In many circles it is a truism that President Donald Trump is an authoritarian whose electoral appeal was largely founded in ethnocentrism. It is also contended that some of the rising right-wing leaders in Europe are authoritarians appealing to ethnocentrism, like Trump. What we are to make of these characterizations must depend at least in part on just what the terms “authoritarian” and “ethnocentrism” are to be taken to mean.

As is usually the case, understanding the history is clarifying.

The term “authoritarian” is of 19^th century origin but came into currency in connection with the rise of fascist régimes in Europe in the 1920s and the simultaneous emergence of the Bolshevik régime in the Soviet Union. It was not intended as a compliment.

(Those who wanted to emphasize their hostility as much as possible called them “totalitarian,” however.)

The horrors of World War II and the discovery of the soulless atrocities committed by the fascists and especially the Nazis extinguished any vestige of sympathy there might have been for authoritarianism. Social scientists eager to understand how the monstrous blight of fascism and communism could have arisen called their subject of study “authoritarianism,” and in doing so announced their lack of sympathy for those exhibiting it. It wasn’t quite the equivalent of setting out to study the origins of utter depravity or ultimate evil, but it came pretty close.

In the late 1940s, when the search began, it seemed natural to expect to find an authoritarian personality type at the root of authoritarianism, and the evidence at that time suggested that personality was largely formed in childhood and solidified in early adulthood. The Wikipedia article on authoritarian personality gives a reasonably good short summary of the early research, up to the 1990s. It’s fair to say that these efforts left many critical issues very unclear.

More recent efforts, however, have produced insights of not only theoretical but immediately practical importance with critical implications for politics. They are most clearly and forcefully presented in two books:

Kinder, Donald R., and Cindy D. Kam. US against THEM: Ethnocentric foundations of American opinion (Chicago: University of Chicago Press, 2009).

Stenner, Karen. The authoritarian dynamic (Cambridge: Cambridge University Press, 2005).

(Note that Stenner wrote first; Kinder & Kam make a good deal of use of her work while she was largely unaware of theirs when she wrote.)

Briefly, Stenner substantially modifies and greatly clarifies the notion of authoritarianism as a social-psychology phenomenon, while drawing out important political implications. Kinder & Kam (“K&K” for short) introduce a related but (as they show) separate quality of ethnocentrism and discuss its (and authoritarianism’s) political implications in greater depth. Together the books explore a grave challenge to liberal democracy and to American democracy in particular, and demonstrate that the remedies usually talked about are unlikely to be effective—and may even make the problem worse.

It’s not accidental or incidental that these books have arrived only recently, for they are both built very largely on one of the quite recent accomplishments of social science, the building of an extensive data base of public opinion research. Public opinion research began in the 1930s but it’s really only in the past few decades that social scientists have developed the resources to probe public attitudes and beliefs in a consistent, rigorous fashion on a broad scale. These authors depend on these accumulated data to a great extent, analyzing them statistically. Their predominant method is ordinary least squares multilinear regression (OLS/MLR) analysis. This is no longer a “cutting edge” technique, but I believe it is well suited to their purposes and the nature of the data. Readers who are not knowledgeable about statistics can relax, for Stenner as well as K&K do a competent job of statistical analysis and interpret their findings accurately in clear language.

(One caveat is that Stenner appears to forget that age and epoch are confounded in the data—if an effect seems to be due to age it might also be due to the different environments in which people of different ages were reared. I cannot see any place that she tests for this.)

The early assumption that authoritarianism must be a fixed personality trait implied that it was always active, always guiding the authoritarian’s thoughts and actions in most if not all circumstances. But as means to measure authoritarianism were developed and applied to large aggregations of people they showed authoritarianism waxing and waning in irregular cycles over time. This seemed to cast the measures themselves into question.

A Canadian professor, Bob Altemeyer, devoted a great deal of effort to refining a survey questionnaire to measure authoritarianism. By the mid-1980s his “Right-Wing Authoritarianism” (RWA) scale had become the widely-accepted standard for studies of authoritarianism. Altemeyer had thoroughly scrubbed it of the various statistical flaws that had clouded earlier scales, and it passed most statistical quality tests, but still the results varied considerably from time to time. It was also quite restrictive, depending on many specific contemporary American cultural references, and thus of limited value in studying authoritarianism elsewhere. (An abbreviated sample version of the RWA questionnaire may be found here.)

