“hate speech” and “counter speech” and, on the other, speech that might not but which was still salient for these groups. Section 2 of this paper describes our research process, including disruptions by COVID-19 that prompted us to focus on case studies of speech as it unfolds on Facebook. Section 3 then discusses findings related to the structure of interactions on Facebook and how these may influence counter-speech on the platform. Section 4 presents some vocabularies of counterspeech that the research encountered. The paper then concludes with an Annex of examples drawn from the research material, in order to support further research and concrete reflection on the analysis discussed in sections two and three. The findings in this paper should be taken as a starting point rather than definitive conclusions. To highlight this fact our analysis is organized in three levels, presenting: 1) observations about the research material; 2) the multiple, sometimes conflicting, plausible interpretations that could explain these observations; and 3) their possible implications for the activities of practitioners and researchers. While the distinctions between these three levels of analysis can blur at times, we hope this presentation is useful for the reader. In order to support further research and encourage concrete reflection on the findings, the paper also concludes with a series of annexes of examples drawn from the research materials that illustrate the observations discussed in Sections 3 and 4. 2. Research process In developing this study, we sought to avoid some problems with orienting research towards the category “hate speech” by instead focusing on the experiences of people who are often targeted by such speech. This could encompass many social categories, which are often intersecting. To give the research a manageable scope we decided to focus on people targeted for their status as members of religious, ethnic, or LGBTQ communities. Our initial research plan involved two phases of research. First: conducting one-on-one interviews and focus group discussions, with no more than three to four people, in Yangon and three other cities, as a means to understand a broader set of perspectives on: 1) the nature of speech that people from these communities find concerning or objectionable, including aspects that may not normally be included within definitions of “hate speech;” and 2) how they manage the fact that such speech is common in general and what they do when they encounter it directly in their everyday lives. Second: gathering texts in which members of marginalized communities utilized counter-speech in some way, such as Burmese-language media coverage of events and social media posts and comment threads. 5