Do variations in authoritarianism explain the left-right divide?

The first frequently asserted but erroneous account of the psychological/biological roots of the Trump movement is that people who are predisposed to authoritarianism are also attracted to leaders such as Donald Trump. Authoritarians prefer “blind submission to authority” or “obedience over liberty” (Norris & Inglehart, Reference Norris and Inglehart2019, p. 69), and at least since the 1950s (Adorno et al., Reference Adorno, Frenkel-Brunswik, Levinson and Sanford1950), this desire to obey authority and to have authority obeyed has been associated with status quo rightists. Claims of an authoritarian-conservative connection were popularized by the work of Robert Altemeyer (Reference Altemeyer1981, Reference Altemeyer1988). In fact, Altemeyer probably took things a step too far by baking conservatism into his conceptualization and measurement of an authoritarian, as his preferred label—“right-wing authoritarian” (RWA)—implies. RWA items such as “You have to admire those who challenged the law by protesting for women’s abortion rights, for animal rights, or to abolish school prayer” do not so much tap who is and who is not authoritarian as who is and who is not conservative on hot-button social issues such as abortion and school prayer. By attaching authoritarian attitudes to particular political beliefs, authoritarianism is assured of being related to political ideology, just as this relationship is assured of being tautological and largely meaningless.

To make matters worse, the behavioral tendencies of Donald Trump and similar nativists around the world have encouraged scholars and pundits to dredge up the older literature on authoritarianism in order to apply the label to Trump’s most dedicated followers. A few of the many works making such a claim include “Trump’s America and the Rise of the Authoritarian Personality” (Linden, Reference Linden2017); “The Rise of Trump: America’s Authoritarian Spring” (MacWilliams, Reference MacWilliams2016); “Group-Based Dominance and Authoritarian Aggression Predict Support for Donald Trump” (Womick et al., Reference Womick, Rothmund, Azevedo, King and Jost2018); “The Rise of American Authoritarianism” (Taub, Reference Taub2016); and “Authoritarian Nightmare: Trump and His Followers” (Dean & Altemeyer, Reference Dean and Altemeyer2020).

Why are these claims mistaken? To begin anecdotally, Trump’s most ardent followers—the ones who stormed the U.S. Capitol on January 6, 2021, and the ones who wanted to tar, feather, and perhaps kill authority figures such as Anthony Fauci, Mike Pence, and Gretchen Whitmer—hardly appear to be authoritarians. In fact, they recoil in the presence of authority. They physically attack rather than adopt a respectful and subservient posture toward police officers and government officials. Many of those who venerate Trump, it would seem, do not value obedience over liberty; if anything, they value liberty over obedience. As has been noted elsewhere, they are much closer to being anti-authoritarians than authoritarians (Lutjen, Reference Lutjen2022).

Systematic evidence comes to the same conclusion. The best empirical studies find little evidence of a link between Trump supporters and authoritarianism. Drawing on a large national survey conducted in 2017, a team of psychologists led by Steven Ludeke found no relationship between Trump support and either authoritarian submission or conventionalism. In fact, the signs of the relationships were in the opposite direction—and in some cases significantly so, suggesting that, if anything, Trump supporters were less authoritarian (Ludeke et al., Reference Ludeke, Klitgaard and Vitriol2018). A second study, this one led by Jake Womick and based on three separate surveys, confirmed Ludeke et al.’s conclusion that there is no relationship between Trump support and either authoritarian submission or conventionalism (Womick et al., Reference Womick, Rothmund, Azevedo, King and Jost2018).

In a related vein, a national survey conducted for me in 2019 reveals how contingent the desire for authority is (Hibbing, Reference Hibbing2020). It showed that Trump supporters only favor authority figures if those authority figures are committed to pursuing a particular set of policies—and, as will be described in detail shortly, the pertinent policies are those that protect societal insiders from the threats believed to be posed by outsiders. One of the survey items asked respondents whether they agreed with the following statement: “Our country desperately needs a forceful, mighty leader who will keep us safe from criminal elements and from foreign powers.” Just 33% of self-identified liberals agreed with that statement, but fully 84% of those who could be considered members of Trump’s base agreed. If we focused only on the result from this single item, it would seem that Trump supporters are indeed authoritarians—after all, they are the ones expressing a desire for “forceful, mighty leaders.”

