voluntary Code of Practice for Disinformation. DIGI, October 16, 2020. https://digi.org.au/wp-content/uploads/2020/10/Discussion-Paper-Final.pdf Valdani, Vicari and Associates. Study for the Assessment of the Implementation of the Code of Practice on Disinformation. Brussels: European Commission, May 2020. https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=66649. Varagur, Krithika. “Activists: Indonesian counterterrorism law threatens civil liberties.” Voa News, May 30, 2018. https://www.voanews.com/east-asia-pacific/activists-indonesian-counterterrorism-law-threatens-civil-liberties Varagur, Krithika. “Indonesia poised to pass Asia’s first ‘right to be forgotten’ law.” Voa News, November 7, 2016. https://www.voanews.com/east-asia/indonesia-poised-pass-asias-first-right-be-forgotten-law Vaswani, Karishma. “Concern over Singapore’s anti-fake news law.” BBC News, April 4, 2019. https://www.bbc.com/news/business-47782470

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 97 Voa News. “Indonesian police intensify crackdown on fake news.” November 21, 2018. https://www.voanews.com/east-asia-pacific/indonesian-police-intensify-crackdown-fake-news Walker, Shaun and Jennifer Rankin. “Hungary passes law that will let Orban rule by decree.” The Guardian, March 31, 2020. https://www.theguardian.com/world/2020/mar/30/hungary-jail-for-coronavirus-misinformation-viktor-orban Warburton, Eve. “Indonesia’s pro-democracy protests cut across deep political cleavages.” New Mandala, October 3, 2019. ttps://www.newmandala.org/indonesias-pro-democracy-protests/ Wardle, Clair and Hossein Derakhshan. Information Disorder: Toward an Interdisciplinary Framework for Research and Policy Making. Council of Europe Report, September 27, 2017. https://rm.coe.int/information-disorder-toward-an-interdisciplinary-framework-for-researc/168076277c Wardle, Claire. “The need for smarter definitions and practical, timely empirical research on information disorder.” Digital Journalism 6, no. 8 (2018): 951-963. DOI: 10.1080/21670811.2018.1502047 Wargadiredja, Arzia Tivany. “Indonesia’s answer to fake news is a threat to free speech.” VICE News, February 6, 2018. https://www.vice.com/en_asia/article/j5bq54/indonesias-answer-to-fake-news-might-be-a-threat-to-free-speech Watin-Augouard, Marc. “Law no. 2018-1202 of 22 December 2018 on combating the manipulation of information.” FIC-Observatory. Published November 1, 2019. https://observatoire-fic.com/en/law-no-2018-1202-of-22-december-2018-on-combating-the- manipulation-of-information-by-general-marc-watin-augouard-founder-of-the-fic/ Wei, Neo Rong. “A look at key changes to Protection from Harassment Act.” TodayOnline, May 7, 2019. https://www.todayonline.com/singapore/look-key-amendments-protection-harassment-bill Wilson, Cameron. “This man’s Facebook page was blocked for spreading false information about the coronavirus.” BuzzFeed News, February 19, 2020. https://www.buzzfeed.com/cameronwilson/singapore-facebook-fake-news-law-alex-tan-coronavirus Wolters, Oliver W. et al. “Indonesia: Religions.” Britannica. Last modified November 11, 2020. https://www.britannica.com/place/Indonesia/Services Wong, Tessa. “Singapore fake news law polices chats and online platforms.” BBC, May 9, 2019. https://www.bbc.com/news/world-asia-48196985 Xiang, Teo Kai. “About.” POFMA’ed. Accessed November 13, 2020. http://pofmaed.com/ Xiang, Teo Kai. “Explainer: What is POFMA?” POFMA’ed. Published July 5, 2020. http://pofmaed.com/explainer-what-is-pofma/ Xu, Terry. “TOC’s submission to Select Committee on Deliberate Online Falsehoods.” The Online Citizen. Published March 7, 2018 https://www.onlinecitizenasia.com/2018/03/07/tocs-submission-to-select-committee-on-deliberate-online-falsehoods/ Yahoo News Singapore. “Thum Ping Tjin, New Naratif to challenge POFMA notice despite compliance.” May 15, 2020. https://sg.news.yahoo.com/thum-ping-tjin-new-naratif-to-challenge-pofma-notice-despite-compliance-071938288.html Yao, Wei and Kenny Chng. “Symposium on POFMA: Reflections on thinking about the POFMA.” Singapore Public Law, (November 2019): n.p. https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=4944&context=sol_research Yong, Nicholas. “GE2020: Workers’ Party is not ‘pulling its punches’ in spite of POFMA, says Pritam Singh.” Yahoo News Singapore. July 7, 2020. https://sg.news.yahoo.com/ge-2020-workers-party-is-not-pulling-its-punches-in-spite-of-pofma- says-pritam-singh-063326754.html?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_ sig=AQAAAIrOu4fGibE2qg0_wGysQZWxGlrMs_D4ufdDdO6OzEkNrRDGDD45KSSR9ldVhmsr66-Mvrh8dfWWIKbYwHBHci5KarXg Rua1iR2bMVxuLWxI9cNhQYQxFivabsbiKXpcqH0JFTv89ncOn32AUssYPlQuGdUBNVyXjXRcmXF-A-5H Yuliawati, Lis and Bayu Nugraha. “Penyebar Hoax Dijerat UU Terorisme, Polri: Tergantung Fakta Hukum.” Viva, March 21, 2019. https://www.viva.co.id/berita/nasional/1132456-penyebar-hoax-dijerat-uu-terorisme-polri-tergantung-fakta-hukum? Zaini, Khairulanwar. “Singapore in 2019: In holding pattern.” Southeast Asian Affairs, no. 1 (2020): 294–322.

