Win Bigly: Persuasion in a World Where Facts Don’t Matter

Adams: Good Morning America, and welcome a fake episode of the Real Coffee podcast. I’m not actually your host, Scott Adams. Listeners, today you are in for a genuine treat: an avatar of the distinguished UCLA political scientist John Zaller is here to discuss his seminal work, The Nature and Origins of Mass Opinion. Welcome, Professor Zaller.

Zaller: Thanks, Scott. And thank you for the laughter that Dilbert has provided over the years.

Adams: Professor, your research deals mostly with public opinion. Why do voters come to hold the political attitudes they do? Where do these ideas come from? How do they change? And, most intriguing, how deeply are these opinions actually believed? Essentially, back in the Eighties you took large-scale survey data and compared it to the amount of liberal and conservative media coverage on various issues, and sort of tried to track the impact. And, not only that, you used cognitive psychology to model how media messaging changes large-scale opinion.

It’s definitely not an easy read, thanks to the amount of statistical modeling. But it does a deep dive on two of my favorite topics. On one hand, it comprehensively illustrates the cognitive nuts and bolts of persuasion. And on the other, it describes how information flows from political actors through the media to voters.

And I think that there are elements of 2016 that really demonstrate how robust your Receive-Accept-Sample model is after all these years. But I also think that Trump broke your model in some important ways, and it needs to be updated to account, not only for changes in media, but for Trump himself.

Zaller: Absolutely. That book was published back in 1992. Most of my data came out of the Reagan Era. The media consisted of Dan Rather reading the news at 6pm. That, and local newspapers. So cable news, social media, and the internet have fundamentally altered the media environment. The information that voters use has become both centralized and decentralized, and that has changed how we think about politics in important ways.

Adams: I’d say it’s also changed how we don’t think about politics in important ways too. The core of The Nature and Origins of Mass Opinion is based on a model of political cognition, the Receive-Accept-Sample model. It really looks at how much – or in most cases, how little – attention the electorate pays to politics. As a result, most of our opinions about politics - and the policy that defines our lives - are opinions that are fairly short-lived. To a large extent, they are a result of what we had last heard on the news. And very often they don’t line up with our self-reported values: conservatives would often have liberal opinions and vice versa, and it was always changing depending on how an issue was being discussed in the news.

Zaller: And this isn’t a neat statistical quirk as much as the preponderant trend within the National Election Survey data that I was using back then.

But political attitudes are a lot more coherent now than they once were. The algorithm exposes voters to a much more consistent set of messages. Liberals are now more homogeneously liberal across different issues, and vice versa for conservatives. Ever since 2015, Trump has made everyone – liberals and conservatives – pay attention. And he’s been able to hold a lot of that attention.

Adams: And that’s what I’d like to discuss today.

But let me start by saying how I admire that your model is incredibly non-partisan: it assumes that liberals and conservatives work the same way. We have the same mental hardware and respond similarly to our media environment. In essence, you suggest some kind of cognitive determinism. The opinions we construct aren’t exactly the product of critical reflection as much as the amount and type of messaging we receive. Even for highly aware voters, you suggest that their policy attitudes are pretty much uncritically borrowed.

Zaller: I think the idea has become a lot more accepted now than it was in the 1990s. But it’s still something that we suspect about our political opponents, and not ourselves. The findings, however, suggested that both liberals and conservatives borrowed their attitudes from the media far more than they thought for themselves.

Adams: Well what it does for me, at least, is undermine what I’ll call the “Inherent Deplorability Thesis.” This suggests that Trump supporters, like myself, are mean-spirited bigots, and that everything we think and feel is somehow an expression of some ugly racism. And this is a really, really offensive argument for people like myself who once saw themselves as forward-thinking liberals. But your argument suggests that opinions are less enduring and shouldn’t necessarily be seen as the product of profound belief.

Zaller: There are better ways to describe Trump’s victory without resorting to calling people names. The beliefs that most people hold may be more ephemeral than they suspect, so there’s no need to write off one half of the electorate as irreconcilably at odds with the other. When you study public opinion as long as I have, you come to see that most voters have more in common than not.

My model demonstrates that opinions can and do change in response to the messaging we come across. I think that if we challenged the taken-for-granted assumption of cognitive autonomy, we might be able to see the determinism at work. If we account for this determinism, we can maximize what little autonomy we have.

Adams: Well, this might be our first disagreement. I’m not as confident that knowing will make any difference. In fact, I’ve argued that being aware of our mental operations doesn’t help when it comes to persuasion.

Zaller: Right, I remember reading that in your book. Before I agreed to this interview I sat down and read Win Bigly: Persuasion in a World Where Facts Don’t Matter. And even though I haven’t read or heard any of your other work, I think it’s an interesting take on the 2016 election.

Basically, you suggest that Trump was so good at getting attention and building trust that he really came to dominate the Republican Primary in ways that only a handful of people – yourself among them – actually predicted. You lay out his persuasion techniques, how he was able to draw attention to himself and to frame issues in vivid, easy-to-recall ways. And you sort of suggest that he was able to exploit these socio-cognitive dynamics in a way that no one had ever done in the past. We can call it the “Theory of the Master Persuader”.

Adams: OK, so now we have three sets of ideas. We have the Inherent Deplorability Thesis, which assumes there are deep kinks in the Republican soul. Let’s put that one aside. We have your Receive-Accept-Sample Model, in which our attitudes are a product of media intensity and individual awareness. And you have the “Theory of the Master Persuader”, in which Trump deployed his “weapons grade” persuasion skills. And these are very complementary. Put them together and we can definitely build a more complete understanding of “political cognitive determinism”.

