Connective Action and Affective Language: Computational Text Analysis of Facebook Comments on Social Movements in South Korea

Shin Haeng Lee, Tae Yun Lim


The present study addresses the question of how the 2016 “candlelight vigil” in South Korea developed through digitally networked activism in opposition to then president Park. Specifically, we examine whether the anti-Park group and its countering pro-Park group manifested different semantic structures in their comments on Facebook pages. To do so, we collected 144,639 and 29,133 comments on the anti-Park and pro-Park pages, respectively, over 5 months (October 2016–March 2017) and constructed semantic networks from the comments made during the heady days of activism. Our analysis of word co-occurrence and community structure for each network demonstrates that the anti-Park activism engaged in random and polysemic patterns of word use. In contrast, the organization-based pro-Park activism exhibited scale-free and semantically coherent organizing. This finding explicates the linguistic mechanism by which the connective logic of engagement affords affective expressions that generate loosely formatted but collaboratively coordinated identities to intensify feelings of engagement.


connective action, affective language, Facebook activism, semantic network, community detection, computational text analysis

Full Text: