From Post Racial to Most Racial? Opposition to Black Lives Matter on Twitter

• In Event: Social Networks and Political Mobilization in the Trump Era

Abstract

Color-blind ideology —which explains “contemporary racial inequality as the outcome of nonracial dynamics” (Bonilla-Silva, 2018, p.2) — pervades the American political landscape in large part because whites across the political spectrum embrace it. This paper focuses on opposition to the Black Lives Matter (#BLM) movement on Twitter as a case in point. We argue that pre-election opposition to #BLM on Twitter (circa 2014-15) was concentrated among left-leaning whites. These liberal/progressive whites tended to deemphasize ongoing racial inequality and frequently used “color-blind” hashtags such as #AllLivesMatter to signal a “loyal opposition” kind of pushback to #BLM.

We’re interested in the specific effect of the 2016 election on this phenomenon. Donald Trump’s candidacy and subsequent win marked a resurgence of old-fashioned racial animus. We hypothesize that it also marked a shift in the concentration of opposition to #BLM on Twitter from left-leaning whites and toward right-leaning whites. We posit that right-leaning whites also tend to deemphasize ongoing racial inequality in their opposition to #BLM, but—compared to their liberal white counterparts—their tweets and retweets more often mention racially charged words like thugs/terrorists/riots (instead of protests). These words, similar to the phrases “welfare handouts” and “get tough on crime,” serve as racial codes that reassert racist tropes of black violence without having to refer to race explicitly at all. We hypothesize that most post-election opposition to #BLM is of this ilk.

This brings us to Ian Haney López, who argues that modern racial pandering is like a dog-whistle in that it operates on two levels: “inaudible and easily denied in one range yet stimulating strong reactions in another” (Haney López, 2014, p.3). In our case, on one level, we have racial codes and dog whistles that are tweeted and retweeted by right-leaning whites who ostensibly reject the idea that race matters and would tell us they don’t have a racist bone in their body. For these people, the fig leaf of “cloaked language hides—even from themselves— the racial character of the overture” (Ibid., p. 5). Meanwhile, on the other level, we have right-leaning whites who clearly perceive a message of racial resentment and react positively to it.

Yet we do not expect to find many racists who identify as such. “Nowadays, except for members of white supremacist organizations, few whites in the U.S. claim to be racist” (Bonilla-Silva, 2018, p.1) and “even members of these organizations now claim that they are not racist, simply pro-white” (Bonilla-Silva, 2018, ch. 1, fn. 1). As such, we are currently working to identify the “pro-whites” in our dataset and to explore their hashtag use. Our analysis relies on a matched panel of 1.8 million Twitter handles to eligible voter data (with attached demographic data), allowing attribution of the aggregate demographics of users of particular hashtags; evaluation of the connection of hashtags via shared users; and robust evaluation of linguistic signatures associated with the use of particular hashtags. Our full dataset goes back to 2017 and a subset goes back to the very first tweet.

To date, nothing has been published on #BLM in relation to hashtags that emerged in the midst or aftermath of the 2016 election, such as #MAGA or #StandForTheFlag. In sum, we posit that liberal whites tended to rely on ostensibly “color-blind” opposition hashtags such as #AllLivesMatter before the 2016 election. During and after the election, however, the locus of opposition to #BLM on Twitter shifted to conservative whites who tend to rely on dog-whistle racial codes and hashtags that are associated with President Trump.

Authors

• Kim M. Williams, Portland State University

• David Lazer, Northeastern University