That conclusion would be premature, however. A parallel survey item asked whether respondents agreed that “our country desperately needs a forceful, mighty leader who will help the poor and save the Earth’s environment.” This focus on helping the poor and saving the environment elicits a radically different attitude toward authoritarian leadership than a focus on keeping us safe from criminals and foreign powers. Liberals suddenly become authoritarians, and Trump supporters cease being authoritarians. Only 35% of Trump supporters want a “forceful, mighty leader” who helps the poor, while 70% of liberals do. Clearly, the key factor is not an indiscriminate desire for a forceful, mighty, authoritarian leader so much as a focused desire for particular policy goals. Trump supporters are hardly authoritarians who favor obedience over liberty regardless of the issues involved.

Do variations in blanket negativity bias explain the left-right divide?

The second, somewhat inaccurate account of the biological underpinnings of the left-right divide that is tempting to apply to the Trump movement is that, compared with people on the political left, those on the right simply are more responsive to negative stimuli generally (for a summary, see Hibbing et al., Reference Hibbing, Smith and Alford2014a). This argument holds that, compared with people on the political left, conservatives are more attuned and responsive to negative stimuli, perhaps because they tend to feel more threatened by them. This notion that there are variations in negativity bias across the ideological spectrum certainly squares with the observation that people on the right often seem more sensitive to disgusting and/or threatening stimuli (Castelli & Cararro, Reference Castelli and Carraro2011; Dodd et al., Reference Dodd, Balzer, Jacobs, Gruszcynski, Smith and Hibbing2012; Inbar et al., Reference Inbar, Pizarro and Bloom2009; Oxley et al., Reference Oxley, Smith, Alford, Hibbing, Miller, Scalora, Hatemi and Hibbing2008). Donald Trump, after all, proudly admits to being a “germaphobe” (Adams, Reference Adams2016; Lippman, Reference Lippman2019), and his followers have been found to score unusually high on measures of “sexual disgust” (Billingsley et al., Reference Billingsley, Lieberman and Tybur2018).

On the other hand, the negativity bias thesis does not always fit. Despite their alleged focus on the negative, individuals on the right are consistently found to be happier than those on the left (Napier & Jost, Reference Napier and Jost2008; Schlenker et al., Reference Schlenker, Chambers and Le2012), they are significantly less, not more, fearful of nearly all stimuli, including natural disasters, clowns, heights, the dark, and the grid collapsing (Hibbing, Reference Hibbing2020, pp. 40–42), and they are less, not more, responsive to negative situations such as environmental degradation and the threat posed by COVID-19 (Klein, Reference Klein2020). None of these results seems consistent with the claim that people on the right are more responsive to the negative aspects of life.

On top of this, a key finding of this research stream—that, compared with people on the left, people on the right display a more noticeable sympathetic nervous system response to negative stimuli, as measured by electrodermal activity—has not replicated well (Bakker et al., Reference Bakker, Schumacher, Gothreau and Arceneaux2020; Knoll et al., Reference Knoll, O’Daniel and Cusato2015). This inconsistent replication, in combination with the problems mentioned earlier, raises questions about the broad generalization that people on the political right are more fearful of the negative aspects of life and serves as a caution against applying the negativity bias thesis to a particularly salient group of people on the right: Trump supporters. A reconsideration seems to be in order.

If Trump’s strongest supporters are not authoritarians, and if they do not have a blanket negativity bias, then what does characterize them in terms of a deeper psychological/biological orientation? Some background regarding the human condition will be beneficial in setting the stage for answering this question.

Bedrock dilemmas of social life

All social units, whether they are relatively primitive hunter-gatherer tribes of a few dozen members or massive, technologically sophisticated, highly developed modern nation-states, must resolve certain bedrock dilemmas (see, e.g., Hibbing et al., Reference Hibbing, Smith and Alford2014b), else they would not be able to function. Given that there are distinct and competing ways of solving these various bedrock dilemmas, intrasocietal disputes over how best to address them will inevitably occur, and the recurring, recognizable nature of these disputes gives politics a commonality quite likely traceable to the universal biological factors characterizing Homo sapiens.

