August 23, 2017

Are Trump Supporters Authoritarians?

Donald Trump won the presidency while flouting long-standing political norms and manifesting gross ignorance of the structure of the federal government, including the separation of powers. These facts, and Trump’s failure to refer to the Constitution even in his inauguration speech, prompted widespread fears that his administration would disregard legal barriers to its objectives, leading to constitutional crises—and that he might even stage a coup d’etat. Would his voters back him up under such circumstances?

This question seemed to be of pressing importance in the early days of Trump’s tenure. But things change fast in the Trumpocene. Perhaps the only pleasant surprise of the new era has been the administration’s adherence to the Constitution. Most strikingly, the president redrafted his proposed travel ban to conform to judicial instructions. The original travel ban was so important to Trump that he made it his first major presidential action, rushing it into place without vetting by the Departments of Justice and Homeland Security. Yet when the courts blocked its implementation, Trump made no move to bypass them or appeal over their heads to “the people.” As Samuel Moyn and David Priestland put it last week in the New York Times, “there is no real evidence that Mr. Trump wants to seize power unconstitutionally, and there is no reason to think he could succeed.” He may well have committed impeachable offenses. But these offenses—such as firing the head of the FBI to protect himself from investigation—are not necessarily violations of the Constitution (although it remains to be seen if Trump violated the emoluments clause), let alone do they amount to a coup.

Yet the possibility of Trump supporters’ authoritarianism is important for the long term, regardless of what happens in the relatively brief time (we can hope) remaining in Trump’s presidency. And there’s troubling evidence on this front. In a recent survey, Ariel Malka of Yeshiva University and Yphtach Lelkes of the University of Pennsylvania found that 52 percent of Republican respondents would support delaying the 2020 presidential election “until the country can make sure that only eligible American citizens can vote.” Fifty-six percent would support a delay if both President Trump and Republicans in Congress backed it.

Such attitudes undercut Moyn and Priestland’s contention that only paranoiacs should worry about threats to “basic freedoms or the rule of law.” Trump himself may be too incompetent and distracted to seize unconstitutional powers, but it would seem that his base might welcome such a move.

The Authoritarian Personality Theory

Even more troubling for the long term—if it’s true—is the hypothesis that significant numbers of Trump supporters are not just loyal to him, or overly preoccupied with vote fraud, but predisposed by their personalities to prefer “strong leaders” to democratic rule and constitutional protections. This theory has been repeatedly featured in the New York Times (here, here, and here) and the Washington Post (here, here, here, and here), and widely publicized elsewhere (e.g., Politico, the Huffington Post, and Vox).

The theory originated in The Authoritarian Personality (1950), a massive, now-discredited study produced by members of the Frankfurt School.[1] The authors of the study attempted to explain the attraction of Nazism by marrying Freud to Marx. The terms of this marriage were best expressed by Theodor Adorno, the lead author: “Cui bono? What purposes within the lives of our subjects are served by anti-Semitic ways of thinking?”[2] Like Marx, Adorno assumed that the sources of malevolent behavior lie under the surface, below the conscious awareness of those involved. But instead of seeing people’s economic interests lurking below, Adorno identified irrational drives. The struggle of id, ego, and superego replaced the struggle of classes.

Contemporary theorists of the authoritarian personality have dropped the Freudian baggage, retaining only the commitment to deep psychological explanations. Thus, the leading theorist of this stripe, Karen Stenner, contends that a primal psychological force “brings together certain traits—obedience to authority, moral absolutism and conformity, intolerance and punitiveness toward dissidents and deviants, animosity and aggression against racial and ethnic out-groups.”[3] This primal force is measured by asking survey respondents which traits are more important in children:

1. independence or respect for their elders?

2. curiosity or good manners?

3. self-reliance or obedience?

4. being considerate or being well-behaved?

