read the article, focusing on the the results and significance

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/227634076

The Presence of Left‐Wing Authoritarianism in Western Europe and Its

Relationship with Conservative Ideology

Article  in  Political Psychology · October 2006

DOI: 10.1111/j.1467-9221.2006.00532.x

CITATIONS

64 READS

1,163

3 authors:

Some of the authors of this publication are also working on these related projects:

Entitlement, belief in life as zero-sum game and subjective well-being: 42-nation study View project

Multitasking View project

Alain Van Hiel

Ghent University

166 PUBLICATIONS   4,558 CITATIONS

SEE PROFILE

Bart Duriez

Karel de Grote-Hogeschool

89 PUBLICATIONS   4,435 CITATIONS

SEE PROFILE

Małgorzata Kossowska

Jagiellonian University

86 PUBLICATIONS   1,222 CITATIONS

SEE PROFILE

All content following this page was uploaded by Bart Duriez on 27 September 2018.

The user has requested enhancement of the downloaded file.

The presence of left-wing authoritarianism in Western Europe

and its relationship with conservative ideology

Alain Van Hiel, Bart Duriez, and Malgorzata Kossowska

Author Note: Alain Van Hiel, Department of Developmental, Personality and Social Psychology, Ghent

University (Belgium); Bart Duriez, Department of Psychology, Katholieke Universiteit

Leuven (Belgium); Malgorzata Kossowska, Institute of Psychology, Krakow (Poland).

Running Head: Left-Wing Authoritarianism

Correspondence should be sent to Alain Van Hiel, Department of Developmental, Personality and Social

Psychology, Henri Dunantlaan 2, B-9000, Ghent, Belgium. E-mail: alain.vanhiel@UGent.be

1

The presence of left-wing authoritarianism in Western Europe

and its relationship with conservative ideology

Abstract

The presence of left-wing authoritarianism (LWA) as well as its relationship with right-wing authoritarianism

(RWA) and conservative ideology was tested in three Flemish samples. In the first study conducted on a

sample of ordinary voters (N = 208), a newly developed LWA scale was found to be internally consistent

and to show high construct validity. In the second study, another voter sample (N = 264) and a sample of

political activists (N = 69) were tested. In the two samples of ordinary voters, only few people obtained high

LWA scores. Moreover, the aggression and submission items did not load on distinct components and LWA

was positively related to RWA and cultural conservatism, and negatively to economic conservatism.

Conversely, in the political activist sample high LWA scores were common among left-wing extremists and

evidence was found for a two-dimensional LWA aggression-submission structure. LWA was negatively

related to RWA, cultural conservatism, and economic conservatism. The concept of LWA and its theoretical

underpinnings are discussed.

Key Words: communism; cultural conservatism; economic conservatism; left-wing authoritarianism;

political beliefs; right-wing authoritarianism

2

The presence of left-wing authoritarianism in Western Europe

and its relationship with conservative ideology

The Authoritarian Personality by Adorno, Frenkel-Brunswik, Levinson, and Sanford (1950) can be

considered as one of the cornerstones of political psychology. Research overviews in Advances of

Experimental Social Psychology (Altemeyer, 1998; Duckitt, 2001) and Political Psychology (Martin, 2001;

special issue edited by G. E. Marcus for Political Psychology) attest the recent interest in the authoritarian-

ism concept. Among authoritarianism researchers, one of the most debated issues is whether

authoritarianism is typical at the extreme right-wing side of the political spectrum only (e.g., Altemeyer,

1996; Stone, 1980; Stone & Smith, 1993), or whether authoritarianism can also be detected among

adherents of extreme left-wing ideology (e.g., Eysenck, 1954, 1981; Ray, 1983). Traditional approaches

tried to develop or identify measures on which left-wing and right-wing extremists would score higher than

moderates (e.g., Eysenck, 1954; Rokeach, 1960; Tetlock, 1983; Sidanius, 1984). However, these attempts

were not entirely successful.

Recently, Altemeyer (1996) constructed a left-wing authoritarianism (LWA) scale that structurally

resembles his right-wing authoritarianism (RWA) scale (Altemeyer, 1981). Both instruments measure the

attitudinal clusters of authoritarian submission, authoritarian aggression, and conventionalism, but whereas

the RWA items refer to established authorities, in the LWA scale these items are embedded in the context

of a left-wing revolutionary cause. After investigating data of 2,544 Canadian participants, Altermeyer (1996)

concluded that he could not identify one single left-wing authoritarian, calling LWA somewhat sceptical the

“Loch Ness monster of political psychology.”

In the present study, we will try to show that LWA does exist, but that it is only present in very

specific groups. That is, we expect that LWA can be found among activists of extremist parties (i.e., among

members of extreme left-wing parties and, to some extent, among anarchists). However, we do not expect

to find LWA in samples of ordinary citizens, nor among activists of “established” political parties. The search

3

for left-wing authoritarians constitutes our first aim, but the ideological correlates of LWA and RWA are

investigated as well. In particular, it is investigated whether RWA and LWA are differentially related to

cultural and economic conservatism.

Classic studies on left-wing authoritarianism

Various authors have criticised The Authoritarian Personality because it was restricted to the

problem of right-wing extremism (e.g., Eysenck, 1954; Rokeach, 1960). These authors asserted that

fascists and communists have many attitudes in common that oppose the value systems of democrats. The

position that communists as well as fascists share traits similar to the ones described in The Authoritarian

Personality has become known as extremism theory or authoritarianism of the left theory (see, Durrheim,

1997-a; Sidanius, 1988). Early contributions in this tradition tried to identify personality dimensions that

characterize extremists of whatever political stance. In order to achieve this aim, Eysenck (1954) extracted

two factors from the correlations among 40 attitudinal statements. The first dimension was interpreted as

liberalism versus conservatism. The second dimension was labelled toughmindedness versus tendermind-

edness. Eysenck (1954; Eysenck & Coulter, 1972) showed that moderates generally obtain low

toughmindedness scores, whereas extremist groups such as communists (N = 43) and especially fascists

(N = 43) obtained higher scores. However, Eysenck’s study has been severely criticized because the F-

scores reported for the moderate group were the lowest obtained so far (Christie, 1956). Moreover,

Rokeach and Hanley (1956) argued that Eysenck’s (1954) results could be explained on the basis of the

content of the toughmindedness scale, which was composed of anti-religiosity and anti-humanitarianism

items. Thus, Rokeach and Hanley (1956) argued that communists obtain high scores on this scale because

they express agreement with the anti-religiosity items, whereas fascists obtain high scores because they

agree with the anti-humanitarianism items. As a consequence, adherents of both extremist groups are likely

to obtain higher toughmindedness scores than moderates who are likely to reject all these statements.

In an attempt to overcome the latter problem, Rokeach (1960) developed the dogmatism scale to

measure ideology-free authoritarianism. He obtained somewhat higher though non-significant dogmatism

4

scores in communists (N = 13). However, in a study of the Italian Parliament, DiRenzo (1967) obtained the

highest dogmatism levels among neo-fascists (N = 24), whereas extreme left-wingers (N = 25) obtained the

lowest scores. Knutson (1974) obtained similar results when studying the governing bodies of six American

political parties, ranging form the Communist Party (N = 11) to the neo-nazi American Socialist White

People’s Party (N = 13). In line with this, Rokeach (1960) obtained high positive correlations (.54 < r < .77)

between dogmatism and the F-scale, and other researchers reported positive correlations between

dogmatism and Altemeyer’s (1981) RWA scale (e.g., Van Hiel & Mervielde, 2002-a).

