Class 23. Racism as a National Security Threat, Part II
Revisiting the active measures playbook, with a digital twist.
If you are just joining The Freedom Academy, welcome! As a reminder, this post is a standalone “lesson” and you do not need to be caught up to follow along. I’ll reference any previous posts that offer relevant background, and you can always visit the syllabus and catch up at your own leisure. All class posts have an audio recording at the bottom, if you prefer to multitask while listening to me lecture!
No single group of Americans was targeted by IRA information operatives more than African-Americans. By far, race and related issues were the preferred target of the information warfare campaign designed to divide the country in 2016. —SSCI Report on Russian Active Measures Campaigns and Interference in the 2016 Election, Vol. 2
In Class 15, I looked at how race was historically exploited by the Soviets during the Cold War. As someone who commented extensively during the Mueller investigation into Russia’s election interference, one of the things that shocked and surprised me the most as the investigation unfolded was the extent to which Russia returned to this tried and true theme in an effort to disenfranchise Black voters in 2016.
In reviewing some of the relevant sources for this post, I thought of Special Counsel Jack Smith’s charges against Trump for January 6, which includes a charge for Conspiracy Against Rights for his attempt to prevent thousands of American votes from being counted by falsely alleging that they were fraudulent (mainly where there was a majority of Black voters). It made me wonder whether Mueller could have included a similar charge against the Internet Research Agency and Kremlin-affiliated individuals whom he indicted in 2018, since the bulk of Russia’s microtargeting efforts were similarly an attempt to interfere with Black voters’ civil rights. (My speculation is purely legal and academic, since, as I discussed previously in Class 20, none of these defendants were ever going to face actual consequences and the charges were ultimately dropped anyway.)
Had Mueller shed more of a spotlight on this facet of Russia’s 2016 election interference, I think it might have gotten more media coverage than it did. (To Mueller’s credit, he did briefly mention it in his 400+ page report that most people didn’t read.) Understanding the specific goals of election disinformation targeting Black voters is important as we barrel into yet another presidential election, since the specific approach differed in important ways from Russia’s disinformation efforts elsewhere.
“Digital Blackface”
It’s commonly understood that in 2016, Russia targeted both the left and right sides of the political spectrum. This is true — most of the narratives on the right involved boosting Trump (and emerged later in the election cycle), and those on the left involved denigrating Hillary Clinton. However, in examining the behavior of Russian trolls in 2016, researchers found that disinformation on either side of the political spectrum was asymmetrical. Specifically, this ideological asymmetry suggests, according to authors of a fascinating study in 2020 that built on this research, “a disproportionate tendency among conservative individuals to encounter, redistribute, and/or believe disinformation.”
These authors tested whether any other asymmetries based on identity existed in 2016, as well. To do this, they desegregated race from ideology — in other words, instead of just “right trolls” and “left trolls,” they created a new category of “black trolls”: Accounts which presented as Black on Facebook and Twitter, a.k.a., “digital blackface.” Their findings were illuminating.
When they did this, they found that digital blackface accounts had significantly higher levels of engagement on social media than either “right” or “left” (non-digital blackface) trolls. In addition to being about three times more likely to tweet than left trolls, these accounts also far exceeded the other troll accounts in the response received from their targeted Black audiences:
The authors conclude that “despite there being substantially more conservative-presenting accounts and tweets than liberal- or Black-presenting ones, when considered on a per-tweet basis, Black-presenting IRA accounts attracted more retweets, likes, and replies than any other identity category.”
This conclusion echoed the findings from research cited by the Senate Intelligence Committee, which included a number of fascinating graphics, like this one from New Knowledge, showing the “media mirage” of digital blackface accounts surrounding Black voters:
Similarly, this graphic from the Computational Propaganda Research Project by the University of Oxford and Graphika shows the relative size of digital blackface accounts relative to other Russian troll accounts on social media (am I the only one surprised and intrigued by the connection between the black identity and “pro-gun politics” and “conservative politics and culture” accounts?):
Importantly, New Knowledge found that “accounts that appeared to be targeting and engaging with Black audiences focused on an entirely different conversation, almost entirely absent any mention of ‘Trump’ or ‘Clinton’” — suggesting that the main purpose was to foster a sense of racial solidarity, which then provided a fertile ground for seeding political narratives.
