The Enigma of Constancy: The Resilience of Trump’s Base


Todd Gitlin

What follows is part two of Todd Gitlin’s deconstruction of the Trump phenomenon in American culture and politics. Part one appeared in our Fall 2019-Winter 2020 issue, number 204-205, under the title “For The Love of Sin: Toward An Understanding of Trump’s Base” (pages 67-94). A version of this essay with full footnotes appears on the Salmagundi Magazine website:

Since the day in June 2015 when he slithered down the escalator into the Trump Tower atrium to announce his candidacy, Trump’s approval ratings have been consistently low—never more than 45 percent—and relatively invariant. Either fact by itself is remarkable. The combination is amazing.

In November 2016, Trump won 46.1 percent of the popular vote. From his inauguration onward, his weekly approval rating, according to Gallup, wobbled between 45 percent and a low of 35 percent in August 2017, shortly after the white supremacist riot in Charlottesville, Virginia. In December of 2018, 82 percent of Republicans approved of Trump. But only 26 percent of Americans identified as Republicans. Independents who approved of Trump made up another 14 percent, and Democrats who approved of him, another two percent. In other words, about 37 percent of the adult population overall approved of Trump—precisely within the range measured by the surveys.

During his first fifty-four months in office, Trump’s approval ratings were confined within a ten percent band. His popularity with a substantial minority—though never a majority—was the most stable thing about him. Normally, public opinion toward presidents wavers far more widely, depending on what is going on at the time. Events intrude. Presidential policies and statements matter considerably. Accordingly, all presidents from Harry S. Truman through Barack Obama have seen their approval ratings fluctuate wildly. Truman’s approval fluctuated as much as 54 points; George W. Bush’s by 39; Ronald Reagan’s by 31.

The Trump anomaly is how little his approval has fluctuated. Unlike his predecessors, Trump started off without a honeymoon. By Gallup’s measure, his largest one-month rally was three percent, in September-October 2018, during his fight to confirm Supreme Court Justice Brett Kavanaugh. A three percent rally was also a record low seen during any administration since Truman’s. According to Gallup’s Jeffrey M. Jones, “The 10-percentage point range in Trump’s approval is the smallest for any president during his first two years in the Oval Office by a significant margin.”

This fact begs to be reckoned with. Anyone who steps outside the right-wing echo chamber knows that he has lied, misled, and half-truthed thousands of times, and been exposed, and proceeded to lie about his lying. The bulk of Trump’s followers don’t care. His approval is largely impregnable. There is no way to tell whether the followers are the same individuals from week to week, month to month, but on balance, no matter how many abandon him, a more or less equal number join his team. Come what may, at least 35 percent of the country remains in his embrace. He is their Trump right or wrong. How is such consistency possible?

Some of it is, no doubt, lockstep partisanship. Partisanship is a hallmark of this era. Having chosen a party, partisans stick with it. Gallup’s Jeffrey M. Jones notes that, on average, during the first two years of his presidency, 85 percent of Republicans approved Trump, as against 8 percent of Democrats. That yawning gulf of 77 points in party preference exceeded even the average under Barack Obama (83 percent Democratic approval as against 13 percent Republican). Party loyalties have hardened into identities that are firm and getting firmer. So it is that, for example, Democrats and Republicans avoid dating each other, don’t want to be neighbors, and don’t want their children to (as it were) intermarry. The two parties do not clearly inhabit distinct territories; there has been neither a legal secession nor a military combat. So there is no civil war. But what has taken place is that the two parties diverge radically not only about their conclusions but about the considerations they bring to bear to reach those conclusions.

