Make your own free website on

South African J. Psychology , 1981, 11: 153-157


J.J. Ray

School of Sociology, University of New South Wales

It is pointed out that authoritarian attitudes and authoritarian behaviour are generally not associated. The few studies that have been made of authoritarian behaviour directly are reviewed. In studies of Nazi Germany, of police and army personnel and of educational methods, no evidence is found that authoritarian behaviour is psychopathological. Ray's (1976) Directiveness scale measuring authoritarian behaviour and a balanced F-scale were presented to a random sample of 95 Sydney people in a doorstep survey. The authoritarian personality as measured by the Directiveness scale was found to go with better mental health as measured by the Eysenck Neuroticism scale. There was no significant correlation between the balanced F-scale and neuroticism. The study made use of extensive controls against acquiescent response set and social desirability response set.


Although Adorno et al. (1950) have implicated authoritarian attitudes in psychopathology, social scientists have done surprisingly little work that examines the status of authoritarian behaviour in this respect. Since the work on attitudes done by Adorno et al. had as its purpose the explanation of authoritarian behaviour, one might have expected that much of the research in the area would have focused on the behaviour directly rather than going by way of the attitudes presumably associated with it.

The reason why this has not been so is not far to seek: attitudes are a lot easier to study than behaviour. Nonetheless this deficit would not be a serious one if we could assume that there was a high relationship between authoritarian attitudes and authoritarian behaviour. The one could be used to tell us something about the other.

Unfortunately it is now clear that, as in so many other fields, in the field of authoritarianism attitudes and behaviour often have no relationship at all. Even if authoritarian attitudes could be conclusively shown to be psychopathological, this would tell us nothing at all about authoritarian behaviour. The evidence for this lack of association has been summarized at length in Ray (1976). Some additional references, however, are Hines (1956), Larsen, Coleman, Forbes and Johnson (1972) and Perry and Cunningham (1975). What little evidence exists on the mental health of people who behave in an authoritarian way is summarized below:

Our first group of studies concerns Nazi Germany. Whatever else they were, the Nazis certainly did behave in an authoritarian way. If authoritarian behaviour is symptomatic of psychological disturbance then they certainly should have shown evidence of such disturbance.

A recent study of Nazi personnel is that by Dicks (1972). His book presents observations and inferences based on interviews with a small sample of former members of the S.S. convicted of crimes against humanity. It includes the life-stories and psychological profiles of the individuals interviewed, and aims at (a) identifying personality traits and relating them to the scales of Nazi fanaticism which were established on the basis of interviews of captured German military personnel during the war; and (b) establishing what forces impelled these particular people to accept these particular roles. The evidence assembled suggests that neither fanaticism nor identifiable psychiatric disorders were crucial amongst these people. They were mostly ordinary, weak-egoed individuals who failed to resist the pressure of appropriate social as well as group climates and sanctions, and thus came to abrogate their customary levels of 'civilized' behaviour.

Even studies of Hitler himself come to some remarkably positive conclusions. Being very careful to exclude value judgements and to rely only on the available evidence, Brink (1974) concludes: `The man in the street believes that Hitler was a madman. However, a careful examination of his life, as portrayed in the reliable sources, fails to support this theory' (p.24). Like modern revisionist historians (Trevor-Roper, 1947; Taylor, 1961) Brink sees something in the view of Hitler as being simply a traditional German leader pursuing traditional German interests. He does show, however, that much of Hitler's life is to be explained as various strategies designed to cope with an initial bad start in life.

The Brink study of Hitler was founded in the methods of Adlerian psychohistory. Kren & Rapaport (1975) apply similar methods to an analysis of the S.S. They conclude that by conventional psychiatric standards, the S.S. men were not seriously maladjusted. They believe that we should look not at the motive for violence, but at the conditions under which the usual moral inhibitions against violence become weakened.

Another famous Nazi leader was Albert Speer, Hitler's personal architect and intimate and also later minister for Armaments and War. In an article by Singer & Wotton (1976) Speer is revealed as an exponent of 'some of the most advanced, participative and "humanistic" management theories being endorsed today' (p.79). He is shown to have been so anti-bureaucratic that he rejoiced when bombing destroyed one of his central record offices. His theory of management he called 'organized improvisation' and he believed in collegial forms of decision-making. He also practised 'a loose or fluid manner of structuring organizations' (p.83). Modern day proponents of such practices do of course believe them to be anything but symptomatic of mental illness.