Zaller: True. But I think there’s a better theory of why Trump succeeded. I’ll call it the Robbers Cave Theory. And even though you don’t realize it, the second half of your book – about your own experiences in the 2016 election – illustrates it perfectly, if inadvertently. You describe what it’s like to be personally and professionally attacked. You talk about your reciprocal bond with your audience, and how you developed a sense of “Team Under Siege”. And I think it helps to explain your transformation from “ultraliberal” to a Trump admirer to a Trump supporter to an unofficial campaign strategist to someone who has made some really offensive and regrettable comments.

Adams: Comments that I have discussed at length. Comments that need to be taken in their proper context.

Zaller: Context is social context, and it has a habit of changing without us noticing. And the internet does a great job of creating contextual bubbles where hurtful things don’t seem hurtful until they are.

When I saw you saying those things on youtube, I had real reservations about accepting your invitation. But I also felt I had to come on this show and talk to you and your audience about this. The same process is at work on both sides, and we can’t let ourselves get played so easily.

Even though it wasn’t your intention, your book illustrates how our instincts for threat and team combine in our minds to drive polarization. This has – somewhat suddenly – become the dominant, determining force in modern politics. And there’s only one person who can really explain it fully. Luckily, he’s got some time to drop by at the end of the show.

Adams: You’re bringing in a surprise guest? John, respectfully, this is my show.

Zaller: Don’t worry. You’ve been waiting to meet this guy for a long time. Trust me.

Adams: This just gets better and better.

So let’s step back for a second. We seem to agree that there are cognitive processes which determine how we think about politics. We may think we’re autonomous and that our opinions are our own, but really this isn’t the case at all.

Zaller: Part of this is based on the assumption that we’re cognitive misers. What we call to mind gets there pretty automatically, without a great deal of explicit “out-loud” thinking. Our brains are far more efficient than they are circumspect.

Adams: And you take it a step further by identifying the relative intensity of media messaging, adding yet another dimension of constraint on what is thinkable. You suggest that elite actors are constantly bombarding us with “policy considerations”. Groups like political parties, think tanks, social movements, lobbyists, and the like. Basically any issue group. They’re trying to make us make a mental connection between our values - about security, democracy, capitalism, fairness, order - and their policy proposals.

Zaller: And the elite group that communicates the most effectively is ultimately the one that most people remember when they go to vote.

Adams: OK. Again, we have the implication that we’re really not the disinterested citizenry that democratic idealism holds us to be. In fact, elite domination is sort of an inevitable fact of representative government. And this is something that, once your empirical research was complete, you were deeply ambivalent. As you mention, one of your colleagues sent you out into the woods with a laptop and a bottle of vodka to resolve this question of elite domination.

Zaller: Something like that. My book begs the uncomfortable question of whether our basic mental processes unavoidably predispose us to elite rule.

Adams: But this wasn’t just an academic problem. You had a detailed, statistically-grounded description of how democracy doesn’t work the way that we assume it does. Your postscript addresses the democratic implications of this with The Parable of the Purple Land. And in the Purple Land live Reds and Blues, who have most things in common but evaluate policies differently. Blue people like short, round policies and Reds like their policies tall and rectangular.

And each group hired experts to advise them on which policies their values would lead them to support. And these experts are engaged in a perpetual debate to resolve policy problems. And working with the experts are politicians, whose job is to sell these policies to as many people as they can. And because the Reds and blues are busy with life, work, family, and baseball, they hire journalists to keep them posted on these debates.

You're suggesting that we, the people, don’t really think very hard or independently about politics, and even less about policy. We’ve deliberately outsourced the work of government to experts who make policy, politicians who sell it, and journalists who tell us about it.

Essentially, our attitudes are given to us by elite political actors. And this, as you observe, flies in the face of the idea that we think about politics autonomously. Far from emerging within us, our attitudes are mostly given to us by competing elite groups.

Zaller: And that’s the key. There isn’t a single, unified, political elite. There are several. And they’re competing hard with one another for space in our memory, trying to connect their ideas about policy to our personal values. So in my parable, voters agree with their expert or elite group, or they would if they took the time and effort to read the minutes of congressional hearings. Which no one, including me, wants to do. So it’s not some conspiratorial kind of elite domination as much as a division of intellectual labor: there are people who think hard about policy because it’s their job, and the rest of us.

Adams: And it works in America.

Zaller: It can work, as long as a few conditions are met. First, policy elites have to disagree with one another. As long as think tanks, chambers of commerce, social movements, or parties are proposing different policies, it means that voters will be likely to have their values represented. Next, the media has to clearly communicate this disagreement to the voting public. Finally, the voting public has to pay enough attention to know which policy elites their values should lead them to support.

It’s not ideal, not by any means, but it is roughly adequate. At very least, it means that the public will get to weigh in on policy disagreements, which is essential. Because only the mass of citizens can force its leaders to do the right thing. Otherwise authority gets concentrated and things go wrong. It’s a pretty minimal version of democratic idealism, but it’s really the only one we’ve got.

And as governance goes, it isn’t bad. Cass Sunstein suggested that this was why America won World War II: unlike Japan and Germany, American leaders were subject to vastly greater scrutiny and criticism, and themselves had access to diverse and dissenting opinions, so they had fewer costly disasters. So if other countries start to use algorithmic forecasting to optimize government policy, we have to remember that our advantage lies both in our diversity of ideas as well as our ability to exert our collective will on our leaders.

Adams: So contrary to the assumption that we’re rational decision-makers, voters are pretty ignorant about politics, even when we’re paying attention.