Journalists gravitate to single-factor theories of why elections turn out the way they do. So does public opinion in general, and 2016 was no exception. One theory, heavily favored in mainstream media, is that what put Trump over the top was a certain economic rationality, a reaction against economic stagnation—in other words, a kick in the shins to “the system,” a rejection of “the establishment”; in particular, against policies that consigned working or middle class voters to economic marginality over recent decades, and against the groups that promoted those policies. Trump voters were estranged because they were ground down and neglected. But the notion that economic opinions elected Trump doesn’t hold up. Contrary to run-of-the-mill journalism, the public at large was not especially angry in 2016. It was Republicans who were angry. According to the political scientists John Sides, Michael Tesler, and Lynn Vavreck, whose book Identity Crisis is the most thorough analysis yet published, “2016 stood out not because voters were angrier but because their improving views of the economy had not much affected their views of Obama and the country.” Trump succeeded in making 46.1% of the voters—strategically located—think of themselves as aggrieved whites.

The research tells us that Trump supporters in 2016 were no more anxious economically than others. The median Trump supporter reported an income “right around the median income for American households overall.” “Trump support was only weakly correlated with whether respondents were worried about losing their own jobs.” Protest voters were voting not against economic policy that had disadvantaged them and their families but against the false view that during Barack Obama’s presidency, racial and ethnic minorities had received benefits that deserved to go to whites.

For this view the case is clear: Trump blared out racial resentment. “[S]upport for Trump was tied most strongly to white grievances: views of immigration, Muslims, and blacks; and liberal views about economic issues. These factors, more than economic anxiety, helped explain Trump’s surprising path to the Republican presidential nomination.” One may characterize Trump’s vote as a protest vote, but a protest vote that stands solid, first, through a long nomination process, and then through a long general campaign, only to segue into unprecedentedly consistent support for a president. This is more than a protest vote—it is a declaration of identity. The question is: How substantial is this identity, and how impermeable?

In particular, is it impermeable to verifiable truth? Do those who stand fast in the bedrock 35 to 40 percent of the population who constitute Trump’s rock-bottom base think that he tells the truth? Not necessarily. So there remains an enigma I highlighted in the first part of this essay: what exactly do his supporters approve, and why?

How well informed are those who support Trump? In late November and early December of 2018, the Washington Post Fact Checker staff asked a national sample to evaluate a series of factual claims without stating whose claims they were. Eleven of these claims were falsehoods to which Trump had subscribed. They included the claim that global warming has natural causes; that the US supplies a majority of NATO’s budget; that North Korea has done more to end its nuclear weapons programs in six months under Trump than in the previous quarter-century, that Democratic Senators support an “open borders” immigration policy. They also included two claims about the 2016 election: that evidence shows that millions of fraudulent votes were cast, and that Russia did not interfere. In all, only 41 percent selected the true responses. To put it another way, assuming that the Republicans who affirmed the false claims were Trump supporters, a majority of Republicans who supported Trump disagreed with him on central questions of fact.

Now, here is another extraordinary thing. Seventy-one percent of the Washington Post sample said they “regularly thought Trump makes misleading claims.” Of that 71 percent, more than two-thirds (49 percent) said they thought these claims were “usually flat-out false,” as against 22 percent who said they were “usually just exaggerations.” If you like, call these “low-information voters,” as political scientists tend to do, and you must still confront the fact that these were voters with above-average education. They are, one way or the other, fact-resistant. Seventy-eight percent of Trump supporters believed Trump’s false claims. Twenty-two percent of them did not believe those claims but supported Trump anyway. Perhaps this critical mass of Republicans did not know they preferred Trump to truth, or perhaps they were indifferent to truth. Either way, a substantial body of Americans refused to let truth stand in their way.

Like the most fervent of Foucault fans, they held a purely instrumental view of truth. The Washington Post poll found that 41 percent of Republicans said false claims were sometimes acceptable “in order to do what’s right for the country.” In 2018, only 49 percent of Republicans thought it “extremely important” for a presidential candidate to be honest, as against 70 percent of Democrats and 66 percent of independents. This question was designed to match precisely a question that had been asked in an AP poll in 2007. Then, the identical 70 percent of Democrats and 66 percent of independents had thought honesty “extremely important.” But by dramatic contrast, in 2007, 71 percent of Republicans had thought honesty “extremely important”—a percentage barely different from Democrats or independents. Something happened to Republicans so that, in 2018, more than half did not much mind dishonesty on the part of their political leader. Forty-one percent of Republicans said false claims were sometimes acceptable “in order to do what’s right for the country.”