98 La Trobe University Appendix A: Tables of laws and regulations, Indonesia

Legislative and Commencement How it defines Relevant actions Stakeholders Outcomes non‑legislative date the problem impacted initiatives

Criminal Code Adopted in 1946 Article 14 refers to The Government Anyone in Indonesia The focus on (KUHP) false information, may arrest and can come under public unrest in the however there punish with lengthy scrutiny. Members KUHP means that doesn’t appear to jail terms or large of the public and it has often been be a definition of fines those that high profile political used in moments what makes it so. are suspected figures alike have of crisis, such as Article 15 states that of spreading faced charges under during natural information or news misinformation, Articles 14 and 15. disasters, mass riots that is uncertain, whether intentionally and recently the This law does not exaggerated or or not, on the COVID-19 pandemic. target online activity incomplete is grounds that it It has been used exclusively. prohibited if it causes, or threatens against criticism of causes, or has the to cause public the Government’s potential to cause, unrest. handling of public unrest. COVID-19. Defamation is There is also defined as the act evidence to suggest of “intentionally it has been used harming someone’s freely against honour or reputation political dissidents, by charging him with including opposition a certain fact.”1 political parties and West Papuan separatists.

1 “Criminal Code [KUHP] (Indonesia), 1946,” accessed November 13, 2020, Article 310(1), https://www.unodc.org/res/cld/document/idn/indonesian_penal_code_html/I.1_ Criminal_Code.pdf

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 99 Appendix A: Tables of laws and regulations, Indonesia

Legislative and Commencement How it defines Relevant actions Stakeholders Outcomes non‑legislative date the problem impacted initiatives

Revised Criminal Tabled in 2018; Like the existing The Bill outlines Given the broad As it hasn’t been Code (RKUHP) it was scheduled Criminal Code, the added powers scope of the bill, passed, there have to be voted on revised version is beyond those it would impact been no outcomes in September largely lacking an granted in KUHP. on everyone in except for mass 2019, which was explicit definition These include Indonesian society. demonstrations postponed. of fake news, the authority to It has been said in of its misinformation or prosecute criticism to be particularly more controversial disinformation. It of judges,6 and “disastrous not clauses relating to does differentiate defamation of the only for women and freedom of speech, between knowingly President or Vice- religious and gender among other things. spreading false President.7 minorities, but for all information,2 Indonesians.”8 spreading information one may suspect to be false,3 and information that is uncertain, exaggerated or incomplete4 – each carry different penalties. Blasphemy – that which discriminates or is hostile “against the religion adopted in Indonesia” – is also prohibited.5

2 Ibid., Article 262(1). 3 Ibid., Article 262(2). 4 Ibid., Article 263. 5 Ibid., Article 304. 6 “Draft Revised Criminal Code [RKUHP] (Indonesia), 2019,” accessed November 13, 2020, Article 281, https://www.hukumonline.com/pusatdata/detail/17797/rancangan- undang-undang-2019# 7 Ibid., Article 218. 8 Andreas Harsono quoted in “RKUHP Explainer: All the controversial articles in Indonesia’s criminal code overhaul,” Coconuts Jakarta, September 19, 2019, https://coconuts.co/bali/features/rkuhp-explainer-all-the-controversial-articles-in-indonesias-criminal-code-overhaul-2/

100 La Trobe University Appendix A: Tables of laws and regulations, Indonesia

Legislative and Commencement How it defines Relevant actions Stakeholders Outcomes non‑legislative date the problem impacted initiatives

Information 2008; revised in Unlike the Criminal Breaches of the law Internet users are UU ITE has had and Electronic 2016 Code, UU ITE relates carry significant obviously impacted several purported Transaction Law directly to electronic jail terms and fines by the law. Ordinary outcomes, such as (UU ITE) information. “False for perpetrators. Indonesians that limited free speech and misleading Suspected have low levels and crackdowns information” here perpetrators are of digital literacy on political dissent. relates to “consumer often arrested and have come under Others have claimed loss in electronic detained prior to scrutiny for sharing that it has resulted in transactions,”9 trial, if one ever misinformation a “revenge” dynamic, however it has been eventuates. UU ITE unwittingly.15 where it is used used much more has been labelled flagrantly against Media professionals broadly than that. the Government’s one’s enemies, and have faced charges “de facto ‘anti-fake has had an overall It prohibits under the law. news’ law” for its chilling effect on information regular use against Self-proclaimed Indonesia’s public intended to incite misinformation.13 atheists have sphere.17 hostility based upon received scrutiny ethnicity, religion The Government for purported hate and race.10 can terminate online speech against access,14 which it Also illegal is the Islam.16 has done in targeted unauthorised areas of the country Critics of the manipulation, in times of unrest. Government are creation and regular targets of the deletion of electronic law. Such as West information to Papuan separatists, make it “seem to who have had be authentic.”11 One their access to the criticism is that it internet blocked. doesn’t properly differentiate between insults and defamation.12