According to one account, the three bedrock dilemmas that are most foundational to politics are how to deal with in-group norm violators, how open to be toward nontraditional ideas and lifestyles, and how to relate to members of other social units (Hibbing, Reference Hibbing2020). Every social entity, regardless of latitude or era, must address these three dilemmas. Moreover, group members’ positions on these dilemmas are likely to follow predictable patterns, thus forming “packages.” Individuals who favor harsh punishment for norm violators also are likely to look askance at new lifestyles and at immigrants from other social groups, just as individuals who want to welcome outsiders are also likely to be open to novel lifestyles and to want to rehabilitate rather than punish norm violators. These logical packages of stances on timeless bedrock dilemmas explain the astounding consistency of what we refer to today as the left-right ideological divide (see Bobbio, Reference Bobbio1996; Jost, Reference Jost2006), with one side of the division welcoming new approaches such as lifestyles involving agriculture and an improved lot for women and the other side championing a more static, militaristic, and traditional approach. Reputable scholars have traced this division not just to postrevolutionary France, but all the way back to Iroquois tribes circa 1,000 CE and even to Neolithic societies in the Mideast more than 10,000 years ago (Graeber & Wengrow, Reference Graeber and Wengrow2021, p. 487). If evidence on intrasocietal divisions were available earlier than the Neolithic period, it would almost certainly reveal the same familiar split between those who feel quite secure in the face of outsiders, norm violators, and new lifestyle devotees and those who feel less secure or even threatened. The particular norms, out-groups, and lifestyles at issue will vary over time and space (e.g., in the United States, Roman Catholics used to be viewed as outsiders, but now they mostly are not; interracial marriage used to be viewed as a dangerous new lifestyle, but now is mostly accepted), but the division between supporters and opponents of whatever out-groups, norm violators, and new lifestyles are salient is a constant.

If this line of thinking is correct, if what is universal is the core division between supporters of out-groups, norm violators, and new lifestyles and opponents of out-groups, norm violators, and new lifestyles, and if Trump supporters and opponents are emblematic of this core division, then issues concerning outsiders and security should be what fundamentally divides Trump supporters and their opponents. Are they?

What distinguishes Trump’s followers?

The issue positions of intense Trump supporters seem to differ in all sorts of ways from the issue positions of others. Of course, if the comparison group is individuals who are well to the left of political center, their issue differences with Trump venerators will be massive and unsurprising. But the more interesting comparison group consists of those who, like Trump supporters, are self-proclaimed conservatives but who are not part of Trump’s base. What are the issues on which ardent Trump supporters differ from their fellow conservatives? Answers can be found using the 2019 U.S. national survey mentioned earlier, which should help us acquire more accurate expectations of the deeper psychological and physiological characteristics that separate people on bedrock dilemmas.

To identify Trump’s most impassioned supporters, I used the following item: “Donald Trump is one of the very best presidents in the entire history of our country.” Respondents could either strongly disagree, disagree, neither agree nor disagree, agree, or strongly agree with this statement. Note that the item sets a high bar. Agreement requires respondents not just to support Trump or to have voted for him but to place him, presumably along with George Washington and Abraham Lincoln, in the pantheon of “best ever” presidents. Even so, in the summer of 2019, 12% of all respondents agreed with this statement and another 17% strongly agreed. Given that my goal is to identify Trump’s most fervent supporters, I classify those who strongly agreed that Trump is one of the best presidents ever as “Trump venerators” and compare their issue positions with those of conservatives who do not venerate Trump (that is, those who do not think he is one of the best ever).

If the theory described here is correct, the kinds of issues that really distinguish fervid Trump supporters will be those relevant to norm violators, new lifestyles, and protection from the threats posed by outsiders. Differences on standard economic and social issues—issues that have not had the same presence throughout the course of human social life—will not be as noticeable. This hypothesis can be tested by grouping policy issues into three categories—economic, social, and nativist/protectionist—and employing three sample issues for each category. The results are presented in Table 1.

Table 1.

What policy stances distinguish Trump venerators?

Notes: NTVCs = non-Trump-venerating conservatives; Trump venerators are those strongly agreeing that Trump is one of the “best presidents ever.”

* p < .05.

N = 187 NTVCs, 170 Trump venerators.

Source: 2019 YouGov survey of a demographically representative sample of U.S. adults.

Compared with conservatives who might not be considered part of Trump’s base, Trump venerators, at best, are only marginally more conservative on economic issues. They are very slightly more supportive of lower taxes across the board and small government, and somewhat less supportive of higher taxes on the rich, but none of these differences is statistically significant at traditionally accepted (p < .05) levels. Differences are even more muted on social issues such as birth control, abortion, and sex education. In fact, on all these issues, Trump venerators are actually slightly (but not significantly) less likely than conservatives who do not venerate Trump to adopt the conservative position, but the bottom line is that the differences in stances on social and economic issues tend to be slight and do not distinguish Trump venerators from other conservatives.

