Make your own free website on

Chapter 10 (pp. 155-173) in: S. Modgil C.M. Modgil (Eds.) "Hans Eysenck: Consensus and controversy" Lewes, E. Sussex, U.K.: Falmer, 1986.



Although it is probably a common impression that Eysenck's work on social attitudes is limited to a single foray, in the form of his 1954 book, The Psychology of Politics, social attitudes in fact constituted one of his very earliest interests -- an interest which continues to this day. To my knowledge his earliest paper on the topic was published during the Second World War (Eysenck, 1944) and he continued to defend his position as recently as 1981 (Eysenck, 1981/82), Nonetheless, it is clear that The Psychology of Politics is his major statement in the area. A collection of his later work, together with some minor updating of his position is, however, available in a recent book (Eysenck and Wilson, 1975), also ambitiously entitled The Psychological Basis of Ideology.

Reading Eysenck's earliest writings is an experience surprising for the discovery that what Eysenck is saying now has changed so little in his lifetime. Whatever else he may be, he is extraordinarily consistent about his basic themes. In one of his very earliest papers (Eysenck, 1940) we see perhaps the first sign of what was to become a besetting habit of thought for Eysenck -- the tendency to describe almost anything in terms of two dimensions. In this paper, presumably written before the war, we find Eysenck explaining the appreciation of poetry in terms of the two dimensions of extraversion and neuroticism. He still uses these variables as major personality descriptors -- though quite recently supplemented by a third dimension: 'P' or 'psychoticism' (Eysenck and Eysenck, 1976).

In his 1944 paper (written in 1942) Eysenck pooled the work of several previous attitude researchers and factor analysts in an endeavour to find what their various results had in common. He concluded that two dimensions of social attitudes could be detected: radicalism-conservatism and 'practical- theoretical'. The latter concept he also identified with the 'tough-mindedness' versus 'tender-mindedness' of William James. He then proceeded to his own factor analysis of some data which he obtained from Flugel and Pryns. These data consisted of responses to a series of contentious issues by members of a number of rather eccentric-sounding special-interest groups in pre-war England. The most amusing of the issues was 'Gymnosophy' -- which turns out to be nudism. Eysenck found three factors but was able (surprise) to give confident interpretations to only two of them. He concluded that these two factors were very similar to those he had just identified in the work of previous authors. If we look at the high-loading items on his second factor as given by Eysenck himself, however, we find that for his largest and apparently least eccentric group of subjects there were only three items that loaded really highly. They were all negative loadings -- indicating that they were 'theoretical' or 'tender-minded' in Eysenck's terms. They were: 'Abstemiousness', 'Vegetarianism' and 'Non-smoking'. I in my naivety when I first looked at these items failed to note the negative sign and thought that Eysenck was proposing 'Abstemiousness', 'Vegetarianism' and 'Non-smoking' as signs of tough-mindedness. Given the strength of will required to give up smoking and the delights of the carnivore, I could see some glimmer of sense in the proposal. It turns out, however, that Eysenck is asserting the opposite. It is apparently 'tender-minded' to be abstemious. The two highest positive loadings were 'Birth-control' and 'Abortion'. Perhaps it is true to say that practising birth control and allowing abortions is tough-minded but surely that is a very incidental judgment rather than being what the factor is about. Australians might have called it the 'wowser' factor. (In Australia 'wowsers' are Methodists, morals campaigners, teetotallers, etc. The best translation into standard English might be 'killjoy'. It is a very dismissive term.) 'Oldfashioned asceticism' or 'Puritanism' might be other reasonable names for what it measures. I cannot to this day see why it is infinitely less tough-minded to be a non-smoker or a vegetarian than it is to practise contraception (the loadings indicate that non-smoking and contraception are opposites on whatever it is that the factor measures) but I can see that it is more Puritan. Already in this 1944 paper, then, we have the first sign of one of Eysenck's traits that is to figure largely later on: an ability to see in factor-analytic results much more than others would be likely to see.

It must be pointed out at this early stage, however, that there has always been a peculiar leniency shown in the psychological literature towards factor analysts. I have already set out at length elsewhere (Ray, 1973c) an example of just how arbitrary factor-naming can be in the psychological literature generally, so I will not repeat it here. The gist of it, however, is that a factor which was initially said to measure 'authoritarianism' in a preliminary version of a paper became a measure of 'Australian chauvinism' in the final version. Although there is no doubt something in common between the two concepts, I would submit that there is also a lot of difference. We have, then, a rather good example of how little rigour there commonly is in interpreting the results of factor analysis. Factors are taken as measuring what the analyst says they do and any questioning of his interpretation is extremely rare. Proof that a factor measures anything at all is even rarer. There has somehow developed a tradition that exempts factor analysts from the rigours of proof (validity demonstrations) that are expected of those who construct scales by other means. Almost anyone who has ever done a factor analysis must know what an odd assortment of items one often finds, all loading high on one factor. Identifying the common thread in these items is almost always a task requiring considerable imagination and creativity; so much so that it is not uncommon for new words to be invented for the purpose. Although he may have other justifications for doing so, Cattell's use of such words as 'surgency' and 'rhathymia' to describe his factors seems a good instance of this. When, therefore, the poor, benighted factor analyst has finally managed to 'identify' his factors, no-one usually has the heart to ask him for proof that his items really do measure what he says they do. The fact that some common thread can be perceived in a purely conceptual way between many disparate items (face validity) is generally counted a sufficient achievement. The analyst is normally not even expected to show that the measure provided by his factor is reliable -- despite the fact that scales derived from factor analysis can easily turn out to be anything but reliable (e.g., Ray, 1971a). Eysenck, therefore, was probably somewhat shielded by this tradition. To have questioned the interpretation of his second factor would have breached one of psychology's guild-rules. Even so, as we shall later see, such questions were finally raised.

As it happened, Eysenck's 'discovery' of two main factors in social attitudes was fortuitous. It formed the basis for what was to become an attractive solution to a considerable puzzle of twentieth century politics: that the further out on the Right or the Left one moved, the more one began to notice that those on the Right and on the Left had a lot in common. Right- and Left-wing extremists, instead of being utterly different from one another, in fact seemed remarkably similar. To go from the extreme Left to the extreme Right was like travelling in a circle. You ended up somewhere remarkably like where you started. This was perhaps more evident before 1945 than it is now. Although there was a certain sense in which one was Right-wing and the other was Left-wing, Hitler's, Germany and Stalin's Russia had striking similarities. This 'same but different' phenomenon was something that anyone with any political consciousness would have had to cope with in the 1930s and 1940s.

Eysenck's habits of thought led to what still is a very clever solution to this puzzle. He maintained that political allegiances should be conceived not on one dimension but on two -- with an addition to the traditional radical-conservative dimension of tough-tender mindedness. Thus Fascists and Communists were the same in that both were high on tough-mindedness but different in that one was radical and the other was conservative. The two major parties traditional in Anglo-Saxon countries, on the other hand, were unified in being much more tender-minded than the totalitarian parties of Europe.

As well as making this powerful proposal for the description of existing political reality, Eysenck took the much bolder step of trying to show that such dimensions existed in the minds of men. His two dimensions were to be not mere political abstractions but factors of social attitudes -- products of the empirical procedures of factor analysis when such were applied to a large body of expressions of social attitudes made by ordinary people. His was a purported discovery about people. His dimensions did not exist just in the mind of some ivory-tower sophist. He proposed, in other words, a congruence between political and psychological reality. There were tough-minded radical governments because there were tough-minded radical people. I, as it happens, concur with Eysenck in believing that Communists tend to be (among other things) tough-minded radicals. I do not, however, believe -- as we shall see -- that Eysenck has succeeded in showing in his own empirical work that Communists are in fact tough-minded radicals.