9 “Law No. 11 of 2008 on Information and Electronic Transactions [UU ITE] (Indonesia), 2008 [revised 2016]” HUKUM Online, accessed November 13, 2020, Article 28(1), https://www.hukumonline.com/pusatdata/detail/lt584a7363785c8/node/lt56b97e5c627c5/uu-no-19-tahun-2016-perubahan-atas-undang-undang-nomor-11-tahun-2008- tentang-informasi-dan-transaksi-elektronik 10 Ibid., Article 28(2). 11 Ibid., Article 35. 12 Usman Hamid, “Indonesia’s Information Law has threatened free speech for more than a decade. This must stop,” The Conversation, November 25, 2019, https://theconversation.com/indonesias-information-law-has-threatened-free-speech-for-more-than-a-decade-this-must-stop-127446. 13 Tapsell, Ross, “Indonesia’s policing of hoax news increasingly politicised,” ISEAS, no. 75 (2019), https://www.iseas.edu.sg/images/pdf/ISEAS_Perspective_2019_75.pdf 14 UU ITE, Article 40. 15 “Indonesia’s policing of hoax news increasingly politicised,” 3. 16 “Indonesian man arrested for Facebook post allegedly insulting Islam,” Mid-Day, May 1, 2018, https://www.mid-day.com/articles/indonesian-man-arrested-for-facebook-post- allegedly-insulting-islam/19374167. 17 “Persoalan UU ITE dan Praktik Pelanggaran Hak Digital di Indonesia,” SAFEnet, published November 14, 2019, https://id.safenet.or.id/wp-content/uploads/2019/11/Persoalan- UU-ITE-dan-Pelanggaran-Hak-Digital-SAFEnet-2019.pdf

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 101 Appendix A: Tables of laws and regulations, Indonesia

Legislative and Commencement How it defines Relevant actions Stakeholders Outcomes non‑legislative date the problem impacted initiatives

Law No. 17 of 2011 The law does not This law authorises This law relates to A recently passed 2011 on State define fake news, the State Intelligence serious criminal law ensures BIN Intelligence however its remit Agency (BIN) activity that is works directly under covers threats to to conduct judged to threaten President Widodo.20 domestic security surveillance, national security, BIN has been used such as information information such as terrorism. for its intelligence and cyber warfare.18 extraction and Nonetheless, expertise to help interception to “hoaxes” have mitigate crisis protect against commonly fallen points, most recently “activities that under this definition during COVID-19. threaten national as they may have the interests and potential to cause security.”19 public disorder. BIN has been used during elections to monitor online information. BIN reports directly to the president.

Terrorism Act 2018 Article 1.4 in the law The terrorism law While senior The Act has not defines the threat of allows the police government officials been used against violence as “speech, to “detain ‘terror have suggested misinformation. writing, picture, suspects’ for as long that perpetrators symbol or body as 21 days without of hoaxes can be language, with or charge, and for caught by the Act “if without electronic… another 200 days if there is an element form which could police need time to of threat of violence create widespread gather evidence.”23 and (they) create an fear.”21 However atmosphere of terror it does not define and widespread misinformation or fear,”24 as of yet disinformation.22 it has not been used to prosecute misinformation.

18 “Law No. 17 of 2011 on State Intelligence (Indonesia), 2011,” accessed November 13, 2020, https://www.icj.org/wp-content/uploads/2011/01/Indonesia-intelligence-law- draft-2011-eng.pdf 19 Ibid., Article 31(a). 20 “New regulation to have intelligence agency work directly under President Jokowi,” The Jakarta Post, July 20, 2020, https://www.thejakartapost.com/news/2020/07/20/new- regulation-to-have-intelligence-agency-work-directly-under-president-jokowi.html 21 Krithika Varagur, “Activists: Indonesian counterterrorism law threatens civil liberties,” Voa News, May 30, 2018, https://www.voanews.com/east-asia-pacific/activists- indonesian-counterterrorism-law-threatens-civil-liberties 22 Fiska Maulidian Nugroho, “Study of fake news dissemination articles on Criminal Code regulations, Law of Information and Electronic Technology, and also Law of Terrorism Criminal Act eradication,” (Paper presented at Fake News and Elections in Asia Conference, Bangkok, Thailand, July 10-12, 2019), 88, accessed November 13, 2020, https://asiacentre.org/wp-content/uploads/2019/09/Fake-News-Conference-Proceeding.pdf 23 “Indonesia passes controversial anti-terror laws to fight ISIL,” Al Jazeera, May 26, 2018, https://www.aljazeera.com/news/2018/05/indonesia-passes-controversial-anti-terror- laws-fight-isil-180525055635674.html 24 Dedi Prasetyo quoted in Lis Yuliawati and Bayu Nugraha, “Penyebar Hoax Dijerat UU Terorisme, Polri: Tergantung Fakta Hukum,” Viva, March 21, 2019, https://www.viva.co.id/ berita/nasional/1132456-penyebar-hoax-dijerat-uu-terorisme-polri-tergantung-fakta-hukum?