Whereas the aforementioned studies tried to identify LWA and RWA through the use of attitudinal

statements, Tetlock (1983, 1984, 1986) and Sidanius (1984, 1988) tried to establish the relationship

between extremism and cognitive functioning. Tetlock conducted a series of studies on the relationship

between political ideology and integrative complexity. Integrative complexity refers to two major structural

characteristics: (1) the degree of differentiation of cognitive elements, and (2) the degree of integration or

interrelatedness of these elements. Tetlock’s (1983, 1984) research, which was conducted on elite samples

from the US senate and the British House of Commons, revealed that advocates of center-left ideology

exhibit higher levels of integrative complexity than ‘extreme conservatives’ and ‘extreme socialists’. In

contrast, two studies conducted by Sidanius (1984, 1988) addressed the hypothesis that extremists show

higher levels of cognitive complexity. Cognitive complexity was measured by the political prediction test in

which participants had to estimate the degree of political rioting and murder likely to occur on the basis of

six items of information. Contrary to Tetlock’s findings, Sidanius reported greater cognitive complexity and

political interest in extremists. Hence, it can be concluded that the cognitive perspective on political

extremism yielded contradictory results (see, Durrheim, 1997-a, 1997-b; Van Hiel & Mervielde, 2003).

In sum, some authors took the existence of LWA for granted whereas others concluded that LWA

is a myth, and a fierce debate developed among scientists on the characteristics of political extremists (e.g.,

Christie, 1956; Eysenck, 1954; 1981; McCloskey & Chong, 1985; Ray, 1983; Rokeach & Hanley, 1956;

Stone, 1980; Stone & Smith, 1993). However, hardly any empirical data were available for this debate and

5

according to Stone and Smith (1993), many political psychologists “base their case on intuitive evidence …

concerning apparent similarities between regimes of the far left and far right, rather than on a systematic

review of the empirical data on any personality and ideology” (p. 154).

Recent developments

Two recent lines of investigation tried to advance the debate on LWA. First, the fall of the commu-

nist regime in the Soviet Union and Eastern Europe in the late 1980s provided extremism theorists with an

excellent chance to prove that they were right. As expected by extremism theory, Hamilton, Sanders, and

McKearney (1995) and McFarland, Ageyev, and Abalakina-Paap (1992) showed that (right-wing)

authoritarianism was positively related to support for communism. In this respect, it is important to know that

current thinking on RWA has evolved to a “new position” quite similar to the position advocated by

extremism theorists. For example, Altemeyer (1996) argues that when he “began talking about “right-wing”

authoritarianism, I was (brazenly) inventing a new sense, a social psychological sense that denotes

submission to the perceived established authorities in one’s life” (p. 218). This definition of authoritarianism

leads to the prediction that adherents of hard-line communist ideology in the former Soviet Union should

evince high RWA levels, whereas extreme left-wingers in Western countries should obtain low scores.

According to this point of view, ideal support for extremism theory would be obtained in samples of extreme

left-wing party members who try to overthrow an established right-wing bourgeoisie regime.

A second line of investigation that might advance the ongoing debate is constituted by Altemeyer’s

(1996) attempt to develop an LWA scale. According to Altemeyer, “psychological right-wingers … support

the perceived established authorities in society, and psychological left-wingers … oppose them” (p. 218).

However, not all psychological left-wingers can be considered authoritarian. Some of them are independent

individuals who want peaceful social reform and do not exhibit the typical authoritarian attitudes, whereas

other psychological left-wingers can be considered true (left-wing) authoritarians who want to seize all the

power themselves and do exhibit authoritarian attitudes.

Altemeyer (1996) defines LWA as the covariation of three attitudinal clusters: (1) Authoritarian

6

submission (= a high degree of submission to authorities who are dedicated to overthrowing the established

authorities), (2) Authoritarian aggression (= general aggressiveness directed against the established

authorities or against persons who are perceived to support those authorities), and (3) Conventionalism (=

adherence to the norms of behavior perceived to be endorsed by the authorities of a revolutionary

movement). These attitudinal clusters are reminiscent of the definition of RWA, which is also constituted by

these three components. Thus, although both left-wing and right-wing authoritarians would indicate that

they submit to their respective authorities, aggress in their names, and adhere to their conventions, their

political ideology is totally incompatible. In other words, although both LWA and RWA tap into a similar

underlying structure, there are huge ideological differences in the content of their items.

Altemeyer (1996) developed an LWA scale that showed sufficient internal reliability (the

Cronbach alpha of his final scale equalled .86 in a voter sample). However, in various Canadian

samples of students (total N = 1,845) and their parents (N = 642), as well as among candidates of

political parties (N = 67), Altemeyer could not identify a single person that could be classified as left-

wing authoritarian according to his LWA scale (not one person scored on average at least 6.0 on 9-point

scales). Moreover, somewhat surprisingly, positive correlations (.11 < r < .18) between LWA and RWA

were obtained as well.

Problems with Altemeyer’s studies on left-wing authoritarianism

Although Altemeyer (1996) succeeded in developing a reliable scale, he did not find one single

high-scoring left-wing authoritarian among 2,000 respondents. In our opinion, two problematic features of

his research may account for this failure. Firstly, Altemeyer did not include adherents of extreme (left-wing)

movements in his sample. Secondly, there are some conceptual problems with Altemeyer’s LWA scale.

Particularly the concept of left-wing conventionalism is unclear.

With respect to the first issue, it is possible that Altemeyer’s failure to identify left-wing

authoritarians was due to the lack of appropriate samples. That is, it is possible that the prevalence of

LWA is limited to adherents of some small “exotic” parties such as, for example, communist movements,

7

whereas LWA is absent among ordinary citizens and activists of “established” political parties. In the

present studies, we therefore administered data in a political activist sample as well (Study 2).

The question arising with respect to the second issue is “what does left-wing conventionalism

mean?” Webster’s (1996) dictionary defines “conventionality” as ”Adherence to established norms,

customs, or usages; conformity to the accepted and traditional” (p. 284). However, Altemeyer (1996)

defines conventionalism as adherence to the norms of behavior perceived to be endorsed by the authorities

of a revolutionary movement, or “to have rules and ‘party discipline’ that must be followed” (pp. 219-220).

However, because obedience to authorities and rules is explicitly mentioned by this definition, the concept

of left-wing conventionalism is hardly distinguishable form the definition of submission.

Moreover, conventionalism is a belief system that essentially opposes ideologies that promote

societal change. Not surprisingly, then, previous political psychological literature on left-wing extremism

barely mentions the conventionalism concept. Conversely, the authoritarian aggression and submission

have been amply discussed in previous accounts of LWA. For example, Eysenck (1980-81) referred to left-

wing extremists in terms of their ruthless oppression of everyone who opposes them and in terms of their

obedience to party discipline (that is, in terms of aggression and submission). We therefore hypothesize

that high scores on both the aggression and submission component are needed in order to diagnose a

person as a true left-wing authoritarian. Anarchists, for example, are expected to fight any regime as long as

they consider it to be authoritarian, even if this regime could be classified as left-wing. They can therefore

only be classified as “left-wing aggressive” but not as left-wing authoritarians, because true left-wing

authoritarians are expected to defend and submit themselves to left-wing regimes, even (or even especially)

when these regimes are authoritarian in nature. According to Altemeyer (1996), “Should their movements

attain power in the flash of a revolution, their strong submission to the new societal authority would make

them psychological right-wing authoritarians on that level” (p. 218). Hence, only people with a high

authoritarian submission level constitute the raw materials of a left-wing authoritarian regime.

8

The present studies

Because conceptual problems with Altemeyer’s (1996) LWA scale may have prevented the

identification of left-wing authoritarians, in the present studies a new LWA scale was developed in order to

test the presence of left-wing authoritarianism in Western Europe (in particular, Flanders, Belgium). In Study

1, the scale’s internal consistency and factorial structure was assessed in a sample of ordinary voters.

Moreover, in order to test whether the scale can be considered a “true authoritarianism measure”, a right-

wing version for each item was written and we computed the correlation between these new items and

RWA. Because it is also important to test this scale in appropriate samples, in Study 2, the LWA scale was

administered in a sample of voters and in a sample of political activists. Study 2 thus allows us to test the

stability of the psychometric analyses conducted in Study 1, but more importantly, we were able to assess

LWA’s validity by the use of groups that can be expected to have high scores on authoritarian aggression

and authoritarian submission. In particular, LWA was assessed among communists (who can be expected

to score high on both of the LWA facets) and anarchists (who can be expected to score high on

authoritarian aggression, but low on authoritarian submission).

