The Propaganda Feedback Loop
The Propaganda Feedback Loop
Abstract and Keywords
This chapter presents a model of the interaction of media outlets, politicians, and the public with an emphasis on the tension between truth-seeking and narratives that confirm partisan identities. This model is used to describe the emergence and mechanics of an insular media ecosystem and how two fundamentally different media ecosystems can coexist. In one, false narratives that reinforce partisan identity not only flourish, but crowd-out true narratives even when these are presented by leading insiders. In the other, false narratives are tested, confronted, and contained by diverse outlets and actors operating in a truth-oriented norms dynamic. Two case studies are analyzed: the first focuses on false reporting on a selection of television networks; the second looks at parallel but politically divergent false rumors—an allegation that Donald Trump raped a 13-yearold and allegations tying Hillary Clinton to pedophilia—and tracks the amplification and resistance these stories faced.
Keywords: asymmetric media ecosystem, partisan identity, political elites, media outlets, propaganda feedback loop, disinformation, false rumors, Donald Trump, Hillary Clinton
IN CHAPTER 2 we showed that the American media ecosystem consists of two distinct, structurally different media ecosystems. One part is the right-wing, dominated by partisan media outlets that are densely interconnected and insular and anchored by Fox News and Breitbart. The other part spans the rest of the spectrum. It includes outlets from the left to historically center-right publications like the Wall Street Journal and is anchored by media organizations on the center and center-left that adhere to professional standards of journalism. There is no distinct left-wing media ecosystem that parallels the right in its internal coherence or insularity from the center.
Throughout the book we document how these segments of the media system operate differently. Here we sketch out a framework for understanding the differences in the incentives, mechanics, and practices between the two parts of the media landscape. The model is a stylized description of the relationships, interactions, and feedback loops between elites, media, and the public. Dynamics on the right tend to reinforce partisan statements, irrespective of their truth, and to punish actors—be they media outlets or politicians and pundits—who insist on speaking truths that are inconsistent with partisan frames and narratives dominant within the ecosystem. Dynamics in the rest of the media landscape, including the left, tend to dampen and contain partisan statements that are demonstrably false. This does not suggest that the center and left are always error-free or pure of heart. In this chapter we describe the reporting on a smear lawsuit alleging that Donald Trump raped a 13-year-old, which suggests that there is ample supply of and demand for nasty falsehoods on the left. In Chapter 6 we will encounter plenty of errors in the mainstream media. What we try to explain here is why on average and at a population level (by which we mean the population of the public, the media outlets, and political elites), we see partisan falsehoods thrive on the right not (p.76) as errors but as design features of the network, and why these design features have made the right-wing ecosystem a richer breeding ground and receptive ecosystem for propagandist efforts, foreign and domestic.
Our basic model initially divides the political landscape into political elites, media outlets, and the public. We assume, for purposes of this basic model, that members of the public consume news because they want to know what is going on in the world and how their political leaders are doing, but also want to get information that confirms them in their worldview and identity. They operate under motivated reasoning, which is to say, the basic psychological model that states that we tend to believe what we want to believe, seek out confirming information, reject or discount disconfirming evidence, and to do otherwise requires hard cognitive and emotional work. Members of the public get some gain from knowing what is really going on and some gain from being reinforced in their beliefs. And they incur some discomfort or cost from being misinformed, and from being challenged in their beliefs. They look for media outlets and politicians that will inform them as best as possible without suffering too much cognitive discomfort. Media outlets seek to attract large audiences, whether to make money, in the case of for-profits, or to better fulfill their mission, in the case of nonprofits. This involves interactions between reporters, editors, and owners, all of whom themselves operate under motivated reasoning and are perfectly able to lie to themselves to reduce the tensions between their often conflicting goals. Politicians want to be elected and achieve their policy goals. Although politicians find ways to reach out directly to their constituents, they must work through media outlets to reach a larger public. Politicians must also be informed both about the state of the world and about their voters’ perceptions of the state of the world, and must balance between these goals when they are in tension. For this purpose, politicians consume the media their voters do. Pundits occupy a position between politicians and media outlets, unconstrained by the need to be re-elected but more constrained by their need to attract audiences to media outlets. In that regard, they are less constrained by facts than politicians, and more like media outlets in their responsiveness to audience preferences. Political activists occupy a spot somewhere between politicians and the public, more focused on substantive victories than on re-election and less sensitive to criticism from media than politicians, but more capable of strategic directed action than members of the public. For simplicity, we focus on the politicians to stand in for political elites more generally.
Media and politicians have the option to serve their audiences and followers by exclusively delivering messages that confirm the prior inclinations of their (p.77) constituents, or by also including true but disconfirming news when the actual state of the world does not conform to partisan beliefs. For media, this is the key distinction between partisan media and objective media. Partisan media are oriented more toward offering identity-confirming information to partisan audiences while objective media strive for accuracy and aspire to neutrality. Audiences in turn can accept or reject the information from politicians and media, and can decide to continue to follow or abandon the politicians and outlets who deliver the news, whether good or bad. The media outlets, politicians, and public all choose each other iteratively, engaging in the sorting and matching game over time that brings together political affinity groups.
Imagine a media ecosystem in what we call a “reality-check dynamic” (Figure 3.1). Here, media outlets more or less follow institutionalized truth-seeking norms, and aim for more-or-less centrist or neutral perspectives to reduce—to the extent possible—the discomfort that their audience experiences when truth-seeking news is disconfirming. Outlets compete on the truth and freshness of their news, and the search for scoops and sensationalism is in tension with the internal norms and the fact that other (p.78) outlets will try to build their own credibility and audience in part by policing them if they get it wrong. They deliver both confirming and disconfirming news to their readers/viewers, and separate news from opinion. Politicians need media outlets to communicate to voters and must navigate this media ecosystem and are constrained in what they can get away with. They try to deliver identity-confirming statements to their voters, but have to keep reasonably close to the truth, at least as reported in media their voters consume, to avoid the reputational harm of being portrayed as dishonest. The media, in turn, report on these politicians in neutral terms and police them for the truth-value of their statements as they police each other. In this dynamic members of the public will receive a mix of truth along with both bias confirmation and disconfirmation. Centrist members of the public are the happiest while media criticism flourishes on the wings—neither of which is getting exactly the trade-off that it wants between partisan confirmation and truth. Levels of trust in any given medium are moderate, because each has occasionally been found in error by its competitors, and each has delivered their audience some disconfirming news.