Zaller: The Ignorance Thesis is a little too simple and really misses some important subtleties. It goes back to the Fifties and Sixties, when political scientists like Philip Converse first noticed it. He called these non-attitudes. “Belief has low centrality for the believer” was how he put it. Not only could voters not name their congressman, but many survey respondents had incoherent and contradictory opinions, except on certain issues, especially race. Generally, respondents had liberal and conservative positions mixed up.

And frankly, I noticed something similar in the Eighties. April’s kick-the-poor conservative would be September’s tax-and-spend liberal. Or how if you change the order of survey questions, respondents would swing from hawk to dove. So as I developed my model, I began to realize that ambivalence might be a little bit more accurate than ignorance.

Arlie Hochschild has made a career of talking to normal people about politics. And she noticed that we confidently answer closed-ended, yes-no, attack-withdraw kind of questions. So much so that it seems that voters have fixed beliefs or an “ideology” of some kind. But the confidence really evaporates as soon as you ask for an explanation, because people start calling to mind all of the things they think about an issue and suddenly realize it’s more complex than they thought. We noticed the same thing with open-ended survey questions: voters were suddenly ambivalent on a lot of issues, compared to simple, closed-ended questions. But again, I was using data sets from the Seventies and Eighties.

Adams: So it’s not that voters are irrational, but that their political attitudes are often made up on the spot, usually based on the last thing they heard in the media.

Zaller: Right. What I saw in the survey data is that most voters don’t possess a single “true attitude”, but several that are often in a state of flux.

First of all, it helps to imagine voters - liberal or conservative - spread out on a spectrum of awareness. Some voters pay a lot of attention to politics. Many voters don’t. And policy elites, or experts - the people at the different think tanks or policy institutes or lobby groups or whatever - are constantly trying to persuade voters that their policy connects to voters' values. It’s sort of like a media shower where voters receive elite messages in direct proportion to the amount of attention they pay.

Adams: So this is the Reception Axiom, the first axiom of the Receive-Accept Sample model.

You go on to say that the mind isn’t necessarily a sponge for these ideas. Voters can recognize which messages don’t actually match their values, again in proportion to their awareness.

Zaller: Yes and no. On one hand, highly-aware voters receive the most messaging, and they can reject policy positions that don’t match their values. People who don’t rate high on awareness tend to accept most elite messages, however temporarily. That’s the first part of the Resistance Axiom.

But awareness doesn’t imply a nuanced, cost-benefit understanding of a policy menu. Basically, if you’re “into politics” you’re better at recognizing partisan cues and filtering messages accordingly.

One example I use is the differing levels of support for the Contras back in the 80s. If conservative respondents were asked about funding anti-Communist guerillas in Latin America, they overwhelmingly supported it. But if they were asked about funding Contras, that support evaporated among low-awareness voters. They could connect policy to their values only if they knew what a Contra was, which was contingent on awareness. So partisan cues matter, especially for people who don’t follow politics.

Adams: But one thing I had a hard time with was your insistence that even high-awareness respondents were still dependent on elite messaging. The only real difference was that they were better at recognizing cues and filtering on them. For you, ideology isn’t a coherent set of well-articulated ideas as much as a way to recognize which ideas we should support.

Zaller: That’s about right. Rather than being able to carefully weigh the different pros and cons of policies, highly-aware voters are more reliably attentive to the messaging of their own party.

Less aware voters, on the other hand, are much more difficult to lead because they pay so little attention to politics. In an election, we’re constantly watching emotionally-dense commercials, and often we lack the awareness to filter out what we, according to our individual beliefs, should disagree with. As a result, we more or less believe everything unless we have a reason not to. But we don’t believe it for long, because it’s usually forgotten during the next commercial break.

Adams: I love this quote by Stephen Earl Bennet about your model:

People of moderate political persuasion are “persuadable,” in this model, not because they are thoughtful, but because they are so ignorant of the positions to which their predispositions should lead them that they can easily be misled. Instead of walking in a straight line, they are buffeted by whatever fragmentary elite messages they happen to have heard in the media. The most ignorant members of the public may (sometimes) walk a straighter line, but only to the extent that they are completely uninformed about elite views that would connect their predispositions to issue positions. When a new issue arises, the ignorance of the least informed is liable to make them walk off a cliff.

Zaller: Again, this isn’t really a function of intelligence or ability. It’s a question of attention and awareness. The cognitive research was pretty unequivocal when it suggested that we automatically recall the ideas that we heard most recently. And that’s what the Accessibility Axiom is based on.

Adams: So parties and candidates are, quite literally, jockeying for space in our memory. If they’re the last thing we hear, we’ll remember them when we go to vote. You suggest that many of us are so overwhelmed by the information deluge that we often make up our minds as we cast our ballots.

Zaller: If you look at the 2000 election you’ll notice that Gore had outspent Bush in advertising during the last month of the election, and he had gained a small but significant lead in swing states. But during the last week, Bush outspent Gore. For every 20 Gore ads that voters saw, they saw 25 Bush ads. Because many undecided voters have short memories, this last-minute spending may have allowed Bush to catch up and make the Florida Recount matter. I think we wrote that “votes cannot be bought, but they can be rented.”

Adams: Again, we come back to the question of whether “ideology” actually matters. Are the voting public‘s political beliefs as profound as we assume? Like you suggest, Bush’s election may not have been the renewal of American Conservatism as much as a successful last-ditch fundraising drive. Maybe we need to question how deeply we believe in the stuff we think we do.