In study 2, we will also investigate the relationship between LWA and other indicators of political

beliefs. In particular, we aimed to test how LWA relates to RWA and cultural and economic conservatism

(see Lipset, 1981; Middendorp, 1978; Van Hiel & Mervielde, 2004; Wilson, 1973). From an economic

perspective, progressive ideology emphasizes equality and rejects inequality of the distribution of

power, income, and opportunities. Economic progressive ideology therefore favors issues such as

worker participation, state economic intervention and trade unionism. Economic conservatism favors

adherence to capitalist ideology, private initiative and unrestricted competition among individuals. In the

cultural domain, progressive ideology stands for the freedom to arrange one’s life according to one’s

own insights, whereas conservatism endorses adherence to traditional values and norms. Cultural

conservatism thus favors authoritarian parent-child relationships, traditional work ethics, and

conventional female roles (see, Middendorp, 1978). Duriez, Van Hiel, and Kossowska (in press) have

9

shown that RWA is predominantly related to cultural conservatism (see also, Duckitt, 2001), whereas its

relationship with economic conservatism is much less pronounced. However, given the anti-capitalist

content of the economic conservatism scale and the historical opposition between communist and

capitalist ideology, one would especially expect a highly negative correlation between LWA and

economic conservatism.

STUDY 1

The items pertaining to Altemeyer’s (1996) LWA scale were composed to simultaneously measure

aggression, submission, and conventionalism, or at least two of the three authoritarianism components.

Because we did not want to include the conventionalism component in our scale, a majority of Altemeyer’s

items were not suited for the present research. Therefore, we selected some items from Altemeyer (1996)

that referred unambiguously to either authoritarian aggression or authoritarian submission. When

necessary, these items were stated in a more direct language.

We selected items (shown in Table 1) that unambiguously promote the use of violence (= the

aggression facet) or the necessity to obey to left-wing leaders (= the submission facet). Eight items were

selected after discussion among the three authors of the present article. Three items were literally drawn

from Altemeyer’s questionnaire (item 3, 5 and 6) and five items were modified (item 1, 2, 4, 7, and 8). This

strategy left us with eight LWA items: Four of which refer to authoritarian aggression (two protrait and two

contrait items), and four of which refer to authoritarian submission (two protrait and two contrait items). The

aggression items explicitly use the term “violence.” These items thus unambiguously champion the use of

violence against the relevant outgroup, and their central idea was about the use of violence (or not using it

in the case of reversed items). Three of the four submission items explicitly used the terms “conformity”,

“obedience” and/or “submission”, whereas the fourth item mentions “to tell how to think and act.” The

principal idea reflected by these items is therefore the concept of authoritarian submission.

Another aim of the present study was to test the construct validity of the LWA scale. An obvious

way to proceed is to compare our scale with Altemeyer’s (1996) LWA measure. However, because

10

Altemeyer did not succeed to identify one single left-wing authoritarian and because he never actually

tested his scale in target groups that can be expected to show heightened levels of LWA, the construct

validity of this scale remains doubtful. In addition, his scale has not been used by other researchers. For

these reasons, in order to test whether the content of our items really reflect “authoritarianism” (and not

something else), we decided to construct a right-wing version of our items. These items were written as

close as possible to our LWA items, with the exception that from a right-wing perspective the targets

belonged to a traditional outgroup (e.g. communists, anarchists). We then computed a correlation between

this right-wing version of our scale and Altemeyer’s traditional measure of right-wing authoritarianism.

Method

Participants

The voter sample was recruited by 24 undergraduate students in political science who asked their

adult neighbours to participate in order to obtain a heterogeneous sample. A total of 240 envelopes and

questionnaires were distributed (10 per student) of which 208 (86.7%) were returned. The students were

told that they would receive extra credit when they collected at least seven questionnaires for a study on

personality, beliefs and political opinions. They were prohibited to make notes on the package given to their

respondents, and were told that the aims of the investigation would be explained to them in detail at the end

of the semester. The accompanying letter for the participants introduced the study as an investigation of

“personality, beliefs, and political viewpoints.” It also read that the study was a joint project of Ghent

University, the Catholic University of Leuven and the Jagiellonian University (Poland). Although anonymity

was explicitly guaranteed, students asked participants’ permission to write their names, addresses and

phone numbers on a separate paper. Participants were asked to answer the booklets individually, and to

return these booklets to the neighbour student in a closed envelope. To make sure that the student didn’t

complete the booklets themselves, we randomly selected 10% of the respondents, gave them a telephone

call, and asked them whether they could remember having filled out our questionnaire. All of the contacted

participants assured us that they had actually taken part in our study.

11

The sample consisted of 51% males, 41% females and 10% who did not indicate their gender. The

mean age was 38.98 years (SD = 16.00). Of these participants, 120 had completed higher education, 66

had completed secondary school, 4 had completed their education at age 14, and 18 did not indicate their

educational level. Fifty-two participants indicated that they had a lower than average income, whereas 109

reported a higher income and 47 did not indicate their income.

Measures

All items were rated on 5-point Likert scales (1 = completely opposed, 3 = neutral, 5 = completely in

agreement).

Left-wing authoritarianism. The Left-Wing Authoritarianism (LWA) scale consisted of 8 items

divided over the 4-item aggression and the 4-item submission facet scales (see Table 1). Analogous to

Altemeyer’s (1996) LWA scale, the items were proceeded by the following statement: “In the items that

follow, the ‘Establishment’ refers to the people in our country who have traditionally had the most power and

the greatest control over the economy: That is, the wealthiest people, the large corporations and the banks,

who are often called the right-wing forces. A ‘revolutionary movement’ denotes a Left-wing movement

dedicated to overthrowing the Establishment, and taking away its power.”

Right-wing version of LWA A right-wing version of our 8 LWA items was constructed by

substituting “the Establishment” by typical right-wing outgroups. For example, item 1 was written: “I agree

with the idea of repressing – with or without violence – individuals who want to overthrow our society in order

to establish a communist regime”, item 2 “The police is justified in using violence against left-wing

revolutionaries and anarchists because they will never give up their struggle peacefully” and item 4 “The

leaders of our country are justified in demanding obedience and conformity of the citizens.”

Right-wing Authoritarianism (RWA) Participants completed an 11-item RWA scale

(Altemeyer, 1981; translated by Meloen into Dutch language, 1991; Cronbach α = .87) that has been

successfully used in previous studies in Flanders (e.g. Duriez, 2004; Duriez & Van Hiel, 2002; Duriez et

al., in press; Kossowska & Van Hiel, 2003; Soenens, Duriez & Goossens, in press).

12

Results

Psychometric analyses of LWA

As can be seen in Table 1, reliability analyses of the 8-item LWA scale revealed sufficient internal

consistency (Cronbach alpha = .77). Given the low number of items for each facet scale and the fact that

these adapted items have not been subjected to psychometric analyses before, the Cronbach alphas for

the aggression (.62) and the submission (.67) facet scale were acceptable. The correlation between the

aggression and submission facet scale was substantial, r = .58, p < .001.

Insert Table 1about here

Next we conducted an explorative Principal Component Analysis on the intercorrelations among the 8 LWA

items. We expected that the aggression and submission items would load on a distinct component. Two

components had an eigenvalue greater than 1.00. The two components were OBLIMIN rotated allowing a

correlation between them. As can be seen in Table 2, two method components emerged of which the first

(explaining 37.14% of the total variance) consisted of positively keyed items and the second (explaining

30.45% of the total variance) consisted of negatively keyed items. Therefore, it must be concluded that

although the complete LWA scale can be used as a reliable measure of left-wing authoritarianism in

samples or voters, the aggression and submission facet scales should not be used as separate subscales.

