Copyright: DFRLab, Atlantic Council, April 17, 2019. With Melissa Hall.
Discussion of white nationalism online was overshadowed by discussion of Candace Owens, a far-right commentator invited by the Republican minority to testify as a part of the U.S. House of Representative’s Judiciary Committee hearing on “Hate Crimes and the Rise of White Nationalism,” according to the DFRLab’s analysis of Twitter sentiment regarding the hearing.
The April 9, 2019, hearing marked the most significant congressional examination of violent white nationalism and identitarianism in the post-9/11 era. As the online reaction to the hearing pivoted from white nationalism in general to Owens, the Twitter conversation was further derailed by discussion of U.S. Rep. Ted Lieu (D-CA), who has a sizeable Twitter following. During the hearing, Lieu devoted his allotted five minutes of questioning to attacking Owens’ credibility.
The DFRLab’s investigation is grounded in a baseline analysis of Twitter traffic related to white nationalism and hate crimes; an analysis of conversation patterns related to Owens and Lieu; and an analysis of baseline discussion of Google and Facebook.
Instead of focusing on the historic nature of the testimony or potential policy solutions, Twitter conversation focused predominantly on a clash between two personalities with a significant online presence. Such framing trivialized the serious issues under discussion and exercised an outsize impact on media coverage of the hearing.
Consequently, questions of white nationalist terrorism and the role of social-media companies received little attention. On the other hand, the feud between Owens and Lieu received altogether too much.
Mentions of Hearing Topics on Twitter
The DFRLab analysis began by establishing a baseline of Twitter conversation regarding white nationalism-related terrorism. Via a Sysomos scan, 660,000 relevant tweets were collected, occurring over the two days prior to the hearing. The selected keywords were “white nationalism,” “white nationalist,” “white supremacy,” “hate crime,” and “hate crimes.”
The morning of the hearing, April 9, showed a significant rise in the quantity of tweets referencing the terms pertaining to white supremacy and hate crimes, peaking around the time the hearing concluded.
For reference, the quantity of tweets pertaining to white nationalism or hate crimes over the previous month, using the same keywords, saw only two other spikes: the first, a dramatic spike on March 15, was the day of the white nationalist terrorist attack in New Zealand that resulted in the murder of 50 Muslim worshippers in a Christchurch mosque; the second, a lesser spike on March 26, occurred the day that charges were droppedagainst Jussie Smollet, a television actor who allegedly faked an attack by white supremacists.
Mentions of Key Figures on Twitter
The DFRLab examined the activity on Twitter pertaining to two individuals, who participated in the hearing: Candace Owens, a conservative activist and communications director for Turning Point USA, a far-right advocacy organization; and U.S. Representative Ted Lieu, a Democrat representing California’s 33rd District. Both Owens and Lieu command large Twitter audiences with 1.23 million and 992,000 followers, respectively.
U.S. media devoted significant coverage to Owens’s and Lieu’s role in the hearing, particularly after Lieu used his allotted time to play a recording in which Owens appeared to defend the genocidal regime of Adolf Hitler. Owens, in turn, used her response time granted by multiple Republicans on the U.S. Judiciary Committee to criticize Lieu furiously. Within 24 hours, the recording of Owens’s response became the most-watched C-SPAN Twitter video of a House hearing, accumulating 4.47 million views.
To measure the public conversation about Owens and Lieu, the DFRLab performed a search in which mentions of “candace owens,” “candaceowens,” “realcandaceo,” and “owens” were used to identify Twitter trends related to Owens and the terms “ted lieu,” “tedlieu,” and “lieu” were used to identify trends related to Lieu.
Similar to the scan of tweets with topics pertaining to white nationalism-related terrorism, the number of tweets with terms pertaining to Owens and Lieu peaked during the time of the hearing.
More notably, the quantity of tweets pertaining to Owens and Lieu peaked above the quantity of tweets mentioning white nationalism-related terrorism, with mentions of Owens being significantly higher than Lieu or the topic keywords.