The story is markedly different, however, when attention shifts to nativist/protectionist policies, as it does in the bottom third of the table. Here we find that Trump venerators are significantly more conservative than non-Trump-venerating conservatives. Even though non-Trump-venerating conservatives are solidly supportive of the death penalty, reducing immigration, and increasing defense spending, Trump venerators are much more likely to embrace these stances. To be specific, compared to their fellow conservatives, Trump venerators are 18 points more supportive of the death penalty, 18 points more supportive of reducing immigration, and 26 points more supportive of increasing defense spending. Consistent with the hypothesis I am advancing, what really distinguishes Trump’s most avid supporters is not their positions on socioeconomic policies but rather their positions on issues that involve in-groups, out-groups, and security from other human beings, especially those human beings who are different from the in-group.

Trump’s politics promise to protect insiders—white, straight, Christian, English-speaking, pledge-reciting Americans—from the threats that he and his followers believe are posed by outsiders—non-heterosexual, non-Christian, non-White, non-anthem-respecting, non-law-abiding, non-Americans. These security, preserve-the-core, identity, in-group/out-group issues are the ones on which Trump’s base differs from the rest of American society—even from many self-identified conservatives (see Table 1). Issues of economic distribution and social behavior are less likely to differentiate—though note that the line between socioeconomic issues on the one hand and nativist/security issues on the other is not always well defined. Some social issues (think trans rights) and economic issues (think distributing government benefits to recent immigrants) are clearly relevant to defending the traditional insider societal core against the intrusions of outsiders and new lifestyles.

The predisposition that undergirds Trump’s base is not a blanket desire for a strong leader—in fact, many Trump supporters worry deeply about centralized power and authority—and it is not a blanket fear of the negative aspects of life—in fact, many Trump supporters are quite sanguine about some negative events such as pandemics and environmental degradation. Rather, the distinguishing predisposition is a desire for their social unit to be homogeneous, united, strong, impervious to outsiders, and well defended against the threats posed by human beings who are outsiders, different, novel, and not beholden to the in-group’s traditional norms, mores, customs, and laws.

A tale of three studies

How did the extensive literatures connecting Trump’s base to authoritarianism and to blanket negativity bias go wrong? With regard to authoritarianism, the explanation is fairly straightforward. Trump and other leaders of his ilk have clear authoritarian tendencies, including a willingness to abuse the judicial system, to constrain the media, to challenge free and fair elections, and to encourage the military to become politically active, so it was natural to assume that the followers of these leaders must crave authoritarianism. Upon further review, it is clear that the followers of these leaders are actually averse to all authority except that which happens to share their policy biases.

With regard to blanket negativity biases, the explanation is more involved, but it can be told by reference to three studies that were conducted in our lab. The first study employed something called the “face-in-the-crowd” paradigm (Mills et al., Reference Mills, Smith, Hibbing and Dodd2014). It works like this: Research participants are shown a computer screen with a large number of small images—perhaps 40 of them—arranged in a grid. With one exception, these images are all of the same thing and that is a human face with a neutral expression. The exception is that in one of the images, the person’s face has a non-neutral expression. In some iterations, this “oddball” expression is angry, and in others, it is happy. Each participant sees dozens of these screens and each time is tasked with hitting the space bar as soon as the oddball expression is spotted. One consistent finding of research employing this protocol is that, on average, participants are faster to hit the space bar when the oddball expression is angry than when it is happy, presumably because from an evolutionary point of view, being able to quickly spot angry people facilitates survival in a way that rapid identification of happy faces does not. Still, there is substantial variation around this average tendency, with some participants being much quicker to spot angry rather than happy faces and others being about the same speed. We wondered what might account for this variation and found a strong correlation with political orientation. Those participants who endorsed a standard array of right-of-center issue positions turned out to be much faster to identify angry faces relative to happy faces than were those participants who preferred left-of-center policy positions.

The second study employed a device known as an eye-tracker that measures precisely where a person’s eyes are looking and thus is an indicator of directed attention. For each participant and for dozens of iterations, we placed four very different smaller images on a computer screen, with each taking up exactly one-fourth of the screen (Dodd et al., Reference Dodd, Balzer, Jacobs, Gruszcynski, Smith and Hibbing2012). Some of these sub-images (beach balls, birthday cakes, people having fun, etc.) previously had been rated positively by a separate group of raters, and some of the images (used toilets, people in distress, wrecked cars, etc.) had been rated negatively by those same raters. We were curious to know whether people devoted more attention to (that is, whether their eyes “dwelled” longer on) the negative than the positive images and found out that they did. But our real interest was in whether, compared to participants with left-of-center policy preferences, participants on the political right were more likely to direct their attention to the negative images, and the evidence indicates that this was indeed the case.

