“Just Asking”: Figleaves and the Mainstreaming of Mis/Disinformation Online

• In Event: Negative Side-Effects of Social Media: Extremism & Hate Speech

Abstract

Before 2016 there was a consensus in political psychology that national politicians in the US could no longer succeed if they were too explicitly racist (Mendelberg 2001). This led politicians to use devices of coded and deniable speech (e.g., dog whistles) to play to voters’ racial resentments without being too obvious. More recently, political success has been achieved by openly and brazenly appealing to racist and extremist sentiments – indeed, this has proven to be a surprisingly effective political strategy. How have overt expressions of racism and falsity come to be accepted in the mainstream? What mechanisms are at play when “saying the unsayable” helps rather than hinders political careers? There has been a great deal of attention in recent years to the dramatic rise in, and spread of blatant racism and obvious falsehood online. However, little empirical attention has been paid to the specific linguistic mechanisms that may facilitate this. This paper explores the rhetoric that allows speakers to “say the quiet part loud” without the perception of doing so, facilitating the mainstreaming of racism and falsehood in online communication.

Saul (2017, 2019, 2021) posited a novel linguistic device theorized to aid in the dissemination and uptake of messages of racial hatred and conspiracism online. After its namesake, a figleaf is a bit of speech that provides cover for that which might otherwise be easily recognized as blatant racism or obvious falsehood. Common figleaves include: “just asking questions,” “people are saying,” “just joking,” and many more. Figleaves raise doubt that a norm has been violated, functioning to reassure uncertain audiences and allow them to deny that deeply held norms against racism and untruthfulness are being violated, thereby facilitating their acceptance and spread to the mainstream. Figleaves pose a particular danger to political discourse because they have the potential to change norms for what counts as acceptable speech. Without the presence of a figleaf, racist or conspiracist messaging would more readily be recognized as unacceptable. Once the audience accepts the figleaf, however, the message no longer seems beyond the bounds of acceptability. Figleaves are especially relevant for uncertain audiences, who may be more readily persuaded to help a racist candidate to succeed where they otherwise might not have, or help a fringe conspiracy theory move to the mainstream. By providing cover for otherwise outrageous speech acts, figleaves can shift the boundaries of acceptable discourse and pave the way for more inflammatory speech.

This is the first empirical study of this novel linguistic device. Drawing on the philosophy of language, digital media analytics, and political communication, we employ a range of data analytic approaches—including NLP and corpus linguistics—to operationalize and validate a measure of figleaves. Using a combination of top-down and data-driven approaches, we identify known, and discover unknown figleaves in emergent racist and conspiracist narratives in a large corpus of news articles and commentary, social media data (Twitter and Facebook), and subreddit posts, drawn from a wide range of mainstream and extremist online communities. In addition, we analyze related content, including actors, sentiment, likes, shares, and support for mis/disinformation. Our methodological approach combines qualitative and quantitative content analytic approaches, state-of-the-art data science, and an innovative corpus-based approach to critical discourse analysis (Baker 2006), which uniquely brings together corpus linguistics and qualitative analysis to uncover discursive meaning across large corpora by analyzing keywords from big data in context.

We assess two hypotheses relating to the theorized role of figleaves in the mainstreaming of racist and conspiracist messages online: 1) figleaves will be more prevalent on mainstream than extremist venues, and 2) messages with (versus without) figleaves will be shared more widely on mainstream venues. This allows us to assess the role of figleaves in the transmission of such messages amongst those who are not members of racist or extremist communities and to identify which figleaves are most effective in spreading harmful mis/disinformation. We can thus better understand how individual claims—e.g., by a politician—are processed in these different environments, and the justifications people create in their positive assessments of, and decisions to share those claims on various venues. In so doing, we will come to a much greater understanding of what is potentially a very important method for the spread of racism and conspiracism. We hope this study will lay the groundwork for the exploration and development of effective techniques to identify and ultimately combat the workings of figleaves online.

Authors

• Lori Young, University of Virginia

• Raymond Drainville

• Jennifer Saul