Now, a new media outlet is launched that adopts a different strategy by emphasizing partisan-confirming news over truth and helping segments of the public to reduce their discomfort by telling them that the outlets providing disconfirming news are not trustworthy. Members of the public who tend to seek confirmation more than truth reward this outlet with attention. Some politicians seek out those outlets and those members. Members of the public now have media outlets and elites confirming their prior beliefs, contrary to what they hear on other media, and are also told by these outlets and elites that other media that contradict what they say are themselves biased and hence untrustworthy. The public that buys into this adjust their levels of trust in other media downward. This reduces the psychological cost of tuning in only to the bias-confirming outlets, as they are now more confident that the partisan good news they hear is true and conflicting news from other outlets is false. Politicians who thrive in this media ecosystem will have done so by aligning their positions and narratives with like-minded publics and supportive media sources or by shifting the narrative in a direction that the public and media are willing to follow. Ideological positions, interpretations of real-world events, and partisan talking points are jointly negotiated by elites, partisan media, pundits, and political activists. News media reject the separation of news and opinion, and compete by policing each other for deviance from identity confirmation, not truth. They similarly align their coverage of politicians to offer favorable coverage to identity-confirming politicians and attacks on (p.79) opponents, and when they police deviance from politicians, it is identity confirmation, not truth, that they police. All these are intended to help sustain a steady flow of identity-confirming news to audiences who tune in to get precisely that from their media. Subsequent politicians who now enter the arena will find it harder to rely on the mainstream media to challenge assertions made by politicians focused on bias-confirming statements. The public that occupies the partisan media ecosystem no longer believes the external lying media. Challengers within the party are forced to use the same partisan media, subject to the same trade-offs between truth and bias confirmation as the incumbent. Incumbents and challengers consume the partisan media more, because they need to understand what their public believes, what they must confirm, and what parameters shape the way in which they can challenge incumbents. We call this dynamic the “propaganda feedback loop,” because once it is set in motion the media, elites, and public are all participants in a self-reinforcing feedback loop that disciplines those who try to step off it with lower attention or votes, and gradually over time increases the costs to everyone of introducing news that is not identity confirming, or challenges the partisan narratives and frames (Figure 3.2). Audiences in this loop will exhibit high trust in identity-confirming media, and low trust in external media. Politicians are not constrained by media policing truth when they (p.80) deliver identity-confirming news to outlets, but rather by media policing the consistency of their statements with party identity.
Propaganda as we define it in Chapter 1—the manipulation of public beliefs, attitudes, and behaviors for political ends, framed in terms that reinforce partisan narratives—is much easier to insert into a system whose audiences, outlets, and elites have nothing to gain by disputing or disbelieving it and everything to lose by doing so, and will survive longer and propagate further in it. The network takes these inputs and converts them into a partisan package delivered to its various constituents: a steady flow of bias-confirming stories that create a shared narrative of the state of the world; a steady flow of audiences, viewers, or clicks for the outlets; and a steady flow of voters highly resilient to arguments made by outsiders on outlets that are outside the network. Outlets within the network are not designed to check or refute propaganda as long as it is consistent with the partisan narrative. There is nothing to gain and everything to lose. Our discussion in Chapter 2 of what happened to Fox News during the primaries and since the election follows these contours precisely. This does not mean that partisan audiences are not exposed to arguments from the other side. Research shows that engaged partisans are indeed aware of the arguments and reporting from the other side,1 but partisan news audiences simply discount opposing views. When the propagandist’s efforts are exposed to external criticism and fact checking, the mechanisms that developed for reducing the cost of disconfirmation—lower attention and lower trust to external media—kick in to insulate the propagandists’ efforts from this external criticism.
By contrast, a media ecosystem that operates under the reality-check dynamic, will tend to be more robust to disinformation operations because each outlet in this system gains from exposing the untruth and loses from being caught in the lie or error. Its audiences are less likely to trust any media source in particular, and more likely to check across different media to see whether a story is, in fact, true. Politicians operating within this type of media system will tend to experience resistance when they lie, even if the lies initially confirm the biases of their voters, because the voters will soon learn that the politician lied and reality is not as rosy as the politician promised. The politicians and outlets face starkly different incentives than politicians who operate in the partisan media ecosystem, and the outlets have developed norms and institutions that reflect those incentives and stabilize the truth-seeking behavior.
Nothing prevents the opposite wing of the political system from developing a parallel partisan media ecosystem, and there are many well-functioning (p.81) democracies with a frankly partisan press, particularly in multipolar political systems. But once one wing has established the strategy of partisan bias confirmation, the centrist media with their truth-seeking institutions and reputations suddenly deliver a new benefit to partisans of the opposite pole—as objective external arbiters they can offer institutionalized credibility to reinforce their view that what their opposition is saying is false. Once one partisan media pole is established, the coverage of existing objective media outlets takes on a partisan flavor without any shift in their own focus on objectivity.
The mainstream media will be able to reconcile their goals of truth-seeking and confirmation from the center with providing a steady flow of partisan-confirming news for the wing in opposition to the wing that is already in the grip of the propaganda feedback loop. The outlets that formed the partisan ecosystem have a first-mover advantage over outlets that try to copy them on the opposite side, because as they decrease the value of the mainstream media to their own audiences, they increase it for the putative audiences of their opponents. The further the first-moving partisan media ecosystem goes down the path of its own propaganda feedback loop, the greater its tendency to produce untrue statements, and the greater the opportunities for reality-check centrist media organizations to deliver news that is both truthful and pleasing to partisans from the other side. Creating a second partisan media ecosystem then becomes more difficult as their potential audience now receives partisan confirmation from centrist objectivity-seeking outlets, and the incremental identity-confirmation benefit they offer their audience is reduced.
In the real world, as opposed to the model, there are of course elements of these two media models found in different media on both sides of the political spectrum. The left includes hyperpartisan media sources, and there are many professional journalists on the right who adhere to standards of objective journalism. The differences lie in the relative power and prominence of the different parts of the system. On the left, politicians and partisans have to navigate the scrutiny and fact checking of objective media sources to reach broader audiences. On the partisan right, the gatekeepers are Fox News, talk radio, Breitbart, and the Drudge Report.
We return in Chapter 11 to examine the historical development of the right-wing media ecosystem, and argue that Rush Limbaugh, and after him Fox News, triggered a propaganda feedback loop that shaped the right-wing media ecosystem as we observe it today. The leading right-wing online media of 2018 were introduced into what was already a decade or two of that feedback loop. Even Breitbart was founded in 2007, almost twenty years after (p.82) Limbaugh became nationally syndicated. The other major right-wing net-native sites like the Daily Caller are of even more recent vintage. By contrast, when the Huffington Post emerged in 2005, the left was still attentive to media functioning roughly within a reality-check dynamic, however strained by failures in coverage of the Iraq War and weapons of mass destruction (WMDs).
Here, we turn to two case studies to put some narrative meat on this rather bare-bones model. The first takes an imperfect but nonetheless best-there-is accounting of how much lying there is on the various television networks, and who does the lying where. The second takes the most directly comparable rumors during the election period—an allegation that Trump raped a 13-year-old and allegations tying Clinton to pedophilia—and traces how each fared in the two parts of the media ecosystem. Both cases offer concrete examples of how, a propaganda feedback loop operated on the right, while on the left, an ample supply and appetite for bias-confirming news was contained by the reality-check dynamic.