102 La Trobe University Appendix B: Tables of laws and regulations, Singapore

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Protection Legislated April (2)(a): “a statement POFMA enables Online users. Due to its brief from Online 2019/enacted of fact is a government lifespan, it is Internet Falsehoods and October 2019 statement which a ministers to issue difficult to gauge intermediaries: Manipulation reasonable person directives to its outcomes. More Act (POFMA) seeing, hearing or correct, remove, y Social media research is needed otherwise perceiving or disable access platforms on purported effects it would consider to to statements it y Direct messaging such as the “chilling” be a representation determines to be platforms4 of free speech and of fact; and false. They may y Internet service self-censorship. make this judgement providers Various actors have (b): a statement is based on whether y Websites. continued to criticise false if it is false or the statement y Technology the Government, misleading, whether is believed to companies. however many have wholly or in part, and be prejudicial to promptly received whether on its own Journalists and Singapore’s public POFMA orders. or in the context in news organisations. interest, security Time will tell if there which it appears.”1 and international Election is a decline in such relations, or to campaigners and forms of public influence elections, political parties, speech. diminish confidence as political ads Some observed in public institutions, must be monitored changes were: or “incite feelings by internet 5 of enmity, hatred intermediaries. y policed speech or ill will between during recent different groups of election persons.”2 y accountability of internet Internet intermediaries intermediaries (ie. Facebook) can be directed to to government remove or disable directives access to false y complication of statements hosted role of judiciary on their platforms.3 vis-à-vis the Government y heavy use of POFMA against oppositional voices, such as opposition politicians and independent journalists.

1 Parliament of Singapore, Protection from Online Falsehoods and Manipulation Act (Parliament of Singapore, 2019), section 2, https://sso.agc.gov.sg/Acts-Supp/18-2019/ Published/20190625?DocDate=20190625 2 Ibid., Section 7. 3 Ibid., Part 4. 4 Adrian, Lim, “Parliament: Fake news law covers closed platforms like chat groups and social media groups, says Edwin Tong,” The Straits Times, May 7, 2019, https://www.straitstimes.com/politics/parliament-fake-news-law-covers-closed-platforms-like-chat-groups-and-social-media-groups 5 Protection from Online Falsehoods and Manipulation Act, Section 48.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 103 Appendix B: Tables of laws and regulations, Singapore

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Protection Legislated March POHA6 applies the Aimed to be used Individuals and The Government from Online 2014/ enacted same definition by “the man on the entities that are says more than Harassment Act November 2014 as POFMA street.”7 found to have 3000 Magistrates (POHA) [revised January (see above box) committed online Complaints were The law’s use 2020] harassment, and made between was judged by the communicated 2014-2019, of which High Court to be falsehoods about 1700 prosecutions available only to other people or were brought and a “person.”8 The entities deemed 900 convictions 2019 amendments as harmful, can made. 500 extended the law’s be ordered to applications were for use to entities such stop, remove protection orders, as companies.9 or correct such of which 200 were Breaking POHA communications. granted. Workplace protection orders Individuals, entities harassment was a can result in arrests. and internet common reason. Police may arrest intermediaries may without a warrant, face protection and take to court, orders or civil anyone deemed lawsuits from the to act against the “victim” of said provisions of the abuse. Act.10 Penalties The State argued include up to a that journalists S$5,000 fine and/ may use it against or 6-12 months jail, harassment.11 which is doubled for repeated offences. Actors outside Singapore can be charged under POHA. ‘Vulnerable people’ and victims of intimate partner violence were given greater protections in 2019 amendments.12 People that commit ‘doxxing’

6 Parliament of Singapore, Protection from Online Harassment Act (Singapore) (Parliament of Singapore, 2014), section 2, https://sso.agc.gov.sg/Act/PHA2014 7 “Parliamentary debates: Protection from Online Falsehoods and Manipulation bill,” Parliament of Singapore, published May 8, 2019, Vol. 94 , Part 2, https://sprs.parl.gov.sg/ search/sprs3topic?reportid=bill-367 8 Human Rights Watch, Kill the Chicken to Scare the Monkeys: Suppression of Free Expression and Assembly in Singapore (Human Rights Watch, 2017), 101, https://www.hrw.org/sites/default/files/report_pdf/singapore1217_web.pdf 9 Neo Rong Wei, “A look at key changes to Protection from Harassment Act,” TodayOnline, May 7, 2019, https://www.todayonline.com/singapore/look-key-amendments- protection-harassment-bill 10 Protection from Online Harassment Act, Section 18. 11 “Laws protecting journalists from online harassment: Singapore,” Library of Congress, accessed November 20, 2020, https://www.loc.gov/law/help/protecting-journalists/ singapore.php 12 Protection from Online Harassment Act, sections 8A & 8B.