Those who choose the second option in the four binaries are diagnosed as suffering from “a deep-seated predisposition, whose primary motives yield a functionally related array of stances concerned with minimizing difference in all its manifestations.” This predisposition toward uniformity, in turn, “is the primary determinant of intolerance of difference across domains, cultures, and time.” That is, “intolerance of racial diversity, political dissent, and moral deviance are all primarily driven” by an authoritarian personality in the contemporary United States and all across the globe, past, present, and future.[4]

Surprisingly, this grandiose theory seems to explain a good deal of Trump’s support. According to survey research by Matthew MacWilliams of UMass Amherst, half of the Republican respondents who chose the second option in the four binaries supported Trump, as did only one in six of those who chose the first option. Similarly, Marc Hetherington of Vanderbilt and Jonathan Weiler of UNC have shown that Republicans tend to score much higher than Democrats on an authoritarian-personality scale consisting of the four items, while Christopher Weber of the University of Arizona, Christopher Federico of the University of Minnesota, and Stanley Feldman of Stony Brook University have correlated the choice of the second of the four pairs of options with racial resentment, sexism, moral traditionalism, and opposition to immigration.

Evidence against the Authoritarian-Personality Theory

If the authoritarian personality hypothesis sounds too good to be true—an explains-everything theory of what’s the matter with Trump supporters and what’s the matter with Republicans, too—that’s because, despite initial appearances, it probably isn’t accurate.

A relatively small criticism has been advanced by Wendy Rahn of the University of Chicago and Eric Oliver of Minnesota. They show that authoritarianism (as defined by those four binaries) was a less important trait among Trump supporters than was mistrust of experts, and that opposition to elites was nearly as important. They also maintain that more Cruz supporters were authoritarian than Trump supporters, and that Rubio supporters were almost as authoritarian. Even with these qualifications, however, Rahn and Oliver’s work allows that a significant slice of Trump supporters is authoritarian, because the author’s don’t challenge the authoritarian-personality construct.

An inadvertent challenge to the construct can be found in Weber, Federico, and Feldman’s research, which shows that in 1996, and again in 2016, Democrats became dramatically less authoritarian, even as Republicans’ authoritarianism drifted slightly higher. There may be a sorting effect here, but not one that is big enough to account for the nearly one-third decline in Democrats’ authoritarianism (from .65 to .46 on a scale compiled from the four child-rearing questions) in the space of 24 years.

Figure 1. Distribution of “Authoritarian Personalities” among Republicans and Democrats

(republished from The Washington Post)

Weber et al. do not notice it, but this finding is devastating to the assumption that the four binaries are grounded in personality structures: such structures are not supposed to change. This changelessness is what makes a personality-based explanation “deep” and confers on it the law-like, immutable qualities that Stenner claims for authoritarian-personality theory. As she puts it, authoritarians are “the kind of people who—by virtue of deep-seated predispositions neither they nor we have much capacity to alter—will always be imperfect democratic citizens.”[5] Personality-based theories are supposed to identify the universal constants underlying certain types of behavior, where “shallower” theories would explain behavior in terms of more transient causes, such as historically contingent beliefs.

The Superficiality of “Deep” Explanations

Beliefs are obvious candidates for “shallow” explanations because they affect behavior profoundly, yet they are malleable.

Take the attitudes about children tapped by the four measures of authoritarianism. The authoritarianism researchers assume that these attitudes spring from psychological sources, but another possibility is that they stem from beliefs about how children should behave held by some parents and schoolteachers but not others. These parents and schoolteachers then impress these beliefs on the children in their charge expressly and tacitly, e.g., by punishing certain infractions and not others, shaping the beliefs about how children should be treated that are held by the children themselves. These beliefs then persist into adulthood, where they can be measured in opinion surveys.

You don’t have to be a deeply insecure fascist-in-the-making to believe that respect for one’s elders, good manners, and so on are desirable traits in children. You may simply have learned these beliefs from your parents or teachers as part of your own childhood training in how to behave. Such beliefs may also be taught or reinforced by what one’s relatives and friends advise when one is rearing one’s own children; or by what one reads about child-rearing in the newspaper or sees on the morning news and afternoon talk shows. Either way, one’s personality may have nothing to do with one’s attitudes about how children should behave.

This belief-based theory would account for all the correlations identified by the “authoritarianism” researchers if we take one additional step: contextualizing beliefs. That is, we need to challenge the disregard of history that is so common in political psychology.

Consider: the psychologists of authoritarianism treat the first options in the four binaries as if they are the natural defaults—timeless values that would manifest themselves in everyone if not for the interference of authoritarian predispositions. Yet non-“authoritarian” ideas about children only began to come into vogue after the appearance of Benjamin Spock’s Baby and Child Care in 1946. By the 1960s, Spock’s views were being widely propagated in the mass media. But it was well known that the venerable Dr. Spock was a prominent antiwar and New Left activist, and his child-care ideas may have been rejected by conservatives for that reason. Moreover, his child-rearing advice fit with a web of progressive beliefs about the natural goodness of human nature that were and are commonly taught on the left but rejected on the right.