More generally, Koomen (1974) shows that pre-war Germany did not even have characteristically authoritarian child-rearing practices. He concludes: 'Secondary analysis of data concerning periods before and after the war showed that before the war, only differences in parental control with regard to daughters could be demonstrated; parental control concerning sons appeared to be approximately the same in the two countries' (Germany and the USA) (p.634). If the men of the S.S. behaved in an authoritarian way it was, then, not because they had harsh upbringings.

Finally, in a comparison of Nazi Germany and Soviet Russia, Unger (1965), an Israeli scholar, finds Germany to have been much less totalitarian. Like Roberts (1938) he believes the Nazi state and its leader to have been basically popular -- unlike its Russian counterpart. It is notable that Hitler came to power with a bigger percentage of the popular vote than that obtained by Britain's last Wilson Labour government. One would normally assume that an electorate would be highly sensitive to the faintest suspicion of deviance in a proffered leader.

Our second body of studies concerns people engaged in the two archetypical authoritarian occupations -- the police and the army. There is no doubt that these people have to behave in an authoritarian way. Is this then an occupation chosen by deviates?

In an extensive review of research on the personality characteristics of policemen, Lefkowitz (1975) concludes: 'Data substantiate, with some qualifications, the existence of a non-pathological modal police personality' (p.3). Fenster & Locke (1973) concluded even more strongly that 'On the whole, policemen scored lower on neuroticism when compared with non-police citizens' (p.358). Their conclusion was based on a study of 548 college and non-college educated New York policemen and civilians.

As far as neuroticism among army voIunteers is concerned, the best evidence available appears to be that by Salas and Richardson (1968) who applied the Eysenck 'N' scale to Australian army conscripts and volunteers. They found neuroticism scores among the volunteers which fell well within the range of the conscript scores and compared favourably with other occupational groups. These data are particularly valuable because Australian conscription laws at the time provided for men to be called up on the basis of a nation-wide ballot of birth dates. Selection was hence completely random. Students were allowed deferred call-up for a few years but were not exempted. The conscript norms do hence provide an excellent control group.

Another interesting study is one by Bourne (1971). He studied altered adrenal function among Vietnam war participants and non-participants in the US armed forces. He found that the level of 17-hydroxycorticosteriod excretion did not vary in either the short or the long term between those engaged in combat and those not engaged in combat. Considered as symptoms of anxiety then, these excretion levels show that the ultimate authoritarian behaviour of combat is still not accompanied by increased anxiety. If this is so, one certainly cannot conclude that a person has to be chronically anxious to engage in authoritarian behaviour. In fact one does not even need to be situationally anxious.

Another group of studies concerns the effect of authoritarian leadership behaviour on school learning and group problem solving tasks. A very well-known early paper by Lippitt & White (1947) asserted without good proof that small groups run along authoritarian lines were less productive than ones run along democratic lines. In a specific test of this assertion, McCurdy & Eber (1953) found no effect on group productivity according to how it was run.

Moving to a school situation, Guetzkow, Kelly & McKeachie (1954) found that students achieved at a higher level in classes taught by autocratic methods as compared to their achievement level when less autocratic methods were used. More recently, in a study of 871 Lancashire primary schools, Bennett (1976) found a consistent advantage in all learning areas for the traditional 'authoritarian' methods. Pupils learnt their '3R's' much more quickly where such methods were used. Ward & Barcher (1975) in the USA also report similar results. If then authoritarian behaviour is a symptom of psychological ill-health, we must apparently regard better learning as one outcome of mentally-ill practices!

Although the above studies generally allow the hypothesis that to behave in an authoritarian way may in fact be symptomatic of superior rather than of inferior psychological health, several authors have also been at pains to stress that much authoritarian behaviour is situationally determined. People often behave in an authoritarian way not because they like to but because it is indicated by the circumstances or required by the milieu. Psychologists, however, have always been particularly interested in authoritarian behaviour that emanates from the nature and preferences of the person. They have been interested in authoritarian behaviour as an expression of an authoritarian personality. The above studies do, then, shed no certain light on the mental health status of authoritarian behaviour when that behaviour is an expression of an authoritarian personality.

The difficulty of examining this question is of course the finding alluded to earlier that our usual authoritarianism attitude scales do not predict behaviour. If they do not predict behaviour, they cannot be valid as measures of personalities that express themselves in authoritarian behaviour. Fortunately, the solution to this dilemma is rather simple. If we wish to study authoritarian personalities, we must use personality scales, and not attitude scales.