104 La Trobe University Appendix B: Tables of laws and regulations, Singapore

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Online News June 2013 As misinformation The Government Websites that post Sites that are Licencing is not explicitly can order licensees on Singapore-related required to become Scheme targeted by this to take down news at least once licensees must (under the framework, it is “prohibited”13 a week on average put up a S$50,000 Broadcasting not defined as material within 24 over a two-month performance bond Act) such. Nonetheless, hours. They can period, and have at to the Government. the publication also be directed to least 50,000 unique This status of “deliberately shut down online visits a month from brings content fabricated articles” sites, and have their Singaporean internet responsibilities upon has been punished licences revoked.14 users.15 them. under this licencing Appears to Websites, such as framework. Content impact“news”-based The Real Singapore, that is “against sites rather than have been ordered to public interest, social media sites, take down prohibited public order, national however there was material and shut harmony, and/or an instance where down their sites offends against the Government completely. There good taste or asked Google to take does not appear decency” is targeted. down a YouTube to be a significant video that it deemed number of such offensive.16 cases. Larger, popular online sites have been forced to abide by this licencing framework.

Defamation 1965 Defamation is The Government Anyone who has Self-censorship has Act (civil [revised 2014] defined in relation may sue anyone that been viewed been reported.18 defamation) to instances of it views has defamed as defaming Some state “libel and slander its reputation. Unlike others including they have taken and other malicious other countries, government. To date, precautionary falsehoods” Singapore’s this Act has seen measures before that may cause defamation voices of dissent, speaking out, such reputational, legislation “does not such as independent as dissipating their financial or other provide a qualified media and individual assets in fear of damage. privilege for criticism internet users face eventually being of government prosecution. sued.19 officials and other public figures. The Civil defamation damages awarded lawsuits have plaintiffs can be so been used by the high as to bankrupt Government and defendants.”17 successive Prime Ministers against political opponents, critical and foreign media for decades.20

13 “Prohibited material is material that is objectionable on the grounds of public interest, public morality, public order, public security, national harmony, or is otherwise prohibited by applicable Singapore laws” (Infocomm Media Development Authority, Internet Code of Practice (Singapore: IMDA, 1997), 2, https://www.imda.gov.sg/-/media/Imda/Files/ Regulations-and-Licensing/Regulations/Codes-of-Practice/Codes-of-Practice-Media/PoliciesandContentGuidelinesInternetInterneCodeOfPractice.pdf) 14 Kill the Chicken to Scare the Monkeys, 49. 15 “What is the licensing framework for online news sites all about?”, Singapore Government, published June 18, 2013, https://www.gov.sg/article/what-is-the-licensing- framework-for-online-news-sites-all-about 16 Alex Au Waipang, “Singapore bloggers wary of news site licence scheme,” Committee to Protect Journalists, published June 4, 2013, https://cpj.org/2013/06/singapores- news-site-license-plan-raises-questions/ 17 Ibid., 96. 18 Kill the Chicken to Scare the Monkeys, 8-10. 19 Ibid. 20 Ibid.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 105 Appendix B: Tables of laws and regulations, Singapore

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Penal Code 1872 “Whoever, by words The Government Political opponents. In the past, the section 499 either spoken or can charge a Government has Independent (criminal intended to be citizen, entity or allegedly used media.24 defamation) read, or by signs, foreign entity with criminal defamation or by visible defamation, the Members of the to make political representations, penalty being a fine, public. opponents makes or publishes up to two years jail ineligible to run for any imputation time, or both.22 parliament.25 concerning any The Government Used by the person, intending to can search private Government less harm, or knowing locations “without often than civil or having reason to a warrant and defamation lawsuits. believe that such seize anything they imputation will harm, A recent case deem relevant to an the reputation of involving The alleged offense.”23 such person, is said, Online Citizen except in the cases demonstrates how hereinafter excepted, it can potentially put to defame that independent media person.”21 in jail.

Sedition Act 1965 As misinformation The Government Members of the Like the use of [revised 2013] is not explicitly can charge alleged public. criminal defamation, targeted by this seditious actors sedition charges Members of the legislation, it is not with a fine of up have seen media. defined as such. to S$5,000 and up independent media However, speech or to three years jail Vocal religious and other outspoken publications with a time.27 figures have been Singaporeans “seditious tendency” charged with jailed for publishing The law enables the is that which inspires sedition for inciting purported Government to jail “hatred,” “contempt” enmity between falsehoods. people it believes or “disaffection” different religious are in breach of toward the groups.29 the Act, even if the Government or the statements had no “administration of seditious intent or justice,” or “promote effects on others.28 ill-will,” “hostility” or “discontent” between the citizens, races and classes of Singapore.26 Falsehoods have fallen under this definition in practice.

21 Parliament of Singapore, Penal Code (Singapore) (Parliament of Singapore, 1972), Section 499, https://sso.agc.gov.sg/Act/PC1871?ProvIds=pr499- 22 Ibid., Section 500. 23 Kill the Chicken to Scare the Monkeys, 9 24 “In December 2018, the executive editor Terry Xu Yuan Chen was charged with defamation, and at the time of writing is awaiting trial. If convicted, Xu will face prison time and a large monetary penalty, common punishments for such ‘’,” (Jason Luger, “Planetary illiberalism and the cybercity-state: in and beyond territory,” Territory, Politics, Governance 8, no. 1 (2020): 12. 25 Kill the Chicken to Scare the Monkeys, 16. 26 Parliament of Singapore, Sedition Act (Singapore) (Parliament of Singapore, 1948), Section 3, https://sso.agc.gov.sg/Act/SA1948?ProvIds=pr3- 27 Ibid., section 4(1). 28 Kill the Chicken to Scare the Monkeys, 44 29 Ibid., 47, 49-50.