Zaller: There’s an idea that attitudes exist “a priori of concrete social interactions, like objects in a drawer that one can take out at will.” I like to think of attitudes as cobbled together in real-time, whenever we need them. The final component of the model, The Response Axiom, goes on to state that we do a sort of weighted average across the different liberal and conservative messages that are called to mind when we think about an issue. And this means that immediate contextual variables – like question order or wording or an intensive media campaign – make a big difference to support for a given policy. The example I used was asking Americans if it was okay to let journalists from enemy countries cover our politics. And most people say no, unless you ask them if it’s okay for American journalists to cover their politics first. Then they are much more supportive.

Adams: Sort of like that experiment by Thomas Gilovich, the one where poli sci students had to read a description of a foreign conflict and then suggest whether America intervene. If the description subtly resembled World War II, they would probably remember Munich and be more likely to recommend intervention. If it had subtle references to Vietnam, they would remember the fall of Saigon and be less likely.

Zaller: Or how holding elections in schools creates support for education funding. But it does explain the persistence of attitude instability. I know that I fulminate high gas prices while I’m filling the tank, but fret about global warming when I hear about hurricanes and forest fires. I complain about taxes and gripe about school cutbacks on the same day, and I usually don’t notice.

Adams: But voters don’t see it that way, subjectively. It feels like a genuine attitude, not something we just made up at the moment. Something that’s a little different each time.

So let’s see if I have this right. The basic idea is that we make up our political attitudes as we need them. The Awareness Axiom posits that there’s a spectrum of awareness, which determines how much elite messaging we get about politics. There’s the Resistance Axiom, which suggests that we can filter elite messages based on our awareness of partisan cues. The less aware you are, the more likely you are to believe whatever you hear. The Accessibility Axiom suggests that – to some large extent – the message we heard most recently shapes our opinion the most. And the Response Axiom means that what we call to mind has a lot to with how things are framed for us.

Zaller: And it all comes together in the Ambivalence Deduction, which is less about ignorance and more about confusion. Many voters, if not most, are literally making it up as they go, based on what they last heard. Again, here we are – or were, back in the Eighties – with our busy lives and we’re often overwhelmed by political messaging, especially at elections. This explains how people hold contradictory positions from one day to the next, or how opinions are not coherently liberal or conservative. It also suggests that the most informed are caught up in a “spiral of conviction” and end up with very different attitudes than a poorly informed member of their own side.

Take Vietnam, for instance. Early in the conflict there was only moderate support from poorly informed liberals. More attentive liberals – the ones who had learned the Johnson Administration’s arguments for escalation – they were the ones that supported the war. But as the war progressed and things got uglier and it was obvious that there was no “light at the end of the tunnel”, that’s when the more attentive liberals began to lose support, as key Democratic figures came out against the war. By this time, however, less-informed liberals had learned the new party line and were likely to support the war. Not surprisingly, Republican support remained pretty consistent, because Republican elites were consistently hawkish.

The same was true during the Persian Gulf War in 1990: it’s striking how polarization is most notable among highly informed Republicans and Democrats, but much, much less so among low-awareness voters who tend to ignore elite disagreement.

It holds for domestic issues as well. I went back to the National Election Studies in the seventies, and found that higher-awareness voters are consistently in line with their partisan elite. When both parties agree on, say, admitting China to the UN or equal seating for black people in restaurants, the public is largely in agreement. But for more controversial issues, like bussing, job guarantees for black people, or if Vietnam was a mistake, there is marked polarization among the more attentive liberals and conservatives.

Adams: I have to say, The Nature and Origins of Mass Opinion is a monumental accomplishment. It actually puts empirical flesh on the bones of Antonio Gramsci’s “War of Ideas” thesis, that power in society comes from the most articulate elite actors.

Zaller: I’m afraid I’m not familiar with him, but yes, that does seem to fit.

And it is worth noting that the model has limits. Monica Lewinsky blew a hole in it early on. The Republican Party really went after Clinton over that scandal. They tried to have him impeached, but voters couldn’t care. There was an economic boom, and conservatives felt rich enough to look the other way. His support dipped and then increased, what I called the “Lewinsky Bump”.

So it’s not a true and perfect model, but it is – or at least it was – a plausible approximation about how we get political information and turn it into the attitudes that guide our votes.

Adams: But it really puts elite actors at the center of political life in a way that people who “believe in” democracy – people like ourselves – should be concerned about. And it lends some truth to the adage that “the best argument against democracy is a five-minute conversation with the average voter.”

Zaller: The best argument for democracy is that it can work. Politics is driven by “intense policy demanders”. These are groups that want things - low taxes, less abortion, civil rights, gun control, more military spending, farm subsidies - and they try to use the government to get it. Parties are, in reality, coalitions of these groups. And the only thing that prevents any one group from dominating the others is the fact that the public decides. Elite domination is a useful division of labor, but only as long as voters understand the disagreement and weigh in on it.

Adams: And I think that there’s some common ground between us. For instance, I accept elite domination as long as it is roughly in line with the will of the electorate. I really don’t think that voters are great at determining policy - at all - and we need experts and elites to define our political reality for us.

Zaller: I believe you refer to it as the Moist Robot Reality filter. We go around assuming that we’re 90% rational and 10% irrational, but you suggest it’s the opposite.

Adams: Well, as a trained hypnotist, it’s easy to see human behavior as 90% irrational. We use what little brainpower and information we have to create our arbitrary realities. So our political realities aren’t real, they’re just stories that we tell ourselves.

And Win Bigly builds off this point. Elites really define our political world for us, and they compete. And Trump won because he had weapons-grade persuasion skills and the competition didn’t. He made the 2016 Republican Primary about himself, and then he was able to manipulate the media to beat Hillary. And we’ve been transfixed for almost a decade.