Depress, Deflect, (Re)Direct
So, what’s new about this information and why do we care? Well, I think it’s important because the specific goals in targeting Black voters differed from those of the general disinformation operation during the election in that they were focused on disenfranchisement. Specifically, Russia figured out that rather than trying to nudge Americans to vote for Trump (either by boosting him or denigrating Clinton directly), it could be more effective to simply discourage Clinton voters from voting at all, or voting in ways in which their votes wouldn’t count.
In this regard, targeting Black voters was the most efficient strategy: Blacks vote overwhelmingly Democratic. In the twelve elections from 1972 to 2016, Democrats averaged 87% of the Black vote, while Republicans got only 10%. In 2020, that disparity increased to 92%-8% in favor of Biden. Using Black voters as a proxy for Democratic votes makes sense in terms of getting more bang for the buck, as it were.
But targeting Black voters also made sense for a more sinister reason: This demographic has history of marginalization and disenfranchisement which can be capitalized upon. It was this thread that Russia exploited most effectively, using three main vectors to influence the Black vote: depress, deflect, and (re)direct.
1. Depress
Many of the Black-targeted accounts engaged in direct voter suppression. In some cases, this was by encouraging voters to stay home because their vote wouldn’t count. For example, the popular IRA account, @Chris_Tegner, tweeted, “Heads Up: If you voted for Bernie in the Primaries, the Election Board will NOT let you vote for Hillary on Nov 8th.” Others suggested that Clinton was undeserving of the Black vote. As detailed by the New Knowledge study, one IRA-backed account, @afrokingdom, posted the following:
Students from St. Augustine’s University and Shaw University, two historically Black colleges, expressed their frustration with Clinton’s tendency to lie. Almost every student doesn’t believe that Hillary is the best candidate, but said that they have to ‘settle’ for Clinton! But I say that this is the BIG mistake! Black people don’t have to vote for Hillary because she is liar! Black people are smart enough to understand that Hillary doesn’t deserve our votes! DON’T VOTE! #williamsandkalvin#awordoftruth #PanAfricanism #BlackNationalism #BlackEmpowerment #AfricanEmpowerment #AfricanAndProud #BlackAndProud #BlackPride #BlackPower #BlackLivesMatter #Amerikkka #UnapologeticallyBlack #UnapologeticallyAfrican #BlackInAmerica #BlackIsBeautiful #JusticeOrElse #ProBlack #dontvote #boycottelection #election2016
Another narrative used to dissuade Black voters was to encourage withholding their vote as a form of protest, as suggested in these memes:
As the Computational Propaganda Research Project observed,
Messaging to African Americans sought to divert their political energy away from established political institutions by preying on anger with structural inequalities faced by African Americans. These campaigns pushed a message that the best way to advance the cause of the African American community was to boycott the election and focus on other issues instead.
2. Deflect
If voter suppression didn’t work, the next best option was to ensure that the vote didn’t end up benefiting Clinton. Russia understood that Black voters were unlikely to be successfully nudged towards Donald Trump, given their strong identification with the Democratic Party. The next best option, then, was to encourage them to cast their vote for a third party “liberal” candidate. Hmmm….who might that be?
Russian accounts dismissed the suggestion that voting for Stein would benefit Republicans; the IRA account @woke_blacks tweeted, “The excuse that a lost Black vote for Hillary is a Trump win is [bullshit]. It could be late, but y’all might want to support Jill Stein instead.” An NBC analysis estimated that IRA accounts tweeted the name “Jill Stein” over 1,000 times close to the election.
Stein, of course, had been a Kremlin darling for a decade prior to the election, pushing its anti-Ukraine talking points. How convenient.
3. (Re)Direct
Finally, if you have a Black voter who is committed to voting, and voting for Hillary no matter what, what do you do? You send their vote into the ether. If you want to understand how Republican tactics that make it harder for Blacks to vote provides fodder for Russian election disinformation, look no further than the IRA narratives which encouraged Black voters to “avoid the lines” — and to “text your vote” or “tweet your vote” instead:
Generating voter confusion about how to cast their vote was, perhaps, the closest parallel to homegrown voter suppression tactics which have existed in the U.S. for decades.
In fact, Russia’s attempt to depress, deflect, and redirect Black votes mirrored the Trump campaign’s own strategy in 2016, which aimed to influence Black voters in an operation called “deterrence.” Using data collected by Cambridge Analytica — which was headed by Steve Bannon — the Trump campaign targeted Black voters with “dark ads” on social media, demonstrating a striking similarity with Russian active measures. (Coincidence? You decide.)
