Twitter Files 15 furthers the misunderstanding of ‘Hamilton 68’
WASHINGTON (TND) — In the latest, and 15th, installment of the “Twitter Files” from Friday, Substack writer Matt Taibbi attempted to lay out a case of “McCarthyism” behavior by the now-defunct web dashboard Hamilton 68.
Hamilton 68 was a dashboard – a “graphical report (as on a website) of various data relevant to a particular business, group, etc.,” as defined by the Merriam-Webster dictionary – created by and run by the Alliance for Securing Democracy (ASD), housed by the German Marshall Fund. The ASD was established in the wake of the 2016 presidential election by conservative and liberal national security experts to counter efforts by Russia, China and other actors aiming to delegitimize democracy.
ASD launched the Hamilton 68 – named after Federalist Papers 68, in which Alexander Hamilton warns about foreign meddling in the American electoral process – dashboard in August 2017, to provide “a near real-time look at Russian propaganda and disinformation efforts online.” At time of launch, the ASD claimed that it was observing 600 Twitter accounts “linked” to Russia activities online. “Some of these accounts are directly controlled by Russia, others are users who on their own initiative reliably repeat and amplify Russian themes,” the think take wrote in its announcement. The group stated that it identified these accounts through three methods: analyzing users promoting purported disinformation that lined up with Russia propaganda outlets; finding openly pro-Russian users; and identifying accounts that appeared to use automation “to boost the signal of other accounts linked to Russian influence operations.”
Taibbi makes the argument, in part based on verifiable discussions within Twitter’s ranks, that the list of 600 accounts was bogus and largely existed to target right-leaning voices. “It was a scam,” Taibbi wrote in one tweet in the “Twitter Files” thread. “Instead of tracking how ‘Russia’ influenced American attitudes, Hamilton 68 simply collected a handful of mostly real, mostly American accounts, and described their organic conversations as Russian scheming.”
Importantly, he highlights something that Twitter staff were equally, highly concerned about: that the ASD did not make public the list of 600 accounts they were tracking. So, Yoel Roth – the former Head of Trust and Safety at Twitter who has fled his home following threats made against him as a result of the “Twitter Files” – reverse engineered the list of 644 accounts being monitored. “The selection of the accounts isbizarre, and seemingly quite arbitrary,” Roth wrote in an internal communication on Oct. 3, 2017. “They appear to strongly preference pro-Trump accounts (which they use to assert that Russia is expressing a preference for Trump even though there’s not good evidence that any of the accounts they selected are or are not actually Russian.” Roth also circulated a Google doc that contained the full list.
Two months later, in Jan. 2018, Roth wrote that he was “increasingly of the opinion that this dashboard is actively damaging and promotes polarization and distrust through its shoddy methodology Real people need to know they’ve been unilaterally labeled Russian stooges without evidence or recourse.”
The “real people” Taibbi chooses to highly include Chicago-based lawyer Dave Shestokas – who ran for Illinois Attorney General in 2022 and whose Twitter feed is full of hyper-partisan, right-wing talking points and videos – conservative media figure Dennis Michael Lynch, independent journalist Joe Lauria and some people like “Oregon native Jacob Levich” whose Twitter accounts Taibbi does not link to.
He then notes how widespread Hamilton 68 was linked to and used as a source by mainstream and left-leaning media organizations, government figures, and universities to discuss alleged continued propaganda attempts and interference by Russia in social and political issues in the U.S.
“The mix of digital McCarthyism and fraud did great damage to American politics and culture,” Taibbi writes in summary of Hamilton 68. “News outlets that don't disavow these stories, or still pay Hamilton vets as analysts, shouldn't be trusted. Every subscriber to those outlets to write to editors about the issue.”
The ASD responded to Taibbi’s claims Friday, publishing a “fact sheet” repeating its methodology in the Hamilton 68 project – which shuttered in December 2018 as the think tank “shifted its approach to analyze the overt information space by tracking official and state-backed social media accounts from Russia, China and Iran – and pushing back with its own fact check of the major allegations in that day’s “Twitter Files.”
While the fact sheet should be taken with a grain of salt, as any organization defending itself with in house fact-checks should, there are important things to note in the document. One is a key to appropriately toning-down Taibbi’s witch-hunt rhetoric. The ASD claims “members of the media, pundits and even some lawmakers often failed to include the appropriate context when using the dashboard’s data, despite ASD experts’ extensive efforts to correct misconceptions at the time.”
