Downplaying Popularity of Protests: Visual Narratives, Authoritarian Control, and the 2019 Hong Kong Social Movement

• In Event: 21066 - Advances in Data and Methods for Collective Behavior/Social Movements Research

Abstract

Protests often pose a threat to authoritarian regimes. In response, authoritarian regimes employ various methods to counter the threat of collective action and sustain their rule. Digital technologies provide authoritarian states with new tools for the dissemination of propaganda and containment of opposition. This study examines the role of visual media, specifically how images and videos are used to shape public opinions on protests in authoritarian contexts. While current studies have primarily utilized text to study protest framing in authoritarian contexts, few have employed an images-as-data approach. By building an original dataset of over 1 million video frames covering the 2019 Hong Kong social movement, this research employs visual machine learning methods to contrast the visual narratives presented by the Chinese government with those from democratic societies. It finds that compared to news outlets in democratic countries, Chinese state media are more likely to portray the social movement as lacking mass support. An extension shows that state media also visually emphasize the larger size of pro-government protests to reinforce its narrative. A survey experiment will be conducted to assess the effectiveness of the visual narrative. The findings shed light on mechanisms through which authoritarian regimes manage dissent, shape public opinion, and maintain control.

Author

• Yuhan Hu, University of Oxford