The Ray (1976) Directiveness scale was designed from the beginning as a measure of authoritarian personality. Instead of asking people whether they approve of people being dominated by others, it asks them whether they themselves tend to dominate others in various situations. It should be no surprise that people might enjoy dominating others but still not approve of it in general. Because of its behaviour-inventory format, the Directiveness scale was found to correlate highly (0.54) with peer-rated authoritarian behaviour. It is then a highly valid measure of the personality sources of authoritarian behaviour. As such, its correlation with mental health measures should represent a considerable advance in the rigour with which we can examine the postulated relationship between mental ill-health and the authoritarian personality.

Ray (1976) has shown that this valid measure of the authoritarian personality does not correlate at all with favourable attitudes to authority. We find then that attitudes, behaviour and personality in this field fall into two clusters - with personality and behaviour in one cluster and attitudes in isolation from the other two. Since it is surely behaviour that we are always ultimately interested in, most of the existing research on authoritarianism has apparently chosen the wrong cluster.

Nonetheless, the study to be reported below will examine the mental health not only of authoritarian personalities, but also of authoritarian attitudes as expressed by the California F-scale. The reason why further study of attitudes is still seen as of some interest is that so much of the previous work has been based on acquiescence-prone unbalanced scales. With the emergence of a balanced F-scale (Ray, 1972), this no longer needs to be so. The BF-scale has not only high reliability, but also a negative correlation between its two halves varying between -0.71 and -0.53. All its items are reversals of F-scale originals. The correlation between this scale and mental health measures would be more conclusive than most of the existing findings.

This correlation becomes of even greater interest if we note that results with previous scales are hardly univocal. Besides the evidence for mental ill-health among those with authoritarian attitudes presented by Adorno et al. (1950) and their successors, there are also many findings showing no relationship between various measures of such attitudes and mental ill-health. Without attempting a detailed review, findings by Masling (1954), Roberts and Jessor (1958), Elms (1970), Richek, Mayo and Puryear (1970), Ray (1971) and Crabbe (1974) may be mentioned.

One qualification that should be made is that in both its original and its balanced form the F-scale is not a very clear measure of authoritarian attitudes, let alone authoritarian behaviour. Relatively few F-scale items make even oblique references to authority. Judged on explicit content alone, the F-scale would have to be taken as measuring some brand of tough-minded conservatism rather than approval of authority. Ray (1974) has, however, shown that the BF-scale correlates 0.539 with acceptance of conventional sources of authority.


The Ray (1976) Directiveness scale was included in a questionnaire together with the short form of the Eysenck Neuroticism scale and a social desirability scale. The social desirability scale was a short form of the Marlow Crowne instrument selected from the item analysis provided by Greenwald and Satow (1970).

The Eysenck Neuroticism scale was selected because of its demonstrated factorial primacy among mental health measures (Eysenck & Eysenck, 1969). The social desirability scale was included for control purposes. Since acquiescence was controlled by the balanced structure of the Directiveness scale, the present work should hence be particularly rigorous in excluding response set interpretations of any results obtained.

Also included in the questionnaire was a question on political party preference -- the answers to which were scored from 5 (most Rightist) to 1 (most Leftist). See Ray (1971) for further details of this scoring method.

The questionnaire was administered to a random sample of people in the Sydney metropolitan area. Cluster sampling was the method used. This method is widely used by commercial public opinion polls - where it generally gives quite accurate results. The sample N was 95.


In the original validation study for the Directiveness scale, results were presented separately for both the full scale and its short form. This practice is continued here. The short form contains only items devised by the present author while the full scale adds items suggested by students. In the original validation study, although these two forms were found to correlate 0.901, there was a noticeable difference in their correlations with social desirability. The full scale correlated -0.244 while the short scale correlated only -0.150. These results were also found on the present occasion. The respective correlations were -0.252 and -0.135. Only the former is significant.

The reliabilities observed were 0.70 for the short Directiveness scale, 0.79 for the full Directiveness scale, 0.73 for the Neuroticism scale, 0.77 for the Social Desirability scale and 0.87 for the Balanced F-scale. The use of short forms has, then, not resulted in any serious lack of reliability.

The correlations with neuroticism were -0.215 for the short Directiveness scale, -0.177 for the full Directiveness scale and -0.109 for the BF-scale. The negative sign indicates that high authoritarianism went with low neuroticism. Only the first two correlations are, however, significant. On correction for social desirability component, the first two correlations drop to -0.195 and -0.138. Only the first of these is significant.

The short and the full Directiveness scales correlated respectively 0.272 and 0.217 with the BF-scale. This indicates that there was on this occasion some tendency for authoritarian attitudes and authoritarian personality to be associated. The correlation between the two halves of the BF-scale was -0.65.