Although Eysenck's theory was something of an intellectual breakthrough, he was unfortunate that the Zeitgeist had changed by the time that he got his 'discovery' into print. The theory was first published as such in 1954 in The Psychology of Politics. By that time the great Nazi monster had fallen and the political scenario no longer mirrored the psychological scenario that Eysenck was putting forward. Indeed, the euphoria of war-time co-operation with Russia made even the Bolshevist monster seem not so bad after all. For some reason this euphoria lasted much longer among intellectuals than it did among the general public. While the Berlin airlift of 1949 probably marked fairly well the end of optimism about Russia as far as the general public was concerned, intellectuals continued for much longer to accept the idealism of Communism at face value. They seemed to see the Russian Communists as simply Leftists who had been particularly successful at implementing their programme. This view was largely shared by the Left generally. I can remember how even into the 1960s Australian Leftists would dismiss accounts of the 30 million who died under Stalin as fabrications of 'the capitalist press'. When it gradually became known through Communist sources that in his 1956 secret speech to the twentieth Communist Party Congress the chairman of the Communist Party of the Soviet Union himself (Nikita Khrushchev) had confirmed the truth of these 'lies', that particular perceptual defence crumbled but others seemed to replace it fairly readily. 'The pressure of external military threat' seems to be a popular excuse for the brutalities of the Soviet system among Marxist intellectuals nowadays, but since it was during war-time that Hitler did away with the Jews, I cannot see why Hitler could not similarly be excused! Before the Second World War many intellectuals found much to admire in Hitler's 'New Germany', so it might be a reasonable hypothesis that intellectuals are just susceptible to idealism of any kind regardless of its real outcomes. In the post-war era, however, the demise of Hitler meant that only Communism remained to draw the loyalties of Western intellectuals. In this climate the need was to distance ourselves from Hitler and to find out how Fascism came about, so that any resurgence of it could be prevented. Military defeat of the Fascists had made their own explanations for themselves dismissible, and explanations of Fascism had to be found which would be so damning that no-one could ever seriously entertain it again. Eysenck's work did little to fill this need.

A work that splendidly filled this need, however, was ironically largely a re-hash of an old Nazi theory -- The Authoritarian Personality by Adorno, Frenkel-Brunswik, Levinson and Sanford (1950). Working mainly in California, this group of American and expatriate German Jews took the Nazi typological theory of Jaensch (1938), reversed its value judgments, added Freudian elements to its originally largely perceptual emphasis and showed that it applied in America as well as in Germany. Eysenck (in Eysenck and Wilson, 1978) records that one of the authors of the 'California' study (Else Frenkel-Brunswik) freely acknowledged to him, the role of Jaensch's theory in forming her own thinking even though Jaensch was not listed in the references of The Authoritarian Personality.

The changed world political scene of the 1950s and 1960s made a one-dimensional account of politics plausible and even desirable. The Authoritarian Personality was nothing if not one-dimensional. According to this book, Fascists were simply extreme conservatives and almost all ills, mental and social, could be traced to or associated with a Rightist ideology. Both Fascism and Conservatism were presented as more or less extreme manifestations of an underlying 'authoritarian' personality (measured by the famous 'F' (for 'Fascism') scale), and such a personality was a 'disease'. One must add that given the role of Marxist intellectuals (such as Adorno) in the composition of this book, the thrust of its conclusions cannot be regarded as entirely surprising. This strange book, then, effectively denied any possibility of such a thing as Left-wing authoritarians and ignored such prominent facts as the struggle against Hitler being led by a Right-wing conservative British leader, Winston Churchill. There could be no-one who throughout all his life opposed more violently all that Hitler had stood for, yet Adorno et al. were saying that the two were really the same. As his pact over the partition of Poland showed, Joseph Stalin had been quite happy to co-operate with Hitler. It was the conservatives, fighting against great odds, who brought about the destruction of Fascism, not the Leftists. Yet from Adorno et al. we would expect Hitler and Churchill to be brothers! Sibling rivalry might have been invoked, but Adorno et al. did not even seem to see that there was anything there to explain. To them psychoanalytic speculation was far more real than the world of everyday politics.

If The Authoritarian Personality was very poor at providing an explanation for the characteristic contempt that pre-war conservatives and Fascists had for one another, Eysenck's The Psychology of Politics made it crystal clear. Fascists were tough-minded and conservatives were tender-minded. Fascists saw conservatives as weak and ineffectual. Conservatives saw Fascists as brutal and aggressive. The opposition between the two arose because they were in fact opposites in important respects. In the circumstances one might be forgiven if one assumed that The Psychology of Politics must have soon eclipsed The Authoritarian Personality in the influence it had upon psychological researchers. Eysenck's work was elegant, clear, careful, objective and in general very 'scientific' by the standards of the day. The Authoritarian Personality, by contrast, was a nightmare of subjectivity, intellectual dishonesty and almost complete lack of scientific caution (Christie and Jahoda, 1954; McKinney, 1973). The fact is, of course, that The Authoritarian Personality had infinitely greater impact on psychologists than Eysenck's work did. In the immediate post-war era intellectuals generally wanted to believe the best of Stalin and the Russian experiment. To accept Eysenck's account would have meant accepting that Communism was in many important ways similar to the now universally decried Nazism. No wonder Eysenck was unpopular! When the world had just suffered so grievously at the hands of a Right-wing tyrant, who wanted to believe that similar perils from a Left-wing tyranny might be in the offing? Social scientists might be able to accept that Leftists can be 'dogmatic' (a surely minor charge -- is not Mother Teresa of Calcutta dogmatic about God's love for the poor?) in Rokeach's (1960) terms, but anything more threatening was rejected.

The scientific means that enabled Eysenck's work to be substantially ignored was a remarkable series of articles in a prestigious American journal (the 1956 Psychological Bulletin) which appears to have caused even Eysenck to abandon the field for many years. Four papers by Rokeach, Hanley and Christie appeared in this volume which mounted most scathing attacks on Eysenck and his work in social attitude and political research. It may be noted that although Rokeach is the author of what might be seen as a 'rival' theory to Eysenck's, Christie at least was even-handed in that he is also a severe critic of The Authoritarian Personality (Christie and Jahoda, 1954; Christie, Havel and Seidenberg, 1956).

The criticisms made of Eysenck all seem to have been well-supported: Eysenck's sampling was rudimentary; he did use an old-fashioned (pre-computer) 'counting' system to score his items which distorts the meaning of non-responses; he did make a number of minor mistakes in the citation details of references he used; he did assign identities to his factors which require considerable imagination to be seen as justified. The point about these criticisms, however, is that similar criticisms could be levelled against almost any other study in the field at that time. A level of rigour and care was demanded of Eysenck that was far in excess of what was demanded of others. Perhaps the clearest indication of how nit-picking the criticism was is that Christie took Eysenck to task in round terms (Christie, 1956b, p. 446) for referring to the California F scale as a measure of 'authoritarianism'! If Christie were to tear out just one hair for every time that someone has referred to the F scale as such a measure, he would have been a bald man twenty years ago!

