Authoritarianism as a concept has been used to describe forms of the government and the state, social organization or structure of groups or cultures, social ideologies and belief systems, styles of leadership and command, and social attitudes, values, and traits characterizing individuals. In social psychology, it is the latter concept – authoritarianism as an individual difference dimension – that has been most intensively studied. This idea of authoritarianism first emerged among social scientists during the 1930s and 1940s to try and explain the rise of fascism and virulent anti-Semitism in Europe at the time. Early explanations drew on both psychoanalysis and Marxism. Wilhelm Reich (1942) proposed that capitalism and sexual repression produced sadomasochistic personalities blending aggression toward the weak and vulnerable with deferential submission to power and authority. Abraham Maslow (1943) and Erich Fromm (1941) also described broadly similar authoritarian personalities whose basic needs attracted them to fascism.

The Theory of the Authoritarian Personality

This idea of an authoritarian personality that was susceptible to right-wing extremism and ethnocentrism, however, remained relatively marginal within the social sciences until the publication in 1950 of an enormously influential volume by Theodor Adorno, Else Frenkel-Brunswick, Daniel Levinson, and R. Nevitt Sanford, titled The Authoritarian Personality. This book reported a program of research that began with the aim of explaining anti-Semitism, but culminated in a far more ambitious theory, which for a time dominated social scientific inquiry into the psychological bases of prejudice and ethnocentrism. Their first major finding was that anti-Semitic attitudes were not held in isolation, but were part of a broader ethnocentric pattern involving a generalized dislike of out-groups and minorities, excessive and uncritical patriotism, and politically conservative attitudes. Their research suggested that this pattern of attitudes seemed to be an expression of a particular personality syndrome consisting of nine tightly covarying traits.

A striking feature of this constellation of nine traits was that they seemed to be directly expressed in “implicitly anti-democratic,” or authoritarian, attitudes and beliefs. On this basis, Adorno et al. (1950) developed their famous F scale consisting of items expressing attitudes which were believed to be direct expressions of each of the nine “traits” of the authoritarian personality syndrome. These nine traits are listed below with gist definitions (in parentheses) and an illustrative F scale item (in quotation marks) for each:

  • Conventionalism (rigid adherence to conventional middle-class values).“A person who has bad manners, habits, and breeding can hardly expect to get along with decent people.”

  • Authoritarian submission (a submissive, uncritical attitude toward authorities). “Obedience and respect for authority are the most important virtues children should learn.”

  • Authoritarian aggression (tendency to condemn, reject, and punish people who violate conventional values). “Homosexuals are hardly better than criminals and ought to be severely punished.”

  • Anti-intraception (opposition to the subjective, imaginative, and tender-minded). “Nowadays more and more people are prying into matters that should remain personal and private.”

  • Superstition and stereotypy (belief in mystical determinants of the individual’s fate, disposition to think in rigid categories). “Someday it will probably be shown that astrology can explain a lot of things.”

  • Power and toughness (preoccupation with the dominance-submission, strong-weak, and leader-follower dimension; identification with power, strength, toughness). “People can be divided into two distinct classes, the weak and the strong.”

  • Destructiveness and cynicism (generalized hostility, vilification of the human). “Human nature being what it is, there will always be war and conflict.”

  • Projectivity (disposition to believe that wild and dangerous things go on in the world, the projection outward of unconscious emotional impulses). “Most people don’t realize how much our lives are controlled by plots hatched in secret places.”

  • Sex (an exaggerated concern with sexual “goings-on”). “The wild sex life of the old Greeks and Romans was tame compared to some of the ‘goings-on’ in this country, even in places where people might least expect it.”

The theory that Adorno et al. (1950) proposed to explain how authoritarian personalities emerged suggested that strict and punitive parental socialization sets up an enduring conflict within the individual. Resentment and hostility toward parental authority, and by extension of all authority, is repressed because of fear of and dependence on the all-powerful parents and replaced by an uncritical submission to and idealization of the parents and therefore for authority in general. The repressed anger and hostility remains but is displaced toward targets seen as sanctioned by conventional authority, such as culturally deviant out-groups and minorities.