Who Lies on TV?
It is hard to conduct a large-scale study of the prevalence of false statements on different media, because it is difficult to find unambiguously objective arbiters, and harder still to find arbiters who make their data available in a way that is readily available to aggregate at the news outlet level. Several studies since the 2016 election have sought to use consensus across several fact-checking sites to identify false stories. The limitation of those approaches is that most sites do not organize their findings in ways that are conducive to studying specific outlets or sources. We used PolitiFact, which is the only major fact-checking site to aggregate its findings by media outlet and which reports stories that they determined to be true as well as false. This therefore allows for at least a baseline sense of what proportion of the statements for a given media outlet that were suspicious enough to check ended up being true or false. PolitiFact was accused of political bias a few years ago after a George Mason study found that PolitiFact rated statements by Republicans as false more often than they rated statements by Democrats as false. This finding, which was not accompanied by any effort to independently assess the fairness of the underlying decisions, was interpreted by conservative media as evidence that PolitiFact was biased, rather than that the prevalence of lying on the right might actually be higher.2 Without valid independent evaluation of the actual truth or falsehood of the claims assessed, it is of course impossible to tell whether (p.83) PolitiFact found more lies on the right than on the left because the site was biased in its selection of stories to check or in its determination of the facts, or whether, in fact, right-wing politicians lie more often than do those on the left.3 That possibility, that the difference is in base rates of lying, rather than the bias of the fact checker, is further supported by some early studies of the Washington Post’s Fact Checker, which also found more lying on the right than on the left.4 Nonetheless, because this basic question of selection bias is not resolvable by outside observers, we offer the following analysis merely as an initial qualitative illustration of this systematic difference. We do not present it as quantitative evidence of its existence, and caution against overinterpreting it in quantitative terms.
PolitiFact is run by professional staff of the Tampa Bay Times, which is owned by the Poynter Institute. The site’s staff select statements to be checked based on whether they are rooted in a verifiable fact, whether the editors believe the statements might be misleading, and whether they are publicly significant. In other words, we should expect to see a relatively high base rate of false statements in the full set, because stories only get checked if they at least raise suspicion in the minds of a professional reporter and fact checker. As a rough measure of the partisan bias of the site, we can compare PolitiFact’s findings on prominent politicians and pundits from the two sides of the aisle. For example, as of December 2017 the Republican and Democratic leadership had remarkably similar proportions of statements rated by PolitiFact as “mostly false” or worse: Mitch McConnell (43 percent), Paul Ryan (43 percent), Nancy Pelosi (41 percent), and Chuck Schumer (42 percent) were practically indistinguishable in terms of the likelihood that the statements they make that seem suspicious enough to warrant fact-checking turn out in fact to be mostly false or worse. Similarly, Rachel Maddow (48 percent) is not systematically different in PolitiFact’s measurements than Bill O’Reilly (53 percent) or Sean Hannity (50 percent), all of whom are quite different from, say, Rush Limbaugh (81 percent) and possibly Glenn Beck (59 percent). These numbers show the high ratio of false statements in the stories selected for checking. The overall rates per channel that we describe below do not represent the proportion of false statements out of all statements on the channel, but only the proportion of those statements treated as fishy enough to check. And while the site indeed checks more statements by these right-wing personalities than by the left-wing personalities, the similarities in findings of falsehood suggest that Politifact is not politically biased in its determinations of truth or falsehood. .
Comparing Fox News to CNN, MSNBC, and ABC is revealing.5 As Figure 3.3 shows, the proportion of Fox News statements that are mostly false (p.84) or worse is almost 50 percent higher than for MSNBC, and more than twice that of CNN.
More revealing yet is the proportion of false statements on each channel that are made by conservative speakers, both paid hosts and pundits and unpaid interviewees. As Figure 3.4 shows, part of the Fox News business model seems to be to pay conservative hosts and pundits to make statements that confirm the biases of its viewers, whether these statements are false or not. MSNBC seems to use a similar approach, but may be less disciplined about it. Most interesting perhaps is that 60 percent of false statements that PolitiFact found on CNN were made by conservative interviewees, some paid pundits, most not, while only 20 percent of such statements were made by liberals. Similarly, nearly 50 percent of the false statements on ABC were made by conservative speakers, and only 21 percent by liberals.
These findings suggest that Fox News and MSNBC are mostly mirror images of each other and follow a fundamentally different model than CNN or ABC. Both Fox News and MSNBC follow a bias-confirming strategy, while CNN and ABC are following roughly a neutral, balanced-perspective strategy. The differential lying by conservatives on CNN and ABC, in turn, is consistent with what we outline in the propaganda feedback loop: politicians who serve a population that is itself trapped in a bias-confirming media ecosystem will tend to lie more often on average than those who serve a constituency that follows neutral channels. For a politician who depends on votes of a public that is mostly inside the propaganda feedback loop, being (p.85) “caught” in making identity-inconsistent statements even on mainstream media is more costly than being “caught” lying on those media. The difference in rates of making false statements on media persists on CNN and ABC, which suggests that MSNBC’s market share is too small, and liberals and Democrats use too many other media that do not follow the bias-confirming strategy to drive liberal politicians and pundits to lie quite as often as their conservative counterparts.
A Case Study in the Effects of the Propaganda Feedback Loop vs. the Reality-Check Dynamic: Clinton Pedophilia vs. Trump Rape in the 2016 Election
There is quite possibly no crime so widely reviled in America as sexual abuse of children. During the campaign, accusations relating to pedophilia were reported against both candidates. Accusations against Clinton ranged from the core story of Bill Clinton (and later Hillary) flying to “orgy island” on the “Lolita Express,” to claims that Clinton’s State Department was involved in child trafficking of 33 children after the Haitian earthquake, to satanic rituals (p.86) and “Pizzagate.” For Trump, they were focused on allegations in a civil suit that the candidate had raped a 13-year-old in 1994. One can get a sense of the vehemence on the left, manifest in an Occupy Democrats headline from October 24, 2016: “Trump allegedly tied-up 13 year-old girl he raped, struck her in the face.” This headline appeared over a story that included sentences such as: “The imagery of a nude, seething, sweating Trump having his way with a defenseless teenager is enough to induce vomiting” leaving little doubt as to the intended effect on readers.6 The Huffington Post analysis of the story, while more measured in tone, argued that the story deserved significant attention, and was itself the most widely shared story of rape or pedophilia thrown at either candidate throughout the entire campaign. In fact, it had over five times as many Facebook shares (over 1.25 million) as the most widely shared story about Clinton pedophilia.7 There was no lack of appetite on the left to hear disgusting stories about Donald Trump.