In other words, a critical mass of Republicans must have approved of Trump while taking issue with important Trump falsehoods. For example, 65 percent of the sample said they thought that Russia tried to interfere with the 2016 presidential election, as against only 15 percent who said (as Trump, citing Vladimir Putin, did) this was false. Sixty-five percent said the “main” cause of global warming was human activity, as against 19 percent who said (as Trump did) the causes were natural, not human. Forty-four percent thought there was no evidence that millions of fraudulent votes were cast in 2016, as against 25 percent who thought (as Trump did) that there was such evidence. Only one-quarter of Americans in the fall of 2018 believed, with Trump, that the migrant “caravan” approaching the Mexican border from the south “included terrorists,” along with an additional 13 percent who were “not sure, but believed this claim was likely to be true.” The sum of that quarter and that eighth—38 percent—takes us into the range of Trump approval overall.

In sum, when it comes to politics, between three and four in ten Americans have cut loose from truth. Some one-third of Americans believe Trump’s falsehoods and support him. More than four in ten think their leader’s false claims are acceptable. Call this postmodernism if you like; I call it nihilism.

This is surely in no small part because of the media they rely on. Asked to name their two main news sources, 29 percent in the Post poll answered: Fox News. In addition, ten percent named local radio, four percent national radio, nine percent Facebook; and 15 percent “other.” Unfortunately, the survey did not get more specific. Is Rush Limbaugh on a local radio station “local” or “national”? How many of those who designated Facebook as their prime source were talking about right-wing Facebook sources in particular? Among adults who said Fox News was one of their top two sources for political news, 33 percent believed in Trump’s false claims tested in the poll, on average, compared with 21 percent of those who said Fox was not a main news source. They did not let their disagreements with his factual claims stand in the way of their support. It was not just that they lived in a bubble, or a silo, or some other metaphor for self-encapsulation. They lived in falsehood. And they didn’t mind. There was something more important to them than truth.

That something was—and remains—comfort. Actually, a particular kind of comfort—the feeling of righteous consistency that derives from avoiding internal conflict. Social psychologists, following Leon Festinger, have long used the term “cognitive dissonance” to describe the mental distress caused by holding contradictory beliefs simultaneously or trying to preserve a belief in the face of contrary evidence. Naturally people prefer to reduce such distress. One way to manage this task is to alter the now discordant belief. Another is to rationalize contradictions. Another is to reframe, or spin, or otherwise evade information that challenges the belief, and occasions when one might have to confront such information. There is, in other words, a need not to know.

One of the foundational findings of the research on cognitive dissonance is that, faced with a conflict between conviction and evidence, the believers will prefer their conviction to the degree that they have deeply invested, emotionally and practically, in it. The landmark study by Festinger and his co-authors—of how a flying saucer cult adapted to the fact that its prediction of flying saucer deliverance from a great flood was never confirmed—concluded that the intensity of a disconfirmed belief might actually increase to the degree that the conviction was strongly held at the start; that it was reinforced by other believers; and that they had invested time, money, and attention—“skin in the game”—such that backing off from the belief would be, to them, prohibitively expensive.

So the will to reduce cognitive dissonance helps the Trump base stay in line. To the true believer, it is better to be wrong with Trump than right with his enemies.

As weird as this may sound, for the purpose of minimizing internal friction it may help not quite to believe one’s beliefs. It is possible to tell a pollster that one “thinks” or “believes” a proposition about which one actually has little to no idea, yet to say that one “thinks” or “believes” because the pollster is an authority who thinks it preferable to believe than not to believe. It is possible to “think” or “believe” shallowly or squishily, as in “I think the American health care system is the best in the world,” or “I think that lowering tax rates on the wealthy encourages them to invest more and thus to create jobs.” For some of his followers, continuing belief in Trump may well be belief that hardens to dogma regardless of whether the reasons for that belief hold up. It is, to paraphrase a Barack Obama slogan, belief that one need not wholly believe in.

