2 M. SCHISSLER common sense is not unfounded. Meta operated with a de facto monopoly in Myanmar during the years immediately before 2016–17, the period of a military “clearance oper­ ation” against Rohingya Muslims that a UN Fact-Finding Mission later concluded war­ ranted prosecution for genocide.2 Changes in Myanmar’s telecommunications infrastructure only five years earlier brought many people online for the first time, the vast majority of whom used Facebook – and not any other platform.3 “For many in Myanmar, Facebook is Google, LinkedIn, Tinder, Tumblr, and Reddit, all in one,” wrote one important analysis.4 With this monopoly as a background, dozens of articles raising concerns about “hate speech” and other forms of content about the Rohingya on Face­ book have been published by scholars, advocates, and major media outlets. The single most influential source linking such content to violence is the UN Fact-Finding Mission following the 2016–17 attacks, which concluded that Facebook played a “significant,”5 even “determining,”6 role in the genocide. A report that Meta commissioned to assess its impacts on human rights in Myanmar, meanwhile, concluded that “Facebook has become a means for those seeking to spread hate and cause harm, and posts have been linked to offline violence.”7 Commensurate with its status in public common sense, the Myanmar case has become an influential template for explaining others, past, present, and future, as well as advocat­ ing for solutions to problems raised by technologies existing and anticipated, including artificial intelligence (AI).8 Yet this template is also extremely narrow: with rare exception, analyses have focused only on content that negatively characterizes the victim group, whether through hate speech, misinformation, or other efforts to construct the Rohingya as foreign, dangerous, and conspiring to conquer Myanmar.9 Such speech should be 2 3 4 5 6 7 8 9 Independent International Fact-Finding Mission on Myanmar, Report of the Detailed Findings of the Independent Inter­ national Fact-Finding Mission on Myanmar (Geneva, 17 September 2018), 178, 241–2, 358–64. Hereafter, “IIFFMM.” During the early the 2010s, the cost of a mobile phone SIM card reduced from nearly USD 1,000 to just over 1 dollar in early 2014. Within less than a year, nearly 60 per cent of the population owned a mobile phone, rising to more than 80 per cent in 2016. Sources cite various figures, from 35 per cent to 85 per cent of phone owners actively using Facebook, often also noting that there were more Myanmar Facebook accounts than people. While messaging appli­ cations such as Viber and WhatsApp were also popular, platforms such as Twitter or Instagram were rarely used; only in 2018–19 did another platform, TikTok, start to become prominent. Yan Naung Oak and Lisa Brooten, “The Tea Shop Meets the 8 O’clock News: Facebook, Convergence and Online Public Spaces,” in Myanmar Media in Transition: Legacies, Challenges and Change, ed. Lisa Brooten, Jane McElhone, and Gayathry Venkiteswaran (Singapore: ISEASYusof Ishak Institute, 2019), 325–65; Thant Sin Oo, Ye Min Thant, and Melyn McKay, Exploring Digital & Mobile Cultures in Myanmar 2019 (Yangon: Phandeeyar, May 2019); Nyi Nyi Kyaw, “Facebooking in Myanmar: From Hate Speech to Fake News to Partisan Political Communication,” ISEAS: Perspective 36 (9 May 2019); Ayesha Zainudeen et al., Mobile Phones, Internet, Information and Knowledge: Myanmar 2016 (Colombo: LIRNE Asia, August 2017). Thant Sin Oo, Ye Min Thant, and McKay, Exploring Digital & Mobile Cultures in Myanmar 2019, 5. IIFFMM, 14. Tom Miles, “U.N. Investigators Cite Facebook Role in Myanmar Crisis,” Reuters, 13 March 2018. Business for Social Responsibility (BSR), Human Rights Impact Assessment: Facebook in Myanmar (October 2018), 24. For example, Paul Gowder, The Networked Leviathan: For Democratic Platforms (Cambridge: Cambridge University Press, 2023); Caroline Crystal, “Facebook, Telegram, and the Ongoing Struggle Against Online Hate Speech” (Carnegie Endowment for International Peace, 7 September 2023); Mark Latonero and Aaina Agarwal, “Human Rights Impact Assessments for AI: Learning from Facebook’s Failure in Myanmar,” Discussion Paper (Carr Center for Human Rights Policy, Harvard Kennedy School, 19 March 2021); Daniel Byman and Aditi Joshi, “Preventing the Next Social-Media Genocide,” Survival 62, no. 6 (2020): 125–52. Scholars have also questioned categories such as “hate speech” and used alternatives, such as Benesch’s “dangerous speech” framework. See for example, Nickey Diamond and Ken MacLean, “Dangerous Speech Cloaked in Saffron Robes: Race, Religion, and Anti-Muslim Violence in Myanmar,” in The Routledge Handbook of Religion, Mass Atrocity, and Genocide, ed. Stephen Smith and Sara Brown (New York: Routledge, 2021), 205–16; Christina Fink, “Dangerous Speech, Anti-Muslim Violence, and Facebook in Myanmar,” Journal of International Affairs 71, no. 1.5 (2018): 43–52. Prasse-Freeman, meanwhile, identified another genre of speech, “hate bait,” which he demonstrated could effec­ tively associate the Rohingya with hatred and danger, such as through posting that, while not “hate speech,”