In particular, the criticisms of Eysenck's sampling were extraordinary. To this day at least 95 per cent of social psychological research that is published in the journals makes no attempt at sampling whatsoever. Psychology as a discipline seems to be characterized by the absurd belief that what is true of an unselected group of American college students will be true of humanity in general (Ray, 1981). In this context Eysenck should be something of a scientific hero. He did at least make a rough attempt to find out what was the case among the general public. Yet he was singled out for a criticism that is politely left unmentioned in almost all other psychological research. I, as it happens, agree fully with Christie that Eysenck's sampling inadequacies totally vitiate the major conclusions he wished to draw from his data, but I would reject most of the literature on authoritarianism as totally worthless for the same reason. Would Christie do the same? His own continued use of student 'samples' would suggest not. It seems clear that Eysenck was taken to task not because his work was particularly bad but because his conclusions were politically unacceptable. Eysenck's biographer also gives very strong support to this conclusion (Gibson, 1981).

Perhaps the most damaging criticism of Eysenck was made by Rokeach and Hanley (1956) -- that Eysenck in some way 'fudged' his data by presenting false mean scores for the various political groups he studied. This was an almost unprecedented direct attack on a scientist's honesty. The way this terminally grave accusation was supported was to attempt a speculative reconstruction of scale means for the various groups from item mean data presented by Eysenck in an earlier publication. They give little weight to the possibilities of rounding error, and they do admit that they have no way of allowing for non-responses. They obtain total score means that differ from Eysenck's and claim as a result that Eysenck's own data reveal him as some sort of crook. Eysenck treats this extravagant inference with the contempt it deserves by simply pointing out that he and his critics scored the same data in different ways so of course the results must differ. He might have been better served in American eyes to stand less on his dignity and become as litigious over the matter as Americans would surely be in similar circumstances. What might seem a properly reserved and dignified response in British eyes could well be seen as an admission of guilt in American eyes.

The major difference between Eysenck and his critics would seem to have been over the content of Eysenck's 'T' scale. Neither side of the argument seemed to be listening to the other at all. There is no doubt that the factor-analytic manoeuvres adopted by Eysenck for the derivation and construction of his T scale were unusual. That is not to say, however, that they were unjustified. Eysenck, like everyone else before and after him, was confronted by the dilemma that although the form of government they practise when in power is terminally authoritarian, Communists will never admit to anything but the most liberal, tolerant and humanistic ideology. Although political practice may quite evidently vary in two dimensions, political ideology seems to vary only in one dimension. The only people who normally have a kind word for any sort for authority are the conservatives and, as both the Second World War and the subsequent Cold War show, it is precisely conservatives who are the most unrelenting opponents of authority when it gets carried away with itself. To me the gap between Communist beliefs and Communist practice is simply the most vivid possible proof of what a vast and dangerous pathology wishful thinking really is. Eysenck shared with the authors of The Authoritarian Personality the strangely naive view that if people behaved in a certain way, then there must somewhere be, in those people, an attitude of some sort which directly corresponded to that behaviour and justified it. The possibility that large slices of humanity could be so misled by wishful thinking as to deny, even to themselves their own real motives (even to the bitter end) could somehow just not be allowed. Freud would certainly have had no such difficulty. Perhaps it was really a sort of arrogance. Both the California authors and Eysenck thought that if people really did have evil but hidden motives, the psychologist with his bag of tricks must surely be able to penetrate the disguise. Perhaps in general there are some grounds for such confidence, but when the people upon whom the bag of tricks are to be used (in this instance Marxists and Communist sympathizers) are taken from among one's own colleagues and students, the tricks cannot be expected to fool anyone very much.

At any event, if there is a finding that all the tough-minded, pro-authority statements are assented to by conservatives, there are two possible interpretations of that finding: (1) one can conclude, as Adorno et al. (1950) did, that this proves that only conservatives are authoritarian, or (2) one can conclude, as Eysenck does, that there is something wrong with the analysis that produced such an absurd conclusion. Both Christie and Eysenck appear to agree that the conclusion is absurd, but they disagree over what to do about it. Christie thinks that if we just look harder we will find attitude statements that will show Communists as tough-minded. Eysenck, who has already looked very hard, knows that that will not work and proposes instead that what we will have to look hard at are not our data but our methods of analyzing them. He therefore quite explicitly describes and justifies at great length an arbitrary rotation of his factors which in his view makes greater sense than some mechanical rotation. He then refers to one of his arbitrarily rotated factors as reflecting 'toughmindedness'. The 'arbitrary' rotation was, however, supported by a very plausible theory. Eysenck said that it was unreasonable to expect that political attitudes could ever be 'just' tough-minded. They had to be either Right-tough or Left-tough. One cannot be tough-minded without some opinion to be tough about. This eminently reasonable proposition, however, seems to have gone in one ear and out the other as far as Eysenck's critics were concerned. They observed that there were no items which loaded on Eysenck's T factor alone (a terrible sin in factor analysis), and seemed in consequence to have thought that they had caught Eysenck in yet another deception. Following orthodox factor-analytic thinking, they concluded that there could not in the circumstances be said to be any T factor there at all. They seemed oblivious to the fact that it was precisely the adequacy of orthodox factor-analytic procedures that Eysenck was calling into question. No-one could have put his proposals with greater energy, clarity and persuasiveness than Eysenck, but they seem to have gone straight over the head of his critics. Orthodox thinking evidently had too strong a grip on their imaginations. Rokeach and Hanley (1956), in other words, presented it as a discovery that Communists scored high on only some of Eysenck's 'tough-minded' items. They seemed to think that in so doing they had caught Eysenck out in some way. Yet Eysenck himself had already demonstrated at great length that this was precisely what his theory required. Their discovery was one that Eysenck had already been shouting from the rooftops.

Where Eysenck was vulnerable was in the content of his T ('tough-mindedness') items. He had not yet shown that any of his items measured tough-mindedness. Had his critics simply hammered this point they might have made a more genuine contribution. The extensive re-tabulation of results they undertook which had only the effect of showing what Eysenck had shown already were superfluous. To extract new generalities from the items loading high on Eysenck's factors, they needed to overcome the traditional indulgence shown over factor-naming (alluded to at the beginning of this paper). As a factor analyst, Eysenck had a certain protective mantle. Faults other than his factor-naming had therefore to be found by his critics.

Amid all their other criticisms Eysenck's critics did manage to make some fairly telling criticisms about factor-naming. They are points that Eysenck has never answered. Had his critics focused their criticisms, instead of criticizing so many things, Eysenck might have been forced to defend his factor-naming.

An abiding criticism of Eysenck's 1954 book, then, is the same as could be made of Eysenck's 1944 work: Eysenck sees depths and meanings in attitude items which run far beyond what it is reasonable to see. This can perhaps best be seen in Eysenck's very latest version of his R and T factors. In Eysenck (1976) the high-loading items on his T ('tough -mindedness') factors are as follows: 'wife-swapping', 'patriotism', 'self-denial', 'moral-training', 'chastity', 'royalty' and 'casual living'. These were the only items with loadings above .40. 'Tradition' and 'divine law' loaded .39 with 'censorship', 'inborn conscience' and 'Bible truth' loading .38. Eysenck is certainly correct in seeing continuities between this factor and his earlier results from 1944 onwards, but how to identify what underlies the items concerned is not nearly as obvious as Eysenck implicitly claims. The very obvious theme in such items is not toughmindedness but old-fashioned morality or religious morality. Religion and morality have always been closely intertwined, and restrictive morality is now increasingly oldfashioned. I would submit that each of the items listed above describes something religious, moral (in the sense of sexual morality) or old-fashioned. That is what the factor is all about. To say it concerns 'tough-mindedness' is a farce.