The correlation between the BF-scale and social desirability was 0.209. When desirability is partialled out, the correlation between BF and neuroticism hence rises to -0.151. This is, however, still non-significant. Authoritarian attitudes were then seen as desirable while authoritarian behaviour was seen as undesirable.


Research into the authoritarian personality has been much plagued by methodological difficulties. When the authoritarian personality is, however, measured in an intrinsically set-free and valid way we find that, far from being pathological, the authoritarian personality is healthier than normal.

Additionally, when controls for acquiescence and social desirability are introduced into the measurement of authoritarian attitudes, we find correlations which, although too low for significance, do at least tend in the opposite direction to that proposed by Adorno et al. (1950). With a larger N, it might be hypothesized that this correlation would reach significance. As it stands, however, the lack of relationship between the BF-scale and neuroticism is entirely consistent with previous results obtained using other scales by Masling (1954), Crabbe (1974), Roberts and Jessop (1959), Richek, Mayo and Puryear (1970), Elms (1970), Fraccia, Sheppard, Pintyr, Crovello and Merlis (1972), Eckhardt, Manning, Morgan, Subotnik and Tinker (1967), Gaensslen, May and Woelpert (1973) and Ray (1971). Even considering authoritarian attitudes alone, we must conclude that after 30 years of research the California case is at the least still 'not proven'.

That authoritarians should be better adjusted than normal is not hard to understand. A dominant personality is probably in a better position to get what it wants than a non-dominant one. Additionally, a personality could probably not become dominant without considerable social skills to start with. Leaders in all spheres of life do seem to be rather scarce. How else can we explain the difficulty political parties have from time to time in finding remotely plausible candidates? When a person who likes to take on leadership tasks is found, therefore, his rewards should be considerable. It is not hard to understand that a higher level of reward should go with reduced anxiety and neurosis.

The unexpected significant correlation between the BF scale and directiveness should be seen in the light of the observation made earlier that the F-scale correlates with attitude to authority rather than measuring it directly. The least inferential conclusion we could draw from the results obtained here would be that we have shown tough-minded conservative attitudes to be related to directiveness in personality. As such, this is hardly a surprising result. Dominant people tend to be tough minded.

That the correlation was owing to the tough-minded component of the BF-scale rather than its conservative component, was suggested by the correlation between the Directiveness scale and political party preference. At -0.008 this was non-significant.

The most certain conclusion we can draw from the present study is that in both the personality and attitude senses, authoritarianism is not psychopathological. In its behaviourally most relevant sense, authoritarianism may even be an indicator of good mental health.


{Articles below by J.J. Ray can generally be accessed simply by clicking on the name of the article. I am however also gradually putting online a lot of abstracts, extracts and summaries from older articles by other authors so if an article not highlighted below seems of particular interest, clicking here or here might just save you a trip to the library}

Adorno, T.W., Frenkel-Brunswik, Else, Levinson, D.J. & Sanford, R.N. The Authoritarian Personality. New York: Harper, 1950.

Bennett, N. Teaching Styles in Pupil Progress. London: Open Books, 1976.

Bourne, P.C. Altered adrenal function in two combat situations in Viet Nam. In Eleftheriou, B.E. & Scott, J.P., The Physiology of Aggression and Defeat. New York: Plenum, 1971.

Brink, T.L. The case of Hitler: An Adlerian perspective on psychohistory. J. Individ. Psychol. 1974, 30, 23 - 31.

Crabbe, B.D. Are authoritarians sick? In Ray, J.J. (Ed.), Conservatism as heresy. Sydney: A.N.Z. Book Co., 1974.

Dicks, H.V. Licensed Mass Murder: A Socio psychological Study of some S.S. Killers. London: Sussex U.P., 1972.

Eckhardt, W., Manning, M., Morgan, C., Subotnik, L. & Tinker, L. J. Militarism in our culture today. J. Human Relations, 1967, 15, 532 - 537.

Elms, A.C. Those little old ladies in tennis shoes are no nuttier than anybody else: It turns out. Psychol. Today, 1970, 3, 27ff.

Eysenck, H.J. & Eysenck, S.B.G. Personality Structure and Measurement. London: Routledge, 1969.

Fenster, C.A. & Locke, B. Neuroticism among policemen: An examination of police personality. J. Applied. Psychol., 1973, 57, 358 - 359.

Fraccia, J., Sheppard, C., Pintyr, J., Crovello, J. & Merlis, S. Personal adjustment and authoritarian attitudes toward the mentally ill. Psychol. Reports, 1972, 31, 483 - 486.

Gaensslen, H., May, F. & Woelpert, F. Relation between Dogmatism and anxiety. Psychol. Reports, 1973, 33, 955 - 958.