Those efforts included ASD voluntarily working with right-wing media publications like The Daily Caller to push back against how many media organizations were using, or misusing, the data. In an article titled “‘We don’t track bots’: what the media’s Russian bot coverage is getting all wrong” from April 9, 2018, Bret Schafer, the ASD’s then-communications director, said that most of the reporting on the dashboard was “inherently inaccurate” and “Most notably, and this is the most common errors, we don’t track bots, or, more specifically, bots are only a small portion of the network that we monitor.”
For example, Schafer and the Caller broke down reports by Business Insider and the Washington Post that reportedly misinterpreted data about Russian bots rallying behind Laura Ingraham following the Parkland shooting in Feb. 2018. While there was a reported 2800 percent increase in a pro-Ingraham hashtag on the Hamilton 68 dashboard’s trending hashtag tracker, that percentage is relative based on the number of accounts tweeting it. “Obviously, no one was using #istandwithlaura before the David Hogg controversy, so the spike there is likely evidence of only around 20-30 tweet,” said Schafer. “Also,” Schafer added, “it’s important to stress that results on the dashboard are meant to be viewed in a nuanced way; i.e., not every URL or hashtag that appears on the dashboard should be interpreted as evidence that pro-Kremlin accounts favor or oppose a certain social or political position.”
It is this point of nuance that the ASD repeated Friday in its fact sheet pushing back against Taibbi’s allegation and emphasized in its self-published guide to the Hamilton methodology. In the section “Understanding the Content” in the methodology guide, author J.M. Berger, an expert on extremism, writes “While the users in the network generally serve to promote Russian influence themes, the content within the network is complex and should be understood in a nuanced way.”
Furthermore, point two of this section may help explain why these 644 accounts were highlighted for observation:
"Content amplified to reflect Russian influence themes. This content is typically produced by third parties, including but not limited to mainstream media, hyperpartisan sites and so-called 'fake news' sites. Third-party content is sometimes amplified because it complements Russian influence themes. At other times, it is amplified for the opposite reason, meaning that users in the network are seeking to attack or discredit the content."
While not knowing the exact content that got folks like Shestokas or Lynch flagged (The National Desk did not have the ability to look back at five-year-old tweets), the tone, caliber and focus of content displayed on their pages today suggests the relevance to the national security community reading up on post-2016 election reports. In a 2020 report by the Brennan Center – “New Evidence Shows How Russia’s Election Interference Has Gotten More Brazen” – analyst Young Mie Kim notes that the Russian-disinformation farm the Internet Research Agency focused on issues like “race, American nationalism/patriotism, immigration, gun control, and LGBT issues” in their campaigns between 2014 and 2017. Considering the conflation of those wellsprings of Russian disinformation with the kinds of conservative talking points the highlighted accounts shared, it is easy to see how they could be identified as part of that Hamilton 68 survey.
The lack of nuance around the bot conversation with regards to Hamilton in the media likely led Roth to express in an Jan. 2018 email his concern that the dashboard “is leading people to assert than any right-leaning content is propagated by Russian bots (because Hamilton said so).”
What neither Twitter nor Taibbi – nor most of the pundits or media outlets – note is the specific disclaimers Berger established at the end of the methodology guide.
- Not all content in this network is “created” by Russia. A significant amount—probably a majority—of content is created by third parties and then amplified by the network because it is relevant to Russian messaging themes.
- Not all content amplified by this network is pro-Russian. The network frequently mobilizes to criticize or attack individuals or news reports that it wishes to discredit.
- Because of the two points above, we emphasize it is NOT CORRECT to describe sites linked by this network as Russian propaganda sites. We are not claiming that content producers linked by this network are Russian propaganda sites. Rather, content linked by this network is RELEVANT to Russian messaging themes.”
While, yes, Hamilton 68 was an imperfect tool, cited by imperfect, error-prone humans that seemed to have not done their research; calling it McCarthyism or fraudulent seems hyperbolic on Taibbi’s part.
The Alliance for Securing Democracy introduced Hamilton 2.0 in Dec. 2022, along similar lines of the original to “increase our understanding of the focus and spread of state-backed government messaging across various information mediums.”
There is another list of 600 accounts identified for tracking as part of the new dashboard. ASD published the full list online and for the public.