So, did it work? There’s no doubt that Black voter turnout was significantly lower in 2016 than in prior elections: More than a third of the voters who voted for Obama in 2012 but stayed home in 2016 — about 1.6 million people — were Black, and election post-mortems suggested that this lower turnout cost Clinton the election.
Still, it’s impossible to say how much of this decline in turnout can be attributed to Russia, which is of course the point. As I discussed in Class 17, the “changing hearts and minds” approach to election interference is imperfect, but also offers more plausible deniability. Especially given Republican voter suppression tactics and the Trump’s micro targeted campaign strategy, it’s hard to know what played the biggest role.
Trump, of course, didn’t care who was responsible, just that he won. He offered a rare display of gratitude to Black voters, stating, “They didn’t come out to vote for Hillary. They didn’t come out. And that was big – so thank you to the African American community.”
Audio here:
Discussion Questions:
What aspects of social media make “digital blackface” especially effective? How does this approach compare with its tactics in the analog age/during the Cold War?
Consider Russia’s targeting of Black voters through the lens of reflexive control (see Class 12). What implicit assumptions or “anchor beliefs” did the IRA employ to get its targets to reach their predetermined decisions?
Do you think the ‘depress, deflect, (re)direct” strategy could be effectively redeployed in 2024, or was it a one-off relative success due to the polarizing nature of the candidates in 2016?
Great piece, Asha. Thanks for the fake pics, so I am informed, but they are disgusting
1. Consider Russia’s targeting of Black voters through the lens of reflexive control (see Class 12). What implicit assumptions or “anchor beliefs” did the IRA employ to get its targets to reach their predetermined decisions?
Because reflexive control has to do with using the assumptions and preconceived notions of the audience to sway the decision-making processes of those so manipulated, the first assumption with black voter suppression must be taken into account as beyond an assumption: It’s that “you are black,” which is factual if a voter is, in fact, African American.
Because the black identity is strongly rooted and constantly reinforced by media outlets in the modern context, the very idea of trolls posing as black people and appealing to the moral value system of the black voting minority is especially intimidating in a world without borders due to the World Wide Web. Even if Donald Trump has chosen to attempt to “wall out” Mexicans crossing the border, Russians can certainly gain access to American domain by way of social media like Facebook. And, so much American media makes its way to Russian “cyber-soil” – including a lot of music that blacks may identify with – that troll farms in foreign nations can certainly piece together narratives that appeal to pro-black people who are already dissatisfied with the American status quo due to racial inequality and socioeconomic discrimination inherent within a top-down system in the United States.
Black pride, while very justified by standards associated with the overcoming of steep obstacles, is the very object used by Russian hackers to influence African Americans to vote alternatively to their typical vote, or not to vote at all. Because most blacks have at least some degree of their identities vested in black pride, when hackers misidentify themselves convincingly as black people pushing a message associated with blacks having been betrayed by those who better represent their needs as a holistic population, some degree of outcome for the foreign actors can be expected. Part of the assumption that is used to manipulate here has to do with false trust being established that the troll is, actually, a black person lobbying his or her sentiment on behalf of his or her own race.
Then, the next assumption that is deployed to complete the manipulation is that if you are black, and you cast your vote for someone who only one-half has you in mind as someone he or she will fight for, then you have not only allowed yourself to be betrayed, but you have betrayed yourself (and you lack sufficient “pride” in condoning insufficient half-measures on behalf of your own kind).
The pride black people demonstrate and ascribe to doesn’t permit this if one is proud to be black. However, because the changing of the mind comes from a foreign troll, it turns out that those victimized black voters have instead been preyed upon by high-tech, transnational crime (as have been those politicians with black voters close to their hearts who are cheated out of election victories that would have benefited the black population as a whole).
The assumption is made here that use of this type of reflexive control over blacks was much more difficult to accomplish for foreign actors when Barack Obama was running for president. Perhaps a study should investigate what levels of reflexive control were used by trolls to manipulate women into voting against Hillary Clinton or not voting at all in 2016 (and instead electing apparent misogyny), Clinton having been a potential, second consecutive landmark Commander in Chief as a “first ever” (and long overdue).
-Matthew Mullaney Doherty, November 10, 2023