Initially this theory was hugely influential, and in the two decades following its publication, the F scale was used in hundreds of studies. These studies did confirm that persons who scored high on the F scale were indeed characterized by right-wing attitudes, political conservatism, nationalism, and generalized prejudice (Duckitt 1992). However, the basic propositions of their theory were not well supported by this research which also revealed important flaws in their F scale. The major criticism of the F scale was all its items were formulated so that “agree” responses indicated higher authoritarianism so that scores were heavily contaminated by the response style of acquiescence (the general tendency for people to agree rather than disagree). When this was corrected, the items of “balanced” versions of the F scale no longer correlated strongly enough with each other to support the idea that they might be measuring a single unitary syndrome or dimension. These criticisms and nonsupportive findings lead to attempts to develop new measures and new theories of the authoritarian personality.

Alternatives to the Theory of the Authoritarian Personality

Three important alternative conceptualizations of an authoritarian personality were proposed by Allport (1954), Rokeach (1954), and Wilson (1973). All three refined the conceptualization of this personality dimension by narrowing its core meaning and discarding Adorno et al.’s (1950) complex psychoanalytic explanation, while Rokeach (1954) and Wilson (1973) also developed new measures to replace the F scale.

Gordon Allport (1954) closely followed Adorno et al.’s description of the kind of personality that would be generally prejudiced against out-groups and minorities but discarded the idea that these traits were the result of inner psychodynamic conflict. Instead he proposed that they stemmed from personal insecurity and fearfulness or “ego weakness” which caused authoritarian personalities to need structure, order, and control in their social environments and to react with hostility to change, deviance, and novelty. Although Allport’s theory never gained the prominence of those of Adorno et al. (1950), his simplified conceptualization of an authoritarian personality was broadly adopted by later theorists, such as Wilson (1973) and Altemeyer (1981).

Milton Rokeach’s (1954) theory of dogmatism proposed that authoritarian personalities were most fundamentally characterized by a rigid cognitive style, that is, “a relatively closed cognitive organization of beliefs … organized around a central set of beliefs about absolute authority” (p. 195). Their dogmatism would attract such individuals to authoritarian movements and beliefs of either the political right or the left and cause them to dislike and reject persons and out-groups with dissimilar beliefs and values. Rokeach (1960) developed a D scale to measure dogmatism in individuals. However, its items did not measure cognitive style directly but were broadly attitudinal like those of the F scale. The items were also all formulated so that agreement indicated higher dogmatism and therefore likely to be contaminated by acquiescent responding. As with the F scale, attempts to balance the D scale (with equal numbers of agree and disagree items) resulted in very little agreement between items suggesting that it will also not be measuring a single unitary personality or cognitive style dimension.

A third alternative theory by Wilson (1973) adopted the term conservatism to describe the authoritarian syndrome. Wilson followed Allport in suggesting that the basic personality characteristic underlying conservatism was a generalized susceptibility to experience threat or anxiety when confronted by uncertainty, resulting in the adoption of authoritarian or conservative social attitudes. Wilson developed a conservatism or C scale to measure this personality and attitudes with items expressing dislike for change, novelty, and diversity and a preference for order, structure, tradition, and convention. While Wilson’s C scale did control acquiescent responding by using equal numbers of agree and disagree items to indicate high conservatism, the correlation between its items was extremely low (r = 0.05), as had been the case for the F and D scales when balanced (Altemeyer 1981). This meant that none of these scales could be validly measuring a unidimensional personality or attitudinal construct.

The major problem plaguing the alternative theories and approaches to authoritarianism was therefore the failure to be able to measure authoritarianism reliably in individuals, and this for a time led to a loss of interest in the construct of authoritarianism in psychology. In the 1980s and 1990s, however, important new research redressed this problem and revived the idea of an authoritarian personality.

Right-Wing Authoritarianism

In 1981 an important book by Bob Altemeyer reported on a program of research leading to the successful development of a fully balanced, unidimensional, and psychometrically sounds right-wing authoritarianism (RWA) scale to measure the authoritarian personality. A crucial finding from this research was that the items of the RWA scale were limited to only three of Adorno et al.’s (1950) original nine traits – authoritarian submission, authoritarian aggression, and conventionalism – because only these three sets of items had covaried strongly enough to be describing a single dimension. Altemeyer suggested that his RWA scale was unidimensional precisely because his successive item analyses had stripped away all items and facets included in the F scale and earlier measures that had been peripheral to these three core traits of the authoritarian personality dimension. His research also confirmed that the RWA scale was a powerful predictor (outperforming earlier measures) of expected effects of authoritarianism in individuals such as generalized prejudice against out-groups and minorities, support for punitive and unjust actions by established authorities against persons or groups regarded as deviant, and holding extreme right-wing attitudes.

