During this summer’s historic Black Lives Matter protests across the United States, social media platforms have been rife with disinformation. That’s not a coincidence. Since 2016, Russian disinformation campaigns have focused particularly on race and social issues related to African Americans, exploiting the fact that race remains a highly volatile area in U.S. politics.
Here’s what you need to know about how Russia may be trying to interfere with the 2020 presidential election.
Here’s the background
In 2018, the Senate Select Committee on Intelligence concluded that Russian information operatives were mainly targeting African Americans. This past April, a CNN investigation discovered Russian troll farms in Ghana and Nigeria that employed African nationals to post content emphasizing U.S. racial divisions. American policymakers were startled by the fact that Russians had outsourced information operations to West Africa. House Intelligence Committee Chairman Adam B. Schiff (D-Calif.) said that “the potential use of continental cutouts” to obscure Russia’s involvement “represents new and inventive ways to cover their tracks and evade detection.”
Here’s how we did our research
While the use of regional intermediaries may be a shift from the previous strategy, our research found that the messaging itself has remained steady. Building on the Senate 2018 report and the recent news, we conducted an exploratory data analysis of purportedly divisive tweets sent from Ghana and Nigeria between June 2017 and March 2020 that had been removed by Twitter, and that are believed to be associated with Kremlin-backed sources. The data set, which was released by Twitter in March, contains 39,964 tweets belonging to 71 accounts that masqueraded as being in the United States. These West African troll accounts pretended to be tweeting from all around the United States, from California to New Jersey. Some accounts reported that they were located abroad in such places as Accra or Geneva but still tried to influence American audiences.
To analyze these, we began by using the Quanteda Lexicoder Sentiment Dictionary to gain a general impression of the balance between positive and negative tweets aimed at influencing U.S. racial politics.
We cross-checked those initial results using a lexicon and rule-based sentiment analysis tool known as VADER, ranking each tweet according to the intensity of its emotion for extra precision. Finally, we identified top hashtags and explored the contexts behind the most-used keywords and “collocations,” or words commonly used together.
Our quantitative text analysis was complemented by a qualitative content analysis of the images and videos that had been included. In total, 6,000 images and 3,713 videos were tweeted. We selected content from the five most active accounts, which tweeted almost 40 percent of the visuals in the data set, for closer analysis. Finally, we identified dominant topics and compared them to the findings of our textual analysis.
Encouraging distrust in black communities
Of all the hashtags we found, the most commonly used was #blacklivesmatter, with 2,109 mentions, followed by #racism (1,381), #policebrutality (1,159), and #blackexcellence (805). We see a similar pattern for word pairs in a tweet. The most used word pairs typically had racial justice themes: “African American,” “police officer,” “lives matter,” “black people” and “human rights.”
Initially, there was a near-equal distribution between positive and negative words when we analyzed the entire data set. However, once we estimated the sentiment of each individual tweet, we found a significantly higher proportion of negative tweets (59.9 percent) than positive (22.4 percent) and neutral (17.7 percent) tweets. The three most common words in the tweets coded as having a negative sentiment were “racism,” “black” and “police,” while the top three most common words in the positive tweets were “love,” “black” and “beautiful.”
Most of the visual content revolved around the plight of African Americans. On the one hand, positive images and videos featured stories about black excellence and achievements that helped further scientific, technological and social progress. On the other, negative visuals typically showed incidents of police brutality or white Americans doing something racist.
In other words, these malicious accounts tweeted a mixture of sentiments to cultivate followers and manipulate U.S. narratives about race, racial tensions and police conduct.
Going forward, Russian-linked trolls could exploit ongoing reports of law enforcement brutality against protesters to further amplify division. The Kremlin’s previous attempts to hijack movements such as Occupy Wall Street and the Standing Rock protests against the Dakota Access pipeline indicate that the Kremlin can skillfully reframe social protests to increase mistrust between U.S. citizens and their government.
If African Americans continue to be the main target, we would expect the Kremlin to increase efforts to shape race-related debates leading up to and during Election Day. Without robust online safety frameworks, it will be difficult to insulate the legacy of recent protests from Russian interference.
Žilvinas Švedkauskas (@ZSvedkauskas) is a PhD researcher at the Institute of Political Science at Eberhard Karls Universität Tübingen in Germany.
Chonlawit Sirikupt (@c_sirikupt) holds an master’s degree in comparative and Middle East politics and society from Eberhard Karls Universität Tübingen.
Michel Salzer (@michelsalzer1) is a final-year master’s candidate in comparative and Middle East politics and society at Eberhard Karls Universität Tübingen.