The late-twentieth-century spread of progressive child-rearing beliefs among liberals would explain why Democrats now tend to share these beliefs, with no need to appeal to their personalities. Conversely, the traditional belief that children should be respectful, well-mannered, obedient, and well-behaved undoubtedly has its own history: a history of behavioral routines and prescriptions handed down over the generations (e.g., “spare the rod, spoil the child”). Such behaviors and prescriptions would have tended to persist in the minds of those to whom they were taught as long as they were not later exposed to convincing presentations of new, progressive beliefs, such as those of Dr. Spock. Since Spock’s beliefs about child-rearing would have been most persuasive to those who already accepted the web of progressive beliefs with which they fit, and since these beliefs (if not Spock’s child-rearing advice itself) tended to be propagated by universities and other organs of high culture, those who were relatively uneducated and outside the cultural avant-garde would have been the least likely to encounter persuasive presentations of these new beliefs. That would include a great many conservative Republicans, especially evangelical Christians.

This belief-based or ideational theory would also explain Weber et al.’s discovery of correlations between the “authoritarian” options among the four binaries and sexism, racism, opposition to immigration, and traditionalism. These are all attitudes that are passed down from generation to generation, and that therefore come to take on the appearance of “common sense.” Inherited attitudes such as these have to be intellectually identified and challenged head-on if they are to be overturned, as often occurs in higher education. The authoritarianism researchers overlook the possibility that old-fashioned webs of beliefs will tend to produce old-fashioned answers not only to child-rearing questions but questions about race, gender, and everything else, regardless of personality. Unlike personalities, though, old-fashioned webs of belief can change if they are displaced by new webs of progressive belief.

Beliefs vs. Psychology as Causes of Political Attitudes

The ideational theory would also explain more detailed political views that have been correlated with “authoritarianism.”

Vox, which was inspired by the psychology literature to commission a survey on the policy-related attitudes of “authoritarians,” found that the respondents who chose the second option in the four pairs deviated from other Americans most decisively in their support for “using military force over diplomacy against countries that threaten the U.S.,” their belief that same-sex marriage is “bad for America,” their support for “airport checks on passengers who appear to be of Middle Eastern descent to curb terrorism,” their feeling of “risk from terrorist organizations such as ISIS,” their belief that building more mosques is “bad for America,” their support for “requiring that all citizens carry a national identity card to show to a police officer on request,” their support for “changing the Constitution to bar citizenship for children of illegal immigrants,” and a host of other attitudes that can be labeled intolerant (or authoritarian) for short. These attitudes, however, are, like the second options among the binaries, “commonsensical” and simplistic responses to perceived social problems. This doesn’t necessarily mean that they’re wrong; a simple answer may be best, depending on the circumstances. But it does mean that they’re the type of attitudes one would expect from people who have had relatively little exposure to the complexities that are often introduced in elite discourse on these issues. At the same time, the “authoritarian” respondents’ views mirror the beliefs propagated on Fox News and conservative talk radio. Thus, they are beliefs that conservatives are far more likely to hear defended than are liberals.

Perhaps anticipating such a response to its survey, Vox points out that the policy-related attitudes of “authoritarians” are “distinct from GOP orthodoxy.” However, this is orthodoxy defined narrowly. It encompasses only the libertarian-inflected ideas common among Republican congressional elites and published in conservative magazines and on the op-ed pages of the Wall Street Journal. There was never any reason to think that these ideas had penetrated to the grass roots, as these publications are read only by conservative elites.

As for the recent increase in progressive beliefs about child-rearing among Democrats, in the absence of further data one can only speculate about its possible ideational (as opposed to psychological) origins. Political scientists have thus far shown almost no interest in researching the ideational sources of people’s beliefs, but innovative researchers might begin by looking into what was being tacitly and expressly taught to schoolchildren about how they should behave in the two decades before the 1990s. The survey results depicted in Figure 1 reflect the views of only those who were 18 or older in the years of the surveys—i.e., those who were no longer being trained at home and in school to be a child of one type or another (traditional or progressive). It stands to reason that the rise in progressive beliefs about children that became evident among Democrats in 1994 may have originated in what was taught to them, in the 18 years before 1994, about how they should behave. By the same token, the sharp uptick in progressive beliefs about children among Democrats seen in 2016 may have originated in what had been taught during the prior 18 years.