Research with the RWA scale also suggested that authoritarianism did not develop in early childhood as a result of punitive parenting but was largely formed through social learning and personal experiences and crystallized during late adolescence (Altemeyer 1981, 1996). Moreover, despite RWA scale scores being generally stable over time, Altemeyer’s research also showed that they could be substantially changed by experiences throughout the life cycle. Thus, RWA decreased with liberal higher education, increased as a result of becoming a parent, and increased in response to societal threats. Research also showed that high scorers on the RWA scale seemed to have been socialized to view the world as a dangerous and threatening place and that this motivated aggressive and punitive authoritarian behaviors.

Social Dominance Orientation

During the 1990s, another important new individual difference construct and measure, social dominance orientation (SDO), was proposed (Pratto et al. 1994; Sidanius and Pratto 1999). The SDO scale taps a “general attitudinal orientation toward intergroup relations, reflecting whether one generally prefers such relations to be equal, versus hierarchical” (Pratto et al. 1994, p. 742). In sharp contrast to the item content of the RWA scale, which expresses beliefs in coercive social control, obedience and respect for existing authorities, and conforming to traditional moral and religious norms and values, SDO items pertain to beliefs supporting social and economic inequality and the right of powerful groups to dominate weaker ones. Not surprisingly, therefore, scores on the SDO and RWA scales were often uncorrelated or only very weakly correlated with each other, indicating that the two scales were measuring different and relatively independent individual difference dimensions.

Despite this, however, it soon became apparent that the SDO scale was as powerful a predictor as the RWA scale of support for authoritarian social phenomena such as nationalism and militarism and support for tough, undemocratic, authoritarian government and of generalized prejudice against out-groups and minorities. Altemeyer (1998) noted that the items of the RWA and SDO scales relate to different sets of the original nine “trait” clusters listed by Adorno et al. (1950) as characterizing the authoritarian personality. He therefore concluded that the RWA and SDO scales measure two different kinds of authoritarian personality dimensions, the submissive and the dominant, respectively.

During the 1990s and the decade that followed, however, an important assumption by Altemeyer began to be seriously challenged. This was his assumption shared by all earlier theorists of authoritarianism that the individual difference dimensions of RWA and SDO were personality dimensions characterized by basic dispositional tendencies to generalized prejudice and authoritarianism.

Reassessing the View of Authoritarianism as Personality

There are a number of reasons why the view of authoritarianism as a personality began to be questioned. First, it was pointed out the items of the RWA and SDO scale, as well as their predecessors, such as the F, D, and C scales, all measured attitudes and beliefs and not behavioral dispositions as the items of personality inventories should (e.g., Duckitt 2001; Feldman and Stenner 1997; Stone et al. 1993). The idea that personality could somehow be inferred from the social attitude and belief items comprising the RWA and SDO scales was therefore merely an assumption that had never been empirically demonstrated or even tested.

Research evidence also supported the contention that authoritarianism measures, such as the RWA Scale, would be better viewed as measuring a dimension of social attitudes and values that might be influenced by personality are not themselves dimensions of personality. First, research by Altemeyer’s (1996) and others has shown that RWA did not seem to be established in childhood as one might expect for a personality trait, but only in late adolescence through a process of social learning. Second, individuals’ level of RWA, although reasonably stable during adulthood, could be markedly altered by changes in their social situations or life experience, such as completing higher education, becoming a parent, or confronting threatening sociopolitical crises. Experimental studies also showed marked change in levels of RWA or equivalent scales as a result of perceived situational threat being manipulated (e.g., Doty et al. 1991; Duckitt and Fisher 2003; Sales 1973) and in SDO by making membership of dominant or high-status social groups salient (e.g., Guimond et al. 2003; Huang and Liu 2005; Schmitt et al. 2003).

