GOING AFTER THE FAMILY 13 importance of commitment and social capital as a priori determinants of activists’ strategic adaptations, rather than assuming that activists’ responses are reactive to regime tactics in a post hoc fashion (e.g., McAdam, 1983). As the first study (to the best of our knowledge) to comparatively and systematically theorize the effects of proxy punishment on diasporas, this paper makes an important contribution to the burgeoning field of transnational repression studies. At the same time, we acknowledge that far more research is needed on how proxy punishment varies by emigrant group, including their particular social movement affiliations and places of residence. For instance, research shows that host-countries with strong economic and geopolitical ties to the home-country are more likely to cooperate with home-countries to enact transnational repression against common enemies, such as members of the Muslim Brotherhood (Darwich, 2017), and facilitate the unlawful extradition of exiles and human rights activists on behalf of foreign allies (Dukalskis, 2021; Schenkkan & Linzer, 2021). Although our analysis does not find variation in the application of proxy punishment by respondents’ countries-of-residence, researchers would benefit from larger, representative sampling to test the influence of the host-country context and its policies on the application of proxy punishment. Host-countries’ condemnations of human rights abuses and related sanctions may pressure regimes to relax their reliance on proxy punishment, and regime actors may be more reticent to apply this sanction when needing to secure economic trade deals and arms sales from particular host-countries. That said, regime officials facing pushback may instead choose to increase proxy punishment as a means to demonstrate their power, steadfastness, and sovereignty in the face of condemnation from great powers such as the United States and the United Kingdom. How bilateral state alliances, regional security agreements, and specific foreign policies impact a group’s vulnerability warrants further attention. Future scholarship can therefore take up these comparisons, both within regions and across them, to better understand whether or not proxy punishment can be mitigated by host-country interventions or bilateral relations. Lastly, while the four country cases examined here demonstrate a common repertoire of proxy punishment, authoritarian regimes are constantly adding new techniques to their repressive toolkits. For instance, the governments of China and Saudi Arabia have recently forced diaspora activists’ close relatives to travel abroad with security agents in order to convince them face-to-face to stop their activism (or return home for prosecution).3 Chinese agents have also called diaspora members via Skype from their parents’ home as a means of intimidation (Gilbert, 2020). Further research is needed to track authoritarian learning in these contexts, and to theorize whether diaspora activists require additional forms of support to contest the pervasive problem of globalized collective punishment. We predict that without a multifaceted counter-response by states, institutions, and organizations with the capacity to speak out, this problem is only likely to worsen in the near future. ACKNOWLEDGEMENTS The authors are deeply grateful to Maria Koinova for encouraging the formation of this paper and to Gerasimos Tsourapas for his invaluable comments on an earlier draft. Research conducted by Dana M. Moss was made possible by funding from the National Science Foundation’s Doctoral Dissertation Improvement Grant 2014–15 (#1433642) and the University of California, Irvine’s Kugelman Citizen Peacebuilding Research Fellowship, the Center for Global Peace and Conflict Studies’ Research Award, the Center for the Study of Democracy, and the Department of Sociology’s Summer Fellowship. Marcus Michaelsen received funding from the European Union’s Horizon 2020 Research and Innovation Program under the Marie Skłodowska-Curie Grant Agreement No. 845988 (DIGIACT) for the writing of this paper. His research was also made possible by the Information Controls Fellowship of the Open Technology Fund. Funding for Gillian Kennedy was granted by The Leverhulme Trust Early Career Fellowship (No. ECF-2017075). All research was conducted in accordance with the authors’ respective institutional permissions and/or Internal Review Board protocols for the use of human subjects in research. CONFLICT OF INTEREST The authors declare no conflict of interest.