Insert Table 2 about here

The presence of LWA

First we checked whether the demographic variables have an effect on LWA. No significant sex

differences for were obtained (F(1, 183) = 0.34, ns.) and the correlation between age and LWA (r = -.12,

ns.) was non-significant. Educational level did yield a significant effect (F(1, 184) = 9.05, p < .001), with

the lower educated obtaining higher LWA scores (M = 2.75) than those who completed secondary (M =

2.26) and higher (M = 1.95) education. However, these results should be interpreted with caution

because the lower educated group consisted of 4 participants only. The mean level of LWA in the present

sample (M = 2.07, SD = 0.59) was rather low.

13

Are there any “true left-wing authoritarians” in this sample? A total of 185 participants (91.6% of

the sample) obtained a score below the theoretical midpoint (i.e., a score below 3.00), whereas only 9

participants (4.5%) had a score on the theoretical midpoint (i.e., a score of 3.00), and only 8 participants

(4.0%) had a score between 3.01 and 3.99 (the highest score being 3.38). In sum, given the lack of high

scoring left-wing authoritarians, the present results were comparable with previous results obtained by

Altemeyer (1996).

The right-wing version of LWA

The internal consistency of the right-wing version of the LWA scale was comparable to the LWA

scale (Cronbach alpha = .76). However, compared to its original counterparts, the aggression facet scale

showed higher internal consistency (Cronbach alpha = .73), whereas the submission facet scale had a

Cronbach alpha of .61 only. The total scale score was highly correlated with RWA (r = .55, p < .001), as well

as with the aggression and submission facet scales (r = .51 and .44, ps < .001). Given the less than perfect

internal consistencies of these scales, correction for attenuation increased the magnitude of these

correlations (r = .68, 64, and .60 respectively). In sum, these analyses reveal that the right-wing version of

our LWA scale is highly correlated with a traditional RWA scale, and suggest that our LWA items sufficiently

tap the authoritarianism concept. In accordance with Altemeyer (1996) we obtained a positive correlation

between LWA and RWA (r = .25, p < .001).

Discussion

The aims of the present study were twofold: (1) to show the internal consistency of the scale

and the validity of its facet scales, and (2) to show its construct validity. Pertaining to the first aim, the

results revealed that our LWA scale showed sufficient internal consistency, but we were not able to

substantiate the hypothesized underlying structure of the aggression and submission facet scales. With

respect to the second goal, the present analyses revealed high convergent correlations between the

right-wing version of our LWA scale and Altemeyer’s RWA scale. In fact, these correlations are quite

impressive given the fact that it was our explicit aim to avoid any explicit reference to conventionalism in

14

our LWA items, whereas conventionality is clearly present in the RWA scale.

STUDY 2

The previous study has shown sufficient reliability and validity of our LWA scale. However,

analogous to the studies of Altemeyer, we did not find any evidence of left-wing authoritarianism. Based

on the accumulating evidence, one could therefore conclude that left-wing authoritarianism probably

does not exist. In this second study, we will try to replicate the findings of the first study. However, a

possible reason for this failure to detect left-wing authoritarianism is the use of a sample of voters. It is

possible that left-wing authoritarianism can be found among members of parties of extreme left-wing

signature only. In a similar vein, Study 1 did not substantiate the two-dimensional aggression-

submission structure of left-wing authoritarianism. However, it is possible that these items constitute a

separate dimensions in a political activist sample. If this is the case, the inclusion of anarchists in such a

sample allows us to validate the two facet scales. Whereas Communists can be expected to obtain

higher scores on both aggression and submission than activists of other political parties, anarchists are

expected to obtain a higher score on authoritarian aggression only.

In Study 2 we tested the following hypotheses.

Hypothesis 1. Given the failure to detect left-wing authoritarians in previous research on

Western samples and in our first study, Hypothesis 1 states that LWA should not (or hardly) be present

in a second convenience sample either. Western countries never had a communist regime, and

communist ideology never attracted a great number of followers. Therefore, it should come as no surprise

that previous research failed to detect left-wing authoritarians in Western countries.

Hypothesis 2a and 2b. Hypothesis 2a states that members of extreme left-wing parties are likely

to obtain higher overall LWA scores than activists of other political parties. In particular, it can be expected

that, in contrast to average citizens, members of communist parties will evince scores above the theoretical

midpoint for LWA and its facet scales. In addition, Hypothesis 2b states that anarchists, who also fight the

established authorities, will evince scores above the theoretical midpoint for the aggression facet scale.

15

However, their scores on authoritarian submission are expected to be low, because anarchists reject

submission to authority in general and the legitimacy of the State in particular. Therefore, they are expected

to fight any regime they consider authoritarian, even if this regime could be classified as left-wing. In short,

hypothesis 2 states that (a) the aggression scale should allow us to differentiate communists and

anarchists from other ideological groups, whereas (b) the submission facet scale should allow us to

differentiate between communists and anarchists.

Hypothesis 3. Duriez et al. (in press) have shown that, whereas RWA is predominantly

related to cultural conservatism (see also, Duckitt, 2001), its relationship with economic conservatism is

much less pronounced. Given the anti-capitalist content of the economic conservatism scale and the

historical opposition between communist and capitalist ideology, one would especially expect a highly

negative correlation between LWA and economic conservatism. Hypothesis 3 therefore states that RWA

should be particularly related to cultural conservatism, whereas LWA should be particularly related to

economic progressivism.

Method

Participants

Following exactly the same procedure as in Study 1, a voter sample was recruited by under-

graduate students. In this way, a total of 320 questionnaires were distributed of which 264 (82.5%) were

returned. This voter sample (N = 264) consisted of 56% males with a mean age of 42.86 years (SD =

14.61). Of these participants, 132 had completed higher education, 84 had completed secondary school, 32

had completed their education at age 14, and 16 did not indicate their educational level.

In addition, a political activist sample was recruited on a voluntarily basis by another group of

political sciences students. These students were asked to individually contact people they knew to be

member of a political organization. In this way, a total of 87 booklets were distributed of which 69 (79.3%)

were returned. The political activist sample (N = 69), consisted of militants of various political movements.

One of our students was acquainted with extreme left-wing organizations and was able to collect 20

16

communist participants. These communists were either affiliated to the Stalinist “Partij Van De Arbeid”

(PVDA; N = 14) or to the neo-marxist Communist Party (N = 6). The anarchists (N = 21) were recruited by

the second author who is well acquainted with the anarchist movement in Leuven, Belgium. The anarchists

were active in the anarchist movement and defined themselves as such. Right-wing extremists were

supporters of the “Vlaams Blok” (N = 11), a party that is very similar to other extreme right-wing European

parties such as the Centrum Partij in the Netherlands, Le Pen’s Front National in France, and the Republi-

kaner in Germany (Ignazi, 1992). Members of all other “traditional” parties (the green, christian democrat,

socialist and conservative party, for a description of these parties and their support, see Van Hiel &

Mervielde, 2002-b) are referred to as “moderates” (N = 17). The questionnaires were distributed on a

personal basis, participants were asked to complete the booklet individually, to put it in an envelope, and to

return it to the person that gave them the booklet. The accompanying letter for the participants introduced

the study as an investigation of “personality, beliefs, and political viewpoints.” It explicitly read that we were

interested in their personal opinions. In fact, although participants were told that they were asked to

collaborate because they were interested in politics, they were unaware of the fact that, in most cases, they

were asked to collaborate because they belonged to a specific ideological movement. Anonymity was

explicitly guaranteed.

Detailed demographic information on this sample was also available. The sample consisted of 74%

males and 26% females. The mean age was 38.96 years (SD = 15.16). Of these participants, 52 had

completed higher education, 15 had completed secondary school, one respondent had completed his

education at age 14, and one person did not indicate his educational level. With respect to their

occupational level, 3 of our participants were unskilled manual employees, 5 were skilled manual

employees, 19 were administrative personnel, 16 were business managers, 8 were self-employed persons,

one was a farmer, and 17 could not be classified in these categories. With respect to their marital status, 28

of them were single, 38 were married or lived together with a partner, two persons reported that their

partner had deceased, and one person was divorced. Thirty-seven of them grew up in a city, 30 on the

17

countryside, and two did not indicate where they grew up.