Third, studies investigating the structure of sociopolitical attitudes and sociocultural values consistently revealed two roughly orthogonal dimensions of social attitudes and values (see, e.g., Duckitt and Sibley 2009, Table 1). One of these social attitude dimensions was typically labeled social conservatism, traditionalism, or collectivism (values of tradition, social conformity, cohesion, social harmony) and correlated very powerfully with RWA. The other was typically labeled in- or anti-egalitarianism, economic conservatism, or belief in inequality or hierarchy (values of power, dominance, achievement) and correlated very strongly with SDO. This again indicated that RWA and SDO would be better viewed as two social attitudinal or ideological dimensions with each rooted in distinct sets of motivational values.

As a result of these findings, most researchers investigating authoritarianism in individuals had by the turn of the century come to abandon the view of authoritarianism as personality. This raised new questions which lead to the emergence of important new theories.

New Theories of Authoritarianism

The new theories of authoritarianism that emerged around the turn of the century have all explicitly conceptualized authoritarianism (typically RWA but sometimes also SDO as well) not as personality but social attitude or ideological attitude dimensions expressing motivationally based values. In contrast to the earlier personality-based theories, these new theories all give greater emphasis to social or group factors, both as causing the motivational values expressed in authoritarian attitudes and in shaping their effects, such as prejudice.

The group cohesion model (Duckitt 1989) was the first clearly systematized social or group approach to authoritarianism and focused only on RWA. It proposed that the three core components of RWA identified by Altemeyer (1981) were not covarying personality traits, but attitudinal clusters which covaried tightly together because all three expressed the motivational goal or value of group cohesion articulated in attitudes favoring the subordination of individual autonomy and self-expression to group cohesion and authority. This model suggested that the degree to which people value group cohesion and therefore hold authoritarian attitudes would be jointly determined by the degree to which they identified with their social group and perceived threats to it. Out-group dislike or prejudice would therefore be caused by perceiving out-groups as threatening in-group cohesion or security. An important implication of this theory was that right-wing authoritarian attitudes in individuals can be held in respect of any social group. Thus, measures such as the RWA scale or the older F scale would have been measuring authoritarian attitudes only in respect of the national or societal groups, since their items referred only to that kind of group. A later elaboration of the group cohesion model proposed by Stellmacher and Petzel (2005) included the development of a general authoritarianism (GA) scale with items that could be applied to any social group and measure authoritarian attitudes held by individuals in relation to any of the groups one might belong to or identify with, such as one’s school, organization, political party, and so forth. Unfortunately, their interesting approach has not yet generated much research.

A second approach, Kreindler’s (2005) dual group processes model (DGPM), also sees authoritarianism as a group phenomenon and also sees RWA caused by being highly identified with a social group and perceiving threat to that group. It differs from the group cohesion model in that it proposes that the threats that really matter are threats to the norms of the group rather than threats to the group’s cohesion. Out-group prejudice is then caused by persons high in RWA being hostile to persons seen as threatening group norms. The DGPM also sees group identification causing SDO. In this case, it is the identification with high-status or dominant groups that causes high group identifiers to value inequality, hierarchy, and group dominance and so to derogate members of lower status groups. Like the group cohesion models, this theory has also not generated much research.

Another theory influenced by the group cohesion model is the interactionist model proposed by Feldman and Stenner (1997), Feldman (2003), and Stenner (2005). This theory focuses only on RWA and its effects and sees RWA as expressing the value of social conformity. Thus, if persons high in RWA perceive threats to social conformity, this will generate authoritarian reactions in them. This interactive hypothesis has been empirically supported in research on authoritarian reactions such as out-group prejudice and support for hard-line right-wing social and political policies (Feldman and Stenner 1997; Feldman 2003; Stenner 2005). Feldman (2003) has also proposed an interesting disaggregation of the three components of RWA identified by Altemeyer (1981, 1996). Thus, he has suggested the core value of social conformity is best expressed in the authoritarian submission component of RWA, that the authoritarian aggression component is best seen as a conceptually distinct authoritarian reaction, and that the conventionalism component is a form of moral conservatism that is not inherently authoritarian.

