"Will Rachel Maddow face a reckoning over her Trump-Russia coverage?” Ross Barkan asked at the Guardian last week. “With Trump has come Russia: two years of conspiracy-mongering about whether the president … conspired with the Russian government to influence the outcome of the 2016 election,” all of which had come to naught. No one had done more mongering than MSNBC's star host, or so Barkan alleged.
In a similar vein, Rolling Stone’s Matt Taibbi identified "Russiagate" as "this generation's WMD," only worse: "As a purely journalistic failure," he wrote, "WMD was a pimple compared to Russiagate." Moreover, he claimed, news that Mueller wouldn’t issue more charges was “a death-blow for the reputation of the American news media.”
These are, to my mind, wildly overheated reactions in the wake of Attorney General Bill Barr’s successful downplaying of special counsel Robert Mueller’s still-secret report, which Barr is now actively hiding. But everything’s overheated nowadays. The question is: What will stick? The idea that there was a lot of sloppy reporting, with inadequate follow-up corrections, which Taibbi’s recounting supports? Or his far more incendiary rhetorical claims and the related idea that “the left,” “Democrats” or the “mainstream media” indulged in a two year witch hunt, chasing conspiracy theories, as Trump would have it?
One challenge in trying to answer that is the ambiguous nature of the term “conspiracy theory.” To me, it is first and foremost about a way of thinking, and perhaps the most fruitful way of understanding it comes from the work of Australian cognitive scientist Stephan Lewandowsky.
On the other hand, Joseph Uscinski, co-author of "American Conspiracy Theories," takes a different view. As that book puts it, “Our basic claim is that conspiracy theories are essentially alarm systems and coping mechanisms to help deal with threats. Consequently, they tend to resonate when groups are suffering from loss, weakness, or disunity.” Uscinski and co-author Joseph Parent describe four elements of conspiracy theory:
(1) a group (2) acting in secret (3) to alter institutions, usurp power, hide truth, or gain utility (4) at the expense of the common good.
Neither the cognitive characteristics nor the question of truth vs. falsehood are central to this conception. While Uscinski’s approach has much to recommend it, for my money Lewandowsky provides a better way for being clear about who’s being clear.
"MSNBC engaged in constant Trump-Russia conspiracy theorizing for the last two years," Uscinski told me. But in saying that, he’s not saying anything about the nature of cognition involved. Given his definition, it’s almost a tautology.
Lewandowsky’s focus on cognitive characteristics, on the other hand, leaves the question open. I’ve written about his work in the past (here and here), and interviewed him for a story three weeks before Trump’s inauguration. “There would be nothing gained if Trump opponents created another conspiratorial parallel world in response to Trump," he warned at the time.
While that certainly happened with a subset of Trump online opponents, as New York magazine has noted, it has not been the dominant response — certainly not in the midterm elections, for example, nor in the early stages of the 2020 presidential race. An October 2018 midterm issues poll from Gallup found that just 45% of voters considered “investigation into Russian involvement in the 2016 U.S. election” to be “extremely" or "very" important, which was far below the top five issues: health care, the economy, immigration, the way women are treated in American society and gun policy (all of which rated above 70%). While the issue may be a major concern for fervent Trump opponents, Democrats simply didn’t run on it in 2018, even in the bluest, safest districts.
It was different in the national media, of course, but that’s not necessarily a reflection of conspiracy theory, for at least three obvious reasons: First, there was prima facie evidence that something was going on — “Russia, if you’re listening” back in 2016, and a highly productive investigation by Mueller since then. Second, investigations lend themselves to the sort of ongoing coverage that builds cable news audiences, which are more central to national political media than ever before. Third, it appealed to political elites who are unified by fears of foreign influence, and by the erosion of American institutions and norms, than by other issues more salient for voters as a whole. This was reflected by the influx of “never Trump” Republicans and ex-Republicans to the airwaves of MSNBC, for example, which far outweighed the presence of actual leftists of the Bernie Sanders or Alexandria Ocasio-Cortez variety.
Still, the potential slide into conspiratorial thinking was and is something to be concerned about. "I think we should do everything we can to restore the cultural authority of facts and reasoned evidence, and so I personally am shying away from anything that might take me into conspiratorial territory,” Lewandowsky said in January 2017. “There is enough known about Trump in broad daylight that opponents can focus on: His failure to release his tax returns or his medical records, his record as a serial sexual predator, and so on. There is no need to resort to anything that might resemble conspiratorial thought.”
With those thoughts in mind, I wanted to check in with Lewandowsky after the Mueller report announcement. Actual conspiracies do exist, as Lewandowsky and his colleagues argued in a 2017 paper, "When THUNCing Trumps Thinking: What Distant Alternative Worlds Can Tell Us About the Real World," and what he calls "conventional cognition" is superior at perceiving them than is "conspiracist cognition."