Measures

All items were rated on 5-point Likert scales (1 = completely opposed, 3 = neutral, 5 = completely in

agreement).

Left-wing and right-wing authoritarianism. The 8-item LWA scale and the 11-item RWA scale that

were used in Study 1 were also administered in this study. Cronbach alpha’s for the complete LWA and

RWA scale were .70 and .88 for the voter sample, and .82 and .94 for the political activists, respectively.

Cultural and economic conservatism. Participants also completed a cultural and economic

conservatism scale (De Witte, 1990; both 12 items). The cultural conservatism scale addresses issues such

as upbringing, work ethic, the position of women in society, premarital sexual intercourse, abortion and

euthanasia. The economic conservatism scale addresses issues such as the desirable impact of trade

unions, level of government interventions in economics and income differences. Cronbach alphas were .88

and .94 for economic conservatism, and 81 and .91 for cultural conservatism in the voter sample and the

political activist sample, respectively.

Results

Psychometric analyses of LWA

As can be seen in Table 1, the total LWA scale yielded a Cronbach alpha of .70 in the voter

sample, but for its facets, low internal consistencies were obtained (Cronbach alpha’s = .62 and .55 for the

aggression and submission scale respectively). In addition, Principal Component Analysis did not confirm

the hypothesized two-dimensional structure (see Table 2). The positively keyed items as well as two

negatively keyed aggression items loaded on the first OBLIMIN rotated component (explaining 32.68% of

the total variance) and the two negatively keyed submission items loaded on the second component

(explaining 20.06% of the total variance). Subsequent analyses with the facet scales should therefore be

interpreted with caution. A strong relationship between aggression and submission (r = .42, p < .001) was

obtained.

18

As can also be seen in Table 1, reliability analyses of the LWA scale revealed sufficient internal

consistency for the political activist sample (Cronbach alpha = .82). The Cronbach alpha’s in the activist

sample for the aggression (.88) and the submission (.77) facet scales were also satisfactory. In addition,

Principal Component Analysis revealed the hypothesized two-dimensional structure (see Table 2). The

aggression items loaded on the first OBLIMIN rotated component (explaining 40.89% of the total variance)

and the submission items loaded on the second component (explaining 32.60% of the total variance). The

correlation between the aggression and submission facet scales was highly positive (r = .31, p < .001).

In sum, the present analyses revealed that the total LWA scale is internally consistent.

However, the aggression and submission facet scales can be used as valid indicators in the political

activist sample, but not in the voter sample. Fortunately, in order to test the various hypotheses (i.e.,

Hypothesis 2a and 2b), the facet scales were only needed in the political activist sample.

Mean level differences and presence of LWA

First, we examined the presence of LWA in the voter sample. No effect of the demographic

variables was obtained (Fs < 1.47, correlation between age and LWA, r = .09, ns.). Thirteen participants

(4.9%) obtained an LWA score equal to the theoretical midpoint, and 9 participants (3.4%) obtained

scores that lied in the 3.01 – 3.99 range. The highest score was 3.63. These results thus indicate a

limited presence of left-wing authoritarianism in the voter sample and therefore corroborate the findings

of Study 1 as well as Hypothesis 1.

Next, we analyzed the presence of LWA according to the various demographic variables

administered in the activist sample. None of them turned out to be significant (Fs < 1.58, correlation

between age and LWA, r = -.16, ns.). Table 3 reports the mean levels of LWA and its facets for the four

ideological groups in the political activist sample, as well as the mean levels of the other political variables.

In general, these results confirm the validity of the LWA scale. Differences between these groups were

highly significant (F(3, 64) = 16.17, p < .001). As expected, the extreme left-wing group was the only group

to obtain scores above the theoretical midpoint of 3.00 (M = 3.51). The six members of the communist party

19

obtained a mean score of 3.00 (95% confidence interval of 2.01 – 3.99), whereas the members of the

Stalinist PVDA obtained a mean score of 3.72 (95% confidence interval of 3.08 – 4.38). The anarchist group

had the second highest LWA score (M = 2.81), followed by the moderate (M = 2.13) and right-wing

extremist group (M = 1.84). Duncan post-hoc analyses revealed that extreme left-wingers had significant

higher LWA scores than the anarchists, who, in turn, obtained significant higher scores than the other

groups.

Insert Table 3 about here

In addition, extreme left-wingers and anarchists obtained authoritarian aggression scores above

the theoretical midpoint (M respectively 3.96 and 3.87), whereas moderates and right-wing extremists (M =

1.90 and 1.66 respectively) (F(3, 64) = 34.01, p < .001) obtained much lower scores. Extreme left-wingers

obtained significant higher submission scores (M = 3.05) than moderates, extreme right-wingers and

anarchists (M = 2.37, 2.02, and 1.75 respectively) (F(3, 64) = 7.56, p < .001). The results on the LWA facet

scales thus reveal that, in line with Hypothesis 2a, extreme left-wingers obtained significantly higher scores

on the aggression and submission facet scales than moderates and right-wing extremists. In line with

Hypothesis 2b, anarchists obtained significantly higher aggression scores than moderates and right-wing

extremists, whereas their scores on the submission and conventionalism facet were not elevated.

This brings us to the question whether there are any “true left-wing authoritarians” among the

adherents of the various ideological groups. Table 4 shows the number of participants who respectively

obtained a score below the theoretical midpoint (i.e., a score below 3.00), a score on the theoretical

midpoint (i.e., a score of 3.00), a score between 3.01 and 3.99, and a score of 4.00 or more. In the extreme

left-wing group, 13 out of 20 participants scored above the theoretical midpoint of the LWA scale. Two

persons even obtained the maximum score of 5.00. In addition, left-wing extremists generally expressed

greater agreement with the authoritarian aggression facet, and about half of them endorsed the items of the

authoritarian submission facet scale.

Insert Table 4 about here

20

In the anarchist group, only 6 of the 21 activists obtained a LWA score higher than the midpoint.

However, a vast majority of the anarchists (N = 18) obtained a score higher than the theoretical midpoint for

the authoritarian aggression facet scale, whereas they expressed much less agreement with the items of

the authoritarian submission and conventionalism facet scales. In sharp contrast, not one single left-wing

authoritarian was detected among the 16 moderates and the 11 right-wing extremists. However, rather

surprisingly, six participants in these groups obtained a score equal or higher than 3.00 for the submission

facet scale.

LWA, RWA and Conservatism

The correlations between LWA, RWA and conservatism are reported in Table 5. The correlation

between LWA and RWA was positive in the voter sample (r = .22, p < .001) corroborating Altemeyer (1996)

who also reported a (rather unexpected) positive RWA – LWA correlation. However, in the political activist

sample a negative correlation between LWA and RWA emerged (r = -.45, p < .001).

Insert Table 5 about here

Next, we analyzed the differential relationships of LWA and RWA with economic and cultural

conservatism. Hypothesis 3 states that LWA is particularly related to economic progressivism, whereas

RWA is particularly related to cultural conservatism. The correlations between these variables are given in

Table 5. Applying a statistic reported by McNemar (1969, p. 158), we checked whether the magnitude of

the (absolute) correlation between LWA and economic conservatism was higher than the correlation

between LWA and cultural conservatism. The corresponding correlations with RWA were also compared. In

the voter sample, the difference in magnitude of the correlation between LWA and economic conservatism

(r = -.13) was not significantly different from its correlation with cultural conservatism (r = .14, t (df = 263) =

.06, ns.), whereas RWA was more strongly related to cultural conservatism (r = .66) than to economic

conservatism (r = .14, t (df = 263) = 9.11, p < .001). In the political activist sample, LWA was more strongly

related to economic conservatism (r = -.63) than to cultural conservatism (r = -.39, t (df = 68) = 2.49, p <

.05), whereas RWA was more strongly related to cultural conservatism (r = .91) than to economic

21

conservatism (r = .71, t (df = 68) = 6.01, p < .001). The results in the political activist sample thus

completely corroborate Hypothesis 3, whereas this hypothesis was only confirmed for the RWA-cultural

conservatism correlation in the voter sample.

