AGGRESSIVENESS, DOMINANCE AND RIGIDITY

These results suggest three alternative hypotheses. It is possible that the differentiation between Communists and Fascists shown by their scores on the aggression and dominance variables is characteristic of Radicals and Conservatives generally, so that Radicals would be found to be directly dominant and indirectly

figure 35

Scores of Fascists, Communists, and Neutral Group with Respect to Dominance and Aggression aggressive, while Conservatives would be indirectly dominant and directly aggressive. It is possible to test this hypothesis by correlat­ing the R scores of the 84 members of the neutral group with the aggression and dominance scores. The results are not in line with the hypothesis, as none of the correlations are significant. As a further test of the hypothesis, D x and D 2 scores were derived for207

the members of the ‘neutral5 group in the manner indicated in the last paragraph; these scores were then correlated with R. Cor­relations again were insignificant. There appears to be little likeli­hood, therefore, of any determination of Radical or Conservative attitudes through the aggressive-dominant complex of traits.The second hypothesis would lead one to speculate along the following lines. Parties change in character according to historical circumstances; their membership is determined by economic and political factors which it would be very difficult to disentangle. Nevertheless, it seems reasonable to suppose that if personality factors such as the need for aggression and dominance are respon­sible to some degree for a person’s choice of party, these traits will be shown most clearly by members of the party when it is relatively small and far from the seats of power. At that time, self-seeking and time-serving motives cannot in the nature of things be present, and the membership will therefore be something of a pure culture of these personality traits for which the party policy is attractive. As the party grows in power, other people will be attracted to it for a great variety of reasons, and this process of dilution will make the average member less and less distinctive from the point of view of his personality structure. The Fascist party is so small as to be almost non-existent; consequently we might expect almost all its members to be characterized by certain personality traits. The Communist party is much larger, and consequently we should ex­pect a certain degree of dilution already in the personality picture. This might explain the fact that while both Communists and Fas­cists show aggressive and dominant traits, as well as rigidity and intolerance of ambiguity (as will be shown later), yet the Fascists show all these traits somewhat more clearly than the Communists.The difference in size between the two parties may also deter­mine certain aspects of their policy. The Communist party, suf­ficiently large to have some influence on the political life of the nation, has nothing to gain and everything to lose from displays of open violence and aggression; the Fascist party, on the other hand, has nothing to lose, and may indeed attract attention, by displays of aggression. Thus the emphasis of teaching and indoctrination is towards direct aggression in the Fascist group, towards indirect aggression in the Communist group. A dramatic illustration of this fact is furnished by the discovery that out of the sample of 43 Fascists studied, four had been members of the Communist party

who had changed over because they had not found appropriate outlets there.It would be difficult to prove an hypothesis of this type; indeed, it would be quite impossible to do so in this country. If we postulate politico-economic causes to account for the size and policy of our two parties, then experimental manipulation of these causes is eo ipso ruled out. It should not be impossible, however, to repeat studies of this kind in other countries-Germany might be an ex­cellent testing-ground of such hypotheses-where the historical and politico-economic pattern is different. Until such further ex­periments are carried out, however, it should be emphasized that while the facts reported here are unlikely to be incorrect, their ex­planation is speculative, and may be altogether along the wrong lines.A further hypothesis makes use of the concept of intelligence. Although no intelligence test was given to the Communist and Fascist group because of pressure of time, results of some of the tests used, as well as observation and discussion, suggested strongly that the Communists were more intelligent than the Fascists. In this connection, it is of interest to note that in the neutral group, where intelligence tests were applied, the following correlations were found: Direct dominance and intelligence = -201; Indirect dominance and intelligence = — 163; direct aggression and in­telligence = —*211; indirect aggression and intelligence = *261. Thus the more intelligent person appears to be characterized by direct dominance and indirect aggression, while the stupid person is characterized by indirect dominance and direct aggression. This picture of the intelligent person’s trait-organization thus agrees with that of the Communist, while that of the stupid person agrees with that of the Fascist. Again, in view of the fact that no actual measurement of intelligence was carried out on the two political groups this hypothesis cannot be regarded as proved; replication of the experiment with the addition of an intelligence test appears to be called for.There is one objection to accepting the facts reported here as being a decisive proof of the hypothesis linking tough-mindedness with aggression and dominance. What we have done, in essence, is to take two minority groups, advocating certain policies which are unorthodox and unusual, and compare them with a neutral or orthodox group. It might be argued, quite reasonably, that