Trump made the RAS model kind of obsolete in that he eradicated ambivalence. He made everyone, or almost everyone, have an opinion. Contrary to the notion that he is either an imbecile or the second coming of Adolf Hitler, Trump is the most persuasive individual we will ever live to see. I saw it early in the Primary.

Zaller: Right. And you say that his Rosie O’Donnell remark was your Come to Jesus moment.

Adams: Like I said, it was a masterstroke of persuasion, timed perfectly, and executed in front of the world. Here’s Megyn Kelly asking him an unanswerable question about all of the horrible things he’s called women over the years. And he cuts her off with “Only Rosie O’Donnell.”

When I heard it I got goosebumps. I was so stunned that I actually got out of my chair and walked toward the television. While almost everyone else took it to be either crass or funny, I knew that it was genius at work. His response was so many things: funny, smart, memorable, visual, thick skinned, provocative, and perfectly on-brand. Not only did he connect with potential supporters on an emotional level, he totally redirected the attention from his past remarks to the problem of political correctness.

Zaller: As you say, “He converted Kelly’s attack into pure energy and then moved that energy where it suited him best. Normal people can’t do that. They wouldn’t even know where to start.”

Adams: Then and there I knew that this guy had a “reality distortion field”. I knew that he was going to walk away with the Republic Primary and possibly the presidency. He may not have been the smartest or most qualified Republican candidate, but he was by far the most persuasive. He was able to capture attention, direct it all towards him, build trust with his supporters, and not get taken down by the stream of bad publicity.

He knew how to make himself memorable. The Rosie O’Donnell moment was one example, but I think the Border Wall illustrates it best. A “big, beautiful wall” is simple. It’s a concrete object that you can actually imagine, especially when you compare it to a concept without an image, like immigration policy. Policy quickly becomes technical and legal and historical. It’s abstract and boring and has a tenth of the emotional resonance.

Trump used the wall to turn sympathizers into supporters. He deliberately leveraged the power of unconscious association. He didn’t care what people knew about immigration, because facts are weak persuasion. He cared what Republicans felt about immigration, and he wanted to connect himself to those feelings. He unequivocally confronted their main anxiety, so that he came to mind first whenever the issue came up.

What most commentators didn’t understand at the time – and won’t admit now – is that Trump deliberately manufactured controversy to get all the attention focused on him. And he did it so elegantly. He played the media like a violin. The controversy that The Wall created – particularly the cost and the impracticality – made it even more memorable. The more it was discussed, the more people were forced to picture it in their minds, the more real it became in their memory, as part of their story. It didn’t exist in concrete reality, but all of the discussion made it mentally real, which was more important.

Zaller: I believe you refer to this technique as Thinking Past the Sale.

Adams: Better yet, he was deliberately wrong and it kept him in the spotlight. Had he admitted to the cost and obvious impracticality, it would have ended the controversy. It would have ended the fixation. Instead, he stuck to his guns and the story was brought into our conscious attention again and again. He let the media - including hostile media - fill in the blanks with mock-ups and drawings, which made the story ever more vivid.

Zaller: You also note that he avoided details, which can be criticized: “But there is one kind of wall that is hard to criticize: the one that is entirely different in each person’s head.”

Adams: It shows the power of a vivid idea that is vague enough to elude specific criticism, but perfectly matches the emotional tone of the people he needed to persuade.

Zaller: I know that my model focuses on recency and frequency of a political message. But your model of political cognition - The Theory of the Master Persuader - seems to prioritize vividness and emotional resonance. I think that it’s a natural complement to the Receive-Accept-Sample model.

He really changed how we receive political messaging. Social media likes outrageous, attention-grabbing content. Trump deliberately provided a non-stop stream of it. Together, they turned millions of low and moderately aware voters into active participants. More people heard more political messaging, and the messaging they heard was centered on Trump himself. As he came to dominate the news cycle, all other Republican contenders were reduced to unimportant has-beens or potential Secretaries of Health. He was just way more interesting and entertaining than anything to do with Chris Christie.

Adams: He also got the Accessibility Axiom to work in his favor. With all of the controversy, he was always the most recent thing on people’s minds, and the most easily accessible mental association.

Zaller: So you might argue that he eradicated ambivalence with a vengeance. Which would, quite fairly, render the Ambivalence Deduction obsolete: suddenly people were paying attention and knew what side they were on, even if they were still hazy on policy specifics.

Adams: This is the key point. Trump did not win because he proposed excellent policy. He didn’t. He seemed to come up with ideas on the fly. That’s why he kept contradicting himself on the details. But it didn’t matter, because policy is weak persuasion.

Trump made Republicans trust him by matching the emotional state of his base. He demonstrated the most important part of leadership: “to unambiguously confront the anxieties of those you lead.” Maybe his suggestions on international security were “unconstitutional, impractical, evil, or something the military would refuse to do.” Maybe his ideas were crazy, but it got people’s attention and – for many – matched how they felt. This activated a similarity bias. Robert Cialdini, the world’s leading expert on persuasion, referred to it as a form of liking. It creates trust, which is the willingness to accept ideas automatically. I call this Pacing and Leading.

Zaller: So Trump fundamentally altered the Resistance Axiom. He became the singular partisan cue. The only issue. He reduced politics to a simple binary: you’re either with him or against him.

Adams: The same way that he built trust with his base, he used Linguistic Kill Shots to paint his opposition as untrustworthy. Again, he used visually compelling pejoratives: Lyin’ Ted Cruz, Lil’ Marco Rubio, Low-Energy Jeb, Crooked Hillary. This was a great play on so many levels. It drove a wedge between his contenders and their supporters. It took the heat for being a liar and a crook off of him and transferred it to his opponents. And it primed us for future confirmation bias: Jeb Bush could do an Iron Man tomorrow and I’d still think he was frail.