As a graduate student Karen Stenner, with the support and stimulus of her thesis advisor, Stanley Feldman, progressively recast the notion of authoritarianism in a wholly new mold. She came to see it not as a fixed, constant aspect of personality but as a latent underlying predisposition which is activated and takes effect in behavior only as it is engaged by specific stimuli. As these stimuli wax and wane, she hypothesizes, so do outward manifestations of authoritarianism. At least in principle this solves one of the greatest problems of earlier conceptualizations such as Altemeyer’s. However, she does not propose or track specific measures of activation stimuli to show that time-varying fluctuations in levels of authoritarianism follow them.

Stenner’s concept of authoritarianism has everything to do with social groups. Humans are deeply and innately social, and automatically form groups with which we identify profoundly. As very small, preverbal children we are aware of others as being in or out of our ingroup, and we take this identification as natural. As K&K say (p. 33)

Early on, children display an inclination to parse the social world into “natural kinds.” They believe that race and sex and ethnicity belong to the living world, and that differences between races or sexes or ethnicities are rooted in biology, or blood, or some such underlying essence. Such differences encompass inner qualities—temperament, intellect, character—as well as outward, physical ones. Children come to these beliefs on their own. They do not need to be taught that race and sex and ethnicity are natural kinds; they know these things themselves. Children are ready, one might say, for ethnocentrism.

As we mature we become more sophisticated and discerning about matters of group identity, yet very, very few of us ever entirely cast off our sense of belonging to and depending upon our ingroup; the group is a critical part of our sense of who we are.

Acute sensitivity to threats to the integrity, uniformity and conformity of the ingroup lies at the heart of authoritarianism, as Stenner conceives it. Intrusion of individuals who do not conform to the ingroup standards, behaviors or attitudes that deviate from strong ingroup norms, or leaders who hesitate to use strong measures against threats to group integrity or conformity are examples of what Stenner terms “normative threats,” sure to excite authoritarian responses among those so predisposed.

In such matters even the best of theories is of little value unless it can be validated through real-world data, and Stenner devoted a great deal of effort to this. Altemeyer’s RWA scale was of very limited value for her purposes. Not only was it specific to Americans of a certain era but it largely depended on asking direct questions about the kinds of attitudes she wanted to be able to predict on the basis of her concept of authoritarian predisposition. She needed an index of the predisposition itself. What she devised is breathtakingly simple and subtle. Stenner’s  scale of authoritarian predisposition derives from just four related expressions of values in childrearing. Here is the question she asks:

Although there are a number of qualities that people feel that children should have, every person thinks that some are more important than others. I am going to read you pairs of desirable qualities. Please tell me which one you think is more important for a child to have:

• Independence, or Respect for elders
• Curiosity, or Good manners
• Obedience, or Self-reliance
• Being considerate, or Well-behaved?

One beauty of this is that it does not depend on time or place. People have been rearing children to live in society time out of mind. You could expect to get a meaningful answer from virtually any human who has lived since the development of languages complex enough to permit the question to be expressed, anywhere in the world.

Moreover, this question (or questions very close to it) has actually been asked in a variety of social-science oriented surveys fielded over the past few decades, all over the world, so that a great deal of data is potentially available for analysis.

(And I observe that when you know someone reasonably well it is frequently possible to make a pretty good estimate of where the person’s priorities would lie, even if circumstances do not permit asking directly.)

On page 128 of her book Stenner asks, “[I]s it generally true that intolerance of difference is better explained by authoritarianism than by any other variable?” Drawing on the multiyear global data of the World Values Survey she comes to an unambiguous answer (page 133): “[A]uthoritarianism is the primary determinant of general intolerance of difference worldwide.” This is pretty sweeping, but it is consistent with the conclusions of many predecessors for decades before.

Stenner essentially is saying that ethnocentrism and racism are simply particular effects of authoritarianism. Kinder and Kam, however, disagree respectfully but strongly. Ethnocentrism, they believe, is a predisposition that is related to authoritarianism, but largely distinct from it. (pp. 225-27.)