Some of the strongest minds of the mid-twentieth century struggled to understand how readily the modern public will accept falsehood. George Orwell used the term “doublethink” to cover the practice of forgetting inconvenient facts and also forgetting that one had forgotten them. More attuned to the sociology of mental surrender was Hannah Arendt who, in The Origins of Totalitarianism, pointed out that totalitarian movements find it useful to practice “a curiously varying mixture of gullibility and cynicism” and that “the whole hierarchical structure of totalitarian movements” enables them to do so. “The essential conviction shared by all ranks,” she wrote, “from fellow-traveler to leader, is that politics is a game of cheating.” If everyone cheats, than there is no moral vantage point from which to oppose a cheat. Arendt went so far as to see a slippery amalgam of dogma and evasion as deeply entrenched in the history of modern society altogether: “A mixture of gullibility and cynicism had been an outstanding characteristic of mob mentality before it became an everyday phenomenon of masses.”

Arendt anticipated a world of verbal garbage that the philosopher Harry G. Frankfurt would famously call bullshit—utterances set forth without regard for whether they are true or not. The liar and the truthteller, Frankfurt argued, are alike in that they both know the difference between truth and falsehood. But bullshitters disrespect truth so much that they neither know nor care whether what they say is true. To the bullshitter, all propositions are moves in a game, assertions of power—no more, no less.

Whether a statement describes reality matters less than whether it has the effect of shoring up the leader’s dominance. In this technical sense, many of Trump’s followers must treat his declarations as bullshit—as words whose semantic meaning is irrelevant but whose practical import is that they are Trump’s. Trump has uttered, or tweeted, them. They feel true; they have what Stephen Colbert memorably called “truthiness.” Trump “owns” them. Your duty is to affiliate with the idol, to stand firm with him against “enemies of the American people.” Loyalty trumps independent thought. Cognitive dissonance underwrites one’s sense of reality. Cynicism triumphs. What’s real is what it’s correct to feel. Truth follows affiliation.

In the land of the tall tale, an instrumental approach to truth comes as no stranger. One may dispute the balance of blame between the political parties, or how much responsibility accrues to Richard Nixon in particular, but it is plain that a turning point occurred with the cult of Ronald Reagan, who, when challenged, was capable of dancing away from some of his more risible proclamations by claiming that he was just kidding. (It would have been hard to believe Nixon was ever kidding.) By the time of “supply side economics,” various combinations of dogma and evasion had become central to Republican politics. What Reagan’s onetime rival George H. W. Bush had called “voodoo economics” has flourished for almost four decades, from Reagan to Paul Ryan. The Republican boilerplate to the effect that budget deficits are unforgivably devastating flourishes when Democrats are presidents but is tucked into the memory hole when Republicans occupy the White House. Republicans cheerily live with the deficits caused by military spending.

But policy positions are one thing and the surrounding tone of a world view is deeper and more dangerous. By 2016, the rejection of well-established science had become a necessity for a career in national Republican politics, for it was a matter of principle, in their eyes, that if disastrous climate change were indeed socially caused, a clamor for government policy to mitigate the damage would arise; but since a government regulation was, by now, anathema, the inner psychologic had to sizable government intervention. Q. E. D.

But the Republican veer toward untruth went, and goes, further. Policy judgments weren’t the half of it. It hardened into an outlook, an identity, a virtual way of mental life. Gradually, in the course of the 1980s and 1990s, the demonization of liberal expertise became foundational—sanctified by religion, fueled by contempt for intellectuals, and funded handsomely by oil companies. In 2000, candidate George W. Bush occasioned much mirth by dismissing vice-president Al Gore as “Ozone Man,” but it was not the rowdy son who originated the putdown; it was his once patrician father who, in 1992, started dismissing vice-presidential candidate Gore in those words for being so bold as to endorse the scientific judgment that the ozone layer was in the process of being destroyed by industrial chemicals. (“You know why I call him Ozone Man?” said the elder Bush. “This guy is so far out in the environmental extreme, we’ll be up to our neck in owls and outta work for every American. He is way out, far out, man.”)