Zaller: Nevertheless, I don’t think other candidates would stoop that low.

Adams: I’m not saying you’re wrong. But like all the other criticism, it never stuck. He spoke with conviction, and he never apologized or admitted to error. This gave his base the “permission” they needed to assume that he wasn’t actually wrong or at fault. He helped them to deploy confirmation bias and to resolve dissonance in his favor.

And he was excellent when it came to redirecting controversy. He could reframe it to whatever he wanted to and move on. It’s what I call the High Ground Maneuver.

Zaller: Like when he was asked about the Pope’s criticism of capitalism. He could either side with the Pope and look like a liberal, or side with capitalism and offend Catholics. And instead he goes off talking about how the Pope should be personally grateful to him for protecting the Vatican from an ISIS invasion.

Adams: It was genius. He didn’t equivocate or apologize or look weak or dwell on the problem. And in the back of every mind was a desire to see Seal Team Six in a shoot-out with ISIS at St. Peter’s.

Throughout the campaign, he just kept moving on, and kept the news cycle moving with him. And this is what made him the Teflon Man. One outrage out of three headlines in a week is bad persuasion. But twenty-five outrages out of twenty-five headlines in a week is excellent persuasion. Voters can’t think deeply about any one of them because there are so many.

Unpredictability kept us looking. Predictable things fade into the background so that we don’t waste energy paying attention to them. But unpredictable things remain visible, and we cannot help but talk and think about them.

Zaller: You compared it to a bed of nails: the controversy was so distributed that it couldn’t hurt him. “He’s creating so many opportunities for disagreement that it’s mentally exhausting. Literally. He’s wearing down his critics, replacing their specific complaints with entire encyclopedias of complaints. And when Trump has created a hundred reasons to complain, do you know what impression will be left with the public? He sure got a lot done. Even if you don’t like it. In only a few days.”

Adams: He “flooded the zone with shit”, purposely. He stirred the pot and created a chaotic environment in which people craved clarity. We had to rely on our biases instead of thinking things through. I love the way Robert Cialdini put it in his book Influence:

With the sophisticated mental apparatus we have used to build world eminence as a species, we have created an environment so complex, fast-paced, and information-laden that we must increasingly deal with it in the fashion of the animals we long ago transcended.

Zaller: That’s Robert Cialdini, professor emeritus of psychology at Arizona State. He was a recurring theme in your book.

Adams: Absolutely. He’s forgotten more about persuasion than anyone will ever know. I call him “Godzilla” in the book, because he can personally make or break campaigns. He was behind Bernie’s “America” ad, easily the best of the 2016 cycle. And once Bernie dropped out, he went over to Clinton and helped her craft her best persuasion offenses against Trump.

Zaller: Specifically that Trump was “dark”.

Adams: Amazing use of unconscious association. Cialdini’s books are simply packed with practical demonstrations of persuasion at work. He really does play three-dimensional chess.

Zaller: Well, I asked him to join us today to round out the model of political cognition that we’re discussing.

Cialdini: Hi John. Hi Scott. Thanks for having me.

Adams: Oh my god! It’s Godzilla himself! In my kitchen! I can’t believe this! Welcome (instantiation of) Professor Cialdini. It’s such an honor.

Cialdini: Well, first let me say thank you for your kind words over the years.

And thanks for quoting me on how we’ve made our world too complex for our limited minds. John’s book demonstrates that being politically savvy has always been hard. And that was in the pre-internet days. The overwhelming “shower” of information has become a firehose. Trump made it even more intense, creating such a chaotic media environment that sustained attention was almost impossible. Nevertheless, our attitudes still feel very genuine.

It’s easy to underestimate the number of mental operations that go into thinking hard. For starters, we have to detect and control our most subtle and fleeting thoughts. And then we have to pay attention to them for an extended period of time, which is both difficult and boring.

Adams: It’s an intentional act of will.

Cialdini: Precisely. But we also have to avoid suppressing inconvenient contradictions. We must force ourselves to dwell on feelings of dissonance and ambiguity, instead of distracting ourselves. Our minds automatically marshall confirming opinions when we’re faced with disagreement.

Adams: But we almost can’t detect it.

Cialdini: It’s hard work to detect and scrutinize the half-substantiated factoids we dredge up out of our memory.

Adams: Those things we use to justify beliefs in order not to think about them.

Zaller: And then we may also have to admit that what we remember is, in fact, partly fabricated, despite the fact that we can see it clearly in our mind’s eye.

Cialdini: We also have to admit that there’s a lot we don’t know and cannot imagine, what Donald Rumsfeld once called “Known Unknowns” and “Unknown Unknowns”, respectively.

Adams: As well as things we take for granted without knowing it, the “Unknown Knowns”.

Zaller: And at the end of all of this, we may come to realize that our attitudes were wrong, so we may have to deal with shame and guilt, which is very, very hard to do. Especially if our online world is full of specious affirmation that lets us feel better about bad choices.

Adams: So we all agree that sophisticated political thought is next to impossible, especially in this media environment.

Cialdini: The problem, Scott, is that it applies to all of us, especially for those who think it can’t be them.

Adams: Absolutely. I couldn’t agree more. Trump short-circuited hard thinking by being so visual, memorable, and overwhelming.

Cialdini: What about you? Were you under the spell of the Master Persuader?

Adams: Hell no. I could see the persuasion for what it was. Did I mention that I’m a trained hypnotist? And I’ve read your books, Dr. Cialidini. That’s why I was able to call Trump’s win so far out. So no, I wasn’t under his spell the way everyone else was. But I certainly admired his talent stack.