I am not saying that certain types of religious or anti-religious sentiment cannot be tough- or tender-minded. If political sentiments can be tough- or tender-minded so surely can religious sentiments be tough- or tender-minded. The point is that we have no evidence that any of the sentiments identified by Eysenck as tough-minded are in fact tough-minded. All we have is Eysenck's word for it. To him the factor exists, can be replicated and therefore must measure something, and he as the factor analyst is the one who must have the decisive word on what that something is. Eysenck writes as if it is obvious what the factor measures, but the emperor in this case has long ago been declared to have no clothes. To put the matter another way: we could say that the items of Eysenck's T factor are either definitionally tough-minded (which is what Eysenck appears to say) or that they are empirically tough-minded. In the latter case we have an empirical hypothesis that can be tested by correlating the T factor with some independently validated measure of tough-mindedness. No-one has attempted this. In the former case we simply have to argue over what the words of the items loading highly on the factor actually say. Do they of themselves embody toughmindedness of some (or any) sort as a consistent and obvious underlying theme, or is some other theme (or themes) apparent? If I believe in casual living and wife-swapping, does that clearly and explicitly make me tough-minded? Might it not be rather more likely to make me other things -- such as uninhibited or decadent? If the items that loaded highly on Eysenck's T factor said such things as 'power', 'authority', 'strength', 'honour', 'aggression', 'destruction', then we might be quite strongly inclined to believe that we were dealing with a tough-mindedness factor, but even then this would be only a preliminary hypothesis. (They might, for instance, measure 'hostility' rather than tough-mindedness.) If we wished to use such a factor to show something about any group in the population, we would then have to produce the sort of validating evidence that is normally expected of any psychological scale -- peer ratings showing that people who assent to such items are in fact seen by others as tough-minded, etc. All this, however, is very far from the situation prevailing with Eysenck's T scale.

The way that Rokeach and Hanley (1956) made much the same point was to show that people are classified by Eysenck as tough-minded Leftists if they oppose Sunday observance, religion and compulsory religious education on the one hand, and on the other hand support abortion, divorce and trial marriage. To call what such a group of opinions have in common anything but religion and morality is fairly inconceivable but, nonetheless, to Eysenck they measure 'tough -mindedness'.

It seems inevitable that we must ask why Eysenck adopts such a peculiar view. Why is it a view that he never even tries to justify? One might say that there are, after all, some items on the T scale which are not directly concerned with religion and morality. These tend to be the low-loading items (i.e., items not very central to the factor) and in any case merely demonstrate that religious people do have some characteristic opinions on non-religious issues. They tend, for instance, to be traditionalists or to be ascetic. Even in such cases, however, the religious element is seldom far away. To be ascetic, for instance, is a very great tradition in many religions and is a quite conventional sign of holiness for many people. To be in favour of royalty is in England not at all irrelevant to religion. Most English people are at least nominal members of the Church of England, and guess who is the Head of the Church of England?

Another possibility is that Eysenck is really laughing up his sleeve. His biographer (Gibson, 1981) believes that Eysenck sometimes writes books (e.g., his first book on race and IQ) with the deliberate expectation that this will bring down opprobrium on his head. If one believes that there is no such thing as bad publicity, this could be a reasonable thing to do. Eysenck is one of the world's most cited authors by his fellow psychologists (Gibson, 1981), and it is my impression that it is the loose and subjective writings in psychology that attract attention. Really rigorous work tends not to attract much interest from one's colleagues. Who, for instance, could imagine a more hilarious proposition than that we should turn to a group of Jewish Marxists for a dispassionate and rigorous account of the sources of Nazism? Would one be tempted to treat Ulster's Rev. Ian Paisley as an authority on the Pope? Yet psychologists to this day still seem to take much of their thinking on Fascism (and hence conservatism) from The Authoritarian Personality -- a book written by authors who were entirely Jewish and at least some of whom were Marxists! I infer that 'attackability' is almost a precondition for relevance in psychology. Oscar Wilde once made an exhortation that ran something along the lines of. 'Remain, as I do, incomprehensible; to be great is to be misunderstood.' Maybe Eysenck has taken that advice to heart.

On the whole, however, the foregoing explanation seems just too Machiavellian. A simpler explanation is that for all his stress on objectivity and scientific rigour Eysenck is human after all and sometimes gets carried away by his passions. Religion and politics have long been notorious for their tendency to expose people as seeing only what they want to see and believing only what they want to believe. Gibson (1981) believes that Eysenck is very political, and I see no reason to disagree. It must be recollected that Prof. Dr. Eysenck is not an Englishman but a Weimar German (perhaps even the last of the great, speculative, German-speaking psychologists). He was born and bred among the very liberal (even libertine) high culture of Weimar Germany. He grew up observing with extreme distaste the roving political street gangs (Right and Left) of pre-war Germany and detests equally their modern-day equivalents. He must find it hard to tell the difference between modern-day anti-Apartheid or anti-Nazi (or even anti-Eysenck) demonstrators and the KPD (Kommunistische Partei Deutschlands) or NSDAP (National Sozialistische Deutsche Arbeiter Partei) street gangs of his youth (see also Ray, 1971b). He has therefore become almost more English than the English. He has embraced almost as a religion the traditionally tolerant, pluralistic, restrained values of English political thought (though the Irish might have a different view of traditional English political practice; many Irish find Adolf Hitler and Oliver Cromwell rather hard to tell apart) and deplores the recurrent brutalities of European political practice with a fervency that would do any Englishman proud. One sign of this extreme Anglophilia is that his post-war letters to his father were written in English -- a language his father hardly understood (Gibson, 1981). I also have a copy of a publication by Eysenck and Levey (1967) in an East German journal which was printed in German bearing the footnote 'Ubersetzung aus dem Englischen von Jurgen Mehl'. In other words, even when he wished to publish in a German journal (had he tried to have it published elsewhere but without success?) he still would not write in German but wrote in English and had someone else translate it for him -- and this from someone who in his youth and teenage years was proud of his literary ability in German! In such circumstances we can see that no-one could be more convinced than Eysenck about how evil political brutality is and how uncompromisingly it must be opposed. His extreme rudeness to Konrad Lorenz when he learned that Lorenz in his day had made his peace with his Nazi overlords (Lorenz was really only interested in ducks and the like; to him a few obeisances to the authorities must have seemed a small price to pay for the peace he needed for his studies of his feathered friends) is an instance of this (Gibson, 1981).

I infer that Eysenck's political passions to this day blind him to the obvious fact that he has failed in his search for psychological evidence of Left-wing authoritarianism. He knows it exists and he knows that he must oppose it. Therefore he concludes that he has in fact found evidence of it and that he has exposed it. He claimed to be able to show that Leftists were 'tough-minded' (a term to him largely synonymous with authoritarianism) but all he managed to show is that they were disrespectful of conventional religion and its concomitant morality. He identifies it as authoritarian to reject religion! Given the way in which men have been (and in some parts of the world still are) enslaved by religious authority for almost all of recorded history, I would have thought quite the opposite -- that to reject religion is normally the first step along the way to a healthy questioning of all authority. With their dogmas, potentates and keys to the kingdom of heaven, it seems to me that conventional Western religions are nothing if not authoritarian. If 'Do as we say or you will go to Hell' is not authoritarian, what is?