The Iran-Contra scandal, the Gulf of Tonkin affair early in the Vietnam conflict and the Tuskegee Syphilis Study are three examples Lewandowsky and his co-authors cite of how conventional cognitive methods have uncovered significant conspiracies. They identify three attributes of conventional cognition: A healthy skepticism, a responsiveness to actual evidence, and an attempt to strive for coherence.
How well have we done in meeting this standard, I wondered? Taibbi certainly documents a lack of healthy skepticism that’s been a recurrent problem with Trump-Russia coverage. But unlike conspiracy theorists, I would argue, journalists as a whole have not doubled down on those mistakes.
When I asked Lewandowsky if he agreed with that, he did what might be expected from a scientist — declined to comment for lack of data. “Given that the number of words written about Russia and Trump probably number in the billions, I cannot claim to have a grasp on what the media collectively has or has not done,” he said. He did offer some observations about “one of the more outlandish claims relating to Russia, namely that the Russians tampered with voting machines during the  election”:
In December 2016, half of Clinton voters believed in this conspiracy).
By March 2018, this had risen to two-thirds of Democrats. Now, there is a subtle difference in the wording of the questions (“Clinton voters” vs. “Democrats”), but even if one adjusts for that, it may well be the case that Democrats have continued to buy into this conspiracy theory.
However, in contrast to this widespread belief among Democrats, I am not aware of much media coverage of this — other than articles that pointed the finger at Democrats for believing this stuff. It seems to me that if the media were guilty of being gullible and uncritical, then they would have run with this voting-machine tampering as well and as far as I can tell, they didn’t really.
Lewandowsky reports that he ran the search string “Russia tampering voting machines,” at the New York Times and got only 23 hits, none of those for an article that strongly endorsed the idea. He got more than 2,700 hits, unsurprisingly, for “Russia Mueller.”
He also noted that the media's tendency to jump to conclusions about the Mueller report is surprising and likely premature: "All we know so far is that a Trump appointee sent a four-page summary to Congress that exonerates Trump of ‘collusion’; however, we don’t know yet what else is in the Mueller report that falls just short of collusion, and from all that’s happened thus far, it’s pretty clear that there was Russian interference that was aided by Americans close to Trump. I am therefore now waiting to see how this unfolds, rather than jump to conclusions.”
That is also, not coincidentally, the same stance that most of the media has taken, albeit after a brief bout of breast-beating and hysteria.
Conspiracist ideation is entirely different from that. In the past, I’ve described a set of six characteristics Lewandowsky has identified, which I think of as falling into three groups. I asked him whether the following organization made sense, and he agreed:
The first two establish a morally self-justifying framework — hidden conspirators are bad (“nefarious intention”), but conspiracy theorists are good (“persecution-victimization”) — which subverts the conspiracists’ purported interest in truth-seeking.
The second pair does this by managing unwanted information: First by ignoring or denying contradictory evidence (“nihilistic skepticism”) and then by building an elaborate fantasy edifice, incorporating random evidence (“nothing happens by accident”).
The last pair subverts truth-seeking by fending off sounder explanations, either by dogmatically insisting there’s a conspiracy even if no coherent account of it can be given ("must be wrong"), or by insisting that evidence against the conspiracy is actually evidence in favor of it — proof of how badly the conspirators want to hide the truth (“self-sealing”).
As I acknowledged in linking to the New York magazine story above, there’s evidence of some online Trump-haters succumbing to this sort of thinking. But that seems sharply at odds with how most of the traditional media has handled these admittedly confusing events. “I agree absolutely,” Lewandowsky responded, observing that he used to follow some of the conspiratorial-minded leftists mentioned in the New York article, and “dropped them pretty quickly because they seemed to turn a peppercorn of evidence into a herd of elephants."
He does not agree that the mainstream media did anything close to that. "I don’t know if they over-egged Mueller and Russia," he said, "but given the intelligence community’s findings and the indictments coming out of Mueller they would have been remiss not to give this story a lot of prominence.”
In particular, Lewandowsky cited Carole Cadwalladr’s response to the Mueller report in the Guardian as "very good.” She argued, in effect, that the most damning part of the report was hiding in plain sight:
In Britain, the news from America should be a huge red flag. Because even while we still do not have the full report, even Barr’s summary of it confirmed something extraordinary and terrifying: incontrovertible evidence of an attack by Russia on America. Mueller’s investigation has laid out how a foreign power had used America’s own media organisations and technology platforms to subvert its own democracy.
After the recent New York Times story revealing that some members of the Mueller team were critical of Barr’s memo, Lewandowsky added a bit more. “Not a surprise to me, really,” he noted. “I always thought the anti-Maddow crowd jumped the gun and will end up with egg on their faces.”
Since his initial work on conspiracist ideation, Lewandowsky has done additional work on another powerful cognitive error he calls “Thinking in Unreflective Counterfactuals” — abbreviated as THUNC, as in the title of his in above-mentioned paper. This involves hypotheses that are built on a nonexistent, counterfactual state of the world, despite the availability of knowledge about the true state of the world.