We can look at the spread of these narratives in online media by counting the number of sentences that mention either the Clinton pedophilia frame or the Trump rape frame. We start by looking at the frequency of these frames on the right and the left that are popular on Facebook but have little or no presence in the open web link economy and Twitter. As we explained in Chapter 2, sites with this attention profile include some of the worst offenders in terms of hyperpartisan bullshit on both sides of the aisle—Bipartisan Report, Addicting Info, and Occupy Democrats on the left, and Truthfeed, Ending the Fed, and Western Journalism, on the right. Looking at these sites (Figure 3.5), we see a symmetric pattern: the sites on the left scream “Trump Rape!”; the sites on the right scream “Clinton Pedophilia!”
The difference between the left and the right is not at the extreme margins, but in the transmission to the mainstream and in the amplification effects of the propaganda feedback loop as opposed to the resistance of the reality-check dynamic. We can run the same textual comparison on the sites that receive the most media inlinks from other sites, which is a useful measure of the sites considered most authoritative (Figure 3.6). The accusations against Trump, although they offered an easy opportunity to report allegations made in publicly available court filings, received a fraction of the attention that the accusations against Clinton received among these top media. In fact, the top media on the left wrote more than twice as many stories, and almost three times as many sentences, about the pedophilia accusations against Clinton as they did about the allegations against Trump. The right, by contrast, wrote consistently more damning stories and sentences about their opponent. (p.87)
(p.88) Overall, the number of stories that the Facebook hyperpartisan sites publish is smaller than the larger, better-staffed sites. The overall magnitude of sentences is therefore lower, but the pattern of symmetric polarization among the hyperpartisan sites is clearly visible in the supply of disgusting, stories intended to provoke partisan rage. The critical difference made clear in the two sets of charts is that on the left, those media higher up the food chain that lend credibility to stories and amplify them to larger audiences refused to pick up the storyline and bring it to the wider public.
We can observe the extent to which the right-wing media was different in its treatment of these incendiary stories by comparing top-performing media by quintile across Facebook, Twitter, and links (Figures 3.7a–c).8 The same picture emerges; the pattern remains stable as we compare the top media by linking, tweeting, and Facebook sharing across the five partisan-attention quintiles. A key difference is that the pattern is less symmetric on Facebook, with substantially more coverage on the right than the left. With the larger number of media outlets included in this view, the craziest sites on the left are diluted, while the bench is deeper on the right. Facebook remains the only medium on which sites on the left give more attention to the Trump rape allegations than the Clinton pedophilia allegations. Systematically the top (p.89) (p.90) right-wing sites repeat the Clinton pedophilia stories several times as often as any other quintile does, and other than on Facebook, there is no symmetrically excessive attention being paid on the left to the mirror-image story intended to elicit visceral disgust toward Donald Trump. Even on Facebook, where we do see the left quintile pushing the Trump rape narrative, it does so to a substantially smaller degree than the right pushes the pedophilia stories on Facebook. The asymmetry we observe in the architecture of attention appears to be replicated here in the emphasis that the sites themselves pay to producing feelings of outrage and disgust in their audience. The top media by all metrics on the right are in the thrall of the propaganda feedback loop. They must participate to retain credibility. The top media in the rest of the media ecosystem are constrained by the inverse dynamic. They cannot participate if they are to retain credibility and audiences. And the reason they cannot participate is revealed when we move from looking at overall emphasis to the network architecture of the two frames. (p.91)
Figure 3.8 describes the network of sites that reported both frames, with edges and node sizes determined by hyperlinks into the story. One feature pops out at anyone who has already been exposed to our maps in Chapter 2. The top sites on the center-left and center that normally occupy the central position in our map—the New York Times, the Washington Post, CNN, The Hill—played a peripheral role in this map. The map is driven by partisan sites, not by mainstream media. The second feature is that the Daily Beast, the Guardian, Gawker, and New York Magazine play a central role on the left and center-left, while Fox News and the Daily Mail are the basin of attraction on the right. This offers us very clear targets to examine the diffusion dynamics in the two networks.
Jane Doe and the Trump Rape Accusations
On June 20, 2016, an anonymous “Jane Doe” filed a lawsuit against Donald Trump, accusing him of having raped her 22 years earlier, when she was 13, at an event organized by Jeffery Epstein. On June 29, Lisa Bloom, legal analyst for NBC News, published a piece on the Huffington Post that went through the complaint and argued that the media should cover the case, rather than turning a blind eye as it did in the Bill Cosby case. Bloom argued that Trump’s overt misogyny, past accusations of sexual assault against him, Trump’s known and publicly admitted long-term relationship with Epstein, and the highly unusual presence of sworn affidavits—one from the victim herself and another from a witness—make the case particularly worthy of attention. Bloom concluded her piece with a clear challenge: “What do you call a nation that refuses to even look at sexual assault claims against a man seeking to lead the free world? Rape culture. We ignore the voices of women at our peril.” Bloom’s piece was shared over a million times on Facebook. It was followed on July 7 by a piece on MSN, which made roughly the same argument as in Bloom’s closing line, and which became the second-most Facebook-shared story addressing either of the two rumors.
On that same day the Guardian published a story that led with, “Lawsuits accusing Donald Trump of sexually assaulting a child in the 1990s appear to have been orchestrated by an eccentric anti-Trump campaigner with a record of making outlandish claims about celebrities.”9 The story highlighted an earlier debunking story on Jezebel, which had framed its own discussion with the statement: “The allegations are beyond the pale, even for Trump, and there’s little sourcing for them beyond the lawsuit itself.”10 Anna Merlan’s story in Jezebel recounted, among other details, how both Gawker and (p.92) Jezebel had been approached by an “Al Taylor,” who put them in touch with a conservative NeverTrump activist, Steve Baer, and tried to sell them on the rape story, and how both outlets decided to reject the proffered “evidence” and not report on the story. The Guardian story then dug deeper and found that “Al Taylor” was likely Norm Lubow, a former producer for the Jerry Springer Show, and proceeded to recount the many times in which Lubow had been involved in generating fake conspiracies. Two weeks later, on the eve of the Republican convention, the Daily Beast dug deeper into the NeverTrump angle, and how the activists created the story but then fell into a cycle of mutual recriminations.11 There were very few stories after the Guardian or Daily Beast stories that reached any prominence at all on the left or center-left, except much later, and those that did outside of the hyperpartisan set like Occupy Democrats all noted and linked back to the Guardian, Daily Beast, and several other debunking or critical stories, such as coverage on LawNewz. And what about Gawker’s prominence in the map? Gawker, which had chosen not to buy into the Trump rape case, found itself a part of the “Clinton pedophilia” frame through the good graces of Fox News.