Perhaps the tragedy of Eysenck's work in the psychology of politics is that although he was not alone in failing to find evidence of psychological authoritarianism among Leftists, he came very close to making the discovery he so earnestly desired. To see this we must first see why such evidence was not found. Generally, the reasons are threefold: psychologists have assumed that what is true of their students will be pretty similar to what is true of the world as a whole (Ray, 1981); they have assumed that ideology and vote will closely correspond; they have assumed that attitude and personality will go fairly closely together. All these assumptions are suspect when stated so baldly, so let us see what happens when we reject each one of them.

Ray (1972, 1974) records a study wherein fresh conscripts into the Australian Army were given what was initially intended as a balanced F scale. Australian Army conscripts were at the time selected by a random birth-date ballot procedure from the entire Australian population of 18-year-old males. 'Dodging' seems to have been very difficult in comparison with the US experience. This sample, then, was vastly more representative in socio-economic and educational terms than the usual student sample (or, more properly, 'non-sample'). The findings of the study were that the especially-written 'Leftist' items which were supposed to 'balance' the items of the original F ('Fascism') scale correlated highly positively with the original F scale items. Anti-authoritarian items turned out to be highly pro-authoritarian! Using non-student subjects, I was able to show that sentiments such as 'Human beings are more important than efficiency', 'Dictatorships are totally wrong', 'All men are equal' and 'Individual freedom is a basic human right' were all authoritarian (assuming that we concede that the Adorno et al. (1950) F scale does measure authoritarianism). Non-students are obviously very unkind to the preconceptions of psychologists! (It might be noted in passing that analyses were also done to exclude acquiescent response bias as an explanation for these results.) There was certainly no difficulty in finding psychological authoritarianism of a Leftist kind with this sample.

Let us now question the relationship between vote and ideology. In 1973 I reported in the European Journal of Social Psychology a study in which I applied successfully balanced (i.e. acquiescence-free) versions of the California F and Rokeach D scales to a random population sample taken in Sydney, Australia, by door-to-door means. I also asked for intended vote from each respondent. I found that neither the F nor the D scale predicted vote. The inference from this is that high F scorers were just as likely to vote Leftist as Rightist. There were just as many authoritarians of the Left as there were of the Right! Again, there was no shortage of psychological authoritarianism among Leftists. Hanson (1975) also records that the F scale often fails to predict vote.

Let us now question the relationship between attitude and behaviour. So far I have accepted for the purposes of the argument the conventional view that the Adorno F scale does measure authoritarianism. This is, however, a highly dubious proposition. For a start the F scale will not seem to predict behaviour reasonably categorized as authoritarian (Titus, 1968; Ray, 1976). People who tend to boss others around are just as likely to get low F scale scores as high. When the F scale does seem to predict the things it ought, the correlations are usually just as well explained by saying that the F scale is measuring nothing more than conservatism (Ray, 1973d, 1983). In these circumstances I devised a new scale in behaviour inventory format -- i.e., a personality scale -- which did predict authoritarian behaviour highly significantly. I called it the 'Directiveness' scale. It is really a very simple-minded construction, containing items like 'Do you tend to boss people around?' and 'If anyone is going to be Top Dog would you rather it be you?' Studies have now shown that this scale does not have any overall relationship either to political ideology or to political party vote (Ray, 1979, 1982). In other words, both Leftists and Rightists are equally likely to be shown by this scale as highly authoritarian. Other findings of authoritarianism in Leftists are the well-known work of Rokeach (1960) and the extensive series of papers by Rothman and Lichter (see Lichter and Rothman, 1981/82).

I have listed briefly three types of finding to show that, although Eysenck's empirical, work was badly flawed, his basic theory is sound. I take that theory to consist of the view that two rather than one dimensions are needed to describe political allegiance and behaviour, and that the two dimensions are radicalism/conservatism and authoritarianism of some sort.

Perhaps sadly, it seems that for most of the time Eysenck was dimly aware of the three conditions I have specified for turning theory into sound empirical findings. He generally made at least some effort to use community rather than student samples; he noted as early as 1951 that there was something funny about the relationship between vote and ideology; and he maintained all along that his second dimension was really a personality dimension rather than an attitude dimension per se. Some further comment about Eysenck's moves in this direction seems appropriate.

Perhaps because of Christie's abrasive criticisms or perhaps because of increasing command of economic resources, Eysenck's later work would appear to be noteworthy for its use of representative sampling. The samples he used in his 1971, 1975 and 1976 studies seem to have been much the sort used by commercial polls. Only in the case of the 1975 sample, does he give us any useful demographic background on the respondents. We find that a quota sample of London contained 215 females and 153 males. It is well-known that females tend to be over-represented in the cities but this seems gross. If the polling organization had such problems getting the sex quota right, one can only imagine what might have happened to the other quotas. In all these studies, the chance of Left-wing authoritarianism emerging was limited by the simple fact that Eysenck used no measures of authoritarianism other than his peculiar T scale.

On the relationship between ideology and vote Eysenck deserves some kudos for anticipating the well-known essay by Lipset (1960) on 'working-class authoritarianism'. Lipset argues that vote is determined by economic self-interest rather than by ideology and that in consequence economic conservatism will separate out from other forms of conservatism -- with the middle class high on economic conservatism and low on general conservatism, while the working class will be low on economic conservatism and high on general conservatism. Precisely this hypothesis was tested in Eysenck (1951) but was rejected. In the same paper, Eysenck concluded that if vote is controlled the workers are more conservative in general ideology. Later in Eysenck (1975) the basic Lipset thesis was accepted (though without making any reference to Lipset). Thus after many years Eysenck has ended up with three rather than two factors in both the attitude and personality domains. Just as he has added a P factor to E and N in the personality domain, he has added economic radicalism to R (general ideological radicalism) and T ('tough-mindedness') in the attitude domain. It does not yet seem to have dawned on him how largely irrelevant this makes all his previous studies of ideology.

Strictly speaking, it is inaccurate to refer to Eysenck now having three attitude factors. Eysenck has always maintained, that the T factor is not a fully fledged attitude factor in its own right. He describes it as a 'projection' onto (or the influence on) the attitude domain of a personality variable (extraversion). He long maintained, that there was only one true dimension in social attitudes -- radicalism-conservatism. This is a position that Wilson and I have also maintained (Wilson, 1973). At any event, Eysenck's thinking here gives rise to a testable hypothesis. Authoritarians should be extraverted. Strangely, in spite of the interest that Eysenck's theories have provoked, no-one but Eysenck seems to have tested this until very recently. Eysenck's own tests used the dubious T scale so are not conclusive. I therefore (Ray, 1980) correlated both the California F scale (which Eysenck acknowledge as a measure of Rightist authoritarianism) and the directiveness scale with measures of Eysenck's E (extraversion) and found the reverse of what Eysenck had predicted. High F scorers were introverted! The directiveness scale showed no significant correlation with E at all.