possibly all minority groups advocating unorthodox policies would be found to be dominant and aggressive; indeed, many writers have commented on the prevalence of these traits in Pacifists, fundamentalists, and other groups having no connection with Communism or Fascism. Fortunately we have a method of proof which obviates this objection. If the quality of tough-mindedness, which distinguishes both’Communists and Fascists, is indeed re-

TABLE xxvi Correlations of Dominance and Aggression with Radicalism, Tender-Mindedness and Rigidity

lated to aggression and dominance, then we should expect that in our neutral group the T-score would show correlations with these personality traits. The correlations between dominance and ag­gression on the one hand, and tough-mindedness on the other, are therefore of crucial importance.Table XXVI gives the correlations between Radicalism, Tender­mindedness, F-scale, a Rigidity and an Intolerance of Ambiguity scale to be described later, and the direct and indirect Dominance and Aggression scores. Correlations are shown separately for the 2 1 0

three groups; the top figure in each case applies to the ‘neutral’ group, the figure on the left below to the Communist, and the figure on the right below to the Fascist group. The following fea­tures of this set of correlations will be noted: (i) There is no con­sistent relation between Radicalism and Aggression, with the pos­sible exception that Radicals may be slightly less directly aggressive and slightly more indirectly aggressive. A repetition of the study would be required to render this conclusion secure. (2) Tender­mindedness shows a strong negative relationship with both domin­ance and aggression. There is only one real exception to this trend, viz. the positive correlation between direct aggression and tender­mindedness in the Communist group. Possible causes of this re­versal have already been discussed. (3) The pattern of correlations with the F-scale is very similar to that shown by the T-scale, though of course all the correlations have opposite signs. This was expected in virtue of the relatively high negative correlations be­tween the F-and T-scales, reported in Table XXIX. (4) A tend­ency is found for rigidity and intolerance of ambiguity to correlate with dominance and aggression; these correlations will be dis­cussed after we have introduced these concepts in some detail. (5) Direct and indirect dominance correlate quite highly together for all groups; they also show somewhat lower positive correlations with direct aggression, and variable correlations of doubtful signi­ficance with indirect aggression. (6) Direct and indirect aggression show negative correlations for the neutral group and the Com­munists, but a positive correlation for the Fascists. It almost ap­pears as if the Fascist group were so full of aggression that instead of showing it in either of these two ways, as do the other two groups, they must show it in both ways at once.The result of this study then strongly confirms those predicted on the basis of previous work and of common observation. Tough­mindedness correlates strongly with aggression and with domin­ance. This tendency can perhaps be seen most clearly in Figure 36 which sets along the abscissa scores on the R-factor and along the ordinate aggressiveness scores derived from combining indirect and direct aggression. It will be seen at a glance that all the Com­munists have aggressiveness scores higher than the mean of the neutral group, while all but five of the Fascists have scores higher than the mean of the neutral group. Comparison with Figure 26 in which we have plotted in a similar manner the relationship be-

tween the three groups along the tough-mindedness axis will show that aggressiveness scores separate out the two tough-minded groups from the neutral group almost as well as does our tough-mindedness scale.One other feature will be noticed in this diagram. Aggressiveness

figure 36

Aggression Scores of Communists, Fascists, and Neutral Group Plotted Against Radicalism-Conservatism 212