What prevented the “Trump raped a 13-year-old” frame from taking off on the left was not a lack of audience desire to receive strong, visceral confirmation of their hatred of Trump. Nothing could be more bias-confirming for Trump opponents. The tremendous success of the Huffington Post and MSN stories during the week of June 29–July 7 exhibits that the desire to believe such a story about Trump existed in spades. It was not the absence of political clickbait fabricators who were trying to push the story to their financial benefit. The passage we quoted from Occupy Democrats and the supply of Trump rape stories from that kind of political clickbait site we showed in Figure 3.5 establish that there were efforts to capitalize on supplying more of this type of story. What prevented this disinformation effort from taking root was the network dynamic whereby diverse sites, many operating on norms dedicated to evaluating the veracity of a story rather than its fit to political purpose or clickbait value, check each other. Even in the absence of the more traditional mainstream press (except the Guardian), the presence and attention of both journalists and readers to diverse sites was enough to enforce a hard constraint on the ability to disseminate politically affirming falsehoods.
Jeffrey Epstein, the Lolita Express, and Orgy Island
The prominence of Fox News in the pedophilia narrative underscores that network propaganda is not primarily a story about marginal sites being (p.93) amplified by big sites, although it sometimes exhibits that dynamic, too. What we see when looking at the Clinton pedophilia frame on the right is a dynamic that rewards the most popular and widely viewed channels at the very top of the media ecosystem for delivering stories, whether true or false, that protect the team, reinforce its beliefs, attack opponents, and refute any claims that might threaten “our” team from outsiders. The Clinton pedophilia story was not primarily a bottom-up, Reddit-imposed, post-truth moment that was then reinforced by higher level media. It was first and foremost the product of a propagandist dynamic between Fox News, as party propaganda organ, and Donald Trump, who, after winning the Indiana primary on May 4, 2016, became the presumptive party nominee. The Clinton pedophilia story was a direct response to the first round of media attacks on Trump for his treatment of women. Fox News’s critical contribution to the Clinton pedophilia narrative was the most widely tweeted, most widely Facebook-shared, and third-most widely linked Fox News story published in May 2016 of any election-related story. Indeed, the story remained the most widely shared Fox News story on Facebook throughout the entire campaign out of nearly 6,500 Fox News stories in our set. And it was in May that Fox News return to the fold, after having been browbeaten by Breitbart and shunned by Trump supporters while the primary was still contested, as we saw in Chapter 2.
On May 13, 2017, Fox News online published a story entitled “Flight logs show Bill Clinton flew on sex offender’s jet much more than previously,”12 a story that Fox sent out for syndication via RSS and apparently on Facebook with the more “clickbaity” title: “JET PURV-EYED FOR BILL Clinton a frequent flyer on sex-offender’s plane.”13 The story resurfaced a story that had been reported by Gawker a year earlier, but embellished it with a new set of “flight logs showing the former president taking at least 26 trips aboard the ‘Lolita Express’—even apparently ditching his Secret Service detail for at least five of the flights, according to records obtained by FoxNews.com.” The story then wove various elements from Epstein’s own prosecution to embellish the details of Orgy Island. That same day, Breitbart reported on the Fox News story to its readers,14 as did the British tabloid the Daily Mail,15 one of the most linked and shared sites among Trump supporters. The Daily Caller16 and Washington Times17 reported on it the next day.
On May 14, the New York Times published online a scathing story entitled “Crossing the Lines: How Donald Trump Behaved with Women in Private,” which started with a description of how Trump had asked a young woman, Rowan Brewer Lane, to change out of her clothes and into a bikini at a pool (p.94) party at Mar-a-Lago, a pattern of behavior the Times called “a debasing face-to-face encounter between Mr. Trump and a young woman he hardly knew,” based on interviews with over 50 women.18 One must assume that the Times had contacted the Trump campaign for its response before the May 14 online publication. The story was printed in the May 15 Sunday Times. In a dynamic typical of the network diversity of the non-right media ecosystem, the next morning, on May 16, MSNBC ran a 17-minute segment of “Morning Joe” eviscerating the New York Times for poor reporting, and leading with a segment from a Fox News interview with Rowan Brewer Lane, where Lane rejected the interpretation of the Times article, saying that Trump “treated her like a gentleman.”19 After about 10 minutes of criticism of the Times’s journalistic decisions in developing and publishing the stories, the conversation shifted to assessing what impact it would have on the campaign, and the panelists all agreed that it would be “gloves off” as far as looking at Bill Clinton sex stories. At that point Donny Deutsch raised the Epstein story and emphasized that it would be a likely target for a significant push. Three hours later, in another MSNBC show, conservative talk show host Hugh Hewitt repeated the Deutsch point.20
Like “Morning Joe” on MSNBC, Sean Hannity ran a segment on his May 16 show, which also began with an interview with Lane, who claimed her words were taken out of context and blamed the Times piece as being a hit piece. Newt Gingrich followed Lane and said:21
GINGRICH: Well, look, The New York Times is totally in the tank for Hillary Clinton. They’re faced with this terrible story about Bill Clinton flying around the world with a convicted pedophile, actually leaving his Secret Service agents behind on at least five trips to go off with his guy by himself for whatever reason. And so they basically wanted to smother a real scandal involving Bill Clinton once again with this made-up story.
