Broadcasting Messages via Telegram: The Case of the 2020 Protests in Belarus

• In Event: Linking Media Content and Consumption to Political Behavior (Virtual Session)

Abstract

We know little about social media communication strategies of either pro-government or opposition-supporting political actors (politicians, activists, protesters, or media in general) in authoritarian countries. Many existing studies about social media use for political communications are Western/English language-oriented and concentrate on democracies. Facebook and Twitter are perhaps the most researched social media concerning political participation, social movements, and protests. Other less common in the West, yet popular in other countries social media and messaging apps such as VK or Telegram, receive less attention despite their role in the organization and coordination of democratic protests in Belarus and Hong Kong.
This paper addresses the gap by focusing on social media use by political actors in Belarus, one of the post-Soviet Russian-speaking countries. The 2020-2021 protests and demonstrations in Belarus provide us with an opportunity to analyze the patterns of media communications by pro-government, neutral, and opposition actors under authoritarianism. The environment of political turmoil allows us to look at the use of new media during mobilization when the opposition movement was advocating for democratic transition. The public Telegram channels, online threads that allow broadcasting of messages to unlimited audiences, became an alternative source of information for the opposition movement. The hybrid nature of the channels that combines one-to-many communications with the speed and low cost of the internet, allowed the messages to be spread to large audiences during the mass mobilization. How does the political affiliation of Telegram channels (pro-opposition, neutral, or pro-government) influence the content of the channels and the reach of their messages? Which messages are the most popular? What are the stylistics of messages? What are the dynamics of leadership on Telegram? Did anti-government channels send mobilizing messages? And what topics are discussed on the channels and what are the dynamics of topic prevalence over time?
To address the research questions, I employ text-as-data techniques to a corpus of Telegram public channel messages. I collected more than 100,000 messages from the 16 most popular, publicly available, and politically oriented Telegram channels in Belarus using Telegram API and the Telethon Python library. The channels are categorized either as opposition, neutral, or pro-government oriented. The corpus includes messages from nine opposition, four neutral, and three pro-government Telegram channels. To analyze the use of Telegram during the protests, I analyze the messages posted between May 1 and December 31 of 2020 using descriptive statistics, Structural Topic Modeling, and word embedding techniques. Even though the largest and most frequent protests took place after the election day (August 9, 2020), the decision to include several previous months in the analysis is based on the timing of political campaigns and the announcement of the date for the election. Overall, the chosen timeframe covers election campaign events, the day of the presidential election, and the months following the election when the largest and the most frequent demonstrations and protests occurred.
The preliminary results show that Telegram, as a broadcast media, played an important role in the organization and coordination of the protests. Pro-opposition channels and some neutral channels were the most active and reached large audiences on the day of the presidential election and after, challenging the government monopoly in the media sector. Pro-opposition channels Nexta Live and Nexta TV sent messages with detailed plans of action to organize and coordinate protests and demonstrations. Some of those messages were among the most popular messages from 16 channels analyzed in this study. Pro-government channels broadcasted their messages only to small audiences. Currently, I work on models of word embeddings that would help to address whether different groups of Telegram channels (pro-government, neutral, and opposition) discuss differently. The study shows that Telegram, an overlooked messaging app, can be an important data source for research on mass mobilization and social movements in authoritarian countries.

Author

• Daria Kuznetsova