K&K see two components to ethnocentrism. First is ingroup favoritism. This is a low bar; most of us have some feelings of comradeship and affection toward members of our ingroup. At the national level it’s expressed as moderate loyalty and patriotism. It’s also seen in attachment to sports teams, almae maters, home towns, etc.

Ethnocentrism becomes problematic when it extends to outgroup hostility. Ethnocentrism per se involves more or less evenhanded hostility toward all outgroups.

Racism, of course, is virulent hostility toward one or more particular outgroups, defined in terms of perceived racial difference. (Note that real anti-Semites regard Jews as somehow “racially” defined, and similarly with anti-Muslims. At one time the Irish were held to be “racially” different from Scots, which is a good illustration of the ultimate absurdity of racial distinctions.) In practice, K&K find (pp. 206-15), essentially all racists are ethnocentric, but the reverse does not always hold.

To test and support these theories K&K have devised a measure of ethnocentrism, or rather two families of measures. The primary measure is based on unfounded pejorative stereotyping of outgroups, such as notions that blacks are “sneaky,” Jews are “pushy,” etc. (Of course there are some stereotypes that are not inherently negative and for which there is at least some real basis, such as the idea that blacks have dark complexions or Asians have epicanthic folds, but this is not what K&K mean, even when the evidence-based stereotypes are not actually as universal as many may imagine.)

Their secondary measure is negative reported feeling tone regarding outgroups. A main attraction of both these measures is again that they can be calculated from answers to widely-used social science opinion research surveys.

A very important question regarding both authoritarianism and ethnocentrism is where do they come from. Is it the case, as lyricist Oscar Hammerstein wrote ironically in his hit South Pacific song, “You’ve Got to be Carefully Taught,” that

You’ve got to be taught before it’s too late,
Before you are six or seven or eight,
To hate all the people your relatives hate

That’s more or less what social scientists assumed until relatively recently; while there was disagreement on details, ethnocentrism and authoritarianism were generally seen as learned traits. But Stenner, Kinder and Kam all argue persuasively that this isn’t so, that these predispositions are very largely genetic. If children tend to follow their parents in these matters it’s at least as much because of genetic inheritance as it is from learning. K&K rely heavily on twin studies—the gold standard for establishing genetic heritability. Stenner depends on more indirect measures but compiles enough of them to make a good case. It’s all quite consistent with a range of recent findings that stable personality-related characteristics are substantially genetic—not in any rigidly deterministic sense but as a strong tendency (see K&K 254n52). In roughly half of cases the child inherits the parents’ levels of ethnocentrism and authoritarianism or very nearly so even if separated from the family at birth. “[T]he transmission of ethnocentrism from one generation to the next would seem to have more to do with genetic inheritance than with social learning.” (K&K 63-64). And pari passu, it would seem, probably for authoritarianism as well.

Because Stenner and K&K both have good statistical measures of their subjects, K&K (writing later) are able to show that ethnocentrism and authoritarianism are not the same thing. They are related, “congenial,” and somewhat correlated, but distinct. (pp. 64, 86-87, 225-27) This has to be borne in mind when reading some of Stenner’s results; she is not separately controlling for ethnocentrism (for which she had no separate measure) and thus finds it easy to misinterpret her results as showing that it’s an aspect of authoritarianism, or largely so, just as many social scientists had long (but incorrectly) believed. The two predispositions often act in parallel (and perhaps mutually reinforcing) ways, but sometimes pull in different directions.

Both authoritarianism and ethnocentrism seem to be pretty widely viewed as distinctly conservative or right-wing traits. The title Altemeyer gave to his widely-used index, “Right-Wing Authoritarianism,” speaks for itself. Where this leaves Josef Stalin and Fidel Castro is a bit of a mystery to me and indeed I’ve known people who were generally identified as “liberals” by themselves and others who to me seemed distinctly authoritarianism. Nor is there much question that many of the laboring left have been pretty hostile to immigrants. And a liberal icon, President Woodrow Wilson, was a notable racist and segregationist.

Taking the question from the other end, Stenner argues that authoritarianism is only one of three main strands of conservativism. In addition to presenting her case at length in her book, she has published a separate article,

Stenner, Karen, “Three Kinds of ‘Conservatism’,” Psychological Inquiry 20, 2-3 (2009): 142–59.