scores of the Fascists range all the way from 8 to 30, in other words there is a tremendous diversity in the Fascist group going from extreme aggressiveness to comparative unaggressiveness. All the Communists’ scores on the other hand lie between 13 and 19. In other words all the Communists in the sample lie within a very narrow range indeed-7 points as compared to 23 for the Fascists!This tendency of Communist groups to show less scatter had also been observed with other variables, in particular the various atti­tude scales which they had filled in, and it had seemed likely that this close agreement might be due to the ‘party line’ and the con­siderable amount of indoctrination which is known to take place in Communist circles. The results shown in Figure 36 make this hypothesis rather unlikely as it does not seem very probable that Communists receive any kind of indoctrination in responding to selected pictures on the T.A.T. The only other hypothesis which suggests itself is that the Communist Party, having a well known, clearly defined programme and policy, tends to attract people who are rather alike with respect to their attitudes and to their degree of aggressiveness. Even if we accept this hypothesis, however, it is still not clear why the same should not be true of the Fascist Party whose strongly aggressive, anti-Semitic and anti-Socialist policy is also well known and would certainly not be a surprise to those taking the trouble to join the Party.*Whatever the answer to this question may be, we may conclude that the data summarized in Figure 36 leave little doubt about the essential correctness of our hypothesis regarding aggressiveness. With respect to dominance, Figure 37 shows combined dominance scores as plotted against Radicalism-Conservatism. It will be ob­served that here differentiation is less successful, although still in line with prediction. The Communist group shows an even greater scatter than does the Fascist group, so that clearly whatever may be the causes responsible for the small scatter of Communist scores on the aggressiveness variable cannot be active in the case of dominance.We must next turn to a consideration of a theory which has re-

ceived widespread acclamation and which, if true, would account for many features of anti-minority prejudice in terms of aggressive­ness. This is the famous scapegoat theory of prejudice.This theory is intimately linked with another originally sug-

Dominance Scores of Communists, Fascists, and Neutral Group Plotted Against Radicalism-Conservatism

gested by McDougall and later adopted by Freud. This theory, the so-called frustration-aggression hypothesis, has been developed ex­tensively by the Yale school, particularly by Dollard and Miller; it maintains in essence that aggression is the inevitable consequence of frustration and that all aggression is due to some kind of frustra­tion.23 The scapegoat theory maintains that the individual high in prejudice has a certain amount of hostility or aggression which he has not been successful in expressing or acting out against the original object of aggression, i.e. the original cause of the frustra­tion which produced the aggression. The theory holds further that the person succeeds in reducing his hostility by displacing or re­directing aggression against the more or less helpless members of minority groups in the form of prejudiced behaviour.We need not take the frustration-aggression hypothesis in its pure form too seriously. (It has, indeed, been revised by its earlier proponents.) We may admit that frustration may have other con­sequences than aggression, and we may admit that aggression may have other causes than frustration. Nevertheless, taking the theory in its most general form it cannot be dismissed on any a priori grounds and certain experiments indeed speak in its favour. Thus in one study, questionnaires concerned with attitudes towards Mexicans and Japanese were administered before and after a long series of difficult examinations. At the same time that the subjects (boys in a summer camp) filled out these questionnaires they were forced to miss a highly prized social activity. After this experi­mental frustration there was a significant increase in the amount of unfavourable opinion expressed against the two minority groups. In this study then we appear to find a certain amount of support for the frustration-aggression theory. It should be noted, however, that this experiment fails to provide evidence for one feature which lies at the heart of the scapegoat theory, namely differential displace­ment. In other words the scapegoat theory demands that those who are high in prejudice should be more liable to displace aggression than those who are low in prejudice; the experiment merely demonstrates that on the average some displacement takes place under the experimental conditions of frustration.An experiment to test this additional hypothesis was carried out by Gardner Lindzey. This author attempted to formulate the scapegoat theory in a reasonably rigorous form; he then made a number of predictions from it and proceeded to test them. His