Greenwald, H.J. & Satow, Y. A short social desirability scale. Psychol. Reports, 1970, 27, 131 -135.

Guetzkow, H., Kelly, E.L. & McKeachie, W.J. An experimental comparison of recitation, discussion and tutorial methods in college teaching. Educ. Psychol, 1954, 45, 193 - 207.

Hines, V.A. F-scale, GAMIN, and public school principal behaviour. J. Educ. Psychol., 1956, 47, 321 - 328.

Koomen, W. A note on the authoritarian German family. J. Marriage and the Family, 1974, 36, 634 - 636.

Kren, G. & Rappoport, L. S.S. atrocities: A psychohistorical perspective. History of Childhood Quarterly: J. Psychohistory, 1975, 3, 130 - 137.

Larsen, K.S., Coleman, D., Forbes, J. & Johnson, R. Is the subject's personality or the experimental situation a better predictor of a subject's willingness to administer shock to a victim. J. Per. & Soc. Psychol., 1972, 22, 287-295.

Lefkowitz, J. Psychological attributes of policemen: A review of research and opinion. J. Social Issues, 1975, 31, 3 - 26.

Lippitt, R. & White, R.K. An experimental study of leadership and group life. In Newcomb, Hartley et al., Readings in Social Psychology. New York: Holt, 1947.

Masling, M. How neurotic is the authoritarian? J. Abnorm. Soc. Psychol., 1954, 49, 316 - 318.

McCurdy, H.G. & Eber, H.W. Democratic versus authoritarian: A further investigation of group problem solving. J. Personality, 1953, 22, 258 - 269.

Perry, A. & Cunningham, W.H. A behavioural test of three F-scales. J. Social Psychol. 1975, 96, 271 - 275.

Ray, J.J. (1971) An "Attitude to Authority" scale. Australian Psychologist, 6, 31-50.

Ray, J.J. (1972) A new balanced F scale -- And its relation to social class. Australian Psychologist 7, 155-166.

Ray, J.J. (1974) Are the workers authoritarian, conservative or both? Ch. 43 in Ray, J.J. (Ed.) Conservatism as heresy Sydney: A.N.Z. Book Co.

Ray, J.J. (1976) Do authoritarians hold authoritarian attitudes? Human Relations, 29, 307-325.

Richek, H.G., Mayo, C.D. & Puryear, H.B. Dogmatism, religiosity and mental health in college students. Mental Hygiene, 1970, 54, 572-574.

Roberts, A.H. & Jessor, R. Authoritarianism, punitiveness and perceived social status. J. Abnorm. Soc. Psychol., 1958, 56, 311 - 314.

Roberts, S.H. The House that Hitler built. New York: Harper, 1938.

Salas, R.G. & Richardson, J.F. Some Australian data on forms A and B of the EPI. Austr. J. Psychol., 1968, 20, 11 - 13.

Singer, E.A. & Wotton, L.M. The triumph and failure of Albert Speer's administrative genius: Implications for current management theory and practice. J. Applied Behav. Sci., 1976, 12, 79103.

Taylor, A.J.P. The Origins of the Second World War. London: H. Hamilton, 1961.

Trevor-Roper, H. The Last Days of Hitler. New York: Macmillan, 1947.

Unger, A.L. Party and state in Soviet Russia and Nazi Germany. Political Q., 1965, 36, 441 -459.

Ward, W.D. & Barcher, P.R. Reading achievement and creativity as related to open classroom experience. J. Educ. Psychol., 1975, 67, 683 - 691.

Go to Index page for this site

Go to John Ray's "Tongue Tied" blog (Backup here)
Go to John Ray's "Dissecting Leftism" blog (Backup here)
Go to John Ray's "Australian Politics" blog (Backup here)
Go to John Ray's "Gun Watch" blog (Backup here)
Go to John Ray's "Education Watch" blog (Backup here)
Go to John Ray's "Socialized Medicine" blog (Backup here)
Go to John Ray's "Political Correctness Watch" blog (Backup here)
Go to John Ray's "Greenie Watch" blog (Backup here)
Go to John Ray's "Leftists as Elitists" blog (Not now regularly updated)
Go to John Ray's "Marx & Engels in their own words" blog (Not now regularly updated)
Go to John Ray's "A scripture blog" (Not now regularly updated)
Go to John Ray's recipe blog (Not now regularly updated -- Backup here)

Go to John Ray's Main academic menu
Go to Menu of recent writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page (Backup here)
Go to Selected pictures from John Ray's blogs (Backup here)
Go to Another picture page (Best with broadband)