It struck me as the most clarifying account I've ever seen of what Donald Trump does virtually every day. Lewandowsky offered a brief explanation.
“Basically, it means that people make an assertion that has outlandish implications,” he said. “Even though the assertion might be true at first glance, consideration of the implications quickly reveals it to be absurd.” As an example, he cited this:
If I were to claim that “climate change is a hoax” (as Donald Trump has done, and as many conservatives are doing), then this assertion might have some prima facie plausibility (e.g., if you believe that Al Gore is the one who is solely responsible for climate change). But if you consider the implications of that claim, it all falls over in a heap. Specifically, you would have to explain why earth scientists would create such a hoax. And why did they start doing this as long ago as 1896 ... ? And why were there no whistle-blowers during the last 120 years that revealed the conspiracy? How would all climate scientists around the world conspire by publishing 14,000 peer-reviewed articles to create climate change? Who coordinates this effort? Why would the same thermometers that everybody trusts to tell us what the weather is like today not also be accurate enough to pick up climate change?
Lewandowsky's paper expands the idea further:
The cases of THUNC are more than violations of the factual world, making them counterfactual, which is easily seen and documented; in addition, the incoherence with accepted beliefs are also not reflected on, both for the originators and the followers of this brand of conspiratorial thinking. It is not just that they are wrong, it is that they do not make sense. Their beliefs are not compatible with one another in a way that is required for rationality.
There’s been no systematic exploration of THUNC, but it seems plausible that it pervades Donald Trump’s thought process, as reflected in the total carelessness with which he seems to pick up and then discard various conspiracy theories, only to pick them up again. It’s not just Trump who is implicated:
Note that the irrationality of these beliefs is shared by any followers of Trump, who also apparently lack the drive to make the set of beliefs coherent with accepted beliefs. The followers are adopting the results of Trump’s incoherent thinking, but perhaps not necessarily his exact same reasoning processes. But they are also guilty of the lack of reflection concerning just how far out this possible world is required to be, to accommodate all the “facts” of the conspiracy. Thus, they also engage in reasoning of the THUNC variety.
Consider what this means: the incoherence shared by Trump and his base is meta-coherent, in that it all fits together in a world where anything goes. The one thing it doesn’t do is converge on a shared clear picture of the world that is subject to empirical verification.
Contrast this with what we can see on MSNBC, for example. I have considerable misgivings about the channel's news coverage — the flood of “never Trumpers” noted above, among other things. It certainly has obsessively focused on the Russia scandal, and part of the reason for that (perhaps unconsciously) is to avoid thinking about why American politics was so vulnerable to manipulation in the first place.
That said, within the framework of what MSNBC does discuss, the channel's dominant tendency is the opposite of that governed by THUNC. Different people with different kinds of expertise seek to make coherent sense together in a way that’s understood in common. Ari Melber on "The Beat" even goes out of his way to get the most pro-Trump people he can entice onto his show — provided they have first-hand relevance.
This isn’t a quantified observation, but it illustrates why I profoundly disbelieve charges that liberal conspiracy theory has run amok. I asked Lewandowsky how he might design a study to test whether Democrats or the traditional media had engaged in conspiracist ideation.
“I would do a big-data content analysis that searches for the six criteria for conspiratorial thought,” he said. “This would be a massive project, of course, but in principle it could be done. You’d need to have a comparison group involving Republicans and their media (Fox, Breitbart), as well as a control that involves an issue on which there is bipartisan consensus. If you can find one!”
Another conspiracy researcher, Kyle Saunders, whose work I have reported on before, proposed a content-based approach:
If we’re just looking at ideation/belief, that’s pretty much doing a survey and straight up asking them whether they believe various Russia conspiracy theories. If you’re talking about the opinion leadership, diffusion of those beliefs and their persuasive capacity, well, that’s a study of social media like Facebook and/or Twitter that’s purely observational via social network and contagion analysis to find the main nodes. But then you’re counting on a lot of assumptions about belief and what a retweet or posting a link means.
Uscinski, for his part, proposed an approach building on his past work, employing the definition of "conspiracy rhetoric" mentioned above and then "coding a body of text from either news organizations or legislators":
The trick is to be even handed and apply the coding evenly and in a way that captures the broad range of conspiracy theories out there. I would do this as an over-time study, to avoid drawing conclusions that might reflect situations rather than central tendencies emanating from the parties. You could compare Republican legislators to Democratic ones, or MSNBC to [Fox News] and CNN. It’s not that hard to do, it’s just that I haven’t seen automated methods that can do this yet (because there is such a broad range of conspiracy theories that use many different words), so it would have to be done by hand for now.
All three of these approaches, in any combination, could help us develop a much better understanding of this phenomenon than we do right now. Although even that would leave a central question unresolved: What is a "conspiracy theory," and how do we know it when we see it? “They” clearly don’t want us to know!