She opens the article by stating,

When people use the terms conservative or rightwing they typically mean one (or problematically, more) of the following: an enduring inclination to favor stability and preservation of the status quo over social change (what I call “status quo conservatism”); a persistent preference for a free market and limited government intervention in the economy (“laissez-faire conservatism”); or an enduring predisposition, in all matters political and social, to favor obedience and conformity (oneness and sameness) over freedom and difference.

The last predisposition, for oneness and sameness, is of course what she calls authoritarianism (although in an aside she remarks that she would be content to call it simply “difference-ism,” if this would suffice to drop the burden of historical and ideological baggage associated with “authoritarianism”). She sees authoritarianism’s traits as “obedience to authority, moral absolutism, intolerance and punitiveness toward dissidents and deviants” (but would continue on to include the “racial and ethnic prejudice” that K&K show to be more the province of ethnocentrism).

(Interestingly, K&K find, on pp. 59-60, that “Americans who think of themselves as conservative are a bit more ethnocentric, on average, than are those who think of themselves as liberal—though the difference is tiny.”)

Stenner argues that these three kinds of “conservatism” do not form a psychological unit but rather are “packaged” together by political and media opinion elites in response to their own functional as well as psychological imperative to array political views on linear left-right scale. She sees no deeper forces joining these three: “neither aversion to social change nor rejection of market intervention implies, necessitates, or tends to produce generalized intolerance of different races, beliefs, and behaviors.”

That is to say that in Stenner’s scheme of things it is open to other opinion elites to pry apart the three (or four) components of “conservatism” to repackage them in alternative arrangements.

It’s likely, I would take it, that most liberals would much prefer not to try to reharness authoritarianism or ethnocentrism to other ends but to diminish them, persuading those affected (or afflicted) by them of the virtues of moderation and inclusiveness. Superficially Stenner offers some hope along these lines, as do Kinder and Kam (pp. 65-66), with their findings that authoritarianism and ethnocentrism are negatively associated with education: the more educated tend to be less authoritarian and ethnocentric.

But the effects of education seem quite limited at best, and may be largely illusory. There appear to be good reasons to believe that the correlation of education with lower scores on authoritarianism and ethnocentrism may have considerably more to do with the kinds of people who seek and accept education than with the effects of the educative process. Moreover, the data offer no guidance regarding what kinds of education might have the desired effect. Does engineering or medical or military or physical education serve? Or must we insist that everyone study social science? (It might be a good thing in some ways, but I cannot envision it as a practical program.)

Given these questions and the underlying genetic heritability of these predispositions, it seems to me that there can be no alternative but to accept them as latent, at least at some level, in substantial portions of our population—conceivably even in ourselves. (Kinder and Kam dampen expectations regarding “reform” of ethnocentrism on pp. 227-28.) If so the implication is clear that those who aspire to political leadership must either build the strength to meet and defeat authoritarianism and ethnocentrism in direct power contests or else find means to keep ethnocentric and authoritarian concerns within limits that our democracy can tolerate. It is important in this regard to remember that these are by no means fixed and eternally active traits but rather predispositions can remain passive and latent as long as they are not activated by perceptions of threat. As Kinder & Kam observe (p. 201), “Without a clear and demonizable adversary, ethnocentrism never becomes engaged.” (Also see K&K 96, 163-79 regarding activation.)

But what about demagogues? No matter what, some will object, a charismatic demagogue can always stride to the stage and conjure threats to activate ethnocentric and authoritarian fears. This is not a concern lightly to be dismissed, surely. But it’s hard to see how we can outlaw demagoguery without summoning even greater threats to our political order.

The founders of the American political order did an amazing job of understanding the threats, based on nothing but study of history and their own experience, and on that foundation installed mechanisms of defense that continue to serve us very well. But in our far larger and more complex polity today we need to harness the insights and tools of modern social science on a much greater and more intensive scale to understand the real nature of all the predispositions and traits that can influence political action and also to understand the various political vulnerabilities they introduce and what further means can be used to limit and constrain them when they threaten to unbalance or overwhelm our democratic institutions.