A Weekly Scan of Information Disorder
Welcome to the inaugural issue of a weekly newsletter on emerging issues and trends in the field of online information disorder, with a near-term focus on mis- and disinformation affecting the 2018 midterm elections. The newsletter is intended for journalists, academics, policymakers, information technologists and other stakeholders. Subscribe here.
Welcome to the Midterm Online Frenzy
The emotion-laden scrutiny of Supreme Court nominee Brett Kavanaugh has nurtured a hothouse for online misperception and mischief. With mainstream political divisions sharpening, internet discourse has grown even more polarized and heated – and more prone to falsehoods – just a month away from midterm elections.
This is the inaugural issue of a weekly newsletter from the Information Disorder Lab (IDLab) at Harvard Kennedy School’s Shorenstein Center on Media, Politics and Public Policy. The IDLab team has identified and assessed scores of examples on both the left and the right of manipulated or misleading social media posts since the Senate Judiciary Committee hearing on Sept. 27. Even when disinformative social media posts are debunked or taken down, they blithely live on in waves of memes, retweets, shares and likes.
A social media tactic we’re seeing frequently is to resurface old allegations and tie them into the Kavanaugh debate. Democrats including former Vice President Joe Biden, Sen. Cory Booker, and Rep. Keith Ellison and Republican Sen. Lindsey Graham all have been cited in Facebook and Twitter posts comparing or linking their actual or alleged behavior to Kavanaugh’s alleged wrongdoings, often with a view toward the coming elections. Kavanaugh has thus become a convenient peg for recycling these old claims, however spurious.
A common target for Kavanaugh-driven online skirmishing is former President Barack Obama and his admitted drug use during high school and college:
Clockwise from top left: fabricated image of Obama with bong “‘I spent the last 2 years of high school in a daze.’ BARACK HUSEIN OBAMA” (22,100 shares on Facebook), image compilation of Obama smoking “Can you say hypocrites” (4,700 shares on Facebook), Donald Trump Junior quote implying unfair double standard between conservatives and liberals (6,500 shares on Facebook- no caption included), meme questioning Obama’s high school records in likely reference to Kavanaugh (4,300 shares on Facebook- no caption included), text-only meme contrasting Obama’s past substance use with Kavanaugh’s (20,600 shares on Facebook- no caption included).
Equal Opportunity Offenders
A major-league falsehood on social media, quickly debunked by Buzzfeed and others, was the tweet from @Alan_Covington on Sept. 28. The account cited the Wall Street Journal as the source for the allegation that the prosecutor who grilled Kavanaugh was silenced by GOP senators because her questions were leading the judge toward a criminal admission. The Journal editor quickly denied that the newspaper ran any such story and the original tweet was soon deleted. Buzzfeed quoted @Alan_Covington as saying later that his account had been hacked and that someone else had posted tweets that day, including the false Journal report. Before the tweet was removed, a number of prominent journalists and commentators retweeted the viral post (Crowdtangle still shows that the tweet had 8,608 interactions generated by social media accounts with collectively 5.6 million followers). Social media posts pushing this fabricated anti-Kavanaugh story persisted throughout the day and can still be found on Twitter today.
Mean Memes Metastasizing
We have been monitoring the rapid evolution of one of the year’s top memes: Nike’s “Believe in Something” campaign featuring football star/activist Colin Kaepernick. A clever variation on the now-familiar meme blends multiple themes into a single message, often conflating false equivalencies. Parodies of Kavanaugh and his accuser Christine Blasey Ford have appeared, skewering the perceived holes in their testimony.
On the left and the right, these memes deliver half-truths with a knowing laugh; they are light lifts, easily produced and reach large numbers of citizens with emotional pitches, competing with fact-based claims for attention. They employ the visual shorthand that has become so familiar thanks to millions of engagements racked up by earlier Nike ad memes featuring everyone from 9/11 first responders, to the Clintons, Alexandria Ocasio-Cortez and even some of history’s more notorious figures.
The online disputes proliferated rapidly beyond sexual assault allegations to questions of truthfulness. In the wake of President Trump’s attack on Sen. Richard Blumenthal (D-CT) for misrepresenting his Vietnam War-era military service, social media exploded on the right with posts and stories perpetuating false elements in Trump’s diatribe, frequently ignoring the fact-checks. In response, left-leaning social publishers employed a further mix of whataboutism and false-equivalency by questioning the President’s military deferments through the epithet “Cadet Bone Spurs;” this spread across a network of junk-news websites and Facebook pages to the tune of 14,715 engagements.