106 La Trobe University Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Action Plan June 22, 2015 Concerned primarily a) Increase European External Established East on Strategic with Russian EU Strategic Action Service StratCom Task Communication coordinated Communication (EEAS), Member Force and the disinformation capacity States, Organisation flagship project campaigns for Security and Co- EUvsDisinfo which b) Work with operation in Europe identifies, compiles partners and (OSCE), Council and exposes development of of Europe, OPEN disinformation networks Neighbourhood, cases. As of 2019, d) Support for journalists the database holds freedom of the and media over 6,500 samples media and freedom representatives, of pro-Kremlin of expression European disinformation. f) Capacity building Regulators Group for for journalists and Audiovisual Media media actors Services (ERGA) h) Engagement with Civil Society i) Increase awareness, develop critical thinking and promote media literacy

Joint Framework April 6, 2016 Hybrid threats Action 2 Creation of EEAS, Member The establishment on Countering encompass an EU Hybrid Fusion States, social of Communication Hybrid Threats: “coercive and Cell, facilitating media specialists, Task Forces A European subversive” cooperation online platforms, for the Eastern Union Response activities that can and secure European Strategic and Southern be economic, communication Communications Neighbourhoods. militaristic, between Member Network. Monitoring technological or States. disinformation, diplomatic. Within Action 3 “Explore awareness-raising this falls “massive with Member States activities in Member disinformation ways to update States about the campaigns, using and coordinate impact of Russian social media to capacities to deliver disinformation. control the political proactive strategic narrative or to The Task Forces communications radicalise, recruit gained support and optimise use of and direct proxy through the media monitoring actors.”1 contribution of and linguistic military advice specialists.” to help counter misinformation campaigns targeted at the EU and individual Member States.

1 European Commission, Joint Framework on Countering Hybrid Threats (Brussels: European Commission, April 6, 2016), 2, https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52016JC0018&from=EN

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 107 Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Tackling Online April 26, 2018 Concerned 3.1.1 Convene a Online platforms, The Code of Disinformation: with large scale multi-stakeholder advertising industry, Practice. Possible A European disinformation, forum on advertisers, if Approach which includes disinformation, and civil society assessment not “misleading or to form the Code representatives. satisfactory. outright false of Practice on information.”2 Disinformation. 3.1.2 Support for Independent Social Observatory the creation of fact checkers, for Disinformation an independent academics, Member and Social Media European network States, civil society. Analysis (SOMA), of fact checkers. launched November Launch a secure 2018. European online Organised a platform for fact checking disinformation conference. to support fact International checker networks Fact-Checking and academics. Network created an EU specific fact checking site FactCheckEu.info

3.1.3 Focus on eIDAS Cooperation Commission reports transparency and Network, online to have mobilised online accountability platforms, the Cooperation via the identification voluntary online Network set up of suppliers of systems, academic under the electronic information. researchers, Identification and technology Authentication companies and fact Service Regulation checkers, Member (eIDAS), raising States. awareness on how eIDAS tools and services could be relied upon to tackle disinformation by increasing accountability and encouraging more responsible behaviour online

3.1.4 Utilise Horizon 2020, Currently, under the Horizon 2020 European Research ERC and Horizon work programme Council projects, 2020, there are to mobilise new traditional media, several actions technologies. Will journalists, online aimed at improving also explore extra platforms. the understanding of support to deploy online disinformation tools to combat and developing new disinformation. tools for content verification. Various projects studying automated systems and their impact on public discourse, and solutions to help traditional media better detect false narratives online.

2 European Commission, Tackling Online Disinformation: A European Approach (Brussels: European Commission, April 26, 2018), 1, https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=CELEX:52018DC0236&qid=1583456906698&from=EN

108 La Trobe University Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Tackling Online April 26, 2018 3.2 Secure and Member States, Commission started Disinformation: resilient election media and a constructive A European process. social media dialogue with Approach representatives, Member States in international orgs. 2018. Delivered a compendium on cyber security of election technology. Raising awareness and strengthening capacities to manage risks to democratic electoral process presented by cyber-attacks and disinformation.

3.3 Fostering OECD, Member Introduce rules for media literacy, States, fact- Member States including steering checkers. and video-sharing a Media Literacy platforms on media Expert Group and literacy and establish supporting all EU a co‑regulatory digital education system. efforts. Fact- Online checkers to provide disinformation educational material and fake news to to schools and be added to OECD educators. ICT familiarity questionnaire.

3.4 Supports the Member states, Implementing practice of quality journalists, media framework for journalism in the representatives. greater investment EU. Member States and funding in encouraged to innovation in consider strategies quality journalism, to address market media freedom and failures hampering plurality. the viability of quality journalism, as well as training for journalists and industry innovation. Will explore increased funding opportunities to support media freedom and pluralism, quality news media and journalism.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 109 Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Joint Framework June 13, 2018 Action 3: Strategic East StratCom Monitoring on Countering communications Task Force, NATO, disinformation, Hybrid Threats Commission, EEAS, awareness- The positive from July 2017 Eastern Partnership raising activities experiences of the to June 2018 (EaP) countries. in EaP countries East StratCom Task and Member Force need to be States about the underpinned and impact of Russian strengthened. disinformation. Task Force set up for Western Balkans. Training for staff in Eastern Partnership countries for enhancing their strategic communications capabilities and their resilience to disinformation.