Discussion

The present results suggest the presence of authoritarianism among Western European adherents

of extreme left-wing parties. Particularly the adherents of the Stalinist party obtained high LWA scores .So, it

seems that we achieved in finding “the Loch Ness Monster of political psychology”. The LWA scale not only

proved to be successful in distinguishing anarchists and extreme left-wingers from the other ideological

groups (the authoritarian aggression facet is most fruitful for this purpose), but also in distinguishing extreme

left-wingers from anarchists (the authoritarian submission facet is most fruitful for this purpose). The

discriminatory power to distinguish between left-wing extremists, anarchists, and other ideological groups

underscores the validity of the aggression and submission facet scales. However, these results also make it

clear that the presence of LWA in Western societies seems to be limited to very specific political

movements that do not elicit much support in the mass public. As such, LWA is a marginal phenomenon in

political life that is not easily found in convenience samples.

Moreover, some left-wing extremists (6 out of 20) did not obtain a score higher than the neutral

point indicating that even among left-wing extremists there are a number of non-authoritarians. Our

analyses revealed that the mean levels of LWA among the adherents of the Communist Party were lower

than among those who supported the Stalinist party. These data suggest that the degree of LWA varies

between the various extreme left-wing movements, and perhaps that there is a greater likeability to find left-

wing authoritarians in “traditional” extreme left-wing movements than in neo-marxist parties. But, even

among the supporters of the Stalinist movement there were 4 (out of 14) non-authoritarians underscoring

the fact that not all left-wing extremists are “by definition” authoritarian.

Finally, the present study also investigated the relationships between LWA, RWA and

conservatism. The results in the activist sample substantiated Hypothesis 3, which states that LWA and

22

economic progressivism are highly related, and that RWA and cultural conservatism are highly related.

These results substantiate the validity of LWA and RWA as two distinctive ideologies. However, in the voter

sample, the results revealed that LWA was not particularly strongly related to economic progressivism.

Moreover, the correlation with cultural conservatism was not in line with the one observed in the political

activist sample.

GENERAL DISCUSSION

An important theme in authoritarianism literature has been the topic of left-wing authoritarianism.

The main objective of the present studies was to identify true left-wing authoritarians. Study 1, conducted on

a voter sample, showed sufficient internal consistency of our LWA scale. Moreover, a right-wing version

of this scale in which the target group was substituted by a typical right-wing outgroup showed high

correlations with the traditional RWA scale (Altemeyer, 1981), attesting its construct validity. However,

analogous to previous studies in Canada (Altemeyer, 1996) we were not able to detect true left-wing

authoritarians. In Study 2 we solicited another sample of voters and a sample of political activist sample.

The main conclusions of Study 1 were replicated in the voter sample. The sample of political activists

was drawn from extremist groups (anarchists, communists and right-wing extremists) and moderate

parties. Several interesting findings were noted. First, unlike the results in the voter samples, the

aggression and submission items loaded on a distinct dimension underscoring the necessity of

distinguishing between these two facets among (left-wing extremist) political activists. Second, extreme

left-wing activists obtained high scores on both left-wing authoritarianism facets, whereas anarchists

obtained high scores on the aggression facet scale only. As a result, extreme left-wingers obtained

significantly higher LWA scores than anarchists, who, in turn, obtained significantly higher LWA scores

than moderates and members of an extreme right-wing party. A substantive part of the left-wing

extremist sample obtained a score higher than the theoretical midpoint of the scale. Third, whereas

LWA was more strongly related to economic progressivism than to cultural conservatism, RWA was

more strongly related to cultural conservatism than to economic conservatism. The latter result shows

23

that LWA and RWA are differentially related to ideology. In the remainder, we will discuss the presence of

LWA in our activist samples, as well as the theoretical implications of the present results for the LWA

concept.

LWA: The Loch Ness Monster of political psychology?

Altemeyer (1996) reported that he could not identify one single left-wing authoritarian in various

Canadian samples of students and their parents, as well as among candidates of moderate political parties.

In the political activist sample, the maximum LWA score on a nine-point scale was 4.92. Only 12 students

out of a total of 546 were found to score above the neutral point of 5.00, with a highest score of no more

than 5.65. In the sample of parents, only 20 participants out of a total of 642 scored above the midpoint,

with a highest score of 5.85. Opposed to Altemeyer’s (1996) results, the highest scoring individual in our

political activist sample obtained the maximum score (5.00). Six participants obtained a score between 4.00

and 5.00, and 7 participants scored between 3.00 and 4.00. In other words, 20% of the participants of this

sample generally agreed with the LWA items, and hence, should be considered as true left-wing

authoritarians. Why did we find so many left-wing authoritarians in this sample, while Altemeyer (1996) did

not find a single one of them in his studies? The answer seems obvious: Whereas Altemeyer (1996) might

have constructed a valid scale, in order to establish LWA, he should have administered this scale to left-

wing extremists.

The left-wing extremists in our political activist sample, who generally obtained LWA scores above

the 3.00 midpoint, were adherents of the PVDA. This political party has its roots in Stalinism, and might thus

be called classic communist instead of neo-marxist. Some statements on the web site of the PVDA

(http://www.pvda.be, Dutch language only) illustrate this. Under the heading “The historical experience of

communism” it reads “Mao’s revolutionary movement has opened the doors to the socialist movement in

the third world.” Chroetsjov is depicted as “The first step of revisionism that undermined the socialist state in

the Soviet Union”. Under the heading “Against who does the PVDA fight?” it is argued that “the PVDA fights

against the world of high finance, banks, holdings and multinationals … which cause exploitation and misery

24

and the rise of fascism”. Moreover, the police is depicted as “a private militia of the capitalists”. Under the

heading “What does the PVDA want?” it reads “the PVDA wants the destruction of the capitalist system and

the foundation of the socialist state which bears on the working class.” These statements reveal an

admiration of harsh regimes known to have terrorized millions of people. However, although these

messages do have an aggressive undertone, they do not call for the actual use of violence. In other words,

the PVDA does not advocate violence, but its members might nevertheless infer the legitimacy of such

actions because the Establishment is depicted as hostile, aggressive and illegitimate. Not surprisingly,

these people obtain high LWA scores, as well as high scores on both the authoritarian aggression and

submission facet.

The anarchist movement in Flanders constitutes loosely organized organizations and

individuals. This movement does, however, have its own information channels. Since most participants

of the anarchist sample read the anarchist magazine “De Nar”, we figured that looking at this

magazine’s website might be informative (http://users.online.be/~pr002099/index2.htm, Dutch language

only). On this website, it is explained that the aim of “De Nar” is to provide information for, and to

stimulate discussions among those who – from an anti-authoritarian viewpoint – attach importance to

participatory democracy, solidarity, and direct action. “De Nar” supports a world in which there is no

room for either political or economical repression. On several pages, it is explained that anarchists loath

authority and the capitalist system (and, in fact, the System in general). Hence, not surprisingly,

anarchists showed levels of anti-Establishment aggression similar to left-wing extremists, but obtained

significantly lower scores on the authoritarian submission facet and, consequently, on LWA in general.

The left-wing authoritarianism concept

It has been shown that LWA does exist and should not be considered a myth, but this study should

only be considered a first small step in validating a theory of LWA. Some of our findings are relevant for

the development of a theoretical framework that links LWA to other political variables. Previous research

abundantly revealed that RWA relates to, for example, prejudice (e.g., Adorno et al., 1950) and poor

25

information processing (e.g., Van Hiel, Pandelaere, & Duriez, 2004). The combination of all these

findings constitutes the theory of right-wing authoritarianism and illustrates the utility and validity of the

RWA concept. For LWA this empirical work remains to be done. However, before proceeding with the

empirical work sustaining such a theory, some problematic points should be considered.