Left-wingers committed distortions, too (the Green Party’s Jill Stein endorsed the cause of anti-vaccine crusaders), but it was overwhelmingly the right that saw the world through partisan lenses. When George W. Bush invaded Iraq, he cited the claim that Saddam Hussein possessed weapons of mass destruction prominently in his rationale. Seventeen years later, it remains an Republican article of faith that the Iraqi dictator possessed those weapons, or “weapons programs,” at the time Bush gave for going to war. According to a 2006 poll, “a full 50 percent of U.S. respondents…said they believe Iraq did have the forbidden arms when U.S. troops invaded in March 2003, an attack whose stated purpose was elimination of supposed WMD.” Later polls have found an enduring American faith in the WMD story. A poll, in early 2015, found that that 50 percent had eroded to 42 percent, although more than 50 percent of Republicans (and viewers of Fox News) thought it “definitely true” or “probably true” that American forces found an active WMD program in Iraq. Seven years after George W. Bush left the White House, and it had been widely reported that Saddam Hussein had eliminated WMD programs after the first Gulf War—in 1991—Bush was still obfuscating and half of Republicans were still aligned with him.

Examples are many. The following are culled from the psychologist John Ehrenreich’s stark 2017 analysis, “Why are Conservatives More Susceptible to Believing Lies”:

As recently as 2016, 45 percent of Republicans still believed that the Affordable Care Act included “death panels” (it doesn’t). A 2015 poll found that 54 percent of GOP primary voters believed then-President Obama to be a Muslim (…he isn’t). Then there are the false beliefs about generally accepted science. Only 25 percent of self-proclaimed Trump voters agree that climate change is caused by human activities. Only 43 percent of Republicans overall believe that humans have evolved over time.

“And then it gets really crazy,” Ehrenreich writes, not overstating. Republican partisanship extends beyond the acceptance of talking points that are directly policy-minded:
Almost 1 in 6 Trump voters, while simultaneously viewing photographs of the crowds at the 2016 inauguration of Donald Trump and at the 2012 inauguration of Barack Obama , insisted that the former were larger. Sixty-six percent of self-described “very conservative” Americans seriously believe that “Muslims are covertly implementing Sharia law in American courts.” Forty-six percent of Trump voters polled just after the 2016 election either thought that Hillary Clinton was connected to a child sex trafficking ring run out of the basement of a pizzeria in Washington, D.C., or weren’t sure if it was true.

Such evidence amply bears out the psychologist Solomon Asch’s findings from experiments of the 1950s. Asch found that his subjects readily revised the evidence of their senses in order to conform to what their peers saw.


In the Enlightenment theory that propelled the expansion of newspapers in the nineteenth century, information and education combine to make up the lifeblood of democracy. The more information, the better. The free flow of information resists the tyranny of the majority and works inexorably toward collective self-improvement. From the continuous sifting and winnowing of information, in a free marketplace of opinion, truth will out. Liberals might for good reason fear the despotism of public opinion, but they could assume that eventually citizens of good will would make effective use of their liberty.

Education was the cure-all. Schooling would swell the knowledgeable majority, who would learn to appreciate the value of circulating minority views so that they could be directly confronted, affording the opportunity to learn even from error. James Madison was promoting education, not a freedom-of-information act, when he wrote, in 1822:

A popular Government, without popular information, or the means of acquiring it, is but a Prologue to a Farce or a Tragedy; or, perhaps both. Knowledge will forever govern ignorance: And a people who mean to be their own Governors, must arm themselves with the power which knowledge gives.

In Madison’s spirit, John Stuart Mill further argued that a majority confronting its dissidents in a rational manner would breathe new life into dead beliefs. All in all, liberty would not only serve the cause of individuality but enable the entire society to function as a permanent learning process. Unification through knowledge would overcome divergent interests.