Action 21: EU Military Staff, Enabling the Intelligence Member States, EU enhancement of capabilities Hybrid Fusion Cell, early warning. extended “by EEAS. contributing military advice to help counter misinformation campaigns targeted at the EU and individual Member States.”3

3 European Commission, Joint Framework on Countering Hybrid Threats from July 2017 to June 2018 (Brussels: European Union, June 13, 2018), 13, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018JC0014&from=EN

110 La Trobe University Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

EU Code of September 26, 2018 Citing a High-Level While specific Digital platforms, Successive Practice on Expert Group paper, measures of advertising bodies, assessments by Disinformation the Code defines how to combat fact checkers and ERGA, independent disinformation as disinformation are researchers. consultant VAA and "’verifiably false to be decided by the Commission or misleading signatories, the five found the Code information’ which, “Pillars” of focus has established cumulatively, include “scrutiny a framework of advertisement for dialogue (a) ‘Is created, placements to between relevant presented and reduce revenues stakeholders, disseminated for of the purveyors monitoring, and economic gain or to of disinformation,” greater transparency intentionally deceive “transparency about and accountability the public’; and political and issue- of platforms’ (b) ‘May cause public based advertising,” operations. harm,’ intended targeting inauthentic Each Pillar had as ‘threats to activity and demonstrated democratic political strengthen integrity improvement, yet and policymaking of services, and significant shortfalls processes as well as empowerment in performance of public goods such of consumers the Code and its as the protection of and the research signatories were EU citizens' health, community. noted – particularly the environment or regarding data security.’” It does not Signatories tasked sharing and the include “misleading with producing empowerment advertising, regular reports of users and reporting errors, on actions taken the research satire and parody, and a framework community. or clearly identified for regular EU partisan news and and independent commentary.” assessment is established.

Directive (EU) November 28, 2018 Article 33a Member States, civil Decrees Member 2018/1808 of Member States society. States to promote the European shall promote and media literacy. Parliament and take measures for Nearly a dozen of the Council the development of Member States media literacy skills. have taken up comprehensive media literacy initiatives.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 111 Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Action Plan December 5, 2018 Action 2 High StratCom Task Increased budget against Representative Forces, online to strengthen the Disinformation to “review the platforms, Union's capabilities mandates of international to detect, analyse the Strategic partners, G7, and expose Communications NATO, Union disinformation as Task Force for Neighbourhoods well as to ensure Western Balkans a coordinated and and South to enable speedy response them to address – ie. Rapid Alert disinformation in System. RAS these regions.” however has received mixed Action 3 “Establish reviews since it has a Rapid Alert not been triggered System for (as of June 2020), addressing and few Member disinformation States have campaigns.” shared information Action 5 through it.5 “Strengthen Fostered closer strategic coordination among communications EU institutions and in the Union's national authorities. neighbourhood.”4

4 European Commission, Action Plan against Disinformation (Brussels: European Commission, December 5, 2018), 8, https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52018JC0036&qid=1583456906698&from=EN 5 James Pamment, The EU’s Role in Fighting Disinformation: Taking Back the Initiative (Washington: Carnegie Endowment for International Peace, June 2020), 9, https://carnegieendowment.org/files/Pamment_-_Future_Threats.pdf

112 La Trobe University Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Action Plan December 5, 2018 Action 6 Close Signatories to the Facebook, Google, against monitoring of the Code Twitter were found Disinformation implementation of to have: Code of Practice, y improved their and “push for scrutiny of ad rapid and effective placements, compliance.” reducing After 12-month malicious click- assessment baiting practices Commission may and advertising propose actions revenues for including regulation. those posting disinformation; y focused on manipulative behaviour to make content more visible through coordinated operations, as well as the abusive use of bots and fake accounts. However the platforms did not make sufficient progress in increasing the transparency of websites hosting ads, partly due to the lack of engagement from the advertising industry. Questions remain about the effectiveness of the transparency measures taken by all signatories. All platforms should also ensure an active and working cooperation with fact checkers in all Member States and empower users to better detect disinformation.

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 113 Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Action Plan December 5, 2018 Action 7 “Targeted European EU institutions against campaigns for the Commission, High launched a number Disinformation public and training Representative, of initiatives to raise for media and public the public, public awareness opinion shapers.” independent media of disinformation Continue support and journalists, across EU for independent academics, Member States, media and quality Member States, including seminars, journalism and fact checkers, conferences and research into researchers, civil media briefings. disinformation. society. Provided proactive Action 8 “Support and multilingual the creation of communication teams of multi- to millions of EU disciplinary citizens through independent social media fact-checkers and and through researchers.” communication campaigns. Action 9 Media Literacy Week, cross The Commission border cooperation worked together with and the “launch of the EEAS and the practical tools for European Parliament the promotion of to produce and media literacy for distribute myth the public.” busting and awareness-raising materials, drawing on both academic and institutional expertise. European Media Literacy Week with more than 320 events in the EU18. The International Fact-Checking Network created a European branch of independent fact checkers covering 14 Member States and launched a website in 11 EU languages.