A first problem for the further study of LWA is its limited presence. That is, the number of people

who are willing to obey to established authorities and to aggress in their names is much higher than

those who would behave accordingly in the name of revolutionary movements. Nevertheless, LWA

seems a to be a promising concept in the context of small extremist groups only. Thus, because of its

limited prevalence, the LWA concept does not seem to hold promise for the psychology of the masses.

Moreover, compared to the activist sample, the results in the voter samples indicate another

pattern of correlations between LWA and other important political variables. Thus, these results suggest

that ordinary voters attach a different meaning to LWA than political activists. One of the most disturbing

findings in the present voter samples as well as in Altemeyer (1996) is the positive correlation between

LWA and RWA. According to Altemeyer (1996), the presence of “wild-card authoritarians” who score high

on both LWA and RWA might explain the existence of hostility, dogmatism and authoritarianism among left-

wing extremists. Moreover, Altemeyer (1996, p. 224) reported the highest levels of ethnocentrism and anti-

gay attitudes among wild-card authoritarians, whom he attributed severe hostility.

LWA’s modest correlation with other indicators of political ideology is another problem. Recent

research on samples of ordinary voters distinguished between two principal dimensions of political

ideology of which the first is constituted by RWA and cultural conservatism and the other one by Social

Dominance Orientation (SDO) and economic conservatism (see, Duckitt, 2001; Duriez et al., in press;

Van Hiel & Kossowska, 2004). Where should we place LWA in this framework? The present results in

the voter sample of Study 2 show that LWA has a modest positive relationship with RWA (r = .22) and

cultural conservatism (r = .14) and therefore does not represent the opposite pole of the RWA – cultural

conservatism dimension. Moreover, its modest correlations with economic conservatism (r = -.13)

26

indicate that LWA is only weakly related to the SDO – economic conservatism axis. In sum, it seems as

if LWA is not well represented in terms of the well-known ideological dimensions and that its meaning

for understanding ideology in the mass public is unclear. In the political activist sample, however, a

clearly different picture is obtained. In this sample, LWA is clearly negatively related to economic

conservatism (r = -.63) suggesting that in these kinds of samples, LWA can be considered as the

opposite pole of the SDO – economic conservatism. 1

Limitations of the present study

Before closing, we should mention two limitations of the present study. A first limitation is that the

sample of political extremists was rather small, and one could argue that the present findings may capitalize

on error variance. However, our results generally corroborated the hypotheses and the magnitude of the

effect sizes was fairly high, attesting the validity of the present findings. Moreover, almost by definition,

“extremists” are not that numerous and they have also been reported to be unwilling to participate in

empirical studies as well (e.g., Rosen, 1951). Not surprisingly then, the number of participants in the present

study was not especially low in comparison with the few previous studies that included a sample of

extremists to test the LWA hypothesis.

A second limitation of the present study is the use of participants from a typical Western

democracy. For example, LWA might have a different meaning in former communist countries. A plausible

reason for this is the recent history of ex-communist countries in which the Establishment made part of the

same left-wing ideology. Because of this historical context, extreme left-wing ideas in such a society might

attract both left-wing and right-wing authoritarians. Thus, further elaboration of the concept of LWA in former

communist countries might be an interesting topic for future research.

1 At the same time, the negative relations between LWA and both RWA (r = -.45) and cultural

conservatism LWA (r = -.39) suggest that LWA can, to a somewhat lesser extent, also be considered an anti-pole of the RWA – cultural conservatism dimension. Apart from rejecting economic conservatism, left-wing authoritarians also strongly reject the (right-wing) conservatism that is implied in the concepts of RWA and cultural conservatism.

27

References Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The Authoritarian

Personality. New York: Harper.

Altemeyer, B. (1981). Right-Wing Authoritarianism. Winnipeg: University of Manitoba Press.

Altemeyer, B. (1996). The Authoritarian Spectre. Cambridge, MA: Harvard University Press.

Altemeyer, B. (1998). The other “authoritarian personality”. In M. P. Zanna (Ed.), Advances in Experimental

Social Psychology (Vol., 30, pp. 47-92). San Diego: Academic Press.

Christie, R. (1956). Eysenck’s treatment of the personality of communists. Psychological Bulletin, 53, 439-

451.

De Witte, H. (1990). Conformisme, radicalisme en machteloosheid: Een onderzoek naar sociaal-culturele

en sociaal-economische opvattingen van arbeiders in Vlaanderen. [Conformism, radicalism and

powerlessness: Examining Flemish workers’ socio-cultural and socio-economic opinions]. Leuven,

Belgium: HIVA-KU Leuven.

DiRenzo, G. J. (1967). Professional politicians and personality structures. American Journal of Sociology,

73, 217-225.

Duckitt, J. (2001). A dual-process cognitive-motivational theory of ideology and prejudice. In M.P. Zanna

(Ed.), Advances in Experimental Social Psychology (Vol. 33, pp. 41-113). San Diego, CA:

Academic Press.

Duriez, B. (2004). A research note on the relation between religiosity and racism: The importance of the

way in which religious contents are being processed. The International Journal for the Psychology

of Religion, 14, 175-189.

Duriez, B. & Van Hiel, A. (2002). The march of modern fascism: A comparison of social dominance

orientation and authoritarianism. Personality and Individual Differences, 32, 1199-1213.

Duriez, B., Van Hiel, A., & Kossowska, M. (in press). Authoritarianism and social dominance in Western and

Eastern Europe. The importance of the socio-political context and of political interest and

28

involvement. Political Psychology.

Durrheim, K. (1997-a). Theoretical conundrum: The politics and science of theorizing authoritarian

cognition. Political Psychology, 18, 625-647.

Durrheim, K. (1997-b). Cognition and ideology: A rhetorical approach to critical theory. Theory and

Psychology, 7, 747-768.

Eysenck, H. J. (1954). The psychology of politics. London: Routledge & Kegan Paul.

Eysenck, H. J. (1980-81). Left-wing authoritarianism: Myth or reality? Political Psychology, 3, 234-239.

Eysenck, H.J. & Coulter, T. (1972). The personality and attitudes of working class British communists and

fascists. Journal of Social Psychology, 87, 59-73.

Hamilton, V. L., Sanders, J., & McKearney, S. J. (1995). Orientations toward authority in an authoritarian

state: Moscow in 1990. Personality and Social Psychology Bulletin, 21, 356-365.

Ignazi, P. (1992). The silent counter-revolution: Hypotheses on the emergence of extreme right-wing parties

in Europe. European Journal of Political Research, 22, 3-34.

Knutson, J. N. (1974). Psychological variables in political recruitment. Berkeley, CA: Wright Institute.

Kossowska, M., & Van Hiel, A. (2003). The relationship between need for closure and conservative beliefs

in Western and Eastern Europe. Political Psychology, 24, 501-518.

Lipset, S. (1981). Political man: The social basis of politics. Baltimore: The John Hopkins University Press.

Martin, J. L. (2001). The authoritarian personality, 50 years later: What lessons are there for political

psychology? Political Psychology, 22, 1-26.

McCloskey, H., & Chong, D. (1985). Similarities and differences between left-wing and right-wing radicals.

British Journal of Political Science, 15, 329-363.

McFarland, S. G., Ageyev, V. S., & Abalakina-Paap, M. A. (1992). Authoritarianism in the former Soviet

Union. Journal of Personality and Social Psychology, 63, 1004-1010.

McNemar, Q. (1969). Psychological Statistics (4th Ed.). New York: Wiley.

Meloen, J. D. (1991). Inventarisatie Nederlandse F-schalen 1959-1990. In P. Scheepers, & R. Eisinga

29

(Eds.), Intolerant en Onderdanig (pp. 186-222). Nijmegen, Netherlands: ITS.

Middendorp, C. (1978). Progressiveness and conservatism. The fundamental dimensions of ideological

controversy and their relationship to social class. Den Haag/Paris/New York: Mouton.