The Daily Caller added another round, reporting that “Former President Bill Clinton is set to campaign in the U.S. Virgin Islands Monday where his friend and convicted pedophile Jeffrey Epstein operated ‘Orgy Island,’ ”22 while Breitbart used the Deutsch’s comment on Morning Joe to publish the headline: “ ‘Morning Joe’ Panel: Clinton Connection with Billionaire Pedophile Jeffrey Epstein Will ‘Blow Up’ Campaign.”23
The next day, May 17, “Special Report with Bret Baier” ran a segment that began with the words “there’s smoke, and where there’s smoke, it’s worthy (p.95) of an investigation.”24 The story essentially repeated for Baier’s TV audience the detailed account that Fox News had published online four days earlier, showing the image of the flight logs “and Orgy Island,” and emphasizing and analyzing the significance of the absence of Secret Service agents on some of these flights.25 And again, on May 18, Gingrich returned to the Epstein “Orgy Island” story on Fox News’s “On the Record” with Greta Van Susteren, an interview that the Daily Caller dutifully reported.26 On May 19, Hannity interviewed Donald Trump for his response to the Times’s story, an interview during which Trump and Hannity emphasized the refutations by several women mentioned in the story and during which Trump underscored that Bill Clinton had been accused of rape, referring to the Juanita Broderick accusations.27 The Epstein “Lolita Express” story returned periodically in the major right-wing outlets over the remainder of the campaign, as a general trope standing in for Bill Clinton’s sexual misbehavior more generally. A June 2016 report on Breitbart excerpted a portion of a transcript from Rush Limbaugh, where Limbaugh says: “if there ever is a man who has had numerous affairs in the Oval Office, outside the Oval Office, in the governor’s office in Arkansas, around the world, palling around with noted pedophile Jeffrey Epstein—why does Bill Clinton get the benefit of the doubt? After all of these years, there ought to be a reaction. Common sense would say, ‘You know what? This many allegations, there has to be something there.’ ”28 In July, as the Clinton Foundation story that we discuss in Chapter 6 was taking off, Malia Zimmerman, who had authored the original May 13 story based on the flight logs, published another piece, entitled “Billionaire Sex Offender Epstein once claimed he co-founded Clinton Foundation,” a claim that the Daily Mail faithfully repeated, crediting Fox News.29
The announcement by James Comey that the FBI was reopening the Clinton email investigation because of emails found on Anthony Weiner’s laptop provided a new opportunity for reviving the pedophilia conspiracy theory. On November 2, True Pundit, a particularly consistent purveyor of disinformation and conspiracy theories, published a story in which it said that “NYPD and FBI sources” confirmed “they now believe Hillary Clinton traveled as Epstein’s guest on at least six occasions, probably more,”30 as well as noting the earlier stories that had “confirmed” that Bill Clinton had traveled more than 20 times to the island and that Huma Abedin and Weiner were cooperating with the FBI. That same day, the story was repeated and linked to by YourNewsWire.31 As we will see in Chapter 8, both True Pundit and YourNewsWire are among the sites that could plausibly be Russian gray propaganda sites, although YourNewsWire seems to be more of a “useful idiot” (p.96) site than a gray site. The following day, November 3, Michael Flynn, then of the Trump campaign and later briefly national security adviser, tweeted out a link to the True Pundit story, attached to a message that read: “U decide—NYPD Blows Whistle on New Hillary Emails: Money Laundering, Sex Crimes w Children, etc. . . MUST READ!”32 On November 4, Erik Prince, founder of Blackwater USA, later Trump transition team member, and brother to Education Secretary Betsy DeVos, spoke in an interview on the Breitbart XM radio show and broadcast the True Pundit story in all its significant details. As Breitbart described the interview, “Prince claimed he had insider knowledge of the investigation that could help explain why FBI Director James Comey had to announce he was reopening the investigation into Clinton’s email server last week.” Prince claimed that his informants told him: “They found State Department emails. They found a lot of other really damning criminal information, including money laundering, including the fact that Hillary went to this sex island with convicted pedophile Jeffrey Epstein. Bill Clinton went there more than 20 times. Hillary Clinton went there at least six times,” he said.33 The story leaves little doubt that Prince’s “inside” sources were likely the True Pundit story—whether he read it directly, picked it up by following Flynn on Twitter, or read the YourNewsWire repeat.
Unlike many of the other stories, which tried to insinuate that Bill Clinton traveled to Orgy Island (a fact that, though oft-repeated, was inconsistent with the flight logs, which showed Clinton on other trips on the plane, none to the Caribbean), Prince’s broadcast of True Pundit’s story took elements of the story developed by Fox News in May and created a new layer of fabrication from them. The numbers used evoke the earlier story lines—more than 20 times clearly connects to the “at least 26 flights” from the Fox News online and Bret Baier stories, and Hillary’s alleged six visits relate to the number of times the Secret Service was not on board. In combination, True Pundit, Flynn, and Prince sent Hillary to Orgy Island repeatedly. The validation of this storyline by Prince and Breitbart helped to make this new twist the primary narrative in the Clinton pedophilia libel. WND, a popular site on the right, cited the Breitbart story on Erik Prince’s radio interview under the headline “Source: FBI Has Evidence Hillary Visited Orgy Island; Huma Abedin said to be cooperating with investigators.”34 Other visible right-wing sites, IJR, Western Journalism, and Conservative Tribune, all linked to and repeated various variations on the story over the following 24 hours, along with several other, more peripheral sites.
If the quintessential moment that captured the sense of epistemic crisis in America was the day on which an armed 20-something walked into a pizza (p.97) parlor in search of the Clinton pedophilia ring, a close study of the ways in which the association between Clinton and pedophilia emerged points the finger at the very top sites in the right-wing media sphere as the primary culprits in feeding and sustaining this crisis. Russians may well have joined in the fun at various points, as we will see in Chapters 7 and 8. Alt-right agitators happily contributed memes and associations, as we show in Chapter 7. But these exotic explanations seem entirely unnecessary when the political leadership and leading media outlets of the American Republican Party were all repeating the lies as truths in outlets watched, heard, and read by millions of people. Newt Gingrich, former Speaker of the House of Representatives, made the accusation on-air, interviewed by Sean Hannity and by Greta Van Susteren on Fox News. Bret Baier, anchoring Fox News’s prime-time news show ran a detailed segment on the accusations of Bill Clinton flying to Orgy Island on the Lolita Express. Fox News online published the underlying materials. Rush Limbaugh discussed the allegation as something everyone knows. Trump’s campaign adviser and national security adviser in waiting, Michael Flynn, tweeted it out. Breitbart, the most widely shared right-wing online site whose on-leave CEO Steve Bannon was then running Trump’s campaign, aired an interview constructed of pure disinformation. It seems highly unlikely that any of the people involved—Prince, Flynn, or the publishers of Breitbart—thought that the accusation that Hillary Clinton had flown six times to Orgy Island was anything other than utterly false, and yet they published it four days before the election on Breitbart’s radio station and online. Not one right-wing outlet came out to criticize and expose this blatant lie for what it was. In the grip of the propaganda feedback loop, the right-wing media ecosystem had no mechanism for self-correction, and instead exhibited dynamics of self-reinforcement, confirmation, and repetition so that readers, viewers, and listeners encountered multiple versions of the same story, over months, to the point that both recall and credibility were enhanced. It is hardly surprising, then, that a YouGov poll from December 2016 found that over 40 percent of Republican respondents thought that it was at least somewhat likely that someone was running a pedophilia ring out of the Clinton campaign.
Our observations about the asymmetric shape of the American networked public sphere show that we do not have a fully polarized media ecosystem, in which both Democrats and Republicans occupy symmetrically closed media bubbles with symmetric, but opposing, self-reinforcing views. Instead, both pre- and post-election, a substantial portion of Republicans and self-identified conservatives occupied a self-reinforcing bubble, while (p.98) Democrats and Independents occupied a media sphere anchored by more traditional media outlets that continue to practice the norms of objective journalism, surrounded by more partisan net-native outlets, many of which also adhere to truth-seeking norms rather than purely partisan advantage. The incentives for elites, and the feedback mechanisms between media, elites, and the public, therefore diverged between the two spheres. The center and left dynamics combine networked media and major media outlets, with the latter playing a moderating effect on partisan bullshit and on politicians who still have to worry about fact-checking sites giving them too many Pinocchios, and the former checking the more traditional media from becoming too complacent or comfortable with conventional wisdom, and were able to adhere to truth-seeking norms because their audience, in turn, was ready and able to distinguish truth from falsehood and reward the former. That still leaves plenty of room for partisanship—in agenda setting and topic selection, in perspective and framing—but it appears to significantly constrain disinformation. The interaction between these diverse media sources, their audiences, and political elites creates resistance to the spread of falsehoods, even when politically convenient. The right followed a distinctly different media model, with outlets and elites providing reinforcement and legitimation to partisan propaganda. In this insular ecosystem, subject to the dynamics of propaganda feedback loops, politicians are more or less immune to fact checking because their core audiences treat the professional fact-checking process as itself partisan, and media outlets, audiences, and elites all discipline each other to remain true to faith and confirm party beliefs and ideology on pain of exclusion and demotion—at the ballot box or in the market for attention.