Cialdini: So you double down on the Theory of the Master Persuader, but you miss the key mechanic in this whole process. Your book is the neatest illustration of it, and you don’t even know it.

Adams: I really don’t see where this is going.

Cialdini: The Receive-Accept-Sample model and the Theory of the Master Persuader explain a great deal, but they don’t explain the intensity of the 2016 election cycle as well as the Robbers Cave Theory. You remember the Robbers Cave Experiment? Back in the fifties, a psychologist named Muzafer Sherif took two groups of boys to a summer camp in Oklahoma.

Adams: Of course. Not without its problems, but still a classic.

Cialdini: Right. So you remember the first week, each group didn’t know of the other group’s existence. And the campfires and swimming are fine, if a little boring after a while. Then at the end of the first week they suddenly find out that another group is there, and immediately there’s this change.

Adams: As I recall, they gave themselves cabin names – the Eagles and the Rattlers – and made flags and challenged each other to a baseball game. They suddenly became more competitive.

Cialdini: Right. So Sherif makes the entire second week an intense, high-stakes tournament. Tug-of-wars, football, baseball, tent-pitching contests, one after another, all for some pocket-knives and a trophy. It didn’t take long for the friendly rivalry to turn into wariness. The cabins turned inward, and everything was about the next competition. They forgot about everything else. And soon there was this deepening animosity. After another epic tug-of-war, they just abandon any pretense of civility. Rude songs were sung, insults hurled, flags burned, food fights engaged, and soon losing a baseball game became a crushing defeat.

Adams: And then it devolved into Lord of the Flies.

Cialdini: That’s right. A rumor of cheating quickly became an established fact, and the Eagles trashed the Rattlers’ cabin. Sherif’s staff had to physically intervene to prevent fist-fights.

So at the beginning of the final week, these cabins want absolutely nothing to do with each other. But then the camp’s water supply stopped unexpectedly, and both the Eagles and the Rattlers had to work together to find the jam and unclog the faucet. When the staff somehow didn’t have enough money to rent a movie, the groups decided to share the cost equitably. When the camp bus mysteriously got stuck in mud, the two groups joined forces to “play tug-of-war against the bus” in order to get it out.

Adams: And the boys end up friends at the end, when they had a common goal. It’s an inspiring story. I’m not sure if it’s all that realistic for the 2024 election cycle.

Cialdini: I think Sherif was trying to disprove the Inherent Deplorability Thesis, way back then. Here’s a guy who had just lived through two world wars and is like “Why are people always fighting?” And he wanted to know if aggression was simply an unalterable part of our psyche, like Freud thought, or if it was more a function of our social environment. If it was, then maybe we could do something to prevent escalation.

Adams: OK, I see where you’re going. Sherif sort of demonstrated that we’re not born hostile_._ Instead, we’re pretty cooperative by nature. But when there’s a threat, we react.

Cialdini: Nothing unleashes the narrow stupidity of human intelligence like conflict. It elevates our capacity to learn, cooperate, and tolerate hardship. We are motivated to think hard and creatively, albeit in the most bounded ways possible. There is no higher purpose than the defense of the collective, and we often rise to the occasion with everything we have.

The Team is a mental shortcut that obviates the hard work of thought. The Team makes some people implicitly trustworthy and others inherently suspect, often on rather flimsy pretenses. It makes us partial to some ideas while rejecting others. It makes us far less critical of our leaders. And it makes us suppress doubts so that ideas become homogeneous.

Zaller: Leaders need conflict, particularly leaders who are weak on policy. People are easier to lead when they’re scared and angry. And if you’re able to manufacture a compelling threat and then offer yourself as the only solution, your supporters support you even more.

Cialdini: But Trump was no Hitler. In fact, he used the opposite strategy. Trump’s talent was not in motivating his followers, but in enraging his enemies. And it worked: for each ounce of contempt he elicited, he gained a pound of support. He triggered righteous indignation in liberals, which he then masterfully used to trigger conservative reactance - that righteous anger that says “Don’t call me bad! Don’t try to control me!” At least, that’s what I got from Arlie Hochschild’s masterful Strangers In Their Own Land.

Zaller: His stunning tweets and comments were the perfect way to direct all of the attention to himself. For liberals, he was so incredibly vulgar and insulting. For everyone else he was so entertaining, so new, so fearless, and above all, so arousing. The more attention he got, the bolder he became, the more attention he drew. Soon, everyone was tuned in to politics, everyone had an opinion, and everyone took a side. Every day cable news got more intense, and Facebook feeds became more shrill and damning, and each side saw the other becoming more menacing. And he did it again and again, creating more indignation and resentment.

And whenever Trump was pilloried for being outrageous, he became a martyr. He suffered for the Team, and became even more credible. To quote Professor Hochschild, once again.

Cialdini: And it relates directly to John’s Resistance Axiom. Conflict makes our own Leaders trustworthy. Whatever they say is implicitly true. Their flaws are forgivable. Their policy proposals may not make a lot of sense, but we’re willing to overlook this. Conflict also makes everyone else – especially the leaders of the other Team – deeply suspect. They’re an easily recognizable partisan cue.

Zaller: The greater the outrage on both sides, the more voters were willing to accept what their leaders told them: that anyone who didn’t trust Hillary was a misogynist white supremacist, and anyone who did was a condescending elite, sailing in their yacht at Martha’s Vineyard, far from the poor they pretended to represent. Liberals saw Trump’s support and figured that Republicans were bigots. Republicans resented being called bigots and doubled down on Trump, as Arlie Hochschild and Irshad Manji illustrate.

Cialdini: The best evidence we have of the Robbers Cave Thesis is your book, Scott.

