A Trope-Based Computational Approach to Participatory Propaganda Analysis

Recent academic research reveals that strategic information operations often co-opt authentic communities to produce <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">participatory propaganda</i>—the manufacturing, curation, and amplification of stories that advance partisan narratives. In this paper, we push the field to move beyond disinformation and rhetorical techniques to a more productive narrative-based approach to propaganda analysis, and propose the first systematic computational method for the study of participatory propaganda. We propose a theory-driven computational framework centered on the concept of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">propaganda trope</i>. Tropes are recurring scenes or stories that effectively anchor and communicate persuasive narratives. We argue that tropes provide a blueprint for impactful propaganda, and offer a valuable analytical lens for understanding how participatory propaganda is created and sustained. We illustrate this argument through an empirical case study examining state media and a state-aligned partisan community on Twitter.