In Chapter 11 we describe how this asymmetric media architectures resulted in highly divergent levels and patterns of trust in media, patterns that offer substantial support for our observations and represent one component of the propaganda feedback loop, on one hand, or the reality-check dynamic, on the other hand. Conservative audiences tend to focus their attention more intensely on fewer sources, which are purely conservative, and to trust these sources much more than liberals do. Consistently conservative respondents trust Fox News, Hannity, Limbaugh, and Glenn Beck much more than they trust ABC, CBS, or NBC, and trust CNN or public radio even less. Consistently liberal audiences, by contrast, spread their attention and trust more broadly, and focus on major mainstream publications and channels.
Parts Two and Three of the book demonstrate that the difference in media ecosystem architecture, the dynamics they set in place, and the divergent patterns of attention, authority, and trust they reflect resulted in different (p.99) susceptibility to manipulation, propaganda, and belief in false factual claims. Russian propaganda seems to have targeted both sides, and Facebook clickbait sites tried to manipulate denizens of all sides of the American public sphere. But, just as we saw in the case of the competing Trump rape and Clinton pedophilia frames, the responsiveness and success appear to have been very different in the two parts of the media ecosystem. In the right-wing media the propaganda feedback loop enabled conspiracy theories, false rumors, and logically implausible claims to perform better, survive longer, and be shared more widely than were parallel efforts aimed at the left.
Our explanation of the difference between the right and the rest is grounded in historically-specific changes in media institutions and cultural practices, rather than in technological or psychological factors. It is not that Republicans are more gullible, or less rational, than Democrats. It is not that technology has destroyed the possibility of shared discourse for all. It is the structure of the media ecosystem within which Republican voters, whether conservatives or right-wing radicals, on the one hand, and Republican politicians, on the other hand, find themselves that made them particularly susceptible to misperception and manipulation, while the media ecosystem that Democrats and their supporters occupied exhibited structural features that were more robust to propaganda efforts and offered more avenues for self-correction and self-healing. Nonetheless, as we will show in Chapter 6, this media ecosystem still suffers from significant failure modes that are tied to the norms of journalism and the very competitive dynamics that usually protect the system from manipulation. And it was precisely those failure modes that enabled Donald Trump the candidate and his allies in Breitbart in particular to frame the Clinton candidacy almost entirely in terms of corruption. In Part Four, after we describe the actual dynamics of disinformation in the different parts of the media ecosystem, we return to describing the institutional, political, and cultural roots of the asymmetry between the right and the rest, and how the propaganda feedback loop took hold in the right wing of American politics. (p.100)
(1.) R. Kelly Garrett, Brian E. Weeks, and Rachel L. Neo, “Driving a wedge between evidence and beliefs: How online ideological news exposure promotes political misperceptions,” Journal of Computer-Mediated Communication 21, no. 5 (2016): 331–348.
(2.) Mark Hemingway, “Once Again, PolitiFact Struggles to Explain Data Showing They Treat GOP Unfairly,” Weekly Standard, May 31, 2013, http://www.weeklystandard.com/once-again-politifact-struggles-to-explain-data-showing-they-treat-gop-unfairly/article/732009.
(3.) Lucas Graves, “What We Can Learn from the Factcheckers’ Ratings,” Columbia Journalism Review, June 4, 2013, http://www.cjr.org/united_states_project/what_we_can_learn_from_the_factcheckers_ratings.php; Paul Krugman, “Facts Have a Well-Known Liberal Bias,” New York Times, December 8, 2017, https://www.nytimes.com/2017/12/08/opinion/facts-have-a-well-known-liberal-bias.html.
(4.) Glenn Kessler, “One Year of Fact Checking—an Accounting,” Washington Post, December 30, 2011, https://www.washingtonpost.com/blogs/fact-checker/post/one-year-of-fact-checking--an-accounting/2011/12/27/gIQAR1taOP_blog.html; Chris Mooney, “Reality Bites Republicans,” The Nation, May 16, 2012, https://www.thenation.com/article/reality-bites-republicans/.
(5.) We do not compare CBS or NBC because there are too few stories checked on those networks to provide a reasonable basis of comparison. NBC is qualitatively similar to ABC.
(6.) Lou Colagiovanni, “Trump Allegedly Tied-Up 13 Year-Old Girl He Raped, Struck Her In Face,” Occupy Democrats, October 24, 2016, http://occupydemocrats.com/2016/10/24/trump-allegedly-tied-13-year-old-girl-raped-struck-face/.
(7.) Lisa Bloom, “Why the New Child Rape Case Filed Against Donald Trump Should Not be Ignored,” Huffington Post, June 26, 2016, https://www.huffingtonpost.in/entry/why-the-new-child-rape-ca_b_10619944.
(8.) Here, we use top 10 sites, excluding candidate sites and sites like WikiLeaks and Wikipedia, which were used as sources by others, and candidate sites as measured by most linked-to, most Twitter shared, and most Facebook shared over the course of the entire 18-month period under observation. Because the last two sites of the most linked-to sites on the right were World Net Daily and Gateway Pundit, two particularly pernicious conspiracy sites, we also ran the analysis without these two, and these reduced the number of sentences from 199 to 170, suggesting that these sites overcontributed slightly, but not to a degree that qualitatively effects the difference between the right and every other quintile observed.
(12.) Malia Zimmerman, “Flight Logs Show Bill Clinton Flew on Sex Offender’s Jet Much More than Previously Known,” Fox News, May 13, 2016, http://www.foxnews.com/us/2016/05/13/flight-logs-show-bill-clinton-flew-on-sex-offenders-jet-much-more-than-previously-known.html.
(13.) The RSS feed was captured on Media Cloud. The Facebook title appeared as of early 2018 on http://app.viralnewschart.com/News/News.aspx?linkId=174625641; but the URL was no longer accessible at the closing of this writing.
(14.) Breitbart News, “Report: Bill Clinton Flew on Jeffrey Epstein’s Jet Without Secret Service Detail,” Breitbart, May 13, 2016, http://www.breitbart.com/ (p.397) big-government/2016/05/13/report-bill-clinton-flew-jeffrey-epsteins-jet-without-secret-service-detail/.
(15.) Geoff Earle and David Martosko, “Bill Clinton Jumped Aboard Disgraced Sex Offender Jeffrey Epstein’s ‘Lolita Express’ Plane for Junkets 26 TIMES in Just Three Years,” Daily Mail Online, May 13, 2016, http://www.dailymail.co.uk/news/article-3589628/Report-Bill-Clinton-jumped-aboard-disgraced-sex-offender-Jeffrey-Epstein-s-Lolita-Express-plane-junkets-26-TIMES-just-three-years.html.