But educational credentials, at least formal ones, have ceased to function as theory prescribed. Not only does class division undergird educational differences, but formal education does not dissolve ignorance. Ehrenreich wrote:

College-educated Republicans are actually more likely than less-educated Republicans to have believed that Barack Obama was a Muslim and that “death panels” were part of the ACA. And for political conservatives, but not for liberals, greater knowledge of science and math is associated with a greater likelihood of dismissing what almost all scientists believe about the human causation of global warming.

It’s also not just misinformation gained from too many hours listening to Fox News, either, because correcting the falsehoods doesn’t change their opinions. For example, nine months following the release of President Obama’s long-form birth certificate, the percentage of Republicans who believed that he was not American-born was actually higher than before the release. Similarly, during the 2012 presidential campaign, Democrats corrected their previous overestimates of the unemployment rate after the Bureau of Labor Statistics released the actual data. Republicans overestimated even more than before.

John Stuart Mill seems to have imagined a republic unified through the expression of differences, evolving, perhaps, toward a confluence of discourse, however contentiously along the way—or even because contentiously along the way. Under the protracted guidance of something like an invisible hand, the republic would eventually consolidate through speech, shaped through presumably endless conversation tending willy-nilly, or asymptomatically, toward some sort of consensus. Alexis de Tocqueville warned about the danger of uniform opinion—tyranny of the majority—, toward the end of the first volume of his Democracy in America, and marveled about the social proximity of “intermingled” Americans whose character would draw ever closer to a common type, even as, politically, “every village forms a sort of republic accustomed to conduct its own affairs.” Associations would protect the majority against its tendency toward self-tyranny. Plurality would underwrite stability.

But Tocqueville was not naïve about America’s prospects. He pointed to one grave limit to the country’s future stability: “the more or less distant but inevitable danger of a conflict between the blacks and the whites of the South of the Union,” which was “a nightmare haunting the American imagination.” However placid the American surface, racial inequality was the inescapable tectonic cleavage. Once again—it is no less true an observation for being banal—Tocqueville was prophetic. No division cuts deeper in American culture today than the one derived from centuries of slavery and white supremacy. No false precision here: no precise calculus is possible; male domination surely grips contemporary culture by its own logic. But white racism set a tone for our other collective secessions from reason and, as I argued in the first half of this essay, still exerts a mighty influence on the polarization of opinion today. No cause looms larger in the American right’s secession from reason than protectiveness of white privilege.

What we confront now throughout much of the world, not least in the United States, is a sealed-off bloc unified by a backlash against liberal culture, a bloc apparently impervious to argument and evidence and fed by an unending flow of reinforcing propaganda. This bloc represents a worldwide cultural backlash toward ethnonationalism throughout the developed world. Each national branch belongs, paradoxically, to an Ethnonationalist International—one might call it an axis. It fervently believes that it is the true nation and fears that, culturally and demographically, it is losing its grip. It pumps up its feelings of righteousness with rallies and rituals—“Lock her up! Build the wall!” That it is rejected by “enemies of the American people”— institutions of higher learning as well as established media—only confirms its rectitude. As John Ehrenreich pointed out, “self-described conservatives tend to place a higher value than those to their left on deference to tradition and authority. They are more likely to value stability, conformity, and order, and have more difficulty tolerating novelty and ambiguity and uncertainty. They are more sensitive than liberals to information suggesting the possibility of danger than to information suggesting benefits. And they are more moralistic and more likely to repress unconscious drives towards unconventional sexuality.”

All signs point to the conclusion that Trump’s base, seized by backlash, amounts to more than one-third of Americans. That is, it numbers about eighty percent of those who, throughout Trump’s first two years in the White House, approved of him. This bloc is overwhelmingly white and disproportionately rural. According to a November 2018 poll, Trump’s support in rural areas was 61 percent, as against 26 percent disapproval, while in cities, only 31 percent approved of him while 59 percent disapproved. In other words, rural net approval was +35, while his urban net was –28. (The same poll found his approve-minus-disapprove numbers among all adults as 43-45, or -2—a considerably smaller number than other polls.)