Eysenck had what I would claim was the correct insight -- that the second variable in the psychology of politics is a personality rather than an attitude variable -- but was limited by his strait-jacketed notions of what constitute the variables of personality in pursuing the insight fully. It might be objected that since extraversion did correlate significantly with the F scale, Eysenck was not too far off -- he just got the sign of the correlation wrong. Unfortunately, the correlation was a low one (-.12 and -.18 with Eysenck's two subfactors of extraversion), and even then it was probably due to the conservatism component of the F scale rather than anything else (Ray, 1980). In the same study introversion was also found to correlate (more highly) with general social conservatism (Ray, 1984).

In justice, it must be said that there may be some basis outside his own work for Eysenck's view that authoritarians are extraverted. In a study of modern-day neo-Nazis (Ray, 1973a), clear indications suggested that such people were quite extraverted. Modern-day neo-Nazis are however a far cry from the historical German article. Some are motivated as much by a love of uniforms and of shocking people as they are by anything else. Inferences from such deviant groups to the population as a whole would be quite untenable.

Eysenck himself has begun to waffle a bit about what the personality variable underlying T really is. In the concluding chapter of The Psychological Basis of Ideology he describes the psychological basis for political authoritarianism as 'the personality variable P (and possibly E)'. E (extraversion) is being edged out. Even the replacement of E by P, however, does not get Eysenck out of trouble with the evidence. In Ray and Bozek (1981) it was shown on a general population sample that Eysenck's P scale correlates negatively with the F scale (where Eysenck would predict a positive correlation).

It takes no theoretical innovation at all to find the elusive personality variable 'underlying' authoritarianism. Why not start with authoritarianism itself? Both Adorno et al. (1950) and Eysenck seem driven by the need to find hidden signs of authoritarianism in attitudes. They both try to measure authoritarianism by attitude scales. Why not just measure authoritarianism directly by a conventional personality scale that asks questions not about great social issues (attitudes) but rather about how the individual person himself feels and behaves (personality)? The directiveness scale is such a scale (Ray, 1976), and had Eysenck adopted such a straightforward approach instead of the devious, over-clever, indirect approach he did adopt he would have had his ideology-free measure of authoritarianism from the beginning. Surely no-one can question Eysenck's view that political decisions will reflect both the attitudes and the personalities of those involved, so where is the difficulty in saying that of the two key dimensions needed to explain political allegiances and actions, one should be measured by an attitude scale and the other by a personality scale?

That Eysenck did not move in this direction is probably due to the fact that probably all psychologists find it hard to live with the gap between attitudes and behaviour. This gap has been well-known at least since the time of La Piere (1934) and has repeatedly been confirmed (Ray, 1971b, 1976). Just because (for example) a person says, 'Hard work is a good thing', there is absolutely no warrant that the person so saying will himself tend to work hard. Psychologists never seem very happy with such a situation. They always seem to think (or even assume) that there should be at least some tendency for those who think hard work is a good thing to work hard themselves. Reality, unfortunately, seems unco-operative. For Adorno et al. (1950) attitude/behaviour congruity was a straight assumption. They assumed throughout that the people who evinced authoritarian attitudes would also be the ones who tended towards authoritarian behaviour. They were, of course, eventually shown to be quite wrong in this (Titus, 1968; Ray, 1976; Ray and Lovejoy, 1983). Eysenck was a better theorist and was more aware of the role of other factors, but even he seems to have been unable to accept that there was no reflection in attitudes of basic behaviour tendency.

In a sense, he was right. Attitudes and behaviour are not totally isolated from one another (Kelman, 1974). The relationship between them, however, is usually far from simple. It is simple or isomorphic relationships we have to question rather than all relationships. Finding the relationship, in other words, is simply another research task. We cannot assume it, and it need not leap out and grab us. It may be subtle rather than obvious. Even the relationship between personality and behaviour may be far from obvious. Because personality is usually measured by asking people how they characteristically behave in various situations, scores on personality scales usually predict actual behaviour rather well (insofar as the .3 or .4 correlation that is all psychologists can usually aspire to can be regarded as 'rather well'). Nonetheless, one finding that seemed unusually fascinating (and relevant) was that a scale of achievement orientation predicted authoritarian behaviour better than did a scale of authoritarian personality (Ray and Lovejoy, 1983). Both scales were orthodox personality scales yet the 'obvious' predictor of the two was not the stronger. In retrospect the finding seems easy enough to explain. One simply has to make the point that human behaviour is characteristically multi-causal (the same behaviour may be emitted to serve several different ends), and acting in a domineering way towards others may be done either because the person likes doing so as such or because he wishes to use his ability to influence the behaviour of others to achieve some other end. In other words, bossing others around may be just one of the many things the achievement-motivated person may have to do in his scramble towards the top. When we observe authoritarian behaviour happening we must be very careful what inference we make about the motives of the person so behaving. It is possible that he is doing it just because he likes it (the authoritarian), but it is more likely that he is doing it in order to achieve some quite separate materialistic goal that he values. If the relationship between personality and behaviour can be complex and not immediately obvious, how much more so must the relationship between attitude and behaviour be complex and non-obvious? Psychologists have always recognized this complexity at least in part. No-one, for instance, would dispute the importance of situational factors in modulating behaviour. The sort of complexity I am suggesting, however, is, I think, much greater than is normally envisaged. When we find that authoritarian behaviour is not associated with authoritarian attitudes but is associated with achievement motivation, I think it becomes clear how exceedingly simplistic almost all the research in the area has been so far. Unfortunately, even very recent work seems to be largely stuck in the same simplistic mould (e.g., Altemeyer, 1981).

In conclusion, we must say that Eysenck's basic theory was supported. Leftists are quite as likely to be authoritarians as are Rightists. Eysenck even had all the basic insights needed to turn this theory into decent empirical findings. In the end he fell short of the goal and resorted to strange stratagems to conceal his failure. When I was a guest in Eysenck's department in 1977 someone (probably Simon Hasleton) remarked to me that Eysenck was really an old-fashioned grand theorist and not an empiricist at all. I think there is some truth in that observation. I at least will always honour his creativity, even if I cannot accept the interpretation he places on his empirical work.

As some sort of epilogue to this chapter, I think it behoves me to point out that there is a well-developed alternative theory to Eysenck's which, perhaps because it has largely been developed by economists and perhaps because it does not put Leftists in a particularly good light, seems virtually unknown among psychologists. This is the libertarian theory as spelt out in a vast range of publications including von Hayek (1944), von Mises (1949) and Friedman (1962). As two of the authors mentioned gained Nobel prizes for work they did in connection with this theory, it seems very strange to me that psychologists know so little of it.

As background to the theory one needs to note that the normal human form of government is tyranny. For most of human history men have been ruled by kings, emperors, caesars; pharaohs, etc. The exceptions in the ancient world (Athenian democracy, the early Roman republic) were fleeting and, in historical terms, the modern-day alternatives are so recent as to be little more than an eyeblink in the total experience of the human race. Even today most of the human race is ruled by dictatorships of one sort or another. Hard though it may be for us to cope with, it is democracy which is the aberration, not tyranny. Libertarians have an explanation for that aberration and fear that it might be a short-lived one. Libertarians deal with Eysenck's problem of Communist/Nazi similarities by denying that there are any important differences between the two. They further fear that although the Fascists may have lost the war, they have won the peace and that we all one day will be Nazis or Communists.