Furthermore, tweets that referenced either Owens or Lieu and which made no reference to keywords regarding white nationalism-related terrorism surpassed the number of tweets that directly referenced the topic of the hearing.
Specifically, tweets referencing Owens peaked on April 9 at approximately 1.8 times the number of tweets referencing the topic keywords. Tweets referencing Lieu peaked at approximately the same number of tweets that referenced the topic keywords.
Mentions of Tech Companies on Twitter
The hearing also included corporate representatives from both Google and Facebook. Several legislators questioned these witnesses regarding the use of digital fingerprinting techniques, content moderation processes, transparency guidelines, and encryption to better monitor the spread of hate crimes and white nationalism-related terrorism.
The DFRLab compared the quantity of tweets referencing Google (using the keyword “google”) and Facebook (using “facebook”) with the traffic pertaining to Owens, Lieu, and white nationalism-related terrorism in order to measure the public’s focus on the role of social-media companies with respect to the hearing. Because Google and Facebook are so frequently mentioned on social media on a normal basis, the DFRLab evaluated the proportional change in traffic, rather than a direct sampling of tweets.
For each two-hour block over four days prior to the hearing (April 4–7), the DFRLab counted the number of tweets mentioning the terms of interest (“Owens,” “Lieu,” “Google,” “Facebook,” and “White Nationalism” et al.), allowing researchers to account for fluctuations in tweeting habits over the day, such as fewer tweets during sleep hours and more during work hours. April 8, the day before the hearing, was not included in the baseline counts to account for any public discussion or hype pertaining to the leadup of the event. These counts were then averaged over the four days to compute a baseline of activity for each two-hour block of the day.
The DFRLab then graphed the number of mentions of the terms of interest during each two-hour block on April 9 divided by the average number of mentions of the terms of interest for that two-hour block during April 4–7, in order evaluate the proportional change in tweet quantities pertaining to each subject.
This method demonstrated that, compared to normal Twitter activity, on the day of the hearing there was an increase in tweets mentioning Owens or Lieu (which increased 20–60x), an order of magnitude higher than the increase in tweets mentioning Google or Facebook (which increased 1–3x). It also demonstrated that, while there was an increase in tweets mentioning white nationalism-related terrorism, it was appreciably less than the increase in those tweets mentioning Owens or Lieu.
The DFRLab’s analysis of the Twitter activity regarding the topics and individuals featured in the U.S. House Judiciary Committee’s April 9 hearing highlight how witnesses and legislators with highly engaged social-media profiles can turn the public’s attention away from the topical focus of congressional testimony.
Despite the historic nature of the April 9 hearing, Twitter discussion of white nationalism-related terrorism was swiftly overshadowed by discussion of Owens and Lieu. In time, this story of narrative hijack became the predominant focus of U.S. media coverage regarding the hearing. (“Candace Owen’s presence turned a serious inquiry into a farce,” Washington Post; “Candace Owens fumes at Rep. Ted Lieu’s use of her Hitler comments,” USA Today; “There Were Also Serious People at That Hearing on White Nationalism,” Rolling Stone.)
This finding carries significant implications for future of U.S. congressional hearings and testimony. Since the first regular television broadcasts of U.S. congressional proceedings in 1979, U.S. legislators have used such publicity to capture media coverage and to focus public attention on under-examined policy issues. But only within the last two years has this broadcasting power of U.S. legislators clashed with the nature of modern internet celebrity, in which a congressional witness may command more online engagement than any (or all) of the legislators who seek to question her.
In the case of April 9’s white nationalism terrorism-related hearing, Owens — invited by the House’s Republican minority — commanded enough online attention to dilute the historic nature of the event. Lieu, in choosing to attack Owens directly, further derailed the narrative from the unprecedented nature of the proceeding. What began as a serious discussion of ideology, terrorism, and content moderation policy swiftly devolved into a clash between Twitter personalities.
The result, thanks to Owens’s invitation and participation, was a highly engaged one-on-one interaction with little topical resonance that consequently stymied policy discussion of a critical national security topic.