(16.) Derek Hunter, “Bill Clinton Took Twice As Many Flights On ‘Pedophile Island’ Billionaire’s ‘Lolita Express’ Than Previously Reported,” Daily Caller, May 14, 2016, http://dailycaller.com/2016/05/14/bill-clinton-took-more-than-twice-as-many-flights-on-pedophile-island-billionaires-lolita-express-than-previously-reported/.
(17.) Douglas Ernst, “Bill Clinton Ditched Secret Service on Multiple ‘Lolita Express’ Flights: Report,” Washington Times, May 14, 2016, https://www.washingtontimes.com/news/2016/may/14/bill-clinton-ditched-secret-service-on-multiple-lo/.
(18.) Megan Twohey and Michael Barbaro, “Crossing the Line: How Donald Trump Behaved With Women in Private,” New York Times, May 14, 2016, https://www.nytimes.com/2016/05/15/us/politics/donald-trump-women.html.
(19.) “Woman in NYT Trump Piece Disputes Report,” NBC News, May 16, 2016, https://www.msnbc.com/morning-joe/watch/woman-in-nyt-s-trump-piece-disputes-report-686463555992.
(20.) “MSNBC Live: MSNBCW: May 16, 2016 12:00pm–1:01pm PDT” (MSNBC, May 16, 2016), https://archive.org/details/MSNBCW_20160516_190000_MSNBC_Live/start/600/end/660; “MSNBC Live” (MSNBC, May 16, 2016), https://archive.org/details/MSNBCW_20160516_190000_MSNBC_Live/start/660/end/720.
(21.) Sean Hannity, “Carson: Press Has Shirked Its Duty to Be Honest; Exclusive: Former Bergdahl Platoon Mates Endorse Trump,” Fox News, May 16, 2016, http://www.foxnews.com/transcript/2016/05/16/carson-press-has-shirked-its-duty-to-be-honest-exclusive-former-bergdahl.html.
(22.) Steve Guest, “Bill Clinton To Campaign Near Pedophile’s ‘Orgy Island’ [VIDEO],” Daily Caller, May 16, 2016, http://dailycaller.com/2016/05/16/bill-clinton-to-campaign-near-pedophiles-orgy-island-video/.
(23.) Trent Baker, “‘Morning Joe’ Panel: Clinton Connection With Billionaire Pedophile Jeffrey Epstein Will ‘Blow Up’ Campaign,” Breitbart, May 16, 2016, http://www.breitbart.com/video/2016/05/16/morning-joe-panel-clinton-connection-with-billionaire-pedophile-jeffrey-epstein-will-blow-up-campaign/.
(24.) “Special Report With Bret Baier: FOXNEWSW: May 17, 2016 3:00pm–4:01pm PDT” (Fox News, May 17, 2016), https://archive.org/details/FOXNEWSW_20160517_220000_Special_Report_With_Bret_Baier/start/2100/end/2160.
(25.) “Special Report With Bret Baier,” https://archive.org/details/FOXNEWSW_20160517_220000_Special_Report_With_Bret_Baier/start/2160/end/2220.
(26.) Steve Guest, “Gingrich: Clinton Is ‘A Target-Rich Environment’ Regarding Relationships With Women [VIDEO],” Daily Caller, May 18, 2016, http://dailycaller.com/2016/05/18/gingrich-clinton-is-a-target-rich-environment-regarding-relationships-with-women-video/.
(27.) Trump Refers to Alleged Bill Clinton Assault as “Rape,” http://www.foxnews.com/politics/2016/05/19/trump-refers-to-alleged-bill-clinton-sexual-indiscretions-as-rape.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+foxnews%2Fpolitics+%28Internal+-+Politics+-+Text%29.
(28.) Breitbart TV, “Limbaugh Rips CNN’s Gergen, ‘Clinton Cash’ Critics for Calling Book ‘Discredited,’ ” Breitbart, June 23, 2016, http://www.breitbart.com/video/2016/06/23/limbaugh-rips-cnns-gergen-clinton-cash-critics-for-calling-book-discredited/.
(29.) Hannah Parry, “Billionaire Sex Offender Jeffrey Epstein Once Claimed He Helped Found Clinton Foundation as He Touted Close Relationship with Former President during Plea Bargain Negotiations,” Daily Mail Online, July 7, 2016, http://www.dailymail.co.uk/news/article-3679023/Billionaire-sex-offender-Jeffrey-Epstein-claimed-helped-Clinton-Foundation-touted-close-relationship-former-president-plea-bargain-negotiations.html.
(30.) “BREAKING BOMBSHELL: NYPD Blows Whistle on New Hillary Emails: Money Laundering, Sex Crimes with Children, Child Exploitation, Pay to Play, Perjury,” True Pundit, November 2, 2016, https://truepundit.com/breaking-bombshell-nypd-blows-whistle-on-new-hillary-emails-money-laundering-sex-crimes-with-children-child-exploitation-pay-to-play-perjury/. Craig Silverman at BuzzFeed locates the origin of the story three days earlier, in a tweet and blog post, which were transposed into a YourNewsWire story, all of which preceded the True Pundit story. Craig Silverman, “How The Bizarre Conspiracy Theory Behind ‘Pizzagate’ Was Spread,” BuzzFeed, November 4, 2016, https://www.buzzfeed.com/craigsilverman/fever-swamp-election?bftwnews&utm_term=.esW37XrXv#.oyY5RoPoW. Because the True Pundit story introduced new elements that were those ultimately picked up by Erik Prince, and because its story was tweeted out by Michael Flynn, we treat the True Pundit story as the core source. But we acknowledge that the earlier tweets could have been the source, and if so, given how early and central the True Pundit story was, the basic overarching implication does not change much.
(31.) Sean Adl-Tabatabai, “NYPD: Hillary Clinton ‘Child Sex Scandal’ About To Break,” Your News Wire, November 2, 2016, http://yournewswire.com/nypd-hillary-clinton-child-sex-scandal/.
(32.) Matthew Rosenberg, “Trump Adviser Has Pushed Clinton Conspiracy Theories,” New York Times, December 5, 2016, https://www.nytimes.com/2016/12/05/us/politics/-michael-flynn-trump-fake-news-clinton.html.
(33.) John Hayward, “Erik Prince: NYPD Ready to Make Arrests in Anthony Weiner Case,” Breitbart, November 4, 2016, http://www.breitbart.com/radio/2016/11/04/erik-prince-nypd-ready-make-arrests-weiner-case/.
(34.) “Source: FBI Has Evidence Hillary Visited ‘Orgy Island,’ ” WND, November 4, 2016, http://www.wnd.com/2016/11/source-fbi-has-evidence-hillary-visited-orgy-island/.