JOURNAL OF GENOCIDE RESEARCH 11 Figure 1. Circa 2012. Note: My thanks to Kyaw Zeyar Win for the images in Figure 1 and Figure 2. The effects of this in-group policing were felt beyond Rakhine State. Images of victims being publicly humiliated as “traitors,” such as those in Figures 1 and 2, were shared on Facebook and could thus be encountered by people across Myanmar. Posts on Facebook also meant that people from elsewhere in Myanmar and their families could still be pub­ licly targeted as “traitors” for their activities in Rakhine State, even after leaving the region. In March 2014, for example, the campaigns against humanitarian organizations escalated into crowd attacks on their offices in Sittwe, forcing the emergency evacuation of staff.59 In the aftermath, photographs likely acquired from hard drives and cameras looted during the attacks began appearing on Facebook. Images of relief activities for Rohingya were posted along with captions furthering narratives about the Rohingya, while images of individual humanitarian workers were posted along with captions condemning them as traitorous members of the in-group (Figure 3). In the image at left, a roster listing the bio­ graphical details of children residing in a camp for displaced Rohingya is captioned, “One could mistake the refugee camp for a maternity ward,” in an obvious reference to antiRohingya narratives about demographic replacement. The image at right features a Burmese man standing in the office of a humanitarian organization, captioned, “An internal traitor who is closely watching the movements of Rakhine people and giving information to NGOs.” The post at left features a negative characterization of the out-group, the Rohingya. The post at right, however, was about not the out-group but a humanitarian worker from the in-group and thus would not fit within the understandings of “hate speech” or anti-Rohingya narratives that structured much research on Facebook in Myanmar. Even so, the aggregate of this research suggests just how widespread and impactful ingroup policing has been. One indicator: in the decade following the 2012 violence in Rakhine State, analyses of Facebook and Myanmar were published nearly every year and, almost without exception, each includes examples that could also illustrate ingroup policing, though this interpretation is often not present.60 In October 2020, for 59 60 “Mobs Attack Offices of UN, Aid Groups in Myanmar’s Rakhine State,” Radio Free Asia, 27 March 2014. Sometimes these reports included content that clearly served the function of in-group policing, while interpreting the examples otherwise. For example, pointing to the way posts characterized or harassed Rohingya and other Muslims, Phandeeyar, Hate Speech Lexicon Research (Yangon, June 2020), 16; Alan Davis, “How Social Media Spurred Myanmar’s Latest Violence” (Institute for War and Peace Reporting, 12 September 2017); Megha Rajagopalan, “Internet Trolls Are Using Facebook To Target Myanmar’s Muslims,” BuzzFeed News, 18 March 2017. Or as demonstrating a “lack of media