Disembodied Deliberation: Algorithmic Curation and Censorship in Virtual Publics

• In Event: Representation and Electoral Systems & Normative Theory Virtual Poster Session
  In Virtual Poster Session: Normative Theory Virtual Poster Session

Abstract

I theorize on how social media and content aggregating platforms such as Facebook, YouTube, and Reddit, employ algorithmic filtering, curating, and censorship in order to moderate speech. I argue certain methods of algorithmic content filtering expose and reify the normative positions of those responsible for employing those algorithms. By analyzing modes of algorithmic censorship in specific contexts, I shed light on the difficulty it presents to pluralistic public deliberation. Although employing algorithmic content moderators removes some of the burden of work for human moderators, in so doing it also removes the option and ability for the public engaged on that platform to do that deliberative work. I argue the use of algorithmic censorship elides an important part of deliberation by automating and thus disembodying the critical work of deciding what is permissible speech and what is not. Just as others note that public discourse needs to be free of structural coercion – I argue it should also be free of algorithmic coercion. Adding to the interdisciplinary literature on algorithmic justice, such as the work of Safiya Noble, I present both a critical analysis and normative theory of public discourse.

Author

• Janet Marie Lawler, University of Virginia