Let us look again at Nazi/Communist similarities. Although both Hitler and Stalin were quite happy to call themselves Nazis and Communists respectively, both formally characterized their regimes as 'socialist'. Libertarians see no reason to reject that description. Both regimes did proclaim a supremacy of the community's needs over individual needs, and even democratic socialists do the same. If any individual (e.g. a businessman) happens to be standing in the way of what are perceived as community needs, he will get short shrift from democratic socialists, Communists and Fascists alike. The only difference is that, although the democratic socialists may imprison him, they are unlikely to kill him. Democratic socialists also, of course, involve more people in the decision over what community needs and interests are in the first place. There are other similarities: Communists claim to speak for 'the people'. Nazis spoke for 'das Volk', which translates roughly as 'the people'. Communism and Nazism might have different bogeymen ('capitalists' versus 'the Jews') but, at least in pre-war Germany, the same individuals would often be caught under either rubric. Hitler wanted to ban Einstein's physics because it was 'Jewish'. Stalin wanted to ban Einstein's physics because it was 'bourgeois' (Eysenck and Wilson, 1978).

In the libertarian view Nazism was simply a less full-blown and more simple-minded version of Communism. (The Israeli scholar, Unger (1965), also argues that Nazism was less totalitarian than Communism.) Both exerted extensive controls over the whole of society and ran the economy at State command. It could be argued that Nazism was in some ways a more successful form of socialism than Communism. Nazis had to rely less on repression within Germany precisely because their methods and rhetoric were more genuinely congenial to the German people than were the methods and rhetoric of Stalin to the Russian people. Even anti-Nazi pre-war writers such as Roberts (1938) acknowledge that Hitler was the most popular man in Germany It the time. The rivalry between Nazism and Communism was sibling rivalry: between national socialism and international (really Russian) socialism.

In a sense, both Hitler and Stalin have had the last laugh. Both stood for the replacement of individual decisions by government, party or bureaucratic decisions. Yet, at the hands of the democratic socialists, all countries of the world have since the Second World War been marching in precisely that direction. Even in the United States the extent to which the government has taken over the spending of the national income since the Second World War is staggering. In England all major industries are owned and run by the government, and government is the major provider of accommodation for the masses. The economy of Britain is now surprisingly similar to that of Eastern bloc countries such as Hungary. So entrenched is socialism in England that even the energetic Mrs Thatcher has been unable to budge it. Even freedom of speech is restricted in the UK for those who have uncongenial ideologies -- as the National Front member found who put up a sign in his front yard saying: 'House for sale: Whites only'. He was imprisoned for his pains. Dislike of blacks is widespread in Britain (Britain is the only predominantly Anglo-Saxon country with an explicitly racist political party mounting major campaigns in national elections) but expression is officially repressed. No doubt all, the things that democratic socialists have done were done with good intentions, but let us remember that Hitler and Stalin also claimed good intentions.

The great prophet of the twentieth century was probably Mussolini. His ideas of course preceded Hitler's, and most of the once-decried features of his corporatist State are now commonplace in advanced 'Western' countries. In a sense what Britain and other 'Western' countries are doing is returning to a historical norm at an only slightly slower pace than that adopted by Communists and Fascists. That norm is collectivism and the form of government that goes with it is, at best, paternalistic. The ancient civilizations of the Nile and the Euphrates were remarkably like ant colonies, and we may end up that way again. Human beings evolved as essentially social, cooperating, labour-sharing animals, and there has always been a perceived need for someone to direct the labour and share out the product. Kingship or aristocracy has been the usual way of legitimating individuals in that role. Nazism had more room in its ideology for quasi-divine leaders and privileged elites, and was more in harmony with historical human practice, but, despite all ideology, Communists have ended up with a similar system. The auguries for other political systems are not good. Collectivism has the sort of universal attraction of a return to the womb: others take the responsibility for one's basic physical needs and in return one forgoes most of one's abilities and opportunities for individualistic, independent, responsible endeavour.

It may seem that the foregoing account has placed far too much emphasis on the economic system as the touchstone of what a society is like. This leading role of the economic system is a basic libertarian thesis. Historically, a few peoples on the fringes of North-West Europe made the transition from a hunting society to civilization rather quickly (at various times in the last 2000 years), and in the process some of the independence of mind more appropriate to a life of hunting carried over into the new life. This independence was only a leaven as the civilizations these people founded did produce from time to time tyrants as powerful and as brutal as any (e.g., England's Henry VIII). Nonetheless, there was generally more decentralization of power (among the barony and others) than in an Oriental despotism, and the individualism this allowed transformed outside influences (particularly the rebirth of learning after the fall of the Byzantine Empire) into first the Protestant Reformation and then finally into the English Industrial Revolution. One of the central pillars of State power (State religion) was undermined and new sources of power (money in the hands of the bourgeoisie) independent of the State were created. Of these two great revolutions, however, only the latter brought about fundamental social changes. After the Reformation, Protestant princes simply took over from where Catholic princes had left off. The system of social hierarchy and patronage remained essentially unaltered. A seed had been sown in that revolution from the bottom up had been shown to be possible but there were to be long travails before the type of social organization actually changed. It is therefore to the economic revolution, the industrial revolution, that we must look for the beginnings of the modern world.

During the time that the power of princes was still weak from the conflicts of the Reformation, there was more opportunity for the individualism of the North-West European peoples to show through in other ways. People began to experiment with new ways and devices for making money. They improved their traditional machines and processes for spinning, weaving, mining and metal-working, and got away with it because the despot was too weak from other conflicts to jump on such apparently minor displays of individualism. The rest, as they say, is history. The process of innovation was enormously profitable and underwent exponential growth. Almost before anyone had realized it, a new phenomenon -- capitalism -- was born. Money and power always walk hand in hand, and all these new holders of substantial wealth (the capitalists) were a new, highly fractionated and quite subversive source of power and influence. They guarded jealously the independence that had made their wealth possible, and traditional society, weakened as it was by religious strife, could not meet such an unprecedented challenge. In the Victorian era laissez-faire capitalism came to open political power. If capitalism took traditional society by surprise, however, the effects of capitalism were also a surprise to most of the bourgeoisie. Not only was a new independent middle class created, but the affluence and its consequent independence after a while trickled down to the working class as well. Giving the worker the opportunity to change his job, his occupation, his employer completely destroyed the basis of feudal power, and it soon destroyed the power of the capitalists as well. Power was again fractionated and the bourgeoisie in turn lost its leading role. Politics became mass politics.

Thus we arrive at the modern era. The workers have no more grasp of the mechanisms of capitalism than did the old land-owning elites, and they hanker for a return to the womb. Twentieth century socialism in its various forms (Bolshevik, democratic and Fascist) is the response. The almost accidental flowering of capitalism transformed the world in a few short years but it was an unnatural flowering by human standards and the heat of disapproval is already withering it. Thus a highly collectivized but popularly accepted society that uses technology but is not dominated by it (such as Hitler's Germany) would seem to be what the future holds for the advanced nations of today.

In all the above little has been said of the role of conservatism. Socialism has been painted in fairly bleak terms. Is conservatism the great white hope? Far from it. Libertarians take a fairly orthodox view of conservatives as simply cautious, careful people whose values are of the past. But since the past contains both libertarian elements (in the economic sphere) and repressive elements (in the sphere of religion and morality), conservatives have no consistent theme and undo the good they do by giving occasional support to liberty in one sphere by trying to repress liberty in other spheres. Socialist repression has an air of inevitability about it. Conservative repression is a functionless shadow of a moral system that once had the important economic function of ensuring that all children were born with a father to provide for them. In time, then, conservatives too will come to accept collectivism as part of what was and therefore must be, and even the limited conservative opposition to collectivism will die.