Adams: Wait. What? My book lays out the Theory of the Master Persuader, not the Robbers Cave Thesis.

Zaller: Think about it. Prior to the election, you were a self-described “ultraliberal” who had grudging admiration for someone with Trump’s talent stack. And you took a lot of liberal heat for going public with this contrarian opinion. Worse, you predicted he would win, so you identified yourself with him.

And as the online conversation got angrier, you got more flack from fellow liberals. More importantly, your audience begins to change: liberals stopped reading your blog, while Trump supporters started to. And they agreed with you even as Democrats and liberals start attacking you personally as well as professionally.

Adams: I lost all of my speaking engagements, which was my primary source of income. It wasn’t a huge issue, because I’ve got plenty of fuck-you money and can say whatever I want, but it still hurt. But you’re right, there were a lot of truly awful things said about me. Partway through I ended up endorsing Hillary out of fear for my own safety.

Zaller: Until she released her death tax proposal. That, along with all of the acrimony, made you endorse Trump again. And not only did you endorse him, you started putting campaign advice on your blog. And his campaign seemed to be following it. You became an active member of Team Trump.

Adams: You’re damn right I felt like I was under siege. Clinton supporters’ attacks were bullying, plain and simple. And Clinton’s death tax was really the final straw. I saw that as theft.

Zaller: Until the Access Hollywood scandal.

Adams: Whatever. Bill Clinton said worse to Trump on the golf course. Anyways, after a couple of weeks the Clinton camp’s bullying was too intense. Not just for me but for the rest of us “deplorables”. The contempt and hostility was too much. Naturally, I re-endorsed Trump. And on Election Night, it felt great to be on the winning side of a Cinderella story, to have persevered in the face of adversity and triumphed.

Zaller: No doubt it felt that way, subjectively. The key, for me, is how you describe the change in your audience, which became steadily pro-Trump. You started writing for a group of people whose dominant emotion was resentment. And they started making you feel important and validated. You shared feelings of fear, anger, and hope. You were comrades in an algorithmic echo chamber. As Robert emphasizes again and again, reciprocity is one of our strongest motivating forces. You had to stand by each other. In essence, you belonged to each other. It was Robbers Cave all over again.

Cialdini: The Theory of the Master Persuader turns on confirmation bias and dissonance reduction. And what’s slightly ironic, I guess, is that your book inadvertently shows how easy it is to resolve dissonance in favor of the Team, rather than admit you’re wrong.

Zaller: You fully pledged yourself to Trump on an emotional level and used an elaborate post-hoc rationalization to deal with the dissonance. You had to rationalize your irrational support for Trump. Like how his sincere dishonesty was just part of New York’s culture.

Cialdini: Or how you write off Trump’s slow denial of the KKK with the line “You can make up your own mind about the ethics of using bad people to get good outcomes.” Seriously.

Zaller: Or how Clinton was deliberately misleading while Trump was merely persuasive. Or Trump’s campaign wasn’t openly hostile, and unlike Clinton he “won’t bully any American citizens who were minding their own business.” This isn’t to endorse Clinton, but to point out a very obvious double standard.

Personally, I think the most galling thing for me was how you started the book with a moving tribute to “We, the People” and by the end you’re saying that voters are too uninformed to decide anything. By pretending that policy doesn’t matter, you could pretend that Trump’s dishonesty is a style, and a clever one at that. Your book is a tribute to cynical manipulation. And you’re okay with it because you were on the winning side. You were so proud to show all those bullies that you were right all along.

Cialdini: Take it easy, John. Remember the key lesson here: we don’t think hard when we feel threatened. It’s harder to reconcile when we’re angry.

Scott, while I was waiting outside I heard you say that knowing how manipulation works doesn’t prevent it. I think that’s true of individuals, but not of groups. Good manipulators are so subtle we can’t perceive it at work on us, in real-time. What we can do is help prevent each other from being taken advantage of by clever manipulators. We can’t see it in ourselves, but we can see it happening to each other. As Edward Wilson said of social evolution: “selfish individuals beat altruistic individuals, while groups of altruists beat groups of selfish individuals.” It’s kind of ironic that autonomy is a collective effort.

What I am asking you to do is find someone who you trust, and who is totally unwilling to provide specious validation. And tell them what I told you, and you ask them if you’re right or merely deploying confirmation bias and dissonance reduction on Trump’s behalf.

Adams: Well, you’re someone I trust. So I have to take it seriously.

Cialdini: That’s a start. If we understand how this cognitive determinism works we can see it in one another. We can help each other recognize it. More importantly, we can avoid triggering it in each other. We might be able to retard the escalation somewhat. Maybe we could confront our own dissonance and get control of this Stone Age tribal defense mechanism.

Zaller: But what if a significant portion of the electorate enjoy being triggered? Think about it. Trump made our opinions matter. He made both liberals and conservatives belong to something urgent. He aroused the intense feelings that are few and far between. Maybe Trump gave voters what the safe and orderly world of pro sports and urban anonymity couldn’t: that deep sense of us who are good and right.

The real issue is social media. Without it, Trump would have remained a wittier, wealthier, more cynical, and less electable Ronald Reagan. But social media companies are like mosquitoes who have struck a major artery. They have found a formula that we cannot resist: sort us into groups, capture our attention with outrage, and then manufacture a threat so clear and present that it connects us to one another. Even though the threat is artificial, the fear and anger it inspires within us is very real.

We have an impulse for solidarity that is easily exploited. We can be made to feel threatened, and to fixate on differences that are more apparent than real. This leads us away from policy solutions to our urgent problems. For reconciliation to be possible, the algorithm must die. I really don’t think there’s another way.