Ray, J. J. (1983). Half of all authoritarians are left-wing: A reply to Eysenck and Stone. Political Psychology,

4, 139-144.

Rokeach, M. (1960). The Open and Closed Mind. New York: Basic Book Inc., Publishers.

Rokeach, M. & Hanley, C. (1956). Eysenck’s tendermindedness dimension: A critique. Psychological

Bulletin, 53, 169-176.

Rosen, E. (1951). Differences between volunteers and non-volunteers for psychological studies. Journal of

Applied Psychology, 35, 185-193.

Sidanius, J. (1984). Political interest, political information search, and ideological homogeneity as a function

of sociopolitical ideology: A tale of three theories. Human Relations, 37, 811-828.

Sidanius, J. (1988). Political sophistication and political deviance: A structural equation examination of con-

text theory. Journal of Personality and Social Psychology, 55, 37-51.

Soenens, B., Duriez, B., & Goossens, L. (in press). Social-psychological profiles of identity styles:

Attitudinal and social-cognitive correlates in late adolescence. Journal of Adolescence.

Stone, W. F. (1980). The myth of left-wing authoritarianism. Political Psychology, 2, 3-19.

Stone, W. F. & Smith, L. D. (1993). Authoritarianism: Left and right. In W.F. Stone, G. Lederer, & R. Christie

(Eds.), Strenghth and weakness: The Authoritarian Personality Today (pp. 144-156). New York:

Springer Verlag.

Tetlock, P. E. (1983). Cognitive style and political ideology. Journal of Personality and Social Psychology,

45, 118-126.

Tetlock, P. E. (1984). Cognitive style and political belief systems in the British House of Commons. Journal

of Personality and Social Psychology, 46, 365-375.

Tetlock, P. E. (1986). A value pluralism model of ideological reasoning. Journal of Personality and Social

30

Psychology, 50, 819-827.

Van Hiel, A., & Kossowska, M. (2004). Contemporary attitudes and their ideological representation in

Flanders (Belgium), Poland and the Ukraine. Manuscript submitted for publication.

Van Hiel, A., & Mervielde, I. (2002-a). Explaining conservative beliefs and political preferences: A

comparison of social dominance orientation and authoritarianism. Journal of Applied Social

Psychology, 32, 965-796.

Van Hiel, A., & Mervielde, I. (2002-b). Social identification among political party members: An empirical test

of Optimal Distinctiveness Theory. Journal of Social Psychology, 142, 202-210.

Van Hiel, A., & Mervielde, I. (2003). The measurement of cognitive complexity and its relationship with

political extremism. Political Psychology, 24, 781-801.

Van Hiel, A., & Mervielde, I. (2004). Openness to Experience and boundaries in the mind: Relationships

with cultural and economic conservatism. Journal of Personality,72, 659-686 .

Van Hiel, A., Pandelaere, M., & Duriez, B. (2004). The impact of need for closure on conservative beliefs

and racism: Differential mediation by authoritarian submission and authoritarian dominance.

Personality and Social Psychology Bulletin, 30, 824-837.

Wilson, G.D. (1973). The psychology of conservatism. London: Academic Press.

31

Table 1. Item-scale correlations and Cronbach alphas of LWA and its aggression and submission facets

Study 1 Study 2

Label Item Voter sample

Voter sample

Activist sample

Item 01 I agree with the basic idea of communism of overthrowing the Establishment – with or without violence – and giving its wealth to the poor………………………………………………………………………………………………………………………………………..

.54

.38

.54

Item 02 A revolutionary movement is justified in using violence because the Establishment will never give up its power peacefully ……………………………………………………………………………………………………………………………………………………….

.67

.61

.70

Item 05 It would be wrong to solve our problems by acts of violence against the conservative Establishment (-)………………….. .47 .45 .59 Item 06 Even though the conservative Establishment who controls our country is repressive and unfair, society should only

be reformed through nonviolent means (-)………………………………………………………………………………………………………….

.24

.21

.67 Cronbach alpha Aggression facet .62 .62 .88 Item 03 The conservative Establishment has so much power and is so unfair that we have to submit to the leaders and

rules of a revolutionary movement in order to destroy them …………………………………………………………………………………

.54

.50

.50 Item 04 A revolutionary movement is justified in demanding obedience and conformity of its members ………………………………. .53 .44 .55 Item 07 Even a revolutionary left-wing movement that fights against the totally unjust right-wing system does not have the

right to tell its members how to think and act (-) ………………………………………………………………………………………………….

.42

.22

.42 Item 08 A left-wing party is not justified in demanding too much conformity and obedience, even after a revolution (-) …………. .51 .33 .34 Cronbach alpha Submission facet .67 .55 .77 Cronbach alpha total LWA scale .77 .70 .82

32

Table 2 Loadings of the LWA items on a two-component solution

Study 1 Study 2 Study 2

Voter sample Voter sample Activist Sample

I II I II I II

Item 01 .77 .02 .78 -.26 .91 -.16 Item 02 .92 -.15 .82 .02 .90 .02 Item 03 .85 .06 .74 -.02 .32 .50 Item 04 .67 .13 .47 .38 .17 .75 Item 05 .04 .68 .56 .17 .78 .05 Item 06 .02 .75 .35 .03 .80 .11 Item 07 -.04 .85 -.04 .74 -.06 .87 Item 08 .00 .44 .03 .80 -.17 .87

Note Loadings > .40 in bold face

33

Table 3 Mean level differences with respect to the political variables in the political activist sample (Study 2)

Construct Left extremist (N=20) Moderates (N=16) Right extremist (N=11) Anarchists (N=21) F(3,64) =

LWA 3.51 a (1.10) 2.13 c (0.49) 1.84 c (0.51) 2.81 b (0.58) 16.17 *** Aggression 3.96 a (1.06) 1.90 b (0.69) 1.66 b (0.56) 3.87 a (0.88) 34.01 *** Submission 3.05 a (1.29) 2.37 b (0.61) 2.02 b (0.83) 1.75 b (0.61) 7.56 *** RWA 1.88 b (0.64) 3.49 a (0.83) 3.70 a (0.36) 1.54 b (0.35) 57.37 *** Cultural conservatism 1.96 b (0.72) 3.08 a (0.76) 3.34 a (0.62) 1.39 c (0.37) 35.02 *** Economic conservatism 1.38 c (0.51) 2.62 b (0.74) 3.45 a (0.70) 1.57 c (0.30) 44.07 ***

Note Standard deviations between parentheses. Mean levels with different subscripts are significantly different at the .05 level. * p < .05, ** p < .01, *** p < .001.

34

Table 4 The presence of LWA in the political activist sample (number of participants by score level)

Left extremists Moderates Right extremists Anarchists

<3 =3 >3 >4 <3 =3 >3 >4 <3 =3 >3 >4 <3 =3 >3 >4

LWA 6 1 4 9 16 – – – 11 – – – 13 2 6 – Aggression 4 2 4 13 15 1 – – 11 – – – 3 – 5 13 Submission 10 – 3 7 12 1 3 – 9 1 1 – 20 1 – –

Note <3 = a score of less than 3.00, =3 = a score of 3.00, >3 = a score between 3.01 and 3.99, and

>4 = a score of 4.00 or more.

35

Table 5 Correlations among the variables included in Study 2

01 02 03 04 05

01. LWA — — 02. Aggression .85 *** — .86 *** — 03. Submission .84 *** .42 *** — .75 *** .31 *** — 04. RWA .22 *** .10 .27 *** — -.45 *** -.69 *** .08 — 05. Cultural conservatism .14 *** .04 .19 ** .66 *** — -.39 *** -.63 *** .09 .91 *** — 06. Economic conservatism -.13 * -.17 ** -.06 .14 * .16 ** -.63 *** -.72 *** -.25 * .71 *** .61 ***

Note Second line and third line figures involve the voter and political activist sample respectively. * p < .05, ** p < .01, *** p < .001.

View publication statsView publication stats