Is there any hope for the individual in our future? Perhaps there is. Perhaps the egg once broken can never be completely put back together again. Human beings constantly do many unnatural things (such as wearing clothes), so the fact that tyranny is man's natural form of government need not mean that it is his inevitable future. Libertarians certainly do what they can to slow down the ever-advancing power of the bureaucratic State (e.g., Green, 1982).

If the Libertarians are right, not even Eysenck's theory can modify the traditional Right-Left view of politics into a workable description of reality. There is no static Right-Left divide or any other static political polarity. Instead all we have is a continuing dynamic progress towards restoring an equilibrium that was accidentally disturbed by a completely unforeseen and unprecedented historical event. When that restoration is accomplished, all politics will revert to what has always been their essential element -- competition for power among individuals.


Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J. and Sanford, R. N. (1950) The Authoritarian Personality. New York: Harper.

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

Christie, R. (1956a) 'Eysenck's treatment of the personality of Communists', Psychological Bulletin, 53, pp. 411-30.

Christie, R. (1956b) 'Some abuses of psychology', Psychological Bulletin, 53, pp. 439-51.

Christie, R. and Jahoda, M. (1954) Studies in the Scope and Method of "The Authoritarian Personality", Glencoe, Ill., Free Press.

Christie, R., Havel, J. and Seidenberg, B. (1956) 'Is the F scale irreversible?', J. Abnorm. Soc. Psychol, 56, pp. 141-58.

Eysenck, H. J. (1940) 'Some factors in the appreciation of poetry, and their relation to temperamental qualities', Character and Personality, 9, pp. 160-7.

Eysenck, H. J. (1944) 'General social attitudes', J. Social Psychol, 19, pp. 207-27,

Eysenck, H. J. (1951) 'Primary social attitudes as related to social class and political party', Brit. J. Sociology, 2, pp. 198-209.

Eysenck, H. J. (1954) The Psychology of Politics. London: Routledge.

Eysenck, H. J. (1971) 'Social attitudes and social class', Brit. J. Soc. Clin. PsycholĄ 10, pp. 210-12.

Eysenck. H. J. (1975) 'The structure of social attitudes', Brit. J. Soc. Clin. Psychol., 14, pp. 323-31.

Eysenck. H. J. (1976) 'Structure of social attitudes', Psychol. Reports, 39, pp. 463-6.

Eysenck, H. J. (1981;82) 'Left-wing authoritarianism: Myth or reality?' Political Psychology, 3, (1/2), pp. 234-238.

Eysenck, H. J. and Eysenck, S. B. G. (1976) Psychoticism as a Dimension of Personality, London, Hodder and Stoughton.

Eysenck, H. J. and Levey, A. (1967) 'Konditionierung, Introversion-extroversion und die Staerke des Nervensystems', Zeitschrift fuer Psychologie, 174, pp. 96-106.

Eysenck, H. J. and Wilson, G. D. (1978) The Psychological Basis of Ideology Lancaster, MTP Press; Baltimore, Md., University Park Press.

Friedman, M. (1962) Capitalism and Freedom, Chicago, Ill., University of Chicago Press.

Gibson, H. B. (1981) Hans Eysenck: The Man and His Work, London, P. Owen.

Green, D. G. (1982) 'Arguments for liberty', Political Quarterly, 53, pp. 418-32.

Hanson, D. J. (1975) 'Authoritarianism as a variable in political research, Il Politico, 40, pp. 700-5.

Hayek, F. A. von (1944) The Road to Serfdom, London, Routledge.

Jaensch, E. R. (1938) Der Gegentypus, Leipzig, Barth.

Kelman, H. C. (1974) 'Attitudes are alive and well and gainfully employed in the sphere of action', Amer. Psychologist, 29, pp. 310-24.

La Piere, R. (1934) 'Attitudes and actions', Social Forces, 13, pp. 230-7.

Lichter, S. R. and Rothman, S. (1981/82) 'Jewish ethnicity and radical culture: A social psychological study of political activity', Political Psychology, 3, pp. 116-57,

Lipset, S. M. (1960) Political Man, New York, Doubleday.

McKinney, D. W. (1973) The Authoritarian Personality Studies, The Hague, Mouton.

Mises, L. von (1949) Human Action, London, Hodge.

Ray, J.J. (1971a) Correspondence: Regarding the Lynn n-Ach test. Bulletin British Psychological Society, 24, 352.

Ray, J.J. (1971b) Ethnocentrism: Attitudes and behaviour. Australian Quarterly, 43, 89-97.

Ray, J.J. (1972) Militarism and psychopathology: A reply to Eckhardt & Newcombe J. Conflict Resolution, 16, 357-362.

Ray, J.J. (1973a) Antisemitic types in Australia. Patterns of Prejudice 7(1), 6-16.

Ray, J.J. (1973b) Dogmatism in relation to sub-types of conservatism: Some Australian data. European J. Social Psychology 3, 221-232.

Ray, J.J. (1973c) Factor analysis and attitude scales. Australian & New Zealand Journal of Sociology 9(3), 11-13.

Ray, J.J. (1973d) Conservatism, authoritarianism and related variables: A review and an empirical study. Ch. 2 in: G.D. Wilson (Ed.) The psychology of conservatism London: Academic Press.

Ray, J.J. (1974) Authoritarian humanism. Ch. 42 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.

Ray, J.J. (1979) Does authoritarianism of personality go with conservatism? Australian Journal of Psychology 31, 9-14.

Ray, J.J. (1980) Are authoritarians extroverted? British Journal of Social & Clinical Psychology 19, 147-148.

Ray, J.J. (1981) Is the ideal sample a non-sample? Bulletin of the British Psychological Society 34, 128-129.

Ray, J.J. (1982) Climate and conservatism in Australia. Journal of Social Psychology, 117, 297-298.

Ray, J.J. (1983) The workers are not authoritarian: Attitude and personality data from six countries. Sociology & Social Research, 67 (2), 166-189.

Ray, J.J. (1984) Political radicals as sensation seekers. J. Social Psychology 122, 293-294.

Ray, J.J. & Bozek, R.S. (1981) Authoritarianism and Eysenck's 'P' scale. Journal of Social Psychology, 113, 231-234.

Ray, J.J. & Lovejoy, F.H. (1983). The behavioral validity of some recent measures of authoritarianism. Journal of Social Psychology, 120, 91-99.

Roberts, S. H. (1938) The House That Hitler Built, New York: Harper.

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

Rokeach, M. and Hanley, C. (1956) 'Eysenck's tender-mindedness dimension: A critique', Psychological Bulletin, 53, pp. 169-76.

Titus, H. E. (1968) 'F scale validity considered against peer nomination criteria', Psychol. Record, 18, pp. 395-403.

Unger, A. L. (1965) 'Party and State in Soviet Russia and Nazi Germany', Political Quarterly, 36, pp. 441-59.

Wilson, G. D. (1973) The Psychology of Politics, London, Academic.


In the 1980s -- when the above was written -- I was a more extreme libertarian than I am now. I tended to think then that a full anarcho-capitalist programme was viable. Since then I have moved back more towards my original free-market conservative views. I therefore no longer see any merit in the two dimensional view of politics that both libertarians and Eysenck espouse and I now accept the normal conservative claim that they are more liberty-oriented than Leftists. For details of why I take that view see here and here and here

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 "Food & Health Skeptic" 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. Backup here)
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)