Conspiracies Here, There
The Kavanaugh hearings and the accusations against him fueled a series of Deep State conspiracy theories, many of which circled back to progressive billionaire financier George Soros: one Twitter handle claimed Soros was financing Blasey Ford’s security; another pointed out that the woman who accosted Sen. Jeff Flake (R-AZ) in an elevator worked for a non-profit funded in part by Soros’ foundation. Disclosure: Soros’ Open Society Foundations is one of the funders of the IDLab. A two-fer: Qanon conspiracists revived claims later shared on Twitter that Soros’ son Alex was in cahoots with Cory Booker (D-NJ) to undermine Trump. And Twitter handles shared a photo said to show Soros with Blasey Ford. In fact, the woman in the photo with Soros is actually Ukrainian activist Lyudmyla Kozlovska.
Going to Extremes
The Kavanaugh climate appears to have encouraged political rivals to trot out accusations of extremism. On the left, a Twitter account with 47,000 followers labeled four GOP senators Nazis; the tweet is no longer published. And South Carolina Republican Lindsey Graham found himself accused of kowtowing to “Trumputin” – implying Trump and Putin hold some Soviet-style “kompromat” over Graham. Some Republicans played the socialist card: supporters of Texas Sen. Ted Cruz used it against Democratic challenger Rep. Beto O’Rourke, and Florida GOP Senate candidate Rick Scott used the socialist claim against Democratic incumbent Sen. Bill Nelson. GOP Rep. Steve King from Iowa posted a long-debunked meme from 2016 of Hillary Clinton giving America’s uranium to Putin for a song in return for funding to the Clinton Foundation.
Platforms are Responding
At the same time, signs are emerging that internet platforms are willing to take stronger action against disinformation and hate speech. Twitter, Facebook and YouTube all have taken steps in recent months, shutting down millions of “inauthentic” accounts and otherwise restraining abusers. Now Reddit has joined the trend, quarantining controversial subreddits around subjects including Holocaust denial, white nationalism, communism and graphic content. These groups have more than 1.3 million users, now forced elsewhere — and importantly, deprived of opportunities to monetize their manipulation.
By generating significant engagement on posts, stories and memes trading in false equivalency, whataboutism and more serious false-context misdirection, social media users on both sides of the political spectrum have deployed their supporters for a disinformation warfare climax to this midterm election season. At the IDLab, we have our sights set on the most contentious House, Senate and gubernatorial races where we have identified and are tracking serious cases of information disorder. We’ll share our analysis with you each week. We welcome your feedback and ideas at firstname.lastname@example.org.
- 4Chan-o-ween: a briefing on information operations developing on the anonymized forum in advance of All-Hallow’s Eve.
- “Shadowbanning” on Twitter: a peek at some of our internal research into who is impacted by Twitter’s efforts to improve the quality of conversations on their platform.
What We’re Reading
- Zeynep Tufekci argues in the New York Times that Russian meddling is a symptom, not the disease.
- In a detailed analysis in Wired, Molly McKew traces the coordination tactics of domestic “information terrorists.”
- On Thursday, the Knight Foundation issued a major study on Twitter’s handling of disinformation since the 2016 election and how it dealt with a “concentrated ‘fake-news’ ecosystem.” Poynter reported on the report.
- The Google News Initiative launched the beta version of a tool designed for fact-checking content. The feature uses the same signals as other Google products, such as Google News, to surface findings from fact-checking organizations like Snopes and PolitiFact.
- A paper from FullFact, the British fact-checking organization, “sets out a framework for a risk-based and proportionate response to the problems of misinformation and disinformation in the UK….The realistic goal is not to eliminate misinformation and disinformation but to build resilience against it.”
- Shorenstein Center Research Fellow Ben Decker is cited in this Quartz article, reporting that the Kavanaugh nomination has prompted unprecedented levels of falsehood and conspiracy theories on Twitter.
Thanks for reading. Let us know what you liked (or didn’t) in this first issue, and what you want more or less of so we improve the newsletter going forward. Email us at email@example.com and subscribe to the newletter below.
–The IDLab Team