A fourth approach, Jost et al.’s (2003) motivated cognition theory of political conservatism, focuses primarily on the causes of conservatism, which it sees as both dispositional and situational. This theory sees conservatism as having two components, with one being attitudinal support for inequality (therefore similar to SDO) and the other being attitudinal resistance to change (therefore an important aspect of RWA). Conservatism, and therefore both resistance to change and support for inequality, is an expression of motivation to manage and reduce threat and uncertainty, which is a joint product of social situational factors likely to activate threat and uncertainty and dispositional factors indexing the strength of personal needs to avoid uncertainty and threat. A meta-analysis by Jost and his colleagues (Jost et al. 2003) showed that indicators of these factors did correlate as expected with RWA, SDO, and other indices of political conservatism (Jost et al. 2003). The findings for SDO in this meta-analysis, however, have been criticized. (Duckitt and Sibley 2010). RWA and related constructs supplied the bulk of the indices used, and their correlations were typically much stronger than those obtained for SDO. Thus, the effects obtained for SDO might have been spurious and due to the positive correlation between RWA and SDO.

Finally, a dual-process motivational (DPM) model provides a broader approach to explaining both RWA and SDO and their effects, which incorporates most of the mechanisms proposed by these new theories. The DPM model proposes that RWA and SDO represent two basic dimensions of social or ideological attitudes with each expressing motivational goals or values made chronically salient for individuals by their worldviews and personalities (Duckitt 2001). High RWA expresses the motivational goal and value of establishing or maintaining collective security, that is, social order, cohesion, and stability. This motivational goal or value is made chronically salient for the individual by the belief that the social world is inherently dangerous and threatening, which is influenced by exposure to and socialization in social environments that are threatening and dangerous. The predisposing personality dimension is social conformity (in Big Five personality dimension terms, low openness and high conscientiousness) which leads individuals to value order, stability, and security as well as influencing their beliefs about how dangerous or threatening their social world may be. Because people high in RWA value collective security and cohesion, they will direct prejudice and hostility toward persons and groups seen as threatening collective security and cohesion.

In contrast, SDO stems from the underlying personality dimension of tough-mindedness versus tender-mindedness (i.e., low agreeableness in terms of the Big Five). Tough-minded personalities view the world as a ruthlessly competitive jungle in which the strong wins and the weak loses. This worldview is influenced by exposure to and socialization in social environments characterized by inequality, group dominance, and competition over power, status, and resources. Being tough-minded and holding this competitive-jungle worldview makes chronically salient the motivational goals and values of power, dominance, and superiority over others, which are expressed in the social attitudes of SDO. Because persons high in SDO value power, dominance, and superiority over others, they will derogate and dislike out-groups low in power or status (in order to justify their own relatively superiority) and be hostile to out-groups competing with their own group over relative power and status. The DPM model of the nature and origins of RWA and SDO and of their effects on prejudice and other authoritarian outcomes has been supported by a great deal of research in the past decade and a half (for recent reviews, see Duckitt and Sibley 2010: Duckitt and Sibley 2017).

Conclusions

The idea of an authoritarian personality arose early in the twentieth century to explain patterns of relatively stable individual differences in a broad range of social, political, and intergroup attitudes which seemed to dispose those individuals to prejudice and ethnocentrism. Initially researchers tried to measure the entire range of attitudes and beliefs originally deemed to comprise the authoritarian personality syndrome on a single psychometric dimension, such as Adorno et al.’s (1950) famous F scale. However, these early attempts failed because the content of the syndrome proved to be simply too heterogeneous to be reflecting a single trait dimension. Later research, such as that of Altemeyer, then revealed that this broad social attitudinal domain comprised two distinct dimensions, which are today best measured by the RWA and SDO scales and seem to comprehensively organize individuals’ social, political, and ideological attitudes.

And finally, a more recent change has involved challenging the conception of authoritarianism as personality, be it one dimension or two. The new approaches see the two dimensions of authoritarianism, RWA and SDO, not as personality dimensions but rather as two distinct social attitudinal dimensions expressing two sets of motivationally based social values. These newer theories focus on trying to clarify the values that lie at the core of RWA, how social environmental factors influence RWA and SDO on their own or in conjunction with personality, and how and why RWA and SDO influence and have their effects on social, political, and intergroup attitudes and reactions.

Cross-References