114 La Trobe University Appendix C: Tables of laws and regulations, European Commission

Legislative and Commencement How it defines the Relevant actions Stakeholders Outcomes non-legislative date problem impacted initiatives

Communication April 30, 2019 Part II. 4 Acting EU27 Leaders, fact These from the together to fight checkers, the public, recommendations Commission disinformation digital platforms and were aimed to guide Europe in May social networks. the actions and Outlined emphasis 2019: Preparing interactions of EU on scaling up for a More leaders prior to their independent fact United, Stronger May 2019 meeting in checking efforts, and More . media literacy and Democratic artificial intelligence Union in an capacities, Increasingly collaboration with Uncertain World online platforms, and media pluralism and freedom, all aimed at improving the public’s access to reliable information.

Tackling June 10, 2020 Acknowledged the Sought to develop Member States, In September 2020, COVID-19 need to expand more robust signatories, the the Commission Disinformation definition of the measures to public, civil society released Code – Getting the problem from counter COVID-19 and researchers signatories’ first Facts Right disinformation, misinformation. baseline reports where intent is key, This includes: on these COVID- to misinformation, 19-related asks. y Improving since the COVID-19 The reports coordination “infodemic” demonstrated between EU demonstrated the signatories had institutions, harm that such increased access including better pandemic-related to and visibility use of the Rapid misinformation of authoritative Alert System, can cause. information.6 civil society and Notably, however, journalists the application of y Working with these measures was the WHO to not uniform across better identify the EU. and counter “misleading and harmful narratives” y Improving public education The most notable measure was to require Code signatories to provide monthly reports on their efforts to counter the infodemic.

6 “First baseline reports – Fighting COVID-19 disinformation monitoring program,” European Commission, published September 10, 2020, https://ec.europa.eu/digital-single- market/en/news/first-baseline-reports-fighting-covid-19-disinformation-monitoring-programme

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 115 Appendix D: Respondent profile and interview schedule

This appendix lists the 14 interviewees who have contributed to the report along with an indicative schedule of the types of questions asked of them. Under the La Trobe University ethics application approved for this research, some interviewees’ identities were not disclosed due to potential adverse consequences. However, some interviewees felt it preferable and, in some instances, safer in their circumstances, to be named. Thus, the table below details the profiles of the respondents interviewed for this project, with more detailed information provided when explicit permission was granted. The interview schedule also varies according to individual circumstances. The Interview Schedule is a semi-structured guide to provide an understanding of the types of questions asked of respondents. Interviews were analysed with the assistance of qualitative data analysis software, NVivo, to locate rich insights in the data, consolidate the main themes that form the findings chapters (chapters five to seven), and to produce defensible findings backed by evidence.

SINGAPORE

Interviewee Name/ Number Position Interview Date

Pingtjin Thum (PJ Thum) Historian, activist and Managing Director of the New Naratif 19 August 2020

Kirsten Han Freelance Journalist 25 August 2020

1 Social Media/Digital Platform Expert 7 August 2020

3 Academic 18 August 2020

4 Social Media/Digital Platform Expert 7 August 2020

11 Academic 20 August 2020

INDONESIA

Interviewee Name/ Number Position Interview Date

Wahyu Dhyatmika Editor-in-Chief of Tempo magazine 24 August 2020

Septiaji Eko Nugroho Chairman of MAFINDO (civil society group tackling hoaxes) 26 August 2020

Andreas Harsono Human Rights activist and Researcher at Amnesty 20 August 2020 International Indonesia Damar Juniarto SAFEnet, Southeast Asia Freedom of Expression, 31 August 2020 Executive Director and activist 7 Academic 19 August 2020

9 Social Media/Digital Platform Expert 8 September 2020

OTHER

Interviewee Name/ Number Position Interview Date

13 EU Social Media/Digital Platform Expert 16 September 2020

14 The Australian Communications Media Authority (ACMA) 18 November 2020

116 La Trobe University Appendix D: Respondent profile and interview schedule Semi-structured interview schedule 1. What is your name? 2. What organisation do you work for? 3. What is your title? 4. What is your career background? 5. How do you define fake news: misinformation and disinformation? a. Any other terms that should be considered? 6. Is misinformation a problem in the region? If so, how widespread is the problem on platforms and social media across the region (APAC)? a. Where is it most challenging in terms of places and spaces online? 7. How have you and/or your organisation dealt with misinformation? 8. Which measures do you think have been most effective? a. What is the evidence for that? 9. How has your government responded to the threat of misinformation on digital platforms and social media? a. What laws, if any, have been passed? b. What penalties are associated with breaches of these laws? 10. What powers do the laws provide to the Government? Do they have limitations? 11. Are there public concerns/controversial aspects of these policy/legal changes? If so, what are they? 12. Who are the people/organisations most concerned about these changes? 13. Through the experience of your work, what do you consider are the main public consequences of ‘fake news’ laws? 14. Can you provide examples of any adverse impacts the fake news laws have had on: a) a person, b) organisations c) others? 15. What is the current state of the debate around misinformation on social media platforms in your country and across APAC? 16. If these laws could be improved, what changes would you like to see implemented? 17. In your experience, which country has the most useful measures to deal with fake news online? Why? 18. Is there anyone in this field who you think I should talk to? 19. Is there anything else you would like to add?

Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific 117 latrobe.edu.au 514270_1220