The third study took things a step further and, in retrospect, perhaps a step too far. This study employed a device capable of measuring electrodermal activity (EDA), which is the best and most frequently employed indicator of the sympathetic nervous system—colloquially known as the “fight-or-flight”—response (Oxley et al., Reference Oxley, Smith, Alford, Hibbing, Miller, Scalora, Hatemi and Hibbing2008). We showed people positive and negative images one at a time and recorded the degree to which their EDA changed when viewing each image. Our results indicated that although participants with left-of-center political preferences tended to respond physiologically nearly as much to the positive as to the negative images, participants with right-of-center political preferences tended to have significantly larger EDA responses to the negative than to the positive images.

At first blush, these three studies seem quite similar in that they all are consistent with the notion that conservatives have more of a negativity bias: they spot negative faces more quickly, they spend more time attending to negative images, and they have a larger “fight of flight” response to negative images, suggesting that they are more fearful of negative stimuli. Upon closer inspection, however, each study indicates something quite different and revealing. The results of the first two studies encourage the conclusion that conservatives are particularly attuned and attentive to negative images such as angry facial expressions. Note that these results do not mean there is a fear response attached, as it is possible to be attentive without being fearful, and in fact this is quite likely the type of response that conservatives display. It is only the third study that claims to find an actual fear response, and, as mentioned previously, the findings reported in this study have not been replicated consistently.

Integrating the findings of these three studies and their follow-ups raises the distinct possibility that what really distinguishes conservatives is not that they have a blanket fear response to anything negative, but only that they are more attuned and attentive to negative stimuli without necessarily being fearful. Rather than engaging in fight or flight, the response instead appears to be to engage in sight and might—to attend and deter rather than to fear and flee.

These studies were all done prior to the political rise of Donald Trump but if they had been conducted later and had attempted to distinguish Trump supporters from non-Trump supporters (rather than generic conservatives from generic liberals), I hypothesize that, pursuant to the face-in-the-crowd study, Trump supporters would be significantly quicker to spot angry faces (and think of how much quicker they would have been if the person with the angry expression was an obvious outsider, perhaps a person with a turban). I further hypothesize that, pursuant to the eye-tracking study, Trump supporters would be more likely than non-Trump supporters to attend to negative images and that this difference would be even starker if the negative images depicted human-induced societal turmoil (e.g., gang violence rather than hurricane damage). With regard to the study on physiological response, however, consistent with the replication findings on conservatives generally, I hypothesize that Trump supporters would not give evidence of a greater fear (EDA) response, perhaps even to threatening outsiders, because much of what drives Trump supporters is not an outright fear of outsiders (though some Trump supporters are clearly threatened) but rather a sense that outsiders simply do not belong and must be detected and attended to.

The theory here is that the physiological study is not replicating because conservatives, and especially Trump supporters, do not have a fight-or-flight sympathetic nervous response to negative stimuli generally and perhaps even to outsider human beings in threatening postures. They are simply keenly aware of and attentive to outsiders. Wariness does not have to include fear. Given that EDA resulting from periods of intense concentration or vigilance has been found to be quite different from EDA resulting from fearful or threatening situations, it is also a mistake to assume that, just because EDA tends to increase under threat, it will increase when participants are paying attention and being vigilant (Davies & Parasuraman, Reference Davies and Parasuraman1982). If I am correct that conservatives, and especially Trump venerators, concentrate intently on outsiders, there may be little reason to expect EDA to go up, and this is precisely what the totality of the literature suggests.

In sum, rather than exposing generically conservative individuals to generically negative images and expecting a generically heightened response, research would be more productive if it instead specified a key type of conservative (fervid Trump supporters), exposed them to a particular type of negative image (one that involved human outsiders), and expected a specific and narrow response (one that involved directed attention rather than a sympathetic nervous system response).

Erstwhile Fox News personality and denizen of the political right Tucker Carlson once said, “The U.S. has too many new people. It doesn’t matter whether they’re good people or bad people. They’re strangers and there are too many of them” (Carlson, Reference Carlson2023). The sentiment Carlson expresses is revealing and puzzling to those on the political left, who assume that Trump supporters must want to cut outsiders out of American society because they fear or hate outsiders. To be sure, some Trump supporters are driven by fear and hatred of outsiders, but many more are driven simply by a deeply ingrained preference for a society that is dominated by insiders who are well protected from challengers. As Michelle Malkin, another right-of-center personality, stated, “Diversity is not our strength; unity is” (Malkin, Reference Malkin2019). Research on the motivations of Trump’s intense supporters would benefit from incorporating this observation.