This bloc is disproportionately evangelical and reliant upon Fox News. In the fall of 2019, “when asked whether they believed that immigrants were ‘invading the country’ and replacing its ethnic and cultural background, 78 percent of Republicans who rely on Fox News said yes, compared with 52 percent of Republicans who do not consider the network their main news source.” The hard core of Trump supporters are self-sequestered, largely in the South, Midwest, and prairie states, along with rural and exurban areas elsewhere. But however advantaged by the Electoral College, they cannot, by themselves, guarantee a majority of electoral votes. They are nowhere near a popular majority. Under favorable circumstances, however, they can prevail. 2016 afforded them the favorable circumstances. Not only did the Clinton campaign frequently blunder, but James Comey’s eleventh-hour intervention was potent. Disinformation emanating from Russia made a difference. So did voter suppression. In a close election, victory has joint authorship.

I am not arguing that all of Trump’s supporters—even all the evangelicals—approved of him for the same reasons, in the same spirit or with the same intensity. I am not arguing that pure faith, ungrounded in material rewards, explains the entirety of Trump’s base of support, or its relative stability, any more than I argued in the first part of this essay that his followers’ projection of their transgressive impulses onto him explains the sum of his appeal. Trump surely accrues support from voters who believe that he kept his promises and specifically that his policies helped either the local or the national economy to the degree that his most fervent promoters claim. I do not want to lump all Trump supporters into the evangelical box, since white voters without college degrees make up a larger share of Trump’s vote than white evangelicals. (So do whites over fifty.) Whether white evangelicals will continue to loom as large in the Republican base as they have to date, or will remain as enthusiastic, is arguable. Perhaps young white evangelicals are beginning to peel away from the old-time religion. The Washington Post columnist Michael Gerson, an important evangelical advisor to George W. Bush and a fierce opponent of Donald Trump, thinks the exodus is underway, pointing to “the massive sell-off of evangelicalism among the young. About 26 percent of Americans 65 and older identify as white evangelical Protestants. Among those ages 18 to 29, the figure is 8 percent.”

It seems unlikely, though possible, that they will return to the fold as they age. Still, many are the complicating variables—and this is only to speak of known unknowns. It is also conceivable that Trump’s power to bind his followers may crack. Insofar as his cultic following relies on material results—in particular, the building of a wall that is more than symbolic—faith in Trump has an empirical limit. The psychohistorian Robert Jay Lifton observes that the fervor of followers, once disappointed, “can be quickly transformed into the anger of the betrayed.” But it strikes me as at least equally likely that the base’s attachment to Trump can outlast many disappointments insofar as they can be blamed on the treachery of his unrelenting enemies. Even a zombie movement can have its life prolonged by the right-wing echo chamber. Anyone who thinks the fervor of zealots is the result of rational calculation is not living on Planet Earth. One must also contemplate the possibility that, after Trump, a less polarizing version of him might emerge to sweep up ever-Trump remnants of a Republican Party, and augment them, given that the party lacks any rival consolidating force beyond the alloy of government-hatred and cultural backlash. Conceivably a near-Trump successor to Trump as a more palatable avatar for backlash—Trump with a human face, if that is not an oxymoronic description—might sweep into the future Republican leadership.

But now we are stretching the limits of prediction to the breaking point. In any given election, the identity of Trump’s Democratic opponent will matter one way or the other. Catastrophic events may always intervene. Campaign strategies and tactics will, as always, matter. If Democrats return to power and rule wisely, they can, over time, trim the size of Trump’s base. But the fervor and loyalty of that base is not near vanishing. For the followers whom Trump called “the Second Amendment people,” any effective turn against their Leader is likely to prove inciting. For some time to come, his decisive evangelicals will be in no mood to collapse quietly. They will be inspired by these words of the influential evangelical organizer George Barna: “God never waits until He has a majority on His side to move forward in power and to claim a decisive victory.”

In other words, the base can be defeated. But they have been roused. They are in no mood to expire quietly.

TODD GITLIN is Professor of Journalism and Sociology, and Chair of the PhD program in Communications at Columbia University. He has published 16 books of non-fiction, fiction and poetry and recently completed a novel, The Opposition, set in the 1960s.