main statement of the theory was as follows: ‘The high in pre­judice have developed a strong preference for reducing aggressive tendencies by means of displacement to an object other than the original instigator of the aggression, while the low in prejudice have not developed such preference for this means of adjustment.5As an alternative hypothesis Lindzey proposes the following: ‘The high in prejudice differ from the low in prejudice not in tendency toward displacement of aggression but rather in the amount of aggression they must reduce or act out. This possibility may be broken down into three special causes, depending upon the place of origin of this surplus of aggression: (a) the greater amount of aggression is a result of constitutional determinants or early developmental experiences that have left the individual with an enduring aggressive need or trait; (b) the greater amount of ag­gressive tendency is a result of chronic and persistent objective frustrations in the environment of the individual that result in his being exposed to excessive frustration; (c) the greater amount of aggressive tendency is the result of the individual’s sensitivity to frustration which results in his being more severely frustrated than the average person in the same objective situation.5Lindzey tested these hypotheses in the following way. Having selected groups of high and low scorers respectively on an ethno-centrism questionnaire he proceeded to measure their aggressive tendency by means of two tests which we have already encoun­tered. One is the Thematic Apperception Test, the other the Picture Frustration Test which consists essentially of a number of pictures like those shown on pages 155 and 201. These measures of aggressive tendency were administered to the subjects at the begin­ning of the study and readministered immediately following the experimental frustration situation which took place about two months after the initial testing. This frustration experience was designed to evoke the maximum amount of frustration and ag­gression. The subjects were made to go without food for from ten to twelve hours; they were induced to drink from a pint to a quart of water and prevented from urinating for approximately three hours; a blood sample was taken with a sterilized spring lancet in such a way as to cause considerable pain. Finally they were made to fail at an assigned task in a group situation when highly moti­vated to succeed, and when their apparent failure caused them to lose status with the other members of the group, all of whom were

accomplices of the experimenter, a circumstance unknown to the subjects of course. This failure in the group situation was the most crucial part of the frustration situation, the earlier physiological assaults being designed chiefly to lower the subject’s threshold of annoyance or frustration.The following deduction made from the hypothesis was con­firmed by the data. ‘Aggressive tendencies that are denied expres­sion against the object originally serving as instigator will be dis­placed so that they will be directed against non-instigating objects.’ On both the T.A.T. and the Picture Frustration test the subjects showed greater aggressiveness after the frustrating situation than before, thus giving evidence that they had displaced the aggres­siveness, which they could not openly show towards their insti­gator, on to the test material. This result is in line with the study already mentioned. The second deduction from the scapegoat theory, however, was not confirmed. This deduction reads as fol­lows: ‘Individuals high in minority group prejudices will show more tendency towards displacement of aggression following frus­tration than those low in minority group prejudices.’ Contrary to this deduction the data actually showed that the high in prejudice displaced less than the low in prejudice, although this difference was not large enough to be statistically significant. ‘The failure to confirm this proposition suggests that the difference between the high and the low in prejudice must lie elsewhere than in the tend­ency towards displacement of aggression.’Lindzey’s third deduction is similar in nature to the propositions already discussed and states: ‘Individuals high in minority group prejudice will show more evidence of outwardly directed aggressive tendencies than those low in minority group prejudice.’ We have quoted a good deal of evidence in favour of this hypothesis and in these data, too, the high in prejudice show more aggression both on the T.A.T. and on the Picture Frustration test than those low in prejudice, although the differences here are not as impressive as they are in our own data.Lindzey’s fourth deduction reads: ‘Individuals high in minority group prejudice will show more frustration susceptibility as mea­sured by the extent to which frustration is experienced as sub­jectively frustrating than those low in minority group prejudice.’ Data to confirm this deduction were obtained by interviewing the subjects, from observing their reactions, and in other similar ways.