This is highly speculative, but it is more logical than supposing that in 2016, or 1994, Democrats’ personalities underwent a sudden transformation.

Disillusionment with Democracy Goes Far Beyond Trump’s Supporters

We’re now better positioned to explain the Malka and Lelkes finding about Republicans’ willingness to suspend elections due to fear of voter fraud.

A justly famous paper by Roberto Stefan Foa and Yascha Mounk undercuts the notion that Republicans or Trump supporters alone are dissatisfied with established political institutions. The youngest cohort in the Mounk and Foa research—precisely the cohort that, in the United States, went for Clinton in a landslide over Trump—is the one that expresses far and away the lowest commitment to democracy:

Figure 2. Declining Support for “Democracy” by Age Cohort

(republished from the New York Times)


It would appear, then, that political disaffection links aging white Republicans and Trump supporters with young, ethnically diverse millennials; and that this disaffection is common not only in the United States but across the industrialized democracies of the West.

The paradox is heightened by the fact that the Trumpist side of the equation is consistent with the most simplistic interpretation of democracy: majority rule. MacWilliams found that Trump supporters are likelier than other Republicans to agree that “it is sometimes necessary to keep other groups in their place,” that “opposition from the minority sometimes needs to be circumscribed,” and that “the minority’s rights” should not necessarily be “protected from the majority’s power.” The Foa and Mounk paper points out that in 2011, an amazing 32 percent of the U.S. respondents thought that it would be “better to have a ‘strong leader’ who does not have to ‘bother with parliament and elections.’” Like so-called authoritarians’ support for traditional norms of child rearing and simplistic public policies, these survey responses may have less to do with a desire to impose uniformity than a desire to solve public problems in a straightforward, “commonsensical” manner—one that is, moreover, consistent with a conventional, uncomplicated understanding of democracy.

On the other side of the equation, an even more amazing 49 percent of U.S. respondents approved of “‘having experts, not government, make decisions according to what they think is best for the country.’”[6] In most other Western countries, the proportion of respondents who favored rule by experts outnumbered those who favored rule by a strong leader by nearly a 2:1 ratio. Foa and Mounk emphasize that support for both populist and expertocratic options has risen over time.

It appears that the rising dissatisfaction with the status quo can go in either of two main directions: unmitigated rule by the majority and unmitigated rule by experts. Neither direction is particularly compatible with constitutional guarantees, the rule of law, or checks and balances: such devices impede the majority, or the experts, from doing whatever they find necessary. But both directions are logical (if ultimately unjustified) responses to the perception that government is failing. Foa and Mounk conclude that “many citizens . . . no longer believe that democracy can deliver on their most pressing needs and preferences.”[7] If we replace the protean term “democracy” with “existing political institutions,” this conclusion seems inescapable.

Foa and Mounk do not speculate about the source of the cross-national malaise with the status quo. The fact that this malaise is increasing dramatically over time virtually rules out a personality-based explanation. The more “commonsensical” explanation blames economic decline. This is the explanation that I’ll examine in my next post.

[1] For a brief overview of scholarly criticisms of The Authoritarian Personality, see Karen Stenner, The Authoritarian Dynamic (Cambridge University Press, 2005), pp. 2-3.

[2] T. W. Adorno, Else Frenkel-Brunswik, Daniel J. Levinson, and R. Nevitt Sanford, The Authoritarian Personality (Harper & Row, 1950), p. 618.

[3] Stenner, p. 3.

[4] Ibid., p. 269.

[5] Ibid., p. 1.

[6] Roberto Stefan Foa and Yascha Mounk, “The Democratic Disconnect,” Journal of Democracy 27(3) (2016): 5-17.

[7] Ibid., p. 16.

Jeffrey Friedman, the Director of the Niskanen Center’s Institute for the Study of Politics and a Visiting Scholar in the Charles and Louise Travers Department of Political Science, University of California, Berkeley, is the editor of Critical Review and of The Nature and Origins of Mass Opinion Reconsidered (Routledge, 2014).